From 8a0df09bcb097cc710fd5c82bf48af0c4797774d Mon Sep 17 00:00:00 2001 From: Francesco Mecca Date: Sat, 4 Jul 2020 12:48:04 +0200 Subject: [PATCH] esercizi --- ...opy_for_students_aa_19_20-checkpoint.ipynb | 12813 +++++++++++++++- .../esercizi/meo/3-clusters.csv | 1 + .../esercizi/meo/CURE-complete.csv | 1 + ...lustering-copy_for_students_aa_19_20.ipynb | 12813 +++++++++++++++- .../esercizi/meo/dataset-DBSCAN.csv | 1 + 5 files changed, 25525 insertions(+), 104 deletions(-) create mode 120000 anno3/apprendimento_automatico/esercizi/meo/3-clusters.csv create mode 120000 anno3/apprendimento_automatico/esercizi/meo/CURE-complete.csv create mode 120000 anno3/apprendimento_automatico/esercizi/meo/dataset-DBSCAN.csv diff --git a/anno3/apprendimento_automatico/esercizi/meo/.ipynb_checkpoints/clustering-copy_for_students_aa_19_20-checkpoint.ipynb b/anno3/apprendimento_automatico/esercizi/meo/.ipynb_checkpoints/clustering-copy_for_students_aa_19_20-checkpoint.ipynb index fea11f6..704609e 100644 --- a/anno3/apprendimento_automatico/esercizi/meo/.ipynb_checkpoints/clustering-copy_for_students_aa_19_20-checkpoint.ipynb +++ b/anno3/apprendimento_automatico/esercizi/meo/.ipynb_checkpoints/clustering-copy_for_students_aa_19_20-checkpoint.ipynb @@ -21,7 +21,7 @@ }, { "cell_type": "code", - "execution_count": 22, + "execution_count": 1, "metadata": {}, "outputs": [ { @@ -35,11 +35,11 @@ "n_samples=\n", "86558\n", "dataset n. 1: n samples, n features\n", - "(150, 2)\n", + "150 2\n", "dataset n. 2: n samples, n features\n", - "(6118, 2)\n", + "6118 2\n", "dataset n. 3: n samples, n features\n", - "(86558, 2)\n" + "86558 2\n" ] } ], @@ -49,6 +49,8 @@ "from os.path import join\n", "\n", "import numpy as np\n", + "color=['b','g','r','c','m','y','k','w']\n", + "\n", " \n", "# this function reads the data file, loads the configuration attributes specifiefd in the heading\n", "# (numer of examples and features), the list of feature names\n", @@ -109,12 +111,14 @@ "outputs": [ { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfgAAAI1CAYAAAA6tuiIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3XuUXWWd5vHnkQjIJSGRmEkIQxgbcSW0F6ak8dZNA7ao\njGHNmtWD02i0WcPoOIq92gtoj7fxQne7HJ3VSxwG1EyD0CzEJm23DnR0dBwFLFAREuhEQQhJSDkI\nAS/I5Td/7F1yrNTlnKrznr3fd38/a9Wqc/bep877VqV+v9rvc/aJI0IAAKAsT2l6AAAAYPho8AAA\nFIgGDwBAgWjwAAAUiAYPAECBaPAAABSIBg9gaFz5rO2f2r6xj+PX2A7bi0Yxvvmy/VLbdzQ9DmAQ\nNHh0lu27bP/C9kO2H7D9LdtvtN3X78WomlPq57G9v+2r6u9H2D5pAV/uJZJeJml1RJwwnBFWbJ9k\ne8cwv2a/IuL/RMSx/Rzb5DiBXjR4dN2/iohDJR0l6QJJ75J0SbNDasQ3JZ0lafcCv85Rku6KiJ8t\nfEgAFoIGD0iKiAcjYpOkfytpg+3jJMn2q2x/1/Ze2/fYfn/Pw75Rf37A9sO2X2j7mba/avv/2f6J\n7ctsHzb5ANvvsn1vvWpwh+1T6u1PsX2e7R/Wj73S9rKZnmfq+G2/v37M/6y/9m22x/qc+68i4hMR\n8U1Jj891vO1VtjfZvt/2dtv/vt5+tqSLJb2wHucHpnnsfrY/Vn9vfiTpVVP2v8H21noOP7L9H+rt\nB0v6sqRV9dd+uB7HCba/Xa/A7LL9V7b3n2Hckysh59jeWR//9p79B9j+RL1vZ337gHrfb5yV16sd\nb7d9i+0Hbf+N7QPnGOd4/e/oPtsfn+v7DCxYRPDBRyc/JN0l6dRptt8t6U317ZMk/baqP4afI+k+\nSWfU+9ZICkmLeh77W6qWqA+QtFxVc/5Eve9YSfdIWtXz+GfWt8+VdL2k1fVj/7uky2d6nmnG/H5J\nv5T0Skn7SfqopOvn8T3ZIemkOY75hqRPSTpQ0vMkTUg6ud73eknfnOWxb5R0u6QjJS2T9LXeualq\n+M+UZEm/J+nnko7v+VnsmPL1/qWkEyUtqr9PWyW9bYbnnvw+Xi7p4PrnOjH5b0DSB+ufwTPqn923\nJP2X6Z67/rdzo6RV9Ty2SnrjLOP8tqTX1rcPkXRi0//++Sj/gzN4YF87VRVtRcT/jogfRMQTEXGL\nqubwezM9MCK2R8R1EfFIRExI+njP8Y+rat5rbT81Iu6KiB/W+94o6T0RsSMiHlHVsP/NgLn7NyPi\nHyLicUl/Lem5Azy2L7aPlPRiSe+KiF9GxPdUnbW/rs8v8Yeq/uC5JyLuV/WHyK9FxN9HxA+j8nVJ\n10p66UxfLCJuiojrI+KxiLhL1R9GM/58ah+IiJ9FxA8kfVbSa+rtfyTpgxGxp/7ZfUDSa2f5Ov8t\nInbW8/g7VX/szORRSb9l+/CIeDgirp9jjMCC0eCBfR0h6X5Jsv07tr9me8L2g6oa8eEzPdD2CttX\n1MvweyVdOnl8RGyX9DZVzXtPfdyq+qFHSfpivdT8gKozwsclrRhg3L35+c8lHZjghXmrJN0fEQ/1\nbPuxqu9Zv4+/Z8pjf832K2xfXy//P6BqRWK27/ezbH/J9u76+/2R2Y6vTX3+yZ/Bqinj6d03nanf\n70NmOfZsSc+SdLvt79g+fY4xAgtGgwd62H6Bqmb1zXrT5yVtknRkRCyR9GlVy8dStdw71Ufq7b8d\nEYtVvXBt8nhFxOcj4iWqGnpI+vN61z2SXhERh/V8HBgR987wPE3ZKWmZ7UN7tv1zSff2+fhdqpbn\nex8rqcrAJX1B0sckrYiIwyT9g2b/fl+oasn/mPr7/e6e42cy9fl31rd3qvq5TLdvEPuMMyK2RcRr\nVC3//7mkq+q8HkiGBg9Isr24Pqu6QtKl9fKtJB2q6oz1l7ZPkPTveh42IekJSf+iZ9uhkh6W9KDt\nIyS9o+c5jrV9ct3IfinpF/XjpeoPhw/bPqo+drnt9bM8z1DVLzA7sL67f/2CsX0aZUTcoyqb/mh9\nzHNUnZ1e2udTXSnprbZX214q6byeffurijAmJD1m+xWS/qBn/32Snm57Sc+2QyXtlfSw7WdLelMf\nY/jPtg+yvU7SGyT9Tb39ckl/Vn/vD5f03gHm1Wufcdo+y/byiHhC0gP15iemfTQwJDR4dN3f2X5I\n1Rn0e1Rl5m/o2f8fJX2wPua9qhqUJCkifi7pw5L+b720fqKq3PZ4SQ9K+ntJV/d8rQNUXYr3E1XL\nu8+QdH6975OqVgqurZ/rekm/M8vzDMT2l22/e5ZD7lD1B8cRkv5XffuoGY59jaoXrO2U9EVJ74uI\nf+xzKP+j/vrfl3Szer4/9bL/W1V9j3+q6o+pTT37b1fVhH9Ufx9WSXp7fdxD9deebNaz+bqk7ZI2\nS/pYRFxbb/+QpHFJt0j6QT2+D/U5r1+bYZynSbrN9sOqftZnRsQvBv3awCAc0abVPwBIw/YaSXdK\nempEPNbsaID0OIMHAKBANHgAAArEEj0AAAXiDB4AgALR4AEAKFCr/w/muRx++OGxZs2apocBAMDI\n3HTTTT+JiOVzHZd1g1+zZo3Gx8ebHgYAACNj+8dzH8USPQAARaLBAwBQIBo8AAAFosEDAFAgGjwA\nAAWiwQMAUCAaPAAABaLBAwBQIBo8AAAFosEDAFAgGjwAAAWiwQMAUCAaPAAABaLBAwBQIBo8AAAF\nosEDAFAgGjwAAAWiwQMAUCAaPAAABaLBAwBQIBo8AGQqInTbzgcVEU0PBS1EgweATG3ZtVdvuvRm\nbdm1t+mhoIVo8ACQqbUrF+vCs47X2pWLmx4KWmhR0wMAAMyPba1btaTpYaClOIMHgMyQvaMfNHgA\nyAzZO/pBgweAzJC9ox80eADIQO+y/GT2brvpYaHFaPAAkAGW5TEoGjwAZIBleQyKy+QAIANcEodB\ncQYPAC3GJXGYLxo8ALQY2TvmiwYPAC1G9o75IoMHgBYje8d8cQYPAC1D7o5hoMEDQMuQu2MYaPAA\n0DLk7hgGMngAaBlydwwDZ/AA0BJk7xgmGjwAtATZO4aJBg8ALUH2jmEigweAliB7xzBxBg8ADSJ3\nRyo0eABoELk7UqHBA0CDyN2RChk8ADSI3B2pcAYPAA0ge0dqNHgAaADZO1KjwQNAA8jekRoZPAA0\ngOwdqXEGDwAjQu6OUUrW4G1/xvYe27f2bPtL27fbvsX2F20f1rPvfNvbbd9h++WpxgUATSF3xyil\nPIP/nKTTpmy7TtJxEfEcSf8k6XxJsr1W0pmS1tWP+ZTt/RKODQBGjtwdo5SswUfENyTdP2XbtRHx\nWH33ekmr69vrJV0REY9ExJ2Stks6IdXYAGCUJpfmJWndqiWy3fCI0AVNZvB/LOnL9e0jJN3Ts29H\nvW0fts+xPW57fGJiIvEQAWDhWJpHExpp8LbfI+kxSZcN+tiIuCgixiJibPny5cMfHAAMGUvzaMLI\nL5Oz/XpJp0s6JZ58Kem9ko7sOWx1vQ0AssclcWjCSM/gbZ8m6Z2SXh0RP+/ZtUnSmbYPsH20pGMk\n3TjKsQHAMHFJHJqW8jK5yyV9W9KxtnfYPlvSX0k6VNJ1tr9n+9OSFBG3SbpS0hZJX5H05oh4PNXY\nACA1cnc0zTn/dTk2Nhbj4+NNDwMA9hER2rJrr9auXMyr5jFUtm+KiLG5juOtagEgAXJ3NI23qgWA\nISJ7R1vQ4AFgiMje0RY0eAAYIq55R1uQwQPAEJG9oy04gweABSJ3RxvR4AFggcjd0UY0eABYIHJ3\ntBEZPAAsELk72ogzeACYB3J3tB0NHgDmgdwdbUeDB4B5IHdH25HBA8A8kLuj7TiDB4ABkL0jFzR4\nABgA2TtyQYMHgAGQvSMXNHgAmEPvsvxk9m676WEBs6LBA8AcWJZHjmjwADAHluWRIy6TA4A5cEkc\ncsQZPADMgEvikDMaPADMgOwdOaPBA8AMyN6RMzJ4AJgB2Ttyxhk8APQgd0cpaPAA0IPcHaWgwQNA\nD3J3lIIMHgB6kLujFJzBA4DI3lEeGjwAiOwd5aHBA4DI3lEeMngAENk7ysMZPIDOIndHyWjwADqL\n3B0lo8ED6Cxyd5SMDB5AZ5G7o2ScwQPoHLJ3dAENHkDnkL2jC2jwADqH7B1dQAYPoHPI3tEFnMED\n6ARyd3QNDR5AJ5C7o2to8AA6gdwdXUODB1C0yaV5SVq3aolsNzwiYDRo8ACKxtI8uooGD6BoLM2j\nq7hMDkDRuCQOXcUZPIDicEkcQIMHUCByd4AGD6BA5O4AGTyAApG7A5zBAygI2TvwJBo8gGKQvQNP\nosEDKAbZO/AkMngAxSB7B57EGTyArJG7A9OjwQPIGrk7MD0aPICskbsD0yODB5A1cndgepzBA8gO\nuTswNxo8gOyQuwNzo8EDyA65OzA3MngA2SF3B+bGGTyAbJC9A/2jwQPIBtk70D8aPIBskL0D/SOD\nB5ANsnegf5zBA2g1cndgfpI1eNufsb3H9q0925bZvs72tvrz0p5959vebvsO2y9PNS4AeSF3B+Yn\n5Rn85ySdNmXbeZI2R8QxkjbX92V7raQzJa2rH/Mp2/slHBuATJC7A/OTrMFHxDck3T9l83pJG+vb\nGyWd0bP9ioh4JCLulLRd0gmpxgag/SaX5iVp3aolst3wiIC8jDqDXxERu+rbuyWtqG8fIemenuN2\n1Nv2Yfsc2+O2xycmJtKNFECjWJoHFqaxF9lF9YqZgV81ExEXRcRYRIwtX748wcgAtAFL88DCjLrB\n32d7pSTVn/fU2++VdGTPcavrbQA6avKSOJbmgfkZdYPfJGlDfXuDpGt6tp9p+wDbR0s6RtKNIx4b\ngIZxSRwwPCkvk7tc0rclHWt7h+2zJV0g6WW2t0k6tb6viLhN0pWStkj6iqQ3R8TjqcYGoJ3I3YHh\ncc5/KY+NjcX4+HjTwwAwJBGhLbv2au3KxSzNAzOwfVNEjM11HG9VC6A1eCtaYHh4q1oAjSN7B4aP\nBg+gcWTvwPDR4AE0jmvegeEjgwfQOLJ3YPg4gwfQCHJ3IC0aPIBGkLsDadHgATSC3B1IiwweQCPI\n3YG0OIMHMFJk78Bo0OABjBTZOzAaNHgAI0X2DowGGTyAkSJ7B0aDM3gAyZG7A6NHgweQHLk7MHo0\neADJkbsDo0eDB5DM5NK8JK1btUS2Gx4R0B00eADJsDQPNIcGDyAZluaB5nCZHIBkuCQOaA5n8ACG\nikvigHagwQMYKnJ3oB1o8ACGitwdaAcyeABDRe4OtANn8ACGguwdaBcaPIChIHsH2oUGD2AoyN6B\ndiGDBzAUZO9Au3AGD2DeyN2B9qLBA5g3cnegvWjwAOaN3B1oLzJ4APNG7g60F2fwAAZG9g60Hw0e\nwMDI3oH2o8EDGBjZO9B+ZPAABkb2DrQfZ/AA+kLuDuSFBg+gL+TuQF5o8AD6Qu4O5IUMHkBfyN2B\nvHAGD2BG5O5AvmjwAGZE7g7kiwYPYEbk7kC+aPAA9jG5NC9J61Ytke2GRwRgUDR4APtgaR7IHw0e\nwD5Ymgfyx2VyAPbBJXFA/jiDByCJS+KA0tDgAUgidwdKQ4MHIIncHSgNGTwASeTuQGk4gwc6juwd\nKBMNHug4snegTDR4oOPI3oEykcEDHUf2DpSJM3igg8jdgfLR4IEOIncHykeDBzqI3B0oHxk80EHk\n7kD5OIMHOoTsHegOGjzQIWTvQHfQ4IEOIXsHuoMMHugQsnegOziDBwpH7g50Ew0eKBy5O9BNNHig\ncOTuQDeRwQOFI3cHuokzeKBQZO9AtzXS4G3/ie3bbN9q+3LbB9peZvs629vqz0ubGBtQCrJ3oNtG\n3uBtHyHprZLGIuI4SftJOlPSeZI2R8QxkjbX9wHME9k70G1NLdEvkvQ024skHSRpp6T1kjbW+zdK\nOqOhsQHZ6l2Wn8zebTc9LAANGHmDj4h7JX1M0t2Sdkl6MCKulbQiInbVh+2WtGK6x9s+x/a47fGJ\niYmRjBnIBcvyACY1sUS/VNXZ+tGSVkk62PZZvcdE9aqgaV8ZFBEXRcRYRIwtX748+XiBnLAsD2BS\nE0v0p0q6MyImIuJRSVdLepGk+2yvlKT6854GxgZkjWV5AJOaaPB3SzrR9kGuqtApkrZK2iRpQ33M\nBknXNDA2AACKMPI3uomIG2xfJelmSY9J+q6kiyQdIulK22dL+rGkPxz12AAAKEUj72QXEe+T9L4p\nmx9RdTYPAAAWiHeyAwCgQDR4AAAKRIMHAKBANHgAAApEgwcAoEA0eAAACkSDBwCgQDR4AAAKRIMH\nAKBANHgAAApEgwcAoEA0eAAACkSDBwCgQDR4AAAKRIMHAKBANHgAAApEgwcAoEA0eAAACkSDBwCg\nQDR4AAAKRIMHAKBANHgAAApEgwcAoEA0eAAACkSDBwCgQDR4AAAKRIMHAKBANHgAAApEgwcAoEA0\neAAACkSDBwCgQDR4AAAKRIMHAKBAczZ422+xvXQUgwEAAMPRzxn8CknfsX2l7dNsO/WgAADAwszZ\n4CPizyQdI+kSSa+XtM32R2w/M/HYAADAPPWVwUdESNpdfzwmaamkq2z/RcKxAQCAeVo01wG2z5X0\nOkk/kXSxpHdExKO2nyJpm6R3ph0iAAAY1JwNXtIySf86In7cuzEinrB9epphAQCAhZizwUfE+2bZ\nt3W4wwEAAMPAdfAAABSIBg8AQIFo8AAAFIgGDwBAgWjwAAAUiAYPAECBaPAAABSIBg8AQIFo8AAA\nFIgGDwBAgWjwAAAUiAYPAECBaPAAABSIBg8AQIFo8AAAFIgGDwBAgWjwAAAUiAYPAECBaPAAABSI\nBg8AQIFo8AAAFIgGDwBAgWjwAAAUiAYPAECBaPAAABSIBg8AQIFo8AAAFIgGDwBAgRpp8LYPs32V\n7dttb7X9QtvLbF9ne1v9eWkTYwMAoARNncF/UtJXIuLZkp4raauk8yRtjohjJG2u7wMAgHkYeYO3\nvUTS70q6RJIi4lcR8YCk9ZI21odtlHTGqMcGAMCwRIRu2/mgIqKR52/iDP5oSROSPmv7u7Yvtn2w\npBURsas+ZrekFdM92PY5tsdtj09MTIxoyAAADGbLrr1606U3a8uuvY08fxMNfpGk4yVdGBHPl/Qz\nTVmOj+rPnWn/5ImIiyJiLCLGli9fnnywAADMx9qVi3XhWcdr7crFjTx/Ew1+h6QdEXFDff8qVQ3/\nPtsrJan+vKeBsQEAMBS2tW7VEtlu5PlH3uAjYreke2wfW286RdIWSZskbai3bZB0zajHBgDAQjWd\nvU9q6lX0b5F0me1bJD1P0kckXSDpZba3STq1vg8AQFaazt4nLWriSSPie5LGptl1yqjHAgDAMDWd\nvU9qpMEDAFCqyey9abxVLTqlLdkYgLK0sbbQ4NEpbcnGAJSljbXFbfprY1BjY2MxPj7e9DCQkYjQ\nll17tXbl4sYuXQFQnlHWFts3RcR0r2P7DWTw6JS2ZGMAytLG2sISPTqhjfkYgPy1ubbQ4NEJbczH\nAOSvzbWFDB6dQPYOIIUmagsZPNCjjfkYgPy1ubawRI9itTkbA5CvXGoLDR7FanM2BiBfudQWMngU\ni9wdQApN1xYyeHRem7MxAPnKpbawRI/i5JKPAchLbrWFBo/i5JKPAchLbrWFDB7FaTofA1CmttSW\nfjN4zuBRhN6ls8l8jOYOYKFyri00eBQht6UzAHnIubawRI8itGXpDEBZ2lhbuEwOnZLLZSsA8pJz\nbWGJHlnL7bIVAHkoobbQ4JG1nPMxAO1VQm0hg0fW2piPAchfm2sLGTw6Ied8DEB7lVBbWKJHdkrI\nxgC0T2m1hQaP7JSQjQFon9JqCxk8stPmbAxAvnKpLWTwKFYJ2RiA9imttrBEj2yUlo8BaIdSawsN\nHtkoLR8D0A6l1hYyeGQjl3wMQF5yqy1k8ChOafkYgHYotbawRI9WKzUbA9CsLtQWGjxardRsDECz\nulBbyODRarllYwDykHNtIYNHEUrNxgA0qwu1hSV6tFIX8jEAo9el2kKDRyt1IR8DMHpdqi1k8Gil\nnPMxAO1VQm0hg0fWupCPARi9LtUWlujRGl3KxgCMTldrCw0erdGlbAzA6HS1tpDBozVKyMYAtE9p\ntaXfDJ4zeDSqd+lsMhsr4RcQQLOoLTR4NKyrS2cA0qK2sESPhpW2dAagHUquLVwmhyx06ZIVAKND\nbWGJHg3p6mUrANKitjyJBo9GkI8BSIHa8iQyeDSi5HwMQHO6UFvI4NFq5GMAUqC2PIkleowM2RiA\nFKgt06PBY2TIxgCkQG2ZHhk8RqYL2RiA0etabSGDR+uQjQFIgdoyPZbokRz5GIAUqC2zo8EjOfIx\nAClQW2ZHBo/kupaPARiNrtYWMni0BvkYgBSoLbNjiR5JkI0BSIHa0j8aPJIgGwOQArWlf2TwSKKr\n2RiAtKgtZPBoGNkYgBSoLf1jiR5DRT4GIAVqy+Bo8Bgq8jEAKVBbBkcGj6EiHwOQArXlSWTwaAT5\nGIAUqC2DY4keC0Y2BiAFasvCNNbgbe9n+7u2v1TfX2b7Otvb6s9LmxobBkM2BiAFasvCNHkGf66k\nrT33z5O0OSKOkbS5vo8MrF25WBeedbzWrlzc9FAAFITasjCNNHjbqyW9StLFPZvXS9pY394o6YxR\njwuDmVw+k6R1q5Z0/oUvAIaD2jIcTZ3Bf0LSOyU90bNtRUTsqm/vlrRiugfaPsf2uO3xiYmJxMPE\nbFg+A5ACtWU4Rt7gbZ8uaU9E3DTTMVG9omLaV1VExEURMRYRY8uXL081TPSB5TMAKVBbhqOJy+Re\nLOnVtl8p6UBJi21fKuk+2ysjYpftlZL2NDA2DIDLVgCkQG0ZjpGfwUfE+RGxOiLWSDpT0lcj4ixJ\nmyRtqA/bIOmaUY8Nc+OyFQApUFuGr03XwV8g6WW2t0k6tb6PliEbA5ACtWX4eKtaDIS3iwSQArWl\nf7xVLZIgGwOQArVl+Nq0RI8WIx8DkAK1JR0aPPpCPgYgBWpLOmTw6Av5GIAUqC2DI4PHUJGPAUiB\n2pIOS/SYEdkYgBSoLaNBg8eMyMYApEBtGQ0yeMyIbAxACtSWhSGDx4KRjQFIgdoyGizRYx/kYwBS\noLaMFg0e+yAfA5ACtWW0yOCxD/IxAClQW4aDDB7zRj4GIAVqy2ixRA9JZGMA0qC2NIcGD0lkYwDS\noLY0hwweksjGAKRBbRm+fjN4zuA7rHfpbDIb4xcQwEJRW9qBBt9hLJ0BSIHa0g4s0XcYS2cAUqC2\npMVlcpgTl6wASIHa0g4s0XcQl60ASIHa0i40+A4iHwOQArWlXcjgO4h8DEAK1JbRIIPHjMjHAKRA\nbWkXlug7gmwMQArUlvaiwXcE2RiAFKgt7UUG3xFkYwBSoLaMHhk8fgPZGIAUqC3txRJ94cjHAKRA\nbWk/GnzhyMcApEBtaT8y+MKRjwFIgdrSHDJ4SCIfA5AGtaX9WKIvENkYgBSoLXmhwReIbAxACtSW\nvJDBF4hsDEAK1JZ2IIPvMLIxAClQW/LCEn1ByMcApEBtyRMNviDkYwBSoLbkiQy+IORjAFKgtrQL\nGXwHkY8BSIHakieW6DNHNgYgBWpL/mjwmSMbA5ACtSV/ZPCZIxsDkAK1pb36zeA5g8/U5PKZJK1b\ntYRfQABDQW0pBw0+UyyfAUiB2lIOlugzxfIZgBSoLe3HZXKF47IVAClQW8rBEn1GuGwFQArUljLR\n4DNCNgYgBWpLmcjgM0I2BiAFakteyOALRDYGIAVqS5lYos8A+RiAFKgtZaPBZ4B8DEAK1JaykcFn\ngHwMQArUljyRwReEfAxACtSWsrFE31JkYwBSoLZ0Bw2+pcjGAKRAbekOMviWIhsDkAK1JX9k8Jkj\nGwOQArWlO1iibxnyMQApUFu6hwbfMuRjAFKgtnQPGXzLkI8BSIHaUg4y+EyRjwFIgdrSPSzRtwDZ\nGIAUqC3dRoNvAbIxAClQW7qNDL4FyMYApEBtKRMZfEbIxgCkQG3pNpboG0Q+BiAFagukBhq87SNt\nf832Ftu32T633r7M9nW2t9Wfl456bKNGPgYgBWoLpAYyeNsrJa2MiJttHyrpJklnSHq9pPsj4gLb\n50laGhHvmu1r5Z7Bk48BSIHaUrZ+M/iRn8FHxK6IuLm+/ZCkrZKOkLRe0sb6sI2qmn5xepfOJvMx\nfgEBLBS1BVM1msHbXiPp+ZJukLQiInbVu3ZLWjHDY86xPW57fGJiYiTjHCaWzgCkQG3BVI1dJmf7\nEElfl/ThiLja9gMRcVjP/p9GxKw5fI5L9CydAUiB2tIdrV2ilyTbT5X0BUmXRcTV9eb76nx+Mqff\n08TYUmPpDEAK1BZM1cSr6C3pEklbI+LjPbs2SdpQ394g6ZpRjy0VLlkBkAK1BbNp4gz+xZJeK+lk\n29+rP14p6QJJL7O9TdKp9f0ikI0BSIHagtnwVrUjQDYGIAVqSzfxVrUtwttFAkiB2oLZ8Fa1CZGP\nAUiB2oJ+0OATIh8DkAK1Bf0gg0+IfAxACtSWbiODbwHyMQApUFvQD5boh4xsDEAK1BYMigY/ZGRj\nAFKgtmBQZPBDRjYGIAVqCyaRwTeEbAxACtQWDIol+iEhHwOQArUF80WDHxLyMQApUFswX2TwQ0I+\nBiAFagumIoMfMfIxAClQWzBfLNEvANkYgBSoLRgGGvwCkI0BSIHagmEgg18AsjEAKVBbMJt+M3jO\n4OdhcvlMktatWsIvIIChoLZgmGjw88DyGYAUqC0YJpbo54HlMwApUFvQDy6TS4jLVgCkQG3BMLFE\n3ycuWwGQArUFqdDg+0Q2BiAFagtSIYPvE9kYgBSoLRgUGfyQkY0BSIHaglRYop8D+RiAFKgtSI0G\nPwfyMQApUFuQGhn8HMjHAKRAbcF8kcEPCfkYgBSoLUiNJfppkI0BSIHaglGiwU+DbAxACtQWjBIZ\n/DTIxgCkQG3BMJDBLwDZGIAUqC0YJZboe5CPAUiB2oIm0OB7kI8BSIHagiaQwfcgHwOQArUFw0QG\nPw/kYwDbqUpcAAAGJ0lEQVRSoLagCSzRAwBQIBo8AAAFosEDAFAgGjwAAAWiwQMAUCAaPAAABaLB\nAwBQIBo8AAAFosEDAFAgGjwAAAWiwQMAUCAaPAAABaLBAwBQIBo8AAAFosEDAFAgGjwAAAWiwQMA\nUCAaPAAABaLBAwBQIBo8AAAFckQ0PYZ5sz0h6cdNj2MeDpf0k6YHMSSlzKWUeUjMpa1KmUsp85Dy\nnctREbF8roOybvC5sj0eEWNNj2MYSplLKfOQmEtblTKXUuYhlTWX6bBEDwBAgWjwAAAUiAbfjIua\nHsAQlTKXUuYhMZe2KmUupcxDKmsu+yCDBwCgQJzBAwBQIBp8QraPtP0121ts32b73Hr7MtvX2d5W\nf17a9Fj7ZXs/29+1/aX6fpZzsX2Y7ats3257q+0X5jgX239S/9u61fbltg/MZR62P2N7j+1be7bN\nOHbb59vebvsO2y9vZtTTm2Euf1n/+7rF9hdtH9azL6u59Oz7U9th+/CebdnNxfZb6p/Nbbb/omd7\na+cyHzT4tB6T9KcRsVbSiZLebHutpPMkbY6IYyRtru/n4lxJW3vu5zqXT0r6SkQ8W9JzVc0pq7nY\nPkLSWyWNRcRxkvaTdKbymcfnJJ02Zdu0Y69/b86UtK5+zKds7ze6oc7pc9p3LtdJOi4iniPpnySd\nL2U7F9k+UtIfSLq7Z1t2c7H9+5LWS3puRKyT9LF6e9vnMjAafEIRsSsibq5vP6SqiRyh6h/Xxvqw\njZLOaGaEg7G9WtKrJF3cszm7udheIul3JV0iSRHxq4h4QBnORdIiSU+zvUjSQZJ2KpN5RMQ3JN0/\nZfNMY18v6YqIeCQi7pS0XdIJIxloH6abS0RcGxGP1Xevl7S6vp3dXGr/VdI7JfW+cCvHubxJ0gUR\n8Uh9zJ56e6vnMh80+BGxvUbS8yXdIGlFROyqd+2WtKKhYQ3qE6p+wZ/o2ZbjXI6WNCHps3XccLHt\ng5XZXCLiXlVnH3dL2iXpwYi4VpnNY4qZxn6EpHt6jttRb8vFH0v6cn07u7nYXi/p3oj4/pRd2c1F\n0rMkvdT2Dba/bvsF9fYc5zIrGvwI2D5E0hckvS0i9vbui+oyhtZfymD7dEl7IuKmmY7JZS6qznqP\nl3RhRDxf0s80ZRk7h7nU+fR6VX+wrJJ0sO2zeo/JYR4zyXnsvWy/R1Vcd1nTY5kP2wdJerek9zY9\nliFZJGmZqtj0HZKutO1mh5QGDT4x209V1dwvi4ir68332V5Z718pac9Mj2+RF0t6te27JF0h6WTb\nlyrPueyQtCMibqjvX6Wq4ec2l1Ml3RkRExHxqKSrJb1I+c2j10xjv1fSkT3Hra63tZrt10s6XdIf\nxZPXJOc2l2eq+iPy+/Xv/2pJN9v+Z8pvLlL1+391VG5UtSJ5uPKcy6xo8AnVfxVeImlrRHy8Z9cm\nSRvq2xskXTPqsQ0qIs6PiNURsUbVC1G+GhFnKc+57JZ0j+1j602nSNqi/OZyt6QTbR9U/1s7RdXr\nPHKbR6+Zxr5J0pm2D7B9tKRjJN3YwPj6Zvs0VZHWqyPi5z27sppLRPwgIp4REWvq3/8dko6vf4+y\nmkvtbyX9viTZfpak/VX9hzM5zmV2EcFHog9JL1G1xHiLpO/VH6+U9HRVrxDeJukfJS1reqwDzusk\nSV+qb2c5F0nPkzRe/2z+VtLSHOci6QOSbpd0q6S/lnRALvOQdLmq1w48qqppnD3b2CW9R9IPJd0h\n6RVNj7+PuWxXlelO/u5/Ote5TNl/l6TDc52LqoZ+af07c7Okk3OYy3w+eCc7AAAKxBI9AAAFosED\nAFAgGjwAAAWiwQMAUCAaPAAABaLBAwBQIBo8AAAFosED6JvtF9T/v/mBtg+u/z/t45oeF4B98UY3\nAAZi+0OSDpT0NFXv6f/RhocEYBo0eAADsb2/pO9I+qWkF0XE4w0PCcA0WKIHMKinSzpE0qGqzuQB\ntBBn8AAGYnuTqv8y+GhJKyPiPzU8JADTWNT0AADkw/brJD0aEZ+3vZ+kb9k+OSK+2vTYAPwmzuAB\nACgQGTwAAAWiwQMAUCAaPAAABaLBAwBQIBo8AAAFosEDAFAgGjwAAAWiwQMAUKD/D/mpeUrYHKI4\nAAAAAElFTkSuQmCC\n", + "image/png": "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\n", "text/plain": [ - "" + "
" ] }, - "metadata": {}, + "metadata": { + "needs_background": "light" + }, "output_type": "display_data" } ], @@ -150,12 +154,14 @@ "outputs": [ { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAABJYAAAKzCAYAAABMJofMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzs3X+0b3ld3/fXJ0MGCD/EKcgKMHEEgRSsWnNFXWqCP5Ki\njOLqylhs/VnqjUrFJDUoavXcNIhtXAl2WWtuMVmpRhET05IaRMRiY4MlQxo1qKkouPhlGIIg+APD\n8Okf53uHM8jMnLPP2fe99+fzeKw16945z3PP3d/963vXZ53z3q33HgAAAAA4qz9WvQEAAAAA7JOF\nJQAAAAAWsbAEAAAAwCIWlgAAAABYxMISAAAAAItYWAIAAABgEQtLAABF2rG/11r77dbaa07x+be0\n1npr7X7XY/uWaq19Zmvt31RvBwCwPgtLAMCptNbe2Fr7/dbae1pr72qt/fPW2te01k7174nrtSiy\n9t/TWvvU1torWmvvbK3d0Vr7sdban1z45T4jyZ9P8pje+1MucDPTWntqa+3NF/k1T6v3/s967088\nzedWbicAcH4WlgCAs/iC3vtDknx0ku9K8k1JfqB2k667j0xyNcktOd4P70ny9xZ+rY9O8sbe++9e\nzKYBAFxfFpYAgDPrvb+79/7SJP9Zkq9orX1ckrTWnt5a+39ba7/TWntTa+3oxB/7vw6/vqu19t7W\n2qe11h7XWvuZ1tq/a629o7X2D1prD7v2B1pr39Rae8vhu6T+TWvtcw4f/2OttW9urf364c++pLV2\n0z39PR+6/a21o8Of+V8PX/t1rbVLp3ztL+u9/1jv/Xd677+X5HuTfPo9fX5r7VGttZcevsPp9a21\nrz58/FlJXpTk0w7beeXD/NkbWmvffdg3v5Hk6R/Sv6q19iuH1/AbrbW/dPj4g5K8LMmjDl/7vYft\neEpr7dWH7zh7W2vte1trN97Ddl/7zq/LrbW3Hj7/G0/0+7fWXnhobz38/v6HdrfvQjp8t9s3ttZ+\nsbX27tbaj7bWHnAf23n74Tz6t621v3WaYwMAXH8WlgCAxXrvr0ny5iSfefjQ7yb58iQPy/EiyNe2\n1r7o0P7s4deH9d4f3Ht/dZKW5AVJHpXkP0xyc5KjJGmtPTHJf53kkw/fJfWfJHnj4Wt8fZIvSvLn\nDn/2t5P8T/fy93w4X5jkxYdtfWmOF4iW+LNJXncv/cU53kePSvIXk3xna+2ze+8/kORrkrz6sJ3f\n8WH+7FcnuTXJf5zk0uHPn/T2Q39okq9K8rdba590+A6oz0vy1sPXfnDv/a1J7kzyV5I8PMmnJfmc\nJF93H6/vs5I8PslfSPJNrbXPPXz8W5N8apJPTPIJSZ6S5Nvu5et8cZKnJfmYJB+f5CvvZTu/J8n3\n9N4fmuRxSV5yH9sIABSxsAQAnNdbk9yUJL33V/Xef6n3/oHe+y8m+ZEcL/58WL331/feX9F7f1/v\n/Y4kf+vE59+Z5P5JntRa++O99zf23n/90L4mybf23t/ce39fjhej/uIZ5yr9XO/9n/be70zygzle\nHDmT1trHJ/n2JH/tHvrNOf5upm/qvf9B7/1f5fi7lL78lH/FFyd5Ye/9Tb33d+Z4Ee4uvfef6L3/\nej/2s0l+Kh9c5Psjeu+v7b3/fO/9/b33Nyb5O7mX43Nwpff+u733X8rxj/x9yeHj/0WSv957f/vh\n2F1J8mX38nX+x977Ww+v45/keEHqnvz7JB/bWnt47/29vfefv49tBACKWFgCAM7r0UnemSSttU9p\nrf2fh6HW787xAtDD7+kPttYe2Vp78eHH3X4nyQ9d+/ze++uT/OUcLxq9/fB5jzr80Y9O8o8PP9L1\nriS/kuOFqEeeYbt/68Tvfy/JA86yMNVa+9gc/xjXN/Te/9k9fNqjkryz9/6eEx/7zRzvs9N4VJI3\nfcifPbkNn9da+/nDj9m9K8nn59739xNaa/9Ha+23Dvv7O+/t8w8+9O+/dgwe9SHbc7J9OB+6vx98\nL5/7rCRPSPKrrbV/0Vq79T62EQAoYmEJAFistfbJOV4k+bnDh344xz9WdnPv/SOSfH+Of9wtSfqH\n+RLfefj4f3T4sacvPfH56b3/cO/9M3K8kNST/PeH9KYkn9d7f9iJ/x7Qe3/LPfw9F6q19tFJfjrJ\nf9d7/8F7+dS3JrmptfaQEx/7U0necsq/6m05/vHAk3/22jbcP8k/SvLdSR7Ze39Ykn+ae9/f/3OS\nX03y+MP+/pYTn39PPvTvf+vh92/N8XH5cO0s/sh29t5/rff+JUk+KsfH/B8e5jEBABtjYQkAOLPW\n2kMP30Xy4iQ/dPgxqSR5SI6/Q+cPWmtPSfKfn/hjdyT5QJLHnvjYQ5K8N8m7W2uPzokfKWutPbG1\n9tmHBZQ/SPL7hz+fHC9YPf+wwJPW2iNaa8+4l7/nwhy282eSfG/v/fvv7XN7729K8s+TvOAwrPrj\nc/zdOD90yr/uJUme01p7TGvtI5N884l2Y45/VPCOJO9vrX1ejucgXfNvk/wHrbWPOPGxhyT5nSTv\nba396SRfe4pt+G9ba3+itfbkHM9x+tHDx38kybcd9v3Dc/wjgad9XSf9ke1srX1pa+0RvfcPJHnX\n4cMf+LB/GgAoZWEJADiLf9Jae0+Ov2PoW3M8E+mrTvSvS/LXD5/z7TkxdPnwBLXnJ/m/Dz/C9qk5\nnsvzSUneneQnkvz4ia91/yTfleQdOf4xqo9K8rxD+54cf2fUTx3+rp9P8in38vecSWvtZa21b7mH\n/F/leNHq6MSTzN57L1/uS5LckuPv5vnHSb6j9/7Tp9yU/yXJy5P8QpJ/mRP75/Djdc/J8T7+7Rwv\n4r30RP/VHC/+/MZhPzwqyTcePu89h699bZHo3vxsktcneWWS7+69/9Th438jye1JfjHJLx2272+c\n8nXd5R6282lJXnfYr9+T5Jm9998/69cGANbXel/9u8UBANiZ1totSd6Q5I/33t9fuzUAwFb5jiUA\nAAAAFrGwBAAAAMAifhQOAAAAgEV8xxIAAAAAi1hYAgAAAGCR+1VvwHk8/OEP77fcckv1ZgAAAAAM\n47Wvfe07eu+POM3n7nph6ZZbbsntt99evRkAAAAAw2it/eZpP9ePwgEAAACwiIUlAAAAABaxsAQA\nAADAIhaWAAAAAFjEwhIAAAAAi1hYAgAAAGARC0sAAAAALGJhCQAAAIBFLCwBAAAAsIiFJQAAAAAW\nsbAEAAAAwCIWlgAAAABYxMISAAAAAItYWAIAAABgEQtLAAAAACxiYQkAAACARSwsAQAAALCIhSUA\nAAAAFrGwBAAAAMAiFpYAAAAAWMTCEgAAAACLWFgCAAAAYBELSwAAAAAsYmEJAAAAgEUsLAEAAACw\niIUlgB175+/+Yf7Oz/563vm7f7jJDgAAjM3CEsCO/djtb8oLXvar+bHb37TJDgAAjO1+1RsAwHK3\nXbr5br9urQMAAGNrvffqbVjs0qVL/fbbb6/eDAAAAIBhtNZe23u/dJrP9aNwAAAAACxiYQkAAACA\nRSwsAexY9VPfPBUOAADmZmEJYMeqn/rmqXAAADA3T4UD2LHqp755KhwAAMzNU+EAAAAAuIunwgEA\nAACwOgtLADtWPZzb8G4AAJibhSWAHasezm14NwAAzM3wboAdqx7ObXg3AADMzfBuAAAAAO5ieDcA\nAAAAq7OwBLBj1cO5De8GAIC5WVgC2LHq4dyGdwMAwNwM7wbYserh3IZ3AwDA3AzvBgAAAOAuhncD\nAAAAsDoLS8CqqodH6/peu8HoAADsgYUlYFXVw6N1fa/dYHQAAPbA8G5gVdXDo3V9r91gdAAA9sDw\nbgAAAADuYng3AAAAAKuzsASsasvDkXV9y93wbgAA9sDCErCqLQ9H1vUtd8O7AQDYA8O7gVVteTiy\nrm+5G94NAMAerDq8u7X2xiTvSXJnkvf33i+11m5K8qNJbknyxiRf3Hv/7cPnPy/Jsw6f/5ze+8vv\n7esb3g0AAABwsbY2vPuzeu+feGKDvjnJK3vvj0/yysP/p7X2pCTPTPLkJE9L8n2ttRuuw/YBAAAA\nsEDFjKVnJPn7h9///SRfdOLjL+69v6/3/oYkr0/ylILtAy7Qlocj6/qWu+HdAADswdoLSz3JT7fW\nXttau3z42CN77287/P63kjzy8PtHJzk5ofTNh4/dTWvtcmvt9tba7Xfcccda2w1ckC0PR9b1LXfD\nuwEA2IO1h3d/Ru/9La21j0ryitbar56MvffeWjvTkKfe+9UkV5PjGUsXt6nAGrY8HFnXt9wN7wYA\nYA9WHd59t7+otaMk703y1Ume2nt/W2vtTyZ5Ve/9iYfB3em9v+Dw+S9PctR7f/U9fU3DuwEAAAAu\n1iaGd7fWHtRae8i13yf5C0n+dZKXJvmKw6d9RZL//fD7lyZ5Zmvt/q21j0ny+CSvWWv7AAAAADif\nNWcsPTLJz7XWfiHHC0Q/0Xv/ySTfleTPt9Z+LcnnHv4/vffXJXlJkl9O8pNJnt17v3PF7QOugy0P\nR9b1LXfDuwEA2IPVFpZ677/Re/+Ew39P7r0///Dxf9d7/5ze++N775/be3/niT/z/N7743rvT+y9\nv2ytbQOuny0PR9b1LXfDuwEA2IO1h3cDk9vycGRd33I3vBsAgD24bsO712B4NwAAAMDF2sTwbgAA\nAADGZmEJBrfl4cS6ri8f3l3dAQAgsbAEw9vycGJd15cP767uAACQGN4Nw9vycGJd15cP767uAACQ\nGN4NAAAAwAmGdwMAAACwOgtLsHPVA3z33s+revv1cfuWtw0AAK6xsAQ7Vz3Ad+/9vKq3Xx+3b3nb\nAADgGsO7YeeqB/juvZ9X9fbr4/YtbxsAAFxjeDcAAAAAdzG8GwAAAIDVWVgCAAAAYBELS1DsvE9e\nqn4y1Nb72qpfnz5u3/K2AQDANRaWoNh5n7xU/WSorfe1Vb8+fdy+5W0DAIBrPBUOip33yUvVT4ba\nel9b9evTx+1b3jYAALjGU+EAAAAAuIunwgEAAACwOgtLsHPVA3z33s+revv1cfuWtw0AAK6xsAQ7\nVz3Ad+/9vKq3Xx+3b3nbAADgGsO7YeeqB/juvZ9X9fbr4/YtbxsAAFxjeDcAAAAAdzG8GwAAAIDV\nWVhiddUDZnVd1/fYt7xtOgAA11hYYnXVA2Z1Xdf32Le8bToAANfccHR0VL0Ni129evXo8uXL1ZvB\nfXjsIx6cmx50Y267dHMeeOMNuq7r+in6lrdNBwAY25UrV952dHR09TSfa3g3AAAAAHcxvBsAAACA\n1VlYYnXVA1Z1Xdf32Le8bToAANdYWGJ11QNWdV3X99i3vG06AADXGN7N6qoHrOq6ru+xb3nbdACA\nsRneDQAAAMAihncDAAAAsDoLS6yuesCqruv6HvuWt00HAOAaC0usrnrAqq7r+h77lrdNBwDgGsO7\nWV31gFVd1/U99i1vmw4AMDbDuwEAAABYxPBuAAAAAFZnYYnVVQ9Y1XVd32Pf8rbpAABcY2GJ1VUP\nWNV1Xd9j3/K26QAAXGN4N6urHrCq67q+x77lbdMBAMZmeDcAAAAAixjeDQAAAMDqLCxNoHrAqa7r\num5492gdYKnq+1d1B8ZjYWkC1QNOdV3XdcO7R+sAS1Xfv6o7MB7DuydQPeBU13VdN7x7tA6wVPX9\nq7oD+2B4NwAAAACLGN4NAAAAwOosLE2gekCfruu6bnj3aB1gqer7V3UHxmNhaQLVA/p0Xdf1s/ct\nb5sOsFz1/au6A+MxvHsC1QP6dF3XdcO7R+sAS1Xfv6o7sA+GdwMAAACwiOHdAAAAAKzOwtIEqgf0\n6bqu64Z3j9YBlqq+f1V3YDwWliZQPaBP13VdP3vf8rbpAMtV37+qOzAew7snUD2gT9d1XTe8e7QO\nsFT1/au6A/tgeDcAAAAAixjeDQAAAMDqLCwBAAAAsIiFpQlUP/lB13Vd91S40TrAUtX3r+oOjMfC\n0gSqn/yg67qun71vedt0gOWq71/VHRiPp8JNoPrJD7qu67qnwo3WAZaqvn9Vd2AfPBUOAAAAgEU8\nFQ4AAACA1VlYmkD1gD5dNyCStVSfX4Z36+5tsD/V1+/sHRiPhaUJVA/o03UDIllL9flleLfu3gb7\nU339zt6B8RjePYHqAX26bkAka6k+vwzv1t3bYH+qr9/ZO7APhncDAAAAsIjh3QAAAACszsLSBKoH\n9Om6AZGspfr8Mrxbd2+D/am+fmfvwHgsLE2gekCfrhsQyVqqzy/Du3X3Ntif6ut39g6Mx/DuCVQP\n6NN1AyJZS/X5ZXi37t4G+1N9/c7egX0wvBsAAACARQzvBgAAAGB1FpYmUD2gT9cNiGQt1eeX4d26\nexuMp/r6Hr0D47GwNIHqAX26bkAka6k+vwzv1t3bYDzV1/foHRiP4d0TqB7Qp+sGRLKW6vPL8G7d\nvQ3GU319j96BfTC8GwAAAIBFDO8GAAAAYHUWliZQPaBP1w2IZC3V55fh3bp7G+xP9fU7ewfGY2Fp\nAtUD+nTdgEjWUn1+Gd6tu7fB/lRfv7N3YDyGd0+gekCfrhsQyVqqzy/Du3X3Ntif6ut39g7sg+Hd\nAAAAACxieDcAAAAAq7OwtAHVA/R03YBGWGbN66P62tV193ZGVX396HN3GJGFpQ2oHqCn6wY0wjJr\nXh/V166uu7czqurrR5+7w4gM796A6gF6um5AIyyz5vVRfe3quns7o6q+fvS5O+yF4d0AAAAALGJ4\nNwAAAACrs7C0AdUD5HTdgEJYZs3rp/ra1XX3dkZVff3oc3cYkYWlDageIKfrBhTCMmteP9XXrq67\ntzOq6utHn7vDiAzv3oDqAXK6bkAhLLPm9VN97eq6ezujqr5+9Lk77MWmhne31m5IcnuSt/Teb22t\nfWKS70/ygCTvT/J1vffXHD73eUmeleTOJM/pvb/83r624d0AAAAAF2trw7u/IcmvnPj//yHJld77\nJyb59sP/p7X2pCTPTPLkJE9L8n2HRSkAAAAANmjVhaXW2mOSPD3Ji058uCd56OH3H5HkrYffPyPJ\ni3vv7+u9vyHJ65M8Zc3t24rqAXK6bkAhrGPN66P62tbn7rBn1dePPneHEa39HUsvTPLcJB848bG/\nnORvttbelOS7kzzv8PFHJzk54ezNh4/dTWvtcmvt9tba7Xfcccc6W32dVQ+Q03UDCmEda14f1de2\nPneHPau+fvS5O4zofmt94dbarUne3nt/bWvtqSfS1yb5K733f9Ra++IkP5Dkc0/7dXvvV5NcTY5n\nLF3gJpe57dLNd/tV10fqMLM1r4/qa1ufu8OeVV8/+twdRrTa8O7W2guSfFmOB3Q/IMc//vbjSb4g\nycN677211pK8u/f+0MPg7vTeX3D48y9PctR7f/U9/R2GdwMAAABcrE0M7+69P6/3/pje+y05Hsr9\nM733L83xTKU/d/i0z07ya4ffvzTJM1tr92+tfUySxyd5zVrbBwAAAMD5rPajcPfiq5N8T2vtfkn+\nIMnlJOm9v6619pIkv5zj73J6du/9zoLtAwAAAOAU1h7enSTpvb+q937r4fc/13v/M733T+i9f0rv\n/bUnPu/5vffH9d6f2Ht/2fXYti2ofjKBro/85Ivq11/dWVfl/q8+t/S5O5xH9fmr6+6fcLGuy8IS\n9676yQS6PvKTL6pff3VnXZX7v/rc0ufucB7V56+uu3/Cxbrh6OioehsWu3r16tHly5erN+PcHvuI\nB+emB92Y2y7dnAfeeIOuD9WrVb/+6s66Kvd/9bmlz93hPKrPX113/4T7duXKlbcdHR1dPc3nrvZU\nuOvBU+EAAAAALtYmngoHAAAAwNgsLG1A9QA5few+u+r9X90ZV/W5pc/d4Tyqz19dd/+Ei2VhaQOq\nB8jpY/fZVe//6s64qs8tfe4O51F9/uq6+ydcLMO7N6B6gJw+dp9d9f6v7oyr+tzS5+5wHtXnr667\nf8J9M7wbAAAAgEUM7wYAAABgdRaWNqB6gJw+dp9d9f6v7oyr+tzS5+5wHtXnr667f8LFsrC0AdUD\n5PSx++yq9391Z1zV55Y+d4fzqD5/dd39Ey6W4d0bUD1ATh+7z656/1d3xlV9bulzdziP6vNX190/\n4b4Z3g0AAADAIoZ3AwAAALA6C0sbUD1ATh+7z656/1d3xlV9bulzdziP6vNX190/4WJZWNqA6gFy\n+th9dtX7v7ozrupzS5+7w3lUn7+67v4JF8vw7g2oHiCnj91nV73/qzvjqj639Lk7nEf1+avr7p9w\n3wzvBgAAAGARw7sBAAAAWJ2FpQ2oHiCnj92rVb9+fe4+sup9q8/d4Tyqz19dn7nDGiwsbUD1ADl9\n7F6t+vXrc/eRVe9bfe4O51F9/ur6zB3WYHj3BlQPkNPH7tWqX78+dx9Z9b7V5+5wHtXnr67P3OG0\nDO8GAAAAYBHDuwEAAABYnYWlDage4KaP3atVv3597s56qo+t7tpiv6rPX732+q5+fbN3WIOFpQ2o\nHuCmj92rVb9+fe7OeqqPre7aYr+qz1+99vqufn2zd1iD4d0bUD3ATR+7V6t+/frcnfVUH1vdtcV+\nVZ+/eu31Xf36Zu9wWoZ3AwAAALCI4d0AAAAArM7C0gZUD3DTx+7Vql+/PndnPdXHVndtsV/V569u\nePfMHdZgYWkDqge46WP3atWvX5+7s57qY6u7ttiv6vNXN7x75g5rMLx7A6oHuOlj92rVr1+fu7Oe\n6mOru7bYr+rzVze8e+YOp2V4NwAAAACLGN4NAAAAwOosLG1A9QA3fexerfr163N31lN9bHXXFvtV\nff7qhnfP3GENFpY2oHqAmz52r1b9+vW5O+upPra6a4v9qj5/dcO7Z+6wBsO7N6B6gJs+dq9W/fr1\nuTvrqT62umuL/ao+f3XDu2fucFqGdwMAAACwiOHdAAAAAKzOwtIGVA9w08fu1apfvz53Zz3Vx1Yf\nu1Or+vjrY19/1a9PH/v8ooaFpQ2oHuCmj92rVb9+fe7OeqqPrT52p1b18dfHvv6qX58+9vlFDcO7\nN6B6gJs+dq9W/fr1uTvrqT62+tidWtXHXx/7+qt+ffrY5xcXx/BuAAAAABYxvBsAAACA1VlYAgAA\nAGARC0sbUD2ZX/fkgzVV73997s56qo+tPnanVvXx18e+/qpfnz72+UUNC0sbUD2ZX/fkgzVV7399\n7s56qo+tPnanVvXx18e+/qpfnz72+UUNT4XbgOrJ/LonH6ypev/rc3fWU31s9bE7taqPvz729Vf9\n+vSxzy8ujqfCAQAAALCIp8IBAAAAsDoLSxtQPUBNN6BuTdX7X5+7s57qY6uP3alVffz1sa+/6ten\nj31+UcPC0gZUD1DTDahbU/X+1+furKf62Opjd2pVH3997Ouv+vXpY59f1DC8ewOqB6jpBtStqXr/\n63N31lN9bPWxO7Wqj78+9vVX/fr0sc8vLo7h3QAAAAAsYng3AAAAAKuzsLQB1QPUdAPq1lS9//W5\nO+upPrb62J1a1cdfH/v6q359+tjnFzUsLG1A9QA13YC6NVXvf33uznqqj60+dqdW9fHXx77+ql+f\nPvb5RQ3DuzegeoCabkDdmqr3vz53Zz3Vx1Yfu1Or+vjrY19/1a9PH/v84uIY3g0AAADAIoZ3AwAA\nALA6C0sbUD1ATTegbk3V+193/rNP1eeu7t4xs+rjP3tnXdXHd/bOmCwsbUD1ADXdgLo1Ve9/3fnP\nPlWfu7p7x8yqj//snXVVH9/ZO2MyvHsDqgeo6QbUral6/+vOf/ap+tzV3TtmVn38Z++sq/r4zt7Z\nD8O7AQAAAFjE8G4AAAAAVmdhaQOqB6jpBtStqXr/685/9qn63NXdO2ZWffxn76yr+vjO3hmThaUN\nqB6gphtQt6bq/a87/9mn6nNXd++YWfXxn72zrurjO3tnTIZ3b0D1ADXdgLo1Ve9/3fnPPlWfu7p7\nx8yqj//snXVVH9/ZO/theDcAAAAAixjeDQAAAMDqLCxtQPUANd2AujVV73/d+c8+VZ+7unvHzKqP\n/+yddVUf39k7Y7KwtAHVA9R0A+rWVL3/dec/+1R97uruHTOrPv6zd9ZVfXxn74zJ8O4NqB6gphtQ\nt6bq/a87/9mn6nNXd++YWfXxn72zrurjO3tnPwzvBgAAAGARw7sBAAAAWJ2FpQ2oHqCmG1C3pur9\nrzv/2afqc1d375hZ9fGfvbOu6uM7e2dMFpY2oHqAmm5A3Zqq97/u/Gefqs9d3b1jZtXHf/bOuqqP\n7+ydMRnevQHVA9R0A+rWVL3/dec/+1R97uruHTOrPv6zd9ZVfXxn7+yH4d0AAAAALGJ4NwAAAACr\ns7C0AdUD1HQD6tZUvf9n72urfn3VnfVUH1vdtTWz6uPv/ALYFwtLG1A9QE03oG5N1ft/9r626tdX\n3VlP9bHVXVszqz7+zi+AfTG8ewOqB6jpBtStqXr/z97XVv36qjvrqT62umtrZtXH3/kFUM/wbgAA\nAAAWMbwbAAAAgNVZWNqA6gGGugGRa6re/7P3tVW/vurOeqqPre7amln18a/uAJyNhaUNqB5gqBsQ\nuabq/T97X1v166vurKf62OqurZlVH//qDsDZGN69AdUDDHUDItdUvf9n72urfn3VnfVUH1vdtTWz\n6uNf3QEwvBsAAACAhQzvBgAAAGB1FpYAAAAAWMTC0gZUP/lC92SRNVXv/9n72qpfX3VnPdXHVndt\nzey8x2ft4+/8A9gWC0sbUP3kC92TRdZUvf9n72urfn3VnfVUH1vdtTWz8x6ftY+/8w9gW+639l/Q\nWrshye1J3tJ7v/Xwsa9P8uwkdyb5id77cw8ff16SZx0+/pze+8vX3r4tuO3SzXf7VR+rz656/8/e\n11b9+qo766k+trpra2bnPT5rH3/nH8C2rP5UuNbaX01yKclDe++3ttY+K8m3Jnl67/19rbWP6r2/\nvbX2pCQ/kuQpSR6V5KeTPKH3fuc9fW1PhQMAAAC4WJt5Klxr7TFJnp7kRSc+/LVJvqv3/r4k6b2/\n/fDxZyR5ce/9fb33NyR5fY4XmQAAAADYoLVnLL0wyXOTfODEx56Q5DNba/9Pa+1nW2uffPj4o5Oc\n/EHnNx/3Taz0AAAgAElEQVQ+NrzqAZm6AZBrqt7/s/e1Vb++6s56qo+t7tqaWfXxr+4AnM1qC0ut\ntVuTvL33/toPSfdLclOST03y15K8pLXWzvB1L7fWbm+t3X7HHXdc3AYXqh6QqRsAuabq/T97X1v1\n66vurKf62OqurZlVH//qDsDZrDm8+9OTfGFr7fOTPCDJQ1trP5Tj70T68X483Ok1rbUPJHl4krck\nOTlB7zGHj91N7/1qkqvJ8YylFbf/uqkekKkbALmm6v0/e19b9eur7qyn+tjqrq2ZVR//6g7A2aw+\nvDtJWmtPTfKNh+HdX5PkUb33b2+tPSHJK5P8qSRPSvLD+eDw7lcmebzh3QAAAADXz1mGd6/5HUv3\n5O8m+buttX+d5A+TfMXhu5de11p7SZJfTvL+JM++t0UlAAAAAGqtPbw7SdJ7f1Xv/dbD7/+w9/6l\nvfeP671/Uu/9Z0583vN774/rvT+x9/6y67FtM6gegDh7n131/tf1vfbZ7y1bPja6904A4IOuy8IS\ntaoHIM7eZ1e9/3V9r332e8uWj43uvRMA+KAbjo6OqrdhsatXrx5dvny5ejM277GPeHBuetCNue3S\nzXngjTfo17nPrnr/6/pe++z3li0fG917JwCM7sqVK287Ojq6eprPvS7Du9dieDcAAADAxTrL8G4/\nCgcAAADAIhaWJlA9gHP2Prvq/a/re+2z31u2fGx0750AwAdZWJpA9QDO2fvsqve/ru+1z35v2fKx\n0b13AgAfZHj3BKoHcM7eZ1e9/3V9r332e8uWj43uvRMARmd4NwAAAACLGN4NAAAAwOosLE2gegDn\n7H121ftf1/faZ7+3bPnY6N47AYAPsrA0geoBnLP32VXvf13fa5/93rLlY6N77wQAPsjw7glUD+Cc\nvc+uev/r+l777PeWLR8b3XsnAIzO8G4AAAAAFjG8GwAAAIDVWViaQPUAztn77Kr3v67vtc9+b9ny\nsdG9dwIAH2RhaQLVAzhn77Or3v+6vtc++71ly8dG994JAHyQ4d0TqB7AOXufXfX+1/W99tnvLVs+\nNrr3TgAYneHdAAAAACxieDcAAAAAq7OwBKyqegCsru+1zz7ceMvHRje8G2Crqu/f+pzvbxaWgFVV\nD4DV9b322Ycbb/nY6IZ3A2xV9f1bn/P9zfBuYFXVA2B1fa999uHGWz42uuHdAFtVff/Wx3l/M7wb\nAAAAgEUM7wYAAABgdRaWgFVVD8jT9b32WYc/XrPlY6MbbgqwVdX3b33O9zcLS8Cqqgfk6fpe+6zD\nH6/Z8rHRDTcF2Krq+7c+5/ub4d3AqqoH5On6XvtIwx+X2PKx0Q03Bdiq6vu3Ps77m+HdAAAAACxi\neDcAAAAAq7OwBKyqekCeru+1zzr88ZotHxvdcFOAraq+f+tzvr9ZWAJWVT0gT9f32mcd/njNlo+N\nbrgpwFZV37/1Od/fDO8GVlU9IE/X99pHGv64xJaPjW64KcBWVd+/9XHe3wzvBgAAAGARw7sBAAAA\nWJ2FJQAAAAAWsbAErKr6yQu6vtc+61NFrtnysdE9NQdgq6rv3/qc728WloBVVT95Qdf32md9qsg1\nWz42uqfmAGxV9f1bn/P9zVPhgFVVP3lB1/faR3qqyBJbPja6p+YAbFX1/Vsf5/3NU+EAAAAAWMRT\n4QAAAABYnYUlYGjVA/r0sTvrqT62umsLYI+q79/6nO9vFpaAoVUP6NPH7qyn+tjqri2APaq+f+tz\nvr8Z3g0MrXpAnz52Zz3Vx1Z3bQHsUfX9Wx/n/c3wbgAAAAAWudDh3a21r2+tfeT5NwsAAACAkZxm\nxtIjk/yL1tpLWmtPa621tTcK4KJUD+jTx+6sp/rY6q4tgD2qvn/rc76/3efCUu/925I8PskPJPnK\nJL/WWvvO1trjVt42gHOrHtCnj91ZT/Wx1V1bAHtUff/W53x/u99pPqn33ltrv5Xkt5K8P8lHJvmH\nrbVX9N6fu+YGApzHbZduvtuvun6RnfVUH1vdtQWwR9X3b33O97f7HN7dWvuGJF+e5B1JXpTkf+u9\n//vW2h9L8mu997LvXDK8GwAAAOBinWV492m+Y+mmJP9p7/03T36w9/6B1tqtSzYQAAAAgP07zYyl\n7/jQRaUT7VcufpMALk71gD597M56qo+t7toC2KPq+7c+5/vbaZ4KB7Bb1QP69LE766k+trprC2CP\nqu/f+pzvbzccHR1Vb8NiV69ePbp8+XL1ZgAb9thHPDg3PejG3Hbp5jzwxht0/UI766k+trprC2CP\nqu/f+jjvb1euXHnb0dHR1dN87n0O794yw7sBAAAALtZZhnf7UTgAAAAAFrGwBAytekCfPnafWfW+\n18fuACxTff/W53x/s7AEDK16QJ8+dp9Z9b7Xx+4ALFN9/9bnfH8zvBsYWvWAPn3sPrPqfa+P3QFY\npvr+rY/z/mZ4NwAAAACLGN4NAAAAwOosLAEAZ1Y9/FIfuwOwTPX9W5/z/c3CEgBwZtXDL/WxOwDL\nVN+/9Tnf3+5XvQEAwP7cdunmu/2q6xfZAVim+v6tz/n+Zng3AAAAAHcxvBsAAACA1VlYAgDOrHr4\npT52B2CZ6vu3Puf7m4UlAODMqodf6mN3AJapvn/rc76/Gd4NAJxZ9fBLfewOwDLV9299zvc3w7sB\nAAAAuIvh3QAAAACszsISAHBm1cMv9bE7AMtU37/1Od/fLCwBAGdWPfxSH7sDsEz1/Vuf8/3N8G4A\n4Myqh1/qY3cAlqm+f+tzvr8Z3g0AAADAXQzvBgAAAGB1FpYAgDOrHn6pj91nV73/9bk7+1Z9/uhz\nXl8WlgCAM6sefqmP3WdXvf/1uTv7Vn3+6HNeXzccHR1Vb8NiV69ePbp8+XL1ZgDAdB77iAfnpgfd\nmNsu3ZwH3niDrl9on131/tfn7uxb9fmjj3N9Xbly5W1HR0dXT/O5hncDAAAAcBfDuwEAAABYnYUl\nAAAAABaxsAQAnFn1U1V0T7UZWfXx1efucB7V5+/W+6gsLAEAZ1b9VBXdU21GVn189bk7nEf1+bv1\nPipPhQMAzqz6qSq6p9qMrPr46nN3OI/q83frfU88FQ4AAACARTwVDgAAAIDVWVgCAM6sevilbvjo\nyKqPrz53h/OoPn+33kdlYQkAOLPq4Ze64aMjqz6++twdzqP6/N16H5Xh3QDAmVUPv9QNHx1Z9fHV\n5+5wHtXn79b7nhjeDQAAAMAihncDAAAAsDoLSwDAmVUPv9QNHx1Z9fHV5+5wHtXn79b7qCwsAQBn\nVj38Ujd8dGTVx1efu8N5VJ+/W++jut/af0Fr7YYktyd5S+/91hMf/2+SfHeSR/Te33H42POSPCvJ\nnUme03t/+drbBwCc3W2Xbr7br/pcnXVVH1997g7nUX3+br2PavXh3a21v5rkUpKHXltYaq3dnORF\nSf50kj/Te39Ha+1JSX4kyVOSPCrJTyd5Qu/9znv62oZ3AwAAAFyszQzvbq09JsnTc7yIdNLfTvLc\nJCdXtZ6R5MW99/f13t+Q5PU5XmQCAAAAYIPWnrH0whwvIH3g2gdaa8/I8Y/F/cKHfO6jk5z8QcQ3\nHz4GAJu05oDH6uGS+rY766o+vrquu38uVb1/9TnPv9UWllprtyZ5e+/9tSc+9ieSfEuSbz/H173c\nWru9tXb7HXfccQFbCgDLrDngsXq4pL7tzrqqj6+u6+6fS1XvX33O82/N4d2fnuQLW2ufn+QBSR6a\n5AeTfEySX2itJcljkvzL1tpTkrwlyckJV485fOxueu9Xk1xNjmcsrbj9AHCv1hzwWD1cUt92Z13V\nx1fXdffPpar3rz7n+bf68O4kaa09Nck3nnwq3OHjb0xy6TC8+8lJfjgfHN79yiSPN7wbAAAA4Po5\ny/DuNb9j6Ux6769rrb0kyS8neX+SZ9/bohIAAAAAtdYe3p0k6b2/6kO/W+nw8Vt67+848f/P770/\nrvf+xN77y67HtgHAUoZ364Z/jqn6+Oq67v65VPX+1ec8/67LwhIAjMjwbt3wzzFVH19d190/l6re\nv/qc598NR0dH1duw2NWrV48uX75cvRkATOqxj3hwbnrQjbnt0s154I03XGhf82vr+++sq/r46rru\n/rlU9f7Vxzn/rly58rajo6Orp/nc6zK8ey2GdwMAAABcrLMM7/ajcAAAAAAsYmEJABYyvFs3/HNM\n1cdX13X3z6Wq968+5/lnYQkAFjK8Wzf8c0zVx1fXdffPpar3rz7n+Wd4NwAsZHi3bvjnmKqPr67r\n7p9LVe9ffZzzz/BuAAAAABYxvBsAAACA1VlYAoCFDO/WDf8cU/Xx1XXd/XOp6v2rz3n+WVgCgIUM\n79YN/xxT9fHVdd39c6nq/avPef4Z3g0ACxnerRv+Oabq46vruvvnUtX7Vx/n/DO8GwAAAIBFDO8G\nAAAAYHUWlgDYreoBjGv2LW+bztqqj6+u6/pe3z+qX58+578fLCwBsFvVAxjX7FveNp21VR9fXdf1\nvb5/VL8+fc5/PxjeDcBuVQ9gNLx73s66qo+vruv6Xt8/ql+fPs6/HwzvBgAAAGARw7sBAAAAWJ2F\nJQAAAAAWsbAEwG5VP9ljzb7lbdNZW/Xx1XVd3+v7R/Xr0+f894OFJQB2q/rJHmv2LW+bztqqj6+u\n6/pe3z+qX58+578fPBUOgN2qfrKHp8LN21lX9fHVdV3f6/tH9evTx/n3g6fCAQAAALCIp8IBAAAA\nsDoLSwDsVvUAxjX7lrdNZ23Vx1fXdX2v7x/Vr0+f898PFpYA2K3qAYxr9i1vm87aqo+vruv6Xt8/\nql+fPue/HwzvBmC3qgcwGt49b2dd1cdX13V9r+8f1a9PH+ffD4Z3AwAAALCI4d0AAAAArM7CEgC7\nVT2Acc2+5W3TWVv18dV1Xd/r+0f169Pn/PeDhSUAdqt6AOOafcvbprO26uOr67q+1/eP6tenz/nv\nB8O7Adit6gGMhnfP21lX9fHVdV3f6/tH9evTx/n3g+HdAAAAACxieDcAAAAAq7OwBMBi1QMQR+5b\n3rYt9NFV71/d+c16qs8vXdfd3y+ahSUAFqsegDhy3/K2baGPrnr/6s5v1lN9fum67v5+0QzvBmCx\n6gGII/ctb9sW+uiq96/u/GY91eeXruvu76dheDcAAAAAixjeDQAAAMDqLCwBsFj1AMSR+5a3bQt9\ndNX7V3d+s57q80vXdff3i2ZhCYDFqgcgjty3vG1b6KOr3r+685v1VJ9fuq67v180w7sBWKx6AOLI\nfcvbtoU+uur9qzu/WU/1+aXruvv7aRjeDQAAAMAihncDAAAAsDoLSwAsVj0AceS+5W3bQh9d9f7V\nnd+sp/r80nXd/f2iWVgCYLHqAYgj9y1v2xb66Kr3r+78Zj3V55eu6+7vF83wbgAWqx6AOHLf8rZt\noY+uev/qzm/WU31+6bru/n4ahncDAAAAsIjh3QAAAACszsISwI5VDyDUDe8edfhl9evT5+6Mrfr8\n0nXd/f2iWVgC2LHqAYS64d2jDr+sfn363J2xVZ9fuq67v180w7sBdqx6AKFuePeowy+rX58+d2ds\n1eeXruvu76dheDcAAAAAixjeDQAAAMDqLCwB7Fj1AEJ93uHdo6vev7rzG0ZVfX3rc3fWYWEJYMeq\nBxDq8w7vHl31/tWd3zCq6utbn7uzDsO7AXasegChPu/w7tFV71/d+Q2jqr6+9bk7p2d4NwAAAACL\nGN4NAAAAwOosLAHsWPUARN3w7lFV71/d+Q2jqr6+9bk767CwBLBj1QMQdcO7R1W9f3XnN4yq+vrW\n5+6sw/BugB2rHoCoG949qur9qzu/YVTV17c+d+f0DO8GAAAAYBHDuwEAAABYnYUlAAAAABaxsASw\nY9VP1tA9FW5U1ftXd37DqKqvb33uzjosLAHsWPWTNXRPhRtV9f7Vnd8wqurrW5+7sw5PhQPYseon\na+ieCjeq6v2rO79hVNXXtz535/Q8FQ4AAACARTwVDgAAAIDVWVgCOIfqAYT6uH3L26bruj5yZ13V\nx1efu7MOC0sA51A9gFAft29523Rd10furKv6+Opzd9ZheDfAOVQPINTH7VveNl3X9ZE766o+vvrc\nndMzvBsAAACARQzvBgAAAGB1FpYAzqF6AKE+bt/68Mkt7zt9/71a9evX5z7/GFv1+T17Zx0WlgDO\noXoAoT5u3/rwyS3vO33/vVr169fnPv8YW/X5PXtnHYZ3A5xD9QBCfdy+9eGTW953+v57terXr899\n/jG26vN79s7pGd4NAAAAwCKGdwMAAACwOgtLAOdQPYBQH7dvffjklvedvv9erfr163Off4yt+vye\nvbMOC0sA51A9gFAft299+OSW952+/16t+vXrc59/jK36/J69sw7DuwHOoXoAoT5u3/rwyS3vO33/\nvVr169fnPv8YW/X5PXvn9AzvBgAAAGARw7sBAAAAWJ2FJYBzqB5AqI/btz58csv7Tt9/r1b9+vW5\nzz/GVn1+z95Zh4UlgHOoHkCoj9u3Pnxyy/tO33+vVv369bnPP8ZWfX7P3lmH4d0A51A9gFAft299\n+OSW952+/16t+vXrc59/jK36/J69c3qGdwMAAACwiOHdAAAAAKzOwhIwtOoBgbpueLeu6/q+Oqyp\n+vzee78vru8aFpaAoVUPCNR1w7t1Xdf31WFN1ef33vt9cX3XMLwbGFr1gEBdN7xb13V9Xx3WVH1+\n773fF9f3xdnU8O7W2g1Jbk/ylt77ra21v5nkC5L8YZJfT/JVvfd3HT73eUmeleTOJM/pvb/83r62\n4d0AAAAAF2trw7u/IcmvnPj/VyT5uN77xyf5/5I8L0laa09K8swkT07ytCTfd1iUAgAAAGCDVl1Y\naq09JsnTk7zo2sd67z/Ve3//4X9/PsljDr9/RpIX997f13t/Q5LXJ3nKmtsHjK96wKA+dp9Z9b7X\nxz73q1+fPncH4GzW/o6lFyZ5bpIP3EP/L5O87PD7Ryc5OWHrzYeP3U1r7XJr7fbW2u133HHHRW4r\nMKDqAYP62H1m1fteH/vcr359+twdgLO531pfuLV2a5K3995f21p76ofp35rk/Un+wVm+bu/9apKr\nyfGMpQvYVGBgt126+W6/6vpF9plV73t97HO/+vXpc3cAzma14d2ttRck+bIcLx49IMlDk/x47/1L\nW2tfmeQvJfmc3vvvHT7/eUnSe3/B4f9fnuSo9/7qe/o7DO8GAAAAuFibGN7de39e7/0xvfdbcjyU\n+2cOi0pPy/GPx33htUWlg5cmeWZr7f6ttY9J8vgkr1lr+wAAAAA4n+vxVLgP9b1JHpLkFa21f9Va\n+/4k6b2/LslLkvxykp9M8uze+50F2wcMpHoAqD52n1n1vtfHPverX58+dwfgbK7LwlLv/VW991sP\nv//Y3vvNvfdPPPz3NSc+7/m998f13p/Ye3/ZPX9FgNOpHgCqj91nVr3v9bHP/erXp8/dATibG46O\njqq3YbGrV68eXb58uXozgA177CMenJsedGNuu3RzHnjjDbp+oX1m1fteH/vcr359+twdgOTKlStv\nOzo6unqaz11tePf1YHg3AAAAwMXaxPBuAAAAAMZmYQkYWvUAUH3sPrPqfa+Pfe5Xvz597g7A2VhY\nAoZWPQBUH7vPrHrf62Of+9WvT5+7A3A2hncDQ6seAKqP3WdWve/1sc/96tenz90BMLwbAAAAgIUM\n7wYAAABgdRaWAAAAAFjEwhIwtOony+hj95lV73td13X3doBtsLAEDK36yTL62H1m1fte13XdvR1g\nGzwVDhha9ZNl9LH7zKr3va7runs7wHo8FQ4AAACARTwVDgAAAIDVWVgChlY9AFQ3gJVlqs+d2c+t\n6v2r665vgP2wsAQMrXoAqG4AK8tUnzuzn1vV+1fXXd8A+2F4NzC06gGgugGsLFN97sx+blXvX113\nfQPUMrwbAAAAgEUM7wYAAABgdRaWgKFVDwDVDWBlmepzZ/Zzq3r/6rrrG2A/LCwBQ6seAKobwMoy\n1efO7OdW9f7Vddc3wH4Y3g0MrXoAqG4AK8tUnzuzn1vV+1fXXd8AtQzvBgAAAGARw7sBAAAAWJ2F\nJWBo1QNAdQNYWab63Jn93Krev7ru+gbYDwtLwNCqB4DqBrCyTPW5M/u5Vb1/dd31DbAfhncDQ6se\nAKobwMoy1efO7OdW9f7Vddc3QC3DuwEAAABYxPBuAAAAAFZnYQkYWvUA0Nk7sEz1tavr3jsAOC0L\nS8DQqgeAzt6BZaqvXV333gHAad2vegMA1nTbpZvv9qt+fTuwTPW1q+uVHYB9MbwbAAAAgLsY3g0A\nAADA6iwsAUOrHkA6eweWqb52dd17BwCnZWEJGFr1ANLZO7BM9bWr6947ADgtw7uBoVUPIJ29A8tU\nX7u6XtkB2BfDuwEAAAC4i+HdAAAAAKzOwhIwtOoBpLN3YJnqa1fXvXcAcFoWloChVQ8gnb0Dy1Rf\nu7ruvQOA0zK8Gxha9QDS2TuwTPW1q+uVHYB9MbwbAAAAgLsY3g0AAADA6iwsAUOrHkA6eweWqb52\ndd17BwCnZWEJGFr1ANLZO7BM9bWr6947ADgtw7uBoVUPIJ29A8tUX7u6XtkB2BfDuwEAAAC4i+Hd\nAAAAAKzOwhLAOVQPONXH7Vsfbrvlfafr+r47APtiYQngHKoHnOrj9q0Pt93yvtN1fd8dgH254ejo\nqHobFrt69erR5cuXqzcDmNhjH/Hg3PSgG3PbpZvzwBtv0PUL6/f1Z6tted/pur7vDkC9K1euvO3o\n6OjqaT7X8G4AAAAA7mJ4NwAAAACrs7AEAAAAwCIWlgDOofrJOfq4fetPTdryvtN1fd8dgH2xsARw\nDtVPztHH7Vt/atKW952u6/vuAOyLp8IBnEP1k3P0cfvWn5q05X2n6/q+OwD1PBUOAAAAgEU8FQ4A\nAACA1VlYAjiH6gGn1QNUq7dv5L714bZb3ne6ru+7A7AvFpYAzqF6wGn1ANXq7Ru5b3247Zb3na7r\n++4A7Ivh3QDnUD3gtHqAavX2jdy3Ptx2y/tO1/V9dwDqGd4NAAAAwCKGdwMAAACwOgtLAOdQPeBU\nH7dvfbjtlvedruv77gDsi4UlgHOoHnCqj9u3Ptx2y/tO1/V9dwD2xfBugHOoHnCqj9u3Ptx2y/tO\n1/V9dwDqGd4NAAAAwCKGdwMAAACwOgtLAOdQPeBUH7dvedt0XR+7A8BZWFgCOIfqAaf6uH3L26br\n+tgdAM7C8G6Ac6gecKqP27e8bbquj90BwPBuAAAAABYxvBsAAACA1VlYAjiH6gGr+rh9y9um6/rY\nHQDOwsISwDlUD1jVx+1b3jZd18fuAHAWhncDnEP1gFV93L7lbdN1fewOAIZ3AwAAALCI4d0AAAAA\nrM7CEsA5VA9Y1cftW942XdfH7gBwFhaWAM6hesCqPm7f8rbpuj52B4CzMLwb4ByqB6zq4/Ytb5uu\n62N3ADC8GwAAAIBFDO8GAAAAYHUWlgDOoXrAqj5u3/K26bo+dgeAs7CwBHAO1QNW9XH7lrdN1/Wx\nOwCcheHdAOdQPWBVH7dvedt0XR+7A4Dh3QAA/P/t3V+oZQd1B+DfYsYkmmitWkLMpGaQISUGqyWK\nVSnBCI11ML40ndK0sSqloBiLIpn6oIISQRH7UAshWgNV02DTOgipSlJpofVPVLAmURuM0UknZvwv\nLVUTVx/ORo8xN97s3H32nHO/72X23mvPzHpY5OasOfu3AQBGEd4NAAAAwOQslgDW2NwBr+rCu9XV\nN7EOAGyfxRLAGps74FVdeLe6+ibWAYDt2zt3AwCM9/vnn/Vzv6pvTv1E7k1dfdPrAMD2TR7eXVV7\nktyc5K7uPlhVj0vy90nOTvLVJJd093eGew8neVmS+5K8qrs/8mB/tvBuAAAAgJ11ooV3X57ktqXz\nK5Lc2N0Hktw4nKeqzk1yKMlTklyU5F3DUgoAAACAE9Cki6Wq2pfkhUmuXrp8cZJrhuNrkrx46fq1\n3f3D7r4jye1JnjllfwDrbu6AW3Xh3erqm1gHALZv6m8svTPJ65L8ZOna6d19bDi+O8npw/GZSZYT\nFI8O135OVf1ZVd1cVTcfP358gpYB1sfcAbfqwrvV1TexDgBs32Th3VV1MMk93f2Zqrrgge7p7q6q\nhxTy1N1XJbkqWWQsPexGAdbY3AG36sK71dU3sQ4AbN9k4d1VdWWSP05yb5JTkjwmyfVJnpHkgu4+\nVlVnJPl4d58zBHenu68cfv9Hkryxu/9jq79DeDcAAADAzjohwru7+3B37+vus7MI5b6puy9NciTJ\nZcNtlyX50HB8JMmhqjq5qvYnOZDkU1P1BwAAAMDDM9mjcA/irUmuq6qXJbkzySVJ0t23VNV1SW7N\n4ltOr+ju+2boDwAAAIBtmDq8O0nS3R/v7oPD8be6+8LuPtDdz+/uby/d95bufnJ3n9PdN6yiN4B1\nNvebk9S9FU5dfRPrAMD2rWSxBMA05n5zkrq3wqmrb2IdANi+OR6FA2CHzP3mJHVvhVNX38Q6ALB9\nk70VbhW8FQ4AAABgZ50Qb4UDAAAAYLNZLAGssbkDbtWFd6urb2IdANg+iyWANTZ3wK268G519U2s\nAwDbJ7wbYI3NHXCrLrxbXX0T6wDA9gnvBgAAAOCnhHcDAAAAMDmLJQBYQ3OHG6sLlwYASCyWAGAt\nzR1urC5cGgAgEd4NAGtp7nBjdeHSAACJ8G4AAAAAlgjvBgAAAGByFksAsIbmDo9WF94NAJBYLAHA\nWpo7PFpdeDcAQCK8GwDW0tzh0erCuwEAEuHdAAAAACwR3g0AAADA5CyWAGANzR0erS68GwAgsVgC\ngLU0d3i0uvBuAIBEeDcArKW5w6PVhXcDACTCuwEAAABYIrwbAAAAgMlZLAHAGpo7PFpdeDcAQGKx\nBABrae7waHXh3QAAifBuAFhLc4dHqwvvBgBIhHcDAAAAsER4NwAAAACTs1gCAAAAYBSLJQAAAABG\nsQcL8PwAAAZtSURBVFgCAAAAYBSLJQAAAABGsVgCAAAAYBSLJQAAAABGsVgCAAAAYBSLJQAAAABG\nsVgCAAAAYBSLJQAAAABGsVgCAAAAYBSLJQAAAABGsVgCAAAAYBSLJQAAAABGsVgCAAAAYBSLJQAA\nAABGsVgCAAAAYBSLJQAAAABGsVgCAAAAYBSLJQAAAABGsVgCAAAAYBSLJQAAAABGsVgCAAAAYBSL\nJQAAAABGsVgCAAAAYJTq7rl7GK2qjie5c5u3PyHJNydsBx6M+WNO5o+5mD3mZP6Yk/ljTuaPnfCk\n7v617dy41oulh6Kqbu7u8+fug93J/DEn88dczB5zMn/MyfwxJ/PHqnkUDgAAAIBRLJYAAAAAGGU3\nLZaumrsBdjXzx5zMH3Mxe8zJ/DEn88eczB8rtWsylgAAAADYWbvpG0sAAAAA7KBdsViqqouq6ktV\ndXtVXTF3P2yuqjqrqv6lqm6tqluq6vLh+uOq6mNV9V/Dr786d69srqraU1Wfq6oPD+fmj5WoqsdW\n1Qer6otVdVtV/bb5YxWq6i+Gn7tfqKoPVNUpZo8pVdV7quqeqvrC0rUtZ66qDg+fRb5UVb87T9ds\ngi1m723Dz97PV9U/VtVjl2pmj8lt/GKpqvYk+eskL0hybpI/rKpz5+2KDXZvktd097lJnpXkFcO8\nXZHkxu4+kOTG4RymcnmS25bOzR+r8ldJ/rm7fyPJb2Yxh+aPSVXVmUleleT87j4vyZ4kh2L2mNZ7\nk1x0v2sPOHPD/wseSvKU4fe8a/iMAmO8N784ex9Lcl53PzXJl5McTsweq7Pxi6Ukz0xye3d/pbt/\nlOTaJBfP3BMbqruPdfdnh+MfZPGh6swsZu6a4bZrkrx4ng7ZdFW1L8kLk1y9dNn8Mbmq+pUkv5Pk\n3UnS3T/q7u/G/LEae5M8sqr2JnlUkv+O2WNC3f2vSb59v8tbzdzFSa7t7h929x1Jbs/iMwo8ZA80\ne9390e6+dzj9RJJ9w7HZYyV2w2LpzCRfXzo/OlyDSVXV2UmenuSTSU7v7mND6e4kp8/UFpvvnUle\nl+QnS9fMH6uwP8nxJH87PIp5dVWdGvPHxLr7riRvT/K1JMeSfK+7Pxqzx+ptNXM+j7BKL01yw3Bs\n9liJ3bBYgpWrqtOS/EOSV3f395drvXgVo9cxsuOq6mCSe7r7M1vdY/6Y0N4kv5Xkb7r76Un+J/d7\n9Mj8MYUhx+biLJabT0xyalVdunyP2WPVzBxzqKrXZxHN8b65e2F32Q2LpbuSnLV0vm+4BpOoqkdk\nsVR6X3dfP1z+RlWdMdTPSHLPXP2x0Z6T5EVV9dUsHvt9XlX9Xcwfq3E0ydHu/uRw/sEsFk3mj6k9\nP8kd3X28u3+c5Pokz47ZY/W2mjmfR5hcVb0kycEkfzQsNhOzx4rshsXSp5McqKr9VXVSFuFlR2bu\niQ1VVZVFvsht3f2OpdKRJJcNx5cl+dCqe2Pzdffh7t7X3Wdn8d+6m7r70pg/VqC7707y9ao6Z7h0\nYZJbY/6Y3teSPKuqHjX8HL4wi4xDs8eqbTVzR5IcqqqTq2p/kgNJPjVDf2yoqrooiyiEF3X3/y6V\nzB4rUT9bZm6uqvq9LHJH9iR5T3e/ZeaW2FBV9dwk/5bkP/OzjJu/zCJn6bokv57kziSXdPf9Ax9h\nx1TVBUle290Hq+rxMX+sQFU9LYvg+JOSfCXJn2bxj1jmj0lV1ZuS/EEWj4B8LsnLk5wWs8dEquoD\nSS5I8oQk30jyhiT/lC1mbnhE6aVZzOiru/uGB/hj4ZfaYvYOJzk5ybeG2z7R3X8+3G/2mNyuWCwB\nAAAAsPN2w6NwAAAAAEzAYgkAAACAUSyWAAAAABjFYgkAAACAUSyWAAAAABjFYgkAAACAUSyWAAAA\nABjFYgkAYEJV9Yyq+nxVnVJVp1bVLVV13tx9AQDshOruuXsAANhoVfXmJKckeWSSo9195cwtAQDs\nCIslAICJVdVJST6d5P+SPLu775u5JQCAHeFROACA6T0+yWlJHp3FN5cAADaCbywBAEysqo4kuTbJ\n/iRndPcrZ24JAGBH7J27AQCATVZVf5Lkx939/qrak+Tfq+p53X3T3L0BADxcvrEEAAAAwCgylgAA\nAAAYxWIJAAAAgFEslgAAAAAYxWIJAAAAgFEslgAAAAAYxWIJAAAAgFEslgAAAAAYxWIJAAAAgFH+\nH+ab4wDoeeRVAAAAAElFTkSuQmCC\n", + "image/png": "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\n", "text/plain": [ - "" + "
" ] }, - "metadata": {}, + "metadata": { + "needs_background": "light" + }, "output_type": "display_data" } ], @@ -191,12 +197,14 @@ "outputs": [ { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfgAAAI1CAYAAAA6tuiIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n", + "image/png": "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\n", "text/plain": [ - "" + "
" ] }, - "metadata": {}, + "metadata": { + "needs_background": "light" + }, "output_type": "display_data" } ], @@ -228,7 +236,7 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 5, "metadata": {}, "outputs": [ { @@ -602,38 +610,44 @@ "outputs": [ { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfgAAAI1CAYAAAA6tuiIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n", + "image/png": "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\n", "text/plain": [ - "" + "
" ] }, - "metadata": {}, + "metadata": { + "needs_background": "light" + }, "output_type": "display_data" } ], "source": [ "import matplotlib.pyplot as plt\n", "\n", - "fig = plt.figure(figsize=(8,8))\n", + "def plot(data, n_samples, kmeans, title):\n", "\n", - "ax = fig.add_subplot(111)\n", - "fig.subplots_adjust(top=1)\n", - "ax.set_title('Clustered points in dataset n. 1')\n", + " fig = plt.figure(figsize=(8,8))\n", "\n", - "ax.set_xlabel('x')\n", - "ax.set_ylabel('y')\n", + " ax = fig.add_subplot(111)\n", + " fig.subplots_adjust(top=1)\n", + " ax.set_title(title)\n", "\n", - "# set the list of colors to be selected when plotting the different clusters\n", - "color=['b','g','r','c','m','y','k','w']\n", + " ax.set_xlabel('x')\n", + " ax.set_ylabel('y')\n", + "\n", + " # set the list of colors to be selected when plotting the different clusters\n", + " color=['b','g','r','c','m','y','k','w']\n", + "\n", + " #plot the dataset\n", + " for clu in range(k):\n", + " # collect the sequence of cooordinates of the points in each given cluster (determined by clu)\n", + " data_list_x = [data[i,0] for i in range(n_samples) if kmeans.labels_[i]==clu]\n", + " data_list_y = [data[i,1] for i in range(n_samples) if kmeans.labels_[i]==clu]\n", + " plt.scatter(data_list_x, data_list_y, s=2, edgecolors='none', c=color[clu], alpha=0.5)\n", + "\n", + " plt.show()\n", " \n", - "#plot the dataset\n", - "for clu in range(k):\n", - " # collect the sequence of cooordinates of the points in each given cluster (determined by clu)\n", - " data_list_x = [data1[i,0] for i in range(n_samples1) if kmeans1.labels_[i]==clu]\n", - " data_list_y = [data1[i,1] for i in range(n_samples1) if kmeans1.labels_[i]==clu]\n", - " plt.scatter(data_list_x, data_list_y, s=2, edgecolors='none', c=color[clu], alpha=0.5)\n", - "\n", - "plt.show()" + "plot(data1, n_samples1, kmeans1, 'Clustered points in dataset n. 1')" ] }, { @@ -647,10 +661,6163 @@ }, { "cell_type": "code", - "execution_count": 16, + "execution_count": 7, "metadata": {}, - "outputs": [], - "source": [] + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Example n. 0 = (5.0, 482.0): cluster 0\n", + "Example n. 1 = (5.0, 481.0): cluster 0\n", + "Example n. 2 = (5.0, 480.0): cluster 0\n", + "Example n. 3 = (5.0, 479.0): cluster 0\n", + "Example n. 4 = (5.0, 478.0): cluster 0\n", + "Example n. 5 = (5.0, 477.0): cluster 0\n", + "Example n. 6 = (5.0, 476.0): cluster 0\n", + "Example n. 7 = (5.0, 475.0): cluster 0\n", + "Example n. 8 = (5.0, 474.0): cluster 0\n", + "Example n. 9 = (5.0, 473.0): cluster 0\n", + "Example n. 10 = (5.0, 472.0): cluster 0\n", + "Example n. 11 = (5.0, 471.0): cluster 0\n", + "Example n. 12 = (5.0, 470.0): cluster 0\n", + "Example n. 13 = (5.0, 469.0): cluster 0\n", + "Example n. 14 = (5.0, 468.0): cluster 0\n", + "Example n. 15 = (5.0, 467.0): cluster 0\n", + "Example n. 16 = (5.0, 466.0): cluster 0\n", + "Example n. 17 = (5.0, 465.0): cluster 0\n", + "Example n. 18 = (5.0, 464.0): cluster 0\n", + "Example n. 19 = (5.0, 463.0): cluster 0\n", + "Example n. 20 = (5.0, 462.0): cluster 0\n", + "Example n. 21 = (5.0, 461.0): cluster 0\n", + "Example n. 22 = (5.0, 460.0): cluster 0\n", + "Example n. 23 = (6.0, 482.0): cluster 0\n", + "Example n. 24 = (6.0, 481.0): cluster 0\n", + "Example n. 25 = (6.0, 480.0): cluster 0\n", + "Example n. 26 = (6.0, 479.0): cluster 0\n", + "Example n. 27 = (6.0, 478.0): cluster 0\n", + "Example n. 28 = (6.0, 477.0): cluster 0\n", + "Example n. 29 = (6.0, 476.0): cluster 0\n", + "Example n. 30 = (6.0, 475.0): cluster 0\n", + "Example n. 31 = (6.0, 474.0): cluster 0\n", + "Example n. 32 = (6.0, 473.0): cluster 0\n", + "Example n. 33 = (6.0, 472.0): cluster 0\n", + "Example n. 34 = (6.0, 471.0): cluster 0\n", + "Example n. 35 = (6.0, 470.0): cluster 0\n", + "Example n. 36 = (6.0, 469.0): cluster 0\n", + "Example n. 37 = (6.0, 468.0): cluster 0\n", + "Example n. 38 = (6.0, 467.0): cluster 0\n", + "Example n. 39 = (6.0, 466.0): cluster 0\n", + "Example n. 40 = (6.0, 465.0): cluster 0\n", + "Example n. 41 = (6.0, 464.0): cluster 0\n", + "Example n. 42 = (6.0, 463.0): cluster 0\n", + "Example n. 43 = (6.0, 462.0): cluster 0\n", + "Example n. 44 = (6.0, 461.0): cluster 0\n", + "Example n. 45 = (6.0, 460.0): cluster 0\n", + "Example n. 46 = (7.0, 482.0): cluster 0\n", + "Example n. 47 = (7.0, 481.0): cluster 0\n", + "Example n. 48 = (7.0, 480.0): cluster 0\n", + "Example n. 49 = (7.0, 479.0): cluster 0\n", + "Example n. 50 = (7.0, 478.0): cluster 0\n", + "Example n. 51 = (7.0, 477.0): cluster 0\n", + "Example n. 52 = (7.0, 476.0): cluster 0\n", + "Example n. 53 = (7.0, 475.0): cluster 0\n", + "Example n. 54 = (7.0, 474.0): cluster 0\n", + "Example n. 55 = (7.0, 473.0): cluster 0\n", + "Example n. 56 = (7.0, 472.0): cluster 0\n", + "Example n. 57 = (7.0, 471.0): cluster 0\n", + "Example n. 58 = (7.0, 470.0): cluster 0\n", + "Example n. 59 = (7.0, 469.0): cluster 0\n", + "Example n. 60 = (7.0, 468.0): cluster 0\n", + "Example n. 61 = (7.0, 467.0): cluster 0\n", + "Example n. 62 = (7.0, 466.0): cluster 0\n", + "Example n. 63 = (7.0, 465.0): cluster 0\n", + "Example n. 64 = (7.0, 464.0): cluster 0\n", + "Example n. 65 = (7.0, 463.0): cluster 0\n", + "Example n. 66 = (7.0, 462.0): cluster 0\n", + "Example n. 67 = (7.0, 461.0): cluster 0\n", + "Example n. 68 = (7.0, 460.0): cluster 0\n", + "Example n. 69 = (8.0, 490.0): cluster 0\n", + "Example n. 70 = (8.0, 489.0): cluster 0\n", + "Example n. 71 = (8.0, 488.0): cluster 0\n", + "Example n. 72 = (8.0, 487.0): cluster 0\n", + "Example n. 73 = (8.0, 486.0): cluster 0\n", + "Example n. 74 = (8.0, 485.0): cluster 0\n", + "Example n. 75 = (8.0, 484.0): cluster 0\n", + "Example n. 76 = (8.0, 483.0): cluster 0\n", + "Example n. 77 = (8.0, 482.0): cluster 0\n", + "Example n. 78 = (8.0, 481.0): cluster 0\n", + "Example n. 79 = (8.0, 480.0): cluster 0\n", + "Example n. 80 = (8.0, 479.0): cluster 0\n", + "Example n. 81 = (8.0, 478.0): cluster 0\n", + "Example n. 82 = (8.0, 477.0): cluster 0\n", + "Example n. 83 = (8.0, 476.0): cluster 0\n", + "Example n. 84 = (8.0, 475.0): cluster 0\n", + "Example n. 85 = (8.0, 474.0): cluster 0\n", + "Example n. 86 = (8.0, 473.0): cluster 0\n", + "Example n. 87 = (8.0, 472.0): cluster 0\n", + "Example n. 88 = (8.0, 471.0): cluster 0\n", + "Example n. 89 = (8.0, 470.0): cluster 0\n", + "Example n. 90 = (8.0, 469.0): cluster 0\n", + "Example n. 91 = (8.0, 468.0): cluster 0\n", + "Example n. 92 = (8.0, 467.0): cluster 0\n", + "Example n. 93 = (8.0, 466.0): cluster 0\n", + "Example n. 94 = (8.0, 465.0): cluster 0\n", + "Example n. 95 = (8.0, 464.0): cluster 0\n", + "Example n. 96 = (8.0, 463.0): cluster 0\n", + "Example n. 97 = (8.0, 462.0): cluster 0\n", + "Example n. 98 = (8.0, 461.0): cluster 0\n", + "Example n. 99 = (8.0, 460.0): cluster 0\n", + "Example n. 100 = (8.0, 459.0): cluster 0\n", + "Example n. 101 = (8.0, 458.0): cluster 0\n", + "Example n. 102 = (8.0, 457.0): cluster 0\n", + "Example n. 103 = (8.0, 456.0): cluster 0\n", + "Example n. 104 = (9.0, 490.0): cluster 0\n", + "Example n. 105 = (9.0, 489.0): cluster 0\n", + "Example n. 106 = (9.0, 488.0): cluster 0\n", + "Example n. 107 = (9.0, 487.0): cluster 0\n", + "Example n. 108 = (9.0, 486.0): cluster 0\n", + "Example n. 109 = (9.0, 485.0): cluster 0\n", + "Example n. 110 = (9.0, 484.0): cluster 0\n", + "Example n. 111 = (9.0, 483.0): cluster 0\n", + "Example n. 112 = (9.0, 482.0): cluster 0\n", + "Example n. 113 = (9.0, 481.0): cluster 0\n", + "Example n. 114 = (9.0, 480.0): cluster 0\n", + "Example n. 115 = (9.0, 479.0): cluster 0\n", + "Example n. 116 = (9.0, 478.0): cluster 0\n", + "Example n. 117 = (9.0, 477.0): cluster 0\n", + "Example n. 118 = (9.0, 476.0): cluster 0\n", + "Example n. 119 = (9.0, 475.0): cluster 0\n", + "Example n. 120 = (9.0, 474.0): cluster 0\n", + "Example n. 121 = (9.0, 473.0): cluster 0\n", + "Example n. 122 = (9.0, 472.0): cluster 0\n", + "Example n. 123 = (9.0, 471.0): cluster 0\n", + "Example n. 124 = (9.0, 470.0): cluster 0\n", + "Example n. 125 = (9.0, 469.0): cluster 0\n", + "Example n. 126 = (9.0, 468.0): cluster 0\n", + "Example n. 127 = (9.0, 467.0): cluster 0\n", + "Example n. 128 = (9.0, 466.0): cluster 0\n", + "Example n. 129 = (9.0, 465.0): cluster 0\n", + "Example n. 130 = (9.0, 464.0): cluster 0\n", + "Example n. 131 = (9.0, 463.0): cluster 0\n", + "Example n. 132 = (9.0, 462.0): cluster 0\n", + "Example n. 133 = (9.0, 461.0): cluster 0\n", + "Example n. 134 = (9.0, 460.0): cluster 0\n", + "Example n. 135 = (9.0, 459.0): cluster 0\n", + "Example n. 136 = (9.0, 458.0): cluster 0\n", + "Example n. 137 = (9.0, 457.0): cluster 0\n", + "Example n. 138 = (9.0, 456.0): cluster 0\n", + "Example n. 139 = (10.0, 490.0): cluster 0\n", + "Example n. 140 = (10.0, 489.0): cluster 0\n", + "Example n. 141 = (10.0, 488.0): cluster 0\n", + "Example n. 142 = (10.0, 487.0): cluster 0\n", + "Example n. 143 = (10.0, 486.0): cluster 0\n", + "Example n. 144 = (10.0, 485.0): cluster 0\n", + "Example n. 145 = (10.0, 484.0): cluster 0\n", + "Example n. 146 = (10.0, 483.0): cluster 0\n", + "Example n. 147 = (10.0, 482.0): cluster 0\n", + "Example n. 148 = (10.0, 481.0): cluster 0\n", + "Example n. 149 = (10.0, 480.0): cluster 0\n", + "Example n. 150 = (10.0, 479.0): cluster 0\n", + "Example n. 151 = (10.0, 478.0): cluster 0\n", + "Example n. 152 = (10.0, 477.0): cluster 0\n", + "Example n. 153 = (10.0, 476.0): cluster 0\n", + "Example n. 154 = (10.0, 475.0): cluster 0\n", + "Example n. 155 = (10.0, 474.0): cluster 0\n", + "Example n. 156 = (10.0, 473.0): cluster 0\n", + "Example n. 157 = (10.0, 472.0): cluster 0\n", + "Example n. 158 = (10.0, 471.0): cluster 0\n", + "Example n. 159 = (10.0, 470.0): cluster 0\n", + "Example n. 160 = (10.0, 469.0): cluster 0\n", + "Example n. 161 = (10.0, 468.0): cluster 0\n", + "Example n. 162 = (10.0, 467.0): cluster 0\n", + "Example n. 163 = (10.0, 466.0): cluster 0\n", + "Example n. 164 = (10.0, 465.0): cluster 0\n", + "Example n. 165 = (10.0, 464.0): cluster 0\n", + "Example n. 166 = (10.0, 463.0): cluster 0\n", + "Example n. 167 = (10.0, 462.0): cluster 0\n", + "Example n. 168 = (10.0, 461.0): cluster 0\n", + "Example n. 169 = (10.0, 460.0): cluster 0\n", + "Example n. 170 = (10.0, 459.0): cluster 0\n", + "Example n. 171 = (10.0, 458.0): cluster 0\n", + "Example n. 172 = (10.0, 457.0): cluster 0\n", + "Example n. 173 = (10.0, 456.0): cluster 0\n", + "Example n. 174 = (11.0, 490.0): cluster 0\n", + "Example n. 175 = (11.0, 489.0): cluster 0\n", + "Example n. 176 = (11.0, 488.0): cluster 0\n", + "Example n. 177 = (11.0, 487.0): cluster 0\n", + "Example n. 178 = (11.0, 486.0): cluster 0\n", + "Example n. 179 = (11.0, 485.0): cluster 0\n", + "Example n. 180 = (11.0, 484.0): cluster 0\n", + "Example n. 181 = (11.0, 483.0): cluster 0\n", + "Example n. 182 = (11.0, 482.0): cluster 0\n", + "Example n. 183 = (11.0, 481.0): cluster 0\n", + "Example n. 184 = (11.0, 480.0): cluster 0\n", + "Example n. 185 = (11.0, 479.0): cluster 0\n", + "Example n. 186 = (11.0, 478.0): cluster 0\n", + "Example n. 187 = (11.0, 477.0): cluster 0\n", + "Example n. 188 = (11.0, 476.0): cluster 0\n", + "Example n. 189 = (11.0, 475.0): cluster 0\n", + "Example n. 190 = (11.0, 474.0): cluster 0\n", + "Example n. 191 = (11.0, 473.0): cluster 0\n", + "Example n. 192 = (11.0, 472.0): cluster 0\n", + "Example n. 193 = (11.0, 471.0): cluster 0\n", + "Example n. 194 = (11.0, 470.0): cluster 0\n", + "Example n. 195 = (11.0, 469.0): cluster 0\n", + "Example n. 196 = (11.0, 468.0): cluster 0\n", + "Example n. 197 = (11.0, 467.0): cluster 0\n", + "Example n. 198 = (11.0, 466.0): cluster 0\n", + "Example n. 199 = (11.0, 465.0): cluster 0\n", + "Example n. 200 = (11.0, 464.0): cluster 0\n", + "Example n. 201 = (11.0, 463.0): cluster 0\n", + "Example n. 202 = (11.0, 462.0): cluster 0\n", + "Example n. 203 = (11.0, 461.0): cluster 0\n", + "Example n. 204 = (11.0, 460.0): cluster 0\n", + "Example n. 205 = (11.0, 459.0): cluster 0\n", + "Example n. 206 = (11.0, 458.0): cluster 0\n", + "Example n. 207 = (11.0, 457.0): cluster 0\n", + "Example n. 208 = (11.0, 456.0): cluster 0\n", + "Example n. 209 = (12.0, 494.0): cluster 0\n", + "Example n. 210 = (12.0, 493.0): cluster 0\n", + "Example n. 211 = (12.0, 492.0): cluster 0\n", + "Example n. 212 = (12.0, 491.0): cluster 0\n", + "Example n. 213 = (12.0, 490.0): cluster 0\n", + "Example n. 214 = (12.0, 489.0): cluster 0\n", + "Example n. 215 = (12.0, 488.0): cluster 0\n", + "Example n. 216 = (12.0, 487.0): cluster 0\n", + "Example n. 217 = (12.0, 486.0): cluster 0\n", + "Example n. 218 = (12.0, 485.0): cluster 0\n", + "Example n. 219 = (12.0, 484.0): cluster 0\n", + "Example n. 220 = (12.0, 483.0): cluster 0\n", + "Example n. 221 = (12.0, 482.0): cluster 0\n", + "Example n. 222 = (12.0, 481.0): cluster 0\n", + "Example n. 223 = (12.0, 480.0): cluster 0\n", + "Example n. 224 = (12.0, 479.0): cluster 0\n", + "Example n. 225 = (12.0, 478.0): cluster 0\n", + "Example n. 226 = (12.0, 477.0): cluster 0\n", + "Example n. 227 = (12.0, 476.0): cluster 0\n", + "Example n. 228 = (12.0, 475.0): cluster 0\n", + "Example n. 229 = (12.0, 474.0): cluster 0\n", + "Example n. 230 = (12.0, 473.0): cluster 0\n", + "Example n. 231 = (12.0, 472.0): cluster 0\n", + "Example n. 232 = (12.0, 471.0): cluster 0\n", + "Example n. 233 = (12.0, 470.0): cluster 0\n", + "Example n. 234 = (12.0, 469.0): cluster 0\n", + "Example n. 235 = (12.0, 468.0): cluster 0\n", + "Example n. 236 = (12.0, 467.0): cluster 0\n", + "Example n. 237 = (12.0, 466.0): cluster 0\n", + "Example n. 238 = (12.0, 465.0): cluster 0\n", + "Example n. 239 = (12.0, 464.0): cluster 0\n", + "Example n. 240 = (12.0, 463.0): cluster 0\n", + "Example n. 241 = (12.0, 462.0): cluster 0\n", + "Example n. 242 = (12.0, 461.0): cluster 0\n", + "Example n. 243 = (12.0, 460.0): cluster 0\n", + "Example n. 244 = (12.0, 459.0): cluster 0\n", + "Example n. 245 = (12.0, 458.0): cluster 0\n", + "Example n. 246 = (12.0, 457.0): cluster 0\n", + "Example n. 247 = (12.0, 456.0): cluster 0\n", + "Example n. 248 = (13.0, 494.0): cluster 0\n", + "Example n. 249 = (13.0, 493.0): cluster 0\n", + "Example n. 250 = (13.0, 492.0): cluster 0\n", + "Example n. 251 = (13.0, 491.0): cluster 0\n", + "Example n. 252 = (13.0, 490.0): cluster 0\n", + "Example n. 253 = (13.0, 489.0): cluster 0\n", + "Example n. 254 = (13.0, 488.0): cluster 0\n", + "Example n. 255 = (13.0, 487.0): cluster 0\n", + "Example n. 256 = (13.0, 486.0): cluster 0\n", + "Example n. 257 = (13.0, 485.0): cluster 0\n", + "Example n. 258 = (13.0, 484.0): cluster 0\n", + "Example n. 259 = (13.0, 483.0): cluster 0\n", + "Example n. 260 = (13.0, 482.0): cluster 0\n", + "Example n. 261 = (13.0, 481.0): cluster 0\n", + "Example n. 262 = (13.0, 480.0): cluster 0\n", + "Example n. 263 = (13.0, 479.0): cluster 0\n", + "Example n. 264 = (13.0, 478.0): cluster 0\n", + "Example n. 265 = (13.0, 477.0): cluster 0\n", + "Example n. 266 = (13.0, 476.0): cluster 0\n", + "Example n. 267 = (13.0, 475.0): cluster 0\n", + "Example n. 268 = (13.0, 474.0): cluster 0\n", + "Example n. 269 = (13.0, 473.0): cluster 0\n", + "Example n. 270 = (13.0, 472.0): cluster 0\n", + "Example n. 271 = (13.0, 471.0): cluster 0\n", + "Example n. 272 = (13.0, 470.0): cluster 0\n", + "Example n. 273 = (13.0, 469.0): cluster 0\n", + "Example n. 274 = (13.0, 468.0): cluster 0\n", + "Example n. 275 = (13.0, 467.0): cluster 0\n", + "Example n. 276 = (13.0, 466.0): cluster 0\n", + "Example n. 277 = (13.0, 465.0): cluster 0\n", + "Example n. 278 = (13.0, 464.0): cluster 0\n", + "Example n. 279 = (13.0, 463.0): cluster 0\n", + "Example n. 280 = (13.0, 462.0): cluster 0\n", + "Example n. 281 = (13.0, 461.0): cluster 0\n", + "Example n. 282 = (13.0, 460.0): cluster 0\n", + "Example n. 283 = (13.0, 459.0): cluster 0\n", + "Example n. 284 = (13.0, 458.0): cluster 0\n", + "Example n. 285 = (13.0, 457.0): cluster 0\n", + "Example n. 286 = (13.0, 456.0): cluster 0\n", + "Example n. 287 = (14.0, 494.0): cluster 0\n", + "Example n. 288 = (14.0, 493.0): cluster 0\n", + "Example n. 289 = (14.0, 492.0): cluster 0\n", + "Example n. 290 = (14.0, 491.0): cluster 0\n", + "Example n. 291 = (14.0, 490.0): cluster 0\n", + "Example n. 292 = (14.0, 489.0): cluster 0\n", + "Example n. 293 = (14.0, 488.0): cluster 0\n", + "Example n. 294 = (14.0, 487.0): cluster 0\n", + "Example n. 295 = (14.0, 486.0): cluster 0\n", + "Example n. 296 = (14.0, 485.0): cluster 0\n", + "Example n. 297 = (14.0, 484.0): cluster 0\n", + "Example n. 298 = (14.0, 483.0): cluster 0\n", + "Example n. 299 = (14.0, 482.0): cluster 0\n", + "Example n. 300 = (14.0, 481.0): cluster 0\n", + "Example n. 301 = (14.0, 480.0): cluster 0\n", + "Example n. 302 = (14.0, 479.0): cluster 0\n", + "Example n. 303 = (14.0, 478.0): cluster 0\n", + "Example n. 304 = (14.0, 477.0): cluster 0\n", + "Example n. 305 = (14.0, 476.0): cluster 0\n", + "Example n. 306 = (14.0, 475.0): cluster 0\n", + "Example n. 307 = (14.0, 474.0): cluster 0\n", + "Example n. 308 = (14.0, 473.0): cluster 0\n", + "Example n. 309 = (14.0, 472.0): cluster 0\n", + "Example n. 310 = (14.0, 471.0): cluster 0\n", + "Example n. 311 = (14.0, 470.0): cluster 0\n", + "Example n. 312 = (14.0, 469.0): cluster 0\n", + "Example n. 313 = (14.0, 468.0): cluster 0\n", + "Example n. 314 = (14.0, 467.0): cluster 0\n", + "Example n. 315 = (14.0, 466.0): cluster 0\n", + "Example n. 316 = (14.0, 465.0): cluster 0\n", + "Example n. 317 = (14.0, 464.0): cluster 0\n", + "Example n. 318 = (14.0, 463.0): cluster 0\n", + "Example n. 319 = (14.0, 462.0): cluster 0\n", + "Example n. 320 = (14.0, 461.0): cluster 0\n", + "Example n. 321 = (14.0, 460.0): cluster 0\n", + "Example n. 322 = (14.0, 459.0): cluster 0\n", + "Example n. 323 = (14.0, 458.0): cluster 0\n", + "Example n. 324 = (14.0, 457.0): cluster 0\n", + "Example n. 325 = (14.0, 456.0): cluster 0\n", + "Example n. 326 = (15.0, 494.0): cluster 0\n", + "Example n. 327 = (15.0, 493.0): cluster 0\n", + "Example n. 328 = (15.0, 492.0): cluster 0\n", + "Example n. 329 = (15.0, 491.0): cluster 0\n", + "Example n. 330 = (15.0, 490.0): cluster 0\n", + "Example n. 331 = (15.0, 489.0): cluster 0\n", + "Example n. 332 = (15.0, 488.0): cluster 0\n", + "Example n. 333 = (15.0, 487.0): cluster 0\n", + "Example n. 334 = (15.0, 486.0): cluster 0\n", + "Example n. 335 = (15.0, 485.0): cluster 0\n", + "Example n. 336 = (15.0, 484.0): cluster 0\n", + "Example n. 337 = (15.0, 483.0): cluster 0\n", + "Example n. 338 = (15.0, 482.0): cluster 0\n", + "Example n. 339 = (15.0, 481.0): cluster 0\n", + "Example n. 340 = (15.0, 480.0): cluster 0\n", + "Example n. 341 = (15.0, 479.0): cluster 0\n", + "Example n. 342 = (15.0, 478.0): cluster 0\n", + "Example n. 343 = (15.0, 477.0): cluster 0\n", + "Example n. 344 = (15.0, 476.0): cluster 0\n", + "Example n. 345 = (15.0, 475.0): cluster 0\n", + "Example n. 346 = (15.0, 474.0): cluster 0\n", + "Example n. 347 = (15.0, 473.0): cluster 0\n", + "Example n. 348 = (15.0, 472.0): cluster 0\n", + "Example n. 349 = (15.0, 471.0): cluster 0\n", + "Example n. 350 = (15.0, 470.0): cluster 0\n", + "Example n. 351 = (15.0, 469.0): cluster 0\n", + "Example n. 352 = (15.0, 468.0): cluster 0\n", + "Example n. 353 = (15.0, 467.0): cluster 0\n", + "Example n. 354 = (15.0, 466.0): cluster 0\n", + "Example n. 355 = (15.0, 465.0): cluster 0\n", + "Example n. 356 = (15.0, 464.0): cluster 0\n", + "Example n. 357 = (15.0, 463.0): cluster 0\n", + "Example n. 358 = (15.0, 462.0): cluster 0\n", + "Example n. 359 = (15.0, 461.0): cluster 0\n", + "Example n. 360 = (15.0, 460.0): cluster 0\n", + "Example n. 361 = (15.0, 459.0): cluster 0\n", + "Example n. 362 = (15.0, 458.0): cluster 0\n", + "Example n. 363 = (15.0, 457.0): cluster 0\n", + "Example n. 364 = (15.0, 456.0): cluster 0\n", + "Example n. 365 = (16.0, 494.0): cluster 0\n", + "Example n. 366 = (16.0, 493.0): cluster 0\n", + "Example n. 367 = (16.0, 492.0): cluster 0\n", + "Example n. 368 = (16.0, 491.0): cluster 0\n", + "Example n. 369 = (16.0, 490.0): cluster 0\n", + "Example n. 370 = (16.0, 489.0): cluster 0\n", + "Example n. 371 = (16.0, 488.0): cluster 0\n", + "Example n. 372 = (16.0, 487.0): cluster 0\n", + "Example n. 373 = (16.0, 486.0): cluster 0\n", + "Example n. 374 = (16.0, 485.0): cluster 0\n", + "Example n. 375 = (16.0, 484.0): cluster 0\n", + "Example n. 376 = (16.0, 483.0): cluster 0\n", + "Example n. 377 = (16.0, 482.0): cluster 0\n", + "Example n. 378 = (16.0, 481.0): cluster 0\n", + "Example n. 379 = (16.0, 480.0): cluster 0\n", + "Example n. 380 = (16.0, 479.0): cluster 0\n", + "Example n. 381 = (16.0, 478.0): cluster 0\n", + "Example n. 382 = (16.0, 477.0): cluster 0\n", + "Example n. 383 = (16.0, 476.0): cluster 0\n", + "Example n. 384 = (16.0, 475.0): cluster 0\n", + "Example n. 385 = (16.0, 474.0): cluster 0\n", + "Example n. 386 = (16.0, 473.0): cluster 0\n", + "Example n. 387 = (16.0, 472.0): cluster 0\n", + "Example n. 388 = (16.0, 471.0): cluster 0\n", + "Example n. 389 = (16.0, 470.0): cluster 0\n", + "Example n. 390 = (16.0, 469.0): cluster 0\n", + "Example n. 391 = (16.0, 468.0): cluster 0\n", + "Example n. 392 = (16.0, 467.0): cluster 0\n", + "Example n. 393 = (16.0, 466.0): cluster 0\n", + "Example n. 394 = (16.0, 465.0): cluster 0\n", + "Example n. 395 = (16.0, 464.0): cluster 0\n", + "Example n. 396 = (16.0, 463.0): cluster 0\n", + "Example n. 397 = (16.0, 462.0): cluster 0\n", + "Example n. 398 = (16.0, 461.0): cluster 0\n", + "Example n. 399 = (16.0, 460.0): cluster 0\n", + "Example n. 400 = (16.0, 459.0): cluster 0\n", + "Example n. 401 = (16.0, 458.0): cluster 0\n", + "Example n. 402 = (16.0, 457.0): cluster 0\n", + "Example n. 403 = (16.0, 456.0): cluster 0\n", + "Example n. 404 = (17.0, 494.0): cluster 0\n", + "Example n. 405 = (17.0, 493.0): cluster 0\n", + "Example n. 406 = (17.0, 492.0): cluster 0\n", + "Example n. 407 = (17.0, 491.0): cluster 0\n", + "Example n. 408 = (17.0, 490.0): cluster 0\n", + "Example n. 409 = (17.0, 489.0): cluster 0\n", + "Example n. 410 = (17.0, 488.0): cluster 0\n", + "Example n. 411 = (17.0, 487.0): cluster 0\n", + "Example n. 412 = (17.0, 486.0): cluster 0\n", + "Example n. 413 = (17.0, 485.0): cluster 0\n", + "Example n. 414 = (17.0, 484.0): cluster 0\n", + "Example n. 415 = (17.0, 483.0): cluster 0\n", + "Example n. 416 = (17.0, 482.0): cluster 0\n", + "Example n. 417 = (17.0, 481.0): cluster 0\n", + "Example n. 418 = (17.0, 480.0): cluster 0\n", + "Example n. 419 = (17.0, 479.0): cluster 0\n", + "Example n. 420 = (17.0, 478.0): cluster 0\n", + "Example n. 421 = (17.0, 477.0): cluster 0\n", + "Example n. 422 = (17.0, 476.0): cluster 0\n", + "Example n. 423 = (17.0, 475.0): cluster 0\n", + "Example n. 424 = (17.0, 474.0): cluster 0\n", + "Example n. 425 = (17.0, 473.0): cluster 0\n", + "Example n. 426 = (17.0, 472.0): cluster 0\n", + "Example n. 427 = (17.0, 471.0): cluster 0\n", + "Example n. 428 = (17.0, 470.0): cluster 0\n", + "Example n. 429 = (17.0, 469.0): cluster 0\n", + "Example n. 430 = (17.0, 468.0): cluster 0\n", + "Example n. 431 = (17.0, 467.0): cluster 0\n", + "Example n. 432 = (17.0, 466.0): cluster 0\n", + "Example n. 433 = (17.0, 465.0): cluster 0\n", + "Example n. 434 = (17.0, 464.0): cluster 0\n", + "Example n. 435 = (17.0, 463.0): cluster 0\n", + "Example n. 436 = (17.0, 462.0): cluster 0\n", + "Example n. 437 = (17.0, 461.0): cluster 0\n", + "Example n. 438 = (17.0, 460.0): cluster 0\n", + "Example n. 439 = (17.0, 459.0): cluster 0\n", + "Example n. 440 = (17.0, 458.0): cluster 0\n", + "Example n. 441 = (17.0, 457.0): cluster 0\n", + "Example n. 442 = (17.0, 456.0): cluster 0\n", + "Example n. 443 = (18.0, 494.0): cluster 0\n", + "Example n. 444 = (18.0, 493.0): cluster 0\n", + "Example n. 445 = (18.0, 492.0): cluster 0\n", + "Example n. 446 = (18.0, 491.0): cluster 0\n", + "Example n. 447 = (18.0, 490.0): cluster 0\n", + "Example n. 448 = (18.0, 489.0): cluster 0\n", + "Example n. 449 = (18.0, 488.0): cluster 0\n", + "Example n. 450 = (18.0, 487.0): cluster 0\n", + "Example n. 451 = (18.0, 486.0): cluster 0\n", + "Example n. 452 = (18.0, 485.0): cluster 0\n", + "Example n. 453 = (18.0, 484.0): cluster 0\n", + "Example n. 454 = (18.0, 483.0): cluster 0\n", + "Example n. 455 = (18.0, 482.0): cluster 0\n", + "Example n. 456 = (18.0, 481.0): cluster 0\n", + "Example n. 457 = (18.0, 480.0): cluster 0\n", + "Example n. 458 = (18.0, 479.0): cluster 0\n", + "Example n. 459 = (18.0, 478.0): cluster 0\n", + "Example n. 460 = (18.0, 477.0): cluster 0\n", + "Example n. 461 = (18.0, 476.0): cluster 0\n", + "Example n. 462 = (18.0, 475.0): cluster 0\n", + "Example n. 463 = (18.0, 474.0): cluster 0\n", + "Example n. 464 = (18.0, 473.0): cluster 0\n", + "Example n. 465 = (18.0, 472.0): cluster 0\n", + "Example n. 466 = (18.0, 471.0): cluster 0\n", + "Example n. 467 = (18.0, 470.0): cluster 0\n", + "Example n. 468 = (18.0, 469.0): cluster 0\n", + "Example n. 469 = (18.0, 468.0): cluster 0\n", + "Example n. 470 = (18.0, 467.0): cluster 0\n", + "Example n. 471 = (18.0, 466.0): cluster 0\n", + "Example n. 472 = (18.0, 465.0): cluster 0\n", + "Example n. 473 = (18.0, 464.0): cluster 0\n", + "Example n. 474 = (18.0, 463.0): cluster 0\n", + "Example n. 475 = (18.0, 462.0): cluster 0\n", + "Example n. 476 = (18.0, 461.0): cluster 0\n", + "Example n. 477 = (18.0, 460.0): cluster 0\n", + "Example n. 478 = (18.0, 459.0): cluster 0\n", + "Example n. 479 = (18.0, 458.0): cluster 0\n", + "Example n. 480 = (18.0, 457.0): cluster 0\n", + "Example n. 481 = (18.0, 456.0): cluster 0\n", + "Example n. 482 = (19.0, 494.0): cluster 0\n", + "Example n. 483 = (19.0, 493.0): cluster 0\n", + "Example n. 484 = (19.0, 492.0): cluster 0\n", + "Example n. 485 = (19.0, 491.0): cluster 0\n", + "Example n. 486 = (19.0, 490.0): cluster 0\n", + "Example n. 487 = (19.0, 489.0): cluster 0\n", + "Example n. 488 = (19.0, 488.0): cluster 0\n", + "Example n. 489 = (19.0, 487.0): cluster 0\n", + "Example n. 490 = (19.0, 486.0): cluster 0\n", + "Example n. 491 = (19.0, 485.0): cluster 0\n", + "Example n. 492 = (19.0, 484.0): cluster 0\n", + "Example n. 493 = (19.0, 483.0): cluster 0\n", + "Example n. 494 = (19.0, 482.0): cluster 0\n", + "Example n. 495 = (19.0, 481.0): cluster 0\n", + "Example n. 496 = (19.0, 480.0): cluster 0\n", + "Example n. 497 = (19.0, 479.0): cluster 0\n", + "Example n. 498 = (19.0, 478.0): cluster 0\n", + "Example n. 499 = (19.0, 477.0): cluster 0\n", + "Example n. 500 = (19.0, 476.0): cluster 0\n", + "Example n. 501 = (19.0, 475.0): cluster 0\n", + "Example n. 502 = (19.0, 474.0): cluster 0\n", + "Example n. 503 = (19.0, 473.0): cluster 0\n", + "Example n. 504 = (19.0, 472.0): cluster 0\n", + "Example n. 505 = (19.0, 471.0): cluster 0\n", + "Example n. 506 = (19.0, 470.0): cluster 0\n", + "Example n. 507 = (19.0, 469.0): cluster 0\n", + "Example n. 508 = (19.0, 468.0): cluster 0\n", + "Example n. 509 = (19.0, 467.0): cluster 0\n", + "Example n. 510 = (19.0, 466.0): cluster 0\n", + "Example n. 511 = (19.0, 465.0): cluster 0\n", + "Example n. 512 = (19.0, 464.0): cluster 0\n", + "Example n. 513 = (19.0, 463.0): cluster 0\n", + "Example n. 514 = (19.0, 462.0): cluster 0\n", + "Example n. 515 = (19.0, 461.0): cluster 0\n", + "Example n. 516 = (19.0, 460.0): cluster 0\n", + "Example n. 517 = (19.0, 459.0): cluster 0\n", + "Example n. 518 = (19.0, 458.0): cluster 0\n", + "Example n. 519 = (19.0, 457.0): cluster 0\n", + "Example n. 520 = (19.0, 456.0): cluster 0\n", + "Example n. 521 = (20.0, 494.0): cluster 0\n", + "Example n. 522 = (20.0, 493.0): cluster 0\n", + "Example n. 523 = (20.0, 492.0): cluster 0\n", + "Example n. 524 = (20.0, 491.0): cluster 0\n", + "Example n. 525 = (20.0, 490.0): cluster 0\n", + "Example n. 526 = (20.0, 489.0): cluster 0\n", + "Example n. 527 = (20.0, 488.0): cluster 0\n", + "Example n. 528 = (20.0, 487.0): cluster 0\n", + "Example n. 529 = (20.0, 486.0): cluster 0\n", + "Example n. 530 = (20.0, 485.0): cluster 0\n", + "Example n. 531 = (20.0, 484.0): cluster 0\n", + "Example n. 532 = (20.0, 483.0): cluster 0\n", + "Example n. 533 = (20.0, 482.0): cluster 0\n", + "Example n. 534 = (20.0, 481.0): cluster 0\n", + "Example n. 535 = (20.0, 480.0): cluster 0\n", + "Example n. 536 = (20.0, 479.0): cluster 0\n", + "Example n. 537 = (20.0, 478.0): cluster 0\n", + "Example n. 538 = (20.0, 477.0): cluster 0\n", + "Example n. 539 = (20.0, 476.0): cluster 0\n", + "Example n. 540 = (20.0, 475.0): cluster 0\n", + "Example n. 541 = (20.0, 474.0): cluster 0\n", + "Example n. 542 = (20.0, 473.0): cluster 0\n", + "Example n. 543 = (20.0, 472.0): cluster 0\n", + "Example n. 544 = (20.0, 471.0): cluster 0\n", + "Example n. 545 = (21.0, 496.0): cluster 0\n", + "Example n. 546 = (21.0, 494.0): cluster 0\n", + "Example n. 547 = (21.0, 493.0): cluster 0\n", + "Example n. 548 = (21.0, 492.0): cluster 0\n", + "Example n. 549 = (21.0, 491.0): cluster 0\n", + "Example n. 550 = (21.0, 490.0): cluster 0\n", + "Example n. 551 = (21.0, 489.0): cluster 0\n", + "Example n. 552 = (21.0, 488.0): cluster 0\n", + "Example n. 553 = (21.0, 487.0): cluster 0\n", + "Example n. 554 = (21.0, 486.0): cluster 0\n", + "Example n. 555 = (21.0, 485.0): cluster 0\n", + "Example n. 556 = (21.0, 484.0): cluster 0\n", + "Example n. 557 = (21.0, 483.0): cluster 0\n", + "Example n. 558 = (21.0, 482.0): cluster 0\n", + "Example n. 559 = (21.0, 481.0): cluster 0\n", + "Example n. 560 = (21.0, 480.0): cluster 0\n", + "Example n. 561 = (21.0, 479.0): cluster 0\n", + "Example n. 562 = (21.0, 478.0): cluster 0\n", + "Example n. 563 = (21.0, 477.0): cluster 0\n", + "Example n. 564 = (21.0, 476.0): cluster 0\n", + "Example n. 565 = (21.0, 475.0): cluster 0\n", + "Example n. 566 = (21.0, 474.0): cluster 0\n", + "Example n. 567 = (21.0, 473.0): cluster 0\n", + "Example n. 568 = (21.0, 472.0): cluster 0\n", + "Example n. 569 = (21.0, 471.0): cluster 0\n", + "Example n. 570 = (22.0, 494.0): cluster 0\n", + "Example n. 571 = (22.0, 493.0): cluster 0\n", + "Example n. 572 = (22.0, 492.0): cluster 0\n", + "Example n. 573 = (22.0, 491.0): cluster 0\n", + "Example n. 574 = (22.0, 490.0): cluster 0\n", + "Example n. 575 = (22.0, 489.0): cluster 0\n", + "Example n. 576 = (22.0, 488.0): cluster 0\n", + "Example n. 577 = (22.0, 487.0): cluster 0\n", + "Example n. 578 = (22.0, 486.0): cluster 0\n", + "Example n. 579 = (22.0, 485.0): cluster 0\n", + "Example n. 580 = (22.0, 484.0): cluster 0\n", + "Example n. 581 = (22.0, 483.0): cluster 0\n", + "Example n. 582 = (22.0, 482.0): cluster 0\n", + "Example n. 583 = (22.0, 481.0): cluster 0\n", + "Example n. 584 = (22.0, 480.0): cluster 0\n", + "Example n. 585 = (22.0, 479.0): cluster 0\n", + "Example n. 586 = (22.0, 478.0): cluster 0\n", + "Example n. 587 = (22.0, 477.0): cluster 0\n", + "Example n. 588 = (22.0, 476.0): cluster 0\n", + "Example n. 589 = (22.0, 475.0): cluster 0\n", + "Example n. 590 = (22.0, 474.0): cluster 0\n", + "Example n. 591 = (22.0, 473.0): cluster 0\n", + "Example n. 592 = (22.0, 472.0): cluster 0\n", + "Example n. 593 = (22.0, 471.0): cluster 0\n", + "Example n. 594 = (23.0, 497.0): cluster 0\n", + "Example n. 595 = (23.0, 496.0): cluster 0\n", + "Example n. 596 = (23.0, 495.0): cluster 0\n", + "Example n. 597 = (23.0, 494.0): cluster 0\n", + "Example n. 598 = (23.0, 493.0): cluster 0\n", + "Example n. 599 = (23.0, 492.0): cluster 0\n", + "Example n. 600 = (23.0, 491.0): cluster 0\n", + "Example n. 601 = (23.0, 490.0): cluster 0\n", + "Example n. 602 = (23.0, 489.0): cluster 0\n", + "Example n. 603 = (23.0, 488.0): cluster 0\n", + "Example n. 604 = (23.0, 487.0): cluster 0\n", + "Example n. 605 = (23.0, 486.0): cluster 0\n", + "Example n. 606 = (23.0, 485.0): cluster 0\n", + "Example n. 607 = (23.0, 484.0): cluster 0\n", + "Example n. 608 = (23.0, 483.0): cluster 0\n", + "Example n. 609 = (23.0, 482.0): cluster 0\n", + "Example n. 610 = (23.0, 481.0): cluster 0\n", + "Example n. 611 = (23.0, 480.0): cluster 0\n", + "Example n. 612 = (23.0, 479.0): cluster 0\n", + "Example n. 613 = (23.0, 478.0): cluster 0\n", + "Example n. 614 = (23.0, 477.0): cluster 0\n", + "Example n. 615 = (23.0, 476.0): cluster 0\n", + "Example n. 616 = (23.0, 475.0): cluster 0\n", + "Example n. 617 = (23.0, 474.0): cluster 0\n", + "Example n. 618 = (23.0, 473.0): cluster 0\n", + "Example n. 619 = (23.0, 472.0): cluster 0\n", + "Example n. 620 = (23.0, 471.0): cluster 0\n", + "Example n. 621 = (24.0, 498.0): cluster 0\n", + "Example n. 622 = (24.0, 497.0): cluster 0\n", + "Example n. 623 = (24.0, 496.0): cluster 0\n", + "Example n. 624 = (24.0, 495.0): cluster 0\n", + "Example n. 625 = (24.0, 494.0): cluster 0\n", + "Example n. 626 = (24.0, 493.0): cluster 0\n", + "Example n. 627 = (24.0, 492.0): cluster 0\n", + "Example n. 628 = (24.0, 491.0): cluster 0\n", + "Example n. 629 = (24.0, 490.0): cluster 0\n", + "Example n. 630 = (24.0, 489.0): cluster 0\n", + "Example n. 631 = (24.0, 488.0): cluster 0\n", + "Example n. 632 = (24.0, 487.0): cluster 0\n", + "Example n. 633 = (24.0, 486.0): cluster 0\n", + "Example n. 634 = (24.0, 485.0): cluster 0\n", + "Example n. 635 = (24.0, 484.0): cluster 0\n", + "Example n. 636 = (24.0, 483.0): cluster 0\n", + "Example n. 637 = (24.0, 482.0): cluster 0\n", + "Example n. 638 = (24.0, 481.0): cluster 0\n", + "Example n. 639 = (24.0, 480.0): cluster 0\n", + "Example n. 640 = (24.0, 479.0): cluster 0\n", + "Example n. 641 = (25.0, 498.0): cluster 0\n", + "Example n. 642 = (25.0, 497.0): cluster 0\n", + "Example n. 643 = (25.0, 496.0): cluster 0\n", + "Example n. 644 = (25.0, 495.0): cluster 0\n", + "Example n. 645 = (25.0, 494.0): cluster 0\n", + "Example n. 646 = (25.0, 493.0): cluster 0\n", + "Example n. 647 = (25.0, 492.0): cluster 0\n", + "Example n. 648 = (25.0, 491.0): cluster 0\n", + "Example n. 649 = (25.0, 490.0): cluster 0\n", + "Example n. 650 = (25.0, 489.0): cluster 0\n", + "Example n. 651 = (25.0, 488.0): cluster 0\n", + "Example n. 652 = (25.0, 487.0): cluster 0\n", + "Example n. 653 = (25.0, 486.0): cluster 0\n", + "Example n. 654 = (25.0, 485.0): cluster 0\n", + "Example n. 655 = (25.0, 484.0): cluster 0\n", + "Example n. 656 = (25.0, 483.0): cluster 0\n", + "Example n. 657 = (25.0, 482.0): cluster 0\n", + "Example n. 658 = (25.0, 481.0): cluster 0\n", + "Example n. 659 = (25.0, 480.0): cluster 0\n", + "Example n. 660 = (25.0, 479.0): cluster 0\n", + "Example n. 661 = (26.0, 498.0): cluster 0\n", + "Example n. 662 = (26.0, 497.0): cluster 0\n", + "Example n. 663 = (26.0, 496.0): cluster 0\n", + "Example n. 664 = (26.0, 495.0): cluster 0\n", + "Example n. 665 = (26.0, 494.0): cluster 0\n", + "Example n. 666 = (26.0, 493.0): cluster 0\n", + "Example n. 667 = (26.0, 492.0): cluster 0\n", + "Example n. 668 = (26.0, 491.0): cluster 0\n", + "Example n. 669 = (26.0, 490.0): cluster 0\n", + "Example n. 670 = (26.0, 489.0): cluster 0\n", + "Example n. 671 = (26.0, 488.0): cluster 0\n", + "Example n. 672 = (26.0, 487.0): cluster 0\n", + "Example n. 673 = (26.0, 486.0): cluster 0\n", + "Example n. 674 = (26.0, 485.0): cluster 0\n", + "Example n. 675 = (26.0, 484.0): cluster 0\n", + "Example n. 676 = (26.0, 483.0): cluster 0\n", + "Example n. 677 = (26.0, 482.0): cluster 0\n", + "Example n. 678 = (26.0, 481.0): cluster 0\n", + "Example n. 679 = (26.0, 480.0): cluster 0\n", + "Example n. 680 = (26.0, 479.0): cluster 0\n", + "Example n. 681 = (27.0, 498.0): cluster 0\n", + "Example n. 682 = (27.0, 497.0): cluster 0\n", + "Example n. 683 = (27.0, 496.0): cluster 0\n", + "Example n. 684 = (27.0, 495.0): cluster 0\n", + "Example n. 685 = (27.0, 494.0): cluster 0\n", + "Example n. 686 = (27.0, 493.0): cluster 0\n", + "Example n. 687 = (27.0, 492.0): cluster 0\n", + "Example n. 688 = (27.0, 491.0): cluster 0\n", + "Example n. 689 = (27.0, 490.0): cluster 0\n", + "Example n. 690 = (27.0, 489.0): cluster 0\n", + "Example n. 691 = (27.0, 488.0): cluster 0\n", + "Example n. 692 = (27.0, 487.0): cluster 0\n", + "Example n. 693 = (27.0, 486.0): cluster 0\n", + "Example n. 694 = (27.0, 485.0): cluster 0\n", + "Example n. 695 = (27.0, 484.0): cluster 0\n", + "Example n. 696 = (27.0, 483.0): cluster 0\n", + "Example n. 697 = (27.0, 482.0): cluster 0\n", + "Example n. 698 = (27.0, 481.0): cluster 0\n", + "Example n. 699 = (27.0, 480.0): cluster 0\n", + "Example n. 700 = (27.0, 479.0): cluster 0\n", + "Example n. 701 = (28.0, 502.0): cluster 0\n", + "Example n. 702 = (28.0, 501.0): cluster 0\n", + "Example n. 703 = (28.0, 500.0): cluster 0\n", + "Example n. 704 = (28.0, 499.0): cluster 0\n", + "Example n. 705 = (28.0, 498.0): cluster 0\n", + "Example n. 706 = (28.0, 497.0): cluster 0\n", + "Example n. 707 = (28.0, 496.0): cluster 0\n", + "Example n. 708 = (28.0, 495.0): cluster 0\n", + "Example n. 709 = (28.0, 494.0): cluster 0\n", + "Example n. 710 = (28.0, 493.0): cluster 0\n", + "Example n. 711 = (28.0, 492.0): cluster 0\n", + "Example n. 712 = (28.0, 491.0): cluster 0\n", + "Example n. 713 = (28.0, 490.0): cluster 0\n", + "Example n. 714 = (28.0, 489.0): cluster 0\n", + "Example n. 715 = (28.0, 488.0): cluster 0\n", + "Example n. 716 = (28.0, 487.0): cluster 0\n", + "Example n. 717 = (28.0, 486.0): cluster 0\n", + "Example n. 718 = (28.0, 485.0): cluster 0\n", + "Example n. 719 = (28.0, 484.0): cluster 0\n", + "Example n. 720 = (28.0, 483.0): cluster 0\n", + "Example n. 721 = (28.0, 475.0): cluster 0\n", + "Example n. 722 = (28.0, 474.0): cluster 0\n", + "Example n. 723 = (28.0, 473.0): cluster 0\n", + "Example n. 724 = (28.0, 472.0): cluster 0\n", + "Example n. 725 = (28.0, 471.0): cluster 0\n", + "Example n. 726 = (28.0, 470.0): cluster 0\n", + "Example n. 727 = (28.0, 469.0): cluster 0\n", + "Example n. 728 = (28.0, 468.0): cluster 0\n", + "Example n. 729 = (28.0, 467.0): cluster 0\n", + "Example n. 730 = (28.0, 466.0): cluster 0\n", + "Example n. 731 = (28.0, 465.0): cluster 0\n", + "Example n. 732 = (28.0, 464.0): cluster 0\n", + "Example n. 733 = (28.0, 463.0): cluster 0\n", + "Example n. 734 = (28.0, 462.0): cluster 0\n", + "Example n. 735 = (28.0, 461.0): cluster 0\n", + "Example n. 736 = (28.0, 460.0): cluster 0\n", + "Example n. 737 = (28.0, 459.0): cluster 0\n", + "Example n. 738 = (28.0, 458.0): cluster 0\n", + "Example n. 739 = (28.0, 457.0): cluster 0\n", + "Example n. 740 = (28.0, 456.0): cluster 0\n", + "Example n. 741 = (28.0, 455.0): cluster 0\n", + "Example n. 742 = (28.0, 454.0): cluster 0\n", + "Example n. 743 = (28.0, 453.0): cluster 0\n", + "Example n. 744 = (28.0, 452.0): cluster 0\n", + "Example n. 745 = (29.0, 502.0): cluster 0\n", + "Example n. 746 = (29.0, 501.0): cluster 0\n", + "Example n. 747 = (29.0, 500.0): cluster 0\n", + "Example n. 748 = (29.0, 499.0): cluster 0\n", + "Example n. 749 = (29.0, 498.0): cluster 0\n", + "Example n. 750 = (29.0, 497.0): cluster 0\n", + "Example n. 751 = (29.0, 496.0): cluster 0\n", + "Example n. 752 = (29.0, 495.0): cluster 0\n", + "Example n. 753 = (29.0, 494.0): cluster 0\n", + "Example n. 754 = (29.0, 493.0): cluster 0\n", + "Example n. 755 = (29.0, 492.0): cluster 0\n", + "Example n. 756 = (29.0, 491.0): cluster 0\n", + "Example n. 757 = (29.0, 490.0): cluster 0\n", + "Example n. 758 = (29.0, 489.0): cluster 0\n", + "Example n. 759 = (29.0, 488.0): cluster 0\n", + "Example n. 760 = (29.0, 487.0): cluster 0\n", + "Example n. 761 = (29.0, 486.0): cluster 0\n", + "Example n. 762 = (29.0, 485.0): cluster 0\n", + "Example n. 763 = (29.0, 484.0): cluster 0\n", + "Example n. 764 = (29.0, 483.0): cluster 0\n", + "Example n. 765 = (29.0, 475.0): cluster 0\n", + "Example n. 766 = (29.0, 474.0): cluster 0\n", + "Example n. 767 = (29.0, 473.0): cluster 0\n", + "Example n. 768 = (29.0, 472.0): cluster 0\n", + "Example n. 769 = (29.0, 471.0): cluster 0\n", + "Example n. 770 = (29.0, 470.0): cluster 0\n", + "Example n. 771 = (29.0, 469.0): cluster 0\n", + "Example n. 772 = (29.0, 468.0): cluster 0\n", + "Example n. 773 = (29.0, 467.0): cluster 0\n", + "Example n. 774 = (29.0, 466.0): cluster 0\n", + "Example n. 775 = (29.0, 465.0): cluster 0\n", + "Example n. 776 = (29.0, 464.0): cluster 0\n", + "Example n. 777 = (29.0, 463.0): cluster 0\n", + "Example n. 778 = (29.0, 462.0): cluster 0\n", + "Example n. 779 = (29.0, 461.0): cluster 0\n", + "Example n. 780 = (29.0, 460.0): cluster 0\n", + "Example n. 781 = (29.0, 459.0): cluster 0\n", + "Example n. 782 = (29.0, 458.0): cluster 0\n", + "Example n. 783 = (29.0, 457.0): cluster 0\n", + "Example n. 784 = (29.0, 456.0): cluster 0\n", + "Example n. 785 = (29.0, 455.0): cluster 0\n", + "Example n. 786 = (29.0, 454.0): cluster 0\n", + "Example n. 787 = (29.0, 453.0): cluster 0\n", + "Example n. 788 = (29.0, 452.0): cluster 0\n", + "Example n. 789 = (30.0, 502.0): cluster 0\n", + "Example n. 790 = (30.0, 501.0): cluster 0\n", + "Example n. 791 = (30.0, 500.0): cluster 0\n", + "Example n. 792 = (30.0, 499.0): cluster 0\n", + "Example n. 793 = (30.0, 498.0): cluster 0\n", + "Example n. 794 = (30.0, 497.0): cluster 0\n", + "Example n. 795 = (30.0, 496.0): cluster 0\n", + "Example n. 796 = (30.0, 495.0): cluster 0\n", + "Example n. 797 = (30.0, 494.0): cluster 0\n", + "Example n. 798 = (30.0, 493.0): cluster 0\n", + "Example n. 799 = (30.0, 492.0): cluster 0\n", + "Example n. 800 = (30.0, 491.0): cluster 0\n", + "Example n. 801 = (30.0, 490.0): cluster 0\n", + "Example n. 802 = (30.0, 489.0): cluster 0\n", + "Example n. 803 = (30.0, 488.0): cluster 0\n", + "Example n. 804 = (30.0, 487.0): cluster 0\n", + "Example n. 805 = (30.0, 486.0): cluster 0\n", + "Example n. 806 = (30.0, 485.0): cluster 0\n", + "Example n. 807 = (30.0, 484.0): cluster 0\n", + "Example n. 808 = (30.0, 483.0): cluster 0\n", + "Example n. 809 = (30.0, 475.0): cluster 0\n", + "Example n. 810 = (30.0, 474.0): cluster 0\n", + "Example n. 811 = (30.0, 473.0): cluster 0\n", + "Example n. 812 = (30.0, 472.0): cluster 0\n", + "Example n. 813 = (30.0, 471.0): cluster 0\n", + "Example n. 814 = (30.0, 470.0): cluster 0\n", + "Example n. 815 = (30.0, 469.0): cluster 0\n", + "Example n. 816 = (30.0, 468.0): cluster 0\n", + "Example n. 817 = (30.0, 467.0): cluster 0\n", + "Example n. 818 = (30.0, 466.0): cluster 0\n", + "Example n. 819 = (30.0, 465.0): cluster 0\n", + "Example n. 820 = (30.0, 464.0): cluster 0\n", + "Example n. 821 = (30.0, 463.0): cluster 0\n", + "Example n. 822 = (30.0, 462.0): cluster 0\n", + "Example n. 823 = (30.0, 461.0): cluster 0\n", + "Example n. 824 = (30.0, 460.0): cluster 0\n", + "Example n. 825 = (30.0, 459.0): cluster 0\n", + "Example n. 826 = (30.0, 458.0): cluster 0\n", + "Example n. 827 = (30.0, 457.0): cluster 0\n", + "Example n. 828 = (30.0, 456.0): cluster 0\n", + "Example n. 829 = (30.0, 455.0): cluster 0\n", + "Example n. 830 = (30.0, 454.0): cluster 0\n", + "Example n. 831 = (30.0, 453.0): cluster 0\n", + "Example n. 832 = (30.0, 452.0): cluster 0\n", + "Example n. 833 = (31.0, 502.0): cluster 0\n", + "Example n. 834 = (31.0, 501.0): cluster 0\n", + "Example n. 835 = (31.0, 500.0): cluster 0\n", + "Example n. 836 = (31.0, 499.0): cluster 0\n", + "Example n. 837 = (31.0, 498.0): cluster 0\n", + "Example n. 838 = (31.0, 497.0): cluster 0\n", + "Example n. 839 = (31.0, 496.0): cluster 0\n", + "Example n. 840 = (31.0, 495.0): cluster 0\n", + "Example n. 841 = (31.0, 494.0): cluster 0\n", + "Example n. 842 = (31.0, 493.0): cluster 0\n", + "Example n. 843 = (31.0, 492.0): cluster 0\n", + "Example n. 844 = (31.0, 491.0): cluster 0\n", + "Example n. 845 = (31.0, 490.0): cluster 0\n", + "Example n. 846 = (31.0, 489.0): cluster 0\n", + "Example n. 847 = (31.0, 488.0): cluster 0\n", + "Example n. 848 = (31.0, 487.0): cluster 0\n", + "Example n. 849 = (31.0, 486.0): cluster 0\n", + "Example n. 850 = (31.0, 485.0): cluster 0\n", + "Example n. 851 = (31.0, 484.0): cluster 0\n", + "Example n. 852 = (31.0, 483.0): cluster 0\n", + "Example n. 853 = (31.0, 475.0): cluster 0\n", + "Example n. 854 = (31.0, 474.0): cluster 0\n", + "Example n. 855 = (31.0, 473.0): cluster 0\n", + "Example n. 856 = (31.0, 472.0): cluster 0\n", + "Example n. 857 = (31.0, 471.0): cluster 0\n", + "Example n. 858 = (31.0, 470.0): cluster 0\n", + "Example n. 859 = (31.0, 469.0): cluster 0\n", + "Example n. 860 = (31.0, 468.0): cluster 0\n", + "Example n. 861 = (31.0, 467.0): cluster 0\n", + "Example n. 862 = (31.0, 466.0): cluster 0\n", + "Example n. 863 = (31.0, 465.0): cluster 0\n", + "Example n. 864 = (31.0, 464.0): cluster 0\n", + "Example n. 865 = (31.0, 463.0): cluster 0\n", + "Example n. 866 = (31.0, 462.0): cluster 0\n", + "Example n. 867 = (31.0, 461.0): cluster 0\n", + "Example n. 868 = (31.0, 460.0): cluster 0\n", + "Example n. 869 = (31.0, 459.0): cluster 0\n", + "Example n. 870 = (31.0, 458.0): cluster 0\n", + "Example n. 871 = (31.0, 457.0): cluster 0\n", + "Example n. 872 = (31.0, 456.0): cluster 0\n", + "Example n. 873 = (31.0, 455.0): cluster 0\n", + "Example n. 874 = (31.0, 454.0): cluster 0\n", + "Example n. 875 = (31.0, 453.0): cluster 0\n", + "Example n. 876 = (31.0, 452.0): cluster 0\n", + "Example n. 877 = (32.0, 502.0): cluster 0\n", + "Example n. 878 = (32.0, 501.0): cluster 0\n", + "Example n. 879 = (32.0, 500.0): cluster 0\n", + "Example n. 880 = (32.0, 499.0): cluster 0\n", + "Example n. 881 = (32.0, 498.0): cluster 0\n", + "Example n. 882 = (32.0, 497.0): cluster 0\n", + "Example n. 883 = (32.0, 495.0): cluster 0\n", + "Example n. 884 = (32.0, 494.0): cluster 0\n", + "Example n. 885 = (32.0, 493.0): cluster 0\n", + "Example n. 886 = (32.0, 492.0): cluster 0\n", + "Example n. 887 = (32.0, 491.0): cluster 0\n", + "Example n. 888 = (32.0, 490.0): cluster 0\n", + "Example n. 889 = (32.0, 489.0): cluster 0\n", + "Example n. 890 = (32.0, 488.0): cluster 0\n", + "Example n. 891 = (32.0, 487.0): cluster 0\n", + "Example n. 892 = (32.0, 479.0): cluster 0\n", + "Example n. 893 = (32.0, 478.0): cluster 0\n", + "Example n. 894 = (32.0, 477.0): cluster 0\n", + "Example n. 895 = (32.0, 476.0): cluster 0\n", + "Example n. 896 = (32.0, 475.0): cluster 0\n", + "Example n. 897 = (32.0, 474.0): cluster 0\n", + "Example n. 898 = (32.0, 473.0): cluster 0\n", + "Example n. 899 = (32.0, 472.0): cluster 0\n", + "Example n. 900 = (32.0, 471.0): cluster 0\n", + "Example n. 901 = (32.0, 470.0): cluster 0\n", + "Example n. 902 = (32.0, 469.0): cluster 0\n", + "Example n. 903 = (32.0, 468.0): cluster 0\n", + "Example n. 904 = (32.0, 467.0): cluster 0\n", + "Example n. 905 = (32.0, 466.0): cluster 0\n", + "Example n. 906 = (32.0, 465.0): cluster 0\n", + "Example n. 907 = (32.0, 464.0): cluster 0\n", + "Example n. 908 = (32.0, 463.0): cluster 0\n", + "Example n. 909 = (32.0, 462.0): cluster 0\n", + "Example n. 910 = (32.0, 461.0): cluster 0\n", + "Example n. 911 = (32.0, 460.0): cluster 0\n", + "Example n. 912 = (32.0, 459.0): cluster 0\n", + "Example n. 913 = (32.0, 458.0): cluster 0\n", + "Example n. 914 = (32.0, 457.0): cluster 0\n", + "Example n. 915 = (32.0, 456.0): cluster 0\n", + "Example n. 916 = (32.0, 455.0): cluster 0\n", + "Example n. 917 = (32.0, 454.0): cluster 0\n", + "Example n. 918 = (32.0, 453.0): cluster 0\n", + "Example n. 919 = (32.0, 452.0): cluster 0\n", + "Example n. 920 = (32.0, 451.0): cluster 0\n", + "Example n. 921 = (32.0, 450.0): cluster 0\n", + "Example n. 922 = (32.0, 449.0): cluster 0\n", + "Example n. 923 = (32.0, 448.0): cluster 0\n", + "Example n. 924 = (32.0, 421.0): cluster 3\n", + "Example n. 925 = (32.0, 420.0): cluster 3\n", + "Example n. 926 = (32.0, 419.0): cluster 3\n", + "Example n. 927 = (32.0, 418.0): cluster 3\n", + "Example n. 928 = (32.0, 417.0): cluster 3\n", + "Example n. 929 = (32.0, 416.0): cluster 3\n", + "Example n. 930 = (32.0, 415.0): cluster 3\n", + "Example n. 931 = (32.0, 414.0): cluster 3\n", + "Example n. 932 = (32.0, 413.0): cluster 3\n", + "Example n. 933 = (32.0, 412.0): cluster 3\n", + "Example n. 934 = (32.0, 411.0): cluster 3\n", + "Example n. 935 = (32.0, 410.0): cluster 3\n", + "Example n. 936 = (33.0, 502.0): cluster 0\n", + "Example n. 937 = (33.0, 501.0): cluster 0\n", + "Example n. 938 = (33.0, 500.0): cluster 0\n", + "Example n. 939 = (33.0, 499.0): cluster 0\n", + "Example n. 940 = (33.0, 498.0): cluster 0\n", + "Example n. 941 = (33.0, 494.0): cluster 0\n", + "Example n. 942 = (33.0, 493.0): cluster 0\n", + "Example n. 943 = (33.0, 492.0): cluster 0\n", + "Example n. 944 = (33.0, 491.0): cluster 0\n", + "Example n. 945 = (33.0, 490.0): cluster 0\n", + "Example n. 946 = (33.0, 489.0): cluster 0\n", + "Example n. 947 = (33.0, 488.0): cluster 0\n", + "Example n. 948 = (33.0, 487.0): cluster 0\n", + "Example n. 949 = (33.0, 479.0): cluster 0\n", + "Example n. 950 = (33.0, 478.0): cluster 0\n", + "Example n. 951 = (33.0, 477.0): cluster 0\n", + "Example n. 952 = (33.0, 476.0): cluster 0\n", + "Example n. 953 = (33.0, 475.0): cluster 0\n", + "Example n. 954 = (33.0, 474.0): cluster 0\n", + "Example n. 955 = (33.0, 473.0): cluster 0\n", + "Example n. 956 = (33.0, 472.0): cluster 0\n", + "Example n. 957 = (33.0, 471.0): cluster 0\n", + "Example n. 958 = (33.0, 470.0): cluster 0\n", + "Example n. 959 = (33.0, 469.0): cluster 0\n", + "Example n. 960 = (33.0, 468.0): cluster 0\n", + "Example n. 961 = (33.0, 467.0): cluster 0\n", + "Example n. 962 = (33.0, 466.0): cluster 0\n", + "Example n. 963 = (33.0, 465.0): cluster 0\n", + "Example n. 964 = (33.0, 464.0): cluster 0\n", + "Example n. 965 = (33.0, 463.0): cluster 0\n", + "Example n. 966 = (33.0, 462.0): cluster 0\n", + "Example n. 967 = (33.0, 461.0): cluster 0\n", + "Example n. 968 = (33.0, 460.0): cluster 0\n", + "Example n. 969 = (33.0, 459.0): cluster 0\n", + "Example n. 970 = (33.0, 458.0): cluster 0\n", + "Example n. 971 = (33.0, 457.0): cluster 0\n", + "Example n. 972 = (33.0, 456.0): cluster 0\n", + "Example n. 973 = (33.0, 455.0): cluster 0\n", + "Example n. 974 = (33.0, 454.0): cluster 0\n", + "Example n. 975 = (33.0, 453.0): cluster 0\n", + "Example n. 976 = (33.0, 452.0): cluster 0\n", + "Example n. 977 = (33.0, 451.0): cluster 0\n", + "Example n. 978 = (33.0, 450.0): cluster 0\n", + "Example n. 979 = (33.0, 449.0): cluster 0\n", + "Example n. 980 = (33.0, 448.0): cluster 0\n", + "Example n. 981 = (33.0, 421.0): cluster 3\n", + "Example n. 982 = (33.0, 420.0): cluster 3\n", + "Example n. 983 = (33.0, 419.0): cluster 3\n", + "Example n. 984 = (33.0, 418.0): cluster 3\n", + "Example n. 985 = (33.0, 417.0): cluster 3\n", + "Example n. 986 = (33.0, 416.0): cluster 3\n", + "Example n. 987 = (33.0, 415.0): cluster 3\n", + "Example n. 988 = (33.0, 414.0): cluster 3\n", + "Example n. 989 = (33.0, 413.0): cluster 3\n", + "Example n. 990 = (33.0, 412.0): cluster 3\n", + "Example n. 991 = (33.0, 411.0): cluster 3\n", + "Example n. 992 = (33.0, 410.0): cluster 3\n", + "Example n. 993 = (34.0, 502.0): cluster 0\n", + "Example n. 994 = (34.0, 501.0): cluster 0\n", + "Example n. 995 = (34.0, 500.0): cluster 0\n", + "Example n. 996 = (34.0, 499.0): cluster 0\n", + "Example n. 997 = (34.0, 498.0): cluster 0\n", + "Example n. 998 = (34.0, 494.0): cluster 0\n", + "Example n. 999 = (34.0, 493.0): cluster 0\n", + "Example n. 1000 = (34.0, 492.0): cluster 0\n", + "Example n. 1001 = (34.0, 491.0): cluster 0\n", + "Example n. 1002 = (34.0, 490.0): cluster 0\n", + "Example n. 1003 = (34.0, 489.0): cluster 0\n", + "Example n. 1004 = (34.0, 488.0): cluster 0\n", + "Example n. 1005 = (34.0, 487.0): cluster 0\n", + "Example n. 1006 = (34.0, 479.0): cluster 0\n", + "Example n. 1007 = (34.0, 478.0): cluster 0\n", + "Example n. 1008 = (34.0, 477.0): cluster 0\n", + "Example n. 1009 = (34.0, 476.0): cluster 0\n", + "Example n. 1010 = (34.0, 475.0): cluster 0\n", + "Example n. 1011 = (34.0, 474.0): cluster 0\n", + "Example n. 1012 = (34.0, 473.0): cluster 0\n", + "Example n. 1013 = (34.0, 472.0): cluster 0\n", + "Example n. 1014 = (34.0, 471.0): cluster 0\n", + "Example n. 1015 = (34.0, 470.0): cluster 0\n", + "Example n. 1016 = (34.0, 469.0): cluster 0\n", + "Example n. 1017 = (34.0, 468.0): cluster 0\n", + "Example n. 1018 = (34.0, 467.0): cluster 0\n", + "Example n. 1019 = (34.0, 466.0): cluster 0\n", + "Example n. 1020 = (34.0, 465.0): cluster 0\n", + "Example n. 1021 = (34.0, 464.0): cluster 0\n", + "Example n. 1022 = (34.0, 463.0): cluster 0\n", + "Example n. 1023 = (34.0, 462.0): cluster 0\n", + "Example n. 1024 = (34.0, 461.0): cluster 0\n", + "Example n. 1025 = (34.0, 460.0): cluster 0\n", + "Example n. 1026 = (34.0, 459.0): cluster 0\n", + "Example n. 1027 = (34.0, 458.0): cluster 0\n", + "Example n. 1028 = (34.0, 457.0): cluster 0\n", + "Example n. 1029 = (34.0, 456.0): cluster 0\n", + "Example n. 1030 = (34.0, 455.0): cluster 0\n", + "Example n. 1031 = (34.0, 454.0): cluster 0\n", + "Example n. 1032 = (34.0, 453.0): cluster 0\n", + "Example n. 1033 = (34.0, 452.0): cluster 0\n", + "Example n. 1034 = (34.0, 451.0): cluster 0\n", + "Example n. 1035 = (34.0, 450.0): cluster 0\n", + "Example n. 1036 = (34.0, 449.0): cluster 0\n", + "Example n. 1037 = (34.0, 448.0): cluster 0\n", + "Example n. 1038 = (34.0, 421.0): cluster 3\n", + "Example n. 1039 = (34.0, 420.0): cluster 3\n", + "Example n. 1040 = (34.0, 419.0): cluster 3\n", + "Example n. 1041 = (34.0, 418.0): cluster 3\n", + "Example n. 1042 = (34.0, 417.0): cluster 3\n", + "Example n. 1043 = (34.0, 416.0): cluster 3\n", + "Example n. 1044 = (34.0, 415.0): cluster 3\n", + "Example n. 1045 = (34.0, 414.0): cluster 3\n", + "Example n. 1046 = (34.0, 413.0): cluster 3\n", + "Example n. 1047 = (34.0, 412.0): cluster 3\n", + "Example n. 1048 = (34.0, 411.0): cluster 3\n", + "Example n. 1049 = (34.0, 410.0): cluster 3\n", + "Example n. 1050 = (35.0, 502.0): cluster 0\n", + "Example n. 1051 = (35.0, 501.0): cluster 0\n", + "Example n. 1052 = (35.0, 500.0): cluster 0\n", + "Example n. 1053 = (35.0, 499.0): cluster 0\n", + "Example n. 1054 = (35.0, 498.0): cluster 0\n", + "Example n. 1055 = (35.0, 497.0): cluster 0\n", + "Example n. 1056 = (35.0, 496.0): cluster 0\n", + "Example n. 1057 = (35.0, 495.0): cluster 0\n", + "Example n. 1058 = (35.0, 494.0): cluster 0\n", + "Example n. 1059 = (35.0, 493.0): cluster 0\n", + "Example n. 1060 = (35.0, 492.0): cluster 0\n", + "Example n. 1061 = (35.0, 491.0): cluster 0\n", + "Example n. 1062 = (35.0, 490.0): cluster 0\n", + "Example n. 1063 = (35.0, 489.0): cluster 0\n", + "Example n. 1064 = (35.0, 488.0): cluster 0\n", + "Example n. 1065 = (35.0, 487.0): cluster 0\n", + "Example n. 1066 = (35.0, 479.0): cluster 0\n", + "Example n. 1067 = (35.0, 478.0): cluster 0\n", + "Example n. 1068 = (35.0, 477.0): cluster 0\n", + "Example n. 1069 = (35.0, 476.0): cluster 0\n", + "Example n. 1070 = (35.0, 475.0): cluster 0\n", + "Example n. 1071 = (35.0, 474.0): cluster 0\n", + "Example n. 1072 = (35.0, 473.0): cluster 0\n", + "Example n. 1073 = (35.0, 472.0): cluster 0\n", + "Example n. 1074 = (35.0, 471.0): cluster 0\n", + "Example n. 1075 = (35.0, 470.0): cluster 0\n", + "Example n. 1076 = (35.0, 469.0): cluster 0\n", + "Example n. 1077 = (35.0, 468.0): cluster 0\n", + "Example n. 1078 = (35.0, 467.0): cluster 0\n", + "Example n. 1079 = (35.0, 466.0): cluster 0\n", + "Example n. 1080 = (35.0, 465.0): cluster 0\n", + "Example n. 1081 = (35.0, 464.0): cluster 0\n", + "Example n. 1082 = (35.0, 463.0): cluster 0\n", + "Example n. 1083 = (35.0, 462.0): cluster 0\n", + "Example n. 1084 = (35.0, 461.0): cluster 0\n", + "Example n. 1085 = (35.0, 460.0): cluster 0\n", + "Example n. 1086 = (35.0, 459.0): cluster 0\n", + "Example n. 1087 = (35.0, 458.0): cluster 0\n", + "Example n. 1088 = (35.0, 457.0): cluster 0\n", + "Example n. 1089 = (35.0, 456.0): cluster 0\n", + "Example n. 1090 = (35.0, 455.0): cluster 0\n", + "Example n. 1091 = (35.0, 454.0): cluster 0\n", + "Example n. 1092 = (35.0, 453.0): cluster 0\n", + "Example n. 1093 = (35.0, 452.0): cluster 0\n", + "Example n. 1094 = (35.0, 451.0): cluster 0\n", + "Example n. 1095 = (35.0, 450.0): cluster 0\n", + "Example n. 1096 = (35.0, 449.0): cluster 1\n", + "Example n. 1097 = (35.0, 448.0): cluster 1\n", + "Example n. 1098 = (35.0, 425.0): cluster 3\n", + "Example n. 1099 = (35.0, 424.0): cluster 3\n", + "Example n. 1100 = (35.0, 423.0): cluster 3\n", + "Example n. 1101 = (35.0, 422.0): cluster 3\n", + "Example n. 1102 = (35.0, 421.0): cluster 3\n", + "Example n. 1103 = (35.0, 420.0): cluster 3\n", + "Example n. 1104 = (35.0, 419.0): cluster 3\n", + "Example n. 1105 = (35.0, 418.0): cluster 3\n", + "Example n. 1106 = (35.0, 417.0): cluster 3\n", + "Example n. 1107 = (35.0, 416.0): cluster 3\n", + "Example n. 1108 = (35.0, 415.0): cluster 3\n", + "Example n. 1109 = (35.0, 414.0): cluster 3\n", + "Example n. 1110 = (35.0, 413.0): cluster 3\n", + "Example n. 1111 = (35.0, 412.0): cluster 3\n", + "Example n. 1112 = (35.0, 411.0): cluster 3\n", + "Example n. 1113 = (35.0, 410.0): cluster 3\n", + "Example n. 1114 = (35.0, 409.0): cluster 3\n", + "Example n. 1115 = (35.0, 408.0): cluster 3\n", + "Example n. 1116 = (35.0, 407.0): cluster 3\n", + "Example n. 1117 = (35.0, 406.0): cluster 3\n", + "Example n. 1118 = (35.0, 405.0): cluster 3\n", + "Example n. 1119 = (35.0, 404.0): cluster 3\n", + "Example n. 1120 = (35.0, 403.0): cluster 3\n", + "Example n. 1121 = (35.0, 402.0): cluster 3\n", + "Example n. 1122 = (36.0, 502.0): cluster 0\n", + "Example n. 1123 = (36.0, 501.0): cluster 0\n", + "Example n. 1124 = (36.0, 500.0): cluster 0\n", + "Example n. 1125 = (36.0, 499.0): cluster 0\n", + "Example n. 1126 = (36.0, 498.0): cluster 0\n", + "Example n. 1127 = (36.0, 497.0): cluster 0\n", + "Example n. 1128 = (36.0, 496.0): cluster 0\n", + "Example n. 1129 = (36.0, 495.0): cluster 0\n", + "Example n. 1130 = (36.0, 494.0): cluster 0\n", + "Example n. 1131 = (36.0, 493.0): cluster 0\n", + "Example n. 1132 = (36.0, 492.0): cluster 0\n", + "Example n. 1133 = (36.0, 491.0): cluster 0\n", + "Example n. 1134 = (36.0, 490.0): cluster 0\n", + "Example n. 1135 = (36.0, 489.0): cluster 0\n", + "Example n. 1136 = (36.0, 488.0): cluster 0\n", + "Example n. 1137 = (36.0, 487.0): cluster 0\n", + "Example n. 1138 = (36.0, 479.0): cluster 0\n", + "Example n. 1139 = (36.0, 478.0): cluster 0\n", + "Example n. 1140 = (36.0, 477.0): cluster 0\n", + "Example n. 1141 = (36.0, 476.0): cluster 0\n", + "Example n. 1142 = (36.0, 475.0): cluster 0\n", + "Example n. 1143 = (36.0, 474.0): cluster 0\n", + "Example n. 1144 = (36.0, 473.0): cluster 0\n", + "Example n. 1145 = (36.0, 472.0): cluster 0\n", + "Example n. 1146 = (36.0, 471.0): cluster 0\n", + "Example n. 1147 = (36.0, 470.0): cluster 0\n", + "Example n. 1148 = (36.0, 469.0): cluster 0\n", + "Example n. 1149 = (36.0, 468.0): cluster 0\n", + "Example n. 1150 = (36.0, 467.0): cluster 0\n", + "Example n. 1151 = (36.0, 466.0): cluster 0\n", + "Example n. 1152 = (36.0, 465.0): cluster 0\n", + "Example n. 1153 = (36.0, 464.0): cluster 0\n", + "Example n. 1154 = (36.0, 463.0): cluster 0\n", + "Example n. 1155 = (36.0, 462.0): cluster 0\n", + "Example n. 1156 = (36.0, 461.0): cluster 0\n", + "Example n. 1157 = (36.0, 460.0): cluster 0\n", + "Example n. 1158 = (36.0, 459.0): cluster 0\n", + "Example n. 1159 = (36.0, 458.0): cluster 0\n", + "Example n. 1160 = (36.0, 457.0): cluster 0\n", + "Example n. 1161 = (36.0, 456.0): cluster 0\n", + "Example n. 1162 = (36.0, 455.0): cluster 0\n", + "Example n. 1163 = (36.0, 454.0): cluster 0\n", + "Example n. 1164 = (36.0, 453.0): cluster 0\n", + "Example n. 1165 = (36.0, 452.0): cluster 1\n", + "Example n. 1166 = (36.0, 451.0): cluster 1\n", + "Example n. 1167 = (36.0, 450.0): cluster 1\n", + "Example n. 1168 = (36.0, 449.0): cluster 1\n", + "Example n. 1169 = (36.0, 448.0): cluster 1\n", + "Example n. 1170 = (36.0, 425.0): cluster 3\n", + "Example n. 1171 = (36.0, 424.0): cluster 3\n", + "Example n. 1172 = (36.0, 423.0): cluster 3\n", + "Example n. 1173 = (36.0, 422.0): cluster 3\n", + "Example n. 1174 = (36.0, 421.0): cluster 3\n", + "Example n. 1175 = (36.0, 420.0): cluster 3\n", + "Example n. 1176 = (36.0, 419.0): cluster 3\n", + "Example n. 1177 = (36.0, 418.0): cluster 3\n", + "Example n. 1178 = (36.0, 417.0): cluster 3\n", + "Example n. 1179 = (36.0, 416.0): cluster 3\n", + "Example n. 1180 = (36.0, 415.0): cluster 3\n", + "Example n. 1181 = (36.0, 414.0): cluster 3\n", + "Example n. 1182 = (36.0, 413.0): cluster 3\n", + "Example n. 1183 = (36.0, 412.0): cluster 3\n", + "Example n. 1184 = (36.0, 411.0): cluster 3\n", + "Example n. 1185 = (36.0, 410.0): cluster 3\n", + "Example n. 1186 = (36.0, 409.0): cluster 3\n", + "Example n. 1187 = (36.0, 408.0): cluster 3\n", + "Example n. 1188 = (36.0, 407.0): cluster 3\n", + "Example n. 1189 = (36.0, 406.0): cluster 3\n", + "Example n. 1190 = (36.0, 405.0): cluster 3\n", + "Example n. 1191 = (36.0, 404.0): cluster 3\n", + "Example n. 1192 = (36.0, 403.0): cluster 3\n", + "Example n. 1193 = (36.0, 402.0): cluster 3\n", + "Example n. 1194 = (37.0, 502.0): cluster 0\n", + "Example n. 1195 = (37.0, 501.0): cluster 0\n", + "Example n. 1196 = (37.0, 500.0): cluster 0\n", + "Example n. 1197 = (37.0, 499.0): cluster 0\n", + "Example n. 1198 = (37.0, 498.0): cluster 0\n", + "Example n. 1199 = (37.0, 497.0): cluster 0\n", + "Example n. 1200 = (37.0, 496.0): cluster 0\n", + "Example n. 1201 = (37.0, 495.0): cluster 0\n", + "Example n. 1202 = (37.0, 494.0): cluster 0\n", + "Example n. 1203 = (37.0, 493.0): cluster 0\n", + "Example n. 1204 = (37.0, 492.0): cluster 0\n", + "Example n. 1205 = (37.0, 491.0): cluster 0\n", + "Example n. 1206 = (37.0, 490.0): cluster 0\n", + "Example n. 1207 = (37.0, 489.0): cluster 0\n", + "Example n. 1208 = (37.0, 488.0): cluster 0\n", + "Example n. 1209 = (37.0, 487.0): cluster 0\n", + "Example n. 1210 = (37.0, 479.0): cluster 0\n", + "Example n. 1211 = (37.0, 478.0): cluster 0\n", + "Example n. 1212 = (37.0, 477.0): cluster 0\n", + "Example n. 1213 = (37.0, 476.0): cluster 0\n", + "Example n. 1214 = (37.0, 475.0): cluster 0\n", + "Example n. 1215 = (37.0, 474.0): cluster 0\n", + "Example n. 1216 = (37.0, 473.0): cluster 0\n", + "Example n. 1217 = (37.0, 472.0): cluster 0\n", + "Example n. 1218 = (37.0, 471.0): cluster 0\n", + "Example n. 1219 = (37.0, 470.0): cluster 0\n", + "Example n. 1220 = (37.0, 469.0): cluster 0\n", + "Example n. 1221 = (37.0, 468.0): cluster 0\n", + "Example n. 1222 = (37.0, 467.0): cluster 0\n", + "Example n. 1223 = (37.0, 466.0): cluster 0\n", + "Example n. 1224 = (37.0, 465.0): cluster 0\n", + "Example n. 1225 = (37.0, 464.0): cluster 0\n", + "Example n. 1226 = (37.0, 463.0): cluster 0\n", + "Example n. 1227 = (37.0, 462.0): cluster 0\n", + "Example n. 1228 = (37.0, 461.0): cluster 0\n", + "Example n. 1229 = (37.0, 460.0): cluster 0\n", + "Example n. 1230 = (37.0, 459.0): cluster 0\n", + "Example n. 1231 = (37.0, 458.0): cluster 0\n", + "Example n. 1232 = (37.0, 457.0): cluster 0\n", + "Example n. 1233 = (37.0, 456.0): cluster 0\n", + "Example n. 1234 = (37.0, 455.0): cluster 1\n", + "Example n. 1235 = (37.0, 454.0): cluster 1\n", + "Example n. 1236 = (37.0, 453.0): cluster 1\n", + "Example n. 1237 = (37.0, 452.0): cluster 1\n", + "Example n. 1238 = (37.0, 451.0): cluster 1\n", + "Example n. 1239 = (37.0, 450.0): cluster 1\n", + "Example n. 1240 = (37.0, 449.0): cluster 1\n", + "Example n. 1241 = (37.0, 448.0): cluster 1\n", + "Example n. 1242 = (37.0, 425.0): cluster 3\n", + "Example n. 1243 = (37.0, 424.0): cluster 3\n", + "Example n. 1244 = (37.0, 423.0): cluster 3\n", + "Example n. 1245 = (37.0, 422.0): cluster 3\n", + "Example n. 1246 = (37.0, 421.0): cluster 3\n", + "Example n. 1247 = (37.0, 420.0): cluster 3\n", + "Example n. 1248 = (37.0, 419.0): cluster 3\n", + "Example n. 1249 = (37.0, 418.0): cluster 3\n", + "Example n. 1250 = (37.0, 417.0): cluster 3\n", + "Example n. 1251 = (37.0, 416.0): cluster 3\n", + "Example n. 1252 = (37.0, 415.0): cluster 3\n", + "Example n. 1253 = (37.0, 414.0): cluster 3\n", + "Example n. 1254 = (37.0, 413.0): cluster 3\n", + "Example n. 1255 = (37.0, 412.0): cluster 3\n", + "Example n. 1256 = (37.0, 411.0): cluster 3\n", + "Example n. 1257 = (37.0, 410.0): cluster 3\n", + "Example n. 1258 = (37.0, 409.0): cluster 3\n", + "Example n. 1259 = (37.0, 408.0): cluster 3\n", + "Example n. 1260 = (37.0, 407.0): cluster 3\n", + "Example n. 1261 = (37.0, 406.0): cluster 3\n", + "Example n. 1262 = (37.0, 405.0): cluster 3\n", + "Example n. 1263 = (37.0, 404.0): cluster 3\n", + "Example n. 1264 = (37.0, 403.0): cluster 3\n", + "Example n. 1265 = (37.0, 402.0): cluster 3\n", + "Example n. 1266 = (38.0, 502.0): cluster 0\n", + "Example n. 1267 = (38.0, 501.0): cluster 0\n", + "Example n. 1268 = (38.0, 500.0): cluster 0\n", + "Example n. 1269 = (38.0, 499.0): cluster 0\n", + "Example n. 1270 = (38.0, 498.0): cluster 0\n", + "Example n. 1271 = (38.0, 497.0): cluster 0\n", + "Example n. 1272 = (38.0, 496.0): cluster 0\n", + "Example n. 1273 = (38.0, 495.0): cluster 0\n", + "Example n. 1274 = (38.0, 494.0): cluster 0\n", + "Example n. 1275 = (38.0, 493.0): cluster 0\n", + "Example n. 1276 = (38.0, 492.0): cluster 0\n", + "Example n. 1277 = (38.0, 491.0): cluster 0\n", + "Example n. 1278 = (38.0, 490.0): cluster 0\n", + "Example n. 1279 = (38.0, 489.0): cluster 0\n", + "Example n. 1280 = (38.0, 488.0): cluster 0\n", + "Example n. 1281 = (38.0, 487.0): cluster 0\n", + "Example n. 1282 = (38.0, 479.0): cluster 0\n", + "Example n. 1283 = (38.0, 478.0): cluster 0\n", + "Example n. 1284 = (38.0, 477.0): cluster 0\n", + "Example n. 1285 = (38.0, 476.0): cluster 0\n", + "Example n. 1286 = (38.0, 475.0): cluster 0\n", + "Example n. 1287 = (38.0, 474.0): cluster 0\n", + "Example n. 1288 = (38.0, 473.0): cluster 0\n", + "Example n. 1289 = (38.0, 472.0): cluster 0\n", + "Example n. 1290 = (38.0, 471.0): cluster 0\n", + "Example n. 1291 = (38.0, 470.0): cluster 0\n", + "Example n. 1292 = (38.0, 469.0): cluster 0\n", + "Example n. 1293 = (38.0, 468.0): cluster 0\n", + "Example n. 1294 = (38.0, 467.0): cluster 0\n", + "Example n. 1295 = (38.0, 466.0): cluster 0\n", + "Example n. 1296 = (38.0, 465.0): cluster 0\n", + "Example n. 1297 = (38.0, 464.0): cluster 0\n", + "Example n. 1298 = (38.0, 463.0): cluster 0\n", + "Example n. 1299 = (38.0, 462.0): cluster 0\n", + "Example n. 1300 = (38.0, 461.0): cluster 0\n", + "Example n. 1301 = (38.0, 460.0): cluster 0\n", + "Example n. 1302 = (38.0, 459.0): cluster 0\n", + "Example n. 1303 = (38.0, 458.0): cluster 0\n", + "Example n. 1304 = (38.0, 457.0): cluster 1\n", + "Example n. 1305 = (38.0, 456.0): cluster 1\n", + "Example n. 1306 = (38.0, 455.0): cluster 1\n", + "Example n. 1307 = (38.0, 454.0): cluster 1\n", + "Example n. 1308 = (38.0, 453.0): cluster 1\n", + "Example n. 1309 = (38.0, 452.0): cluster 1\n", + "Example n. 1310 = (38.0, 451.0): cluster 1\n", + "Example n. 1311 = (38.0, 450.0): cluster 1\n", + "Example n. 1312 = (38.0, 449.0): cluster 1\n", + "Example n. 1313 = (38.0, 448.0): cluster 1\n", + "Example n. 1314 = (38.0, 425.0): cluster 3\n", + "Example n. 1315 = (38.0, 424.0): cluster 3\n", + "Example n. 1316 = (38.0, 423.0): cluster 3\n", + "Example n. 1317 = (38.0, 422.0): cluster 3\n", + "Example n. 1318 = (38.0, 421.0): cluster 3\n", + "Example n. 1319 = (38.0, 420.0): cluster 3\n", + "Example n. 1320 = (38.0, 419.0): cluster 3\n", + "Example n. 1321 = (38.0, 418.0): cluster 3\n", + "Example n. 1322 = (38.0, 417.0): cluster 3\n", + "Example n. 1323 = (38.0, 416.0): cluster 3\n", + "Example n. 1324 = (38.0, 415.0): cluster 3\n", + "Example n. 1325 = (38.0, 414.0): cluster 3\n", + "Example n. 1326 = (38.0, 413.0): cluster 3\n", + "Example n. 1327 = (38.0, 412.0): cluster 3\n", + "Example n. 1328 = (38.0, 411.0): cluster 3\n", + "Example n. 1329 = (38.0, 410.0): cluster 3\n", + "Example n. 1330 = (38.0, 409.0): cluster 3\n", + "Example n. 1331 = (38.0, 408.0): cluster 3\n", + "Example n. 1332 = (38.0, 407.0): cluster 3\n", + "Example n. 1333 = (38.0, 406.0): cluster 3\n", + "Example n. 1334 = (38.0, 405.0): cluster 3\n", + "Example n. 1335 = (38.0, 404.0): cluster 3\n", + "Example n. 1336 = (38.0, 403.0): cluster 3\n", + "Example n. 1337 = (38.0, 402.0): cluster 3\n", + "Example n. 1338 = (39.0, 506.0): cluster 0\n", + "Example n. 1339 = (39.0, 505.0): cluster 0\n", + "Example n. 1340 = (39.0, 504.0): cluster 0\n", + "Example n. 1341 = (39.0, 503.0): cluster 0\n", + "Example n. 1342 = (39.0, 502.0): cluster 0\n", + "Example n. 1343 = (39.0, 501.0): cluster 0\n", + "Example n. 1344 = (39.0, 500.0): cluster 0\n", + "Example n. 1345 = (39.0, 499.0): cluster 0\n", + "Example n. 1346 = (39.0, 498.0): cluster 0\n", + "Example n. 1347 = (39.0, 497.0): cluster 0\n", + "Example n. 1348 = (39.0, 496.0): cluster 0\n", + "Example n. 1349 = (39.0, 495.0): cluster 0\n", + "Example n. 1350 = (39.0, 494.0): cluster 0\n", + "Example n. 1351 = (39.0, 493.0): cluster 0\n", + "Example n. 1352 = (39.0, 492.0): cluster 0\n", + "Example n. 1353 = (39.0, 491.0): cluster 0\n", + "Example n. 1354 = (39.0, 490.0): cluster 0\n", + "Example n. 1355 = (39.0, 489.0): cluster 0\n", + "Example n. 1356 = (39.0, 488.0): cluster 0\n", + "Example n. 1357 = (39.0, 487.0): cluster 0\n", + "Example n. 1358 = (39.0, 479.0): cluster 0\n", + "Example n. 1359 = (39.0, 478.0): cluster 0\n", + "Example n. 1360 = (39.0, 477.0): cluster 0\n", + "Example n. 1361 = (39.0, 476.0): cluster 0\n", + "Example n. 1362 = (39.0, 475.0): cluster 0\n", + "Example n. 1363 = (39.0, 474.0): cluster 0\n", + "Example n. 1364 = (39.0, 473.0): cluster 0\n", + "Example n. 1365 = (39.0, 472.0): cluster 0\n", + "Example n. 1366 = (39.0, 471.0): cluster 0\n", + "Example n. 1367 = (39.0, 470.0): cluster 0\n", + "Example n. 1368 = (39.0, 469.0): cluster 0\n", + "Example n. 1369 = (39.0, 468.0): cluster 0\n", + "Example n. 1370 = (39.0, 467.0): cluster 0\n", + "Example n. 1371 = (39.0, 466.0): cluster 0\n", + "Example n. 1372 = (39.0, 465.0): cluster 0\n", + "Example n. 1373 = (39.0, 464.0): cluster 0\n", + "Example n. 1374 = (39.0, 463.0): cluster 0\n", + "Example n. 1375 = (39.0, 462.0): cluster 0\n", + "Example n. 1376 = (39.0, 461.0): cluster 0\n", + "Example n. 1377 = (39.0, 460.0): cluster 1\n", + "Example n. 1378 = (39.0, 459.0): cluster 1\n", + "Example n. 1379 = (39.0, 458.0): cluster 1\n", + "Example n. 1380 = (39.0, 457.0): cluster 1\n", + "Example n. 1381 = (39.0, 456.0): cluster 1\n", + "Example n. 1382 = (39.0, 455.0): cluster 1\n", + "Example n. 1383 = (39.0, 454.0): cluster 1\n", + "Example n. 1384 = (39.0, 453.0): cluster 1\n", + "Example n. 1385 = (39.0, 452.0): cluster 1\n", + "Example n. 1386 = (39.0, 451.0): cluster 1\n", + "Example n. 1387 = (39.0, 450.0): cluster 1\n", + "Example n. 1388 = (39.0, 449.0): cluster 1\n", + "Example n. 1389 = (39.0, 448.0): cluster 1\n", + "Example n. 1390 = (39.0, 429.0): cluster 3\n", + "Example n. 1391 = (39.0, 428.0): cluster 3\n", + "Example n. 1392 = (39.0, 427.0): cluster 3\n", + "Example n. 1393 = (39.0, 426.0): cluster 3\n", + "Example n. 1394 = (39.0, 425.0): cluster 3\n", + "Example n. 1395 = (39.0, 424.0): cluster 3\n", + "Example n. 1396 = (39.0, 423.0): cluster 3\n", + "Example n. 1397 = (39.0, 422.0): cluster 3\n", + "Example n. 1398 = (39.0, 421.0): cluster 3\n", + "Example n. 1399 = (39.0, 420.0): cluster 3\n", + "Example n. 1400 = (39.0, 419.0): cluster 3\n", + "Example n. 1401 = (39.0, 418.0): cluster 3\n", + "Example n. 1402 = (39.0, 417.0): cluster 3\n", + "Example n. 1403 = (39.0, 416.0): cluster 3\n", + "Example n. 1404 = (39.0, 415.0): cluster 3\n", + "Example n. 1405 = (39.0, 414.0): cluster 3\n", + "Example n. 1406 = (39.0, 413.0): cluster 3\n", + "Example n. 1407 = (39.0, 412.0): cluster 3\n", + "Example n. 1408 = (39.0, 411.0): cluster 3\n", + "Example n. 1409 = (39.0, 410.0): cluster 3\n", + "Example n. 1410 = (39.0, 409.0): cluster 3\n", + "Example n. 1411 = (39.0, 408.0): cluster 3\n", + "Example n. 1412 = (39.0, 407.0): cluster 3\n", + "Example n. 1413 = (39.0, 406.0): cluster 3\n", + "Example n. 1414 = (39.0, 405.0): cluster 3\n", + "Example n. 1415 = (39.0, 404.0): cluster 3\n", + "Example n. 1416 = (39.0, 403.0): cluster 3\n", + "Example n. 1417 = (39.0, 402.0): cluster 3\n", + "Example n. 1418 = (39.0, 401.0): cluster 3\n", + "Example n. 1419 = (39.0, 400.0): cluster 3\n", + "Example n. 1420 = (39.0, 399.0): cluster 3\n", + "Example n. 1421 = (39.0, 398.0): cluster 3\n", + "Example n. 1422 = (40.0, 506.0): cluster 0\n", + "Example n. 1423 = (40.0, 505.0): cluster 0\n", + "Example n. 1424 = (40.0, 504.0): cluster 0\n", + "Example n. 1425 = (40.0, 503.0): cluster 0\n", + "Example n. 1426 = (40.0, 502.0): cluster 0\n", + "Example n. 1427 = (40.0, 501.0): cluster 0\n", + "Example n. 1428 = (40.0, 500.0): cluster 0\n", + "Example n. 1429 = (40.0, 499.0): cluster 0\n", + "Example n. 1430 = (40.0, 498.0): cluster 0\n", + "Example n. 1431 = (40.0, 497.0): cluster 0\n", + "Example n. 1432 = (40.0, 496.0): cluster 0\n", + "Example n. 1433 = (40.0, 495.0): cluster 0\n", + "Example n. 1434 = (40.0, 494.0): cluster 0\n", + "Example n. 1435 = (40.0, 493.0): cluster 0\n", + "Example n. 1436 = (40.0, 492.0): cluster 0\n", + "Example n. 1437 = (40.0, 491.0): cluster 0\n", + "Example n. 1438 = (40.0, 490.0): cluster 0\n", + "Example n. 1439 = (40.0, 489.0): cluster 0\n", + "Example n. 1440 = (40.0, 488.0): cluster 0\n", + "Example n. 1441 = (40.0, 487.0): cluster 0\n", + "Example n. 1442 = (40.0, 475.0): cluster 0\n", + "Example n. 1443 = (40.0, 474.0): cluster 0\n", + "Example n. 1444 = (40.0, 473.0): cluster 0\n", + "Example n. 1445 = (40.0, 472.0): cluster 0\n", + "Example n. 1446 = (40.0, 471.0): cluster 0\n", + "Example n. 1447 = (40.0, 470.0): cluster 0\n", + "Example n. 1448 = (40.0, 469.0): cluster 0\n", + "Example n. 1449 = (40.0, 468.0): cluster 0\n", + "Example n. 1450 = (40.0, 467.0): cluster 0\n", + "Example n. 1451 = (40.0, 466.0): cluster 0\n", + "Example n. 1452 = (40.0, 465.0): cluster 0\n", + "Example n. 1453 = (40.0, 464.0): cluster 0\n", + "Example n. 1454 = (40.0, 459.0): cluster 1\n", + "Example n. 1455 = (40.0, 458.0): cluster 1\n", + "Example n. 1456 = (40.0, 457.0): cluster 1\n", + "Example n. 1457 = (40.0, 456.0): cluster 1\n", + "Example n. 1458 = (40.0, 455.0): cluster 1\n", + "Example n. 1459 = (40.0, 454.0): cluster 1\n", + "Example n. 1460 = (40.0, 453.0): cluster 1\n", + "Example n. 1461 = (40.0, 452.0): cluster 1\n", + "Example n. 1462 = (40.0, 451.0): cluster 1\n", + "Example n. 1463 = (40.0, 450.0): cluster 1\n", + "Example n. 1464 = (40.0, 449.0): cluster 1\n", + "Example n. 1465 = (40.0, 448.0): cluster 1\n", + "Example n. 1466 = (40.0, 429.0): cluster 3\n", + "Example n. 1467 = (40.0, 428.0): cluster 3\n", + "Example n. 1468 = (40.0, 427.0): cluster 3\n", + "Example n. 1469 = (40.0, 426.0): cluster 3\n", + "Example n. 1470 = (40.0, 425.0): cluster 3\n", + "Example n. 1471 = (40.0, 424.0): cluster 3\n", + "Example n. 1472 = (40.0, 423.0): cluster 3\n", + "Example n. 1473 = (40.0, 422.0): cluster 3\n", + "Example n. 1474 = (40.0, 421.0): cluster 3\n", + "Example n. 1475 = (40.0, 420.0): cluster 3\n", + "Example n. 1476 = (40.0, 419.0): cluster 3\n", + "Example n. 1477 = (40.0, 418.0): cluster 3\n", + "Example n. 1478 = (40.0, 417.0): cluster 3\n", + "Example n. 1479 = (40.0, 416.0): cluster 3\n", + "Example n. 1480 = (40.0, 415.0): cluster 3\n", + "Example n. 1481 = (40.0, 414.0): cluster 3\n", + "Example n. 1482 = (40.0, 413.0): cluster 3\n", + "Example n. 1483 = (40.0, 412.0): cluster 3\n", + "Example n. 1484 = (40.0, 411.0): cluster 3\n", + "Example n. 1485 = (40.0, 410.0): cluster 3\n", + "Example n. 1486 = (40.0, 409.0): cluster 3\n", + "Example n. 1487 = (40.0, 408.0): cluster 3\n", + "Example n. 1488 = (40.0, 407.0): cluster 3\n", + "Example n. 1489 = (40.0, 406.0): cluster 3\n", + "Example n. 1490 = (40.0, 405.0): cluster 3\n", + "Example n. 1491 = (40.0, 404.0): cluster 3\n", + "Example n. 1492 = (40.0, 403.0): cluster 3\n", + "Example n. 1493 = (40.0, 402.0): cluster 3\n", + "Example n. 1494 = (40.0, 401.0): cluster 3\n", + "Example n. 1495 = (40.0, 400.0): cluster 3\n", + "Example n. 1496 = (40.0, 399.0): cluster 3\n", + "Example n. 1497 = (40.0, 398.0): cluster 3\n", + "Example n. 1498 = (41.0, 506.0): cluster 0\n", + "Example n. 1499 = (41.0, 505.0): cluster 0\n", + "Example n. 1500 = (41.0, 504.0): cluster 0\n", + "Example n. 1501 = (41.0, 503.0): cluster 0\n", + "Example n. 1502 = (41.0, 502.0): cluster 0\n", + "Example n. 1503 = (41.0, 501.0): cluster 0\n", + "Example n. 1504 = (41.0, 500.0): cluster 0\n", + "Example n. 1505 = (41.0, 499.0): cluster 0\n", + "Example n. 1506 = (41.0, 498.0): cluster 0\n", + "Example n. 1507 = (41.0, 497.0): cluster 0\n", + "Example n. 1508 = (41.0, 496.0): cluster 0\n", + "Example n. 1509 = (41.0, 495.0): cluster 0\n", + "Example n. 1510 = (41.0, 494.0): cluster 0\n", + "Example n. 1511 = (41.0, 493.0): cluster 0\n", + "Example n. 1512 = (41.0, 492.0): cluster 0\n", + "Example n. 1513 = (41.0, 491.0): cluster 0\n", + "Example n. 1514 = (41.0, 490.0): cluster 0\n", + "Example n. 1515 = (41.0, 489.0): cluster 0\n", + "Example n. 1516 = (41.0, 488.0): cluster 0\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Example n. 1517 = (41.0, 487.0): cluster 0\n", + "Example n. 1518 = (41.0, 475.0): cluster 0\n", + "Example n. 1519 = (41.0, 474.0): cluster 0\n", + "Example n. 1520 = (41.0, 473.0): cluster 0\n", + "Example n. 1521 = (41.0, 472.0): cluster 0\n", + "Example n. 1522 = (41.0, 471.0): cluster 0\n", + "Example n. 1523 = (41.0, 470.0): cluster 0\n", + "Example n. 1524 = (41.0, 469.0): cluster 0\n", + "Example n. 1525 = (41.0, 468.0): cluster 0\n", + "Example n. 1526 = (41.0, 467.0): cluster 0\n", + "Example n. 1527 = (41.0, 466.0): cluster 0\n", + "Example n. 1528 = (41.0, 465.0): cluster 1\n", + "Example n. 1529 = (41.0, 464.0): cluster 1\n", + "Example n. 1530 = (41.0, 459.0): cluster 1\n", + "Example n. 1531 = (41.0, 458.0): cluster 1\n", + "Example n. 1532 = (41.0, 457.0): cluster 1\n", + "Example n. 1533 = (41.0, 456.0): cluster 1\n", + "Example n. 1534 = (41.0, 455.0): cluster 1\n", + "Example n. 1535 = (41.0, 454.0): cluster 1\n", + "Example n. 1536 = (41.0, 453.0): cluster 1\n", + "Example n. 1537 = (41.0, 452.0): cluster 1\n", + "Example n. 1538 = (41.0, 451.0): cluster 1\n", + "Example n. 1539 = (41.0, 450.0): cluster 1\n", + "Example n. 1540 = (41.0, 449.0): cluster 1\n", + "Example n. 1541 = (41.0, 448.0): cluster 1\n", + "Example n. 1542 = (41.0, 429.0): cluster 3\n", + "Example n. 1543 = (41.0, 428.0): cluster 3\n", + "Example n. 1544 = (41.0, 427.0): cluster 3\n", + "Example n. 1545 = (41.0, 426.0): cluster 3\n", + "Example n. 1546 = (41.0, 425.0): cluster 3\n", + "Example n. 1547 = (41.0, 424.0): cluster 3\n", + "Example n. 1548 = (41.0, 423.0): cluster 3\n", + "Example n. 1549 = (41.0, 422.0): cluster 3\n", + "Example n. 1550 = (41.0, 421.0): cluster 3\n", + "Example n. 1551 = (41.0, 420.0): cluster 3\n", + "Example n. 1552 = (41.0, 419.0): cluster 3\n", + "Example n. 1553 = (41.0, 418.0): cluster 3\n", + "Example n. 1554 = (41.0, 417.0): cluster 3\n", + "Example n. 1555 = (41.0, 416.0): cluster 3\n", + "Example n. 1556 = (41.0, 415.0): cluster 3\n", + "Example n. 1557 = (41.0, 414.0): cluster 3\n", + "Example n. 1558 = (41.0, 413.0): cluster 3\n", + "Example n. 1559 = (41.0, 412.0): cluster 3\n", + "Example n. 1560 = (41.0, 411.0): cluster 3\n", + "Example n. 1561 = (41.0, 410.0): cluster 3\n", + "Example n. 1562 = (41.0, 409.0): cluster 3\n", + "Example n. 1563 = (41.0, 408.0): cluster 3\n", + "Example n. 1564 = (41.0, 407.0): cluster 3\n", + "Example n. 1565 = (41.0, 406.0): cluster 3\n", + "Example n. 1566 = (41.0, 405.0): cluster 3\n", + "Example n. 1567 = (41.0, 404.0): cluster 3\n", + "Example n. 1568 = (41.0, 403.0): cluster 3\n", + "Example n. 1569 = (41.0, 402.0): cluster 3\n", + "Example n. 1570 = (41.0, 401.0): cluster 3\n", + "Example n. 1571 = (41.0, 400.0): cluster 3\n", + "Example n. 1572 = (41.0, 399.0): cluster 3\n", + "Example n. 1573 = (41.0, 398.0): cluster 3\n", + "Example n. 1574 = (42.0, 506.0): cluster 0\n", + "Example n. 1575 = (42.0, 505.0): cluster 0\n", + "Example n. 1576 = (42.0, 504.0): cluster 0\n", + "Example n. 1577 = (42.0, 503.0): cluster 0\n", + "Example n. 1578 = (42.0, 502.0): cluster 0\n", + "Example n. 1579 = (42.0, 501.0): cluster 0\n", + "Example n. 1580 = (42.0, 500.0): cluster 0\n", + "Example n. 1581 = (42.0, 499.0): cluster 0\n", + "Example n. 1582 = (42.0, 498.0): cluster 0\n", + "Example n. 1583 = (42.0, 497.0): cluster 0\n", + "Example n. 1584 = (42.0, 496.0): cluster 0\n", + "Example n. 1585 = (42.0, 495.0): cluster 0\n", + "Example n. 1586 = (42.0, 494.0): cluster 0\n", + "Example n. 1587 = (42.0, 493.0): cluster 0\n", + "Example n. 1588 = (42.0, 492.0): cluster 0\n", + "Example n. 1589 = (42.0, 491.0): cluster 0\n", + "Example n. 1590 = (42.0, 490.0): cluster 0\n", + "Example n. 1591 = (42.0, 489.0): cluster 0\n", + "Example n. 1592 = (42.0, 488.0): cluster 0\n", + "Example n. 1593 = (42.0, 487.0): cluster 0\n", + "Example n. 1594 = (42.0, 475.0): cluster 0\n", + "Example n. 1595 = (42.0, 474.0): cluster 0\n", + "Example n. 1596 = (42.0, 473.0): cluster 0\n", + "Example n. 1597 = (42.0, 472.0): cluster 0\n", + "Example n. 1598 = (42.0, 471.0): cluster 0\n", + "Example n. 1599 = (42.0, 470.0): cluster 0\n", + "Example n. 1600 = (42.0, 469.0): cluster 0\n", + "Example n. 1601 = (42.0, 468.0): cluster 0\n", + "Example n. 1602 = (42.0, 467.0): cluster 1\n", + "Example n. 1603 = (42.0, 466.0): cluster 1\n", + "Example n. 1604 = (42.0, 465.0): cluster 1\n", + "Example n. 1605 = (42.0, 464.0): cluster 1\n", + "Example n. 1606 = (42.0, 459.0): cluster 1\n", + "Example n. 1607 = (42.0, 458.0): cluster 1\n", + "Example n. 1608 = (42.0, 457.0): cluster 1\n", + "Example n. 1609 = (42.0, 456.0): cluster 1\n", + "Example n. 1610 = (42.0, 455.0): cluster 1\n", + "Example n. 1611 = (42.0, 454.0): cluster 1\n", + "Example n. 1612 = (42.0, 453.0): cluster 1\n", + "Example n. 1613 = (42.0, 452.0): cluster 1\n", + "Example n. 1614 = (42.0, 451.0): cluster 1\n", + "Example n. 1615 = (42.0, 450.0): cluster 1\n", + "Example n. 1616 = (42.0, 449.0): cluster 1\n", + "Example n. 1617 = (42.0, 448.0): cluster 1\n", + "Example n. 1618 = (42.0, 429.0): cluster 3\n", + "Example n. 1619 = (42.0, 428.0): cluster 3\n", + "Example n. 1620 = (42.0, 427.0): cluster 3\n", + "Example n. 1621 = (42.0, 426.0): cluster 3\n", + "Example n. 1622 = (42.0, 425.0): cluster 3\n", + "Example n. 1623 = (42.0, 424.0): cluster 3\n", + "Example n. 1624 = (42.0, 423.0): cluster 3\n", + "Example n. 1625 = (42.0, 422.0): cluster 3\n", + "Example n. 1626 = (42.0, 421.0): cluster 3\n", + "Example n. 1627 = (42.0, 420.0): cluster 3\n", + "Example n. 1628 = (42.0, 419.0): cluster 3\n", + "Example n. 1629 = (42.0, 418.0): cluster 3\n", + "Example n. 1630 = (42.0, 417.0): cluster 3\n", + "Example n. 1631 = (42.0, 416.0): cluster 3\n", + "Example n. 1632 = (42.0, 415.0): cluster 3\n", + "Example n. 1633 = (42.0, 414.0): cluster 3\n", + "Example n. 1634 = (42.0, 413.0): cluster 3\n", + "Example n. 1635 = (42.0, 412.0): cluster 3\n", + "Example n. 1636 = (42.0, 411.0): cluster 3\n", + "Example n. 1637 = (42.0, 410.0): cluster 3\n", + "Example n. 1638 = (42.0, 409.0): cluster 3\n", + "Example n. 1639 = (42.0, 408.0): cluster 3\n", + "Example n. 1640 = (42.0, 406.0): cluster 3\n", + "Example n. 1641 = (42.0, 405.0): cluster 3\n", + "Example n. 1642 = (42.0, 404.0): cluster 3\n", + "Example n. 1643 = (42.0, 403.0): cluster 3\n", + "Example n. 1644 = (42.0, 402.0): cluster 3\n", + "Example n. 1645 = (42.0, 401.0): cluster 3\n", + "Example n. 1646 = (42.0, 400.0): cluster 3\n", + "Example n. 1647 = (42.0, 399.0): cluster 3\n", + "Example n. 1648 = (42.0, 398.0): cluster 3\n", + "Example n. 1649 = (43.0, 506.0): cluster 0\n", + "Example n. 1650 = (43.0, 505.0): cluster 0\n", + "Example n. 1651 = (43.0, 504.0): cluster 0\n", + "Example n. 1652 = (43.0, 503.0): cluster 0\n", + "Example n. 1653 = (43.0, 502.0): cluster 0\n", + "Example n. 1654 = (43.0, 501.0): cluster 0\n", + "Example n. 1655 = (43.0, 500.0): cluster 0\n", + "Example n. 1656 = (43.0, 499.0): cluster 0\n", + "Example n. 1657 = (43.0, 498.0): cluster 0\n", + "Example n. 1658 = (43.0, 497.0): cluster 0\n", + "Example n. 1659 = (43.0, 496.0): cluster 0\n", + "Example n. 1660 = (43.0, 495.0): cluster 0\n", + "Example n. 1661 = (43.0, 494.0): cluster 0\n", + "Example n. 1662 = (43.0, 493.0): cluster 0\n", + "Example n. 1663 = (43.0, 492.0): cluster 0\n", + "Example n. 1664 = (43.0, 491.0): cluster 0\n", + "Example n. 1665 = (43.0, 490.0): cluster 0\n", + "Example n. 1666 = (43.0, 489.0): cluster 0\n", + "Example n. 1667 = (43.0, 488.0): cluster 0\n", + "Example n. 1668 = (43.0, 487.0): cluster 0\n", + "Example n. 1669 = (43.0, 475.0): cluster 0\n", + "Example n. 1670 = (43.0, 474.0): cluster 0\n", + "Example n. 1671 = (43.0, 473.0): cluster 0\n", + "Example n. 1672 = (43.0, 472.0): cluster 0\n", + "Example n. 1673 = (43.0, 471.0): cluster 0\n", + "Example n. 1674 = (43.0, 470.0): cluster 1\n", + "Example n. 1675 = (43.0, 469.0): cluster 1\n", + "Example n. 1676 = (43.0, 468.0): cluster 1\n", + "Example n. 1677 = (43.0, 467.0): cluster 1\n", + "Example n. 1678 = (43.0, 466.0): cluster 1\n", + "Example n. 1679 = (43.0, 465.0): cluster 1\n", + "Example n. 1680 = (43.0, 464.0): cluster 1\n", + "Example n. 1681 = (43.0, 463.0): cluster 1\n", + "Example n. 1682 = (43.0, 462.0): cluster 1\n", + "Example n. 1683 = (43.0, 461.0): cluster 1\n", + "Example n. 1684 = (43.0, 460.0): cluster 1\n", + "Example n. 1685 = (43.0, 459.0): cluster 1\n", + "Example n. 1686 = (43.0, 458.0): cluster 1\n", + "Example n. 1687 = (43.0, 457.0): cluster 1\n", + "Example n. 1688 = (43.0, 456.0): cluster 1\n", + "Example n. 1689 = (43.0, 455.0): cluster 1\n", + "Example n. 1690 = (43.0, 454.0): cluster 1\n", + "Example n. 1691 = (43.0, 453.0): cluster 1\n", + "Example n. 1692 = (43.0, 452.0): cluster 1\n", + "Example n. 1693 = (43.0, 451.0): cluster 1\n", + "Example n. 1694 = (43.0, 450.0): cluster 1\n", + "Example n. 1695 = (43.0, 449.0): cluster 1\n", + "Example n. 1696 = (43.0, 448.0): cluster 1\n", + "Example n. 1697 = (43.0, 432.0): cluster 3\n", + "Example n. 1698 = (43.0, 431.0): cluster 3\n", + "Example n. 1699 = (43.0, 430.0): cluster 3\n", + "Example n. 1700 = (43.0, 429.0): cluster 3\n", + "Example n. 1701 = (43.0, 428.0): cluster 3\n", + "Example n. 1702 = (43.0, 427.0): cluster 3\n", + "Example n. 1703 = (43.0, 426.0): cluster 3\n", + "Example n. 1704 = (43.0, 425.0): cluster 3\n", + "Example n. 1705 = (43.0, 424.0): cluster 3\n", + "Example n. 1706 = (43.0, 423.0): cluster 3\n", + "Example n. 1707 = (43.0, 422.0): cluster 3\n", + "Example n. 1708 = (43.0, 421.0): cluster 3\n", + "Example n. 1709 = (43.0, 420.0): cluster 3\n", + "Example n. 1710 = (43.0, 419.0): cluster 3\n", + "Example n. 1711 = (43.0, 418.0): cluster 3\n", + "Example n. 1712 = (43.0, 417.0): cluster 3\n", + "Example n. 1713 = (43.0, 416.0): cluster 3\n", + "Example n. 1714 = (43.0, 415.0): cluster 3\n", + "Example n. 1715 = (43.0, 414.0): cluster 3\n", + "Example n. 1716 = (43.0, 413.0): cluster 3\n", + "Example n. 1717 = (43.0, 412.0): cluster 3\n", + "Example n. 1718 = (43.0, 411.0): cluster 3\n", + "Example n. 1719 = (43.0, 410.0): cluster 3\n", + "Example n. 1720 = (43.0, 409.0): cluster 3\n", + "Example n. 1721 = (43.0, 408.0): cluster 3\n", + "Example n. 1722 = (43.0, 407.0): cluster 3\n", + "Example n. 1723 = (43.0, 406.0): cluster 3\n", + "Example n. 1724 = (43.0, 405.0): cluster 3\n", + "Example n. 1725 = (43.0, 404.0): cluster 3\n", + "Example n. 1726 = (43.0, 403.0): cluster 3\n", + "Example n. 1727 = (43.0, 402.0): cluster 3\n", + "Example n. 1728 = (43.0, 401.0): cluster 3\n", + "Example n. 1729 = (43.0, 400.0): cluster 3\n", + "Example n. 1730 = (43.0, 399.0): cluster 3\n", + "Example n. 1731 = (43.0, 398.0): cluster 3\n", + "Example n. 1732 = (44.0, 502.0): cluster 0\n", + "Example n. 1733 = (44.0, 501.0): cluster 0\n", + "Example n. 1734 = (44.0, 500.0): cluster 0\n", + "Example n. 1735 = (44.0, 499.0): cluster 0\n", + "Example n. 1736 = (44.0, 498.0): cluster 0\n", + "Example n. 1737 = (44.0, 497.0): cluster 0\n", + "Example n. 1738 = (44.0, 496.0): cluster 0\n", + "Example n. 1739 = (44.0, 495.0): cluster 0\n", + "Example n. 1740 = (44.0, 494.0): cluster 0\n", + "Example n. 1741 = (44.0, 493.0): cluster 0\n", + "Example n. 1742 = (44.0, 492.0): cluster 0\n", + "Example n. 1743 = (44.0, 491.0): cluster 0\n", + "Example n. 1744 = (44.0, 490.0): cluster 0\n", + "Example n. 1745 = (44.0, 489.0): cluster 0\n", + "Example n. 1746 = (44.0, 488.0): cluster 0\n", + "Example n. 1747 = (44.0, 487.0): cluster 0\n", + "Example n. 1748 = (44.0, 475.0): cluster 0\n", + "Example n. 1749 = (44.0, 474.0): cluster 0\n", + "Example n. 1750 = (44.0, 473.0): cluster 0\n", + "Example n. 1751 = (44.0, 472.0): cluster 1\n", + "Example n. 1752 = (44.0, 471.0): cluster 1\n", + "Example n. 1753 = (44.0, 470.0): cluster 1\n", + "Example n. 1754 = (44.0, 469.0): cluster 1\n", + "Example n. 1755 = (44.0, 468.0): cluster 1\n", + "Example n. 1756 = (44.0, 463.0): cluster 1\n", + "Example n. 1757 = (44.0, 462.0): cluster 1\n", + "Example n. 1758 = (44.0, 461.0): cluster 1\n", + "Example n. 1759 = (44.0, 460.0): cluster 1\n", + "Example n. 1760 = (44.0, 459.0): cluster 1\n", + "Example n. 1761 = (44.0, 458.0): cluster 1\n", + "Example n. 1762 = (44.0, 457.0): cluster 1\n", + "Example n. 1763 = (44.0, 456.0): cluster 1\n", + "Example n. 1764 = (44.0, 455.0): cluster 1\n", + "Example n. 1765 = (44.0, 454.0): cluster 1\n", + "Example n. 1766 = (44.0, 453.0): cluster 1\n", + "Example n. 1767 = (44.0, 452.0): cluster 1\n", + "Example n. 1768 = (44.0, 451.0): cluster 1\n", + "Example n. 1769 = (44.0, 450.0): cluster 1\n", + "Example n. 1770 = (44.0, 449.0): cluster 1\n", + "Example n. 1771 = (44.0, 448.0): cluster 1\n", + "Example n. 1772 = (44.0, 432.0): cluster 3\n", + "Example n. 1773 = (44.0, 431.0): cluster 3\n", + "Example n. 1774 = (44.0, 430.0): cluster 3\n", + "Example n. 1775 = (44.0, 429.0): cluster 3\n", + "Example n. 1776 = (44.0, 428.0): cluster 3\n", + "Example n. 1777 = (44.0, 427.0): cluster 3\n", + "Example n. 1778 = (44.0, 426.0): cluster 3\n", + "Example n. 1779 = (44.0, 425.0): cluster 3\n", + "Example n. 1780 = (44.0, 424.0): cluster 3\n", + "Example n. 1781 = (44.0, 423.0): cluster 3\n", + "Example n. 1782 = (44.0, 422.0): cluster 3\n", + "Example n. 1783 = (44.0, 421.0): cluster 3\n", + "Example n. 1784 = (44.0, 420.0): cluster 3\n", + "Example n. 1785 = (44.0, 419.0): cluster 3\n", + "Example n. 1786 = (44.0, 418.0): cluster 3\n", + "Example n. 1787 = (44.0, 417.0): cluster 3\n", + "Example n. 1788 = (44.0, 416.0): cluster 3\n", + "Example n. 1789 = (44.0, 415.0): cluster 3\n", + "Example n. 1790 = (44.0, 414.0): cluster 3\n", + "Example n. 1791 = (44.0, 409.0): cluster 3\n", + "Example n. 1792 = (44.0, 408.0): cluster 3\n", + "Example n. 1793 = (44.0, 407.0): cluster 3\n", + "Example n. 1794 = (44.0, 406.0): cluster 3\n", + "Example n. 1795 = (44.0, 405.0): cluster 3\n", + "Example n. 1796 = (44.0, 404.0): cluster 3\n", + "Example n. 1797 = (44.0, 403.0): cluster 3\n", + "Example n. 1798 = (44.0, 402.0): cluster 3\n", + "Example n. 1799 = (44.0, 401.0): cluster 3\n", + "Example n. 1800 = (44.0, 400.0): cluster 3\n", + "Example n. 1801 = (44.0, 399.0): cluster 3\n", + "Example n. 1802 = (44.0, 398.0): cluster 3\n", + "Example n. 1803 = (45.0, 502.0): cluster 0\n", + "Example n. 1804 = (45.0, 501.0): cluster 0\n", + "Example n. 1805 = (45.0, 500.0): cluster 0\n", + "Example n. 1806 = (45.0, 499.0): cluster 0\n", + "Example n. 1807 = (45.0, 498.0): cluster 0\n", + "Example n. 1808 = (45.0, 497.0): cluster 0\n", + "Example n. 1809 = (45.0, 496.0): cluster 0\n", + "Example n. 1810 = (45.0, 495.0): cluster 0\n", + "Example n. 1811 = (45.0, 494.0): cluster 0\n", + "Example n. 1812 = (45.0, 493.0): cluster 0\n", + "Example n. 1813 = (45.0, 492.0): cluster 0\n", + "Example n. 1814 = (45.0, 491.0): cluster 0\n", + "Example n. 1815 = (45.0, 490.0): cluster 0\n", + "Example n. 1816 = (45.0, 489.0): cluster 0\n", + "Example n. 1817 = (45.0, 488.0): cluster 0\n", + "Example n. 1818 = (45.0, 487.0): cluster 0\n", + "Example n. 1819 = (45.0, 475.0): cluster 1\n", + "Example n. 1820 = (45.0, 474.0): cluster 1\n", + "Example n. 1821 = (45.0, 473.0): cluster 1\n", + "Example n. 1822 = (45.0, 472.0): cluster 1\n", + "Example n. 1823 = (45.0, 471.0): cluster 1\n", + "Example n. 1824 = (45.0, 470.0): cluster 1\n", + "Example n. 1825 = (45.0, 469.0): cluster 1\n", + "Example n. 1826 = (45.0, 468.0): cluster 1\n", + "Example n. 1827 = (45.0, 463.0): cluster 1\n", + "Example n. 1828 = (45.0, 462.0): cluster 1\n", + "Example n. 1829 = (45.0, 461.0): cluster 1\n", + "Example n. 1830 = (45.0, 460.0): cluster 1\n", + "Example n. 1831 = (45.0, 459.0): cluster 1\n", + "Example n. 1832 = (45.0, 458.0): cluster 1\n", + "Example n. 1833 = (45.0, 457.0): cluster 1\n", + "Example n. 1834 = (45.0, 456.0): cluster 1\n", + "Example n. 1835 = (45.0, 455.0): cluster 1\n", + "Example n. 1836 = (45.0, 454.0): cluster 1\n", + "Example n. 1837 = (45.0, 453.0): cluster 1\n", + "Example n. 1838 = (45.0, 452.0): cluster 1\n", + "Example n. 1839 = (45.0, 451.0): cluster 1\n", + "Example n. 1840 = (45.0, 450.0): cluster 1\n", + "Example n. 1841 = (45.0, 449.0): cluster 1\n", + "Example n. 1842 = (45.0, 448.0): cluster 1\n", + "Example n. 1843 = (45.0, 432.0): cluster 3\n", + "Example n. 1844 = (45.0, 431.0): cluster 3\n", + "Example n. 1845 = (45.0, 430.0): cluster 3\n", + "Example n. 1846 = (45.0, 429.0): cluster 3\n", + "Example n. 1847 = (45.0, 428.0): cluster 3\n", + "Example n. 1848 = (45.0, 427.0): cluster 3\n", + "Example n. 1849 = (45.0, 426.0): cluster 3\n", + "Example n. 1850 = (45.0, 425.0): cluster 3\n", + "Example n. 1851 = (45.0, 424.0): cluster 3\n", + "Example n. 1852 = (45.0, 423.0): cluster 3\n", + "Example n. 1853 = (45.0, 422.0): cluster 3\n", + "Example n. 1854 = (45.0, 421.0): cluster 3\n", + "Example n. 1855 = (45.0, 420.0): cluster 3\n", + "Example n. 1856 = (45.0, 419.0): cluster 3\n", + "Example n. 1857 = (45.0, 418.0): cluster 3\n", + "Example n. 1858 = (45.0, 417.0): cluster 3\n", + "Example n. 1859 = (45.0, 416.0): cluster 3\n", + "Example n. 1860 = (45.0, 415.0): cluster 3\n", + "Example n. 1861 = (45.0, 414.0): cluster 3\n", + "Example n. 1862 = (45.0, 409.0): cluster 3\n", + "Example n. 1863 = (45.0, 408.0): cluster 3\n", + "Example n. 1864 = (45.0, 407.0): cluster 3\n", + "Example n. 1865 = (45.0, 406.0): cluster 3\n", + "Example n. 1866 = (45.0, 405.0): cluster 3\n", + "Example n. 1867 = (45.0, 404.0): cluster 3\n", + "Example n. 1868 = (45.0, 403.0): cluster 3\n", + "Example n. 1869 = (45.0, 402.0): cluster 3\n", + "Example n. 1870 = (45.0, 401.0): cluster 3\n", + "Example n. 1871 = (45.0, 400.0): cluster 3\n", + "Example n. 1872 = (45.0, 399.0): cluster 3\n", + "Example n. 1873 = (45.0, 398.0): cluster 3\n", + "Example n. 1874 = (46.0, 502.0): cluster 0\n", + "Example n. 1875 = (46.0, 501.0): cluster 0\n", + "Example n. 1876 = (46.0, 500.0): cluster 0\n", + "Example n. 1877 = (46.0, 499.0): cluster 0\n", + "Example n. 1878 = (46.0, 498.0): cluster 0\n", + "Example n. 1879 = (46.0, 497.0): cluster 0\n", + "Example n. 1880 = (46.0, 496.0): cluster 0\n", + "Example n. 1881 = (46.0, 495.0): cluster 0\n", + "Example n. 1882 = (46.0, 494.0): cluster 0\n", + "Example n. 1883 = (46.0, 493.0): cluster 0\n", + "Example n. 1884 = (46.0, 492.0): cluster 0\n", + "Example n. 1885 = (46.0, 491.0): cluster 0\n", + "Example n. 1886 = (46.0, 490.0): cluster 0\n", + "Example n. 1887 = (46.0, 489.0): cluster 0\n", + "Example n. 1888 = (46.0, 488.0): cluster 0\n", + "Example n. 1889 = (46.0, 487.0): cluster 0\n", + "Example n. 1890 = (46.0, 475.0): cluster 1\n", + "Example n. 1891 = (46.0, 474.0): cluster 1\n", + "Example n. 1892 = (46.0, 473.0): cluster 1\n", + "Example n. 1893 = (46.0, 472.0): cluster 1\n", + "Example n. 1894 = (46.0, 471.0): cluster 1\n", + "Example n. 1895 = (46.0, 470.0): cluster 1\n", + "Example n. 1896 = (46.0, 469.0): cluster 1\n", + "Example n. 1897 = (46.0, 468.0): cluster 1\n", + "Example n. 1898 = (46.0, 463.0): cluster 1\n", + "Example n. 1899 = (46.0, 462.0): cluster 1\n", + "Example n. 1900 = (46.0, 461.0): cluster 1\n", + "Example n. 1901 = (46.0, 460.0): cluster 1\n", + "Example n. 1902 = (46.0, 459.0): cluster 1\n", + "Example n. 1903 = (46.0, 458.0): cluster 1\n", + "Example n. 1904 = (46.0, 457.0): cluster 1\n", + "Example n. 1905 = (46.0, 456.0): cluster 1\n", + "Example n. 1906 = (46.0, 455.0): cluster 1\n", + "Example n. 1907 = (46.0, 454.0): cluster 1\n", + "Example n. 1908 = (46.0, 453.0): cluster 1\n", + "Example n. 1909 = (46.0, 452.0): cluster 1\n", + "Example n. 1910 = (46.0, 451.0): cluster 1\n", + "Example n. 1911 = (46.0, 450.0): cluster 1\n", + "Example n. 1912 = (46.0, 449.0): cluster 1\n", + "Example n. 1913 = (46.0, 448.0): cluster 1\n", + "Example n. 1914 = (46.0, 432.0): cluster 3\n", + "Example n. 1915 = (46.0, 431.0): cluster 3\n", + "Example n. 1916 = (46.0, 430.0): cluster 3\n", + "Example n. 1917 = (46.0, 429.0): cluster 3\n", + "Example n. 1918 = (46.0, 428.0): cluster 3\n", + "Example n. 1919 = (46.0, 427.0): cluster 3\n", + "Example n. 1920 = (46.0, 426.0): cluster 3\n", + "Example n. 1921 = (46.0, 425.0): cluster 3\n", + "Example n. 1922 = (46.0, 424.0): cluster 3\n", + "Example n. 1923 = (46.0, 423.0): cluster 3\n", + "Example n. 1924 = (46.0, 422.0): cluster 3\n", + "Example n. 1925 = (46.0, 421.0): cluster 3\n", + "Example n. 1926 = (46.0, 420.0): cluster 3\n", + "Example n. 1927 = (46.0, 419.0): cluster 3\n", + "Example n. 1928 = (46.0, 418.0): cluster 3\n", + "Example n. 1929 = (46.0, 417.0): cluster 3\n", + "Example n. 1930 = (46.0, 416.0): cluster 3\n", + "Example n. 1931 = (46.0, 415.0): cluster 3\n", + "Example n. 1932 = (46.0, 414.0): cluster 3\n", + "Example n. 1933 = (46.0, 409.0): cluster 3\n", + "Example n. 1934 = (46.0, 408.0): cluster 3\n", + "Example n. 1935 = (46.0, 407.0): cluster 3\n", + "Example n. 1936 = (46.0, 406.0): cluster 3\n", + "Example n. 1937 = (46.0, 405.0): cluster 3\n", + "Example n. 1938 = (46.0, 404.0): cluster 3\n", + "Example n. 1939 = (46.0, 403.0): cluster 3\n", + "Example n. 1940 = (46.0, 402.0): cluster 3\n", + "Example n. 1941 = (46.0, 401.0): cluster 3\n", + "Example n. 1942 = (46.0, 400.0): cluster 3\n", + "Example n. 1943 = (46.0, 399.0): cluster 3\n", + "Example n. 1944 = (46.0, 398.0): cluster 3\n", + "Example n. 1945 = (47.0, 502.0): cluster 0\n", + "Example n. 1946 = (47.0, 501.0): cluster 0\n", + "Example n. 1947 = (47.0, 500.0): cluster 0\n", + "Example n. 1948 = (47.0, 499.0): cluster 0\n", + "Example n. 1949 = (47.0, 498.0): cluster 0\n", + "Example n. 1950 = (47.0, 497.0): cluster 0\n", + "Example n. 1951 = (47.0, 496.0): cluster 0\n", + "Example n. 1952 = (47.0, 495.0): cluster 0\n", + "Example n. 1953 = (47.0, 494.0): cluster 0\n", + "Example n. 1954 = (47.0, 493.0): cluster 0\n", + "Example n. 1955 = (47.0, 492.0): cluster 0\n", + "Example n. 1956 = (47.0, 491.0): cluster 0\n", + "Example n. 1957 = (47.0, 490.0): cluster 0\n", + "Example n. 1958 = (47.0, 489.0): cluster 0\n", + "Example n. 1959 = (47.0, 488.0): cluster 0\n", + "Example n. 1960 = (47.0, 487.0): cluster 0\n", + "Example n. 1961 = (47.0, 486.0): cluster 0\n", + "Example n. 1962 = (47.0, 485.0): cluster 0\n", + "Example n. 1963 = (47.0, 484.0): cluster 0\n", + "Example n. 1964 = (47.0, 483.0): cluster 0\n", + "Example n. 1965 = (47.0, 475.0): cluster 1\n", + "Example n. 1966 = (47.0, 474.0): cluster 1\n", + "Example n. 1967 = (47.0, 473.0): cluster 1\n", + "Example n. 1968 = (47.0, 472.0): cluster 1\n", + "Example n. 1969 = (47.0, 471.0): cluster 1\n", + "Example n. 1970 = (47.0, 470.0): cluster 1\n", + "Example n. 1971 = (47.0, 469.0): cluster 1\n", + "Example n. 1972 = (47.0, 468.0): cluster 1\n", + "Example n. 1973 = (47.0, 463.0): cluster 1\n", + "Example n. 1974 = (47.0, 462.0): cluster 1\n", + "Example n. 1975 = (47.0, 461.0): cluster 1\n", + "Example n. 1976 = (47.0, 460.0): cluster 1\n", + "Example n. 1977 = (47.0, 459.0): cluster 1\n", + "Example n. 1978 = (47.0, 458.0): cluster 1\n", + "Example n. 1979 = (47.0, 457.0): cluster 1\n", + "Example n. 1980 = (47.0, 456.0): cluster 1\n", + "Example n. 1981 = (47.0, 455.0): cluster 1\n", + "Example n. 1982 = (47.0, 454.0): cluster 1\n", + "Example n. 1983 = (47.0, 453.0): cluster 1\n", + "Example n. 1984 = (47.0, 452.0): cluster 1\n", + "Example n. 1985 = (47.0, 451.0): cluster 1\n", + "Example n. 1986 = (47.0, 450.0): cluster 1\n", + "Example n. 1987 = (47.0, 449.0): cluster 1\n", + "Example n. 1988 = (47.0, 448.0): cluster 1\n", + "Example n. 1989 = (47.0, 436.0): cluster 3\n", + "Example n. 1990 = (47.0, 435.0): cluster 3\n", + "Example n. 1991 = (47.0, 434.0): cluster 3\n", + "Example n. 1992 = (47.0, 433.0): cluster 3\n", + "Example n. 1993 = (47.0, 432.0): cluster 3\n", + "Example n. 1994 = (47.0, 431.0): cluster 3\n", + "Example n. 1995 = (47.0, 430.0): cluster 3\n", + "Example n. 1996 = (47.0, 429.0): cluster 3\n", + "Example n. 1997 = (47.0, 428.0): cluster 3\n", + "Example n. 1998 = (47.0, 427.0): cluster 3\n", + "Example n. 1999 = (47.0, 426.0): cluster 3\n", + "Example n. 2000 = (47.0, 425.0): cluster 3\n", + "Example n. 2001 = (47.0, 424.0): cluster 3\n", + "Example n. 2002 = (47.0, 423.0): cluster 3\n", + "Example n. 2003 = (47.0, 422.0): cluster 3\n", + "Example n. 2004 = (47.0, 421.0): cluster 3\n", + "Example n. 2005 = (47.0, 420.0): cluster 3\n", + "Example n. 2006 = (47.0, 419.0): cluster 3\n", + "Example n. 2007 = (47.0, 418.0): cluster 3\n", + "Example n. 2008 = (47.0, 417.0): cluster 3\n", + "Example n. 2009 = (47.0, 416.0): cluster 3\n", + "Example n. 2010 = (47.0, 415.0): cluster 3\n", + "Example n. 2011 = (47.0, 414.0): cluster 3\n", + "Example n. 2012 = (47.0, 409.0): cluster 3\n", + "Example n. 2013 = (47.0, 408.0): cluster 3\n", + "Example n. 2014 = (47.0, 407.0): cluster 3\n", + "Example n. 2015 = (47.0, 406.0): cluster 3\n", + "Example n. 2016 = (47.0, 405.0): cluster 3\n", + "Example n. 2017 = (47.0, 404.0): cluster 3\n", + "Example n. 2018 = (47.0, 403.0): cluster 3\n", + "Example n. 2019 = (47.0, 402.0): cluster 3\n", + "Example n. 2020 = (47.0, 401.0): cluster 3\n", + "Example n. 2021 = (47.0, 400.0): cluster 3\n", + "Example n. 2022 = (47.0, 399.0): cluster 3\n", + "Example n. 2023 = (47.0, 398.0): cluster 3\n", + "Example n. 2024 = (48.0, 502.0): cluster 0\n", + "Example n. 2025 = (48.0, 501.0): cluster 0\n", + "Example n. 2026 = (48.0, 500.0): cluster 0\n", + "Example n. 2027 = (48.0, 499.0): cluster 0\n", + "Example n. 2028 = (48.0, 498.0): cluster 0\n", + "Example n. 2029 = (48.0, 497.0): cluster 0\n", + "Example n. 2030 = (48.0, 496.0): cluster 0\n", + "Example n. 2031 = (48.0, 495.0): cluster 0\n", + "Example n. 2032 = (48.0, 494.0): cluster 0\n", + "Example n. 2033 = (48.0, 493.0): cluster 0\n", + "Example n. 2034 = (48.0, 492.0): cluster 0\n", + "Example n. 2035 = (48.0, 491.0): cluster 0\n", + "Example n. 2036 = (48.0, 490.0): cluster 0\n", + "Example n. 2037 = (48.0, 489.0): cluster 0\n", + "Example n. 2038 = (48.0, 488.0): cluster 0\n", + "Example n. 2039 = (48.0, 487.0): cluster 0\n", + "Example n. 2040 = (48.0, 486.0): cluster 0\n", + "Example n. 2041 = (48.0, 485.0): cluster 0\n", + "Example n. 2042 = (48.0, 484.0): cluster 0\n", + "Example n. 2043 = (48.0, 483.0): cluster 0\n", + "Example n. 2044 = (48.0, 463.0): cluster 1\n", + "Example n. 2045 = (48.0, 462.0): cluster 1\n", + "Example n. 2046 = (48.0, 461.0): cluster 1\n", + "Example n. 2047 = (48.0, 460.0): cluster 1\n", + "Example n. 2048 = (48.0, 459.0): cluster 1\n", + "Example n. 2049 = (48.0, 458.0): cluster 1\n", + "Example n. 2050 = (48.0, 457.0): cluster 1\n", + "Example n. 2051 = (48.0, 456.0): cluster 1\n", + "Example n. 2052 = (48.0, 455.0): cluster 1\n", + "Example n. 2053 = (48.0, 454.0): cluster 1\n", + "Example n. 2054 = (48.0, 453.0): cluster 1\n", + "Example n. 2055 = (48.0, 452.0): cluster 1\n", + "Example n. 2056 = (48.0, 451.0): cluster 1\n", + "Example n. 2057 = (48.0, 450.0): cluster 1\n", + "Example n. 2058 = (48.0, 449.0): cluster 1\n", + "Example n. 2059 = (48.0, 448.0): cluster 1\n", + "Example n. 2060 = (48.0, 436.0): cluster 3\n", + "Example n. 2061 = (48.0, 435.0): cluster 3\n", + "Example n. 2062 = (48.0, 434.0): cluster 3\n", + "Example n. 2063 = (48.0, 433.0): cluster 3\n", + "Example n. 2064 = (48.0, 432.0): cluster 3\n", + "Example n. 2065 = (48.0, 431.0): cluster 3\n", + "Example n. 2066 = (48.0, 430.0): cluster 3\n", + "Example n. 2067 = (48.0, 429.0): cluster 3\n", + "Example n. 2068 = (48.0, 428.0): cluster 3\n", + "Example n. 2069 = (48.0, 427.0): cluster 3\n", + "Example n. 2070 = (48.0, 426.0): cluster 3\n", + "Example n. 2071 = (48.0, 425.0): cluster 3\n", + "Example n. 2072 = (48.0, 424.0): cluster 3\n", + "Example n. 2073 = (48.0, 423.0): cluster 3\n", + "Example n. 2074 = (48.0, 422.0): cluster 3\n", + "Example n. 2075 = (48.0, 421.0): cluster 3\n", + "Example n. 2076 = (48.0, 420.0): cluster 3\n", + "Example n. 2077 = (48.0, 419.0): cluster 3\n", + "Example n. 2078 = (48.0, 418.0): cluster 3\n", + "Example n. 2079 = (48.0, 417.0): cluster 3\n", + "Example n. 2080 = (48.0, 409.0): cluster 3\n", + "Example n. 2081 = (48.0, 408.0): cluster 3\n", + "Example n. 2082 = (48.0, 407.0): cluster 3\n", + "Example n. 2083 = (48.0, 406.0): cluster 3\n", + "Example n. 2084 = (48.0, 405.0): cluster 3\n", + "Example n. 2085 = (48.0, 404.0): cluster 3\n", + "Example n. 2086 = (48.0, 403.0): cluster 3\n", + "Example n. 2087 = (48.0, 402.0): cluster 3\n", + "Example n. 2088 = (48.0, 401.0): cluster 3\n", + "Example n. 2089 = (48.0, 400.0): cluster 3\n", + "Example n. 2090 = (48.0, 399.0): cluster 3\n", + "Example n. 2091 = (48.0, 398.0): cluster 3\n", + "Example n. 2092 = (49.0, 502.0): cluster 0\n", + "Example n. 2093 = (49.0, 501.0): cluster 0\n", + "Example n. 2094 = (49.0, 500.0): cluster 0\n", + "Example n. 2095 = (49.0, 499.0): cluster 0\n", + "Example n. 2096 = (49.0, 498.0): cluster 0\n", + "Example n. 2097 = (49.0, 497.0): cluster 0\n", + "Example n. 2098 = (49.0, 496.0): cluster 0\n", + "Example n. 2099 = (49.0, 495.0): cluster 0\n", + "Example n. 2100 = (49.0, 494.0): cluster 0\n", + "Example n. 2101 = (49.0, 493.0): cluster 0\n", + "Example n. 2102 = (49.0, 492.0): cluster 0\n", + "Example n. 2103 = (49.0, 491.0): cluster 0\n", + "Example n. 2104 = (49.0, 490.0): cluster 0\n", + "Example n. 2105 = (49.0, 489.0): cluster 0\n", + "Example n. 2106 = (49.0, 488.0): cluster 0\n", + "Example n. 2107 = (49.0, 487.0): cluster 0\n", + "Example n. 2108 = (49.0, 486.0): cluster 0\n", + "Example n. 2109 = (49.0, 485.0): cluster 1\n", + "Example n. 2110 = (49.0, 484.0): cluster 1\n", + "Example n. 2111 = (49.0, 483.0): cluster 1\n", + "Example n. 2112 = (49.0, 463.0): cluster 1\n", + "Example n. 2113 = (49.0, 462.0): cluster 1\n", + "Example n. 2114 = (49.0, 461.0): cluster 1\n", + "Example n. 2115 = (49.0, 460.0): cluster 1\n", + "Example n. 2116 = (49.0, 459.0): cluster 1\n", + "Example n. 2117 = (49.0, 458.0): cluster 1\n", + "Example n. 2118 = (49.0, 457.0): cluster 1\n", + "Example n. 2119 = (49.0, 456.0): cluster 1\n", + "Example n. 2120 = (49.0, 455.0): cluster 1\n", + "Example n. 2121 = (49.0, 454.0): cluster 1\n", + "Example n. 2122 = (49.0, 453.0): cluster 1\n", + "Example n. 2123 = (49.0, 452.0): cluster 1\n", + "Example n. 2124 = (49.0, 451.0): cluster 1\n", + "Example n. 2125 = (49.0, 450.0): cluster 1\n", + "Example n. 2126 = (49.0, 449.0): cluster 1\n", + "Example n. 2127 = (49.0, 448.0): cluster 1\n", + "Example n. 2128 = (49.0, 436.0): cluster 3\n", + "Example n. 2129 = (49.0, 435.0): cluster 3\n", + "Example n. 2130 = (49.0, 434.0): cluster 3\n", + "Example n. 2131 = (49.0, 433.0): cluster 3\n", + "Example n. 2132 = (49.0, 432.0): cluster 3\n", + "Example n. 2133 = (49.0, 431.0): cluster 3\n", + "Example n. 2134 = (49.0, 430.0): cluster 3\n", + "Example n. 2135 = (49.0, 429.0): cluster 3\n", + "Example n. 2136 = (49.0, 428.0): cluster 3\n", + "Example n. 2137 = (49.0, 427.0): cluster 3\n", + "Example n. 2138 = (49.0, 426.0): cluster 3\n", + "Example n. 2139 = (49.0, 425.0): cluster 3\n", + "Example n. 2140 = (49.0, 424.0): cluster 3\n", + "Example n. 2141 = (49.0, 423.0): cluster 3\n", + "Example n. 2142 = (49.0, 422.0): cluster 3\n", + "Example n. 2143 = (49.0, 421.0): cluster 3\n", + "Example n. 2144 = (49.0, 420.0): cluster 3\n", + "Example n. 2145 = (49.0, 419.0): cluster 3\n", + "Example n. 2146 = (49.0, 418.0): cluster 3\n", + "Example n. 2147 = (49.0, 417.0): cluster 3\n", + "Example n. 2148 = (49.0, 409.0): cluster 3\n", + "Example n. 2149 = (49.0, 408.0): cluster 3\n", + "Example n. 2150 = (49.0, 407.0): cluster 3\n", + "Example n. 2151 = (49.0, 406.0): cluster 3\n", + "Example n. 2152 = (49.0, 405.0): cluster 3\n", + "Example n. 2153 = (49.0, 404.0): cluster 3\n", + "Example n. 2154 = (49.0, 403.0): cluster 3\n", + "Example n. 2155 = (49.0, 402.0): cluster 3\n", + "Example n. 2156 = (49.0, 401.0): cluster 3\n", + "Example n. 2157 = (49.0, 400.0): cluster 3\n", + "Example n. 2158 = (49.0, 399.0): cluster 3\n", + "Example n. 2159 = (49.0, 398.0): cluster 3\n", + "Example n. 2160 = (50.0, 502.0): cluster 0\n", + "Example n. 2161 = (50.0, 501.0): cluster 0\n", + "Example n. 2162 = (50.0, 500.0): cluster 0\n", + "Example n. 2163 = (50.0, 499.0): cluster 0\n", + "Example n. 2164 = (50.0, 498.0): cluster 0\n", + "Example n. 2165 = (50.0, 497.0): cluster 0\n", + "Example n. 2166 = (50.0, 496.0): cluster 0\n", + "Example n. 2167 = (50.0, 495.0): cluster 0\n", + "Example n. 2168 = (50.0, 494.0): cluster 0\n", + "Example n. 2169 = (50.0, 493.0): cluster 0\n", + "Example n. 2170 = (50.0, 492.0): cluster 0\n", + "Example n. 2171 = (50.0, 491.0): cluster 0\n", + "Example n. 2172 = (50.0, 490.0): cluster 0\n", + "Example n. 2173 = (50.0, 489.0): cluster 0\n", + "Example n. 2174 = (50.0, 488.0): cluster 1\n", + "Example n. 2175 = (50.0, 487.0): cluster 1\n", + "Example n. 2176 = (50.0, 486.0): cluster 1\n", + "Example n. 2177 = (50.0, 485.0): cluster 1\n", + "Example n. 2178 = (50.0, 484.0): cluster 1\n", + "Example n. 2179 = (50.0, 483.0): cluster 1\n", + "Example n. 2180 = (50.0, 463.0): cluster 1\n", + "Example n. 2181 = (50.0, 462.0): cluster 1\n", + "Example n. 2182 = (50.0, 461.0): cluster 1\n", + "Example n. 2183 = (50.0, 460.0): cluster 1\n", + "Example n. 2184 = (50.0, 459.0): cluster 1\n", + "Example n. 2185 = (50.0, 458.0): cluster 1\n", + "Example n. 2186 = (50.0, 457.0): cluster 1\n", + "Example n. 2187 = (50.0, 456.0): cluster 1\n", + "Example n. 2188 = (50.0, 455.0): cluster 1\n", + "Example n. 2189 = (50.0, 454.0): cluster 1\n", + "Example n. 2190 = (50.0, 453.0): cluster 1\n", + "Example n. 2191 = (50.0, 452.0): cluster 1\n", + "Example n. 2192 = (50.0, 451.0): cluster 1\n", + "Example n. 2193 = (50.0, 450.0): cluster 1\n", + "Example n. 2194 = (50.0, 449.0): cluster 1\n", + "Example n. 2195 = (50.0, 448.0): cluster 1\n", + "Example n. 2196 = (50.0, 436.0): cluster 3\n", + "Example n. 2197 = (50.0, 435.0): cluster 3\n", + "Example n. 2198 = (50.0, 434.0): cluster 3\n", + "Example n. 2199 = (50.0, 433.0): cluster 3\n", + "Example n. 2200 = (50.0, 432.0): cluster 3\n", + "Example n. 2201 = (50.0, 431.0): cluster 3\n", + "Example n. 2202 = (50.0, 430.0): cluster 3\n", + "Example n. 2203 = (50.0, 429.0): cluster 3\n", + "Example n. 2204 = (50.0, 428.0): cluster 3\n", + "Example n. 2205 = (50.0, 427.0): cluster 3\n", + "Example n. 2206 = (50.0, 426.0): cluster 3\n", + "Example n. 2207 = (50.0, 425.0): cluster 3\n", + "Example n. 2208 = (50.0, 424.0): cluster 3\n", + "Example n. 2209 = (50.0, 423.0): cluster 3\n", + "Example n. 2210 = (50.0, 422.0): cluster 3\n", + "Example n. 2211 = (50.0, 421.0): cluster 3\n", + "Example n. 2212 = (50.0, 420.0): cluster 3\n", + "Example n. 2213 = (50.0, 419.0): cluster 3\n", + "Example n. 2214 = (50.0, 418.0): cluster 3\n", + "Example n. 2215 = (50.0, 417.0): cluster 3\n", + "Example n. 2216 = (50.0, 409.0): cluster 3\n", + "Example n. 2217 = (50.0, 408.0): cluster 3\n", + "Example n. 2218 = (50.0, 407.0): cluster 3\n", + "Example n. 2219 = (50.0, 406.0): cluster 3\n", + "Example n. 2220 = (50.0, 405.0): cluster 3\n", + "Example n. 2221 = (50.0, 404.0): cluster 3\n", + "Example n. 2222 = (50.0, 403.0): cluster 3\n", + "Example n. 2223 = (50.0, 402.0): cluster 3\n", + "Example n. 2224 = (50.0, 401.0): cluster 3\n", + "Example n. 2225 = (50.0, 400.0): cluster 3\n", + "Example n. 2226 = (50.0, 399.0): cluster 3\n", + "Example n. 2227 = (50.0, 398.0): cluster 3\n", + "Example n. 2228 = (51.0, 502.0): cluster 0\n", + "Example n. 2229 = (51.0, 501.0): cluster 0\n", + "Example n. 2230 = (51.0, 500.0): cluster 0\n", + "Example n. 2231 = (51.0, 499.0): cluster 0\n", + "Example n. 2232 = (51.0, 498.0): cluster 0\n", + "Example n. 2233 = (51.0, 497.0): cluster 0\n", + "Example n. 2234 = (51.0, 496.0): cluster 0\n", + "Example n. 2235 = (51.0, 495.0): cluster 0\n", + "Example n. 2236 = (51.0, 494.0): cluster 0\n", + "Example n. 2237 = (51.0, 493.0): cluster 0\n", + "Example n. 2238 = (51.0, 492.0): cluster 0\n", + "Example n. 2239 = (51.0, 491.0): cluster 0\n", + "Example n. 2240 = (51.0, 490.0): cluster 1\n", + "Example n. 2241 = (51.0, 489.0): cluster 1\n", + "Example n. 2242 = (51.0, 488.0): cluster 1\n", + "Example n. 2243 = (51.0, 487.0): cluster 1\n", + "Example n. 2244 = (51.0, 486.0): cluster 1\n", + "Example n. 2245 = (51.0, 485.0): cluster 1\n", + "Example n. 2246 = (51.0, 484.0): cluster 1\n", + "Example n. 2247 = (51.0, 483.0): cluster 1\n", + "Example n. 2248 = (51.0, 482.0): cluster 1\n", + "Example n. 2249 = (51.0, 481.0): cluster 1\n", + "Example n. 2250 = (51.0, 479.0): cluster 1\n", + "Example n. 2251 = (51.0, 459.0): cluster 1\n", + "Example n. 2252 = (51.0, 458.0): cluster 1\n", + "Example n. 2253 = (51.0, 457.0): cluster 1\n", + "Example n. 2254 = (51.0, 456.0): cluster 1\n", + "Example n. 2255 = (51.0, 455.0): cluster 1\n", + "Example n. 2256 = (51.0, 454.0): cluster 1\n", + "Example n. 2257 = (51.0, 453.0): cluster 1\n", + "Example n. 2258 = (51.0, 452.0): cluster 1\n", + "Example n. 2259 = (51.0, 440.0): cluster 3\n", + "Example n. 2260 = (51.0, 439.0): cluster 3\n", + "Example n. 2261 = (51.0, 438.0): cluster 3\n", + "Example n. 2262 = (51.0, 437.0): cluster 3\n", + "Example n. 2263 = (51.0, 436.0): cluster 3\n", + "Example n. 2264 = (51.0, 435.0): cluster 3\n", + "Example n. 2265 = (51.0, 434.0): cluster 3\n", + "Example n. 2266 = (51.0, 433.0): cluster 3\n", + "Example n. 2267 = (51.0, 432.0): cluster 3\n", + "Example n. 2268 = (51.0, 431.0): cluster 3\n", + "Example n. 2269 = (51.0, 430.0): cluster 3\n", + "Example n. 2270 = (51.0, 429.0): cluster 3\n", + "Example n. 2271 = (51.0, 428.0): cluster 3\n", + "Example n. 2272 = (51.0, 427.0): cluster 3\n", + "Example n. 2273 = (51.0, 426.0): cluster 3\n", + "Example n. 2274 = (51.0, 425.0): cluster 3\n", + "Example n. 2275 = (51.0, 424.0): cluster 3\n", + "Example n. 2276 = (51.0, 423.0): cluster 3\n", + "Example n. 2277 = (51.0, 422.0): cluster 3\n", + "Example n. 2278 = (51.0, 421.0): cluster 3\n", + "Example n. 2279 = (51.0, 409.0): cluster 3\n", + "Example n. 2280 = (51.0, 408.0): cluster 3\n", + "Example n. 2281 = (51.0, 407.0): cluster 3\n", + "Example n. 2282 = (51.0, 406.0): cluster 3\n", + "Example n. 2283 = (51.0, 405.0): cluster 3\n", + "Example n. 2284 = (51.0, 404.0): cluster 3\n", + "Example n. 2285 = (51.0, 403.0): cluster 3\n", + "Example n. 2286 = (51.0, 402.0): cluster 3\n", + "Example n. 2287 = (51.0, 401.0): cluster 3\n", + "Example n. 2288 = (51.0, 400.0): cluster 3\n", + "Example n. 2289 = (51.0, 399.0): cluster 3\n", + "Example n. 2290 = (51.0, 398.0): cluster 3\n", + "Example n. 2291 = (52.0, 502.0): cluster 0\n", + "Example n. 2292 = (52.0, 501.0): cluster 0\n", + "Example n. 2293 = (52.0, 500.0): cluster 0\n", + "Example n. 2294 = (52.0, 499.0): cluster 0\n", + "Example n. 2295 = (52.0, 498.0): cluster 0\n", + "Example n. 2296 = (52.0, 497.0): cluster 0\n", + "Example n. 2297 = (52.0, 496.0): cluster 0\n", + "Example n. 2298 = (52.0, 495.0): cluster 0\n", + "Example n. 2299 = (52.0, 494.0): cluster 0\n", + "Example n. 2300 = (52.0, 493.0): cluster 1\n", + "Example n. 2301 = (52.0, 492.0): cluster 1\n", + "Example n. 2302 = (52.0, 491.0): cluster 1\n", + "Example n. 2303 = (52.0, 490.0): cluster 1\n", + "Example n. 2304 = (52.0, 489.0): cluster 1\n", + "Example n. 2305 = (52.0, 488.0): cluster 1\n", + "Example n. 2306 = (52.0, 487.0): cluster 1\n", + "Example n. 2307 = (52.0, 486.0): cluster 1\n", + "Example n. 2308 = (52.0, 485.0): cluster 1\n", + "Example n. 2309 = (52.0, 484.0): cluster 1\n", + "Example n. 2310 = (52.0, 483.0): cluster 1\n", + "Example n. 2311 = (52.0, 482.0): cluster 1\n", + "Example n. 2312 = (52.0, 481.0): cluster 1\n", + "Example n. 2313 = (52.0, 480.0): cluster 1\n", + "Example n. 2314 = (52.0, 459.0): cluster 1\n", + "Example n. 2315 = (52.0, 458.0): cluster 1\n", + "Example n. 2316 = (52.0, 457.0): cluster 1\n", + "Example n. 2317 = (52.0, 456.0): cluster 1\n", + "Example n. 2318 = (52.0, 455.0): cluster 1\n", + "Example n. 2319 = (52.0, 454.0): cluster 1\n", + "Example n. 2320 = (52.0, 453.0): cluster 1\n", + "Example n. 2321 = (52.0, 452.0): cluster 1\n", + "Example n. 2322 = (52.0, 440.0): cluster 3\n", + "Example n. 2323 = (52.0, 439.0): cluster 3\n", + "Example n. 2324 = (52.0, 438.0): cluster 3\n", + "Example n. 2325 = (52.0, 437.0): cluster 3\n", + "Example n. 2326 = (52.0, 436.0): cluster 3\n", + "Example n. 2327 = (52.0, 435.0): cluster 3\n", + "Example n. 2328 = (52.0, 434.0): cluster 3\n", + "Example n. 2329 = (52.0, 433.0): cluster 3\n", + "Example n. 2330 = (52.0, 432.0): cluster 3\n", + "Example n. 2331 = (52.0, 431.0): cluster 3\n", + "Example n. 2332 = (52.0, 430.0): cluster 3\n", + "Example n. 2333 = (52.0, 429.0): cluster 3\n", + "Example n. 2334 = (52.0, 428.0): cluster 3\n", + "Example n. 2335 = (52.0, 427.0): cluster 3\n", + "Example n. 2336 = (52.0, 426.0): cluster 3\n", + "Example n. 2337 = (52.0, 425.0): cluster 3\n", + "Example n. 2338 = (52.0, 424.0): cluster 3\n", + "Example n. 2339 = (52.0, 423.0): cluster 3\n", + "Example n. 2340 = (52.0, 422.0): cluster 3\n", + "Example n. 2341 = (52.0, 421.0): cluster 3\n", + "Example n. 2342 = (52.0, 409.0): cluster 3\n", + "Example n. 2343 = (52.0, 408.0): cluster 3\n", + "Example n. 2344 = (52.0, 407.0): cluster 3\n", + "Example n. 2345 = (52.0, 406.0): cluster 3\n", + "Example n. 2346 = (52.0, 405.0): cluster 3\n", + "Example n. 2347 = (52.0, 404.0): cluster 3\n", + "Example n. 2348 = (52.0, 403.0): cluster 3\n", + "Example n. 2349 = (52.0, 402.0): cluster 3\n", + "Example n. 2350 = (52.0, 401.0): cluster 3\n", + "Example n. 2351 = (52.0, 400.0): cluster 3\n", + "Example n. 2352 = (52.0, 399.0): cluster 3\n", + "Example n. 2353 = (52.0, 398.0): cluster 3\n", + "Example n. 2354 = (53.0, 502.0): cluster 0\n", + "Example n. 2355 = (53.0, 501.0): cluster 0\n", + "Example n. 2356 = (53.0, 500.0): cluster 0\n", + "Example n. 2357 = (53.0, 499.0): cluster 0\n", + "Example n. 2358 = (53.0, 498.0): cluster 0\n", + "Example n. 2359 = (53.0, 497.0): cluster 0\n", + "Example n. 2360 = (53.0, 496.0): cluster 0\n", + "Example n. 2361 = (53.0, 495.0): cluster 1\n", + "Example n. 2362 = (53.0, 494.0): cluster 1\n", + "Example n. 2363 = (53.0, 493.0): cluster 1\n", + "Example n. 2364 = (53.0, 492.0): cluster 1\n", + "Example n. 2365 = (53.0, 491.0): cluster 1\n", + "Example n. 2366 = (53.0, 490.0): cluster 1\n", + "Example n. 2367 = (53.0, 489.0): cluster 1\n", + "Example n. 2368 = (53.0, 488.0): cluster 1\n", + "Example n. 2369 = (53.0, 487.0): cluster 1\n", + "Example n. 2370 = (53.0, 486.0): cluster 1\n", + "Example n. 2371 = (53.0, 485.0): cluster 1\n", + "Example n. 2372 = (53.0, 484.0): cluster 1\n", + "Example n. 2373 = (53.0, 483.0): cluster 1\n", + "Example n. 2374 = (53.0, 481.0): cluster 1\n", + "Example n. 2375 = (53.0, 459.0): cluster 1\n", + "Example n. 2376 = (53.0, 458.0): cluster 1\n", + "Example n. 2377 = (53.0, 457.0): cluster 1\n", + "Example n. 2378 = (53.0, 456.0): cluster 1\n", + "Example n. 2379 = (53.0, 455.0): cluster 1\n", + "Example n. 2380 = (53.0, 454.0): cluster 1\n", + "Example n. 2381 = (53.0, 453.0): cluster 1\n", + "Example n. 2382 = (53.0, 452.0): cluster 1\n", + "Example n. 2383 = (53.0, 440.0): cluster 3\n", + "Example n. 2384 = (53.0, 439.0): cluster 3\n", + "Example n. 2385 = (53.0, 438.0): cluster 3\n", + "Example n. 2386 = (53.0, 437.0): cluster 3\n", + "Example n. 2387 = (53.0, 436.0): cluster 3\n", + "Example n. 2388 = (53.0, 435.0): cluster 3\n", + "Example n. 2389 = (53.0, 434.0): cluster 3\n", + "Example n. 2390 = (53.0, 433.0): cluster 3\n", + "Example n. 2391 = (53.0, 432.0): cluster 3\n", + "Example n. 2392 = (53.0, 431.0): cluster 3\n", + "Example n. 2393 = (53.0, 430.0): cluster 3\n", + "Example n. 2394 = (53.0, 429.0): cluster 3\n", + "Example n. 2395 = (53.0, 428.0): cluster 3\n", + "Example n. 2396 = (53.0, 427.0): cluster 3\n", + "Example n. 2397 = (53.0, 426.0): cluster 3\n", + "Example n. 2398 = (53.0, 425.0): cluster 3\n", + "Example n. 2399 = (53.0, 424.0): cluster 3\n", + "Example n. 2400 = (53.0, 423.0): cluster 3\n", + "Example n. 2401 = (53.0, 422.0): cluster 3\n", + "Example n. 2402 = (53.0, 421.0): cluster 3\n", + "Example n. 2403 = (53.0, 409.0): cluster 3\n", + "Example n. 2404 = (53.0, 408.0): cluster 3\n", + "Example n. 2405 = (53.0, 407.0): cluster 3\n", + "Example n. 2406 = (53.0, 406.0): cluster 3\n", + "Example n. 2407 = (53.0, 405.0): cluster 3\n", + "Example n. 2408 = (53.0, 404.0): cluster 3\n", + "Example n. 2409 = (53.0, 403.0): cluster 3\n", + "Example n. 2410 = (53.0, 402.0): cluster 3\n", + "Example n. 2411 = (53.0, 401.0): cluster 3\n", + "Example n. 2412 = (53.0, 400.0): cluster 3\n", + "Example n. 2413 = (53.0, 399.0): cluster 3\n", + "Example n. 2414 = (53.0, 398.0): cluster 3\n", + "Example n. 2415 = (54.0, 502.0): cluster 0\n", + "Example n. 2416 = (54.0, 501.0): cluster 0\n", + "Example n. 2417 = (54.0, 500.0): cluster 0\n", + "Example n. 2418 = (54.0, 499.0): cluster 0\n", + "Example n. 2419 = (54.0, 498.0): cluster 1\n", + "Example n. 2420 = (54.0, 497.0): cluster 1\n", + "Example n. 2421 = (54.0, 496.0): cluster 1\n", + "Example n. 2422 = (54.0, 495.0): cluster 1\n", + "Example n. 2423 = (54.0, 494.0): cluster 1\n", + "Example n. 2424 = (54.0, 493.0): cluster 1\n", + "Example n. 2425 = (54.0, 492.0): cluster 1\n", + "Example n. 2426 = (54.0, 491.0): cluster 1\n", + "Example n. 2427 = (54.0, 490.0): cluster 1\n", + "Example n. 2428 = (54.0, 489.0): cluster 1\n", + "Example n. 2429 = (54.0, 488.0): cluster 1\n", + "Example n. 2430 = (54.0, 487.0): cluster 1\n", + "Example n. 2431 = (54.0, 486.0): cluster 1\n", + "Example n. 2432 = (54.0, 485.0): cluster 1\n", + "Example n. 2433 = (54.0, 484.0): cluster 1\n", + "Example n. 2434 = (54.0, 483.0): cluster 1\n", + "Example n. 2435 = (54.0, 482.0): cluster 1\n", + "Example n. 2436 = (54.0, 481.0): cluster 1\n", + "Example n. 2437 = (54.0, 459.0): cluster 1\n", + "Example n. 2438 = (54.0, 458.0): cluster 1\n", + "Example n. 2439 = (54.0, 457.0): cluster 1\n", + "Example n. 2440 = (54.0, 456.0): cluster 1\n", + "Example n. 2441 = (54.0, 455.0): cluster 1\n", + "Example n. 2442 = (54.0, 454.0): cluster 1\n", + "Example n. 2443 = (54.0, 453.0): cluster 1\n", + "Example n. 2444 = (54.0, 452.0): cluster 1\n", + "Example n. 2445 = (54.0, 440.0): cluster 3\n", + "Example n. 2446 = (54.0, 439.0): cluster 3\n", + "Example n. 2447 = (54.0, 438.0): cluster 3\n", + "Example n. 2448 = (54.0, 437.0): cluster 3\n", + "Example n. 2449 = (54.0, 436.0): cluster 3\n", + "Example n. 2450 = (54.0, 435.0): cluster 3\n", + "Example n. 2451 = (54.0, 434.0): cluster 3\n", + "Example n. 2452 = (54.0, 433.0): cluster 3\n", + "Example n. 2453 = (54.0, 432.0): cluster 3\n", + "Example n. 2454 = (54.0, 431.0): cluster 3\n", + "Example n. 2455 = (54.0, 430.0): cluster 3\n", + "Example n. 2456 = (54.0, 429.0): cluster 3\n", + "Example n. 2457 = (54.0, 428.0): cluster 3\n", + "Example n. 2458 = (54.0, 427.0): cluster 3\n", + "Example n. 2459 = (54.0, 426.0): cluster 3\n", + "Example n. 2460 = (54.0, 425.0): cluster 3\n", + "Example n. 2461 = (54.0, 424.0): cluster 3\n", + "Example n. 2462 = (54.0, 423.0): cluster 3\n", + "Example n. 2463 = (54.0, 422.0): cluster 3\n", + "Example n. 2464 = (54.0, 421.0): cluster 3\n", + "Example n. 2465 = (54.0, 409.0): cluster 3\n", + "Example n. 2466 = (54.0, 408.0): cluster 3\n", + "Example n. 2467 = (54.0, 407.0): cluster 3\n", + "Example n. 2468 = (54.0, 406.0): cluster 3\n", + "Example n. 2469 = (54.0, 405.0): cluster 3\n", + "Example n. 2470 = (54.0, 404.0): cluster 3\n", + "Example n. 2471 = (54.0, 403.0): cluster 3\n", + "Example n. 2472 = (54.0, 402.0): cluster 3\n", + "Example n. 2473 = (54.0, 401.0): cluster 3\n", + "Example n. 2474 = (54.0, 400.0): cluster 3\n", + "Example n. 2475 = (54.0, 399.0): cluster 3\n", + "Example n. 2476 = (54.0, 398.0): cluster 3\n", + "Example n. 2477 = (55.0, 498.0): cluster 1\n", + "Example n. 2478 = (55.0, 497.0): cluster 1\n", + "Example n. 2479 = (55.0, 496.0): cluster 1\n", + "Example n. 2480 = (55.0, 495.0): cluster 1\n", + "Example n. 2481 = (55.0, 494.0): cluster 1\n", + "Example n. 2482 = (55.0, 493.0): cluster 1\n", + "Example n. 2483 = (55.0, 492.0): cluster 1\n", + "Example n. 2484 = (55.0, 491.0): cluster 1\n", + "Example n. 2485 = (55.0, 490.0): cluster 1\n", + "Example n. 2486 = (55.0, 489.0): cluster 1\n", + "Example n. 2487 = (55.0, 488.0): cluster 1\n", + "Example n. 2488 = (55.0, 487.0): cluster 1\n", + "Example n. 2489 = (55.0, 486.0): cluster 1\n", + "Example n. 2490 = (55.0, 485.0): cluster 1\n", + "Example n. 2491 = (55.0, 484.0): cluster 1\n", + "Example n. 2492 = (55.0, 483.0): cluster 1\n", + "Example n. 2493 = (55.0, 482.0): cluster 1\n", + "Example n. 2494 = (55.0, 481.0): cluster 1\n", + "Example n. 2495 = (55.0, 480.0): cluster 1\n", + "Example n. 2496 = (55.0, 479.0): cluster 1\n", + "Example n. 2497 = (55.0, 478.0): cluster 1\n", + "Example n. 2498 = (55.0, 477.0): cluster 1\n", + "Example n. 2499 = (55.0, 476.0): cluster 1\n", + "Example n. 2500 = (55.0, 475.0): cluster 1\n", + "Example n. 2501 = (55.0, 440.0): cluster 3\n", + "Example n. 2502 = (55.0, 439.0): cluster 3\n", + "Example n. 2503 = (55.0, 438.0): cluster 3\n", + "Example n. 2504 = (55.0, 437.0): cluster 3\n", + "Example n. 2505 = (55.0, 436.0): cluster 3\n", + "Example n. 2506 = (55.0, 435.0): cluster 3\n", + "Example n. 2507 = (55.0, 434.0): cluster 3\n", + "Example n. 2508 = (55.0, 433.0): cluster 3\n", + "Example n. 2509 = (55.0, 432.0): cluster 3\n", + "Example n. 2510 = (55.0, 431.0): cluster 3\n", + "Example n. 2511 = (55.0, 430.0): cluster 3\n", + "Example n. 2512 = (55.0, 429.0): cluster 3\n", + "Example n. 2513 = (55.0, 428.0): cluster 3\n", + "Example n. 2514 = (55.0, 427.0): cluster 3\n", + "Example n. 2515 = (55.0, 426.0): cluster 3\n", + "Example n. 2516 = (55.0, 425.0): cluster 3\n", + "Example n. 2517 = (55.0, 405.0): cluster 3\n", + "Example n. 2518 = (55.0, 404.0): cluster 3\n", + "Example n. 2519 = (55.0, 403.0): cluster 3\n", + "Example n. 2520 = (55.0, 402.0): cluster 3\n", + "Example n. 2521 = (55.0, 401.0): cluster 3\n", + "Example n. 2522 = (55.0, 400.0): cluster 3\n", + "Example n. 2523 = (55.0, 399.0): cluster 3\n", + "Example n. 2524 = (55.0, 398.0): cluster 3\n", + "Example n. 2525 = (55.0, 397.0): cluster 3\n", + "Example n. 2526 = (55.0, 396.0): cluster 3\n", + "Example n. 2527 = (55.0, 395.0): cluster 3\n", + "Example n. 2528 = (55.0, 394.0): cluster 3\n", + "Example n. 2529 = (56.0, 498.0): cluster 1\n", + "Example n. 2530 = (56.0, 497.0): cluster 1\n", + "Example n. 2531 = (56.0, 496.0): cluster 1\n", + "Example n. 2532 = (56.0, 495.0): cluster 1\n", + "Example n. 2533 = (56.0, 494.0): cluster 1\n", + "Example n. 2534 = (56.0, 493.0): cluster 1\n", + "Example n. 2535 = (56.0, 492.0): cluster 1\n", + "Example n. 2536 = (56.0, 491.0): cluster 1\n", + "Example n. 2537 = (56.0, 490.0): cluster 1\n", + "Example n. 2538 = (56.0, 489.0): cluster 1\n", + "Example n. 2539 = (56.0, 488.0): cluster 1\n", + "Example n. 2540 = (56.0, 487.0): cluster 1\n", + "Example n. 2541 = (56.0, 486.0): cluster 1\n", + "Example n. 2542 = (56.0, 485.0): cluster 1\n", + "Example n. 2543 = (56.0, 484.0): cluster 1\n", + "Example n. 2544 = (56.0, 483.0): cluster 1\n", + "Example n. 2545 = (56.0, 482.0): cluster 1\n", + "Example n. 2546 = (56.0, 481.0): cluster 1\n", + "Example n. 2547 = (56.0, 480.0): cluster 1\n", + "Example n. 2548 = (56.0, 479.0): cluster 1\n", + "Example n. 2549 = (56.0, 478.0): cluster 1\n", + "Example n. 2550 = (56.0, 477.0): cluster 1\n", + "Example n. 2551 = (56.0, 476.0): cluster 1\n", + "Example n. 2552 = (56.0, 475.0): cluster 1\n", + "Example n. 2553 = (56.0, 440.0): cluster 3\n", + "Example n. 2554 = (56.0, 439.0): cluster 3\n", + "Example n. 2555 = (56.0, 438.0): cluster 3\n", + "Example n. 2556 = (56.0, 437.0): cluster 3\n", + "Example n. 2557 = (56.0, 436.0): cluster 3\n", + "Example n. 2558 = (56.0, 435.0): cluster 3\n", + "Example n. 2559 = (56.0, 434.0): cluster 3\n", + "Example n. 2560 = (56.0, 433.0): cluster 3\n", + "Example n. 2561 = (56.0, 432.0): cluster 3\n", + "Example n. 2562 = (56.0, 431.0): cluster 3\n", + "Example n. 2563 = (56.0, 430.0): cluster 3\n", + "Example n. 2564 = (56.0, 429.0): cluster 3\n", + "Example n. 2565 = (56.0, 428.0): cluster 3\n", + "Example n. 2566 = (56.0, 427.0): cluster 3\n", + "Example n. 2567 = (56.0, 426.0): cluster 3\n", + "Example n. 2568 = (56.0, 425.0): cluster 3\n", + "Example n. 2569 = (56.0, 405.0): cluster 3\n", + "Example n. 2570 = (56.0, 404.0): cluster 3\n", + "Example n. 2571 = (56.0, 403.0): cluster 3\n", + "Example n. 2572 = (56.0, 402.0): cluster 3\n", + "Example n. 2573 = (56.0, 401.0): cluster 3\n", + "Example n. 2574 = (56.0, 400.0): cluster 3\n", + "Example n. 2575 = (56.0, 399.0): cluster 3\n", + "Example n. 2576 = (56.0, 398.0): cluster 3\n", + "Example n. 2577 = (56.0, 397.0): cluster 3\n", + "Example n. 2578 = (56.0, 396.0): cluster 3\n", + "Example n. 2579 = (56.0, 395.0): cluster 3\n", + "Example n. 2580 = (56.0, 394.0): cluster 3\n", + "Example n. 2581 = (57.0, 498.0): cluster 1\n", + "Example n. 2582 = (57.0, 497.0): cluster 1\n", + "Example n. 2583 = (57.0, 496.0): cluster 1\n", + "Example n. 2584 = (57.0, 495.0): cluster 1\n", + "Example n. 2585 = (57.0, 494.0): cluster 1\n", + "Example n. 2586 = (57.0, 493.0): cluster 1\n", + "Example n. 2587 = (57.0, 492.0): cluster 1\n", + "Example n. 2588 = (57.0, 491.0): cluster 1\n", + "Example n. 2589 = (57.0, 490.0): cluster 1\n", + "Example n. 2590 = (57.0, 489.0): cluster 1\n", + "Example n. 2591 = (57.0, 488.0): cluster 1\n", + "Example n. 2592 = (57.0, 487.0): cluster 1\n", + "Example n. 2593 = (57.0, 486.0): cluster 1\n", + "Example n. 2594 = (57.0, 485.0): cluster 1\n", + "Example n. 2595 = (57.0, 484.0): cluster 1\n", + "Example n. 2596 = (57.0, 483.0): cluster 1\n", + "Example n. 2597 = (57.0, 482.0): cluster 1\n", + "Example n. 2598 = (57.0, 481.0): cluster 1\n", + "Example n. 2599 = (57.0, 480.0): cluster 1\n", + "Example n. 2600 = (57.0, 479.0): cluster 1\n", + "Example n. 2601 = (57.0, 478.0): cluster 1\n", + "Example n. 2602 = (57.0, 477.0): cluster 1\n", + "Example n. 2603 = (57.0, 476.0): cluster 1\n", + "Example n. 2604 = (57.0, 475.0): cluster 1\n", + "Example n. 2605 = (57.0, 440.0): cluster 3\n", + "Example n. 2606 = (57.0, 439.0): cluster 3\n", + "Example n. 2607 = (57.0, 438.0): cluster 3\n", + "Example n. 2608 = (57.0, 437.0): cluster 3\n", + "Example n. 2609 = (57.0, 436.0): cluster 3\n", + "Example n. 2610 = (57.0, 435.0): cluster 3\n", + "Example n. 2611 = (57.0, 434.0): cluster 3\n", + "Example n. 2612 = (57.0, 433.0): cluster 3\n", + "Example n. 2613 = (57.0, 432.0): cluster 3\n", + "Example n. 2614 = (57.0, 431.0): cluster 3\n", + "Example n. 2615 = (57.0, 430.0): cluster 3\n", + "Example n. 2616 = (57.0, 429.0): cluster 3\n", + "Example n. 2617 = (57.0, 428.0): cluster 3\n", + "Example n. 2618 = (57.0, 427.0): cluster 3\n", + "Example n. 2619 = (57.0, 426.0): cluster 3\n", + "Example n. 2620 = (57.0, 425.0): cluster 3\n", + "Example n. 2621 = (57.0, 405.0): cluster 3\n", + "Example n. 2622 = (57.0, 404.0): cluster 3\n", + "Example n. 2623 = (57.0, 403.0): cluster 3\n", + "Example n. 2624 = (57.0, 402.0): cluster 3\n", + "Example n. 2625 = (57.0, 401.0): cluster 3\n", + "Example n. 2626 = (57.0, 400.0): cluster 3\n", + "Example n. 2627 = (57.0, 399.0): cluster 3\n", + "Example n. 2628 = (57.0, 398.0): cluster 3\n", + "Example n. 2629 = (57.0, 397.0): cluster 3\n", + "Example n. 2630 = (57.0, 396.0): cluster 3\n", + "Example n. 2631 = (57.0, 395.0): cluster 3\n", + "Example n. 2632 = (57.0, 394.0): cluster 3\n", + "Example n. 2633 = (58.0, 498.0): cluster 1\n", + "Example n. 2634 = (58.0, 497.0): cluster 1\n", + "Example n. 2635 = (58.0, 496.0): cluster 1\n", + "Example n. 2636 = (58.0, 495.0): cluster 1\n", + "Example n. 2637 = (58.0, 494.0): cluster 1\n", + "Example n. 2638 = (58.0, 493.0): cluster 1\n", + "Example n. 2639 = (58.0, 492.0): cluster 1\n", + "Example n. 2640 = (58.0, 491.0): cluster 1\n", + "Example n. 2641 = (58.0, 490.0): cluster 1\n", + "Example n. 2642 = (58.0, 489.0): cluster 1\n", + "Example n. 2643 = (58.0, 488.0): cluster 1\n", + "Example n. 2644 = (58.0, 487.0): cluster 1\n", + "Example n. 2645 = (58.0, 486.0): cluster 1\n", + "Example n. 2646 = (58.0, 485.0): cluster 1\n", + "Example n. 2647 = (58.0, 484.0): cluster 1\n", + "Example n. 2648 = (58.0, 483.0): cluster 1\n", + "Example n. 2649 = (58.0, 482.0): cluster 1\n", + "Example n. 2650 = (58.0, 481.0): cluster 1\n", + "Example n. 2651 = (58.0, 480.0): cluster 1\n", + "Example n. 2652 = (58.0, 479.0): cluster 1\n", + "Example n. 2653 = (58.0, 478.0): cluster 1\n", + "Example n. 2654 = (58.0, 477.0): cluster 1\n", + "Example n. 2655 = (58.0, 476.0): cluster 1\n", + "Example n. 2656 = (58.0, 475.0): cluster 1\n", + "Example n. 2657 = (58.0, 440.0): cluster 3\n", + "Example n. 2658 = (58.0, 439.0): cluster 3\n", + "Example n. 2659 = (58.0, 438.0): cluster 3\n", + "Example n. 2660 = (58.0, 437.0): cluster 3\n", + "Example n. 2661 = (58.0, 436.0): cluster 3\n", + "Example n. 2662 = (58.0, 435.0): cluster 3\n", + "Example n. 2663 = (58.0, 434.0): cluster 3\n", + "Example n. 2664 = (58.0, 433.0): cluster 3\n", + "Example n. 2665 = (58.0, 432.0): cluster 3\n", + "Example n. 2666 = (58.0, 431.0): cluster 3\n", + "Example n. 2667 = (58.0, 430.0): cluster 3\n", + "Example n. 2668 = (58.0, 429.0): cluster 3\n", + "Example n. 2669 = (58.0, 428.0): cluster 3\n", + "Example n. 2670 = (58.0, 427.0): cluster 3\n", + "Example n. 2671 = (58.0, 426.0): cluster 3\n", + "Example n. 2672 = (58.0, 425.0): cluster 3\n", + "Example n. 2673 = (58.0, 405.0): cluster 3\n", + "Example n. 2674 = (58.0, 404.0): cluster 3\n", + "Example n. 2675 = (58.0, 403.0): cluster 3\n", + "Example n. 2676 = (58.0, 402.0): cluster 3\n", + "Example n. 2677 = (58.0, 401.0): cluster 3\n", + "Example n. 2678 = (58.0, 400.0): cluster 3\n", + "Example n. 2679 = (58.0, 399.0): cluster 3\n", + "Example n. 2680 = (58.0, 398.0): cluster 3\n", + "Example n. 2681 = (58.0, 397.0): cluster 3\n", + "Example n. 2682 = (58.0, 396.0): cluster 3\n", + "Example n. 2683 = (58.0, 395.0): cluster 3\n", + "Example n. 2684 = (58.0, 394.0): cluster 3\n", + "Example n. 2685 = (59.0, 494.0): cluster 1\n", + "Example n. 2686 = (59.0, 493.0): cluster 1\n", + "Example n. 2687 = (59.0, 492.0): cluster 1\n", + "Example n. 2688 = (59.0, 491.0): cluster 1\n", + "Example n. 2689 = (59.0, 490.0): cluster 1\n", + "Example n. 2690 = (59.0, 489.0): cluster 1\n", + "Example n. 2691 = (59.0, 488.0): cluster 1\n", + "Example n. 2692 = (59.0, 487.0): cluster 1\n", + "Example n. 2693 = (59.0, 486.0): cluster 1\n", + "Example n. 2694 = (59.0, 485.0): cluster 1\n", + "Example n. 2695 = (59.0, 484.0): cluster 1\n", + "Example n. 2696 = (59.0, 483.0): cluster 1\n", + "Example n. 2697 = (59.0, 482.0): cluster 1\n", + "Example n. 2698 = (59.0, 481.0): cluster 1\n", + "Example n. 2699 = (59.0, 480.0): cluster 1\n", + "Example n. 2700 = (59.0, 479.0): cluster 1\n", + "Example n. 2701 = (59.0, 478.0): cluster 1\n", + "Example n. 2702 = (59.0, 477.0): cluster 1\n", + "Example n. 2703 = (59.0, 476.0): cluster 1\n", + "Example n. 2704 = (59.0, 475.0): cluster 1\n", + "Example n. 2705 = (59.0, 474.0): cluster 1\n", + "Example n. 2706 = (59.0, 473.0): cluster 1\n", + "Example n. 2707 = (59.0, 472.0): cluster 1\n", + "Example n. 2708 = (59.0, 471.0): cluster 1\n", + "Example n. 2709 = (59.0, 470.0): cluster 1\n", + "Example n. 2710 = (59.0, 469.0): cluster 1\n", + "Example n. 2711 = (59.0, 468.0): cluster 1\n", + "Example n. 2712 = (59.0, 448.0): cluster 1\n", + "Example n. 2713 = (59.0, 447.0): cluster 1\n", + "Example n. 2714 = (59.0, 446.0): cluster 1\n", + "Example n. 2715 = (59.0, 445.0): cluster 1\n", + "Example n. 2716 = (59.0, 444.0): cluster 1\n", + "Example n. 2717 = (59.0, 443.0): cluster 1\n", + "Example n. 2718 = (59.0, 442.0): cluster 1\n", + "Example n. 2719 = (59.0, 441.0): cluster 3\n", + "Example n. 2720 = (59.0, 440.0): cluster 3\n", + "Example n. 2721 = (59.0, 439.0): cluster 3\n", + "Example n. 2722 = (59.0, 438.0): cluster 3\n", + "Example n. 2723 = (59.0, 437.0): cluster 3\n", + "Example n. 2724 = (59.0, 436.0): cluster 3\n", + "Example n. 2725 = (59.0, 435.0): cluster 3\n", + "Example n. 2726 = (59.0, 434.0): cluster 3\n", + "Example n. 2727 = (59.0, 433.0): cluster 3\n", + "Example n. 2728 = (59.0, 432.0): cluster 3\n", + "Example n. 2729 = (59.0, 431.0): cluster 3\n", + "Example n. 2730 = (59.0, 430.0): cluster 3\n", + "Example n. 2731 = (59.0, 429.0): cluster 3\n", + "Example n. 2732 = (59.0, 425.0): cluster 3\n", + "Example n. 2733 = (59.0, 424.0): cluster 3\n", + "Example n. 2734 = (59.0, 423.0): cluster 3\n", + "Example n. 2735 = (59.0, 422.0): cluster 3\n", + "Example n. 2736 = (59.0, 421.0): cluster 3\n", + "Example n. 2737 = (59.0, 420.0): cluster 3\n", + "Example n. 2738 = (59.0, 419.0): cluster 3\n", + "Example n. 2739 = (59.0, 418.0): cluster 3\n", + "Example n. 2740 = (59.0, 417.0): cluster 3\n", + "Example n. 2741 = (59.0, 409.0): cluster 3\n", + "Example n. 2742 = (59.0, 408.0): cluster 3\n", + "Example n. 2743 = (59.0, 407.0): cluster 3\n", + "Example n. 2744 = (59.0, 406.0): cluster 3\n", + "Example n. 2745 = (59.0, 405.0): cluster 3\n", + "Example n. 2746 = (59.0, 404.0): cluster 3\n", + "Example n. 2747 = (59.0, 403.0): cluster 3\n", + "Example n. 2748 = (59.0, 402.0): cluster 3\n", + "Example n. 2749 = (59.0, 401.0): cluster 3\n", + "Example n. 2750 = (59.0, 400.0): cluster 3\n", + "Example n. 2751 = (59.0, 399.0): cluster 3\n", + "Example n. 2752 = (59.0, 398.0): cluster 3\n", + "Example n. 2753 = (59.0, 397.0): cluster 3\n", + "Example n. 2754 = (59.0, 396.0): cluster 3\n", + "Example n. 2755 = (59.0, 395.0): cluster 3\n", + "Example n. 2756 = (59.0, 394.0): cluster 3\n", + "Example n. 2757 = (60.0, 494.0): cluster 1\n", + "Example n. 2758 = (60.0, 493.0): cluster 1\n", + "Example n. 2759 = (60.0, 492.0): cluster 1\n", + "Example n. 2760 = (60.0, 491.0): cluster 1\n", + "Example n. 2761 = (60.0, 490.0): cluster 1\n", + "Example n. 2762 = (60.0, 489.0): cluster 1\n", + "Example n. 2763 = (60.0, 488.0): cluster 1\n", + "Example n. 2764 = (60.0, 487.0): cluster 1\n", + "Example n. 2765 = (60.0, 486.0): cluster 1\n", + "Example n. 2766 = (60.0, 485.0): cluster 1\n", + "Example n. 2767 = (60.0, 484.0): cluster 1\n", + "Example n. 2768 = (60.0, 483.0): cluster 1\n", + "Example n. 2769 = (60.0, 482.0): cluster 1\n", + "Example n. 2770 = (60.0, 481.0): cluster 1\n", + "Example n. 2771 = (60.0, 480.0): cluster 1\n", + "Example n. 2772 = (60.0, 479.0): cluster 1\n", + "Example n. 2773 = (60.0, 478.0): cluster 1\n", + "Example n. 2774 = (60.0, 477.0): cluster 1\n", + "Example n. 2775 = (60.0, 476.0): cluster 1\n", + "Example n. 2776 = (60.0, 475.0): cluster 1\n", + "Example n. 2777 = (60.0, 474.0): cluster 1\n", + "Example n. 2778 = (60.0, 473.0): cluster 1\n", + "Example n. 2779 = (60.0, 472.0): cluster 1\n", + "Example n. 2780 = (60.0, 471.0): cluster 1\n", + "Example n. 2781 = (60.0, 470.0): cluster 1\n", + "Example n. 2782 = (60.0, 469.0): cluster 1\n", + "Example n. 2783 = (60.0, 468.0): cluster 1\n", + "Example n. 2784 = (60.0, 448.0): cluster 1\n", + "Example n. 2785 = (60.0, 447.0): cluster 1\n", + "Example n. 2786 = (60.0, 446.0): cluster 1\n", + "Example n. 2787 = (60.0, 445.0): cluster 1\n", + "Example n. 2788 = (60.0, 444.0): cluster 1\n", + "Example n. 2789 = (60.0, 443.0): cluster 1\n", + "Example n. 2790 = (60.0, 442.0): cluster 1\n", + "Example n. 2791 = (60.0, 441.0): cluster 3\n", + "Example n. 2792 = (60.0, 440.0): cluster 3\n", + "Example n. 2793 = (60.0, 439.0): cluster 3\n", + "Example n. 2794 = (60.0, 438.0): cluster 3\n", + "Example n. 2795 = (60.0, 437.0): cluster 3\n", + "Example n. 2796 = (60.0, 436.0): cluster 3\n", + "Example n. 2797 = (60.0, 435.0): cluster 3\n", + "Example n. 2798 = (60.0, 434.0): cluster 3\n", + "Example n. 2799 = (60.0, 433.0): cluster 3\n", + "Example n. 2800 = (60.0, 432.0): cluster 3\n", + "Example n. 2801 = (60.0, 431.0): cluster 3\n", + "Example n. 2802 = (60.0, 430.0): cluster 3\n", + "Example n. 2803 = (60.0, 429.0): cluster 3\n", + "Example n. 2804 = (60.0, 425.0): cluster 3\n", + "Example n. 2805 = (60.0, 424.0): cluster 3\n", + "Example n. 2806 = (60.0, 423.0): cluster 3\n", + "Example n. 2807 = (60.0, 422.0): cluster 3\n", + "Example n. 2808 = (60.0, 421.0): cluster 3\n", + "Example n. 2809 = (60.0, 420.0): cluster 3\n", + "Example n. 2810 = (60.0, 419.0): cluster 3\n", + "Example n. 2811 = (60.0, 418.0): cluster 3\n", + "Example n. 2812 = (60.0, 417.0): cluster 3\n", + "Example n. 2813 = (60.0, 409.0): cluster 3\n", + "Example n. 2814 = (60.0, 408.0): cluster 3\n", + "Example n. 2815 = (60.0, 407.0): cluster 3\n", + "Example n. 2816 = (60.0, 406.0): cluster 3\n", + "Example n. 2817 = (60.0, 405.0): cluster 3\n", + "Example n. 2818 = (60.0, 404.0): cluster 3\n", + "Example n. 2819 = (60.0, 403.0): cluster 3\n", + "Example n. 2820 = (60.0, 402.0): cluster 3\n", + "Example n. 2821 = (60.0, 401.0): cluster 3\n", + "Example n. 2822 = (60.0, 400.0): cluster 3\n", + "Example n. 2823 = (60.0, 399.0): cluster 3\n", + "Example n. 2824 = (60.0, 398.0): cluster 3\n", + "Example n. 2825 = (60.0, 397.0): cluster 3\n", + "Example n. 2826 = (60.0, 396.0): cluster 3\n", + "Example n. 2827 = (60.0, 395.0): cluster 3\n", + "Example n. 2828 = (60.0, 394.0): cluster 3\n", + "Example n. 2829 = (61.0, 494.0): cluster 1\n", + "Example n. 2830 = (61.0, 493.0): cluster 1\n", + "Example n. 2831 = (61.0, 492.0): cluster 1\n", + "Example n. 2832 = (61.0, 491.0): cluster 1\n", + "Example n. 2833 = (61.0, 490.0): cluster 1\n", + "Example n. 2834 = (61.0, 489.0): cluster 1\n", + "Example n. 2835 = (61.0, 488.0): cluster 1\n", + "Example n. 2836 = (61.0, 487.0): cluster 1\n", + "Example n. 2837 = (61.0, 486.0): cluster 1\n", + "Example n. 2838 = (61.0, 485.0): cluster 1\n", + "Example n. 2839 = (61.0, 484.0): cluster 1\n", + "Example n. 2840 = (61.0, 483.0): cluster 1\n", + "Example n. 2841 = (61.0, 482.0): cluster 1\n", + "Example n. 2842 = (61.0, 481.0): cluster 1\n", + "Example n. 2843 = (61.0, 480.0): cluster 1\n", + "Example n. 2844 = (61.0, 479.0): cluster 1\n", + "Example n. 2845 = (61.0, 478.0): cluster 1\n", + "Example n. 2846 = (61.0, 477.0): cluster 1\n", + "Example n. 2847 = (61.0, 476.0): cluster 1\n", + "Example n. 2848 = (61.0, 475.0): cluster 1\n", + "Example n. 2849 = (61.0, 474.0): cluster 1\n", + "Example n. 2850 = (61.0, 473.0): cluster 1\n", + "Example n. 2851 = (61.0, 472.0): cluster 1\n", + "Example n. 2852 = (61.0, 471.0): cluster 1\n", + "Example n. 2853 = (61.0, 470.0): cluster 1\n", + "Example n. 2854 = (61.0, 469.0): cluster 1\n", + "Example n. 2855 = (61.0, 468.0): cluster 1\n", + "Example n. 2856 = (61.0, 448.0): cluster 1\n", + "Example n. 2857 = (61.0, 447.0): cluster 1\n", + "Example n. 2858 = (61.0, 446.0): cluster 1\n", + "Example n. 2859 = (61.0, 445.0): cluster 1\n", + "Example n. 2860 = (61.0, 444.0): cluster 1\n", + "Example n. 2861 = (61.0, 443.0): cluster 1\n", + "Example n. 2862 = (61.0, 442.0): cluster 1\n", + "Example n. 2863 = (61.0, 441.0): cluster 3\n", + "Example n. 2864 = (61.0, 440.0): cluster 3\n", + "Example n. 2865 = (61.0, 439.0): cluster 3\n", + "Example n. 2866 = (61.0, 438.0): cluster 3\n", + "Example n. 2867 = (61.0, 437.0): cluster 3\n", + "Example n. 2868 = (61.0, 436.0): cluster 3\n", + "Example n. 2869 = (61.0, 435.0): cluster 3\n", + "Example n. 2870 = (61.0, 434.0): cluster 3\n", + "Example n. 2871 = (61.0, 433.0): cluster 3\n", + "Example n. 2872 = (61.0, 432.0): cluster 3\n", + "Example n. 2873 = (61.0, 431.0): cluster 3\n", + "Example n. 2874 = (61.0, 430.0): cluster 3\n", + "Example n. 2875 = (61.0, 429.0): cluster 3\n", + "Example n. 2876 = (61.0, 425.0): cluster 3\n", + "Example n. 2877 = (61.0, 424.0): cluster 3\n", + "Example n. 2878 = (61.0, 423.0): cluster 3\n", + "Example n. 2879 = (61.0, 422.0): cluster 3\n", + "Example n. 2880 = (61.0, 421.0): cluster 3\n", + "Example n. 2881 = (61.0, 420.0): cluster 3\n", + "Example n. 2882 = (61.0, 419.0): cluster 3\n", + "Example n. 2883 = (61.0, 418.0): cluster 3\n", + "Example n. 2884 = (61.0, 417.0): cluster 3\n", + "Example n. 2885 = (61.0, 409.0): cluster 3\n", + "Example n. 2886 = (61.0, 408.0): cluster 3\n", + "Example n. 2887 = (61.0, 407.0): cluster 3\n", + "Example n. 2888 = (61.0, 406.0): cluster 3\n", + "Example n. 2889 = (61.0, 405.0): cluster 3\n", + "Example n. 2890 = (61.0, 404.0): cluster 3\n", + "Example n. 2891 = (61.0, 403.0): cluster 3\n", + "Example n. 2892 = (61.0, 402.0): cluster 3\n", + "Example n. 2893 = (61.0, 401.0): cluster 3\n", + "Example n. 2894 = (61.0, 400.0): cluster 3\n", + "Example n. 2895 = (61.0, 399.0): cluster 3\n", + "Example n. 2896 = (61.0, 398.0): cluster 3\n", + "Example n. 2897 = (61.0, 397.0): cluster 3\n", + "Example n. 2898 = (61.0, 396.0): cluster 3\n", + "Example n. 2899 = (61.0, 395.0): cluster 3\n", + "Example n. 2900 = (61.0, 394.0): cluster 3\n", + "Example n. 2901 = (62.0, 494.0): cluster 1\n", + "Example n. 2902 = (62.0, 493.0): cluster 1\n", + "Example n. 2903 = (62.0, 492.0): cluster 1\n", + "Example n. 2904 = (62.0, 491.0): cluster 1\n", + "Example n. 2905 = (62.0, 490.0): cluster 1\n", + "Example n. 2906 = (62.0, 489.0): cluster 1\n", + "Example n. 2907 = (62.0, 488.0): cluster 1\n", + "Example n. 2908 = (62.0, 487.0): cluster 1\n", + "Example n. 2909 = (62.0, 486.0): cluster 1\n", + "Example n. 2910 = (62.0, 485.0): cluster 1\n", + "Example n. 2911 = (62.0, 484.0): cluster 1\n", + "Example n. 2912 = (62.0, 483.0): cluster 1\n", + "Example n. 2913 = (62.0, 482.0): cluster 1\n", + "Example n. 2914 = (62.0, 481.0): cluster 1\n", + "Example n. 2915 = (62.0, 480.0): cluster 1\n", + "Example n. 2916 = (62.0, 479.0): cluster 1\n", + "Example n. 2917 = (62.0, 478.0): cluster 1\n", + "Example n. 2918 = (62.0, 477.0): cluster 1\n", + "Example n. 2919 = (62.0, 476.0): cluster 1\n", + "Example n. 2920 = (62.0, 475.0): cluster 1\n", + "Example n. 2921 = (62.0, 474.0): cluster 1\n", + "Example n. 2922 = (62.0, 473.0): cluster 1\n", + "Example n. 2923 = (62.0, 472.0): cluster 1\n", + "Example n. 2924 = (62.0, 471.0): cluster 1\n", + "Example n. 2925 = (62.0, 470.0): cluster 1\n", + "Example n. 2926 = (62.0, 469.0): cluster 1\n", + "Example n. 2927 = (62.0, 468.0): cluster 1\n", + "Example n. 2928 = (62.0, 467.0): cluster 1\n", + "Example n. 2929 = (62.0, 466.0): cluster 1\n", + "Example n. 2930 = (62.0, 465.0): cluster 1\n", + "Example n. 2931 = (62.0, 464.0): cluster 1\n", + "Example n. 2932 = (62.0, 463.0): cluster 1\n", + "Example n. 2933 = (62.0, 462.0): cluster 1\n", + "Example n. 2934 = (62.0, 461.0): cluster 1\n", + "Example n. 2935 = (62.0, 460.0): cluster 1\n", + "Example n. 2936 = (62.0, 459.0): cluster 1\n", + "Example n. 2937 = (62.0, 458.0): cluster 1\n", + "Example n. 2938 = (62.0, 457.0): cluster 1\n", + "Example n. 2939 = (62.0, 456.0): cluster 1\n", + "Example n. 2940 = (62.0, 455.0): cluster 1\n", + "Example n. 2941 = (62.0, 454.0): cluster 1\n", + "Example n. 2942 = (62.0, 453.0): cluster 1\n", + "Example n. 2943 = (62.0, 452.0): cluster 1\n", + "Example n. 2944 = (62.0, 451.0): cluster 1\n", + "Example n. 2945 = (62.0, 450.0): cluster 1\n", + "Example n. 2946 = (62.0, 449.0): cluster 1\n", + "Example n. 2947 = (62.0, 448.0): cluster 1\n", + "Example n. 2948 = (62.0, 447.0): cluster 1\n", + "Example n. 2949 = (62.0, 446.0): cluster 1\n", + "Example n. 2950 = (62.0, 445.0): cluster 1\n", + "Example n. 2951 = (62.0, 444.0): cluster 1\n", + "Example n. 2952 = (62.0, 443.0): cluster 1\n", + "Example n. 2953 = (62.0, 442.0): cluster 1\n", + "Example n. 2954 = (62.0, 441.0): cluster 3\n", + "Example n. 2955 = (62.0, 440.0): cluster 3\n", + "Example n. 2956 = (62.0, 439.0): cluster 3\n", + "Example n. 2957 = (62.0, 438.0): cluster 3\n", + "Example n. 2958 = (62.0, 437.0): cluster 3\n", + "Example n. 2959 = (62.0, 436.0): cluster 3\n", + "Example n. 2960 = (62.0, 435.0): cluster 3\n", + "Example n. 2961 = (62.0, 434.0): cluster 3\n", + "Example n. 2962 = (62.0, 433.0): cluster 3\n", + "Example n. 2963 = (62.0, 432.0): cluster 3\n", + "Example n. 2964 = (62.0, 431.0): cluster 3\n", + "Example n. 2965 = (62.0, 430.0): cluster 3\n", + "Example n. 2966 = (62.0, 429.0): cluster 3\n", + "Example n. 2967 = (62.0, 425.0): cluster 3\n", + "Example n. 2968 = (62.0, 424.0): cluster 3\n", + "Example n. 2969 = (62.0, 423.0): cluster 3\n", + "Example n. 2970 = (62.0, 422.0): cluster 3\n", + "Example n. 2971 = (62.0, 421.0): cluster 3\n", + "Example n. 2972 = (62.0, 420.0): cluster 3\n", + "Example n. 2973 = (62.0, 419.0): cluster 3\n", + "Example n. 2974 = (62.0, 418.0): cluster 3\n", + "Example n. 2975 = (62.0, 417.0): cluster 3\n", + "Example n. 2976 = (62.0, 416.0): cluster 3\n", + "Example n. 2977 = (62.0, 415.0): cluster 3\n", + "Example n. 2978 = (62.0, 414.0): cluster 3\n", + "Example n. 2979 = (62.0, 409.0): cluster 3\n", + "Example n. 2980 = (62.0, 408.0): cluster 3\n", + "Example n. 2981 = (62.0, 407.0): cluster 3\n", + "Example n. 2982 = (62.0, 406.0): cluster 3\n", + "Example n. 2983 = (62.0, 405.0): cluster 3\n", + "Example n. 2984 = (62.0, 404.0): cluster 3\n", + "Example n. 2985 = (62.0, 403.0): cluster 3\n", + "Example n. 2986 = (62.0, 402.0): cluster 3\n", + "Example n. 2987 = (62.0, 401.0): cluster 3\n", + "Example n. 2988 = (62.0, 400.0): cluster 3\n", + "Example n. 2989 = (62.0, 399.0): cluster 3\n", + "Example n. 2990 = (62.0, 398.0): cluster 3\n", + "Example n. 2991 = (62.0, 397.0): cluster 3\n", + "Example n. 2992 = (62.0, 396.0): cluster 3\n", + "Example n. 2993 = (62.0, 395.0): cluster 3\n", + "Example n. 2994 = (62.0, 394.0): cluster 3\n", + "Example n. 2995 = (63.0, 494.0): cluster 1\n", + "Example n. 2996 = (63.0, 493.0): cluster 1\n", + "Example n. 2997 = (63.0, 492.0): cluster 1\n", + "Example n. 2998 = (63.0, 491.0): cluster 1\n", + "Example n. 2999 = (63.0, 490.0): cluster 1\n", + "Example n. 3000 = (63.0, 489.0): cluster 1\n", + "Example n. 3001 = (63.0, 488.0): cluster 1\n", + "Example n. 3002 = (63.0, 487.0): cluster 1\n", + "Example n. 3003 = (63.0, 486.0): cluster 1\n", + "Example n. 3004 = (63.0, 485.0): cluster 1\n", + "Example n. 3005 = (63.0, 484.0): cluster 1\n", + "Example n. 3006 = (63.0, 483.0): cluster 1\n", + "Example n. 3007 = (63.0, 482.0): cluster 1\n", + "Example n. 3008 = (63.0, 481.0): cluster 1\n", + "Example n. 3009 = (63.0, 480.0): cluster 1\n", + "Example n. 3010 = (63.0, 479.0): cluster 1\n", + "Example n. 3011 = (63.0, 478.0): cluster 1\n", + "Example n. 3012 = (63.0, 477.0): cluster 1\n", + "Example n. 3013 = (63.0, 476.0): cluster 1\n", + "Example n. 3014 = (63.0, 475.0): cluster 1\n", + "Example n. 3015 = (63.0, 474.0): cluster 1\n", + "Example n. 3016 = (63.0, 473.0): cluster 1\n", + "Example n. 3017 = (63.0, 472.0): cluster 1\n", + "Example n. 3018 = (63.0, 471.0): cluster 1\n", + "Example n. 3019 = (63.0, 470.0): cluster 1\n", + "Example n. 3020 = (63.0, 469.0): cluster 1\n", + "Example n. 3021 = (63.0, 468.0): cluster 1\n", + "Example n. 3022 = (63.0, 467.0): cluster 1\n", + "Example n. 3023 = (63.0, 466.0): cluster 1\n", + "Example n. 3024 = (63.0, 465.0): cluster 1\n", + "Example n. 3025 = (63.0, 464.0): cluster 1\n", + "Example n. 3026 = (63.0, 463.0): cluster 1\n", + "Example n. 3027 = (63.0, 462.0): cluster 1\n", + "Example n. 3028 = (63.0, 461.0): cluster 1\n", + "Example n. 3029 = (63.0, 460.0): cluster 1\n", + "Example n. 3030 = (63.0, 459.0): cluster 1\n", + "Example n. 3031 = (63.0, 458.0): cluster 1\n", + "Example n. 3032 = (63.0, 457.0): cluster 1\n", + "Example n. 3033 = (63.0, 456.0): cluster 1\n", + "Example n. 3034 = (63.0, 455.0): cluster 1\n", + "Example n. 3035 = (63.0, 454.0): cluster 1\n", + "Example n. 3036 = (63.0, 453.0): cluster 1\n", + "Example n. 3037 = (63.0, 452.0): cluster 1\n", + "Example n. 3038 = (63.0, 451.0): cluster 1\n", + "Example n. 3039 = (63.0, 450.0): cluster 1\n", + "Example n. 3040 = (63.0, 449.0): cluster 1\n", + "Example n. 3041 = (63.0, 448.0): cluster 1\n", + "Example n. 3042 = (63.0, 447.0): cluster 1\n", + "Example n. 3043 = (63.0, 446.0): cluster 1\n", + "Example n. 3044 = (63.0, 445.0): cluster 1\n", + "Example n. 3045 = (63.0, 444.0): cluster 1\n", + "Example n. 3046 = (63.0, 443.0): cluster 1\n", + "Example n. 3047 = (63.0, 442.0): cluster 1\n", + "Example n. 3048 = (63.0, 441.0): cluster 3\n", + "Example n. 3049 = (63.0, 440.0): cluster 3\n", + "Example n. 3050 = (63.0, 439.0): cluster 3\n", + "Example n. 3051 = (63.0, 438.0): cluster 3\n", + "Example n. 3052 = (63.0, 437.0): cluster 3\n", + "Example n. 3053 = (63.0, 436.0): cluster 3\n", + "Example n. 3054 = (63.0, 435.0): cluster 3\n", + "Example n. 3055 = (63.0, 434.0): cluster 3\n", + "Example n. 3056 = (63.0, 433.0): cluster 3\n", + "Example n. 3057 = (63.0, 425.0): cluster 3\n", + "Example n. 3058 = (63.0, 424.0): cluster 3\n", + "Example n. 3059 = (63.0, 423.0): cluster 3\n", + "Example n. 3060 = (63.0, 422.0): cluster 3\n", + "Example n. 3061 = (63.0, 421.0): cluster 3\n", + "Example n. 3062 = (63.0, 420.0): cluster 3\n", + "Example n. 3063 = (63.0, 419.0): cluster 3\n", + "Example n. 3064 = (63.0, 418.0): cluster 3\n", + "Example n. 3065 = (63.0, 417.0): cluster 3\n", + "Example n. 3066 = (63.0, 416.0): cluster 3\n", + "Example n. 3067 = (63.0, 415.0): cluster 3\n", + "Example n. 3068 = (63.0, 414.0): cluster 3\n", + "Example n. 3069 = (63.0, 409.0): cluster 3\n", + "Example n. 3070 = (63.0, 408.0): cluster 3\n", + "Example n. 3071 = (63.0, 407.0): cluster 3\n", + "Example n. 3072 = (63.0, 406.0): cluster 3\n", + "Example n. 3073 = (63.0, 405.0): cluster 3\n", + "Example n. 3074 = (63.0, 404.0): cluster 3\n", + "Example n. 3075 = (63.0, 403.0): cluster 3\n", + "Example n. 3076 = (63.0, 402.0): cluster 3\n", + "Example n. 3077 = (63.0, 401.0): cluster 3\n", + "Example n. 3078 = (63.0, 400.0): cluster 3\n", + "Example n. 3079 = (63.0, 399.0): cluster 3\n", + "Example n. 3080 = (63.0, 398.0): cluster 3\n", + "Example n. 3081 = (63.0, 397.0): cluster 3\n", + "Example n. 3082 = (63.0, 396.0): cluster 3\n", + "Example n. 3083 = (63.0, 395.0): cluster 3\n", + "Example n. 3084 = (63.0, 394.0): cluster 3\n", + "Example n. 3085 = (64.0, 494.0): cluster 1\n", + "Example n. 3086 = (64.0, 493.0): cluster 1\n", + "Example n. 3087 = (64.0, 492.0): cluster 1\n", + "Example n. 3088 = (64.0, 491.0): cluster 1\n", + "Example n. 3089 = (64.0, 490.0): cluster 1\n", + "Example n. 3090 = (64.0, 489.0): cluster 1\n", + "Example n. 3091 = (64.0, 488.0): cluster 1\n", + "Example n. 3092 = (64.0, 487.0): cluster 1\n", + "Example n. 3093 = (64.0, 486.0): cluster 1\n", + "Example n. 3094 = (64.0, 485.0): cluster 1\n", + "Example n. 3095 = (64.0, 484.0): cluster 1\n", + "Example n. 3096 = (64.0, 483.0): cluster 1\n", + "Example n. 3097 = (64.0, 482.0): cluster 1\n", + "Example n. 3098 = (64.0, 481.0): cluster 1\n", + "Example n. 3099 = (64.0, 480.0): cluster 1\n", + "Example n. 3100 = (64.0, 479.0): cluster 1\n", + "Example n. 3101 = (64.0, 478.0): cluster 1\n", + "Example n. 3102 = (64.0, 477.0): cluster 1\n", + "Example n. 3103 = (64.0, 476.0): cluster 1\n", + "Example n. 3104 = (64.0, 475.0): cluster 1\n", + "Example n. 3105 = (64.0, 474.0): cluster 1\n", + "Example n. 3106 = (64.0, 473.0): cluster 1\n", + "Example n. 3107 = (64.0, 472.0): cluster 1\n", + "Example n. 3108 = (64.0, 471.0): cluster 1\n", + "Example n. 3109 = (64.0, 470.0): cluster 1\n", + "Example n. 3110 = (64.0, 469.0): cluster 1\n", + "Example n. 3111 = (64.0, 468.0): cluster 1\n", + "Example n. 3112 = (64.0, 467.0): cluster 1\n", + "Example n. 3113 = (64.0, 466.0): cluster 1\n", + "Example n. 3114 = (64.0, 465.0): cluster 1\n", + "Example n. 3115 = (64.0, 464.0): cluster 1\n", + "Example n. 3116 = (64.0, 463.0): cluster 1\n", + "Example n. 3117 = (64.0, 462.0): cluster 1\n", + "Example n. 3118 = (64.0, 461.0): cluster 1\n", + "Example n. 3119 = (64.0, 460.0): cluster 1\n", + "Example n. 3120 = (64.0, 459.0): cluster 1\n", + "Example n. 3121 = (64.0, 458.0): cluster 1\n", + "Example n. 3122 = (64.0, 457.0): cluster 1\n", + "Example n. 3123 = (64.0, 456.0): cluster 1\n", + "Example n. 3124 = (64.0, 455.0): cluster 1\n", + "Example n. 3125 = (64.0, 454.0): cluster 1\n", + "Example n. 3126 = (64.0, 453.0): cluster 1\n", + "Example n. 3127 = (64.0, 452.0): cluster 1\n", + "Example n. 3128 = (64.0, 451.0): cluster 1\n", + "Example n. 3129 = (64.0, 450.0): cluster 1\n", + "Example n. 3130 = (64.0, 449.0): cluster 1\n", + "Example n. 3131 = (64.0, 448.0): cluster 1\n", + "Example n. 3132 = (64.0, 447.0): cluster 1\n", + "Example n. 3133 = (64.0, 446.0): cluster 1\n", + "Example n. 3134 = (64.0, 445.0): cluster 1\n", + "Example n. 3135 = (64.0, 444.0): cluster 1\n", + "Example n. 3136 = (64.0, 443.0): cluster 1\n", + "Example n. 3137 = (64.0, 442.0): cluster 1\n", + "Example n. 3138 = (64.0, 441.0): cluster 3\n", + "Example n. 3139 = (64.0, 440.0): cluster 3\n", + "Example n. 3140 = (64.0, 439.0): cluster 3\n", + "Example n. 3141 = (64.0, 438.0): cluster 3\n", + "Example n. 3142 = (64.0, 437.0): cluster 3\n", + "Example n. 3143 = (64.0, 436.0): cluster 3\n", + "Example n. 3144 = (64.0, 435.0): cluster 3\n", + "Example n. 3145 = (64.0, 434.0): cluster 3\n", + "Example n. 3146 = (64.0, 433.0): cluster 3\n", + "Example n. 3147 = (64.0, 425.0): cluster 3\n", + "Example n. 3148 = (64.0, 424.0): cluster 3\n", + "Example n. 3149 = (64.0, 423.0): cluster 3\n", + "Example n. 3150 = (64.0, 422.0): cluster 3\n", + "Example n. 3151 = (64.0, 421.0): cluster 3\n", + "Example n. 3152 = (64.0, 420.0): cluster 3\n", + "Example n. 3153 = (64.0, 419.0): cluster 3\n", + "Example n. 3154 = (64.0, 418.0): cluster 3\n", + "Example n. 3155 = (64.0, 417.0): cluster 3\n", + "Example n. 3156 = (64.0, 416.0): cluster 3\n", + "Example n. 3157 = (64.0, 415.0): cluster 3\n", + "Example n. 3158 = (64.0, 414.0): cluster 3\n", + "Example n. 3159 = (64.0, 409.0): cluster 3\n", + "Example n. 3160 = (64.0, 408.0): cluster 3\n", + "Example n. 3161 = (64.0, 407.0): cluster 3\n", + "Example n. 3162 = (64.0, 406.0): cluster 3\n", + "Example n. 3163 = (64.0, 405.0): cluster 3\n", + "Example n. 3164 = (64.0, 404.0): cluster 3\n", + "Example n. 3165 = (64.0, 403.0): cluster 3\n", + "Example n. 3166 = (64.0, 402.0): cluster 3\n", + "Example n. 3167 = (64.0, 401.0): cluster 3\n", + "Example n. 3168 = (64.0, 400.0): cluster 3\n", + "Example n. 3169 = (64.0, 399.0): cluster 3\n", + "Example n. 3170 = (64.0, 398.0): cluster 3\n", + "Example n. 3171 = (64.0, 397.0): cluster 3\n", + "Example n. 3172 = (64.0, 396.0): cluster 3\n", + "Example n. 3173 = (64.0, 395.0): cluster 3\n", + "Example n. 3174 = (64.0, 394.0): cluster 3\n", + "Example n. 3175 = (65.0, 494.0): cluster 1\n", + "Example n. 3176 = (65.0, 493.0): cluster 1\n", + "Example n. 3177 = (65.0, 492.0): cluster 1\n", + "Example n. 3178 = (65.0, 491.0): cluster 1\n", + "Example n. 3179 = (65.0, 490.0): cluster 1\n", + "Example n. 3180 = (65.0, 489.0): cluster 1\n", + "Example n. 3181 = (65.0, 488.0): cluster 1\n", + "Example n. 3182 = (65.0, 487.0): cluster 1\n", + "Example n. 3183 = (65.0, 486.0): cluster 1\n", + "Example n. 3184 = (65.0, 485.0): cluster 1\n", + "Example n. 3185 = (65.0, 484.0): cluster 1\n", + "Example n. 3186 = (65.0, 483.0): cluster 1\n", + "Example n. 3187 = (65.0, 482.0): cluster 1\n", + "Example n. 3188 = (65.0, 481.0): cluster 1\n", + "Example n. 3189 = (65.0, 480.0): cluster 1\n", + "Example n. 3190 = (65.0, 479.0): cluster 1\n", + "Example n. 3191 = (65.0, 478.0): cluster 1\n", + "Example n. 3192 = (65.0, 477.0): cluster 1\n", + "Example n. 3193 = (65.0, 476.0): cluster 1\n", + "Example n. 3194 = (65.0, 475.0): cluster 1\n", + "Example n. 3195 = (65.0, 474.0): cluster 1\n", + "Example n. 3196 = (65.0, 473.0): cluster 1\n", + "Example n. 3197 = (65.0, 472.0): cluster 1\n", + "Example n. 3198 = (65.0, 471.0): cluster 1\n", + "Example n. 3199 = (65.0, 470.0): cluster 1\n", + "Example n. 3200 = (65.0, 469.0): cluster 1\n", + "Example n. 3201 = (65.0, 468.0): cluster 1\n", + "Example n. 3202 = (65.0, 467.0): cluster 1\n", + "Example n. 3203 = (65.0, 466.0): cluster 1\n", + "Example n. 3204 = (65.0, 465.0): cluster 1\n", + "Example n. 3205 = (65.0, 464.0): cluster 1\n", + "Example n. 3206 = (65.0, 463.0): cluster 1\n", + "Example n. 3207 = (65.0, 462.0): cluster 1\n", + "Example n. 3208 = (65.0, 461.0): cluster 1\n", + "Example n. 3209 = (65.0, 460.0): cluster 1\n", + "Example n. 3210 = (65.0, 459.0): cluster 1\n", + "Example n. 3211 = (65.0, 458.0): cluster 1\n", + "Example n. 3212 = (65.0, 457.0): cluster 1\n", + "Example n. 3213 = (65.0, 456.0): cluster 1\n", + "Example n. 3214 = (65.0, 455.0): cluster 1\n", + "Example n. 3215 = (65.0, 454.0): cluster 1\n", + "Example n. 3216 = (65.0, 453.0): cluster 1\n", + "Example n. 3217 = (65.0, 452.0): cluster 1\n", + "Example n. 3218 = (65.0, 451.0): cluster 1\n", + "Example n. 3219 = (65.0, 450.0): cluster 1\n", + "Example n. 3220 = (65.0, 449.0): cluster 1\n", + "Example n. 3221 = (65.0, 448.0): cluster 1\n", + "Example n. 3222 = (65.0, 447.0): cluster 1\n", + "Example n. 3223 = (65.0, 446.0): cluster 1\n", + "Example n. 3224 = (65.0, 445.0): cluster 1\n", + "Example n. 3225 = (65.0, 444.0): cluster 1\n", + "Example n. 3226 = (65.0, 443.0): cluster 1\n", + "Example n. 3227 = (65.0, 442.0): cluster 1\n", + "Example n. 3228 = (65.0, 441.0): cluster 3\n", + "Example n. 3229 = (65.0, 440.0): cluster 3\n", + "Example n. 3230 = (65.0, 439.0): cluster 3\n", + "Example n. 3231 = (65.0, 438.0): cluster 3\n", + "Example n. 3232 = (65.0, 437.0): cluster 3\n", + "Example n. 3233 = (65.0, 436.0): cluster 3\n", + "Example n. 3234 = (65.0, 435.0): cluster 3\n", + "Example n. 3235 = (65.0, 434.0): cluster 3\n", + "Example n. 3236 = (65.0, 433.0): cluster 3\n", + "Example n. 3237 = (65.0, 425.0): cluster 3\n", + "Example n. 3238 = (65.0, 424.0): cluster 3\n", + "Example n. 3239 = (65.0, 423.0): cluster 3\n", + "Example n. 3240 = (65.0, 422.0): cluster 3\n", + "Example n. 3241 = (65.0, 421.0): cluster 3\n", + "Example n. 3242 = (65.0, 420.0): cluster 3\n", + "Example n. 3243 = (65.0, 419.0): cluster 3\n", + "Example n. 3244 = (65.0, 418.0): cluster 3\n", + "Example n. 3245 = (65.0, 417.0): cluster 3\n", + "Example n. 3246 = (65.0, 416.0): cluster 3\n", + "Example n. 3247 = (65.0, 415.0): cluster 3\n", + "Example n. 3248 = (65.0, 414.0): cluster 3\n", + "Example n. 3249 = (65.0, 409.0): cluster 3\n", + "Example n. 3250 = (65.0, 408.0): cluster 3\n", + "Example n. 3251 = (65.0, 407.0): cluster 3\n", + "Example n. 3252 = (65.0, 406.0): cluster 3\n", + "Example n. 3253 = (65.0, 405.0): cluster 3\n", + "Example n. 3254 = (65.0, 404.0): cluster 3\n", + "Example n. 3255 = (65.0, 403.0): cluster 3\n", + "Example n. 3256 = (65.0, 402.0): cluster 3\n", + "Example n. 3257 = (65.0, 401.0): cluster 3\n", + "Example n. 3258 = (65.0, 400.0): cluster 3\n", + "Example n. 3259 = (65.0, 399.0): cluster 3\n", + "Example n. 3260 = (65.0, 398.0): cluster 3\n", + "Example n. 3261 = (65.0, 397.0): cluster 3\n", + "Example n. 3262 = (65.0, 396.0): cluster 3\n", + "Example n. 3263 = (65.0, 395.0): cluster 3\n", + "Example n. 3264 = (65.0, 394.0): cluster 3\n", + "Example n. 3265 = (66.0, 494.0): cluster 1\n", + "Example n. 3266 = (66.0, 493.0): cluster 1\n", + "Example n. 3267 = (66.0, 492.0): cluster 1\n", + "Example n. 3268 = (66.0, 491.0): cluster 1\n", + "Example n. 3269 = (66.0, 490.0): cluster 1\n", + "Example n. 3270 = (66.0, 489.0): cluster 1\n", + "Example n. 3271 = (66.0, 488.0): cluster 1\n", + "Example n. 3272 = (66.0, 487.0): cluster 1\n", + "Example n. 3273 = (66.0, 486.0): cluster 1\n", + "Example n. 3274 = (66.0, 485.0): cluster 1\n", + "Example n. 3275 = (66.0, 484.0): cluster 1\n", + "Example n. 3276 = (66.0, 483.0): cluster 1\n", + "Example n. 3277 = (66.0, 482.0): cluster 1\n", + "Example n. 3278 = (66.0, 481.0): cluster 1\n", + "Example n. 3279 = (66.0, 480.0): cluster 1\n", + "Example n. 3280 = (66.0, 479.0): cluster 1\n", + "Example n. 3281 = (66.0, 478.0): cluster 1\n", + "Example n. 3282 = (66.0, 477.0): cluster 1\n", + "Example n. 3283 = (66.0, 476.0): cluster 1\n", + "Example n. 3284 = (66.0, 475.0): cluster 1\n", + "Example n. 3285 = (66.0, 474.0): cluster 1\n", + "Example n. 3286 = (66.0, 473.0): cluster 1\n", + "Example n. 3287 = (66.0, 472.0): cluster 1\n", + "Example n. 3288 = (66.0, 471.0): cluster 1\n", + "Example n. 3289 = (66.0, 470.0): cluster 1\n", + "Example n. 3290 = (66.0, 469.0): cluster 1\n", + "Example n. 3291 = (66.0, 468.0): cluster 1\n", + "Example n. 3292 = (66.0, 467.0): cluster 1\n", + "Example n. 3293 = (66.0, 466.0): cluster 1\n", + "Example n. 3294 = (66.0, 465.0): cluster 1\n", + "Example n. 3295 = (66.0, 464.0): cluster 1\n", + "Example n. 3296 = (66.0, 463.0): cluster 1\n", + "Example n. 3297 = (66.0, 462.0): cluster 1\n", + "Example n. 3298 = (66.0, 461.0): cluster 1\n", + "Example n. 3299 = (66.0, 460.0): cluster 1\n", + "Example n. 3300 = (66.0, 459.0): cluster 1\n", + "Example n. 3301 = (66.0, 458.0): cluster 1\n", + "Example n. 3302 = (66.0, 457.0): cluster 1\n", + "Example n. 3303 = (66.0, 456.0): cluster 1\n", + "Example n. 3304 = (66.0, 455.0): cluster 1\n", + "Example n. 3305 = (66.0, 454.0): cluster 1\n", + "Example n. 3306 = (66.0, 453.0): cluster 1\n", + "Example n. 3307 = (66.0, 452.0): cluster 1\n", + "Example n. 3308 = (66.0, 451.0): cluster 1\n", + "Example n. 3309 = (66.0, 450.0): cluster 1\n", + "Example n. 3310 = (66.0, 449.0): cluster 1\n", + "Example n. 3311 = (66.0, 448.0): cluster 1\n", + "Example n. 3312 = (66.0, 447.0): cluster 1\n", + "Example n. 3313 = (66.0, 446.0): cluster 1\n", + "Example n. 3314 = (66.0, 445.0): cluster 1\n", + "Example n. 3315 = (66.0, 444.0): cluster 1\n", + "Example n. 3316 = (66.0, 443.0): cluster 1\n", + "Example n. 3317 = (66.0, 442.0): cluster 1\n", + "Example n. 3318 = (66.0, 441.0): cluster 3\n", + "Example n. 3319 = (66.0, 440.0): cluster 3\n", + "Example n. 3320 = (66.0, 439.0): cluster 3\n", + "Example n. 3321 = (66.0, 438.0): cluster 3\n", + "Example n. 3322 = (66.0, 437.0): cluster 3\n", + "Example n. 3323 = (66.0, 436.0): cluster 3\n", + "Example n. 3324 = (66.0, 435.0): cluster 3\n", + "Example n. 3325 = (66.0, 434.0): cluster 3\n", + "Example n. 3326 = (66.0, 433.0): cluster 3\n", + "Example n. 3327 = (66.0, 429.0): cluster 3\n", + "Example n. 3328 = (66.0, 428.0): cluster 3\n", + "Example n. 3329 = (66.0, 427.0): cluster 3\n", + "Example n. 3330 = (66.0, 426.0): cluster 3\n", + "Example n. 3331 = (66.0, 425.0): cluster 3\n", + "Example n. 3332 = (66.0, 424.0): cluster 3\n", + "Example n. 3333 = (66.0, 423.0): cluster 3\n", + "Example n. 3334 = (66.0, 422.0): cluster 3\n", + "Example n. 3335 = (66.0, 421.0): cluster 3\n", + "Example n. 3336 = (66.0, 420.0): cluster 3\n", + "Example n. 3337 = (66.0, 419.0): cluster 3\n", + "Example n. 3338 = (66.0, 418.0): cluster 3\n", + "Example n. 3339 = (66.0, 417.0): cluster 3\n", + "Example n. 3340 = (66.0, 416.0): cluster 3\n", + "Example n. 3341 = (66.0, 415.0): cluster 3\n", + "Example n. 3342 = (66.0, 414.0): cluster 3\n", + "Example n. 3343 = (66.0, 409.0): cluster 3\n", + "Example n. 3344 = (66.0, 408.0): cluster 3\n", + "Example n. 3345 = (66.0, 407.0): cluster 3\n", + "Example n. 3346 = (66.0, 406.0): cluster 3\n", + "Example n. 3347 = (66.0, 405.0): cluster 3\n", + "Example n. 3348 = (66.0, 404.0): cluster 3\n", + "Example n. 3349 = (66.0, 403.0): cluster 3\n", + "Example n. 3350 = (66.0, 402.0): cluster 3\n", + "Example n. 3351 = (66.0, 401.0): cluster 3\n", + "Example n. 3352 = (66.0, 400.0): cluster 3\n", + "Example n. 3353 = (66.0, 399.0): cluster 3\n", + "Example n. 3354 = (66.0, 398.0): cluster 3\n", + "Example n. 3355 = (66.0, 397.0): cluster 3\n", + "Example n. 3356 = (66.0, 396.0): cluster 3\n", + "Example n. 3357 = (66.0, 395.0): cluster 3\n", + "Example n. 3358 = (66.0, 394.0): cluster 3\n", + "Example n. 3359 = (67.0, 486.0): cluster 1\n", + "Example n. 3360 = (67.0, 485.0): cluster 1\n", + "Example n. 3361 = (67.0, 484.0): cluster 1\n", + "Example n. 3362 = (67.0, 483.0): cluster 1\n", + "Example n. 3363 = (67.0, 482.0): cluster 1\n", + "Example n. 3364 = (67.0, 481.0): cluster 1\n", + "Example n. 3365 = (67.0, 480.0): cluster 1\n", + "Example n. 3366 = (67.0, 479.0): cluster 1\n", + "Example n. 3367 = (67.0, 478.0): cluster 1\n", + "Example n. 3368 = (67.0, 477.0): cluster 1\n", + "Example n. 3369 = (67.0, 476.0): cluster 1\n", + "Example n. 3370 = (67.0, 475.0): cluster 1\n", + "Example n. 3371 = (67.0, 474.0): cluster 1\n", + "Example n. 3372 = (67.0, 473.0): cluster 1\n", + "Example n. 3373 = (67.0, 472.0): cluster 1\n", + "Example n. 3374 = (67.0, 471.0): cluster 1\n", + "Example n. 3375 = (67.0, 470.0): cluster 1\n", + "Example n. 3376 = (67.0, 469.0): cluster 1\n", + "Example n. 3377 = (67.0, 468.0): cluster 1\n", + "Example n. 3378 = (67.0, 467.0): cluster 1\n", + "Example n. 3379 = (67.0, 466.0): cluster 1\n", + "Example n. 3380 = (67.0, 465.0): cluster 1\n", + "Example n. 3381 = (67.0, 464.0): cluster 1\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Example n. 3382 = (67.0, 463.0): cluster 1\n", + "Example n. 3383 = (67.0, 462.0): cluster 1\n", + "Example n. 3384 = (67.0, 461.0): cluster 1\n", + "Example n. 3385 = (67.0, 460.0): cluster 1\n", + "Example n. 3386 = (67.0, 459.0): cluster 1\n", + "Example n. 3387 = (67.0, 458.0): cluster 1\n", + "Example n. 3388 = (67.0, 457.0): cluster 1\n", + "Example n. 3389 = (67.0, 456.0): cluster 1\n", + "Example n. 3390 = (67.0, 455.0): cluster 1\n", + "Example n. 3391 = (67.0, 454.0): cluster 1\n", + "Example n. 3392 = (67.0, 453.0): cluster 1\n", + "Example n. 3393 = (67.0, 452.0): cluster 1\n", + "Example n. 3394 = (67.0, 451.0): cluster 1\n", + "Example n. 3395 = (67.0, 450.0): cluster 1\n", + "Example n. 3396 = (67.0, 449.0): cluster 1\n", + "Example n. 3397 = (67.0, 448.0): cluster 1\n", + "Example n. 3398 = (67.0, 447.0): cluster 1\n", + "Example n. 3399 = (67.0, 446.0): cluster 1\n", + "Example n. 3400 = (67.0, 445.0): cluster 1\n", + "Example n. 3401 = (67.0, 444.0): cluster 1\n", + "Example n. 3402 = (67.0, 443.0): cluster 1\n", + "Example n. 3403 = (67.0, 442.0): cluster 1\n", + "Example n. 3404 = (67.0, 441.0): cluster 3\n", + "Example n. 3405 = (67.0, 440.0): cluster 3\n", + "Example n. 3406 = (67.0, 439.0): cluster 3\n", + "Example n. 3407 = (67.0, 438.0): cluster 3\n", + "Example n. 3408 = (67.0, 437.0): cluster 3\n", + "Example n. 3409 = (67.0, 429.0): cluster 3\n", + "Example n. 3410 = (67.0, 428.0): cluster 3\n", + "Example n. 3411 = (67.0, 427.0): cluster 3\n", + "Example n. 3412 = (67.0, 426.0): cluster 3\n", + "Example n. 3413 = (67.0, 425.0): cluster 3\n", + "Example n. 3414 = (67.0, 424.0): cluster 3\n", + "Example n. 3415 = (67.0, 423.0): cluster 3\n", + "Example n. 3416 = (67.0, 422.0): cluster 3\n", + "Example n. 3417 = (67.0, 421.0): cluster 3\n", + "Example n. 3418 = (67.0, 420.0): cluster 3\n", + "Example n. 3419 = (67.0, 419.0): cluster 3\n", + "Example n. 3420 = (67.0, 418.0): cluster 3\n", + "Example n. 3421 = (67.0, 417.0): cluster 3\n", + "Example n. 3422 = (67.0, 409.0): cluster 3\n", + "Example n. 3423 = (67.0, 408.0): cluster 3\n", + "Example n. 3424 = (67.0, 407.0): cluster 3\n", + "Example n. 3425 = (67.0, 406.0): cluster 3\n", + "Example n. 3426 = (67.0, 405.0): cluster 3\n", + "Example n. 3427 = (67.0, 404.0): cluster 3\n", + "Example n. 3428 = (67.0, 403.0): cluster 3\n", + "Example n. 3429 = (67.0, 402.0): cluster 3\n", + "Example n. 3430 = (67.0, 401.0): cluster 3\n", + "Example n. 3431 = (67.0, 400.0): cluster 3\n", + "Example n. 3432 = (67.0, 399.0): cluster 3\n", + "Example n. 3433 = (67.0, 398.0): cluster 3\n", + "Example n. 3434 = (67.0, 397.0): cluster 3\n", + "Example n. 3435 = (67.0, 396.0): cluster 3\n", + "Example n. 3436 = (67.0, 395.0): cluster 3\n", + "Example n. 3437 = (67.0, 394.0): cluster 3\n", + "Example n. 3438 = (68.0, 486.0): cluster 1\n", + "Example n. 3439 = (68.0, 485.0): cluster 1\n", + "Example n. 3440 = (68.0, 484.0): cluster 1\n", + "Example n. 3441 = (68.0, 483.0): cluster 1\n", + "Example n. 3442 = (68.0, 482.0): cluster 1\n", + "Example n. 3443 = (68.0, 481.0): cluster 1\n", + "Example n. 3444 = (68.0, 480.0): cluster 1\n", + "Example n. 3445 = (68.0, 479.0): cluster 1\n", + "Example n. 3446 = (68.0, 478.0): cluster 1\n", + "Example n. 3447 = (68.0, 477.0): cluster 1\n", + "Example n. 3448 = (68.0, 476.0): cluster 1\n", + "Example n. 3449 = (68.0, 475.0): cluster 1\n", + "Example n. 3450 = (68.0, 474.0): cluster 1\n", + "Example n. 3451 = (68.0, 473.0): cluster 1\n", + "Example n. 3452 = (68.0, 472.0): cluster 1\n", + "Example n. 3453 = (68.0, 471.0): cluster 1\n", + "Example n. 3454 = (68.0, 470.0): cluster 1\n", + "Example n. 3455 = (68.0, 469.0): cluster 1\n", + "Example n. 3456 = (68.0, 468.0): cluster 1\n", + "Example n. 3457 = (68.0, 467.0): cluster 1\n", + "Example n. 3458 = (68.0, 466.0): cluster 1\n", + "Example n. 3459 = (68.0, 465.0): cluster 1\n", + "Example n. 3460 = (68.0, 464.0): cluster 1\n", + "Example n. 3461 = (68.0, 463.0): cluster 1\n", + "Example n. 3462 = (68.0, 462.0): cluster 1\n", + "Example n. 3463 = (68.0, 461.0): cluster 1\n", + "Example n. 3464 = (68.0, 460.0): cluster 1\n", + "Example n. 3465 = (68.0, 459.0): cluster 1\n", + "Example n. 3466 = (68.0, 458.0): cluster 1\n", + "Example n. 3467 = (68.0, 457.0): cluster 1\n", + "Example n. 3468 = (68.0, 456.0): cluster 1\n", + "Example n. 3469 = (68.0, 455.0): cluster 1\n", + "Example n. 3470 = (68.0, 454.0): cluster 1\n", + "Example n. 3471 = (68.0, 453.0): cluster 1\n", + "Example n. 3472 = (68.0, 452.0): cluster 1\n", + "Example n. 3473 = (68.0, 451.0): cluster 1\n", + "Example n. 3474 = (68.0, 450.0): cluster 1\n", + "Example n. 3475 = (68.0, 449.0): cluster 1\n", + "Example n. 3476 = (68.0, 448.0): cluster 1\n", + "Example n. 3477 = (68.0, 447.0): cluster 1\n", + "Example n. 3478 = (68.0, 446.0): cluster 1\n", + "Example n. 3479 = (68.0, 445.0): cluster 1\n", + "Example n. 3480 = (68.0, 444.0): cluster 1\n", + "Example n. 3481 = (68.0, 443.0): cluster 1\n", + "Example n. 3482 = (68.0, 442.0): cluster 1\n", + "Example n. 3483 = (68.0, 441.0): cluster 3\n", + "Example n. 3484 = (68.0, 440.0): cluster 3\n", + "Example n. 3485 = (68.0, 439.0): cluster 3\n", + "Example n. 3486 = (68.0, 438.0): cluster 3\n", + "Example n. 3487 = (68.0, 437.0): cluster 3\n", + "Example n. 3488 = (68.0, 429.0): cluster 3\n", + "Example n. 3489 = (68.0, 428.0): cluster 3\n", + "Example n. 3490 = (68.0, 427.0): cluster 3\n", + "Example n. 3491 = (68.0, 426.0): cluster 3\n", + "Example n. 3492 = (68.0, 425.0): cluster 3\n", + "Example n. 3493 = (68.0, 424.0): cluster 3\n", + "Example n. 3494 = (68.0, 423.0): cluster 3\n", + "Example n. 3495 = (68.0, 422.0): cluster 3\n", + "Example n. 3496 = (68.0, 421.0): cluster 3\n", + "Example n. 3497 = (68.0, 420.0): cluster 3\n", + "Example n. 3498 = (68.0, 419.0): cluster 3\n", + "Example n. 3499 = (68.0, 418.0): cluster 3\n", + "Example n. 3500 = (68.0, 417.0): cluster 3\n", + "Example n. 3501 = (68.0, 409.0): cluster 3\n", + "Example n. 3502 = (68.0, 408.0): cluster 3\n", + "Example n. 3503 = (68.0, 407.0): cluster 3\n", + "Example n. 3504 = (68.0, 406.0): cluster 3\n", + "Example n. 3505 = (68.0, 405.0): cluster 3\n", + "Example n. 3506 = (68.0, 404.0): cluster 3\n", + "Example n. 3507 = (68.0, 403.0): cluster 3\n", + "Example n. 3508 = (68.0, 402.0): cluster 3\n", + "Example n. 3509 = (68.0, 401.0): cluster 3\n", + "Example n. 3510 = (68.0, 400.0): cluster 3\n", + "Example n. 3511 = (68.0, 399.0): cluster 3\n", + "Example n. 3512 = (68.0, 398.0): cluster 3\n", + "Example n. 3513 = (68.0, 397.0): cluster 3\n", + "Example n. 3514 = (68.0, 396.0): cluster 3\n", + "Example n. 3515 = (68.0, 395.0): cluster 3\n", + "Example n. 3516 = (68.0, 394.0): cluster 3\n", + "Example n. 3517 = (69.0, 486.0): cluster 1\n", + "Example n. 3518 = (69.0, 485.0): cluster 1\n", + "Example n. 3519 = (69.0, 484.0): cluster 1\n", + "Example n. 3520 = (69.0, 483.0): cluster 1\n", + "Example n. 3521 = (69.0, 482.0): cluster 1\n", + "Example n. 3522 = (69.0, 481.0): cluster 1\n", + "Example n. 3523 = (69.0, 480.0): cluster 1\n", + "Example n. 3524 = (69.0, 479.0): cluster 1\n", + "Example n. 3525 = (69.0, 478.0): cluster 1\n", + "Example n. 3526 = (69.0, 477.0): cluster 1\n", + "Example n. 3527 = (69.0, 476.0): cluster 1\n", + "Example n. 3528 = (69.0, 475.0): cluster 1\n", + "Example n. 3529 = (69.0, 474.0): cluster 1\n", + "Example n. 3530 = (69.0, 473.0): cluster 1\n", + "Example n. 3531 = (69.0, 472.0): cluster 1\n", + "Example n. 3532 = (69.0, 471.0): cluster 1\n", + "Example n. 3533 = (69.0, 470.0): cluster 1\n", + "Example n. 3534 = (69.0, 469.0): cluster 1\n", + "Example n. 3535 = (69.0, 468.0): cluster 1\n", + "Example n. 3536 = (69.0, 467.0): cluster 1\n", + "Example n. 3537 = (69.0, 466.0): cluster 1\n", + "Example n. 3538 = (69.0, 465.0): cluster 1\n", + "Example n. 3539 = (69.0, 464.0): cluster 1\n", + "Example n. 3540 = (69.0, 463.0): cluster 1\n", + "Example n. 3541 = (69.0, 462.0): cluster 1\n", + "Example n. 3542 = (69.0, 461.0): cluster 1\n", + "Example n. 3543 = (69.0, 460.0): cluster 1\n", + "Example n. 3544 = (69.0, 459.0): cluster 1\n", + "Example n. 3545 = (69.0, 458.0): cluster 1\n", + "Example n. 3546 = (69.0, 457.0): cluster 1\n", + "Example n. 3547 = (69.0, 456.0): cluster 1\n", + "Example n. 3548 = (69.0, 455.0): cluster 1\n", + "Example n. 3549 = (69.0, 454.0): cluster 1\n", + "Example n. 3550 = (69.0, 453.0): cluster 1\n", + "Example n. 3551 = (69.0, 452.0): cluster 1\n", + "Example n. 3552 = (69.0, 451.0): cluster 1\n", + "Example n. 3553 = (69.0, 450.0): cluster 1\n", + "Example n. 3554 = (69.0, 449.0): cluster 1\n", + "Example n. 3555 = (69.0, 448.0): cluster 1\n", + "Example n. 3556 = (69.0, 447.0): cluster 1\n", + "Example n. 3557 = (69.0, 446.0): cluster 1\n", + "Example n. 3558 = (69.0, 445.0): cluster 1\n", + "Example n. 3559 = (69.0, 444.0): cluster 1\n", + "Example n. 3560 = (69.0, 443.0): cluster 1\n", + "Example n. 3561 = (69.0, 442.0): cluster 1\n", + "Example n. 3562 = (69.0, 441.0): cluster 3\n", + "Example n. 3563 = (69.0, 440.0): cluster 3\n", + "Example n. 3564 = (69.0, 439.0): cluster 3\n", + "Example n. 3565 = (69.0, 438.0): cluster 3\n", + "Example n. 3566 = (69.0, 437.0): cluster 3\n", + "Example n. 3567 = (69.0, 429.0): cluster 3\n", + "Example n. 3568 = (69.0, 428.0): cluster 3\n", + "Example n. 3569 = (69.0, 427.0): cluster 3\n", + "Example n. 3570 = (69.0, 426.0): cluster 3\n", + "Example n. 3571 = (69.0, 425.0): cluster 3\n", + "Example n. 3572 = (69.0, 424.0): cluster 3\n", + "Example n. 3573 = (69.0, 423.0): cluster 3\n", + "Example n. 3574 = (69.0, 422.0): cluster 3\n", + "Example n. 3575 = (69.0, 421.0): cluster 3\n", + "Example n. 3576 = (69.0, 420.0): cluster 3\n", + "Example n. 3577 = (69.0, 419.0): cluster 3\n", + "Example n. 3578 = (69.0, 418.0): cluster 3\n", + "Example n. 3579 = (69.0, 417.0): cluster 3\n", + "Example n. 3580 = (69.0, 409.0): cluster 3\n", + "Example n. 3581 = (69.0, 408.0): cluster 3\n", + "Example n. 3582 = (69.0, 407.0): cluster 3\n", + "Example n. 3583 = (69.0, 406.0): cluster 3\n", + "Example n. 3584 = (69.0, 405.0): cluster 3\n", + "Example n. 3585 = (69.0, 404.0): cluster 3\n", + "Example n. 3586 = (69.0, 403.0): cluster 3\n", + "Example n. 3587 = (69.0, 402.0): cluster 3\n", + "Example n. 3588 = (69.0, 401.0): cluster 3\n", + "Example n. 3589 = (69.0, 400.0): cluster 3\n", + "Example n. 3590 = (69.0, 399.0): cluster 3\n", + "Example n. 3591 = (69.0, 398.0): cluster 3\n", + "Example n. 3592 = (69.0, 397.0): cluster 3\n", + "Example n. 3593 = (69.0, 396.0): cluster 3\n", + "Example n. 3594 = (69.0, 395.0): cluster 3\n", + "Example n. 3595 = (69.0, 394.0): cluster 3\n", + "Example n. 3596 = (70.0, 486.0): cluster 1\n", + "Example n. 3597 = (70.0, 485.0): cluster 1\n", + "Example n. 3598 = (70.0, 484.0): cluster 1\n", + "Example n. 3599 = (70.0, 483.0): cluster 1\n", + "Example n. 3600 = (70.0, 482.0): cluster 1\n", + "Example n. 3601 = (70.0, 481.0): cluster 1\n", + "Example n. 3602 = (70.0, 480.0): cluster 1\n", + "Example n. 3603 = (70.0, 479.0): cluster 1\n", + "Example n. 3604 = (70.0, 478.0): cluster 1\n", + "Example n. 3605 = (70.0, 477.0): cluster 1\n", + "Example n. 3606 = (70.0, 476.0): cluster 1\n", + "Example n. 3607 = (70.0, 475.0): cluster 1\n", + "Example n. 3608 = (70.0, 474.0): cluster 1\n", + "Example n. 3609 = (70.0, 473.0): cluster 1\n", + "Example n. 3610 = (70.0, 472.0): cluster 1\n", + "Example n. 3611 = (70.0, 471.0): cluster 1\n", + "Example n. 3612 = (70.0, 470.0): cluster 1\n", + "Example n. 3613 = (70.0, 469.0): cluster 1\n", + "Example n. 3614 = (70.0, 468.0): cluster 1\n", + "Example n. 3615 = (70.0, 467.0): cluster 1\n", + "Example n. 3616 = (70.0, 466.0): cluster 1\n", + "Example n. 3617 = (70.0, 465.0): cluster 1\n", + "Example n. 3618 = (70.0, 464.0): cluster 1\n", + "Example n. 3619 = (70.0, 463.0): cluster 1\n", + "Example n. 3620 = (70.0, 462.0): cluster 1\n", + "Example n. 3621 = (70.0, 461.0): cluster 1\n", + "Example n. 3622 = (70.0, 460.0): cluster 1\n", + "Example n. 3623 = (70.0, 459.0): cluster 1\n", + "Example n. 3624 = (70.0, 458.0): cluster 1\n", + "Example n. 3625 = (70.0, 457.0): cluster 1\n", + "Example n. 3626 = (70.0, 456.0): cluster 1\n", + "Example n. 3627 = (70.0, 455.0): cluster 1\n", + "Example n. 3628 = (70.0, 454.0): cluster 1\n", + "Example n. 3629 = (70.0, 453.0): cluster 1\n", + "Example n. 3630 = (70.0, 452.0): cluster 1\n", + "Example n. 3631 = (70.0, 451.0): cluster 1\n", + "Example n. 3632 = (70.0, 450.0): cluster 1\n", + "Example n. 3633 = (70.0, 449.0): cluster 1\n", + "Example n. 3634 = (70.0, 448.0): cluster 1\n", + "Example n. 3635 = (70.0, 447.0): cluster 1\n", + "Example n. 3636 = (70.0, 446.0): cluster 1\n", + "Example n. 3637 = (70.0, 445.0): cluster 1\n", + "Example n. 3638 = (70.0, 444.0): cluster 1\n", + "Example n. 3639 = (70.0, 443.0): cluster 1\n", + "Example n. 3640 = (70.0, 442.0): cluster 1\n", + "Example n. 3641 = (70.0, 441.0): cluster 3\n", + "Example n. 3642 = (70.0, 440.0): cluster 3\n", + "Example n. 3643 = (70.0, 439.0): cluster 3\n", + "Example n. 3644 = (70.0, 438.0): cluster 3\n", + "Example n. 3645 = (70.0, 437.0): cluster 3\n", + "Example n. 3646 = (70.0, 432.0): cluster 3\n", + "Example n. 3647 = (70.0, 431.0): cluster 3\n", + "Example n. 3648 = (70.0, 430.0): cluster 3\n", + "Example n. 3649 = (70.0, 429.0): cluster 3\n", + "Example n. 3650 = (70.0, 428.0): cluster 3\n", + "Example n. 3651 = (70.0, 427.0): cluster 3\n", + "Example n. 3652 = (70.0, 426.0): cluster 3\n", + "Example n. 3653 = (70.0, 425.0): cluster 3\n", + "Example n. 3654 = (70.0, 424.0): cluster 3\n", + "Example n. 3655 = (70.0, 423.0): cluster 3\n", + "Example n. 3656 = (70.0, 422.0): cluster 3\n", + "Example n. 3657 = (70.0, 421.0): cluster 3\n", + "Example n. 3658 = (70.0, 420.0): cluster 3\n", + "Example n. 3659 = (70.0, 419.0): cluster 3\n", + "Example n. 3660 = (70.0, 418.0): cluster 3\n", + "Example n. 3661 = (70.0, 417.0): cluster 3\n", + "Example n. 3662 = (70.0, 413.0): cluster 3\n", + "Example n. 3663 = (70.0, 412.0): cluster 3\n", + "Example n. 3664 = (70.0, 411.0): cluster 3\n", + "Example n. 3665 = (70.0, 410.0): cluster 3\n", + "Example n. 3666 = (70.0, 409.0): cluster 3\n", + "Example n. 3667 = (70.0, 408.0): cluster 3\n", + "Example n. 3668 = (70.0, 407.0): cluster 3\n", + "Example n. 3669 = (70.0, 406.0): cluster 3\n", + "Example n. 3670 = (70.0, 405.0): cluster 3\n", + "Example n. 3671 = (70.0, 404.0): cluster 3\n", + "Example n. 3672 = (70.0, 403.0): cluster 3\n", + "Example n. 3673 = (70.0, 402.0): cluster 3\n", + "Example n. 3674 = (70.0, 401.0): cluster 3\n", + "Example n. 3675 = (70.0, 400.0): cluster 3\n", + "Example n. 3676 = (70.0, 399.0): cluster 3\n", + "Example n. 3677 = (70.0, 398.0): cluster 3\n", + "Example n. 3678 = (70.0, 397.0): cluster 3\n", + "Example n. 3679 = (70.0, 396.0): cluster 3\n", + "Example n. 3680 = (70.0, 395.0): cluster 3\n", + "Example n. 3681 = (70.0, 394.0): cluster 3\n", + "Example n. 3682 = (71.0, 482.0): cluster 1\n", + "Example n. 3683 = (71.0, 481.0): cluster 1\n", + "Example n. 3684 = (71.0, 480.0): cluster 1\n", + "Example n. 3685 = (71.0, 479.0): cluster 1\n", + "Example n. 3686 = (71.0, 478.0): cluster 1\n", + "Example n. 3687 = (71.0, 477.0): cluster 1\n", + "Example n. 3688 = (71.0, 476.0): cluster 1\n", + "Example n. 3689 = (71.0, 475.0): cluster 1\n", + "Example n. 3690 = (71.0, 474.0): cluster 1\n", + "Example n. 3691 = (71.0, 473.0): cluster 1\n", + "Example n. 3692 = (71.0, 472.0): cluster 1\n", + "Example n. 3693 = (71.0, 471.0): cluster 1\n", + "Example n. 3694 = (71.0, 470.0): cluster 1\n", + "Example n. 3695 = (71.0, 469.0): cluster 1\n", + "Example n. 3696 = (71.0, 468.0): cluster 1\n", + "Example n. 3697 = (71.0, 467.0): cluster 1\n", + "Example n. 3698 = (71.0, 466.0): cluster 1\n", + "Example n. 3699 = (71.0, 465.0): cluster 1\n", + "Example n. 3700 = (71.0, 464.0): cluster 1\n", + "Example n. 3701 = (71.0, 463.0): cluster 1\n", + "Example n. 3702 = (71.0, 462.0): cluster 1\n", + "Example n. 3703 = (71.0, 461.0): cluster 1\n", + "Example n. 3704 = (71.0, 460.0): cluster 1\n", + "Example n. 3705 = (71.0, 459.0): cluster 1\n", + "Example n. 3706 = (71.0, 458.0): cluster 1\n", + "Example n. 3707 = (71.0, 457.0): cluster 1\n", + "Example n. 3708 = (71.0, 456.0): cluster 1\n", + "Example n. 3709 = (71.0, 455.0): cluster 1\n", + "Example n. 3710 = (71.0, 454.0): cluster 1\n", + "Example n. 3711 = (71.0, 453.0): cluster 1\n", + "Example n. 3712 = (71.0, 452.0): cluster 1\n", + "Example n. 3713 = (71.0, 451.0): cluster 1\n", + "Example n. 3714 = (71.0, 450.0): cluster 1\n", + "Example n. 3715 = (71.0, 449.0): cluster 1\n", + "Example n. 3716 = (71.0, 448.0): cluster 1\n", + "Example n. 3717 = (71.0, 447.0): cluster 1\n", + "Example n. 3718 = (71.0, 446.0): cluster 1\n", + "Example n. 3719 = (71.0, 445.0): cluster 1\n", + "Example n. 3720 = (71.0, 444.0): cluster 1\n", + "Example n. 3721 = (71.0, 443.0): cluster 1\n", + "Example n. 3722 = (71.0, 442.0): cluster 1\n", + "Example n. 3723 = (71.0, 441.0): cluster 3\n", + "Example n. 3724 = (71.0, 432.0): cluster 3\n", + "Example n. 3725 = (71.0, 431.0): cluster 3\n", + "Example n. 3726 = (71.0, 430.0): cluster 3\n", + "Example n. 3727 = (71.0, 429.0): cluster 3\n", + "Example n. 3728 = (71.0, 428.0): cluster 3\n", + "Example n. 3729 = (71.0, 427.0): cluster 3\n", + "Example n. 3730 = (71.0, 426.0): cluster 3\n", + "Example n. 3731 = (71.0, 425.0): cluster 3\n", + "Example n. 3732 = (71.0, 424.0): cluster 3\n", + "Example n. 3733 = (71.0, 423.0): cluster 3\n", + "Example n. 3734 = (71.0, 422.0): cluster 3\n", + "Example n. 3735 = (71.0, 421.0): cluster 3\n", + "Example n. 3736 = (71.0, 420.0): cluster 3\n", + "Example n. 3737 = (71.0, 419.0): cluster 3\n", + "Example n. 3738 = (71.0, 418.0): cluster 3\n", + "Example n. 3739 = (71.0, 417.0): cluster 3\n", + "Example n. 3740 = (71.0, 413.0): cluster 3\n", + "Example n. 3741 = (71.0, 412.0): cluster 3\n", + "Example n. 3742 = (71.0, 411.0): cluster 3\n", + "Example n. 3743 = (71.0, 410.0): cluster 3\n", + "Example n. 3744 = (71.0, 409.0): cluster 3\n", + "Example n. 3745 = (71.0, 408.0): cluster 3\n", + "Example n. 3746 = (71.0, 407.0): cluster 3\n", + "Example n. 3747 = (71.0, 406.0): cluster 3\n", + "Example n. 3748 = (71.0, 405.0): cluster 3\n", + "Example n. 3749 = (71.0, 404.0): cluster 3\n", + "Example n. 3750 = (71.0, 403.0): cluster 3\n", + "Example n. 3751 = (71.0, 402.0): cluster 3\n", + "Example n. 3752 = (71.0, 401.0): cluster 3\n", + "Example n. 3753 = (71.0, 400.0): cluster 3\n", + "Example n. 3754 = (71.0, 399.0): cluster 3\n", + "Example n. 3755 = (71.0, 398.0): cluster 3\n", + "Example n. 3756 = (72.0, 482.0): cluster 1\n", + "Example n. 3757 = (72.0, 481.0): cluster 1\n", + "Example n. 3758 = (72.0, 480.0): cluster 1\n", + "Example n. 3759 = (72.0, 479.0): cluster 1\n", + "Example n. 3760 = (72.0, 478.0): cluster 1\n", + "Example n. 3761 = (72.0, 477.0): cluster 1\n", + "Example n. 3762 = (72.0, 476.0): cluster 1\n", + "Example n. 3763 = (72.0, 475.0): cluster 1\n", + "Example n. 3764 = (72.0, 474.0): cluster 1\n", + "Example n. 3765 = (72.0, 473.0): cluster 1\n", + "Example n. 3766 = (72.0, 472.0): cluster 1\n", + "Example n. 3767 = (72.0, 471.0): cluster 1\n", + "Example n. 3768 = (72.0, 470.0): cluster 1\n", + "Example n. 3769 = (72.0, 469.0): cluster 1\n", + "Example n. 3770 = (72.0, 468.0): cluster 1\n", + "Example n. 3771 = (72.0, 467.0): cluster 1\n", + "Example n. 3772 = (72.0, 466.0): cluster 1\n", + "Example n. 3773 = (72.0, 465.0): cluster 1\n", + "Example n. 3774 = (72.0, 464.0): cluster 1\n", + "Example n. 3775 = (72.0, 463.0): cluster 1\n", + "Example n. 3776 = (72.0, 462.0): cluster 1\n", + "Example n. 3777 = (72.0, 461.0): cluster 1\n", + "Example n. 3778 = (72.0, 460.0): cluster 1\n", + "Example n. 3779 = (72.0, 459.0): cluster 1\n", + "Example n. 3780 = (72.0, 458.0): cluster 1\n", + "Example n. 3781 = (72.0, 457.0): cluster 1\n", + "Example n. 3782 = (72.0, 456.0): cluster 1\n", + "Example n. 3783 = (72.0, 455.0): cluster 1\n", + "Example n. 3784 = (72.0, 454.0): cluster 1\n", + "Example n. 3785 = (72.0, 453.0): cluster 1\n", + "Example n. 3786 = (72.0, 452.0): cluster 1\n", + "Example n. 3787 = (72.0, 451.0): cluster 1\n", + "Example n. 3788 = (72.0, 450.0): cluster 1\n", + "Example n. 3789 = (72.0, 449.0): cluster 1\n", + "Example n. 3790 = (72.0, 448.0): cluster 1\n", + "Example n. 3791 = (72.0, 447.0): cluster 1\n", + "Example n. 3792 = (72.0, 446.0): cluster 1\n", + "Example n. 3793 = (72.0, 445.0): cluster 1\n", + "Example n. 3794 = (72.0, 444.0): cluster 1\n", + "Example n. 3795 = (72.0, 443.0): cluster 1\n", + "Example n. 3796 = (72.0, 442.0): cluster 1\n", + "Example n. 3797 = (72.0, 441.0): cluster 3\n", + "Example n. 3798 = (72.0, 432.0): cluster 3\n", + "Example n. 3799 = (72.0, 431.0): cluster 3\n", + "Example n. 3800 = (72.0, 430.0): cluster 3\n", + "Example n. 3801 = (72.0, 429.0): cluster 3\n", + "Example n. 3802 = (72.0, 428.0): cluster 3\n", + "Example n. 3803 = (72.0, 427.0): cluster 3\n", + "Example n. 3804 = (72.0, 426.0): cluster 3\n", + "Example n. 3805 = (72.0, 425.0): cluster 3\n", + "Example n. 3806 = (72.0, 424.0): cluster 3\n", + "Example n. 3807 = (72.0, 423.0): cluster 3\n", + "Example n. 3808 = (72.0, 422.0): cluster 3\n", + "Example n. 3809 = (72.0, 421.0): cluster 3\n", + "Example n. 3810 = (72.0, 420.0): cluster 3\n", + "Example n. 3811 = (72.0, 419.0): cluster 3\n", + "Example n. 3812 = (72.0, 418.0): cluster 3\n", + "Example n. 3813 = (72.0, 417.0): cluster 3\n", + "Example n. 3814 = (72.0, 413.0): cluster 3\n", + "Example n. 3815 = (72.0, 412.0): cluster 3\n", + "Example n. 3816 = (72.0, 411.0): cluster 3\n", + "Example n. 3817 = (72.0, 410.0): cluster 3\n", + "Example n. 3818 = (72.0, 409.0): cluster 3\n", + "Example n. 3819 = (72.0, 408.0): cluster 3\n", + "Example n. 3820 = (72.0, 407.0): cluster 3\n", + "Example n. 3821 = (72.0, 406.0): cluster 3\n", + "Example n. 3822 = (72.0, 405.0): cluster 3\n", + "Example n. 3823 = (72.0, 404.0): cluster 3\n", + "Example n. 3824 = (72.0, 403.0): cluster 3\n", + "Example n. 3825 = (72.0, 402.0): cluster 3\n", + "Example n. 3826 = (72.0, 401.0): cluster 3\n", + "Example n. 3827 = (72.0, 400.0): cluster 3\n", + "Example n. 3828 = (72.0, 399.0): cluster 3\n", + "Example n. 3829 = (72.0, 398.0): cluster 3\n", + "Example n. 3830 = (73.0, 482.0): cluster 1\n", + "Example n. 3831 = (73.0, 481.0): cluster 1\n", + "Example n. 3832 = (73.0, 480.0): cluster 1\n", + "Example n. 3833 = (73.0, 479.0): cluster 1\n", + "Example n. 3834 = (73.0, 478.0): cluster 1\n", + "Example n. 3835 = (73.0, 477.0): cluster 1\n", + "Example n. 3836 = (73.0, 476.0): cluster 1\n", + "Example n. 3837 = (73.0, 475.0): cluster 1\n", + "Example n. 3838 = (73.0, 474.0): cluster 1\n", + "Example n. 3839 = (73.0, 473.0): cluster 1\n", + "Example n. 3840 = (73.0, 472.0): cluster 1\n", + "Example n. 3841 = (73.0, 471.0): cluster 1\n", + "Example n. 3842 = (73.0, 470.0): cluster 1\n", + "Example n. 3843 = (73.0, 469.0): cluster 1\n", + "Example n. 3844 = (73.0, 468.0): cluster 1\n", + "Example n. 3845 = (73.0, 467.0): cluster 1\n", + "Example n. 3846 = (73.0, 466.0): cluster 1\n", + "Example n. 3847 = (73.0, 465.0): cluster 1\n", + "Example n. 3848 = (73.0, 464.0): cluster 1\n", + "Example n. 3849 = (73.0, 463.0): cluster 1\n", + "Example n. 3850 = (73.0, 462.0): cluster 1\n", + "Example n. 3851 = (73.0, 461.0): cluster 1\n", + "Example n. 3852 = (73.0, 460.0): cluster 1\n", + "Example n. 3853 = (73.0, 459.0): cluster 1\n", + "Example n. 3854 = (73.0, 458.0): cluster 1\n", + "Example n. 3855 = (73.0, 457.0): cluster 1\n", + "Example n. 3856 = (73.0, 456.0): cluster 1\n", + "Example n. 3857 = (73.0, 455.0): cluster 1\n", + "Example n. 3858 = (73.0, 454.0): cluster 1\n", + "Example n. 3859 = (73.0, 453.0): cluster 1\n", + "Example n. 3860 = (73.0, 452.0): cluster 1\n", + "Example n. 3861 = (73.0, 451.0): cluster 1\n", + "Example n. 3862 = (73.0, 450.0): cluster 1\n", + "Example n. 3863 = (73.0, 449.0): cluster 1\n", + "Example n. 3864 = (73.0, 448.0): cluster 1\n", + "Example n. 3865 = (73.0, 447.0): cluster 1\n", + "Example n. 3866 = (73.0, 446.0): cluster 1\n", + "Example n. 3867 = (73.0, 445.0): cluster 1\n", + "Example n. 3868 = (73.0, 444.0): cluster 1\n", + "Example n. 3869 = (73.0, 443.0): cluster 1\n", + "Example n. 3870 = (73.0, 442.0): cluster 1\n", + "Example n. 3871 = (73.0, 441.0): cluster 3\n", + "Example n. 3872 = (73.0, 432.0): cluster 3\n", + "Example n. 3873 = (73.0, 431.0): cluster 3\n", + "Example n. 3874 = (73.0, 430.0): cluster 3\n", + "Example n. 3875 = (73.0, 429.0): cluster 3\n", + "Example n. 3876 = (73.0, 428.0): cluster 3\n", + "Example n. 3877 = (73.0, 427.0): cluster 3\n", + "Example n. 3878 = (73.0, 426.0): cluster 3\n", + "Example n. 3879 = (73.0, 425.0): cluster 3\n", + "Example n. 3880 = (73.0, 424.0): cluster 3\n", + "Example n. 3881 = (73.0, 423.0): cluster 3\n", + "Example n. 3882 = (73.0, 422.0): cluster 3\n", + "Example n. 3883 = (73.0, 421.0): cluster 3\n", + "Example n. 3884 = (73.0, 420.0): cluster 3\n", + "Example n. 3885 = (73.0, 419.0): cluster 3\n", + "Example n. 3886 = (73.0, 418.0): cluster 3\n", + "Example n. 3887 = (73.0, 417.0): cluster 3\n", + "Example n. 3888 = (73.0, 413.0): cluster 3\n", + "Example n. 3889 = (73.0, 412.0): cluster 3\n", + "Example n. 3890 = (73.0, 411.0): cluster 3\n", + "Example n. 3891 = (73.0, 410.0): cluster 3\n", + "Example n. 3892 = (73.0, 409.0): cluster 3\n", + "Example n. 3893 = (73.0, 408.0): cluster 3\n", + "Example n. 3894 = (73.0, 407.0): cluster 3\n", + "Example n. 3895 = (73.0, 406.0): cluster 3\n", + "Example n. 3896 = (73.0, 405.0): cluster 3\n", + "Example n. 3897 = (73.0, 404.0): cluster 3\n", + "Example n. 3898 = (73.0, 403.0): cluster 3\n", + "Example n. 3899 = (73.0, 402.0): cluster 3\n", + "Example n. 3900 = (73.0, 401.0): cluster 3\n", + "Example n. 3901 = (73.0, 400.0): cluster 3\n", + "Example n. 3902 = (73.0, 399.0): cluster 3\n", + "Example n. 3903 = (73.0, 398.0): cluster 3\n", + "Example n. 3904 = (74.0, 482.0): cluster 1\n", + "Example n. 3905 = (74.0, 481.0): cluster 1\n", + "Example n. 3906 = (74.0, 480.0): cluster 1\n", + "Example n. 3907 = (74.0, 479.0): cluster 1\n", + "Example n. 3908 = (74.0, 478.0): cluster 1\n", + "Example n. 3909 = (74.0, 477.0): cluster 1\n", + "Example n. 3910 = (74.0, 476.0): cluster 1\n", + "Example n. 3911 = (74.0, 475.0): cluster 1\n", + "Example n. 3912 = (74.0, 474.0): cluster 1\n", + "Example n. 3913 = (74.0, 473.0): cluster 1\n", + "Example n. 3914 = (74.0, 472.0): cluster 1\n", + "Example n. 3915 = (74.0, 471.0): cluster 1\n", + "Example n. 3916 = (74.0, 470.0): cluster 1\n", + "Example n. 3917 = (74.0, 469.0): cluster 1\n", + "Example n. 3918 = (74.0, 468.0): cluster 1\n", + "Example n. 3919 = (74.0, 467.0): cluster 1\n", + "Example n. 3920 = (74.0, 466.0): cluster 1\n", + "Example n. 3921 = (74.0, 465.0): cluster 1\n", + "Example n. 3922 = (74.0, 464.0): cluster 1\n", + "Example n. 3923 = (74.0, 463.0): cluster 1\n", + "Example n. 3924 = (74.0, 462.0): cluster 1\n", + "Example n. 3925 = (74.0, 461.0): cluster 1\n", + "Example n. 3926 = (74.0, 460.0): cluster 1\n", + "Example n. 3927 = (74.0, 459.0): cluster 1\n", + "Example n. 3928 = (74.0, 458.0): cluster 1\n", + "Example n. 3929 = (74.0, 457.0): cluster 1\n", + "Example n. 3930 = (74.0, 456.0): cluster 1\n", + "Example n. 3931 = (74.0, 455.0): cluster 1\n", + "Example n. 3932 = (74.0, 454.0): cluster 1\n", + "Example n. 3933 = (74.0, 453.0): cluster 1\n", + "Example n. 3934 = (74.0, 452.0): cluster 1\n", + "Example n. 3935 = (74.0, 451.0): cluster 1\n", + "Example n. 3936 = (74.0, 450.0): cluster 1\n", + "Example n. 3937 = (74.0, 449.0): cluster 1\n", + "Example n. 3938 = (74.0, 448.0): cluster 1\n", + "Example n. 3939 = (74.0, 447.0): cluster 1\n", + "Example n. 3940 = (74.0, 446.0): cluster 1\n", + "Example n. 3941 = (74.0, 445.0): cluster 1\n", + "Example n. 3942 = (74.0, 444.0): cluster 1\n", + "Example n. 3943 = (74.0, 443.0): cluster 1\n", + "Example n. 3944 = (74.0, 442.0): cluster 1\n", + "Example n. 3945 = (74.0, 441.0): cluster 3\n", + "Example n. 3946 = (74.0, 432.0): cluster 3\n", + "Example n. 3947 = (74.0, 431.0): cluster 3\n", + "Example n. 3948 = (74.0, 430.0): cluster 3\n", + "Example n. 3949 = (74.0, 429.0): cluster 3\n", + "Example n. 3950 = (74.0, 428.0): cluster 3\n", + "Example n. 3951 = (74.0, 427.0): cluster 3\n", + "Example n. 3952 = (74.0, 426.0): cluster 3\n", + "Example n. 3953 = (74.0, 425.0): cluster 3\n", + "Example n. 3954 = (74.0, 424.0): cluster 3\n", + "Example n. 3955 = (74.0, 423.0): cluster 3\n", + "Example n. 3956 = (74.0, 422.0): cluster 3\n", + "Example n. 3957 = (74.0, 421.0): cluster 3\n", + "Example n. 3958 = (74.0, 420.0): cluster 3\n", + "Example n. 3959 = (74.0, 419.0): cluster 3\n", + "Example n. 3960 = (74.0, 418.0): cluster 3\n", + "Example n. 3961 = (74.0, 417.0): cluster 3\n", + "Example n. 3962 = (74.0, 416.0): cluster 3\n", + "Example n. 3963 = (74.0, 415.0): cluster 3\n", + "Example n. 3964 = (74.0, 414.0): cluster 3\n", + "Example n. 3965 = (74.0, 413.0): cluster 3\n", + "Example n. 3966 = (74.0, 412.0): cluster 3\n", + "Example n. 3967 = (74.0, 411.0): cluster 3\n", + "Example n. 3968 = (74.0, 410.0): cluster 3\n", + "Example n. 3969 = (74.0, 409.0): cluster 3\n", + "Example n. 3970 = (74.0, 408.0): cluster 3\n", + "Example n. 3971 = (74.0, 407.0): cluster 3\n", + "Example n. 3972 = (74.0, 406.0): cluster 3\n", + "Example n. 3973 = (74.0, 405.0): cluster 3\n", + "Example n. 3974 = (74.0, 404.0): cluster 3\n", + "Example n. 3975 = (74.0, 403.0): cluster 3\n", + "Example n. 3976 = (74.0, 402.0): cluster 3\n", + "Example n. 3977 = (74.0, 401.0): cluster 3\n", + "Example n. 3978 = (74.0, 400.0): cluster 3\n", + "Example n. 3979 = (74.0, 399.0): cluster 3\n", + "Example n. 3980 = (74.0, 398.0): cluster 3\n", + "Example n. 3981 = (75.0, 471.0): cluster 1\n", + "Example n. 3982 = (75.0, 470.0): cluster 1\n", + "Example n. 3983 = (75.0, 469.0): cluster 1\n", + "Example n. 3984 = (75.0, 468.0): cluster 1\n", + "Example n. 3985 = (75.0, 467.0): cluster 1\n", + "Example n. 3986 = (75.0, 466.0): cluster 1\n", + "Example n. 3987 = (75.0, 465.0): cluster 1\n", + "Example n. 3988 = (75.0, 464.0): cluster 1\n", + "Example n. 3989 = (75.0, 463.0): cluster 1\n", + "Example n. 3990 = (75.0, 462.0): cluster 1\n", + "Example n. 3991 = (75.0, 461.0): cluster 1\n", + "Example n. 3992 = (75.0, 460.0): cluster 1\n", + "Example n. 3993 = (75.0, 459.0): cluster 1\n", + "Example n. 3994 = (75.0, 458.0): cluster 1\n", + "Example n. 3995 = (75.0, 457.0): cluster 1\n", + "Example n. 3996 = (75.0, 456.0): cluster 1\n", + "Example n. 3997 = (75.0, 455.0): cluster 1\n", + "Example n. 3998 = (75.0, 454.0): cluster 1\n", + "Example n. 3999 = (75.0, 453.0): cluster 1\n", + "Example n. 4000 = (75.0, 452.0): cluster 1\n", + "Example n. 4001 = (75.0, 451.0): cluster 1\n", + "Example n. 4002 = (75.0, 450.0): cluster 1\n", + "Example n. 4003 = (75.0, 449.0): cluster 1\n", + "Example n. 4004 = (75.0, 448.0): cluster 1\n", + "Example n. 4005 = (75.0, 447.0): cluster 1\n", + "Example n. 4006 = (75.0, 446.0): cluster 1\n", + "Example n. 4007 = (75.0, 445.0): cluster 1\n", + "Example n. 4008 = (75.0, 444.0): cluster 1\n", + "Example n. 4009 = (75.0, 432.0): cluster 3\n", + "Example n. 4010 = (75.0, 431.0): cluster 3\n", + "Example n. 4011 = (75.0, 430.0): cluster 3\n", + "Example n. 4012 = (75.0, 429.0): cluster 3\n", + "Example n. 4013 = (75.0, 428.0): cluster 3\n", + "Example n. 4014 = (75.0, 427.0): cluster 3\n", + "Example n. 4015 = (75.0, 426.0): cluster 3\n", + "Example n. 4016 = (75.0, 425.0): cluster 3\n", + "Example n. 4017 = (75.0, 424.0): cluster 3\n", + "Example n. 4018 = (75.0, 423.0): cluster 3\n", + "Example n. 4019 = (75.0, 422.0): cluster 3\n", + "Example n. 4020 = (75.0, 421.0): cluster 3\n", + "Example n. 4021 = (75.0, 417.0): cluster 3\n", + "Example n. 4022 = (75.0, 416.0): cluster 3\n", + "Example n. 4023 = (75.0, 415.0): cluster 3\n", + "Example n. 4024 = (75.0, 414.0): cluster 3\n", + "Example n. 4025 = (75.0, 413.0): cluster 3\n", + "Example n. 4026 = (75.0, 412.0): cluster 3\n", + "Example n. 4027 = (75.0, 411.0): cluster 3\n", + "Example n. 4028 = (75.0, 410.0): cluster 3\n", + "Example n. 4029 = (75.0, 409.0): cluster 3\n", + "Example n. 4030 = (75.0, 408.0): cluster 3\n", + "Example n. 4031 = (75.0, 407.0): cluster 3\n", + "Example n. 4032 = (75.0, 406.0): cluster 3\n", + "Example n. 4033 = (75.0, 405.0): cluster 3\n", + "Example n. 4034 = (75.0, 404.0): cluster 3\n", + "Example n. 4035 = (75.0, 403.0): cluster 3\n", + "Example n. 4036 = (75.0, 402.0): cluster 3\n", + "Example n. 4037 = (75.0, 401.0): cluster 3\n", + "Example n. 4038 = (75.0, 400.0): cluster 3\n", + "Example n. 4039 = (75.0, 399.0): cluster 3\n", + "Example n. 4040 = (75.0, 398.0): cluster 3\n", + "Example n. 4041 = (76.0, 471.0): cluster 1\n", + "Example n. 4042 = (76.0, 470.0): cluster 1\n", + "Example n. 4043 = (76.0, 469.0): cluster 1\n", + "Example n. 4044 = (76.0, 468.0): cluster 1\n", + "Example n. 4045 = (76.0, 467.0): cluster 1\n", + "Example n. 4046 = (76.0, 466.0): cluster 1\n", + "Example n. 4047 = (76.0, 465.0): cluster 1\n", + "Example n. 4048 = (76.0, 464.0): cluster 1\n", + "Example n. 4049 = (76.0, 463.0): cluster 1\n", + "Example n. 4050 = (76.0, 462.0): cluster 1\n", + "Example n. 4051 = (76.0, 461.0): cluster 1\n", + "Example n. 4052 = (76.0, 460.0): cluster 1\n", + "Example n. 4053 = (76.0, 459.0): cluster 1\n", + "Example n. 4054 = (76.0, 458.0): cluster 1\n", + "Example n. 4055 = (76.0, 457.0): cluster 1\n", + "Example n. 4056 = (76.0, 456.0): cluster 1\n", + "Example n. 4057 = (76.0, 455.0): cluster 1\n", + "Example n. 4058 = (76.0, 454.0): cluster 1\n", + "Example n. 4059 = (76.0, 453.0): cluster 1\n", + "Example n. 4060 = (76.0, 452.0): cluster 1\n", + "Example n. 4061 = (76.0, 451.0): cluster 1\n", + "Example n. 4062 = (76.0, 450.0): cluster 1\n", + "Example n. 4063 = (76.0, 449.0): cluster 1\n", + "Example n. 4064 = (76.0, 448.0): cluster 1\n", + "Example n. 4065 = (76.0, 447.0): cluster 1\n", + "Example n. 4066 = (76.0, 446.0): cluster 1\n", + "Example n. 4067 = (76.0, 445.0): cluster 1\n", + "Example n. 4068 = (76.0, 444.0): cluster 1\n", + "Example n. 4069 = (76.0, 432.0): cluster 3\n", + "Example n. 4070 = (76.0, 431.0): cluster 3\n", + "Example n. 4071 = (76.0, 430.0): cluster 3\n", + "Example n. 4072 = (76.0, 429.0): cluster 3\n", + "Example n. 4073 = (76.0, 428.0): cluster 3\n", + "Example n. 4074 = (76.0, 427.0): cluster 3\n", + "Example n. 4075 = (76.0, 426.0): cluster 3\n", + "Example n. 4076 = (76.0, 425.0): cluster 3\n", + "Example n. 4077 = (76.0, 424.0): cluster 3\n", + "Example n. 4078 = (76.0, 423.0): cluster 3\n", + "Example n. 4079 = (76.0, 422.0): cluster 3\n", + "Example n. 4080 = (76.0, 421.0): cluster 3\n", + "Example n. 4081 = (76.0, 417.0): cluster 3\n", + "Example n. 4082 = (76.0, 416.0): cluster 3\n", + "Example n. 4083 = (76.0, 415.0): cluster 3\n", + "Example n. 4084 = (76.0, 414.0): cluster 3\n", + "Example n. 4085 = (76.0, 413.0): cluster 3\n", + "Example n. 4086 = (76.0, 412.0): cluster 3\n", + "Example n. 4087 = (76.0, 411.0): cluster 3\n", + "Example n. 4088 = (76.0, 410.0): cluster 3\n", + "Example n. 4089 = (76.0, 409.0): cluster 3\n", + "Example n. 4090 = (76.0, 408.0): cluster 3\n", + "Example n. 4091 = (76.0, 407.0): cluster 3\n", + "Example n. 4092 = (76.0, 406.0): cluster 3\n", + "Example n. 4093 = (76.0, 405.0): cluster 3\n", + "Example n. 4094 = (76.0, 404.0): cluster 3\n", + "Example n. 4095 = (76.0, 403.0): cluster 3\n", + "Example n. 4096 = (76.0, 402.0): cluster 3\n", + "Example n. 4097 = (76.0, 401.0): cluster 3\n", + "Example n. 4098 = (76.0, 400.0): cluster 3\n", + "Example n. 4099 = (76.0, 399.0): cluster 3\n", + "Example n. 4100 = (76.0, 398.0): cluster 3\n", + "Example n. 4101 = (77.0, 471.0): cluster 1\n", + "Example n. 4102 = (77.0, 470.0): cluster 1\n", + "Example n. 4103 = (77.0, 469.0): cluster 1\n", + "Example n. 4104 = (77.0, 468.0): cluster 1\n", + "Example n. 4105 = (77.0, 467.0): cluster 1\n", + "Example n. 4106 = (77.0, 466.0): cluster 1\n", + "Example n. 4107 = (77.0, 465.0): cluster 1\n", + "Example n. 4108 = (77.0, 464.0): cluster 1\n", + "Example n. 4109 = (77.0, 463.0): cluster 1\n", + "Example n. 4110 = (77.0, 462.0): cluster 1\n", + "Example n. 4111 = (77.0, 461.0): cluster 1\n", + "Example n. 4112 = (77.0, 460.0): cluster 1\n", + "Example n. 4113 = (77.0, 459.0): cluster 1\n", + "Example n. 4114 = (77.0, 458.0): cluster 1\n", + "Example n. 4115 = (77.0, 457.0): cluster 1\n", + "Example n. 4116 = (77.0, 456.0): cluster 1\n", + "Example n. 4117 = (77.0, 455.0): cluster 1\n", + "Example n. 4118 = (77.0, 454.0): cluster 1\n", + "Example n. 4119 = (77.0, 453.0): cluster 1\n", + "Example n. 4120 = (77.0, 452.0): cluster 1\n", + "Example n. 4121 = (77.0, 451.0): cluster 1\n", + "Example n. 4122 = (77.0, 450.0): cluster 1\n", + "Example n. 4123 = (77.0, 449.0): cluster 1\n", + "Example n. 4124 = (77.0, 448.0): cluster 1\n", + "Example n. 4125 = (77.0, 447.0): cluster 1\n", + "Example n. 4126 = (77.0, 446.0): cluster 1\n", + "Example n. 4127 = (77.0, 445.0): cluster 1\n", + "Example n. 4128 = (77.0, 444.0): cluster 1\n", + "Example n. 4129 = (77.0, 432.0): cluster 3\n", + "Example n. 4130 = (77.0, 431.0): cluster 3\n", + "Example n. 4131 = (77.0, 430.0): cluster 3\n", + "Example n. 4132 = (77.0, 429.0): cluster 3\n", + "Example n. 4133 = (77.0, 428.0): cluster 3\n", + "Example n. 4134 = (77.0, 427.0): cluster 3\n", + "Example n. 4135 = (77.0, 426.0): cluster 3\n", + "Example n. 4136 = (77.0, 425.0): cluster 3\n", + "Example n. 4137 = (77.0, 424.0): cluster 3\n", + "Example n. 4138 = (77.0, 423.0): cluster 3\n", + "Example n. 4139 = (77.0, 422.0): cluster 3\n", + "Example n. 4140 = (77.0, 421.0): cluster 3\n", + "Example n. 4141 = (77.0, 417.0): cluster 3\n", + "Example n. 4142 = (77.0, 416.0): cluster 3\n", + "Example n. 4143 = (77.0, 415.0): cluster 3\n", + "Example n. 4144 = (77.0, 414.0): cluster 3\n", + "Example n. 4145 = (77.0, 413.0): cluster 3\n", + "Example n. 4146 = (77.0, 412.0): cluster 3\n", + "Example n. 4147 = (77.0, 411.0): cluster 3\n", + "Example n. 4148 = (77.0, 410.0): cluster 3\n", + "Example n. 4149 = (77.0, 409.0): cluster 3\n", + "Example n. 4150 = (77.0, 408.0): cluster 3\n", + "Example n. 4151 = (77.0, 407.0): cluster 3\n", + "Example n. 4152 = (77.0, 406.0): cluster 3\n", + "Example n. 4153 = (77.0, 405.0): cluster 3\n", + "Example n. 4154 = (77.0, 404.0): cluster 3\n", + "Example n. 4155 = (77.0, 403.0): cluster 3\n", + "Example n. 4156 = (77.0, 402.0): cluster 3\n", + "Example n. 4157 = (77.0, 401.0): cluster 3\n", + "Example n. 4158 = (77.0, 400.0): cluster 3\n", + "Example n. 4159 = (77.0, 399.0): cluster 3\n", + "Example n. 4160 = (77.0, 398.0): cluster 3\n", + "Example n. 4161 = (78.0, 436.0): cluster 3\n", + "Example n. 4162 = (78.0, 435.0): cluster 3\n", + "Example n. 4163 = (78.0, 434.0): cluster 3\n", + "Example n. 4164 = (78.0, 433.0): cluster 3\n", + "Example n. 4165 = (78.0, 432.0): cluster 3\n", + "Example n. 4166 = (78.0, 431.0): cluster 3\n", + "Example n. 4167 = (78.0, 430.0): cluster 3\n", + "Example n. 4168 = (78.0, 429.0): cluster 3\n", + "Example n. 4169 = (78.0, 428.0): cluster 3\n", + "Example n. 4170 = (78.0, 427.0): cluster 3\n", + "Example n. 4171 = (78.0, 426.0): cluster 3\n", + "Example n. 4172 = (78.0, 425.0): cluster 3\n", + "Example n. 4173 = (78.0, 421.0): cluster 3\n", + "Example n. 4174 = (78.0, 420.0): cluster 3\n", + "Example n. 4175 = (78.0, 419.0): cluster 3\n", + "Example n. 4176 = (78.0, 418.0): cluster 3\n", + "Example n. 4177 = (78.0, 417.0): cluster 3\n", + "Example n. 4178 = (78.0, 416.0): cluster 3\n", + "Example n. 4179 = (78.0, 415.0): cluster 3\n", + "Example n. 4180 = (78.0, 414.0): cluster 3\n", + "Example n. 4181 = (78.0, 413.0): cluster 3\n", + "Example n. 4182 = (78.0, 412.0): cluster 3\n", + "Example n. 4183 = (78.0, 411.0): cluster 3\n", + "Example n. 4184 = (78.0, 410.0): cluster 3\n", + "Example n. 4185 = (78.0, 409.0): cluster 3\n", + "Example n. 4186 = (78.0, 408.0): cluster 3\n", + "Example n. 4187 = (78.0, 407.0): cluster 3\n", + "Example n. 4188 = (78.0, 406.0): cluster 3\n", + "Example n. 4189 = (78.0, 405.0): cluster 3\n", + "Example n. 4190 = (78.0, 404.0): cluster 3\n", + "Example n. 4191 = (78.0, 403.0): cluster 3\n", + "Example n. 4192 = (78.0, 402.0): cluster 3\n", + "Example n. 4193 = (78.0, 401.0): cluster 3\n", + "Example n. 4194 = (78.0, 400.0): cluster 3\n", + "Example n. 4195 = (78.0, 399.0): cluster 3\n", + "Example n. 4196 = (78.0, 398.0): cluster 3\n", + "Example n. 4197 = (79.0, 436.0): cluster 3\n", + "Example n. 4198 = (79.0, 435.0): cluster 3\n", + "Example n. 4199 = (79.0, 434.0): cluster 3\n", + "Example n. 4200 = (79.0, 433.0): cluster 3\n", + "Example n. 4201 = (79.0, 432.0): cluster 3\n", + "Example n. 4202 = (79.0, 431.0): cluster 3\n", + "Example n. 4203 = (79.0, 430.0): cluster 3\n", + "Example n. 4204 = (79.0, 429.0): cluster 3\n", + "Example n. 4205 = (79.0, 428.0): cluster 3\n", + "Example n. 4206 = (79.0, 427.0): cluster 3\n", + "Example n. 4207 = (79.0, 426.0): cluster 3\n", + "Example n. 4208 = (79.0, 425.0): cluster 3\n", + "Example n. 4209 = (79.0, 421.0): cluster 3\n", + "Example n. 4210 = (79.0, 420.0): cluster 3\n", + "Example n. 4211 = (79.0, 419.0): cluster 3\n", + "Example n. 4212 = (79.0, 418.0): cluster 3\n", + "Example n. 4213 = (79.0, 417.0): cluster 3\n", + "Example n. 4214 = (79.0, 416.0): cluster 3\n", + "Example n. 4215 = (79.0, 415.0): cluster 3\n", + "Example n. 4216 = (79.0, 414.0): cluster 3\n", + "Example n. 4217 = (79.0, 413.0): cluster 3\n", + "Example n. 4218 = (79.0, 412.0): cluster 3\n", + "Example n. 4219 = (79.0, 411.0): cluster 3\n", + "Example n. 4220 = (79.0, 410.0): cluster 3\n", + "Example n. 4221 = (79.0, 409.0): cluster 3\n", + "Example n. 4222 = (79.0, 408.0): cluster 3\n", + "Example n. 4223 = (79.0, 407.0): cluster 3\n", + "Example n. 4224 = (79.0, 406.0): cluster 3\n", + "Example n. 4225 = (79.0, 405.0): cluster 3\n", + "Example n. 4226 = (79.0, 404.0): cluster 3\n", + "Example n. 4227 = (79.0, 403.0): cluster 3\n", + "Example n. 4228 = (79.0, 402.0): cluster 3\n", + "Example n. 4229 = (79.0, 401.0): cluster 3\n", + "Example n. 4230 = (79.0, 400.0): cluster 3\n", + "Example n. 4231 = (79.0, 399.0): cluster 3\n", + "Example n. 4232 = (79.0, 398.0): cluster 3\n", + "Example n. 4233 = (80.0, 436.0): cluster 3\n", + "Example n. 4234 = (80.0, 435.0): cluster 3\n", + "Example n. 4235 = (80.0, 434.0): cluster 3\n", + "Example n. 4236 = (80.0, 433.0): cluster 3\n", + "Example n. 4237 = (80.0, 432.0): cluster 3\n", + "Example n. 4238 = (80.0, 431.0): cluster 3\n", + "Example n. 4239 = (80.0, 430.0): cluster 3\n", + "Example n. 4240 = (80.0, 429.0): cluster 3\n", + "Example n. 4241 = (80.0, 428.0): cluster 3\n", + "Example n. 4242 = (80.0, 427.0): cluster 3\n", + "Example n. 4243 = (80.0, 426.0): cluster 3\n", + "Example n. 4244 = (80.0, 425.0): cluster 3\n", + "Example n. 4245 = (80.0, 421.0): cluster 3\n", + "Example n. 4246 = (80.0, 420.0): cluster 3\n", + "Example n. 4247 = (80.0, 419.0): cluster 3\n", + "Example n. 4248 = (80.0, 418.0): cluster 3\n", + "Example n. 4249 = (80.0, 417.0): cluster 3\n", + "Example n. 4250 = (80.0, 416.0): cluster 3\n", + "Example n. 4251 = (80.0, 415.0): cluster 3\n", + "Example n. 4252 = (80.0, 414.0): cluster 3\n", + "Example n. 4253 = (80.0, 413.0): cluster 3\n", + "Example n. 4254 = (80.0, 412.0): cluster 3\n", + "Example n. 4255 = (80.0, 411.0): cluster 3\n", + "Example n. 4256 = (80.0, 410.0): cluster 3\n", + "Example n. 4257 = (80.0, 409.0): cluster 3\n", + "Example n. 4258 = (80.0, 408.0): cluster 3\n", + "Example n. 4259 = (80.0, 407.0): cluster 3\n", + "Example n. 4260 = (80.0, 406.0): cluster 3\n", + "Example n. 4261 = (80.0, 405.0): cluster 3\n", + "Example n. 4262 = (80.0, 404.0): cluster 3\n", + "Example n. 4263 = (80.0, 403.0): cluster 3\n", + "Example n. 4264 = (80.0, 402.0): cluster 3\n", + "Example n. 4265 = (80.0, 401.0): cluster 3\n", + "Example n. 4266 = (80.0, 400.0): cluster 3\n", + "Example n. 4267 = (80.0, 399.0): cluster 3\n", + "Example n. 4268 = (80.0, 398.0): cluster 3\n", + "Example n. 4269 = (81.0, 436.0): cluster 3\n", + "Example n. 4270 = (81.0, 435.0): cluster 3\n", + "Example n. 4271 = (81.0, 434.0): cluster 3\n", + "Example n. 4272 = (81.0, 433.0): cluster 3\n", + "Example n. 4273 = (81.0, 432.0): cluster 3\n", + "Example n. 4274 = (81.0, 431.0): cluster 3\n", + "Example n. 4275 = (81.0, 430.0): cluster 3\n", + "Example n. 4276 = (81.0, 429.0): cluster 3\n", + "Example n. 4277 = (81.0, 428.0): cluster 3\n", + "Example n. 4278 = (81.0, 427.0): cluster 3\n", + "Example n. 4279 = (81.0, 426.0): cluster 3\n", + "Example n. 4280 = (81.0, 425.0): cluster 3\n", + "Example n. 4281 = (81.0, 421.0): cluster 3\n", + "Example n. 4282 = (81.0, 420.0): cluster 3\n", + "Example n. 4283 = (81.0, 419.0): cluster 3\n", + "Example n. 4284 = (81.0, 418.0): cluster 3\n", + "Example n. 4285 = (81.0, 417.0): cluster 3\n", + "Example n. 4286 = (81.0, 416.0): cluster 3\n", + "Example n. 4287 = (81.0, 415.0): cluster 3\n", + "Example n. 4288 = (81.0, 414.0): cluster 3\n", + "Example n. 4289 = (81.0, 413.0): cluster 3\n", + "Example n. 4290 = (81.0, 412.0): cluster 3\n", + "Example n. 4291 = (81.0, 411.0): cluster 3\n", + "Example n. 4292 = (81.0, 410.0): cluster 3\n", + "Example n. 4293 = (81.0, 409.0): cluster 3\n", + "Example n. 4294 = (81.0, 408.0): cluster 3\n", + "Example n. 4295 = (81.0, 407.0): cluster 3\n", + "Example n. 4296 = (81.0, 406.0): cluster 3\n", + "Example n. 4297 = (81.0, 405.0): cluster 3\n", + "Example n. 4298 = (81.0, 404.0): cluster 3\n", + "Example n. 4299 = (81.0, 403.0): cluster 3\n", + "Example n. 4300 = (81.0, 402.0): cluster 3\n", + "Example n. 4301 = (81.0, 401.0): cluster 3\n", + "Example n. 4302 = (81.0, 400.0): cluster 3\n", + "Example n. 4303 = (81.0, 399.0): cluster 3\n", + "Example n. 4304 = (81.0, 398.0): cluster 3\n", + "Example n. 4305 = (82.0, 440.0): cluster 3\n", + "Example n. 4306 = (82.0, 439.0): cluster 3\n", + "Example n. 4307 = (82.0, 438.0): cluster 3\n", + "Example n. 4308 = (82.0, 437.0): cluster 3\n", + "Example n. 4309 = (82.0, 436.0): cluster 3\n", + "Example n. 4310 = (82.0, 435.0): cluster 3\n", + "Example n. 4311 = (82.0, 434.0): cluster 3\n", + "Example n. 4312 = (82.0, 433.0): cluster 3\n", + "Example n. 4313 = (82.0, 432.0): cluster 3\n", + "Example n. 4314 = (82.0, 431.0): cluster 3\n", + "Example n. 4315 = (82.0, 430.0): cluster 3\n", + "Example n. 4316 = (82.0, 429.0): cluster 3\n", + "Example n. 4317 = (82.0, 425.0): cluster 3\n", + "Example n. 4318 = (82.0, 424.0): cluster 3\n", + "Example n. 4319 = (82.0, 423.0): cluster 3\n", + "Example n. 4320 = (82.0, 422.0): cluster 3\n", + "Example n. 4321 = (82.0, 421.0): cluster 3\n", + "Example n. 4322 = (82.0, 420.0): cluster 3\n", + "Example n. 4323 = (82.0, 419.0): cluster 3\n", + "Example n. 4324 = (82.0, 418.0): cluster 3\n", + "Example n. 4325 = (82.0, 417.0): cluster 3\n", + "Example n. 4326 = (82.0, 416.0): cluster 3\n", + "Example n. 4327 = (82.0, 415.0): cluster 3\n", + "Example n. 4328 = (82.0, 414.0): cluster 3\n", + "Example n. 4329 = (82.0, 413.0): cluster 3\n", + "Example n. 4330 = (82.0, 412.0): cluster 3\n", + "Example n. 4331 = (82.0, 411.0): cluster 3\n", + "Example n. 4332 = (82.0, 410.0): cluster 3\n", + "Example n. 4333 = (82.0, 409.0): cluster 3\n", + "Example n. 4334 = (82.0, 408.0): cluster 3\n", + "Example n. 4335 = (82.0, 407.0): cluster 3\n", + "Example n. 4336 = (82.0, 406.0): cluster 3\n", + "Example n. 4337 = (82.0, 405.0): cluster 3\n", + "Example n. 4338 = (82.0, 404.0): cluster 3\n", + "Example n. 4339 = (82.0, 403.0): cluster 3\n", + "Example n. 4340 = (82.0, 402.0): cluster 3\n", + "Example n. 4341 = (83.0, 440.0): cluster 2\n", + "Example n. 4342 = (83.0, 439.0): cluster 3\n", + "Example n. 4343 = (83.0, 438.0): cluster 3\n", + "Example n. 4344 = (83.0, 437.0): cluster 3\n", + "Example n. 4345 = (83.0, 436.0): cluster 3\n", + "Example n. 4346 = (83.0, 435.0): cluster 3\n", + "Example n. 4347 = (83.0, 434.0): cluster 3\n", + "Example n. 4348 = (83.0, 433.0): cluster 3\n", + "Example n. 4349 = (83.0, 432.0): cluster 3\n", + "Example n. 4350 = (83.0, 431.0): cluster 3\n", + "Example n. 4351 = (83.0, 430.0): cluster 3\n", + "Example n. 4352 = (83.0, 429.0): cluster 3\n", + "Example n. 4353 = (83.0, 425.0): cluster 3\n", + "Example n. 4354 = (83.0, 424.0): cluster 3\n", + "Example n. 4355 = (83.0, 423.0): cluster 3\n", + "Example n. 4356 = (83.0, 422.0): cluster 3\n", + "Example n. 4357 = (83.0, 421.0): cluster 3\n", + "Example n. 4358 = (83.0, 420.0): cluster 3\n", + "Example n. 4359 = (83.0, 419.0): cluster 3\n", + "Example n. 4360 = (83.0, 418.0): cluster 3\n", + "Example n. 4361 = (83.0, 417.0): cluster 3\n", + "Example n. 4362 = (83.0, 416.0): cluster 3\n", + "Example n. 4363 = (83.0, 415.0): cluster 3\n", + "Example n. 4364 = (83.0, 414.0): cluster 3\n", + "Example n. 4365 = (83.0, 413.0): cluster 3\n", + "Example n. 4366 = (83.0, 412.0): cluster 3\n", + "Example n. 4367 = (83.0, 411.0): cluster 3\n", + "Example n. 4368 = (83.0, 410.0): cluster 3\n", + "Example n. 4369 = (83.0, 409.0): cluster 3\n", + "Example n. 4370 = (83.0, 408.0): cluster 3\n", + "Example n. 4371 = (83.0, 407.0): cluster 3\n", + "Example n. 4372 = (83.0, 406.0): cluster 3\n", + "Example n. 4373 = (83.0, 405.0): cluster 3\n", + "Example n. 4374 = (83.0, 404.0): cluster 3\n", + "Example n. 4375 = (83.0, 403.0): cluster 3\n", + "Example n. 4376 = (83.0, 402.0): cluster 3\n", + "Example n. 4377 = (84.0, 440.0): cluster 2\n", + "Example n. 4378 = (84.0, 439.0): cluster 2\n", + "Example n. 4379 = (84.0, 438.0): cluster 3\n", + "Example n. 4380 = (84.0, 437.0): cluster 3\n", + "Example n. 4381 = (84.0, 436.0): cluster 3\n", + "Example n. 4382 = (84.0, 435.0): cluster 3\n", + "Example n. 4383 = (84.0, 434.0): cluster 3\n", + "Example n. 4384 = (84.0, 433.0): cluster 3\n", + "Example n. 4385 = (84.0, 432.0): cluster 3\n", + "Example n. 4386 = (84.0, 431.0): cluster 3\n", + "Example n. 4387 = (84.0, 430.0): cluster 3\n", + "Example n. 4388 = (84.0, 429.0): cluster 3\n", + "Example n. 4389 = (84.0, 425.0): cluster 3\n", + "Example n. 4390 = (84.0, 424.0): cluster 3\n", + "Example n. 4391 = (84.0, 423.0): cluster 3\n", + "Example n. 4392 = (84.0, 422.0): cluster 3\n", + "Example n. 4393 = (84.0, 421.0): cluster 3\n", + "Example n. 4394 = (84.0, 420.0): cluster 3\n", + "Example n. 4395 = (84.0, 419.0): cluster 3\n", + "Example n. 4396 = (84.0, 418.0): cluster 3\n", + "Example n. 4397 = (84.0, 417.0): cluster 3\n", + "Example n. 4398 = (84.0, 416.0): cluster 3\n", + "Example n. 4399 = (84.0, 415.0): cluster 3\n", + "Example n. 4400 = (84.0, 414.0): cluster 3\n", + "Example n. 4401 = (84.0, 413.0): cluster 3\n", + "Example n. 4402 = (84.0, 412.0): cluster 3\n", + "Example n. 4403 = (84.0, 411.0): cluster 3\n", + "Example n. 4404 = (84.0, 410.0): cluster 3\n", + "Example n. 4405 = (84.0, 409.0): cluster 3\n", + "Example n. 4406 = (84.0, 408.0): cluster 3\n", + "Example n. 4407 = (84.0, 407.0): cluster 3\n", + "Example n. 4408 = (84.0, 406.0): cluster 3\n", + "Example n. 4409 = (84.0, 405.0): cluster 3\n", + "Example n. 4410 = (84.0, 404.0): cluster 3\n", + "Example n. 4411 = (84.0, 403.0): cluster 3\n", + "Example n. 4412 = (84.0, 402.0): cluster 3\n", + "Example n. 4413 = (85.0, 440.0): cluster 2\n", + "Example n. 4414 = (85.0, 439.0): cluster 2\n", + "Example n. 4415 = (85.0, 438.0): cluster 2\n", + "Example n. 4416 = (85.0, 437.0): cluster 2\n", + "Example n. 4417 = (85.0, 436.0): cluster 3\n", + "Example n. 4418 = (85.0, 435.0): cluster 3\n", + "Example n. 4419 = (85.0, 434.0): cluster 3\n", + "Example n. 4420 = (85.0, 433.0): cluster 3\n", + "Example n. 4421 = (85.0, 432.0): cluster 3\n", + "Example n. 4422 = (85.0, 431.0): cluster 3\n", + "Example n. 4423 = (85.0, 430.0): cluster 3\n", + "Example n. 4424 = (85.0, 429.0): cluster 3\n", + "Example n. 4425 = (85.0, 425.0): cluster 3\n", + "Example n. 4426 = (85.0, 424.0): cluster 3\n", + "Example n. 4427 = (85.0, 423.0): cluster 3\n", + "Example n. 4428 = (85.0, 422.0): cluster 3\n", + "Example n. 4429 = (85.0, 421.0): cluster 3\n", + "Example n. 4430 = (85.0, 420.0): cluster 3\n", + "Example n. 4431 = (85.0, 419.0): cluster 3\n", + "Example n. 4432 = (85.0, 418.0): cluster 3\n", + "Example n. 4433 = (85.0, 417.0): cluster 3\n", + "Example n. 4434 = (85.0, 416.0): cluster 3\n", + "Example n. 4435 = (85.0, 415.0): cluster 3\n", + "Example n. 4436 = (85.0, 414.0): cluster 3\n", + "Example n. 4437 = (85.0, 413.0): cluster 3\n", + "Example n. 4438 = (85.0, 412.0): cluster 3\n", + "Example n. 4439 = (85.0, 411.0): cluster 3\n", + "Example n. 4440 = (85.0, 410.0): cluster 3\n", + "Example n. 4441 = (85.0, 409.0): cluster 3\n", + "Example n. 4442 = (85.0, 408.0): cluster 3\n", + "Example n. 4443 = (85.0, 407.0): cluster 3\n", + "Example n. 4444 = (85.0, 406.0): cluster 3\n", + "Example n. 4445 = (85.0, 405.0): cluster 3\n", + "Example n. 4446 = (85.0, 404.0): cluster 3\n", + "Example n. 4447 = (85.0, 403.0): cluster 3\n", + "Example n. 4448 = (85.0, 402.0): cluster 3\n", + "Example n. 4449 = (86.0, 444.0): cluster 2\n", + "Example n. 4450 = (86.0, 443.0): cluster 2\n", + "Example n. 4451 = (86.0, 442.0): cluster 2\n", + "Example n. 4452 = (86.0, 441.0): cluster 2\n", + "Example n. 4453 = (86.0, 440.0): cluster 2\n", + "Example n. 4454 = (86.0, 439.0): cluster 2\n", + "Example n. 4455 = (86.0, 438.0): cluster 2\n", + "Example n. 4456 = (86.0, 437.0): cluster 2\n", + "Example n. 4457 = (86.0, 436.0): cluster 2\n", + "Example n. 4458 = (86.0, 435.0): cluster 3\n", + "Example n. 4459 = (86.0, 434.0): cluster 3\n", + "Example n. 4460 = (86.0, 433.0): cluster 3\n", + "Example n. 4461 = (86.0, 425.0): cluster 3\n", + "Example n. 4462 = (86.0, 424.0): cluster 3\n", + "Example n. 4463 = (86.0, 423.0): cluster 3\n", + "Example n. 4464 = (86.0, 422.0): cluster 3\n", + "Example n. 4465 = (86.0, 421.0): cluster 3\n", + "Example n. 4466 = (86.0, 420.0): cluster 3\n", + "Example n. 4467 = (86.0, 419.0): cluster 3\n", + "Example n. 4468 = (86.0, 418.0): cluster 3\n", + "Example n. 4469 = (86.0, 417.0): cluster 3\n", + "Example n. 4470 = (86.0, 416.0): cluster 3\n", + "Example n. 4471 = (86.0, 415.0): cluster 3\n", + "Example n. 4472 = (86.0, 414.0): cluster 3\n", + "Example n. 4473 = (86.0, 413.0): cluster 3\n", + "Example n. 4474 = (86.0, 412.0): cluster 3\n", + "Example n. 4475 = (86.0, 411.0): cluster 3\n", + "Example n. 4476 = (86.0, 410.0): cluster 3\n", + "Example n. 4477 = (86.0, 409.0): cluster 3\n", + "Example n. 4478 = (86.0, 408.0): cluster 3\n", + "Example n. 4479 = (86.0, 407.0): cluster 3\n", + "Example n. 4480 = (86.0, 406.0): cluster 3\n", + "Example n. 4481 = (86.0, 405.0): cluster 3\n", + "Example n. 4482 = (86.0, 404.0): cluster 3\n", + "Example n. 4483 = (86.0, 403.0): cluster 3\n", + "Example n. 4484 = (86.0, 402.0): cluster 3\n", + "Example n. 4485 = (87.0, 444.0): cluster 2\n", + "Example n. 4486 = (87.0, 443.0): cluster 2\n", + "Example n. 4487 = (87.0, 442.0): cluster 2\n", + "Example n. 4488 = (87.0, 441.0): cluster 2\n", + "Example n. 4489 = (87.0, 440.0): cluster 2\n", + "Example n. 4490 = (87.0, 439.0): cluster 2\n", + "Example n. 4491 = (87.0, 438.0): cluster 2\n", + "Example n. 4492 = (87.0, 437.0): cluster 2\n", + "Example n. 4493 = (87.0, 436.0): cluster 2\n", + "Example n. 4494 = (87.0, 435.0): cluster 2\n", + "Example n. 4495 = (87.0, 434.0): cluster 3\n", + "Example n. 4496 = (87.0, 433.0): cluster 3\n", + "Example n. 4497 = (87.0, 425.0): cluster 3\n", + "Example n. 4498 = (87.0, 424.0): cluster 3\n", + "Example n. 4499 = (87.0, 423.0): cluster 3\n", + "Example n. 4500 = (87.0, 422.0): cluster 3\n", + "Example n. 4501 = (87.0, 421.0): cluster 3\n", + "Example n. 4502 = (87.0, 420.0): cluster 3\n", + "Example n. 4503 = (87.0, 419.0): cluster 3\n", + "Example n. 4504 = (87.0, 418.0): cluster 3\n", + "Example n. 4505 = (87.0, 417.0): cluster 3\n", + "Example n. 4506 = (87.0, 416.0): cluster 3\n", + "Example n. 4507 = (87.0, 415.0): cluster 3\n", + "Example n. 4508 = (87.0, 414.0): cluster 3\n", + "Example n. 4509 = (87.0, 413.0): cluster 3\n", + "Example n. 4510 = (87.0, 412.0): cluster 3\n", + "Example n. 4511 = (87.0, 411.0): cluster 3\n", + "Example n. 4512 = (87.0, 410.0): cluster 3\n", + "Example n. 4513 = (87.0, 409.0): cluster 3\n", + "Example n. 4514 = (87.0, 408.0): cluster 3\n", + "Example n. 4515 = (87.0, 407.0): cluster 3\n", + "Example n. 4516 = (87.0, 406.0): cluster 3\n", + "Example n. 4517 = (87.0, 405.0): cluster 3\n", + "Example n. 4518 = (87.0, 404.0): cluster 3\n", + "Example n. 4519 = (87.0, 403.0): cluster 3\n", + "Example n. 4520 = (87.0, 402.0): cluster 3\n", + "Example n. 4521 = (88.0, 444.0): cluster 2\n", + "Example n. 4522 = (88.0, 443.0): cluster 2\n", + "Example n. 4523 = (88.0, 442.0): cluster 2\n", + "Example n. 4524 = (88.0, 441.0): cluster 2\n", + "Example n. 4525 = (88.0, 440.0): cluster 2\n", + "Example n. 4526 = (88.0, 439.0): cluster 2\n", + "Example n. 4527 = (88.0, 438.0): cluster 2\n", + "Example n. 4528 = (88.0, 437.0): cluster 2\n", + "Example n. 4529 = (88.0, 436.0): cluster 2\n", + "Example n. 4530 = (88.0, 435.0): cluster 2\n", + "Example n. 4531 = (88.0, 434.0): cluster 2\n", + "Example n. 4532 = (88.0, 433.0): cluster 2\n", + "Example n. 4533 = (88.0, 425.0): cluster 3\n", + "Example n. 4534 = (88.0, 424.0): cluster 3\n", + "Example n. 4535 = (88.0, 423.0): cluster 3\n", + "Example n. 4536 = (88.0, 422.0): cluster 3\n", + "Example n. 4537 = (88.0, 421.0): cluster 3\n", + "Example n. 4538 = (88.0, 420.0): cluster 3\n", + "Example n. 4539 = (88.0, 419.0): cluster 3\n", + "Example n. 4540 = (88.0, 418.0): cluster 3\n", + "Example n. 4541 = (88.0, 417.0): cluster 3\n", + "Example n. 4542 = (88.0, 416.0): cluster 3\n", + "Example n. 4543 = (88.0, 415.0): cluster 3\n", + "Example n. 4544 = (88.0, 414.0): cluster 3\n", + "Example n. 4545 = (88.0, 413.0): cluster 3\n", + "Example n. 4546 = (88.0, 412.0): cluster 3\n", + "Example n. 4547 = (88.0, 411.0): cluster 3\n", + "Example n. 4548 = (88.0, 410.0): cluster 3\n", + "Example n. 4549 = (88.0, 409.0): cluster 3\n", + "Example n. 4550 = (88.0, 408.0): cluster 3\n", + "Example n. 4551 = (88.0, 407.0): cluster 3\n", + "Example n. 4552 = (88.0, 406.0): cluster 3\n", + "Example n. 4553 = (88.0, 405.0): cluster 3\n", + "Example n. 4554 = (88.0, 404.0): cluster 3\n", + "Example n. 4555 = (88.0, 403.0): cluster 3\n", + "Example n. 4556 = (88.0, 402.0): cluster 3\n", + "Example n. 4557 = (89.0, 444.0): cluster 2\n", + "Example n. 4558 = (89.0, 443.0): cluster 2\n", + "Example n. 4559 = (89.0, 442.0): cluster 2\n", + "Example n. 4560 = (89.0, 441.0): cluster 2\n", + "Example n. 4561 = (89.0, 440.0): cluster 2\n", + "Example n. 4562 = (89.0, 439.0): cluster 2\n", + "Example n. 4563 = (89.0, 438.0): cluster 2\n", + "Example n. 4564 = (89.0, 437.0): cluster 2\n", + "Example n. 4565 = (89.0, 436.0): cluster 2\n", + "Example n. 4566 = (89.0, 435.0): cluster 2\n", + "Example n. 4567 = (89.0, 434.0): cluster 2\n", + "Example n. 4568 = (89.0, 433.0): cluster 2\n", + "Example n. 4569 = (89.0, 425.0): cluster 3\n", + "Example n. 4570 = (89.0, 424.0): cluster 3\n", + "Example n. 4571 = (89.0, 423.0): cluster 3\n", + "Example n. 4572 = (89.0, 422.0): cluster 3\n", + "Example n. 4573 = (89.0, 421.0): cluster 3\n", + "Example n. 4574 = (89.0, 420.0): cluster 3\n", + "Example n. 4575 = (89.0, 419.0): cluster 3\n", + "Example n. 4576 = (89.0, 418.0): cluster 3\n", + "Example n. 4577 = (89.0, 417.0): cluster 3\n", + "Example n. 4578 = (89.0, 416.0): cluster 3\n", + "Example n. 4579 = (89.0, 415.0): cluster 3\n", + "Example n. 4580 = (89.0, 414.0): cluster 3\n", + "Example n. 4581 = (89.0, 413.0): cluster 3\n", + "Example n. 4582 = (89.0, 412.0): cluster 3\n", + "Example n. 4583 = (89.0, 411.0): cluster 3\n", + "Example n. 4584 = (89.0, 410.0): cluster 3\n", + "Example n. 4585 = (89.0, 409.0): cluster 3\n", + "Example n. 4586 = (89.0, 408.0): cluster 3\n", + "Example n. 4587 = (89.0, 407.0): cluster 3\n", + "Example n. 4588 = (89.0, 406.0): cluster 3\n", + "Example n. 4589 = (89.0, 405.0): cluster 3\n", + "Example n. 4590 = (89.0, 404.0): cluster 3\n", + "Example n. 4591 = (89.0, 403.0): cluster 3\n", + "Example n. 4592 = (89.0, 402.0): cluster 3\n", + "Example n. 4593 = (90.0, 471.0): cluster 2\n", + "Example n. 4594 = (90.0, 470.0): cluster 2\n", + "Example n. 4595 = (90.0, 469.0): cluster 2\n", + "Example n. 4596 = (90.0, 468.0): cluster 2\n", + "Example n. 4597 = (90.0, 467.0): cluster 2\n", + "Example n. 4598 = (90.0, 466.0): cluster 2\n", + "Example n. 4599 = (90.0, 465.0): cluster 2\n", + "Example n. 4600 = (90.0, 464.0): cluster 2\n", + "Example n. 4601 = (90.0, 463.0): cluster 2\n", + "Example n. 4602 = (90.0, 462.0): cluster 2\n", + "Example n. 4603 = (90.0, 461.0): cluster 2\n", + "Example n. 4604 = (90.0, 460.0): cluster 2\n", + "Example n. 4605 = (90.0, 444.0): cluster 2\n", + "Example n. 4606 = (90.0, 443.0): cluster 2\n", + "Example n. 4607 = (90.0, 442.0): cluster 2\n", + "Example n. 4608 = (90.0, 441.0): cluster 2\n", + "Example n. 4609 = (90.0, 440.0): cluster 2\n", + "Example n. 4610 = (90.0, 439.0): cluster 2\n", + "Example n. 4611 = (90.0, 438.0): cluster 2\n", + "Example n. 4612 = (90.0, 437.0): cluster 2\n", + "Example n. 4613 = (90.0, 436.0): cluster 2\n", + "Example n. 4614 = (90.0, 435.0): cluster 2\n", + "Example n. 4615 = (90.0, 434.0): cluster 2\n", + "Example n. 4616 = (90.0, 433.0): cluster 2\n", + "Example n. 4617 = (90.0, 425.0): cluster 3\n", + "Example n. 4618 = (90.0, 424.0): cluster 3\n", + "Example n. 4619 = (90.0, 423.0): cluster 3\n", + "Example n. 4620 = (90.0, 422.0): cluster 3\n", + "Example n. 4621 = (90.0, 421.0): cluster 3\n", + "Example n. 4622 = (90.0, 420.0): cluster 3\n", + "Example n. 4623 = (90.0, 419.0): cluster 3\n", + "Example n. 4624 = (90.0, 418.0): cluster 3\n", + "Example n. 4625 = (90.0, 417.0): cluster 3\n", + "Example n. 4626 = (90.0, 416.0): cluster 3\n", + "Example n. 4627 = (90.0, 415.0): cluster 3\n", + "Example n. 4628 = (90.0, 414.0): cluster 3\n", + "Example n. 4629 = (90.0, 413.0): cluster 3\n", + "Example n. 4630 = (90.0, 412.0): cluster 3\n", + "Example n. 4631 = (90.0, 411.0): cluster 3\n", + "Example n. 4632 = (90.0, 410.0): cluster 3\n", + "Example n. 4633 = (90.0, 409.0): cluster 3\n", + "Example n. 4634 = (90.0, 408.0): cluster 3\n", + "Example n. 4635 = (90.0, 407.0): cluster 3\n", + "Example n. 4636 = (90.0, 406.0): cluster 3\n", + "Example n. 4637 = (91.0, 471.0): cluster 2\n", + "Example n. 4638 = (91.0, 470.0): cluster 2\n", + "Example n. 4639 = (91.0, 469.0): cluster 2\n", + "Example n. 4640 = (91.0, 468.0): cluster 2\n", + "Example n. 4641 = (91.0, 467.0): cluster 2\n", + "Example n. 4642 = (91.0, 466.0): cluster 2\n", + "Example n. 4643 = (91.0, 465.0): cluster 2\n", + "Example n. 4644 = (91.0, 464.0): cluster 2\n", + "Example n. 4645 = (91.0, 463.0): cluster 2\n", + "Example n. 4646 = (91.0, 462.0): cluster 2\n", + "Example n. 4647 = (91.0, 460.0): cluster 2\n", + "Example n. 4648 = (91.0, 444.0): cluster 2\n", + "Example n. 4649 = (91.0, 443.0): cluster 2\n", + "Example n. 4650 = (91.0, 442.0): cluster 2\n", + "Example n. 4651 = (91.0, 441.0): cluster 2\n", + "Example n. 4652 = (91.0, 440.0): cluster 2\n", + "Example n. 4653 = (91.0, 439.0): cluster 2\n", + "Example n. 4654 = (91.0, 438.0): cluster 2\n", + "Example n. 4655 = (91.0, 437.0): cluster 2\n", + "Example n. 4656 = (91.0, 436.0): cluster 2\n", + "Example n. 4657 = (91.0, 435.0): cluster 2\n", + "Example n. 4658 = (91.0, 434.0): cluster 2\n", + "Example n. 4659 = (91.0, 433.0): cluster 2\n", + "Example n. 4660 = (91.0, 425.0): cluster 3\n", + "Example n. 4661 = (91.0, 424.0): cluster 3\n", + "Example n. 4662 = (91.0, 423.0): cluster 3\n", + "Example n. 4663 = (91.0, 422.0): cluster 3\n", + "Example n. 4664 = (91.0, 421.0): cluster 3\n", + "Example n. 4665 = (91.0, 420.0): cluster 3\n", + "Example n. 4666 = (91.0, 419.0): cluster 3\n", + "Example n. 4667 = (91.0, 418.0): cluster 3\n", + "Example n. 4668 = (91.0, 417.0): cluster 3\n", + "Example n. 4669 = (91.0, 416.0): cluster 3\n", + "Example n. 4670 = (91.0, 415.0): cluster 3\n", + "Example n. 4671 = (91.0, 414.0): cluster 3\n", + "Example n. 4672 = (91.0, 413.0): cluster 3\n", + "Example n. 4673 = (91.0, 412.0): cluster 3\n", + "Example n. 4674 = (91.0, 411.0): cluster 3\n", + "Example n. 4675 = (91.0, 410.0): cluster 3\n", + "Example n. 4676 = (91.0, 409.0): cluster 3\n", + "Example n. 4677 = (91.0, 408.0): cluster 3\n", + "Example n. 4678 = (91.0, 407.0): cluster 3\n", + "Example n. 4679 = (91.0, 406.0): cluster 3\n", + "Example n. 4680 = (92.0, 471.0): cluster 2\n", + "Example n. 4681 = (92.0, 470.0): cluster 2\n", + "Example n. 4682 = (92.0, 469.0): cluster 2\n", + "Example n. 4683 = (92.0, 468.0): cluster 2\n", + "Example n. 4684 = (92.0, 467.0): cluster 2\n", + "Example n. 4685 = (92.0, 466.0): cluster 2\n", + "Example n. 4686 = (92.0, 465.0): cluster 2\n", + "Example n. 4687 = (92.0, 464.0): cluster 2\n", + "Example n. 4688 = (92.0, 463.0): cluster 2\n", + "Example n. 4689 = (92.0, 462.0): cluster 2\n", + "Example n. 4690 = (92.0, 461.0): cluster 2\n", + "Example n. 4691 = (92.0, 460.0): cluster 2\n", + "Example n. 4692 = (92.0, 444.0): cluster 2\n", + "Example n. 4693 = (92.0, 443.0): cluster 2\n", + "Example n. 4694 = (92.0, 442.0): cluster 2\n", + "Example n. 4695 = (92.0, 441.0): cluster 2\n", + "Example n. 4696 = (92.0, 440.0): cluster 2\n", + "Example n. 4697 = (92.0, 439.0): cluster 2\n", + "Example n. 4698 = (92.0, 438.0): cluster 2\n", + "Example n. 4699 = (92.0, 437.0): cluster 2\n", + "Example n. 4700 = (92.0, 436.0): cluster 2\n", + "Example n. 4701 = (92.0, 435.0): cluster 2\n", + "Example n. 4702 = (92.0, 434.0): cluster 2\n", + "Example n. 4703 = (92.0, 433.0): cluster 2\n", + "Example n. 4704 = (92.0, 425.0): cluster 3\n", + "Example n. 4705 = (92.0, 424.0): cluster 3\n", + "Example n. 4706 = (92.0, 423.0): cluster 3\n", + "Example n. 4707 = (92.0, 422.0): cluster 3\n", + "Example n. 4708 = (92.0, 421.0): cluster 3\n", + "Example n. 4709 = (92.0, 420.0): cluster 3\n", + "Example n. 4710 = (92.0, 419.0): cluster 3\n", + "Example n. 4711 = (92.0, 418.0): cluster 3\n", + "Example n. 4712 = (92.0, 417.0): cluster 3\n", + "Example n. 4713 = (92.0, 416.0): cluster 3\n", + "Example n. 4714 = (92.0, 415.0): cluster 3\n", + "Example n. 4715 = (92.0, 414.0): cluster 3\n", + "Example n. 4716 = (92.0, 413.0): cluster 3\n", + "Example n. 4717 = (92.0, 412.0): cluster 3\n", + "Example n. 4718 = (92.0, 411.0): cluster 3\n", + "Example n. 4719 = (92.0, 410.0): cluster 3\n", + "Example n. 4720 = (92.0, 409.0): cluster 3\n", + "Example n. 4721 = (92.0, 408.0): cluster 3\n", + "Example n. 4722 = (92.0, 407.0): cluster 3\n", + "Example n. 4723 = (92.0, 406.0): cluster 3\n", + "Example n. 4724 = (93.0, 479.0): cluster 2\n", + "Example n. 4725 = (93.0, 478.0): cluster 2\n", + "Example n. 4726 = (93.0, 477.0): cluster 2\n", + "Example n. 4727 = (93.0, 476.0): cluster 2\n", + "Example n. 4728 = (93.0, 475.0): cluster 2\n", + "Example n. 4729 = (93.0, 474.0): cluster 2\n", + "Example n. 4730 = (93.0, 473.0): cluster 2\n", + "Example n. 4731 = (93.0, 472.0): cluster 2\n", + "Example n. 4732 = (93.0, 471.0): cluster 2\n", + "Example n. 4733 = (93.0, 470.0): cluster 2\n", + "Example n. 4734 = (93.0, 469.0): cluster 2\n", + "Example n. 4735 = (93.0, 468.0): cluster 2\n", + "Example n. 4736 = (93.0, 467.0): cluster 2\n", + "Example n. 4737 = (93.0, 466.0): cluster 2\n", + "Example n. 4738 = (93.0, 465.0): cluster 2\n", + "Example n. 4739 = (93.0, 464.0): cluster 2\n", + "Example n. 4740 = (93.0, 463.0): cluster 2\n", + "Example n. 4741 = (93.0, 462.0): cluster 2\n", + "Example n. 4742 = (93.0, 461.0): cluster 2\n", + "Example n. 4743 = (93.0, 460.0): cluster 2\n", + "Example n. 4744 = (93.0, 444.0): cluster 2\n", + "Example n. 4745 = (93.0, 443.0): cluster 2\n", + "Example n. 4746 = (93.0, 442.0): cluster 2\n", + "Example n. 4747 = (93.0, 441.0): cluster 2\n", + "Example n. 4748 = (93.0, 440.0): cluster 2\n", + "Example n. 4749 = (93.0, 439.0): cluster 2\n", + "Example n. 4750 = (93.0, 438.0): cluster 2\n", + "Example n. 4751 = (93.0, 437.0): cluster 2\n", + "Example n. 4752 = (93.0, 436.0): cluster 2\n", + "Example n. 4753 = (93.0, 435.0): cluster 2\n", + "Example n. 4754 = (93.0, 434.0): cluster 2\n", + "Example n. 4755 = (93.0, 433.0): cluster 2\n", + "Example n. 4756 = (93.0, 425.0): cluster 3\n", + "Example n. 4757 = (93.0, 424.0): cluster 3\n", + "Example n. 4758 = (93.0, 423.0): cluster 3\n", + "Example n. 4759 = (93.0, 422.0): cluster 3\n", + "Example n. 4760 = (93.0, 421.0): cluster 3\n", + "Example n. 4761 = (93.0, 420.0): cluster 3\n", + "Example n. 4762 = (93.0, 419.0): cluster 3\n", + "Example n. 4763 = (93.0, 418.0): cluster 3\n", + "Example n. 4764 = (93.0, 417.0): cluster 3\n", + "Example n. 4765 = (93.0, 416.0): cluster 3\n", + "Example n. 4766 = (93.0, 415.0): cluster 3\n", + "Example n. 4767 = (93.0, 414.0): cluster 3\n", + "Example n. 4768 = (93.0, 413.0): cluster 3\n", + "Example n. 4769 = (93.0, 412.0): cluster 3\n", + "Example n. 4770 = (93.0, 411.0): cluster 3\n", + "Example n. 4771 = (93.0, 410.0): cluster 3\n", + "Example n. 4772 = (93.0, 409.0): cluster 3\n", + "Example n. 4773 = (93.0, 408.0): cluster 3\n", + "Example n. 4774 = (93.0, 407.0): cluster 3\n", + "Example n. 4775 = (93.0, 406.0): cluster 3\n", + "Example n. 4776 = (94.0, 479.0): cluster 2\n", + "Example n. 4777 = (94.0, 478.0): cluster 2\n", + "Example n. 4778 = (94.0, 477.0): cluster 2\n", + "Example n. 4779 = (94.0, 476.0): cluster 2\n", + "Example n. 4780 = (94.0, 475.0): cluster 2\n", + "Example n. 4781 = (94.0, 474.0): cluster 2\n", + "Example n. 4782 = (94.0, 473.0): cluster 2\n", + "Example n. 4783 = (94.0, 472.0): cluster 2\n", + "Example n. 4784 = (94.0, 471.0): cluster 2\n", + "Example n. 4785 = (94.0, 470.0): cluster 2\n", + "Example n. 4786 = (94.0, 469.0): cluster 2\n", + "Example n. 4787 = (94.0, 468.0): cluster 2\n", + "Example n. 4788 = (94.0, 467.0): cluster 2\n", + "Example n. 4789 = (94.0, 466.0): cluster 2\n", + "Example n. 4790 = (94.0, 465.0): cluster 2\n", + "Example n. 4791 = (94.0, 464.0): cluster 2\n", + "Example n. 4792 = (94.0, 463.0): cluster 2\n", + "Example n. 4793 = (94.0, 462.0): cluster 2\n", + "Example n. 4794 = (94.0, 460.0): cluster 2\n", + "Example n. 4795 = (94.0, 444.0): cluster 2\n", + "Example n. 4796 = (94.0, 443.0): cluster 2\n", + "Example n. 4797 = (94.0, 442.0): cluster 2\n", + "Example n. 4798 = (94.0, 441.0): cluster 2\n", + "Example n. 4799 = (94.0, 440.0): cluster 2\n", + "Example n. 4800 = (94.0, 439.0): cluster 2\n", + "Example n. 4801 = (94.0, 438.0): cluster 2\n", + "Example n. 4802 = (94.0, 437.0): cluster 2\n", + "Example n. 4803 = (94.0, 436.0): cluster 2\n", + "Example n. 4804 = (94.0, 435.0): cluster 2\n", + "Example n. 4805 = (94.0, 434.0): cluster 2\n", + "Example n. 4806 = (94.0, 433.0): cluster 2\n", + "Example n. 4807 = (94.0, 425.0): cluster 2\n", + "Example n. 4808 = (94.0, 424.0): cluster 3\n", + "Example n. 4809 = (94.0, 423.0): cluster 3\n", + "Example n. 4810 = (94.0, 422.0): cluster 3\n", + "Example n. 4811 = (94.0, 421.0): cluster 3\n", + "Example n. 4812 = (94.0, 420.0): cluster 3\n", + "Example n. 4813 = (94.0, 419.0): cluster 3\n", + "Example n. 4814 = (94.0, 418.0): cluster 3\n", + "Example n. 4815 = (94.0, 417.0): cluster 3\n", + "Example n. 4816 = (94.0, 416.0): cluster 3\n", + "Example n. 4817 = (94.0, 415.0): cluster 3\n", + "Example n. 4818 = (94.0, 414.0): cluster 3\n", + "Example n. 4819 = (95.0, 479.0): cluster 2\n", + "Example n. 4820 = (95.0, 478.0): cluster 2\n", + "Example n. 4821 = (95.0, 477.0): cluster 2\n", + "Example n. 4822 = (95.0, 476.0): cluster 2\n", + "Example n. 4823 = (95.0, 475.0): cluster 2\n", + "Example n. 4824 = (95.0, 474.0): cluster 2\n", + "Example n. 4825 = (95.0, 473.0): cluster 2\n", + "Example n. 4826 = (95.0, 472.0): cluster 2\n", + "Example n. 4827 = (95.0, 471.0): cluster 2\n", + "Example n. 4828 = (95.0, 470.0): cluster 2\n", + "Example n. 4829 = (95.0, 469.0): cluster 2\n", + "Example n. 4830 = (95.0, 468.0): cluster 2\n", + "Example n. 4831 = (95.0, 467.0): cluster 2\n", + "Example n. 4832 = (95.0, 466.0): cluster 2\n", + "Example n. 4833 = (95.0, 465.0): cluster 2\n", + "Example n. 4834 = (95.0, 464.0): cluster 2\n", + "Example n. 4835 = (95.0, 463.0): cluster 2\n", + "Example n. 4836 = (95.0, 462.0): cluster 2\n", + "Example n. 4837 = (95.0, 461.0): cluster 2\n", + "Example n. 4838 = (95.0, 460.0): cluster 2\n", + "Example n. 4839 = (95.0, 444.0): cluster 2\n", + "Example n. 4840 = (95.0, 443.0): cluster 2\n", + "Example n. 4841 = (95.0, 442.0): cluster 2\n", + "Example n. 4842 = (95.0, 441.0): cluster 2\n", + "Example n. 4843 = (95.0, 440.0): cluster 2\n", + "Example n. 4844 = (95.0, 439.0): cluster 2\n", + "Example n. 4845 = (95.0, 438.0): cluster 2\n", + "Example n. 4846 = (95.0, 437.0): cluster 2\n", + "Example n. 4847 = (95.0, 436.0): cluster 2\n", + "Example n. 4848 = (95.0, 435.0): cluster 2\n", + "Example n. 4849 = (95.0, 434.0): cluster 2\n", + "Example n. 4850 = (95.0, 433.0): cluster 2\n", + "Example n. 4851 = (95.0, 425.0): cluster 2\n", + "Example n. 4852 = (95.0, 424.0): cluster 2\n", + "Example n. 4853 = (95.0, 423.0): cluster 3\n", + "Example n. 4854 = (95.0, 422.0): cluster 3\n", + "Example n. 4855 = (95.0, 421.0): cluster 3\n", + "Example n. 4856 = (95.0, 420.0): cluster 3\n", + "Example n. 4857 = (95.0, 419.0): cluster 3\n", + "Example n. 4858 = (95.0, 418.0): cluster 3\n", + "Example n. 4859 = (95.0, 417.0): cluster 3\n", + "Example n. 4860 = (95.0, 416.0): cluster 3\n", + "Example n. 4861 = (95.0, 415.0): cluster 3\n", + "Example n. 4862 = (95.0, 414.0): cluster 3\n", + "Example n. 4863 = (96.0, 479.0): cluster 2\n", + "Example n. 4864 = (96.0, 478.0): cluster 2\n", + "Example n. 4865 = (96.0, 477.0): cluster 2\n", + "Example n. 4866 = (96.0, 476.0): cluster 2\n", + "Example n. 4867 = (96.0, 475.0): cluster 2\n", + "Example n. 4868 = (96.0, 474.0): cluster 2\n", + "Example n. 4869 = (96.0, 473.0): cluster 2\n", + "Example n. 4870 = (96.0, 472.0): cluster 2\n", + "Example n. 4871 = (96.0, 471.0): cluster 2\n", + "Example n. 4872 = (96.0, 470.0): cluster 2\n", + "Example n. 4873 = (96.0, 469.0): cluster 2\n", + "Example n. 4874 = (96.0, 468.0): cluster 2\n", + "Example n. 4875 = (96.0, 467.0): cluster 2\n", + "Example n. 4876 = (96.0, 466.0): cluster 2\n", + "Example n. 4877 = (96.0, 465.0): cluster 2\n", + "Example n. 4878 = (96.0, 464.0): cluster 2\n", + "Example n. 4879 = (96.0, 463.0): cluster 2\n", + "Example n. 4880 = (96.0, 462.0): cluster 2\n", + "Example n. 4881 = (96.0, 461.0): cluster 2\n", + "Example n. 4882 = (96.0, 460.0): cluster 2\n", + "Example n. 4883 = (96.0, 444.0): cluster 2\n", + "Example n. 4884 = (96.0, 443.0): cluster 2\n", + "Example n. 4885 = (96.0, 442.0): cluster 2\n", + "Example n. 4886 = (96.0, 441.0): cluster 2\n", + "Example n. 4887 = (96.0, 440.0): cluster 2\n", + "Example n. 4888 = (96.0, 439.0): cluster 2\n", + "Example n. 4889 = (96.0, 438.0): cluster 2\n", + "Example n. 4890 = (96.0, 437.0): cluster 2\n", + "Example n. 4891 = (96.0, 436.0): cluster 2\n", + "Example n. 4892 = (96.0, 435.0): cluster 2\n", + "Example n. 4893 = (96.0, 434.0): cluster 2\n", + "Example n. 4894 = (96.0, 433.0): cluster 2\n", + "Example n. 4895 = (96.0, 425.0): cluster 2\n", + "Example n. 4896 = (96.0, 424.0): cluster 2\n", + "Example n. 4897 = (96.0, 423.0): cluster 2\n", + "Example n. 4898 = (96.0, 422.0): cluster 3\n", + "Example n. 4899 = (96.0, 421.0): cluster 3\n", + "Example n. 4900 = (96.0, 420.0): cluster 3\n", + "Example n. 4901 = (96.0, 419.0): cluster 3\n", + "Example n. 4902 = (96.0, 418.0): cluster 3\n", + "Example n. 4903 = (96.0, 417.0): cluster 3\n", + "Example n. 4904 = (96.0, 416.0): cluster 3\n", + "Example n. 4905 = (96.0, 415.0): cluster 3\n", + "Example n. 4906 = (96.0, 414.0): cluster 3\n", + "Example n. 4907 = (97.0, 486.0): cluster 2\n", + "Example n. 4908 = (97.0, 485.0): cluster 2\n", + "Example n. 4909 = (97.0, 483.0): cluster 2\n", + "Example n. 4910 = (97.0, 482.0): cluster 2\n", + "Example n. 4911 = (97.0, 481.0): cluster 2\n", + "Example n. 4912 = (97.0, 480.0): cluster 2\n", + "Example n. 4913 = (97.0, 479.0): cluster 2\n", + "Example n. 4914 = (97.0, 478.0): cluster 2\n", + "Example n. 4915 = (97.0, 477.0): cluster 2\n", + "Example n. 4916 = (97.0, 476.0): cluster 2\n", + "Example n. 4917 = (97.0, 475.0): cluster 2\n", + "Example n. 4918 = (97.0, 474.0): cluster 2\n", + "Example n. 4919 = (97.0, 473.0): cluster 2\n", + "Example n. 4920 = (97.0, 472.0): cluster 2\n", + "Example n. 4921 = (97.0, 471.0): cluster 2\n", + "Example n. 4922 = (97.0, 470.0): cluster 2\n", + "Example n. 4923 = (97.0, 469.0): cluster 2\n", + "Example n. 4924 = (97.0, 468.0): cluster 2\n", + "Example n. 4925 = (97.0, 467.0): cluster 2\n", + "Example n. 4926 = (97.0, 466.0): cluster 2\n", + "Example n. 4927 = (97.0, 465.0): cluster 2\n", + "Example n. 4928 = (97.0, 464.0): cluster 2\n", + "Example n. 4929 = (97.0, 462.0): cluster 2\n", + "Example n. 4930 = (97.0, 461.0): cluster 2\n", + "Example n. 4931 = (97.0, 460.0): cluster 2\n", + "Example n. 4932 = (97.0, 444.0): cluster 2\n", + "Example n. 4933 = (97.0, 443.0): cluster 2\n", + "Example n. 4934 = (97.0, 442.0): cluster 2\n", + "Example n. 4935 = (97.0, 441.0): cluster 2\n", + "Example n. 4936 = (97.0, 440.0): cluster 2\n", + "Example n. 4937 = (97.0, 439.0): cluster 2\n", + "Example n. 4938 = (97.0, 438.0): cluster 2\n", + "Example n. 4939 = (97.0, 437.0): cluster 2\n", + "Example n. 4940 = (97.0, 436.0): cluster 2\n", + "Example n. 4941 = (97.0, 435.0): cluster 2\n", + "Example n. 4942 = (97.0, 434.0): cluster 2\n", + "Example n. 4943 = (97.0, 433.0): cluster 2\n", + "Example n. 4944 = (97.0, 425.0): cluster 2\n", + "Example n. 4945 = (97.0, 424.0): cluster 2\n", + "Example n. 4946 = (97.0, 423.0): cluster 2\n", + "Example n. 4947 = (97.0, 422.0): cluster 2\n", + "Example n. 4948 = (97.0, 421.0): cluster 2\n", + "Example n. 4949 = (97.0, 420.0): cluster 3\n", + "Example n. 4950 = (97.0, 419.0): cluster 3\n", + "Example n. 4951 = (97.0, 418.0): cluster 3\n", + "Example n. 4952 = (97.0, 417.0): cluster 3\n", + "Example n. 4953 = (97.0, 416.0): cluster 3\n", + "Example n. 4954 = (97.0, 415.0): cluster 3\n", + "Example n. 4955 = (97.0, 414.0): cluster 3\n", + "Example n. 4956 = (98.0, 486.0): cluster 2\n", + "Example n. 4957 = (98.0, 485.0): cluster 2\n", + "Example n. 4958 = (98.0, 484.0): cluster 2\n", + "Example n. 4959 = (98.0, 483.0): cluster 2\n", + "Example n. 4960 = (98.0, 482.0): cluster 2\n", + "Example n. 4961 = (98.0, 481.0): cluster 2\n", + "Example n. 4962 = (98.0, 480.0): cluster 2\n", + "Example n. 4963 = (98.0, 479.0): cluster 2\n", + "Example n. 4964 = (98.0, 478.0): cluster 2\n", + "Example n. 4965 = (98.0, 477.0): cluster 2\n", + "Example n. 4966 = (98.0, 476.0): cluster 2\n", + "Example n. 4967 = (98.0, 475.0): cluster 2\n", + "Example n. 4968 = (98.0, 474.0): cluster 2\n", + "Example n. 4969 = (98.0, 473.0): cluster 2\n", + "Example n. 4970 = (98.0, 472.0): cluster 2\n", + "Example n. 4971 = (98.0, 471.0): cluster 2\n", + "Example n. 4972 = (98.0, 470.0): cluster 2\n", + "Example n. 4973 = (98.0, 469.0): cluster 2\n", + "Example n. 4974 = (98.0, 468.0): cluster 2\n", + "Example n. 4975 = (98.0, 467.0): cluster 2\n", + "Example n. 4976 = (98.0, 466.0): cluster 2\n", + "Example n. 4977 = (98.0, 465.0): cluster 2\n", + "Example n. 4978 = (98.0, 464.0): cluster 2\n", + "Example n. 4979 = (98.0, 444.0): cluster 2\n", + "Example n. 4980 = (98.0, 443.0): cluster 2\n", + "Example n. 4981 = (98.0, 442.0): cluster 2\n", + "Example n. 4982 = (98.0, 441.0): cluster 2\n", + "Example n. 4983 = (98.0, 440.0): cluster 2\n", + "Example n. 4984 = (98.0, 439.0): cluster 2\n", + "Example n. 4985 = (98.0, 438.0): cluster 2\n", + "Example n. 4986 = (98.0, 437.0): cluster 2\n", + "Example n. 4987 = (98.0, 436.0): cluster 2\n", + "Example n. 4988 = (98.0, 435.0): cluster 2\n", + "Example n. 4989 = (98.0, 434.0): cluster 2\n", + "Example n. 4990 = (98.0, 433.0): cluster 2\n", + "Example n. 4991 = (99.0, 486.0): cluster 2\n", + "Example n. 4992 = (99.0, 485.0): cluster 2\n", + "Example n. 4993 = (99.0, 484.0): cluster 2\n", + "Example n. 4994 = (99.0, 483.0): cluster 2\n", + "Example n. 4995 = (99.0, 482.0): cluster 2\n", + "Example n. 4996 = (99.0, 481.0): cluster 2\n", + "Example n. 4997 = (99.0, 480.0): cluster 2\n", + "Example n. 4998 = (99.0, 479.0): cluster 2\n", + "Example n. 4999 = (99.0, 478.0): cluster 2\n", + "Example n. 5000 = (99.0, 477.0): cluster 2\n", + "Example n. 5001 = (99.0, 476.0): cluster 2\n", + "Example n. 5002 = (99.0, 475.0): cluster 2\n", + "Example n. 5003 = (99.0, 474.0): cluster 2\n", + "Example n. 5004 = (99.0, 473.0): cluster 2\n", + "Example n. 5005 = (99.0, 472.0): cluster 2\n", + "Example n. 5006 = (99.0, 471.0): cluster 2\n", + "Example n. 5007 = (99.0, 470.0): cluster 2\n", + "Example n. 5008 = (99.0, 469.0): cluster 2\n", + "Example n. 5009 = (99.0, 468.0): cluster 2\n", + "Example n. 5010 = (99.0, 467.0): cluster 2\n", + "Example n. 5011 = (99.0, 466.0): cluster 2\n", + "Example n. 5012 = (99.0, 465.0): cluster 2\n", + "Example n. 5013 = (99.0, 464.0): cluster 2\n", + "Example n. 5014 = (99.0, 444.0): cluster 2\n", + "Example n. 5015 = (99.0, 443.0): cluster 2\n", + "Example n. 5016 = (99.0, 442.0): cluster 2\n", + "Example n. 5017 = (99.0, 441.0): cluster 2\n", + "Example n. 5018 = (99.0, 440.0): cluster 2\n", + "Example n. 5019 = (99.0, 439.0): cluster 2\n", + "Example n. 5020 = (99.0, 438.0): cluster 2\n", + "Example n. 5021 = (99.0, 437.0): cluster 2\n", + "Example n. 5022 = (99.0, 436.0): cluster 2\n", + "Example n. 5023 = (99.0, 435.0): cluster 2\n", + "Example n. 5024 = (99.0, 434.0): cluster 2\n", + "Example n. 5025 = (99.0, 433.0): cluster 2\n", + "Example n. 5026 = (100.0, 486.0): cluster 2\n", + "Example n. 5027 = (100.0, 485.0): cluster 2\n", + "Example n. 5028 = (100.0, 484.0): cluster 2\n", + "Example n. 5029 = (100.0, 483.0): cluster 2\n", + "Example n. 5030 = (100.0, 482.0): cluster 2\n", + "Example n. 5031 = (100.0, 481.0): cluster 2\n", + "Example n. 5032 = (100.0, 480.0): cluster 2\n", + "Example n. 5033 = (100.0, 479.0): cluster 2\n", + "Example n. 5034 = (100.0, 478.0): cluster 2\n", + "Example n. 5035 = (100.0, 477.0): cluster 2\n", + "Example n. 5036 = (100.0, 476.0): cluster 2\n", + "Example n. 5037 = (100.0, 475.0): cluster 2\n", + "Example n. 5038 = (100.0, 474.0): cluster 2\n", + "Example n. 5039 = (100.0, 473.0): cluster 2\n", + "Example n. 5040 = (100.0, 472.0): cluster 2\n", + "Example n. 5041 = (100.0, 471.0): cluster 2\n", + "Example n. 5042 = (100.0, 470.0): cluster 2\n", + "Example n. 5043 = (100.0, 469.0): cluster 2\n", + "Example n. 5044 = (100.0, 468.0): cluster 2\n", + "Example n. 5045 = (100.0, 467.0): cluster 2\n", + "Example n. 5046 = (100.0, 466.0): cluster 2\n", + "Example n. 5047 = (100.0, 465.0): cluster 2\n", + "Example n. 5048 = (100.0, 464.0): cluster 2\n", + "Example n. 5049 = (100.0, 444.0): cluster 2\n", + "Example n. 5050 = (100.0, 443.0): cluster 2\n", + "Example n. 5051 = (100.0, 442.0): cluster 2\n", + "Example n. 5052 = (100.0, 441.0): cluster 2\n", + "Example n. 5053 = (100.0, 440.0): cluster 2\n", + "Example n. 5054 = (100.0, 439.0): cluster 2\n", + "Example n. 5055 = (100.0, 438.0): cluster 2\n", + "Example n. 5056 = (100.0, 437.0): cluster 2\n", + "Example n. 5057 = (100.0, 436.0): cluster 2\n", + "Example n. 5058 = (100.0, 435.0): cluster 2\n", + "Example n. 5059 = (100.0, 434.0): cluster 2\n", + "Example n. 5060 = (100.0, 433.0): cluster 2\n", + "Example n. 5061 = (101.0, 490.0): cluster 2\n", + "Example n. 5062 = (101.0, 489.0): cluster 2\n", + "Example n. 5063 = (101.0, 488.0): cluster 2\n", + "Example n. 5064 = (101.0, 487.0): cluster 2\n", + "Example n. 5065 = (101.0, 486.0): cluster 2\n", + "Example n. 5066 = (101.0, 485.0): cluster 2\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Example n. 5067 = (101.0, 484.0): cluster 2\n", + "Example n. 5068 = (101.0, 483.0): cluster 2\n", + "Example n. 5069 = (101.0, 482.0): cluster 2\n", + "Example n. 5070 = (101.0, 481.0): cluster 2\n", + "Example n. 5071 = (101.0, 480.0): cluster 2\n", + "Example n. 5072 = (101.0, 479.0): cluster 2\n", + "Example n. 5073 = (101.0, 478.0): cluster 2\n", + "Example n. 5074 = (101.0, 477.0): cluster 2\n", + "Example n. 5075 = (101.0, 476.0): cluster 2\n", + "Example n. 5076 = (101.0, 475.0): cluster 2\n", + "Example n. 5077 = (101.0, 474.0): cluster 2\n", + "Example n. 5078 = (101.0, 473.0): cluster 2\n", + "Example n. 5079 = (101.0, 472.0): cluster 2\n", + "Example n. 5080 = (101.0, 471.0): cluster 2\n", + "Example n. 5081 = (101.0, 470.0): cluster 2\n", + "Example n. 5082 = (101.0, 469.0): cluster 2\n", + "Example n. 5083 = (101.0, 468.0): cluster 2\n", + "Example n. 5084 = (101.0, 467.0): cluster 2\n", + "Example n. 5085 = (101.0, 466.0): cluster 2\n", + "Example n. 5086 = (101.0, 465.0): cluster 2\n", + "Example n. 5087 = (101.0, 464.0): cluster 2\n", + "Example n. 5088 = (101.0, 444.0): cluster 2\n", + "Example n. 5089 = (101.0, 443.0): cluster 2\n", + "Example n. 5090 = (101.0, 442.0): cluster 2\n", + "Example n. 5091 = (101.0, 441.0): cluster 2\n", + "Example n. 5092 = (101.0, 440.0): cluster 2\n", + "Example n. 5093 = (101.0, 439.0): cluster 2\n", + "Example n. 5094 = (101.0, 438.0): cluster 2\n", + "Example n. 5095 = (101.0, 437.0): cluster 2\n", + "Example n. 5096 = (101.0, 436.0): cluster 2\n", + "Example n. 5097 = (101.0, 435.0): cluster 2\n", + "Example n. 5098 = (101.0, 434.0): cluster 2\n", + "Example n. 5099 = (101.0, 433.0): cluster 2\n", + "Example n. 5100 = (101.0, 432.0): cluster 2\n", + "Example n. 5101 = (101.0, 431.0): cluster 2\n", + "Example n. 5102 = (101.0, 430.0): cluster 2\n", + "Example n. 5103 = (101.0, 429.0): cluster 2\n", + "Example n. 5104 = (102.0, 490.0): cluster 2\n", + "Example n. 5105 = (102.0, 489.0): cluster 2\n", + "Example n. 5106 = (102.0, 488.0): cluster 2\n", + "Example n. 5107 = (102.0, 487.0): cluster 2\n", + "Example n. 5108 = (102.0, 486.0): cluster 2\n", + "Example n. 5109 = (102.0, 485.0): cluster 2\n", + "Example n. 5110 = (102.0, 484.0): cluster 2\n", + "Example n. 5111 = (102.0, 483.0): cluster 2\n", + "Example n. 5112 = (102.0, 482.0): cluster 2\n", + "Example n. 5113 = (102.0, 481.0): cluster 2\n", + "Example n. 5114 = (102.0, 480.0): cluster 2\n", + "Example n. 5115 = (102.0, 479.0): cluster 2\n", + "Example n. 5116 = (102.0, 478.0): cluster 2\n", + "Example n. 5117 = (102.0, 477.0): cluster 2\n", + "Example n. 5118 = (102.0, 476.0): cluster 2\n", + "Example n. 5119 = (102.0, 475.0): cluster 2\n", + "Example n. 5120 = (102.0, 474.0): cluster 2\n", + "Example n. 5121 = (102.0, 473.0): cluster 2\n", + "Example n. 5122 = (102.0, 472.0): cluster 2\n", + "Example n. 5123 = (102.0, 471.0): cluster 2\n", + "Example n. 5124 = (102.0, 470.0): cluster 2\n", + "Example n. 5125 = (102.0, 469.0): cluster 2\n", + "Example n. 5126 = (102.0, 468.0): cluster 2\n", + "Example n. 5127 = (102.0, 444.0): cluster 2\n", + "Example n. 5128 = (102.0, 443.0): cluster 2\n", + "Example n. 5129 = (102.0, 442.0): cluster 2\n", + "Example n. 5130 = (102.0, 441.0): cluster 2\n", + "Example n. 5131 = (102.0, 440.0): cluster 2\n", + "Example n. 5132 = (102.0, 439.0): cluster 2\n", + "Example n. 5133 = (102.0, 438.0): cluster 2\n", + "Example n. 5134 = (102.0, 437.0): cluster 2\n", + "Example n. 5135 = (102.0, 436.0): cluster 2\n", + "Example n. 5136 = (102.0, 435.0): cluster 2\n", + "Example n. 5137 = (102.0, 434.0): cluster 2\n", + "Example n. 5138 = (102.0, 433.0): cluster 2\n", + "Example n. 5139 = (102.0, 432.0): cluster 2\n", + "Example n. 5140 = (102.0, 431.0): cluster 2\n", + "Example n. 5141 = (102.0, 430.0): cluster 2\n", + "Example n. 5142 = (102.0, 429.0): cluster 2\n", + "Example n. 5143 = (103.0, 490.0): cluster 2\n", + "Example n. 5144 = (103.0, 489.0): cluster 2\n", + "Example n. 5145 = (103.0, 488.0): cluster 2\n", + "Example n. 5146 = (103.0, 487.0): cluster 2\n", + "Example n. 5147 = (103.0, 486.0): cluster 2\n", + "Example n. 5148 = (103.0, 485.0): cluster 2\n", + "Example n. 5149 = (103.0, 484.0): cluster 2\n", + "Example n. 5150 = (103.0, 483.0): cluster 2\n", + "Example n. 5151 = (103.0, 482.0): cluster 2\n", + "Example n. 5152 = (103.0, 481.0): cluster 2\n", + "Example n. 5153 = (103.0, 480.0): cluster 2\n", + "Example n. 5154 = (103.0, 479.0): cluster 2\n", + "Example n. 5155 = (103.0, 478.0): cluster 2\n", + "Example n. 5156 = (103.0, 477.0): cluster 2\n", + "Example n. 5157 = (103.0, 476.0): cluster 2\n", + "Example n. 5158 = (103.0, 475.0): cluster 2\n", + "Example n. 5159 = (103.0, 474.0): cluster 2\n", + "Example n. 5160 = (103.0, 473.0): cluster 2\n", + "Example n. 5161 = (103.0, 472.0): cluster 2\n", + "Example n. 5162 = (103.0, 471.0): cluster 2\n", + "Example n. 5163 = (103.0, 470.0): cluster 2\n", + "Example n. 5164 = (103.0, 469.0): cluster 2\n", + "Example n. 5165 = (103.0, 468.0): cluster 2\n", + "Example n. 5166 = (103.0, 444.0): cluster 2\n", + "Example n. 5167 = (103.0, 443.0): cluster 2\n", + "Example n. 5168 = (103.0, 442.0): cluster 2\n", + "Example n. 5169 = (103.0, 441.0): cluster 2\n", + "Example n. 5170 = (103.0, 440.0): cluster 2\n", + "Example n. 5171 = (103.0, 439.0): cluster 2\n", + "Example n. 5172 = (103.0, 438.0): cluster 2\n", + "Example n. 5173 = (103.0, 437.0): cluster 2\n", + "Example n. 5174 = (103.0, 436.0): cluster 2\n", + "Example n. 5175 = (103.0, 435.0): cluster 2\n", + "Example n. 5176 = (103.0, 434.0): cluster 2\n", + "Example n. 5177 = (103.0, 433.0): cluster 2\n", + "Example n. 5178 = (103.0, 432.0): cluster 2\n", + "Example n. 5179 = (103.0, 431.0): cluster 2\n", + "Example n. 5180 = (103.0, 430.0): cluster 2\n", + "Example n. 5181 = (103.0, 429.0): cluster 2\n", + "Example n. 5182 = (104.0, 490.0): cluster 2\n", + "Example n. 5183 = (104.0, 489.0): cluster 2\n", + "Example n. 5184 = (104.0, 488.0): cluster 2\n", + "Example n. 5185 = (104.0, 487.0): cluster 2\n", + "Example n. 5186 = (104.0, 486.0): cluster 2\n", + "Example n. 5187 = (104.0, 485.0): cluster 2\n", + "Example n. 5188 = (104.0, 484.0): cluster 2\n", + "Example n. 5189 = (104.0, 483.0): cluster 2\n", + "Example n. 5190 = (104.0, 482.0): cluster 2\n", + "Example n. 5191 = (104.0, 481.0): cluster 2\n", + "Example n. 5192 = (104.0, 480.0): cluster 2\n", + "Example n. 5193 = (104.0, 479.0): cluster 2\n", + "Example n. 5194 = (104.0, 478.0): cluster 2\n", + "Example n. 5195 = (104.0, 477.0): cluster 2\n", + "Example n. 5196 = (104.0, 476.0): cluster 2\n", + "Example n. 5197 = (104.0, 475.0): cluster 2\n", + "Example n. 5198 = (104.0, 474.0): cluster 2\n", + "Example n. 5199 = (104.0, 473.0): cluster 2\n", + "Example n. 5200 = (104.0, 472.0): cluster 2\n", + "Example n. 5201 = (104.0, 471.0): cluster 2\n", + "Example n. 5202 = (104.0, 470.0): cluster 2\n", + "Example n. 5203 = (104.0, 469.0): cluster 2\n", + "Example n. 5204 = (104.0, 468.0): cluster 2\n", + "Example n. 5205 = (104.0, 444.0): cluster 2\n", + "Example n. 5206 = (104.0, 443.0): cluster 2\n", + "Example n. 5207 = (104.0, 442.0): cluster 2\n", + "Example n. 5208 = (104.0, 441.0): cluster 2\n", + "Example n. 5209 = (104.0, 440.0): cluster 2\n", + "Example n. 5210 = (104.0, 439.0): cluster 2\n", + "Example n. 5211 = (104.0, 438.0): cluster 2\n", + "Example n. 5212 = (104.0, 437.0): cluster 2\n", + "Example n. 5213 = (104.0, 436.0): cluster 2\n", + "Example n. 5214 = (104.0, 435.0): cluster 2\n", + "Example n. 5215 = (104.0, 434.0): cluster 2\n", + "Example n. 5216 = (104.0, 433.0): cluster 2\n", + "Example n. 5217 = (104.0, 432.0): cluster 2\n", + "Example n. 5218 = (104.0, 431.0): cluster 2\n", + "Example n. 5219 = (104.0, 430.0): cluster 2\n", + "Example n. 5220 = (104.0, 429.0): cluster 2\n", + "Example n. 5221 = (105.0, 490.0): cluster 2\n", + "Example n. 5222 = (105.0, 489.0): cluster 2\n", + "Example n. 5223 = (105.0, 488.0): cluster 2\n", + "Example n. 5224 = (105.0, 487.0): cluster 2\n", + "Example n. 5225 = (105.0, 486.0): cluster 2\n", + "Example n. 5226 = (105.0, 485.0): cluster 2\n", + "Example n. 5227 = (105.0, 484.0): cluster 2\n", + "Example n. 5228 = (105.0, 483.0): cluster 2\n", + "Example n. 5229 = (105.0, 482.0): cluster 2\n", + "Example n. 5230 = (105.0, 481.0): cluster 2\n", + "Example n. 5231 = (105.0, 480.0): cluster 2\n", + "Example n. 5232 = (105.0, 479.0): cluster 2\n", + "Example n. 5233 = (105.0, 478.0): cluster 2\n", + "Example n. 5234 = (105.0, 477.0): cluster 2\n", + "Example n. 5235 = (105.0, 476.0): cluster 2\n", + "Example n. 5236 = (105.0, 475.0): cluster 2\n", + "Example n. 5237 = (105.0, 474.0): cluster 2\n", + "Example n. 5238 = (105.0, 473.0): cluster 2\n", + "Example n. 5239 = (105.0, 472.0): cluster 2\n", + "Example n. 5240 = (105.0, 471.0): cluster 2\n", + "Example n. 5241 = (105.0, 470.0): cluster 2\n", + "Example n. 5242 = (105.0, 469.0): cluster 2\n", + "Example n. 5243 = (105.0, 468.0): cluster 2\n", + "Example n. 5244 = (105.0, 444.0): cluster 2\n", + "Example n. 5245 = (105.0, 443.0): cluster 2\n", + "Example n. 5246 = (105.0, 442.0): cluster 2\n", + "Example n. 5247 = (105.0, 441.0): cluster 2\n", + "Example n. 5248 = (105.0, 440.0): cluster 2\n", + "Example n. 5249 = (105.0, 439.0): cluster 2\n", + "Example n. 5250 = (105.0, 438.0): cluster 2\n", + "Example n. 5251 = (105.0, 437.0): cluster 2\n", + "Example n. 5252 = (105.0, 436.0): cluster 2\n", + "Example n. 5253 = (105.0, 435.0): cluster 2\n", + "Example n. 5254 = (105.0, 434.0): cluster 2\n", + "Example n. 5255 = (105.0, 433.0): cluster 2\n", + "Example n. 5256 = (105.0, 432.0): cluster 2\n", + "Example n. 5257 = (105.0, 431.0): cluster 2\n", + "Example n. 5258 = (105.0, 430.0): cluster 2\n", + "Example n. 5259 = (105.0, 429.0): cluster 2\n", + "Example n. 5260 = (106.0, 490.0): cluster 2\n", + "Example n. 5261 = (106.0, 489.0): cluster 2\n", + "Example n. 5262 = (106.0, 488.0): cluster 2\n", + "Example n. 5263 = (106.0, 487.0): cluster 2\n", + "Example n. 5264 = (106.0, 486.0): cluster 2\n", + "Example n. 5265 = (106.0, 485.0): cluster 2\n", + "Example n. 5266 = (106.0, 484.0): cluster 2\n", + "Example n. 5267 = (106.0, 483.0): cluster 2\n", + "Example n. 5268 = (106.0, 482.0): cluster 2\n", + "Example n. 5269 = (106.0, 481.0): cluster 2\n", + "Example n. 5270 = (106.0, 480.0): cluster 2\n", + "Example n. 5271 = (106.0, 479.0): cluster 2\n", + "Example n. 5272 = (106.0, 478.0): cluster 2\n", + "Example n. 5273 = (106.0, 477.0): cluster 2\n", + "Example n. 5274 = (106.0, 476.0): cluster 2\n", + "Example n. 5275 = (106.0, 475.0): cluster 2\n", + "Example n. 5276 = (106.0, 440.0): cluster 2\n", + "Example n. 5277 = (106.0, 439.0): cluster 2\n", + "Example n. 5278 = (106.0, 438.0): cluster 2\n", + "Example n. 5279 = (106.0, 437.0): cluster 2\n", + "Example n. 5280 = (106.0, 436.0): cluster 2\n", + "Example n. 5281 = (106.0, 435.0): cluster 2\n", + "Example n. 5282 = (106.0, 434.0): cluster 2\n", + "Example n. 5283 = (106.0, 433.0): cluster 2\n", + "Example n. 5284 = (106.0, 432.0): cluster 2\n", + "Example n. 5285 = (106.0, 431.0): cluster 2\n", + "Example n. 5286 = (106.0, 430.0): cluster 2\n", + "Example n. 5287 = (106.0, 429.0): cluster 2\n", + "Example n. 5288 = (107.0, 490.0): cluster 2\n", + "Example n. 5289 = (107.0, 489.0): cluster 2\n", + "Example n. 5290 = (107.0, 488.0): cluster 2\n", + "Example n. 5291 = (107.0, 487.0): cluster 2\n", + "Example n. 5292 = (107.0, 486.0): cluster 2\n", + "Example n. 5293 = (107.0, 485.0): cluster 2\n", + "Example n. 5294 = (107.0, 484.0): cluster 2\n", + "Example n. 5295 = (107.0, 483.0): cluster 2\n", + "Example n. 5296 = (107.0, 482.0): cluster 2\n", + "Example n. 5297 = (107.0, 481.0): cluster 2\n", + "Example n. 5298 = (107.0, 480.0): cluster 2\n", + "Example n. 5299 = (107.0, 479.0): cluster 2\n", + "Example n. 5300 = (107.0, 478.0): cluster 2\n", + "Example n. 5301 = (107.0, 477.0): cluster 2\n", + "Example n. 5302 = (107.0, 476.0): cluster 2\n", + "Example n. 5303 = (107.0, 475.0): cluster 2\n", + "Example n. 5304 = (107.0, 440.0): cluster 2\n", + "Example n. 5305 = (107.0, 439.0): cluster 2\n", + "Example n. 5306 = (107.0, 438.0): cluster 2\n", + "Example n. 5307 = (107.0, 437.0): cluster 2\n", + "Example n. 5308 = (107.0, 436.0): cluster 2\n", + "Example n. 5309 = (107.0, 435.0): cluster 2\n", + "Example n. 5310 = (107.0, 434.0): cluster 2\n", + "Example n. 5311 = (107.0, 433.0): cluster 2\n", + "Example n. 5312 = (107.0, 432.0): cluster 2\n", + "Example n. 5313 = (107.0, 431.0): cluster 2\n", + "Example n. 5314 = (107.0, 430.0): cluster 2\n", + "Example n. 5315 = (107.0, 429.0): cluster 2\n", + "Example n. 5316 = (108.0, 490.0): cluster 2\n", + "Example n. 5317 = (108.0, 489.0): cluster 2\n", + "Example n. 5318 = (108.0, 488.0): cluster 2\n", + "Example n. 5319 = (108.0, 487.0): cluster 2\n", + "Example n. 5320 = (108.0, 486.0): cluster 2\n", + "Example n. 5321 = (108.0, 485.0): cluster 2\n", + "Example n. 5322 = (108.0, 484.0): cluster 2\n", + "Example n. 5323 = (108.0, 483.0): cluster 2\n", + "Example n. 5324 = (108.0, 482.0): cluster 2\n", + "Example n. 5325 = (108.0, 481.0): cluster 2\n", + "Example n. 5326 = (108.0, 480.0): cluster 2\n", + "Example n. 5327 = (108.0, 479.0): cluster 2\n", + "Example n. 5328 = (108.0, 478.0): cluster 2\n", + "Example n. 5329 = (108.0, 477.0): cluster 2\n", + "Example n. 5330 = (108.0, 476.0): cluster 2\n", + "Example n. 5331 = (108.0, 475.0): cluster 2\n", + "Example n. 5332 = (108.0, 440.0): cluster 2\n", + "Example n. 5333 = (108.0, 439.0): cluster 2\n", + "Example n. 5334 = (108.0, 438.0): cluster 2\n", + "Example n. 5335 = (108.0, 437.0): cluster 2\n", + "Example n. 5336 = (108.0, 436.0): cluster 2\n", + "Example n. 5337 = (108.0, 435.0): cluster 2\n", + "Example n. 5338 = (108.0, 434.0): cluster 2\n", + "Example n. 5339 = (108.0, 433.0): cluster 2\n", + "Example n. 5340 = (108.0, 432.0): cluster 2\n", + "Example n. 5341 = (108.0, 431.0): cluster 2\n", + "Example n. 5342 = (108.0, 430.0): cluster 2\n", + "Example n. 5343 = (108.0, 429.0): cluster 2\n", + "Example n. 5344 = (109.0, 482.0): cluster 2\n", + "Example n. 5345 = (109.0, 481.0): cluster 2\n", + "Example n. 5346 = (109.0, 480.0): cluster 2\n", + "Example n. 5347 = (109.0, 479.0): cluster 2\n", + "Example n. 5348 = (109.0, 478.0): cluster 2\n", + "Example n. 5349 = (109.0, 477.0): cluster 2\n", + "Example n. 5350 = (109.0, 476.0): cluster 2\n", + "Example n. 5351 = (109.0, 475.0): cluster 2\n", + "Example n. 5352 = (109.0, 474.0): cluster 2\n", + "Example n. 5353 = (109.0, 473.0): cluster 2\n", + "Example n. 5354 = (109.0, 472.0): cluster 2\n", + "Example n. 5355 = (109.0, 471.0): cluster 2\n", + "Example n. 5356 = (109.0, 470.0): cluster 2\n", + "Example n. 5357 = (109.0, 469.0): cluster 2\n", + "Example n. 5358 = (109.0, 468.0): cluster 2\n", + "Example n. 5359 = (109.0, 440.0): cluster 2\n", + "Example n. 5360 = (109.0, 439.0): cluster 2\n", + "Example n. 5361 = (109.0, 438.0): cluster 2\n", + "Example n. 5362 = (109.0, 437.0): cluster 2\n", + "Example n. 5363 = (109.0, 436.0): cluster 2\n", + "Example n. 5364 = (109.0, 435.0): cluster 2\n", + "Example n. 5365 = (109.0, 434.0): cluster 2\n", + "Example n. 5366 = (109.0, 433.0): cluster 2\n", + "Example n. 5367 = (109.0, 432.0): cluster 2\n", + "Example n. 5368 = (109.0, 431.0): cluster 2\n", + "Example n. 5369 = (109.0, 430.0): cluster 2\n", + "Example n. 5370 = (109.0, 429.0): cluster 2\n", + "Example n. 5371 = (109.0, 428.0): cluster 2\n", + "Example n. 5372 = (109.0, 427.0): cluster 2\n", + "Example n. 5373 = (109.0, 426.0): cluster 2\n", + "Example n. 5374 = (109.0, 425.0): cluster 2\n", + "Example n. 5375 = (110.0, 482.0): cluster 2\n", + "Example n. 5376 = (110.0, 481.0): cluster 2\n", + "Example n. 5377 = (110.0, 480.0): cluster 2\n", + "Example n. 5378 = (110.0, 479.0): cluster 2\n", + "Example n. 5379 = (110.0, 478.0): cluster 2\n", + "Example n. 5380 = (110.0, 477.0): cluster 2\n", + "Example n. 5381 = (110.0, 476.0): cluster 2\n", + "Example n. 5382 = (110.0, 475.0): cluster 2\n", + "Example n. 5383 = (110.0, 474.0): cluster 2\n", + "Example n. 5384 = (110.0, 473.0): cluster 2\n", + "Example n. 5385 = (110.0, 472.0): cluster 2\n", + "Example n. 5386 = (110.0, 471.0): cluster 2\n", + "Example n. 5387 = (110.0, 470.0): cluster 2\n", + "Example n. 5388 = (110.0, 469.0): cluster 2\n", + "Example n. 5389 = (110.0, 468.0): cluster 2\n", + "Example n. 5390 = (110.0, 440.0): cluster 2\n", + "Example n. 5391 = (110.0, 439.0): cluster 2\n", + "Example n. 5392 = (110.0, 438.0): cluster 2\n", + "Example n. 5393 = (110.0, 437.0): cluster 2\n", + "Example n. 5394 = (110.0, 436.0): cluster 2\n", + "Example n. 5395 = (110.0, 435.0): cluster 2\n", + "Example n. 5396 = (110.0, 434.0): cluster 2\n", + "Example n. 5397 = (110.0, 433.0): cluster 2\n", + "Example n. 5398 = (110.0, 432.0): cluster 2\n", + "Example n. 5399 = (110.0, 431.0): cluster 2\n", + "Example n. 5400 = (110.0, 430.0): cluster 2\n", + "Example n. 5401 = (110.0, 429.0): cluster 2\n", + "Example n. 5402 = (110.0, 428.0): cluster 2\n", + "Example n. 5403 = (110.0, 427.0): cluster 2\n", + "Example n. 5404 = (110.0, 426.0): cluster 2\n", + "Example n. 5405 = (110.0, 425.0): cluster 2\n", + "Example n. 5406 = (111.0, 482.0): cluster 2\n", + "Example n. 5407 = (111.0, 481.0): cluster 2\n", + "Example n. 5408 = (111.0, 480.0): cluster 2\n", + "Example n. 5409 = (111.0, 479.0): cluster 2\n", + "Example n. 5410 = (111.0, 478.0): cluster 2\n", + "Example n. 5411 = (111.0, 477.0): cluster 2\n", + "Example n. 5412 = (111.0, 476.0): cluster 2\n", + "Example n. 5413 = (111.0, 475.0): cluster 2\n", + "Example n. 5414 = (111.0, 474.0): cluster 2\n", + "Example n. 5415 = (111.0, 473.0): cluster 2\n", + "Example n. 5416 = (111.0, 472.0): cluster 2\n", + "Example n. 5417 = (111.0, 471.0): cluster 2\n", + "Example n. 5418 = (111.0, 470.0): cluster 2\n", + "Example n. 5419 = (111.0, 469.0): cluster 2\n", + "Example n. 5420 = (111.0, 468.0): cluster 2\n", + "Example n. 5421 = (111.0, 440.0): cluster 2\n", + "Example n. 5422 = (111.0, 439.0): cluster 2\n", + "Example n. 5423 = (111.0, 438.0): cluster 2\n", + "Example n. 5424 = (111.0, 437.0): cluster 2\n", + "Example n. 5425 = (111.0, 436.0): cluster 2\n", + "Example n. 5426 = (111.0, 435.0): cluster 2\n", + "Example n. 5427 = (111.0, 434.0): cluster 2\n", + "Example n. 5428 = (111.0, 433.0): cluster 2\n", + "Example n. 5429 = (111.0, 432.0): cluster 2\n", + "Example n. 5430 = (111.0, 431.0): cluster 2\n", + "Example n. 5431 = (111.0, 430.0): cluster 2\n", + "Example n. 5432 = (111.0, 429.0): cluster 2\n", + "Example n. 5433 = (111.0, 428.0): cluster 2\n", + "Example n. 5434 = (111.0, 427.0): cluster 2\n", + "Example n. 5435 = (111.0, 426.0): cluster 2\n", + "Example n. 5436 = (111.0, 425.0): cluster 2\n", + "Example n. 5437 = (112.0, 482.0): cluster 2\n", + "Example n. 5438 = (112.0, 481.0): cluster 2\n", + "Example n. 5439 = (112.0, 480.0): cluster 2\n", + "Example n. 5440 = (112.0, 479.0): cluster 2\n", + "Example n. 5441 = (112.0, 478.0): cluster 2\n", + "Example n. 5442 = (112.0, 477.0): cluster 2\n", + "Example n. 5443 = (112.0, 476.0): cluster 2\n", + "Example n. 5444 = (112.0, 475.0): cluster 2\n", + "Example n. 5445 = (112.0, 474.0): cluster 2\n", + "Example n. 5446 = (112.0, 473.0): cluster 2\n", + "Example n. 5447 = (112.0, 472.0): cluster 2\n", + "Example n. 5448 = (112.0, 471.0): cluster 2\n", + "Example n. 5449 = (112.0, 470.0): cluster 2\n", + "Example n. 5450 = (112.0, 469.0): cluster 2\n", + "Example n. 5451 = (112.0, 468.0): cluster 2\n", + "Example n. 5452 = (112.0, 440.0): cluster 2\n", + "Example n. 5453 = (112.0, 439.0): cluster 2\n", + "Example n. 5454 = (112.0, 438.0): cluster 2\n", + "Example n. 5455 = (112.0, 437.0): cluster 2\n", + "Example n. 5456 = (112.0, 436.0): cluster 2\n", + "Example n. 5457 = (112.0, 435.0): cluster 2\n", + "Example n. 5458 = (112.0, 434.0): cluster 2\n", + "Example n. 5459 = (112.0, 433.0): cluster 2\n", + "Example n. 5460 = (112.0, 432.0): cluster 2\n", + "Example n. 5461 = (112.0, 431.0): cluster 2\n", + "Example n. 5462 = (112.0, 430.0): cluster 2\n", + "Example n. 5463 = (112.0, 429.0): cluster 2\n", + "Example n. 5464 = (112.0, 428.0): cluster 2\n", + "Example n. 5465 = (112.0, 427.0): cluster 2\n", + "Example n. 5466 = (112.0, 426.0): cluster 2\n", + "Example n. 5467 = (112.0, 425.0): cluster 2\n", + "Example n. 5468 = (113.0, 482.0): cluster 2\n", + "Example n. 5469 = (113.0, 481.0): cluster 2\n", + "Example n. 5470 = (113.0, 480.0): cluster 2\n", + "Example n. 5471 = (113.0, 479.0): cluster 2\n", + "Example n. 5472 = (113.0, 478.0): cluster 2\n", + "Example n. 5473 = (113.0, 477.0): cluster 2\n", + "Example n. 5474 = (113.0, 476.0): cluster 2\n", + "Example n. 5475 = (113.0, 475.0): cluster 2\n", + "Example n. 5476 = (113.0, 474.0): cluster 2\n", + "Example n. 5477 = (113.0, 473.0): cluster 2\n", + "Example n. 5478 = (113.0, 472.0): cluster 2\n", + "Example n. 5479 = (113.0, 471.0): cluster 2\n", + "Example n. 5480 = (113.0, 470.0): cluster 2\n", + "Example n. 5481 = (113.0, 469.0): cluster 2\n", + "Example n. 5482 = (113.0, 468.0): cluster 2\n", + "Example n. 5483 = (113.0, 467.0): cluster 2\n", + "Example n. 5484 = (113.0, 466.0): cluster 2\n", + "Example n. 5485 = (113.0, 465.0): cluster 2\n", + "Example n. 5486 = (113.0, 464.0): cluster 2\n", + "Example n. 5487 = (113.0, 436.0): cluster 2\n", + "Example n. 5488 = (113.0, 435.0): cluster 2\n", + "Example n. 5489 = (113.0, 434.0): cluster 2\n", + "Example n. 5490 = (113.0, 433.0): cluster 2\n", + "Example n. 5491 = (113.0, 432.0): cluster 2\n", + "Example n. 5492 = (113.0, 431.0): cluster 2\n", + "Example n. 5493 = (113.0, 430.0): cluster 2\n", + "Example n. 5494 = (113.0, 429.0): cluster 2\n", + "Example n. 5495 = (113.0, 428.0): cluster 2\n", + "Example n. 5496 = (113.0, 427.0): cluster 2\n", + "Example n. 5497 = (113.0, 426.0): cluster 2\n", + "Example n. 5498 = (113.0, 425.0): cluster 2\n", + "Example n. 5499 = (114.0, 482.0): cluster 2\n", + "Example n. 5500 = (114.0, 481.0): cluster 2\n", + "Example n. 5501 = (114.0, 480.0): cluster 2\n", + "Example n. 5502 = (114.0, 479.0): cluster 2\n", + "Example n. 5503 = (114.0, 478.0): cluster 2\n", + "Example n. 5504 = (114.0, 477.0): cluster 2\n", + "Example n. 5505 = (114.0, 476.0): cluster 2\n", + "Example n. 5506 = (114.0, 475.0): cluster 2\n", + "Example n. 5507 = (114.0, 474.0): cluster 2\n", + "Example n. 5508 = (114.0, 473.0): cluster 2\n", + "Example n. 5509 = (114.0, 472.0): cluster 2\n", + "Example n. 5510 = (114.0, 471.0): cluster 2\n", + "Example n. 5511 = (114.0, 470.0): cluster 2\n", + "Example n. 5512 = (114.0, 469.0): cluster 2\n", + "Example n. 5513 = (114.0, 468.0): cluster 2\n", + "Example n. 5514 = (114.0, 467.0): cluster 2\n", + "Example n. 5515 = (114.0, 466.0): cluster 2\n", + "Example n. 5516 = (114.0, 465.0): cluster 2\n", + "Example n. 5517 = (114.0, 464.0): cluster 2\n", + "Example n. 5518 = (114.0, 436.0): cluster 2\n", + "Example n. 5519 = (114.0, 435.0): cluster 2\n", + "Example n. 5520 = (114.0, 434.0): cluster 2\n", + "Example n. 5521 = (114.0, 433.0): cluster 2\n", + "Example n. 5522 = (114.0, 432.0): cluster 2\n", + "Example n. 5523 = (114.0, 431.0): cluster 2\n", + "Example n. 5524 = (114.0, 430.0): cluster 2\n", + "Example n. 5525 = (114.0, 429.0): cluster 2\n", + "Example n. 5526 = (114.0, 428.0): cluster 2\n", + "Example n. 5527 = (114.0, 427.0): cluster 2\n", + "Example n. 5528 = (114.0, 426.0): cluster 2\n", + "Example n. 5529 = (114.0, 425.0): cluster 2\n", + "Example n. 5530 = (115.0, 482.0): cluster 2\n", + "Example n. 5531 = (115.0, 481.0): cluster 2\n", + "Example n. 5532 = (115.0, 480.0): cluster 2\n", + "Example n. 5533 = (115.0, 479.0): cluster 2\n", + "Example n. 5534 = (115.0, 478.0): cluster 2\n", + "Example n. 5535 = (115.0, 477.0): cluster 2\n", + "Example n. 5536 = (115.0, 476.0): cluster 2\n", + "Example n. 5537 = (115.0, 475.0): cluster 2\n", + "Example n. 5538 = (115.0, 474.0): cluster 2\n", + "Example n. 5539 = (115.0, 473.0): cluster 2\n", + "Example n. 5540 = (115.0, 472.0): cluster 2\n", + "Example n. 5541 = (115.0, 471.0): cluster 2\n", + "Example n. 5542 = (115.0, 470.0): cluster 2\n", + "Example n. 5543 = (115.0, 469.0): cluster 2\n", + "Example n. 5544 = (115.0, 468.0): cluster 2\n", + "Example n. 5545 = (115.0, 467.0): cluster 2\n", + "Example n. 5546 = (115.0, 466.0): cluster 2\n", + "Example n. 5547 = (115.0, 465.0): cluster 2\n", + "Example n. 5548 = (115.0, 464.0): cluster 2\n", + "Example n. 5549 = (115.0, 436.0): cluster 2\n", + "Example n. 5550 = (115.0, 435.0): cluster 2\n", + "Example n. 5551 = (115.0, 434.0): cluster 2\n", + "Example n. 5552 = (115.0, 433.0): cluster 2\n", + "Example n. 5553 = (115.0, 432.0): cluster 2\n", + "Example n. 5554 = (115.0, 431.0): cluster 2\n", + "Example n. 5555 = (115.0, 430.0): cluster 2\n", + "Example n. 5556 = (115.0, 429.0): cluster 2\n", + "Example n. 5557 = (115.0, 428.0): cluster 2\n", + "Example n. 5558 = (115.0, 427.0): cluster 2\n", + "Example n. 5559 = (115.0, 426.0): cluster 2\n", + "Example n. 5560 = (115.0, 425.0): cluster 2\n", + "Example n. 5561 = (116.0, 482.0): cluster 2\n", + "Example n. 5562 = (116.0, 481.0): cluster 2\n", + "Example n. 5563 = (116.0, 480.0): cluster 2\n", + "Example n. 5564 = (116.0, 479.0): cluster 2\n", + "Example n. 5565 = (116.0, 478.0): cluster 2\n", + "Example n. 5566 = (116.0, 477.0): cluster 2\n", + "Example n. 5567 = (116.0, 476.0): cluster 2\n", + "Example n. 5568 = (116.0, 475.0): cluster 2\n", + "Example n. 5569 = (116.0, 474.0): cluster 2\n", + "Example n. 5570 = (116.0, 473.0): cluster 2\n", + "Example n. 5571 = (116.0, 472.0): cluster 2\n", + "Example n. 5572 = (116.0, 471.0): cluster 2\n", + "Example n. 5573 = (116.0, 470.0): cluster 2\n", + "Example n. 5574 = (116.0, 469.0): cluster 2\n", + "Example n. 5575 = (116.0, 468.0): cluster 2\n", + "Example n. 5576 = (116.0, 467.0): cluster 2\n", + "Example n. 5577 = (116.0, 466.0): cluster 2\n", + "Example n. 5578 = (116.0, 465.0): cluster 2\n", + "Example n. 5579 = (116.0, 464.0): cluster 2\n", + "Example n. 5580 = (116.0, 436.0): cluster 2\n", + "Example n. 5581 = (116.0, 435.0): cluster 2\n", + "Example n. 5582 = (116.0, 434.0): cluster 2\n", + "Example n. 5583 = (116.0, 433.0): cluster 2\n", + "Example n. 5584 = (116.0, 432.0): cluster 2\n", + "Example n. 5585 = (116.0, 431.0): cluster 2\n", + "Example n. 5586 = (116.0, 430.0): cluster 2\n", + "Example n. 5587 = (116.0, 429.0): cluster 2\n", + "Example n. 5588 = (116.0, 428.0): cluster 2\n", + "Example n. 5589 = (116.0, 427.0): cluster 2\n", + "Example n. 5590 = (116.0, 426.0): cluster 2\n", + "Example n. 5591 = (116.0, 425.0): cluster 2\n", + "Example n. 5592 = (117.0, 475.0): cluster 2\n", + "Example n. 5593 = (117.0, 474.0): cluster 2\n", + "Example n. 5594 = (117.0, 473.0): cluster 2\n", + "Example n. 5595 = (117.0, 472.0): cluster 2\n", + "Example n. 5596 = (117.0, 471.0): cluster 2\n", + "Example n. 5597 = (117.0, 470.0): cluster 2\n", + "Example n. 5598 = (117.0, 469.0): cluster 2\n", + "Example n. 5599 = (117.0, 468.0): cluster 2\n", + "Example n. 5600 = (117.0, 467.0): cluster 2\n", + "Example n. 5601 = (117.0, 466.0): cluster 2\n", + "Example n. 5602 = (117.0, 465.0): cluster 2\n", + "Example n. 5603 = (117.0, 464.0): cluster 2\n", + "Example n. 5604 = (117.0, 432.0): cluster 2\n", + "Example n. 5605 = (117.0, 431.0): cluster 2\n", + "Example n. 5606 = (117.0, 430.0): cluster 2\n", + "Example n. 5607 = (117.0, 429.0): cluster 2\n", + "Example n. 5608 = (117.0, 428.0): cluster 2\n", + "Example n. 5609 = (117.0, 427.0): cluster 2\n", + "Example n. 5610 = (117.0, 426.0): cluster 2\n", + "Example n. 5611 = (117.0, 425.0): cluster 2\n", + "Example n. 5612 = (117.0, 424.0): cluster 2\n", + "Example n. 5613 = (117.0, 423.0): cluster 2\n", + "Example n. 5614 = (117.0, 422.0): cluster 2\n", + "Example n. 5615 = (117.0, 421.0): cluster 2\n", + "Example n. 5616 = (118.0, 475.0): cluster 2\n", + "Example n. 5617 = (118.0, 474.0): cluster 2\n", + "Example n. 5618 = (118.0, 473.0): cluster 2\n", + "Example n. 5619 = (118.0, 472.0): cluster 2\n", + "Example n. 5620 = (118.0, 471.0): cluster 2\n", + "Example n. 5621 = (118.0, 470.0): cluster 2\n", + "Example n. 5622 = (118.0, 469.0): cluster 2\n", + "Example n. 5623 = (118.0, 468.0): cluster 2\n", + "Example n. 5624 = (118.0, 467.0): cluster 2\n", + "Example n. 5625 = (118.0, 466.0): cluster 2\n", + "Example n. 5626 = (118.0, 465.0): cluster 2\n", + "Example n. 5627 = (118.0, 464.0): cluster 2\n", + "Example n. 5628 = (118.0, 432.0): cluster 2\n", + "Example n. 5629 = (118.0, 431.0): cluster 2\n", + "Example n. 5630 = (118.0, 430.0): cluster 2\n", + "Example n. 5631 = (118.0, 429.0): cluster 2\n", + "Example n. 5632 = (118.0, 428.0): cluster 2\n", + "Example n. 5633 = (118.0, 427.0): cluster 2\n", + "Example n. 5634 = (118.0, 426.0): cluster 2\n", + "Example n. 5635 = (118.0, 425.0): cluster 2\n", + "Example n. 5636 = (118.0, 424.0): cluster 2\n", + "Example n. 5637 = (118.0, 423.0): cluster 2\n", + "Example n. 5638 = (118.0, 422.0): cluster 2\n", + "Example n. 5639 = (118.0, 421.0): cluster 2\n", + "Example n. 5640 = (119.0, 475.0): cluster 2\n", + "Example n. 5641 = (119.0, 474.0): cluster 2\n", + "Example n. 5642 = (119.0, 473.0): cluster 2\n", + "Example n. 5643 = (119.0, 472.0): cluster 2\n", + "Example n. 5644 = (119.0, 471.0): cluster 2\n", + "Example n. 5645 = (119.0, 470.0): cluster 2\n", + "Example n. 5646 = (119.0, 469.0): cluster 2\n", + "Example n. 5647 = (119.0, 468.0): cluster 2\n", + "Example n. 5648 = (119.0, 467.0): cluster 2\n", + "Example n. 5649 = (119.0, 466.0): cluster 2\n", + "Example n. 5650 = (119.0, 465.0): cluster 2\n", + "Example n. 5651 = (119.0, 464.0): cluster 2\n", + "Example n. 5652 = (119.0, 432.0): cluster 2\n", + "Example n. 5653 = (119.0, 431.0): cluster 2\n", + "Example n. 5654 = (119.0, 430.0): cluster 2\n", + "Example n. 5655 = (119.0, 429.0): cluster 2\n", + "Example n. 5656 = (119.0, 428.0): cluster 2\n", + "Example n. 5657 = (119.0, 427.0): cluster 2\n", + "Example n. 5658 = (119.0, 426.0): cluster 2\n", + "Example n. 5659 = (119.0, 425.0): cluster 2\n", + "Example n. 5660 = (119.0, 424.0): cluster 2\n", + "Example n. 5661 = (119.0, 423.0): cluster 2\n", + "Example n. 5662 = (119.0, 422.0): cluster 2\n", + "Example n. 5663 = (119.0, 421.0): cluster 2\n", + "Example n. 5664 = (120.0, 475.0): cluster 2\n", + "Example n. 5665 = (120.0, 474.0): cluster 2\n", + "Example n. 5666 = (120.0, 473.0): cluster 2\n", + "Example n. 5667 = (120.0, 472.0): cluster 2\n", + "Example n. 5668 = (120.0, 471.0): cluster 2\n", + "Example n. 5669 = (120.0, 470.0): cluster 2\n", + "Example n. 5670 = (120.0, 469.0): cluster 2\n", + "Example n. 5671 = (120.0, 468.0): cluster 2\n", + "Example n. 5672 = (120.0, 467.0): cluster 2\n", + "Example n. 5673 = (120.0, 466.0): cluster 2\n", + "Example n. 5674 = (120.0, 465.0): cluster 2\n", + "Example n. 5675 = (120.0, 464.0): cluster 2\n", + "Example n. 5676 = (120.0, 455.0): cluster 2\n", + "Example n. 5677 = (120.0, 454.0): cluster 2\n", + "Example n. 5678 = (120.0, 453.0): cluster 2\n", + "Example n. 5679 = (120.0, 452.0): cluster 2\n", + "Example n. 5680 = (120.0, 451.0): cluster 2\n", + "Example n. 5681 = (120.0, 450.0): cluster 2\n", + "Example n. 5682 = (120.0, 449.0): cluster 2\n", + "Example n. 5683 = (120.0, 448.0): cluster 2\n", + "Example n. 5684 = (120.0, 447.0): cluster 2\n", + "Example n. 5685 = (120.0, 446.0): cluster 2\n", + "Example n. 5686 = (120.0, 445.0): cluster 2\n", + "Example n. 5687 = (120.0, 444.0): cluster 2\n", + "Example n. 5688 = (120.0, 440.0): cluster 2\n", + "Example n. 5689 = (120.0, 439.0): cluster 2\n", + "Example n. 5690 = (120.0, 438.0): cluster 2\n", + "Example n. 5691 = (120.0, 437.0): cluster 2\n", + "Example n. 5692 = (120.0, 436.0): cluster 2\n", + "Example n. 5693 = (120.0, 435.0): cluster 2\n", + "Example n. 5694 = (120.0, 434.0): cluster 2\n", + "Example n. 5695 = (120.0, 433.0): cluster 2\n", + "Example n. 5696 = (120.0, 432.0): cluster 2\n", + "Example n. 5697 = (120.0, 431.0): cluster 2\n", + "Example n. 5698 = (120.0, 430.0): cluster 2\n", + "Example n. 5699 = (120.0, 429.0): cluster 2\n", + "Example n. 5700 = (120.0, 428.0): cluster 2\n", + "Example n. 5701 = (120.0, 427.0): cluster 2\n", + "Example n. 5702 = (120.0, 426.0): cluster 2\n", + "Example n. 5703 = (120.0, 425.0): cluster 2\n", + "Example n. 5704 = (120.0, 424.0): cluster 2\n", + "Example n. 5705 = (120.0, 423.0): cluster 2\n", + "Example n. 5706 = (120.0, 422.0): cluster 2\n", + "Example n. 5707 = (120.0, 421.0): cluster 2\n", + "Example n. 5708 = (121.0, 471.0): cluster 2\n", + "Example n. 5709 = (121.0, 470.0): cluster 2\n", + "Example n. 5710 = (121.0, 469.0): cluster 2\n", + "Example n. 5711 = (121.0, 468.0): cluster 2\n", + "Example n. 5712 = (121.0, 467.0): cluster 2\n", + "Example n. 5713 = (121.0, 466.0): cluster 2\n", + "Example n. 5714 = (121.0, 465.0): cluster 2\n", + "Example n. 5715 = (121.0, 464.0): cluster 2\n", + "Example n. 5716 = (121.0, 455.0): cluster 2\n", + "Example n. 5717 = (121.0, 454.0): cluster 2\n", + "Example n. 5718 = (121.0, 453.0): cluster 2\n", + "Example n. 5719 = (121.0, 452.0): cluster 2\n", + "Example n. 5720 = (121.0, 451.0): cluster 2\n", + "Example n. 5721 = (121.0, 450.0): cluster 2\n", + "Example n. 5722 = (121.0, 449.0): cluster 2\n", + "Example n. 5723 = (121.0, 448.0): cluster 2\n", + "Example n. 5724 = (121.0, 447.0): cluster 2\n", + "Example n. 5725 = (121.0, 446.0): cluster 2\n", + "Example n. 5726 = (121.0, 445.0): cluster 2\n", + "Example n. 5727 = (121.0, 444.0): cluster 2\n", + "Example n. 5728 = (121.0, 440.0): cluster 2\n", + "Example n. 5729 = (121.0, 439.0): cluster 2\n", + "Example n. 5730 = (121.0, 438.0): cluster 2\n", + "Example n. 5731 = (121.0, 437.0): cluster 2\n", + "Example n. 5732 = (121.0, 436.0): cluster 2\n", + "Example n. 5733 = (121.0, 435.0): cluster 2\n", + "Example n. 5734 = (121.0, 434.0): cluster 2\n", + "Example n. 5735 = (121.0, 433.0): cluster 2\n", + "Example n. 5736 = (121.0, 432.0): cluster 2\n", + "Example n. 5737 = (121.0, 431.0): cluster 2\n", + "Example n. 5738 = (121.0, 430.0): cluster 2\n", + "Example n. 5739 = (121.0, 429.0): cluster 2\n", + "Example n. 5740 = (121.0, 428.0): cluster 2\n", + "Example n. 5741 = (121.0, 427.0): cluster 2\n", + "Example n. 5742 = (121.0, 426.0): cluster 2\n", + "Example n. 5743 = (121.0, 425.0): cluster 2\n", + "Example n. 5744 = (121.0, 424.0): cluster 2\n", + "Example n. 5745 = (121.0, 423.0): cluster 2\n", + "Example n. 5746 = (121.0, 422.0): cluster 2\n", + "Example n. 5747 = (121.0, 421.0): cluster 2\n", + "Example n. 5748 = (122.0, 471.0): cluster 2\n", + "Example n. 5749 = (122.0, 470.0): cluster 2\n", + "Example n. 5750 = (122.0, 469.0): cluster 2\n", + "Example n. 5751 = (122.0, 468.0): cluster 2\n", + "Example n. 5752 = (122.0, 467.0): cluster 2\n", + "Example n. 5753 = (122.0, 466.0): cluster 2\n", + "Example n. 5754 = (122.0, 465.0): cluster 2\n", + "Example n. 5755 = (122.0, 464.0): cluster 2\n", + "Example n. 5756 = (122.0, 455.0): cluster 2\n", + "Example n. 5757 = (122.0, 454.0): cluster 2\n", + "Example n. 5758 = (122.0, 453.0): cluster 2\n", + "Example n. 5759 = (122.0, 452.0): cluster 2\n", + "Example n. 5760 = (122.0, 451.0): cluster 2\n", + "Example n. 5761 = (122.0, 450.0): cluster 2\n", + "Example n. 5762 = (122.0, 449.0): cluster 2\n", + "Example n. 5763 = (122.0, 448.0): cluster 2\n", + "Example n. 5764 = (122.0, 447.0): cluster 2\n", + "Example n. 5765 = (122.0, 446.0): cluster 2\n", + "Example n. 5766 = (122.0, 445.0): cluster 2\n", + "Example n. 5767 = (122.0, 444.0): cluster 2\n", + "Example n. 5768 = (122.0, 440.0): cluster 2\n", + "Example n. 5769 = (122.0, 439.0): cluster 2\n", + "Example n. 5770 = (122.0, 438.0): cluster 2\n", + "Example n. 5771 = (122.0, 437.0): cluster 2\n", + "Example n. 5772 = (122.0, 436.0): cluster 2\n", + "Example n. 5773 = (122.0, 435.0): cluster 2\n", + "Example n. 5774 = (122.0, 434.0): cluster 2\n", + "Example n. 5775 = (122.0, 433.0): cluster 2\n", + "Example n. 5776 = (122.0, 432.0): cluster 2\n", + "Example n. 5777 = (122.0, 431.0): cluster 2\n", + "Example n. 5778 = (122.0, 430.0): cluster 2\n", + "Example n. 5779 = (122.0, 429.0): cluster 2\n", + "Example n. 5780 = (122.0, 428.0): cluster 2\n", + "Example n. 5781 = (122.0, 427.0): cluster 2\n", + "Example n. 5782 = (122.0, 426.0): cluster 2\n", + "Example n. 5783 = (122.0, 425.0): cluster 2\n", + "Example n. 5784 = (122.0, 424.0): cluster 2\n", + "Example n. 5785 = (122.0, 423.0): cluster 2\n", + "Example n. 5786 = (122.0, 422.0): cluster 2\n", + "Example n. 5787 = (122.0, 421.0): cluster 2\n", + "Example n. 5788 = (123.0, 471.0): cluster 2\n", + "Example n. 5789 = (123.0, 470.0): cluster 2\n", + "Example n. 5790 = (123.0, 469.0): cluster 2\n", + "Example n. 5791 = (123.0, 468.0): cluster 2\n", + "Example n. 5792 = (123.0, 467.0): cluster 2\n", + "Example n. 5793 = (123.0, 466.0): cluster 2\n", + "Example n. 5794 = (123.0, 465.0): cluster 2\n", + "Example n. 5795 = (123.0, 464.0): cluster 2\n", + "Example n. 5796 = (123.0, 455.0): cluster 2\n", + "Example n. 5797 = (123.0, 454.0): cluster 2\n", + "Example n. 5798 = (123.0, 453.0): cluster 2\n", + "Example n. 5799 = (123.0, 452.0): cluster 2\n", + "Example n. 5800 = (123.0, 451.0): cluster 2\n", + "Example n. 5801 = (123.0, 450.0): cluster 2\n", + "Example n. 5802 = (123.0, 449.0): cluster 2\n", + "Example n. 5803 = (123.0, 448.0): cluster 2\n", + "Example n. 5804 = (123.0, 447.0): cluster 2\n", + "Example n. 5805 = (123.0, 446.0): cluster 2\n", + "Example n. 5806 = (123.0, 445.0): cluster 2\n", + "Example n. 5807 = (123.0, 444.0): cluster 2\n", + "Example n. 5808 = (123.0, 440.0): cluster 2\n", + "Example n. 5809 = (123.0, 439.0): cluster 2\n", + "Example n. 5810 = (123.0, 438.0): cluster 2\n", + "Example n. 5811 = (123.0, 437.0): cluster 2\n", + "Example n. 5812 = (123.0, 436.0): cluster 2\n", + "Example n. 5813 = (123.0, 435.0): cluster 2\n", + "Example n. 5814 = (123.0, 434.0): cluster 2\n", + "Example n. 5815 = (123.0, 433.0): cluster 2\n", + "Example n. 5816 = (123.0, 432.0): cluster 2\n", + "Example n. 5817 = (123.0, 431.0): cluster 2\n", + "Example n. 5818 = (123.0, 430.0): cluster 2\n", + "Example n. 5819 = (123.0, 429.0): cluster 2\n", + "Example n. 5820 = (123.0, 428.0): cluster 2\n", + "Example n. 5821 = (123.0, 427.0): cluster 2\n", + "Example n. 5822 = (123.0, 426.0): cluster 2\n", + "Example n. 5823 = (123.0, 425.0): cluster 2\n", + "Example n. 5824 = (123.0, 424.0): cluster 2\n", + "Example n. 5825 = (123.0, 423.0): cluster 2\n", + "Example n. 5826 = (123.0, 422.0): cluster 2\n", + "Example n. 5827 = (123.0, 421.0): cluster 2\n", + "Example n. 5828 = (124.0, 471.0): cluster 2\n", + "Example n. 5829 = (124.0, 470.0): cluster 2\n", + "Example n. 5830 = (124.0, 469.0): cluster 2\n", + "Example n. 5831 = (124.0, 468.0): cluster 2\n", + "Example n. 5832 = (124.0, 467.0): cluster 2\n", + "Example n. 5833 = (124.0, 466.0): cluster 2\n", + "Example n. 5834 = (124.0, 465.0): cluster 2\n", + "Example n. 5835 = (124.0, 464.0): cluster 2\n", + "Example n. 5836 = (124.0, 455.0): cluster 2\n", + "Example n. 5837 = (124.0, 454.0): cluster 2\n", + "Example n. 5838 = (124.0, 453.0): cluster 2\n", + "Example n. 5839 = (124.0, 452.0): cluster 2\n", + "Example n. 5840 = (124.0, 451.0): cluster 2\n", + "Example n. 5841 = (124.0, 450.0): cluster 2\n", + "Example n. 5842 = (124.0, 449.0): cluster 2\n", + "Example n. 5843 = (124.0, 448.0): cluster 2\n", + "Example n. 5844 = (124.0, 447.0): cluster 2\n", + "Example n. 5845 = (124.0, 446.0): cluster 2\n", + "Example n. 5846 = (124.0, 445.0): cluster 2\n", + "Example n. 5847 = (124.0, 444.0): cluster 2\n", + "Example n. 5848 = (124.0, 443.0): cluster 2\n", + "Example n. 5849 = (124.0, 442.0): cluster 2\n", + "Example n. 5850 = (124.0, 441.0): cluster 2\n", + "Example n. 5851 = (124.0, 440.0): cluster 2\n", + "Example n. 5852 = (124.0, 439.0): cluster 2\n", + "Example n. 5853 = (124.0, 438.0): cluster 2\n", + "Example n. 5854 = (124.0, 437.0): cluster 2\n", + "Example n. 5855 = (124.0, 436.0): cluster 2\n", + "Example n. 5856 = (124.0, 435.0): cluster 2\n", + "Example n. 5857 = (124.0, 434.0): cluster 2\n", + "Example n. 5858 = (124.0, 433.0): cluster 2\n", + "Example n. 5859 = (124.0, 432.0): cluster 2\n", + "Example n. 5860 = (124.0, 431.0): cluster 2\n", + "Example n. 5861 = (124.0, 430.0): cluster 2\n", + "Example n. 5862 = (124.0, 429.0): cluster 2\n", + "Example n. 5863 = (124.0, 428.0): cluster 2\n", + "Example n. 5864 = (124.0, 427.0): cluster 2\n", + "Example n. 5865 = (124.0, 426.0): cluster 2\n", + "Example n. 5866 = (124.0, 425.0): cluster 2\n", + "Example n. 5867 = (124.0, 424.0): cluster 2\n", + "Example n. 5868 = (124.0, 423.0): cluster 2\n", + "Example n. 5869 = (124.0, 422.0): cluster 2\n", + "Example n. 5870 = (124.0, 421.0): cluster 2\n", + "Example n. 5871 = (125.0, 455.0): cluster 2\n", + "Example n. 5872 = (125.0, 454.0): cluster 2\n", + "Example n. 5873 = (125.0, 453.0): cluster 2\n", + "Example n. 5874 = (125.0, 452.0): cluster 2\n", + "Example n. 5875 = (125.0, 451.0): cluster 2\n", + "Example n. 5876 = (125.0, 450.0): cluster 2\n", + "Example n. 5877 = (125.0, 449.0): cluster 2\n", + "Example n. 5878 = (125.0, 448.0): cluster 2\n", + "Example n. 5879 = (125.0, 447.0): cluster 2\n", + "Example n. 5880 = (125.0, 446.0): cluster 2\n", + "Example n. 5881 = (125.0, 445.0): cluster 2\n", + "Example n. 5882 = (125.0, 444.0): cluster 2\n", + "Example n. 5883 = (125.0, 443.0): cluster 2\n", + "Example n. 5884 = (125.0, 442.0): cluster 2\n", + "Example n. 5885 = (125.0, 441.0): cluster 2\n", + "Example n. 5886 = (125.0, 440.0): cluster 2\n", + "Example n. 5887 = (125.0, 439.0): cluster 2\n", + "Example n. 5888 = (125.0, 438.0): cluster 2\n", + "Example n. 5889 = (125.0, 437.0): cluster 2\n", + "Example n. 5890 = (125.0, 436.0): cluster 2\n", + "Example n. 5891 = (125.0, 435.0): cluster 2\n", + "Example n. 5892 = (125.0, 434.0): cluster 2\n", + "Example n. 5893 = (125.0, 433.0): cluster 2\n", + "Example n. 5894 = (125.0, 432.0): cluster 2\n", + "Example n. 5895 = (125.0, 431.0): cluster 2\n", + "Example n. 5896 = (125.0, 430.0): cluster 2\n", + "Example n. 5897 = (125.0, 429.0): cluster 2\n", + "Example n. 5898 = (125.0, 428.0): cluster 2\n", + "Example n. 5899 = (125.0, 427.0): cluster 2\n", + "Example n. 5900 = (125.0, 426.0): cluster 2\n", + "Example n. 5901 = (125.0, 425.0): cluster 2\n", + "Example n. 5902 = (125.0, 424.0): cluster 2\n", + "Example n. 5903 = (125.0, 423.0): cluster 2\n", + "Example n. 5904 = (125.0, 422.0): cluster 2\n", + "Example n. 5905 = (125.0, 421.0): cluster 2\n", + "Example n. 5906 = (126.0, 455.0): cluster 2\n", + "Example n. 5907 = (126.0, 454.0): cluster 2\n", + "Example n. 5908 = (126.0, 453.0): cluster 2\n", + "Example n. 5909 = (126.0, 452.0): cluster 2\n", + "Example n. 5910 = (126.0, 451.0): cluster 2\n", + "Example n. 5911 = (126.0, 450.0): cluster 2\n", + "Example n. 5912 = (126.0, 449.0): cluster 2\n", + "Example n. 5913 = (126.0, 448.0): cluster 2\n", + "Example n. 5914 = (126.0, 447.0): cluster 2\n", + "Example n. 5915 = (126.0, 446.0): cluster 2\n", + "Example n. 5916 = (126.0, 445.0): cluster 2\n", + "Example n. 5917 = (126.0, 444.0): cluster 2\n", + "Example n. 5918 = (126.0, 443.0): cluster 2\n", + "Example n. 5919 = (126.0, 442.0): cluster 2\n", + "Example n. 5920 = (126.0, 441.0): cluster 2\n", + "Example n. 5921 = (126.0, 440.0): cluster 2\n", + "Example n. 5922 = (126.0, 439.0): cluster 2\n", + "Example n. 5923 = (126.0, 438.0): cluster 2\n", + "Example n. 5924 = (126.0, 437.0): cluster 2\n", + "Example n. 5925 = (126.0, 436.0): cluster 2\n", + "Example n. 5926 = (126.0, 435.0): cluster 2\n", + "Example n. 5927 = (126.0, 434.0): cluster 2\n", + "Example n. 5928 = (126.0, 433.0): cluster 2\n", + "Example n. 5929 = (126.0, 432.0): cluster 2\n", + "Example n. 5930 = (126.0, 431.0): cluster 2\n", + "Example n. 5931 = (126.0, 430.0): cluster 2\n", + "Example n. 5932 = (126.0, 429.0): cluster 2\n", + "Example n. 5933 = (126.0, 428.0): cluster 2\n", + "Example n. 5934 = (126.0, 427.0): cluster 2\n", + "Example n. 5935 = (126.0, 426.0): cluster 2\n", + "Example n. 5936 = (126.0, 425.0): cluster 2\n", + "Example n. 5937 = (126.0, 424.0): cluster 2\n", + "Example n. 5938 = (126.0, 423.0): cluster 2\n", + "Example n. 5939 = (126.0, 422.0): cluster 2\n", + "Example n. 5940 = (126.0, 421.0): cluster 2\n", + "Example n. 5941 = (127.0, 455.0): cluster 2\n", + "Example n. 5942 = (127.0, 454.0): cluster 2\n", + "Example n. 5943 = (127.0, 453.0): cluster 2\n", + "Example n. 5944 = (127.0, 452.0): cluster 2\n", + "Example n. 5945 = (127.0, 451.0): cluster 2\n", + "Example n. 5946 = (127.0, 450.0): cluster 2\n", + "Example n. 5947 = (127.0, 449.0): cluster 2\n", + "Example n. 5948 = (127.0, 448.0): cluster 2\n", + "Example n. 5949 = (127.0, 447.0): cluster 2\n", + "Example n. 5950 = (127.0, 446.0): cluster 2\n", + "Example n. 5951 = (127.0, 445.0): cluster 2\n", + "Example n. 5952 = (127.0, 444.0): cluster 2\n", + "Example n. 5953 = (127.0, 443.0): cluster 2\n", + "Example n. 5954 = (127.0, 442.0): cluster 2\n", + "Example n. 5955 = (127.0, 441.0): cluster 2\n", + "Example n. 5956 = (127.0, 440.0): cluster 2\n", + "Example n. 5957 = (127.0, 439.0): cluster 2\n", + "Example n. 5958 = (127.0, 438.0): cluster 2\n", + "Example n. 5959 = (127.0, 437.0): cluster 2\n", + "Example n. 5960 = (127.0, 436.0): cluster 2\n", + "Example n. 5961 = (127.0, 435.0): cluster 2\n", + "Example n. 5962 = (127.0, 434.0): cluster 2\n", + "Example n. 5963 = (127.0, 433.0): cluster 2\n", + "Example n. 5964 = (127.0, 432.0): cluster 2\n", + "Example n. 5965 = (127.0, 431.0): cluster 2\n", + "Example n. 5966 = (127.0, 430.0): cluster 2\n", + "Example n. 5967 = (127.0, 429.0): cluster 2\n", + "Example n. 5968 = (127.0, 428.0): cluster 2\n", + "Example n. 5969 = (127.0, 427.0): cluster 2\n", + "Example n. 5970 = (127.0, 426.0): cluster 2\n", + "Example n. 5971 = (127.0, 425.0): cluster 2\n", + "Example n. 5972 = (127.0, 424.0): cluster 2\n", + "Example n. 5973 = (127.0, 423.0): cluster 2\n", + "Example n. 5974 = (127.0, 422.0): cluster 2\n", + "Example n. 5975 = (128.0, 455.0): cluster 2\n", + "Example n. 5976 = (128.0, 454.0): cluster 2\n", + "Example n. 5977 = (128.0, 453.0): cluster 2\n", + "Example n. 5978 = (128.0, 452.0): cluster 2\n", + "Example n. 5979 = (128.0, 451.0): cluster 2\n", + "Example n. 5980 = (128.0, 450.0): cluster 2\n", + "Example n. 5981 = (128.0, 449.0): cluster 2\n", + "Example n. 5982 = (128.0, 448.0): cluster 2\n", + "Example n. 5983 = (128.0, 447.0): cluster 2\n", + "Example n. 5984 = (128.0, 446.0): cluster 2\n", + "Example n. 5985 = (128.0, 445.0): cluster 2\n", + "Example n. 5986 = (128.0, 444.0): cluster 2\n", + "Example n. 5987 = (128.0, 443.0): cluster 2\n", + "Example n. 5988 = (128.0, 442.0): cluster 2\n", + "Example n. 5989 = (128.0, 441.0): cluster 2\n", + "Example n. 5990 = (128.0, 440.0): cluster 2\n", + "Example n. 5991 = (128.0, 439.0): cluster 2\n", + "Example n. 5992 = (128.0, 438.0): cluster 2\n", + "Example n. 5993 = (128.0, 437.0): cluster 2\n", + "Example n. 5994 = (128.0, 436.0): cluster 2\n", + "Example n. 5995 = (128.0, 435.0): cluster 2\n", + "Example n. 5996 = (128.0, 434.0): cluster 2\n", + "Example n. 5997 = (128.0, 433.0): cluster 2\n", + "Example n. 5998 = (128.0, 432.0): cluster 2\n", + "Example n. 5999 = (128.0, 431.0): cluster 2\n", + "Example n. 6000 = (128.0, 430.0): cluster 2\n", + "Example n. 6001 = (128.0, 429.0): cluster 2\n", + "Example n. 6002 = (128.0, 428.0): cluster 2\n", + "Example n. 6003 = (128.0, 427.0): cluster 2\n", + "Example n. 6004 = (128.0, 426.0): cluster 2\n", + "Example n. 6005 = (128.0, 425.0): cluster 2\n", + "Example n. 6006 = (128.0, 424.0): cluster 2\n", + "Example n. 6007 = (128.0, 423.0): cluster 2\n", + "Example n. 6008 = (128.0, 422.0): cluster 2\n", + "Example n. 6009 = (128.0, 421.0): cluster 2\n", + "Example n. 6010 = (129.0, 455.0): cluster 2\n", + "Example n. 6011 = (129.0, 454.0): cluster 2\n", + "Example n. 6012 = (129.0, 453.0): cluster 2\n", + "Example n. 6013 = (129.0, 452.0): cluster 2\n", + "Example n. 6014 = (129.0, 451.0): cluster 2\n", + "Example n. 6015 = (129.0, 450.0): cluster 2\n", + "Example n. 6016 = (129.0, 449.0): cluster 2\n", + "Example n. 6017 = (129.0, 448.0): cluster 2\n", + "Example n. 6018 = (129.0, 447.0): cluster 2\n", + "Example n. 6019 = (129.0, 446.0): cluster 2\n", + "Example n. 6020 = (129.0, 445.0): cluster 2\n", + "Example n. 6021 = (129.0, 444.0): cluster 2\n", + "Example n. 6022 = (129.0, 443.0): cluster 2\n", + "Example n. 6023 = (129.0, 442.0): cluster 2\n", + "Example n. 6024 = (129.0, 441.0): cluster 2\n", + "Example n. 6025 = (129.0, 440.0): cluster 2\n", + "Example n. 6026 = (129.0, 439.0): cluster 2\n", + "Example n. 6027 = (129.0, 438.0): cluster 2\n", + "Example n. 6028 = (129.0, 437.0): cluster 2\n", + "Example n. 6029 = (129.0, 436.0): cluster 2\n", + "Example n. 6030 = (129.0, 435.0): cluster 2\n", + "Example n. 6031 = (129.0, 434.0): cluster 2\n", + "Example n. 6032 = (129.0, 433.0): cluster 2\n", + "Example n. 6033 = (129.0, 432.0): cluster 2\n", + "Example n. 6034 = (129.0, 431.0): cluster 2\n", + "Example n. 6035 = (129.0, 430.0): cluster 2\n", + "Example n. 6036 = (129.0, 429.0): cluster 2\n", + "Example n. 6037 = (130.0, 455.0): cluster 2\n", + "Example n. 6038 = (130.0, 454.0): cluster 2\n", + "Example n. 6039 = (130.0, 453.0): cluster 2\n", + "Example n. 6040 = (130.0, 452.0): cluster 2\n", + "Example n. 6041 = (130.0, 451.0): cluster 2\n", + "Example n. 6042 = (130.0, 450.0): cluster 2\n", + "Example n. 6043 = (130.0, 449.0): cluster 2\n", + "Example n. 6044 = (130.0, 448.0): cluster 2\n", + "Example n. 6045 = (130.0, 447.0): cluster 2\n", + "Example n. 6046 = (130.0, 446.0): cluster 2\n", + "Example n. 6047 = (130.0, 445.0): cluster 2\n", + "Example n. 6048 = (130.0, 444.0): cluster 2\n", + "Example n. 6049 = (130.0, 443.0): cluster 2\n", + "Example n. 6050 = (130.0, 442.0): cluster 2\n", + "Example n. 6051 = (130.0, 441.0): cluster 2\n", + "Example n. 6052 = (130.0, 440.0): cluster 2\n", + "Example n. 6053 = (130.0, 439.0): cluster 2\n", + "Example n. 6054 = (130.0, 438.0): cluster 2\n", + "Example n. 6055 = (130.0, 437.0): cluster 2\n", + "Example n. 6056 = (130.0, 436.0): cluster 2\n", + "Example n. 6057 = (130.0, 435.0): cluster 2\n", + "Example n. 6058 = (130.0, 434.0): cluster 2\n", + "Example n. 6059 = (130.0, 433.0): cluster 2\n", + "Example n. 6060 = (130.0, 432.0): cluster 2\n", + "Example n. 6061 = (130.0, 431.0): cluster 2\n", + "Example n. 6062 = (130.0, 430.0): cluster 2\n", + "Example n. 6063 = (130.0, 429.0): cluster 2\n", + "Example n. 6064 = (131.0, 455.0): cluster 2\n", + "Example n. 6065 = (131.0, 454.0): cluster 2\n", + "Example n. 6066 = (131.0, 453.0): cluster 2\n", + "Example n. 6067 = (131.0, 452.0): cluster 2\n", + "Example n. 6068 = (131.0, 451.0): cluster 2\n", + "Example n. 6069 = (131.0, 450.0): cluster 2\n", + "Example n. 6070 = (131.0, 449.0): cluster 2\n", + "Example n. 6071 = (131.0, 448.0): cluster 2\n", + "Example n. 6072 = (131.0, 447.0): cluster 2\n", + "Example n. 6073 = (131.0, 446.0): cluster 2\n", + "Example n. 6074 = (131.0, 445.0): cluster 2\n", + "Example n. 6075 = (131.0, 444.0): cluster 2\n", + "Example n. 6076 = (131.0, 443.0): cluster 2\n", + "Example n. 6077 = (131.0, 442.0): cluster 2\n", + "Example n. 6078 = (131.0, 441.0): cluster 2\n", + "Example n. 6079 = (131.0, 440.0): cluster 2\n", + "Example n. 6080 = (131.0, 439.0): cluster 2\n", + "Example n. 6081 = (131.0, 438.0): cluster 2\n", + "Example n. 6082 = (131.0, 437.0): cluster 2\n", + "Example n. 6083 = (131.0, 436.0): cluster 2\n", + "Example n. 6084 = (131.0, 435.0): cluster 2\n", + "Example n. 6085 = (131.0, 434.0): cluster 2\n", + "Example n. 6086 = (131.0, 433.0): cluster 2\n", + "Example n. 6087 = (131.0, 432.0): cluster 2\n", + "Example n. 6088 = (131.0, 431.0): cluster 2\n", + "Example n. 6089 = (131.0, 430.0): cluster 2\n", + "Example n. 6090 = (131.0, 429.0): cluster 2\n", + "Example n. 6091 = (132.0, 455.0): cluster 2\n", + "Example n. 6092 = (132.0, 454.0): cluster 2\n", + "Example n. 6093 = (132.0, 453.0): cluster 2\n", + "Example n. 6094 = (132.0, 452.0): cluster 2\n", + "Example n. 6095 = (132.0, 451.0): cluster 2\n", + "Example n. 6096 = (132.0, 450.0): cluster 2\n", + "Example n. 6097 = (132.0, 449.0): cluster 2\n", + "Example n. 6098 = (132.0, 448.0): cluster 2\n", + "Example n. 6099 = (132.0, 447.0): cluster 2\n", + "Example n. 6100 = (132.0, 446.0): cluster 2\n", + "Example n. 6101 = (132.0, 445.0): cluster 2\n", + "Example n. 6102 = (132.0, 444.0): cluster 2\n", + "Example n. 6103 = (132.0, 443.0): cluster 2\n", + "Example n. 6104 = (132.0, 442.0): cluster 2\n", + "Example n. 6105 = (132.0, 441.0): cluster 2\n", + "Example n. 6106 = (132.0, 440.0): cluster 2\n", + "Example n. 6107 = (132.0, 439.0): cluster 2\n", + "Example n. 6108 = (132.0, 438.0): cluster 2\n", + "Example n. 6109 = (132.0, 437.0): cluster 2\n", + "Example n. 6110 = (132.0, 436.0): cluster 2\n", + "Example n. 6111 = (132.0, 435.0): cluster 2\n", + "Example n. 6112 = (132.0, 434.0): cluster 2\n", + "Example n. 6113 = (132.0, 433.0): cluster 2\n", + "Example n. 6114 = (132.0, 432.0): cluster 2\n", + "Example n. 6115 = (132.0, 431.0): cluster 2\n", + "Example n. 6116 = (132.0, 430.0): cluster 2\n", + "Example n. 6117 = (132.0, 429.0): cluster 2\n" + ] + } + ], + "source": [ + "data2,feature_names2,n_samples2,n_features2 = load_data(file_path, file_name2)\n", + "\n", + "np.random.seed(5)\n", + "\n", + "k=4\n", + "km2 = KMeans(n_clusters=k, random_state=0).fit(data2)\n", + "\n", + "for i in range(n_samples2):\n", + " print(f'Example n. {i} = ({(data2[i,0])}, {data2[i,1]}): cluster {km2.labels_[i]}')" + ] }, { "cell_type": "markdown", @@ -661,10 +6828,25 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 8, "metadata": {}, - "outputs": [], - "source": [] + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "plot(data2, n_samples2, km2, 'Clustered points in dataset n. 2')" + ] }, { "cell_type": "markdown", @@ -675,10 +6857,6174 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 9, "metadata": {}, - "outputs": [], - "source": [] + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Example n. 0 = (9.0, 480.0): cluster 0\n", + "Example n. 1 = (9.0, 479.0): cluster 0\n", + "Example n. 2 = (9.0, 478.0): cluster 0\n", + "Example n. 3 = (9.0, 477.0): cluster 0\n", + "Example n. 4 = (9.0, 467.0): cluster 0\n", + "Example n. 5 = (9.0, 466.0): cluster 0\n", + "Example n. 6 = (9.0, 465.0): cluster 0\n", + "Example n. 7 = (9.0, 464.0): cluster 0\n", + "Example n. 8 = (9.0, 392.0): cluster 0\n", + "Example n. 9 = (9.0, 391.0): cluster 0\n", + "Example n. 10 = (9.0, 390.0): cluster 0\n", + "Example n. 11 = (9.0, 350.0): cluster 0\n", + "Example n. 12 = (9.0, 349.0): cluster 0\n", + "Example n. 13 = (9.0, 348.0): cluster 0\n", + "Example n. 14 = (9.0, 347.0): cluster 0\n", + "Example n. 15 = (9.0, 290.0): cluster 0\n", + "Example n. 16 = (9.0, 289.0): cluster 0\n", + "Example n. 17 = (9.0, 288.0): cluster 0\n", + "Example n. 18 = (9.0, 287.0): cluster 0\n", + "Example n. 19 = (9.0, 240.0): cluster 0\n", + "Example n. 20 = (9.0, 239.0): cluster 0\n", + "Example n. 21 = (9.0, 238.0): cluster 0\n", + "Example n. 22 = (9.0, 231.0): cluster 0\n", + "Example n. 23 = (9.0, 230.0): cluster 0\n", + "Example n. 24 = (9.0, 229.0): cluster 0\n", + "Example n. 25 = (9.0, 228.0): cluster 0\n", + "Example n. 26 = (9.0, 129.0): cluster 0\n", + "Example n. 27 = (9.0, 128.0): cluster 0\n", + "Example n. 28 = (9.0, 127.0): cluster 0\n", + "Example n. 29 = (9.0, 126.0): cluster 0\n", + "Example n. 30 = (10.0, 480.0): cluster 0\n", + "Example n. 31 = (10.0, 479.0): cluster 0\n", + "Example n. 32 = (10.0, 478.0): cluster 0\n", + "Example n. 33 = (10.0, 477.0): cluster 0\n", + "Example n. 34 = (10.0, 476.0): cluster 0\n", + "Example n. 35 = (10.0, 468.0): cluster 0\n", + "Example n. 36 = (10.0, 467.0): cluster 0\n", + "Example n. 37 = (10.0, 466.0): cluster 0\n", + "Example n. 38 = (10.0, 465.0): cluster 0\n", + "Example n. 39 = (10.0, 464.0): cluster 0\n", + "Example n. 40 = (10.0, 463.0): cluster 0\n", + "Example n. 41 = (10.0, 393.0): cluster 0\n", + "Example n. 42 = (10.0, 392.0): cluster 0\n", + "Example n. 43 = (10.0, 391.0): cluster 0\n", + "Example n. 44 = (10.0, 390.0): cluster 0\n", + "Example n. 45 = (10.0, 389.0): cluster 0\n", + "Example n. 46 = (10.0, 351.0): cluster 0\n", + "Example n. 47 = (10.0, 350.0): cluster 0\n", + "Example n. 48 = (10.0, 349.0): cluster 0\n", + "Example n. 49 = (10.0, 348.0): cluster 0\n", + "Example n. 50 = (10.0, 347.0): cluster 0\n", + "Example n. 51 = (10.0, 346.0): cluster 0\n", + "Example n. 52 = (10.0, 291.0): cluster 0\n", + "Example n. 53 = (10.0, 290.0): cluster 0\n", + "Example n. 54 = (10.0, 289.0): cluster 0\n", + "Example n. 55 = (10.0, 288.0): cluster 0\n", + "Example n. 56 = (10.0, 287.0): cluster 0\n", + "Example n. 57 = (10.0, 286.0): cluster 0\n", + "Example n. 58 = (10.0, 242.0): cluster 0\n", + "Example n. 59 = (10.0, 241.0): cluster 0\n", + "Example n. 60 = (10.0, 240.0): cluster 0\n", + "Example n. 61 = (10.0, 239.0): cluster 0\n", + "Example n. 62 = (10.0, 238.0): cluster 0\n", + "Example n. 63 = (10.0, 237.0): cluster 0\n", + "Example n. 64 = (10.0, 236.0): cluster 0\n", + "Example n. 65 = (10.0, 232.0): cluster 0\n", + "Example n. 66 = (10.0, 231.0): cluster 0\n", + "Example n. 67 = (10.0, 230.0): cluster 0\n", + "Example n. 68 = (10.0, 229.0): cluster 0\n", + "Example n. 69 = (10.0, 228.0): cluster 0\n", + "Example n. 70 = (10.0, 227.0): cluster 0\n", + "Example n. 71 = (10.0, 226.0): cluster 0\n", + "Example n. 72 = (10.0, 225.0): cluster 0\n", + "Example n. 73 = (10.0, 129.0): cluster 0\n", + "Example n. 74 = (10.0, 128.0): cluster 0\n", + "Example n. 75 = (10.0, 127.0): cluster 0\n", + "Example n. 76 = (10.0, 126.0): cluster 0\n", + "Example n. 77 = (10.0, 125.0): cluster 0\n", + "Example n. 78 = (11.0, 481.0): cluster 0\n", + "Example n. 79 = (11.0, 480.0): cluster 0\n", + "Example n. 80 = (11.0, 479.0): cluster 0\n", + "Example n. 81 = (11.0, 478.0): cluster 0\n", + "Example n. 82 = (11.0, 477.0): cluster 0\n", + "Example n. 83 = (11.0, 476.0): cluster 0\n", + "Example n. 84 = (11.0, 468.0): cluster 0\n", + "Example n. 85 = (11.0, 467.0): cluster 0\n", + "Example n. 86 = (11.0, 466.0): cluster 0\n", + "Example n. 87 = (11.0, 465.0): cluster 0\n", + "Example n. 88 = (11.0, 464.0): cluster 0\n", + "Example n. 89 = (11.0, 463.0): cluster 0\n", + "Example n. 90 = (11.0, 452.0): cluster 0\n", + "Example n. 91 = (11.0, 451.0): cluster 0\n", + "Example n. 92 = (11.0, 408.0): cluster 0\n", + "Example n. 93 = (11.0, 407.0): cluster 0\n", + "Example n. 94 = (11.0, 406.0): cluster 0\n", + "Example n. 95 = (11.0, 393.0): cluster 0\n", + "Example n. 96 = (11.0, 392.0): cluster 0\n", + "Example n. 97 = (11.0, 391.0): cluster 0\n", + "Example n. 98 = (11.0, 390.0): cluster 0\n", + "Example n. 99 = (11.0, 389.0): cluster 0\n", + "Example n. 100 = (11.0, 388.0): cluster 0\n", + "Example n. 101 = (11.0, 351.0): cluster 0\n", + "Example n. 102 = (11.0, 350.0): cluster 0\n", + "Example n. 103 = (11.0, 349.0): cluster 0\n", + "Example n. 104 = (11.0, 348.0): cluster 0\n", + "Example n. 105 = (11.0, 347.0): cluster 0\n", + "Example n. 106 = (11.0, 346.0): cluster 0\n", + "Example n. 107 = (11.0, 291.0): cluster 0\n", + "Example n. 108 = (11.0, 290.0): cluster 0\n", + "Example n. 109 = (11.0, 289.0): cluster 0\n", + "Example n. 110 = (11.0, 288.0): cluster 0\n", + "Example n. 111 = (11.0, 287.0): cluster 0\n", + "Example n. 112 = (11.0, 286.0): cluster 0\n", + "Example n. 113 = (11.0, 242.0): cluster 0\n", + "Example n. 114 = (11.0, 241.0): cluster 0\n", + "Example n. 115 = (11.0, 240.0): cluster 0\n", + "Example n. 116 = (11.0, 239.0): cluster 0\n", + "Example n. 117 = (11.0, 238.0): cluster 0\n", + "Example n. 118 = (11.0, 237.0): cluster 0\n", + "Example n. 119 = (11.0, 236.0): cluster 0\n", + "Example n. 120 = (11.0, 235.0): cluster 0\n", + "Example n. 121 = (11.0, 232.0): cluster 0\n", + "Example n. 122 = (11.0, 231.0): cluster 0\n", + "Example n. 123 = (11.0, 230.0): cluster 0\n", + "Example n. 124 = (11.0, 229.0): cluster 0\n", + "Example n. 125 = (11.0, 228.0): cluster 0\n", + "Example n. 126 = (11.0, 227.0): cluster 0\n", + "Example n. 127 = (11.0, 226.0): cluster 0\n", + "Example n. 128 = (11.0, 225.0): cluster 0\n", + "Example n. 129 = (11.0, 224.0): cluster 0\n", + "Example n. 130 = (11.0, 130.0): cluster 0\n", + "Example n. 131 = (11.0, 129.0): cluster 0\n", + "Example n. 132 = (11.0, 128.0): cluster 0\n", + "Example n. 133 = (11.0, 127.0): cluster 0\n", + "Example n. 134 = (11.0, 126.0): cluster 0\n", + "Example n. 135 = (11.0, 125.0): cluster 0\n", + "Example n. 136 = (12.0, 480.0): cluster 0\n", + "Example n. 137 = (12.0, 479.0): cluster 0\n", + "Example n. 138 = (12.0, 478.0): cluster 0\n", + "Example n. 139 = (12.0, 477.0): cluster 0\n", + "Example n. 140 = (12.0, 476.0): cluster 0\n", + "Example n. 141 = (12.0, 468.0): cluster 0\n", + "Example n. 142 = (12.0, 467.0): cluster 0\n", + "Example n. 143 = (12.0, 466.0): cluster 0\n", + "Example n. 144 = (12.0, 465.0): cluster 0\n", + "Example n. 145 = (12.0, 464.0): cluster 0\n", + "Example n. 146 = (12.0, 463.0): cluster 0\n", + "Example n. 147 = (12.0, 453.0): cluster 0\n", + "Example n. 148 = (12.0, 452.0): cluster 0\n", + "Example n. 149 = (12.0, 451.0): cluster 0\n", + "Example n. 150 = (12.0, 450.0): cluster 0\n", + "Example n. 151 = (12.0, 409.0): cluster 0\n", + "Example n. 152 = (12.0, 408.0): cluster 0\n", + "Example n. 153 = (12.0, 407.0): cluster 0\n", + "Example n. 154 = (12.0, 406.0): cluster 0\n", + "Example n. 155 = (12.0, 405.0): cluster 0\n", + "Example n. 156 = (12.0, 393.0): cluster 0\n", + "Example n. 157 = (12.0, 392.0): cluster 0\n", + "Example n. 158 = (12.0, 391.0): cluster 0\n", + "Example n. 159 = (12.0, 390.0): cluster 0\n", + "Example n. 160 = (12.0, 389.0): cluster 0\n", + "Example n. 161 = (12.0, 388.0): cluster 0\n", + "Example n. 162 = (12.0, 351.0): cluster 0\n", + "Example n. 163 = (12.0, 350.0): cluster 0\n", + "Example n. 164 = (12.0, 349.0): cluster 0\n", + "Example n. 165 = (12.0, 348.0): cluster 0\n", + "Example n. 166 = (12.0, 347.0): cluster 0\n", + "Example n. 167 = (12.0, 346.0): cluster 0\n", + "Example n. 168 = (12.0, 291.0): cluster 0\n", + "Example n. 169 = (12.0, 290.0): cluster 0\n", + "Example n. 170 = (12.0, 289.0): cluster 0\n", + "Example n. 171 = (12.0, 288.0): cluster 0\n", + "Example n. 172 = (12.0, 287.0): cluster 0\n", + "Example n. 173 = (12.0, 256.0): cluster 0\n", + "Example n. 174 = (12.0, 255.0): cluster 0\n", + "Example n. 175 = (12.0, 254.0): cluster 0\n", + "Example n. 176 = (12.0, 252.0): cluster 0\n", + "Example n. 177 = (12.0, 251.0): cluster 0\n", + "Example n. 178 = (12.0, 250.0): cluster 0\n", + "Example n. 179 = (12.0, 249.0): cluster 0\n", + "Example n. 180 = (12.0, 242.0): cluster 0\n", + "Example n. 181 = (12.0, 241.0): cluster 0\n", + "Example n. 182 = (12.0, 240.0): cluster 0\n", + "Example n. 183 = (12.0, 239.0): cluster 0\n", + "Example n. 184 = (12.0, 238.0): cluster 0\n", + "Example n. 185 = (12.0, 237.0): cluster 0\n", + "Example n. 186 = (12.0, 236.0): cluster 0\n", + "Example n. 187 = (12.0, 235.0): cluster 0\n", + "Example n. 188 = (12.0, 232.0): cluster 0\n", + "Example n. 189 = (12.0, 231.0): cluster 0\n", + "Example n. 190 = (12.0, 230.0): cluster 0\n", + "Example n. 191 = (12.0, 229.0): cluster 0\n", + "Example n. 192 = (12.0, 228.0): cluster 0\n", + "Example n. 193 = (12.0, 227.0): cluster 0\n", + "Example n. 194 = (12.0, 226.0): cluster 0\n", + "Example n. 195 = (12.0, 225.0): cluster 0\n", + "Example n. 196 = (12.0, 224.0): cluster 0\n", + "Example n. 197 = (12.0, 223.0): cluster 0\n", + "Example n. 198 = (12.0, 206.0): cluster 0\n", + "Example n. 199 = (12.0, 205.0): cluster 0\n", + "Example n. 200 = (12.0, 204.0): cluster 0\n", + "Example n. 201 = (12.0, 200.0): cluster 0\n", + "Example n. 202 = (12.0, 129.0): cluster 0\n", + "Example n. 203 = (12.0, 128.0): cluster 0\n", + "Example n. 204 = (12.0, 127.0): cluster 0\n", + "Example n. 205 = (12.0, 126.0): cluster 0\n", + "Example n. 206 = (12.0, 125.0): cluster 0\n", + "Example n. 207 = (13.0, 480.0): cluster 0\n", + "Example n. 208 = (13.0, 479.0): cluster 0\n", + "Example n. 209 = (13.0, 478.0): cluster 0\n", + "Example n. 210 = (13.0, 477.0): cluster 0\n", + "Example n. 211 = (13.0, 467.0): cluster 0\n", + "Example n. 212 = (13.0, 466.0): cluster 0\n", + "Example n. 213 = (13.0, 465.0): cluster 0\n", + "Example n. 214 = (13.0, 464.0): cluster 0\n", + "Example n. 215 = (13.0, 454.0): cluster 0\n", + "Example n. 216 = (13.0, 453.0): cluster 0\n", + "Example n. 217 = (13.0, 452.0): cluster 0\n", + "Example n. 218 = (13.0, 451.0): cluster 0\n", + "Example n. 219 = (13.0, 450.0): cluster 0\n", + "Example n. 220 = (13.0, 449.0): cluster 0\n", + "Example n. 221 = (13.0, 409.0): cluster 0\n", + "Example n. 222 = (13.0, 408.0): cluster 0\n", + "Example n. 223 = (13.0, 407.0): cluster 0\n", + "Example n. 224 = (13.0, 406.0): cluster 0\n", + "Example n. 225 = (13.0, 405.0): cluster 0\n", + "Example n. 226 = (13.0, 393.0): cluster 0\n", + "Example n. 227 = (13.0, 392.0): cluster 0\n", + "Example n. 228 = (13.0, 391.0): cluster 0\n", + "Example n. 229 = (13.0, 390.0): cluster 0\n", + "Example n. 230 = (13.0, 389.0): cluster 0\n", + "Example n. 231 = (13.0, 350.0): cluster 0\n", + "Example n. 232 = (13.0, 349.0): cluster 0\n", + "Example n. 233 = (13.0, 348.0): cluster 0\n", + "Example n. 234 = (13.0, 347.0): cluster 0\n", + "Example n. 235 = (13.0, 290.0): cluster 0\n", + "Example n. 236 = (13.0, 289.0): cluster 0\n", + "Example n. 237 = (13.0, 288.0): cluster 0\n", + "Example n. 238 = (13.0, 287.0): cluster 0\n", + "Example n. 239 = (13.0, 266.0): cluster 0\n", + "Example n. 240 = (13.0, 265.0): cluster 0\n", + "Example n. 241 = (13.0, 264.0): cluster 0\n", + "Example n. 242 = (13.0, 263.0): cluster 0\n", + "Example n. 243 = (13.0, 257.0): cluster 0\n", + "Example n. 244 = (13.0, 256.0): cluster 0\n", + "Example n. 245 = (13.0, 255.0): cluster 0\n", + "Example n. 246 = (13.0, 254.0): cluster 0\n", + "Example n. 247 = (13.0, 253.0): cluster 0\n", + "Example n. 248 = (13.0, 252.0): cluster 0\n", + "Example n. 249 = (13.0, 251.0): cluster 0\n", + "Example n. 250 = (13.0, 250.0): cluster 0\n", + "Example n. 251 = (13.0, 249.0): cluster 0\n", + "Example n. 252 = (13.0, 248.0): cluster 0\n", + "Example n. 253 = (13.0, 241.0): cluster 0\n", + "Example n. 254 = (13.0, 240.0): cluster 0\n", + "Example n. 255 = (13.0, 239.0): cluster 0\n", + "Example n. 256 = (13.0, 238.0): cluster 0\n", + "Example n. 257 = (13.0, 237.0): cluster 0\n", + "Example n. 258 = (13.0, 236.0): cluster 0\n", + "Example n. 259 = (13.0, 235.0): cluster 0\n", + "Example n. 260 = (13.0, 231.0): cluster 0\n", + "Example n. 261 = (13.0, 230.0): cluster 0\n", + "Example n. 262 = (13.0, 229.0): cluster 0\n", + "Example n. 263 = (13.0, 228.0): cluster 0\n", + "Example n. 264 = (13.0, 227.0): cluster 0\n", + "Example n. 265 = (13.0, 226.0): cluster 0\n", + "Example n. 266 = (13.0, 225.0): cluster 0\n", + "Example n. 267 = (13.0, 224.0): cluster 0\n", + "Example n. 268 = (13.0, 223.0): cluster 0\n", + "Example n. 269 = (13.0, 222.0): cluster 0\n", + "Example n. 270 = (13.0, 207.0): cluster 0\n", + "Example n. 271 = (13.0, 206.0): cluster 0\n", + "Example n. 272 = (13.0, 205.0): cluster 0\n", + "Example n. 273 = (13.0, 204.0): cluster 0\n", + "Example n. 274 = (13.0, 203.0): cluster 0\n", + "Example n. 275 = (13.0, 202.0): cluster 0\n", + "Example n. 276 = (13.0, 201.0): cluster 0\n", + "Example n. 277 = (13.0, 200.0): cluster 0\n", + "Example n. 278 = (13.0, 199.0): cluster 0\n", + "Example n. 279 = (13.0, 198.0): cluster 0\n", + "Example n. 280 = (13.0, 129.0): cluster 0\n", + "Example n. 281 = (13.0, 128.0): cluster 0\n", + "Example n. 282 = (13.0, 127.0): cluster 0\n", + "Example n. 283 = (13.0, 126.0): cluster 0\n", + "Example n. 284 = (13.0, 121.0): cluster 0\n", + "Example n. 285 = (13.0, 120.0): cluster 0\n", + "Example n. 286 = (13.0, 119.0): cluster 0\n", + "Example n. 287 = (13.0, 118.0): cluster 0\n", + "Example n. 288 = (14.0, 454.0): cluster 0\n", + "Example n. 289 = (14.0, 453.0): cluster 0\n", + "Example n. 290 = (14.0, 452.0): cluster 0\n", + "Example n. 291 = (14.0, 451.0): cluster 0\n", + "Example n. 292 = (14.0, 450.0): cluster 0\n", + "Example n. 293 = (14.0, 449.0): cluster 0\n", + "Example n. 294 = (14.0, 409.0): cluster 0\n", + "Example n. 295 = (14.0, 408.0): cluster 0\n", + "Example n. 296 = (14.0, 407.0): cluster 0\n", + "Example n. 297 = (14.0, 406.0): cluster 0\n", + "Example n. 298 = (14.0, 405.0): cluster 0\n", + "Example n. 299 = (14.0, 392.0): cluster 0\n", + "Example n. 300 = (14.0, 391.0): cluster 0\n", + "Example n. 301 = (14.0, 390.0): cluster 0\n", + "Example n. 302 = (14.0, 267.0): cluster 0\n", + "Example n. 303 = (14.0, 266.0): cluster 0\n", + "Example n. 304 = (14.0, 265.0): cluster 0\n", + "Example n. 305 = (14.0, 264.0): cluster 0\n", + "Example n. 306 = (14.0, 263.0): cluster 0\n", + "Example n. 307 = (14.0, 262.0): cluster 0\n", + "Example n. 308 = (14.0, 257.0): cluster 0\n", + "Example n. 309 = (14.0, 256.0): cluster 0\n", + "Example n. 310 = (14.0, 255.0): cluster 0\n", + "Example n. 311 = (14.0, 254.0): cluster 0\n", + "Example n. 312 = (14.0, 253.0): cluster 0\n", + "Example n. 313 = (14.0, 252.0): cluster 0\n", + "Example n. 314 = (14.0, 251.0): cluster 0\n", + "Example n. 315 = (14.0, 250.0): cluster 0\n", + "Example n. 316 = (14.0, 249.0): cluster 0\n", + "Example n. 317 = (14.0, 248.0): cluster 0\n", + "Example n. 318 = (14.0, 244.0): cluster 0\n", + "Example n. 319 = (14.0, 243.0): cluster 0\n", + "Example n. 320 = (14.0, 242.0): cluster 0\n", + "Example n. 321 = (14.0, 241.0): cluster 0\n", + "Example n. 322 = (14.0, 240.0): cluster 0\n", + "Example n. 323 = (14.0, 239.0): cluster 0\n", + "Example n. 324 = (14.0, 238.0): cluster 0\n", + "Example n. 325 = (14.0, 237.0): cluster 0\n", + "Example n. 326 = (14.0, 236.0): cluster 0\n", + "Example n. 327 = (14.0, 228.0): cluster 0\n", + "Example n. 328 = (14.0, 227.0): cluster 0\n", + "Example n. 329 = (14.0, 226.0): cluster 0\n", + "Example n. 330 = (14.0, 225.0): cluster 0\n", + "Example n. 331 = (14.0, 224.0): cluster 0\n", + "Example n. 332 = (14.0, 223.0): cluster 0\n", + "Example n. 333 = (14.0, 222.0): cluster 0\n", + "Example n. 334 = (14.0, 219.0): cluster 0\n", + "Example n. 335 = (14.0, 218.0): cluster 0\n", + "Example n. 336 = (14.0, 207.0): cluster 0\n", + "Example n. 337 = (14.0, 206.0): cluster 0\n", + "Example n. 338 = (14.0, 205.0): cluster 0\n", + "Example n. 339 = (14.0, 204.0): cluster 0\n", + "Example n. 340 = (14.0, 203.0): cluster 0\n", + "Example n. 341 = (14.0, 202.0): cluster 0\n", + "Example n. 342 = (14.0, 201.0): cluster 0\n", + "Example n. 343 = (14.0, 200.0): cluster 0\n", + "Example n. 344 = (14.0, 199.0): cluster 0\n", + "Example n. 345 = (14.0, 198.0): cluster 0\n", + "Example n. 346 = (14.0, 195.0): cluster 0\n", + "Example n. 347 = (14.0, 194.0): cluster 0\n", + "Example n. 348 = (14.0, 127.0): cluster 0\n", + "Example n. 349 = (14.0, 122.0): cluster 0\n", + "Example n. 350 = (14.0, 121.0): cluster 0\n", + "Example n. 351 = (14.0, 120.0): cluster 0\n", + "Example n. 352 = (14.0, 119.0): cluster 0\n", + "Example n. 353 = (14.0, 118.0): cluster 0\n", + "Example n. 354 = (14.0, 117.0): cluster 0\n", + "Example n. 355 = (15.0, 453.0): cluster 0\n", + "Example n. 356 = (15.0, 452.0): cluster 0\n", + "Example n. 357 = (15.0, 451.0): cluster 0\n", + "Example n. 358 = (15.0, 450.0): cluster 0\n", + "Example n. 359 = (15.0, 449.0): cluster 0\n", + "Example n. 360 = (15.0, 409.0): cluster 0\n", + "Example n. 361 = (15.0, 408.0): cluster 0\n", + "Example n. 362 = (15.0, 407.0): cluster 0\n", + "Example n. 363 = (15.0, 406.0): cluster 0\n", + "Example n. 364 = (15.0, 405.0): cluster 0\n", + "Example n. 365 = (15.0, 310.0): cluster 0\n", + "Example n. 366 = (15.0, 309.0): cluster 0\n", + "Example n. 367 = (15.0, 276.0): cluster 0\n", + "Example n. 368 = (15.0, 275.0): cluster 0\n", + "Example n. 369 = (15.0, 268.0): cluster 0\n", + "Example n. 370 = (15.0, 267.0): cluster 0\n", + "Example n. 371 = (15.0, 266.0): cluster 0\n", + "Example n. 372 = (15.0, 265.0): cluster 0\n", + "Example n. 373 = (15.0, 264.0): cluster 0\n", + "Example n. 374 = (15.0, 263.0): cluster 0\n", + "Example n. 375 = (15.0, 262.0): cluster 0\n", + "Example n. 376 = (15.0, 261.0): cluster 0\n", + "Example n. 377 = (15.0, 258.0): cluster 0\n", + "Example n. 378 = (15.0, 257.0): cluster 0\n", + "Example n. 379 = (15.0, 256.0): cluster 0\n", + "Example n. 380 = (15.0, 255.0): cluster 0\n", + "Example n. 381 = (15.0, 254.0): cluster 0\n", + "Example n. 382 = (15.0, 253.0): cluster 0\n", + "Example n. 383 = (15.0, 252.0): cluster 0\n", + "Example n. 384 = (15.0, 251.0): cluster 0\n", + "Example n. 385 = (15.0, 250.0): cluster 0\n", + "Example n. 386 = (15.0, 249.0): cluster 0\n", + "Example n. 387 = (15.0, 248.0): cluster 0\n", + "Example n. 388 = (15.0, 245.0): cluster 0\n", + "Example n. 389 = (15.0, 244.0): cluster 0\n", + "Example n. 390 = (15.0, 243.0): cluster 0\n", + "Example n. 391 = (15.0, 242.0): cluster 0\n", + "Example n. 392 = (15.0, 241.0): cluster 0\n", + "Example n. 393 = (15.0, 238.0): cluster 0\n", + "Example n. 394 = (15.0, 237.0): cluster 0\n", + "Example n. 395 = (15.0, 227.0): cluster 0\n", + "Example n. 396 = (15.0, 226.0): cluster 0\n", + "Example n. 397 = (15.0, 225.0): cluster 0\n", + "Example n. 398 = (15.0, 224.0): cluster 0\n", + "Example n. 399 = (15.0, 223.0): cluster 0\n", + "Example n. 400 = (15.0, 222.0): cluster 0\n", + "Example n. 401 = (15.0, 221.0): cluster 0\n", + "Example n. 402 = (15.0, 220.0): cluster 0\n", + "Example n. 403 = (15.0, 219.0): cluster 0\n", + "Example n. 404 = (15.0, 218.0): cluster 0\n", + "Example n. 405 = (15.0, 217.0): cluster 0\n", + "Example n. 406 = (15.0, 216.0): cluster 0\n", + "Example n. 407 = (15.0, 207.0): cluster 0\n", + "Example n. 408 = (15.0, 206.0): cluster 0\n", + "Example n. 409 = (15.0, 205.0): cluster 0\n", + "Example n. 410 = (15.0, 204.0): cluster 0\n", + "Example n. 411 = (15.0, 203.0): cluster 0\n", + "Example n. 412 = (15.0, 202.0): cluster 0\n", + "Example n. 413 = (15.0, 201.0): cluster 0\n", + "Example n. 414 = (15.0, 200.0): cluster 0\n", + "Example n. 415 = (15.0, 199.0): cluster 0\n", + "Example n. 416 = (15.0, 198.0): cluster 0\n", + "Example n. 417 = (15.0, 197.0): cluster 0\n", + "Example n. 418 = (15.0, 196.0): cluster 0\n", + "Example n. 419 = (15.0, 195.0): cluster 0\n", + "Example n. 420 = (15.0, 194.0): cluster 0\n", + "Example n. 421 = (15.0, 193.0): cluster 0\n", + "Example n. 422 = (15.0, 192.0): cluster 0\n", + "Example n. 423 = (15.0, 122.0): cluster 0\n", + "Example n. 424 = (15.0, 121.0): cluster 0\n", + "Example n. 425 = (15.0, 120.0): cluster 0\n", + "Example n. 426 = (15.0, 119.0): cluster 0\n", + "Example n. 427 = (15.0, 118.0): cluster 0\n", + "Example n. 428 = (15.0, 117.0): cluster 0\n", + "Example n. 429 = (16.0, 453.0): cluster 0\n", + "Example n. 430 = (16.0, 452.0): cluster 0\n", + "Example n. 431 = (16.0, 451.0): cluster 0\n", + "Example n. 432 = (16.0, 450.0): cluster 0\n", + "Example n. 433 = (16.0, 407.0): cluster 0\n", + "Example n. 434 = (16.0, 311.0): cluster 0\n", + "Example n. 435 = (16.0, 310.0): cluster 0\n", + "Example n. 436 = (16.0, 309.0): cluster 0\n", + "Example n. 437 = (16.0, 308.0): cluster 0\n", + "Example n. 438 = (16.0, 278.0): cluster 0\n", + "Example n. 439 = (16.0, 277.0): cluster 0\n", + "Example n. 440 = (16.0, 276.0): cluster 0\n", + "Example n. 441 = (16.0, 275.0): cluster 0\n", + "Example n. 442 = (16.0, 274.0): cluster 0\n", + "Example n. 443 = (16.0, 268.0): cluster 0\n", + "Example n. 444 = (16.0, 267.0): cluster 0\n", + "Example n. 445 = (16.0, 266.0): cluster 0\n", + "Example n. 446 = (16.0, 265.0): cluster 0\n", + "Example n. 447 = (16.0, 264.0): cluster 0\n", + "Example n. 448 = (16.0, 263.0): cluster 0\n", + "Example n. 449 = (16.0, 262.0): cluster 0\n", + "Example n. 450 = (16.0, 261.0): cluster 0\n", + "Example n. 451 = (16.0, 258.0): cluster 0\n", + "Example n. 452 = (16.0, 257.0): cluster 0\n", + "Example n. 453 = (16.0, 256.0): cluster 0\n", + "Example n. 454 = (16.0, 255.0): cluster 0\n", + "Example n. 455 = (16.0, 254.0): cluster 0\n", + "Example n. 456 = (16.0, 253.0): cluster 0\n", + "Example n. 457 = (16.0, 252.0): cluster 0\n", + "Example n. 458 = (16.0, 251.0): cluster 0\n", + "Example n. 459 = (16.0, 250.0): cluster 0\n", + "Example n. 460 = (16.0, 249.0): cluster 0\n", + "Example n. 461 = (16.0, 246.0): cluster 0\n", + "Example n. 462 = (16.0, 245.0): cluster 0\n", + "Example n. 463 = (16.0, 244.0): cluster 0\n", + "Example n. 464 = (16.0, 243.0): cluster 0\n", + "Example n. 465 = (16.0, 242.0): cluster 0\n", + "Example n. 466 = (16.0, 241.0): cluster 0\n", + "Example n. 467 = (16.0, 228.0): cluster 0\n", + "Example n. 468 = (16.0, 227.0): cluster 0\n", + "Example n. 469 = (16.0, 226.0): cluster 0\n", + "Example n. 470 = (16.0, 225.0): cluster 0\n", + "Example n. 471 = (16.0, 224.0): cluster 0\n", + "Example n. 472 = (16.0, 223.0): cluster 0\n", + "Example n. 473 = (16.0, 222.0): cluster 0\n", + "Example n. 474 = (16.0, 221.0): cluster 0\n", + "Example n. 475 = (16.0, 220.0): cluster 0\n", + "Example n. 476 = (16.0, 219.0): cluster 0\n", + "Example n. 477 = (16.0, 218.0): cluster 0\n", + "Example n. 478 = (16.0, 217.0): cluster 0\n", + "Example n. 479 = (16.0, 216.0): cluster 0\n", + "Example n. 480 = (16.0, 207.0): cluster 0\n", + "Example n. 481 = (16.0, 206.0): cluster 0\n", + "Example n. 482 = (16.0, 205.0): cluster 0\n", + "Example n. 483 = (16.0, 204.0): cluster 0\n", + "Example n. 484 = (16.0, 203.0): cluster 0\n", + "Example n. 485 = (16.0, 202.0): cluster 0\n", + "Example n. 486 = (16.0, 201.0): cluster 0\n", + "Example n. 487 = (16.0, 200.0): cluster 0\n", + "Example n. 488 = (16.0, 199.0): cluster 0\n", + "Example n. 489 = (16.0, 198.0): cluster 0\n", + "Example n. 490 = (16.0, 197.0): cluster 0\n", + "Example n. 491 = (16.0, 196.0): cluster 0\n", + "Example n. 492 = (16.0, 195.0): cluster 0\n", + "Example n. 493 = (16.0, 194.0): cluster 0\n", + "Example n. 494 = (16.0, 193.0): cluster 0\n", + "Example n. 495 = (16.0, 192.0): cluster 0\n", + "Example n. 496 = (16.0, 122.0): cluster 0\n", + "Example n. 497 = (16.0, 121.0): cluster 0\n", + "Example n. 498 = (16.0, 120.0): cluster 0\n", + "Example n. 499 = (16.0, 119.0): cluster 0\n", + "Example n. 500 = (16.0, 118.0): cluster 0\n", + "Example n. 501 = (16.0, 117.0): cluster 0\n", + "Example n. 502 = (17.0, 500.0): cluster 0\n", + "Example n. 503 = (17.0, 499.0): cluster 0\n", + "Example n. 504 = (17.0, 498.0): cluster 0\n", + "Example n. 505 = (17.0, 312.0): cluster 0\n", + "Example n. 506 = (17.0, 311.0): cluster 0\n", + "Example n. 507 = (17.0, 310.0): cluster 0\n", + "Example n. 508 = (17.0, 309.0): cluster 0\n", + "Example n. 509 = (17.0, 308.0): cluster 0\n", + "Example n. 510 = (17.0, 307.0): cluster 0\n", + "Example n. 511 = (17.0, 279.0): cluster 0\n", + "Example n. 512 = (17.0, 278.0): cluster 0\n", + "Example n. 513 = (17.0, 277.0): cluster 0\n", + "Example n. 514 = (17.0, 276.0): cluster 0\n", + "Example n. 515 = (17.0, 275.0): cluster 0\n", + "Example n. 516 = (17.0, 274.0): cluster 0\n", + "Example n. 517 = (17.0, 273.0): cluster 0\n", + "Example n. 518 = (17.0, 270.0): cluster 0\n", + "Example n. 519 = (17.0, 269.0): cluster 0\n", + "Example n. 520 = (17.0, 268.0): cluster 0\n", + "Example n. 521 = (17.0, 267.0): cluster 0\n", + "Example n. 522 = (17.0, 266.0): cluster 0\n", + "Example n. 523 = (17.0, 265.0): cluster 0\n", + "Example n. 524 = (17.0, 264.0): cluster 0\n", + "Example n. 525 = (17.0, 263.0): cluster 0\n", + "Example n. 526 = (17.0, 262.0): cluster 0\n", + "Example n. 527 = (17.0, 261.0): cluster 0\n", + "Example n. 528 = (17.0, 260.0): cluster 0\n", + "Example n. 529 = (17.0, 259.0): cluster 0\n", + "Example n. 530 = (17.0, 258.0): cluster 0\n", + "Example n. 531 = (17.0, 257.0): cluster 0\n", + "Example n. 532 = (17.0, 256.0): cluster 0\n", + "Example n. 533 = (17.0, 255.0): cluster 0\n", + "Example n. 534 = (17.0, 254.0): cluster 0\n", + "Example n. 535 = (17.0, 246.0): cluster 0\n", + "Example n. 536 = (17.0, 245.0): cluster 0\n", + "Example n. 537 = (17.0, 244.0): cluster 0\n", + "Example n. 538 = (17.0, 243.0): cluster 0\n", + "Example n. 539 = (17.0, 242.0): cluster 0\n", + "Example n. 540 = (17.0, 241.0): cluster 0\n", + "Example n. 541 = (17.0, 228.0): cluster 0\n", + "Example n. 542 = (17.0, 227.0): cluster 0\n", + "Example n. 543 = (17.0, 226.0): cluster 0\n", + "Example n. 544 = (17.0, 225.0): cluster 0\n", + "Example n. 545 = (17.0, 224.0): cluster 0\n", + "Example n. 546 = (17.0, 223.0): cluster 0\n", + "Example n. 547 = (17.0, 222.0): cluster 0\n", + "Example n. 548 = (17.0, 221.0): cluster 0\n", + "Example n. 549 = (17.0, 220.0): cluster 0\n", + "Example n. 550 = (17.0, 219.0): cluster 0\n", + "Example n. 551 = (17.0, 218.0): cluster 0\n", + "Example n. 552 = (17.0, 217.0): cluster 0\n", + "Example n. 553 = (17.0, 216.0): cluster 0\n", + "Example n. 554 = (17.0, 207.0): cluster 0\n", + "Example n. 555 = (17.0, 206.0): cluster 0\n", + "Example n. 556 = (17.0, 205.0): cluster 0\n", + "Example n. 557 = (17.0, 204.0): cluster 0\n", + "Example n. 558 = (17.0, 203.0): cluster 0\n", + "Example n. 559 = (17.0, 202.0): cluster 0\n", + "Example n. 560 = (17.0, 201.0): cluster 0\n", + "Example n. 561 = (17.0, 200.0): cluster 0\n", + "Example n. 562 = (17.0, 199.0): cluster 0\n", + "Example n. 563 = (17.0, 198.0): cluster 0\n", + "Example n. 564 = (17.0, 197.0): cluster 0\n", + "Example n. 565 = (17.0, 196.0): cluster 0\n", + "Example n. 566 = (17.0, 195.0): cluster 0\n", + "Example n. 567 = (17.0, 194.0): cluster 0\n", + "Example n. 568 = (17.0, 193.0): cluster 0\n", + "Example n. 569 = (17.0, 192.0): cluster 0\n", + "Example n. 570 = (17.0, 148.0): cluster 0\n", + "Example n. 571 = (17.0, 147.0): cluster 0\n", + "Example n. 572 = (17.0, 142.0): cluster 0\n", + "Example n. 573 = (17.0, 132.0): cluster 0\n", + "Example n. 574 = (17.0, 131.0): cluster 0\n", + "Example n. 575 = (17.0, 130.0): cluster 0\n", + "Example n. 576 = (17.0, 121.0): cluster 0\n", + "Example n. 577 = (17.0, 120.0): cluster 0\n", + "Example n. 578 = (17.0, 119.0): cluster 0\n", + "Example n. 579 = (17.0, 118.0): cluster 0\n", + "Example n. 580 = (17.0, 92.0): cluster 0\n", + "Example n. 581 = (18.0, 501.0): cluster 0\n", + "Example n. 582 = (18.0, 500.0): cluster 0\n", + "Example n. 583 = (18.0, 499.0): cluster 0\n", + "Example n. 584 = (18.0, 498.0): cluster 0\n", + "Example n. 585 = (18.0, 497.0): cluster 0\n", + "Example n. 586 = (18.0, 312.0): cluster 0\n", + "Example n. 587 = (18.0, 311.0): cluster 0\n", + "Example n. 588 = (18.0, 310.0): cluster 0\n", + "Example n. 589 = (18.0, 309.0): cluster 0\n", + "Example n. 590 = (18.0, 308.0): cluster 0\n", + "Example n. 591 = (18.0, 307.0): cluster 0\n", + "Example n. 592 = (18.0, 281.0): cluster 0\n", + "Example n. 593 = (18.0, 280.0): cluster 0\n", + "Example n. 594 = (18.0, 279.0): cluster 0\n", + "Example n. 595 = (18.0, 278.0): cluster 0\n", + "Example n. 596 = (18.0, 277.0): cluster 0\n", + "Example n. 597 = (18.0, 276.0): cluster 0\n", + "Example n. 598 = (18.0, 275.0): cluster 0\n", + "Example n. 599 = (18.0, 274.0): cluster 0\n", + "Example n. 600 = (18.0, 273.0): cluster 0\n", + "Example n. 601 = (18.0, 271.0): cluster 0\n", + "Example n. 602 = (18.0, 270.0): cluster 0\n", + "Example n. 603 = (18.0, 269.0): cluster 0\n", + "Example n. 604 = (18.0, 268.0): cluster 0\n", + "Example n. 605 = (18.0, 267.0): cluster 0\n", + "Example n. 606 = (18.0, 266.0): cluster 0\n", + "Example n. 607 = (18.0, 265.0): cluster 0\n", + "Example n. 608 = (18.0, 264.0): cluster 0\n", + "Example n. 609 = (18.0, 263.0): cluster 0\n", + "Example n. 610 = (18.0, 262.0): cluster 0\n", + "Example n. 611 = (18.0, 261.0): cluster 0\n", + "Example n. 612 = (18.0, 260.0): cluster 0\n", + "Example n. 613 = (18.0, 259.0): cluster 0\n", + "Example n. 614 = (18.0, 258.0): cluster 0\n", + "Example n. 615 = (18.0, 257.0): cluster 0\n", + "Example n. 616 = (18.0, 256.0): cluster 0\n", + "Example n. 617 = (18.0, 255.0): cluster 0\n", + "Example n. 618 = (18.0, 254.0): cluster 0\n", + "Example n. 619 = (18.0, 246.0): cluster 0\n", + "Example n. 620 = (18.0, 245.0): cluster 0\n", + "Example n. 621 = (18.0, 244.0): cluster 0\n", + "Example n. 622 = (18.0, 243.0): cluster 0\n", + "Example n. 623 = (18.0, 242.0): cluster 0\n", + "Example n. 624 = (18.0, 241.0): cluster 0\n", + "Example n. 625 = (18.0, 234.0): cluster 0\n", + "Example n. 626 = (18.0, 228.0): cluster 0\n", + "Example n. 627 = (18.0, 227.0): cluster 0\n", + "Example n. 628 = (18.0, 226.0): cluster 0\n", + "Example n. 629 = (18.0, 225.0): cluster 0\n", + "Example n. 630 = (18.0, 224.0): cluster 0\n", + "Example n. 631 = (18.0, 223.0): cluster 0\n", + "Example n. 632 = (18.0, 222.0): cluster 0\n", + "Example n. 633 = (18.0, 221.0): cluster 0\n", + "Example n. 634 = (18.0, 220.0): cluster 0\n", + "Example n. 635 = (18.0, 219.0): cluster 0\n", + "Example n. 636 = (18.0, 218.0): cluster 0\n", + "Example n. 637 = (18.0, 217.0): cluster 0\n", + "Example n. 638 = (18.0, 216.0): cluster 0\n", + "Example n. 639 = (18.0, 215.0): cluster 0\n", + "Example n. 640 = (18.0, 214.0): cluster 0\n", + "Example n. 641 = (18.0, 208.0): cluster 0\n", + "Example n. 642 = (18.0, 207.0): cluster 0\n", + "Example n. 643 = (18.0, 206.0): cluster 0\n", + "Example n. 644 = (18.0, 205.0): cluster 0\n", + "Example n. 645 = (18.0, 204.0): cluster 0\n", + "Example n. 646 = (18.0, 203.0): cluster 0\n", + "Example n. 647 = (18.0, 202.0): cluster 0\n", + "Example n. 648 = (18.0, 201.0): cluster 0\n", + "Example n. 649 = (18.0, 200.0): cluster 0\n", + "Example n. 650 = (18.0, 199.0): cluster 0\n", + "Example n. 651 = (18.0, 198.0): cluster 0\n", + "Example n. 652 = (18.0, 197.0): cluster 0\n", + "Example n. 653 = (18.0, 196.0): cluster 0\n", + "Example n. 654 = (18.0, 195.0): cluster 0\n", + "Example n. 655 = (18.0, 194.0): cluster 0\n", + "Example n. 656 = (18.0, 193.0): cluster 0\n", + "Example n. 657 = (18.0, 192.0): cluster 0\n", + "Example n. 658 = (18.0, 191.0): cluster 0\n", + "Example n. 659 = (18.0, 190.0): cluster 0\n", + "Example n. 660 = (18.0, 150.0): cluster 0\n", + "Example n. 661 = (18.0, 149.0): cluster 0\n", + "Example n. 662 = (18.0, 148.0): cluster 0\n", + "Example n. 663 = (18.0, 147.0): cluster 0\n", + "Example n. 664 = (18.0, 146.0): cluster 0\n", + "Example n. 665 = (18.0, 144.0): cluster 0\n", + "Example n. 666 = (18.0, 143.0): cluster 0\n", + "Example n. 667 = (18.0, 142.0): cluster 0\n", + "Example n. 668 = (18.0, 141.0): cluster 0\n", + "Example n. 669 = (18.0, 140.0): cluster 0\n", + "Example n. 670 = (18.0, 133.0): cluster 0\n", + "Example n. 671 = (18.0, 132.0): cluster 0\n", + "Example n. 672 = (18.0, 131.0): cluster 0\n", + "Example n. 673 = (18.0, 130.0): cluster 0\n", + "Example n. 674 = (18.0, 129.0): cluster 0\n", + "Example n. 675 = (18.0, 121.0): cluster 0\n", + "Example n. 676 = (18.0, 120.0): cluster 0\n", + "Example n. 677 = (18.0, 119.0): cluster 0\n", + "Example n. 678 = (18.0, 118.0): cluster 0\n", + "Example n. 679 = (18.0, 117.0): cluster 0\n", + "Example n. 680 = (18.0, 93.0): cluster 0\n", + "Example n. 681 = (18.0, 92.0): cluster 0\n", + "Example n. 682 = (18.0, 91.0): cluster 0\n", + "Example n. 683 = (18.0, 90.0): cluster 0\n", + "Example n. 684 = (19.0, 502.0): cluster 0\n", + "Example n. 685 = (19.0, 501.0): cluster 0\n", + "Example n. 686 = (19.0, 500.0): cluster 0\n", + "Example n. 687 = (19.0, 499.0): cluster 0\n", + "Example n. 688 = (19.0, 498.0): cluster 0\n", + "Example n. 689 = (19.0, 497.0): cluster 0\n", + "Example n. 690 = (19.0, 382.0): cluster 0\n", + "Example n. 691 = (19.0, 381.0): cluster 0\n", + "Example n. 692 = (19.0, 380.0): cluster 0\n", + "Example n. 693 = (19.0, 350.0): cluster 0\n", + "Example n. 694 = (19.0, 349.0): cluster 0\n", + "Example n. 695 = (19.0, 348.0): cluster 0\n", + "Example n. 696 = (19.0, 347.0): cluster 0\n", + "Example n. 697 = (19.0, 312.0): cluster 0\n", + "Example n. 698 = (19.0, 311.0): cluster 0\n", + "Example n. 699 = (19.0, 310.0): cluster 0\n", + "Example n. 700 = (19.0, 309.0): cluster 0\n", + "Example n. 701 = (19.0, 308.0): cluster 0\n", + "Example n. 702 = (19.0, 307.0): cluster 0\n", + "Example n. 703 = (19.0, 282.0): cluster 0\n", + "Example n. 704 = (19.0, 281.0): cluster 0\n", + "Example n. 705 = (19.0, 280.0): cluster 0\n", + "Example n. 706 = (19.0, 279.0): cluster 0\n", + "Example n. 707 = (19.0, 278.0): cluster 0\n", + "Example n. 708 = (19.0, 277.0): cluster 0\n", + "Example n. 709 = (19.0, 276.0): cluster 0\n", + "Example n. 710 = (19.0, 275.0): cluster 0\n", + "Example n. 711 = (19.0, 274.0): cluster 0\n", + "Example n. 712 = (19.0, 273.0): cluster 0\n", + "Example n. 713 = (19.0, 272.0): cluster 0\n", + "Example n. 714 = (19.0, 271.0): cluster 0\n", + "Example n. 715 = (19.0, 270.0): cluster 0\n", + "Example n. 716 = (19.0, 269.0): cluster 0\n", + "Example n. 717 = (19.0, 268.0): cluster 0\n", + "Example n. 718 = (19.0, 267.0): cluster 0\n", + "Example n. 719 = (19.0, 266.0): cluster 0\n", + "Example n. 720 = (19.0, 265.0): cluster 0\n", + "Example n. 721 = (19.0, 264.0): cluster 0\n", + "Example n. 722 = (19.0, 263.0): cluster 0\n", + "Example n. 723 = (19.0, 262.0): cluster 0\n", + "Example n. 724 = (19.0, 261.0): cluster 0\n", + "Example n. 725 = (19.0, 260.0): cluster 0\n", + "Example n. 726 = (19.0, 259.0): cluster 0\n", + "Example n. 727 = (19.0, 258.0): cluster 0\n", + "Example n. 728 = (19.0, 257.0): cluster 0\n", + "Example n. 729 = (19.0, 256.0): cluster 0\n", + "Example n. 730 = (19.0, 255.0): cluster 0\n", + "Example n. 731 = (19.0, 254.0): cluster 0\n", + "Example n. 732 = (19.0, 253.0): cluster 0\n", + "Example n. 733 = (19.0, 251.0): cluster 0\n", + "Example n. 734 = (19.0, 250.0): cluster 0\n", + "Example n. 735 = (19.0, 249.0): cluster 0\n", + "Example n. 736 = (19.0, 246.0): cluster 0\n", + "Example n. 737 = (19.0, 245.0): cluster 0\n", + "Example n. 738 = (19.0, 244.0): cluster 0\n", + "Example n. 739 = (19.0, 243.0): cluster 0\n", + "Example n. 740 = (19.0, 242.0): cluster 0\n", + "Example n. 741 = (19.0, 236.0): cluster 0\n", + "Example n. 742 = (19.0, 235.0): cluster 0\n", + "Example n. 743 = (19.0, 234.0): cluster 0\n", + "Example n. 744 = (19.0, 233.0): cluster 0\n", + "Example n. 745 = (19.0, 232.0): cluster 0\n", + "Example n. 746 = (19.0, 228.0): cluster 0\n", + "Example n. 747 = (19.0, 227.0): cluster 0\n", + "Example n. 748 = (19.0, 226.0): cluster 0\n", + "Example n. 749 = (19.0, 225.0): cluster 0\n", + "Example n. 750 = (19.0, 224.0): cluster 0\n", + "Example n. 751 = (19.0, 223.0): cluster 0\n", + "Example n. 752 = (19.0, 222.0): cluster 0\n", + "Example n. 753 = (19.0, 221.0): cluster 0\n", + "Example n. 754 = (19.0, 220.0): cluster 0\n", + "Example n. 755 = (19.0, 219.0): cluster 0\n", + "Example n. 756 = (19.0, 218.0): cluster 0\n", + "Example n. 757 = (19.0, 217.0): cluster 0\n", + "Example n. 758 = (19.0, 216.0): cluster 0\n", + "Example n. 759 = (19.0, 215.0): cluster 0\n", + "Example n. 760 = (19.0, 214.0): cluster 0\n", + "Example n. 761 = (19.0, 208.0): cluster 0\n", + "Example n. 762 = (19.0, 207.0): cluster 0\n", + "Example n. 763 = (19.0, 206.0): cluster 0\n", + "Example n. 764 = (19.0, 205.0): cluster 0\n", + "Example n. 765 = (19.0, 204.0): cluster 0\n", + "Example n. 766 = (19.0, 203.0): cluster 0\n", + "Example n. 767 = (19.0, 202.0): cluster 0\n", + "Example n. 768 = (19.0, 201.0): cluster 0\n", + "Example n. 769 = (19.0, 200.0): cluster 0\n", + "Example n. 770 = (19.0, 199.0): cluster 0\n", + "Example n. 771 = (19.0, 198.0): cluster 0\n", + "Example n. 772 = (19.0, 197.0): cluster 0\n", + "Example n. 773 = (19.0, 196.0): cluster 0\n", + "Example n. 774 = (19.0, 195.0): cluster 0\n", + "Example n. 775 = (19.0, 194.0): cluster 0\n", + "Example n. 776 = (19.0, 193.0): cluster 0\n", + "Example n. 777 = (19.0, 192.0): cluster 0\n", + "Example n. 778 = (19.0, 191.0): cluster 0\n", + "Example n. 779 = (19.0, 190.0): cluster 0\n", + "Example n. 780 = (19.0, 189.0): cluster 0\n", + "Example n. 781 = (19.0, 150.0): cluster 0\n", + "Example n. 782 = (19.0, 149.0): cluster 0\n", + "Example n. 783 = (19.0, 148.0): cluster 0\n", + "Example n. 784 = (19.0, 147.0): cluster 0\n", + "Example n. 785 = (19.0, 146.0): cluster 0\n", + "Example n. 786 = (19.0, 145.0): cluster 0\n", + "Example n. 787 = (19.0, 144.0): cluster 0\n", + "Example n. 788 = (19.0, 143.0): cluster 0\n", + "Example n. 789 = (19.0, 142.0): cluster 0\n", + "Example n. 790 = (19.0, 141.0): cluster 0\n", + "Example n. 791 = (19.0, 140.0): cluster 0\n", + "Example n. 792 = (19.0, 133.0): cluster 0\n", + "Example n. 793 = (19.0, 132.0): cluster 0\n", + "Example n. 794 = (19.0, 131.0): cluster 0\n", + "Example n. 795 = (19.0, 130.0): cluster 0\n", + "Example n. 796 = (19.0, 129.0): cluster 0\n", + "Example n. 797 = (19.0, 121.0): cluster 0\n", + "Example n. 798 = (19.0, 120.0): cluster 0\n", + "Example n. 799 = (19.0, 119.0): cluster 0\n", + "Example n. 800 = (19.0, 118.0): cluster 0\n", + "Example n. 801 = (19.0, 117.0): cluster 0\n", + "Example n. 802 = (19.0, 94.0): cluster 0\n", + "Example n. 803 = (19.0, 93.0): cluster 0\n", + "Example n. 804 = (19.0, 92.0): cluster 0\n", + "Example n. 805 = (19.0, 91.0): cluster 0\n", + "Example n. 806 = (19.0, 90.0): cluster 0\n", + "Example n. 807 = (19.0, 89.0): cluster 0\n", + "Example n. 808 = (20.0, 502.0): cluster 0\n", + "Example n. 809 = (20.0, 501.0): cluster 0\n", + "Example n. 810 = (20.0, 500.0): cluster 0\n", + "Example n. 811 = (20.0, 499.0): cluster 0\n", + "Example n. 812 = (20.0, 498.0): cluster 0\n", + "Example n. 813 = (20.0, 497.0): cluster 0\n", + "Example n. 814 = (20.0, 383.0): cluster 0\n", + "Example n. 815 = (20.0, 382.0): cluster 0\n", + "Example n. 816 = (20.0, 381.0): cluster 0\n", + "Example n. 817 = (20.0, 380.0): cluster 0\n", + "Example n. 818 = (20.0, 379.0): cluster 0\n", + "Example n. 819 = (20.0, 350.0): cluster 0\n", + "Example n. 820 = (20.0, 349.0): cluster 0\n", + "Example n. 821 = (20.0, 348.0): cluster 0\n", + "Example n. 822 = (20.0, 347.0): cluster 0\n", + "Example n. 823 = (20.0, 346.0): cluster 0\n", + "Example n. 824 = (20.0, 311.0): cluster 0\n", + "Example n. 825 = (20.0, 310.0): cluster 0\n", + "Example n. 826 = (20.0, 309.0): cluster 0\n", + "Example n. 827 = (20.0, 308.0): cluster 0\n", + "Example n. 828 = (20.0, 287.0): cluster 0\n", + "Example n. 829 = (20.0, 286.0): cluster 0\n", + "Example n. 830 = (20.0, 285.0): cluster 0\n", + "Example n. 831 = (20.0, 284.0): cluster 0\n", + "Example n. 832 = (20.0, 283.0): cluster 0\n", + "Example n. 833 = (20.0, 282.0): cluster 0\n", + "Example n. 834 = (20.0, 281.0): cluster 0\n", + "Example n. 835 = (20.0, 280.0): cluster 0\n", + "Example n. 836 = (20.0, 279.0): cluster 0\n", + "Example n. 837 = (20.0, 278.0): cluster 0\n", + "Example n. 838 = (20.0, 277.0): cluster 0\n", + "Example n. 839 = (20.0, 276.0): cluster 0\n", + "Example n. 840 = (20.0, 275.0): cluster 0\n", + "Example n. 841 = (20.0, 274.0): cluster 0\n", + "Example n. 842 = (20.0, 272.0): cluster 0\n", + "Example n. 843 = (20.0, 271.0): cluster 0\n", + "Example n. 844 = (20.0, 270.0): cluster 0\n", + "Example n. 845 = (20.0, 269.0): cluster 0\n", + "Example n. 846 = (20.0, 268.0): cluster 0\n", + "Example n. 847 = (20.0, 267.0): cluster 0\n", + "Example n. 848 = (20.0, 266.0): cluster 0\n", + "Example n. 849 = (20.0, 265.0): cluster 0\n", + "Example n. 850 = (20.0, 263.0): cluster 0\n", + "Example n. 851 = (20.0, 262.0): cluster 0\n", + "Example n. 852 = (20.0, 261.0): cluster 0\n", + "Example n. 853 = (20.0, 260.0): cluster 0\n", + "Example n. 854 = (20.0, 259.0): cluster 0\n", + "Example n. 855 = (20.0, 258.0): cluster 0\n", + "Example n. 856 = (20.0, 257.0): cluster 0\n", + "Example n. 857 = (20.0, 256.0): cluster 0\n", + "Example n. 858 = (20.0, 255.0): cluster 0\n", + "Example n. 859 = (20.0, 254.0): cluster 0\n", + "Example n. 860 = (20.0, 253.0): cluster 0\n", + "Example n. 861 = (20.0, 252.0): cluster 0\n", + "Example n. 862 = (20.0, 251.0): cluster 0\n", + "Example n. 863 = (20.0, 250.0): cluster 0\n", + "Example n. 864 = (20.0, 249.0): cluster 0\n", + "Example n. 865 = (20.0, 248.0): cluster 0\n", + "Example n. 866 = (20.0, 246.0): cluster 0\n", + "Example n. 867 = (20.0, 245.0): cluster 0\n", + "Example n. 868 = (20.0, 244.0): cluster 0\n", + "Example n. 869 = (20.0, 243.0): cluster 0\n", + "Example n. 870 = (20.0, 242.0): cluster 0\n", + "Example n. 871 = (20.0, 237.0): cluster 0\n", + "Example n. 872 = (20.0, 236.0): cluster 0\n", + "Example n. 873 = (20.0, 235.0): cluster 0\n", + "Example n. 874 = (20.0, 234.0): cluster 0\n", + "Example n. 875 = (20.0, 233.0): cluster 0\n", + "Example n. 876 = (20.0, 232.0): cluster 0\n", + "Example n. 877 = (20.0, 226.0): cluster 0\n", + "Example n. 878 = (20.0, 225.0): cluster 0\n", + "Example n. 879 = (20.0, 224.0): cluster 0\n", + "Example n. 880 = (20.0, 223.0): cluster 0\n", + "Example n. 881 = (20.0, 222.0): cluster 0\n", + "Example n. 882 = (20.0, 221.0): cluster 0\n", + "Example n. 883 = (20.0, 220.0): cluster 0\n", + "Example n. 884 = (20.0, 219.0): cluster 0\n", + "Example n. 885 = (20.0, 218.0): cluster 0\n", + "Example n. 886 = (20.0, 217.0): cluster 0\n", + "Example n. 887 = (20.0, 216.0): cluster 0\n", + "Example n. 888 = (20.0, 215.0): cluster 0\n", + "Example n. 889 = (20.0, 214.0): cluster 0\n", + "Example n. 890 = (20.0, 208.0): cluster 0\n", + "Example n. 891 = (20.0, 207.0): cluster 0\n", + "Example n. 892 = (20.0, 206.0): cluster 0\n", + "Example n. 893 = (20.0, 205.0): cluster 0\n", + "Example n. 894 = (20.0, 204.0): cluster 0\n", + "Example n. 895 = (20.0, 203.0): cluster 0\n", + "Example n. 896 = (20.0, 202.0): cluster 0\n", + "Example n. 897 = (20.0, 201.0): cluster 0\n", + "Example n. 898 = (20.0, 200.0): cluster 0\n", + "Example n. 899 = (20.0, 199.0): cluster 0\n", + "Example n. 900 = (20.0, 198.0): cluster 0\n", + "Example n. 901 = (20.0, 197.0): cluster 0\n", + "Example n. 902 = (20.0, 196.0): cluster 0\n", + "Example n. 903 = (20.0, 195.0): cluster 0\n", + "Example n. 904 = (20.0, 194.0): cluster 0\n", + "Example n. 905 = (20.0, 193.0): cluster 0\n", + "Example n. 906 = (20.0, 192.0): cluster 0\n", + "Example n. 907 = (20.0, 191.0): cluster 0\n", + "Example n. 908 = (20.0, 190.0): cluster 0\n", + "Example n. 909 = (20.0, 189.0): cluster 0\n", + "Example n. 910 = (20.0, 188.0): cluster 0\n", + "Example n. 911 = (20.0, 150.0): cluster 0\n", + "Example n. 912 = (20.0, 149.0): cluster 0\n", + "Example n. 913 = (20.0, 148.0): cluster 0\n", + "Example n. 914 = (20.0, 147.0): cluster 0\n", + "Example n. 915 = (20.0, 146.0): cluster 0\n", + "Example n. 916 = (20.0, 145.0): cluster 0\n", + "Example n. 917 = (20.0, 144.0): cluster 0\n", + "Example n. 918 = (20.0, 143.0): cluster 0\n", + "Example n. 919 = (20.0, 142.0): cluster 0\n", + "Example n. 920 = (20.0, 141.0): cluster 0\n", + "Example n. 921 = (20.0, 140.0): cluster 0\n", + "Example n. 922 = (20.0, 133.0): cluster 0\n", + "Example n. 923 = (20.0, 132.0): cluster 0\n", + "Example n. 924 = (20.0, 131.0): cluster 3\n", + "Example n. 925 = (20.0, 130.0): cluster 3\n", + "Example n. 926 = (20.0, 129.0): cluster 3\n", + "Example n. 927 = (20.0, 125.0): cluster 3\n", + "Example n. 928 = (20.0, 124.0): cluster 3\n", + "Example n. 929 = (20.0, 123.0): cluster 3\n", + "Example n. 930 = (20.0, 122.0): cluster 3\n", + "Example n. 931 = (20.0, 121.0): cluster 3\n", + "Example n. 932 = (20.0, 120.0): cluster 3\n", + "Example n. 933 = (20.0, 119.0): cluster 3\n", + "Example n. 934 = (20.0, 118.0): cluster 3\n", + "Example n. 935 = (20.0, 117.0): cluster 3\n", + "Example n. 936 = (20.0, 94.0): cluster 0\n", + "Example n. 937 = (20.0, 93.0): cluster 0\n", + "Example n. 938 = (20.0, 92.0): cluster 0\n", + "Example n. 939 = (20.0, 91.0): cluster 0\n", + "Example n. 940 = (20.0, 90.0): cluster 0\n", + "Example n. 941 = (20.0, 89.0): cluster 0\n", + "Example n. 942 = (21.0, 501.0): cluster 0\n", + "Example n. 943 = (21.0, 500.0): cluster 0\n", + "Example n. 944 = (21.0, 499.0): cluster 0\n", + "Example n. 945 = (21.0, 498.0): cluster 0\n", + "Example n. 946 = (21.0, 497.0): cluster 0\n", + "Example n. 947 = (21.0, 383.0): cluster 0\n", + "Example n. 948 = (21.0, 382.0): cluster 0\n", + "Example n. 949 = (21.0, 381.0): cluster 0\n", + "Example n. 950 = (21.0, 380.0): cluster 0\n", + "Example n. 951 = (21.0, 379.0): cluster 0\n", + "Example n. 952 = (21.0, 369.0): cluster 0\n", + "Example n. 953 = (21.0, 351.0): cluster 0\n", + "Example n. 954 = (21.0, 350.0): cluster 0\n", + "Example n. 955 = (21.0, 349.0): cluster 0\n", + "Example n. 956 = (21.0, 348.0): cluster 0\n", + "Example n. 957 = (21.0, 347.0): cluster 0\n", + "Example n. 958 = (21.0, 346.0): cluster 0\n", + "Example n. 959 = (21.0, 307.0): cluster 0\n", + "Example n. 960 = (21.0, 306.0): cluster 0\n", + "Example n. 961 = (21.0, 305.0): cluster 0\n", + "Example n. 962 = (21.0, 287.0): cluster 0\n", + "Example n. 963 = (21.0, 286.0): cluster 0\n", + "Example n. 964 = (21.0, 285.0): cluster 0\n", + "Example n. 965 = (21.0, 284.0): cluster 0\n", + "Example n. 966 = (21.0, 283.0): cluster 0\n", + "Example n. 967 = (21.0, 282.0): cluster 0\n", + "Example n. 968 = (21.0, 281.0): cluster 0\n", + "Example n. 969 = (21.0, 280.0): cluster 0\n", + "Example n. 970 = (21.0, 279.0): cluster 0\n", + "Example n. 971 = (21.0, 278.0): cluster 0\n", + "Example n. 972 = (21.0, 277.0): cluster 0\n", + "Example n. 973 = (21.0, 276.0): cluster 0\n", + "Example n. 974 = (21.0, 275.0): cluster 0\n", + "Example n. 975 = (21.0, 271.0): cluster 0\n", + "Example n. 976 = (21.0, 270.0): cluster 0\n", + "Example n. 977 = (21.0, 269.0): cluster 0\n", + "Example n. 978 = (21.0, 268.0): cluster 0\n", + "Example n. 979 = (21.0, 267.0): cluster 0\n", + "Example n. 980 = (21.0, 262.0): cluster 0\n", + "Example n. 981 = (21.0, 261.0): cluster 3\n", + "Example n. 982 = (21.0, 260.0): cluster 3\n", + "Example n. 983 = (21.0, 259.0): cluster 3\n", + "Example n. 984 = (21.0, 258.0): cluster 3\n", + "Example n. 985 = (21.0, 257.0): cluster 3\n", + "Example n. 986 = (21.0, 256.0): cluster 3\n", + "Example n. 987 = (21.0, 255.0): cluster 3\n", + "Example n. 988 = (21.0, 254.0): cluster 3\n", + "Example n. 989 = (21.0, 253.0): cluster 3\n", + "Example n. 990 = (21.0, 252.0): cluster 3\n", + "Example n. 991 = (21.0, 251.0): cluster 3\n", + "Example n. 992 = (21.0, 250.0): cluster 3\n", + "Example n. 993 = (21.0, 249.0): cluster 0\n", + "Example n. 994 = (21.0, 248.0): cluster 0\n", + "Example n. 995 = (21.0, 247.0): cluster 0\n", + "Example n. 996 = (21.0, 244.0): cluster 0\n", + "Example n. 997 = (21.0, 237.0): cluster 0\n", + "Example n. 998 = (21.0, 236.0): cluster 0\n", + "Example n. 999 = (21.0, 235.0): cluster 0\n", + "Example n. 1000 = (21.0, 234.0): cluster 0\n", + "Example n. 1001 = (21.0, 233.0): cluster 0\n", + "Example n. 1002 = (21.0, 232.0): cluster 0\n", + "Example n. 1003 = (21.0, 231.0): cluster 0\n", + "Example n. 1004 = (21.0, 226.0): cluster 0\n", + "Example n. 1005 = (21.0, 225.0): cluster 0\n", + "Example n. 1006 = (21.0, 224.0): cluster 0\n", + "Example n. 1007 = (21.0, 223.0): cluster 0\n", + "Example n. 1008 = (21.0, 222.0): cluster 0\n", + "Example n. 1009 = (21.0, 221.0): cluster 0\n", + "Example n. 1010 = (21.0, 220.0): cluster 0\n", + "Example n. 1011 = (21.0, 219.0): cluster 0\n", + "Example n. 1012 = (21.0, 218.0): cluster 0\n", + "Example n. 1013 = (21.0, 217.0): cluster 0\n", + "Example n. 1014 = (21.0, 216.0): cluster 0\n", + "Example n. 1015 = (21.0, 215.0): cluster 0\n", + "Example n. 1016 = (21.0, 214.0): cluster 0\n", + "Example n. 1017 = (21.0, 213.0): cluster 0\n", + "Example n. 1018 = (21.0, 212.0): cluster 0\n", + "Example n. 1019 = (21.0, 208.0): cluster 0\n", + "Example n. 1020 = (21.0, 207.0): cluster 0\n", + "Example n. 1021 = (21.0, 206.0): cluster 0\n", + "Example n. 1022 = (21.0, 205.0): cluster 0\n", + "Example n. 1023 = (21.0, 204.0): cluster 0\n", + "Example n. 1024 = (21.0, 203.0): cluster 0\n", + "Example n. 1025 = (21.0, 202.0): cluster 0\n", + "Example n. 1026 = (21.0, 201.0): cluster 0\n", + "Example n. 1027 = (21.0, 200.0): cluster 0\n", + "Example n. 1028 = (21.0, 199.0): cluster 0\n", + "Example n. 1029 = (21.0, 198.0): cluster 0\n", + "Example n. 1030 = (21.0, 197.0): cluster 0\n", + "Example n. 1031 = (21.0, 196.0): cluster 0\n", + "Example n. 1032 = (21.0, 195.0): cluster 0\n", + "Example n. 1033 = (21.0, 194.0): cluster 0\n", + "Example n. 1034 = (21.0, 193.0): cluster 0\n", + "Example n. 1035 = (21.0, 192.0): cluster 0\n", + "Example n. 1036 = (21.0, 191.0): cluster 0\n", + "Example n. 1037 = (21.0, 190.0): cluster 0\n", + "Example n. 1038 = (21.0, 189.0): cluster 3\n", + "Example n. 1039 = (21.0, 188.0): cluster 3\n", + "Example n. 1040 = (21.0, 150.0): cluster 3\n", + "Example n. 1041 = (21.0, 149.0): cluster 3\n", + "Example n. 1042 = (21.0, 148.0): cluster 3\n", + "Example n. 1043 = (21.0, 147.0): cluster 3\n", + "Example n. 1044 = (21.0, 146.0): cluster 3\n", + "Example n. 1045 = (21.0, 144.0): cluster 3\n", + "Example n. 1046 = (21.0, 143.0): cluster 3\n", + "Example n. 1047 = (21.0, 142.0): cluster 3\n", + "Example n. 1048 = (21.0, 141.0): cluster 3\n", + "Example n. 1049 = (21.0, 140.0): cluster 3\n", + "Example n. 1050 = (21.0, 133.0): cluster 0\n", + "Example n. 1051 = (21.0, 132.0): cluster 0\n", + "Example n. 1052 = (21.0, 131.0): cluster 0\n", + "Example n. 1053 = (21.0, 130.0): cluster 0\n", + "Example n. 1054 = (21.0, 129.0): cluster 0\n", + "Example n. 1055 = (21.0, 126.0): cluster 0\n", + "Example n. 1056 = (21.0, 125.0): cluster 0\n", + "Example n. 1057 = (21.0, 124.0): cluster 0\n", + "Example n. 1058 = (21.0, 123.0): cluster 0\n", + "Example n. 1059 = (21.0, 122.0): cluster 0\n", + "Example n. 1060 = (21.0, 121.0): cluster 0\n", + "Example n. 1061 = (21.0, 120.0): cluster 0\n", + "Example n. 1062 = (21.0, 119.0): cluster 0\n", + "Example n. 1063 = (21.0, 118.0): cluster 0\n", + "Example n. 1064 = (21.0, 117.0): cluster 0\n", + "Example n. 1065 = (21.0, 94.0): cluster 0\n", + "Example n. 1066 = (21.0, 93.0): cluster 0\n", + "Example n. 1067 = (21.0, 92.0): cluster 0\n", + "Example n. 1068 = (21.0, 91.0): cluster 0\n", + "Example n. 1069 = (21.0, 90.0): cluster 0\n", + "Example n. 1070 = (22.0, 500.0): cluster 0\n", + "Example n. 1071 = (22.0, 499.0): cluster 0\n", + "Example n. 1072 = (22.0, 498.0): cluster 0\n", + "Example n. 1073 = (22.0, 383.0): cluster 0\n", + "Example n. 1074 = (22.0, 382.0): cluster 0\n", + "Example n. 1075 = (22.0, 381.0): cluster 0\n", + "Example n. 1076 = (22.0, 380.0): cluster 0\n", + "Example n. 1077 = (22.0, 379.0): cluster 0\n", + "Example n. 1078 = (22.0, 371.0): cluster 0\n", + "Example n. 1079 = (22.0, 370.0): cluster 0\n", + "Example n. 1080 = (22.0, 369.0): cluster 0\n", + "Example n. 1081 = (22.0, 368.0): cluster 0\n", + "Example n. 1082 = (22.0, 350.0): cluster 0\n", + "Example n. 1083 = (22.0, 349.0): cluster 0\n", + "Example n. 1084 = (22.0, 348.0): cluster 0\n", + "Example n. 1085 = (22.0, 347.0): cluster 0\n", + "Example n. 1086 = (22.0, 346.0): cluster 0\n", + "Example n. 1087 = (22.0, 308.0): cluster 0\n", + "Example n. 1088 = (22.0, 307.0): cluster 0\n", + "Example n. 1089 = (22.0, 306.0): cluster 0\n", + "Example n. 1090 = (22.0, 305.0): cluster 0\n", + "Example n. 1091 = (22.0, 304.0): cluster 0\n", + "Example n. 1092 = (22.0, 302.0): cluster 0\n", + "Example n. 1093 = (22.0, 301.0): cluster 0\n", + "Example n. 1094 = (22.0, 300.0): cluster 0\n", + "Example n. 1095 = (22.0, 295.0): cluster 0\n", + "Example n. 1096 = (22.0, 294.0): cluster 1\n", + "Example n. 1097 = (22.0, 293.0): cluster 1\n", + "Example n. 1098 = (22.0, 292.0): cluster 3\n", + "Example n. 1099 = (22.0, 288.0): cluster 3\n", + "Example n. 1100 = (22.0, 287.0): cluster 3\n", + "Example n. 1101 = (22.0, 286.0): cluster 3\n", + "Example n. 1102 = (22.0, 285.0): cluster 3\n", + "Example n. 1103 = (22.0, 284.0): cluster 3\n", + "Example n. 1104 = (22.0, 283.0): cluster 3\n", + "Example n. 1105 = (22.0, 282.0): cluster 3\n", + "Example n. 1106 = (22.0, 281.0): cluster 3\n", + "Example n. 1107 = (22.0, 280.0): cluster 3\n", + "Example n. 1108 = (22.0, 279.0): cluster 3\n", + "Example n. 1109 = (22.0, 278.0): cluster 3\n", + "Example n. 1110 = (22.0, 277.0): cluster 3\n", + "Example n. 1111 = (22.0, 276.0): cluster 3\n", + "Example n. 1112 = (22.0, 270.0): cluster 3\n", + "Example n. 1113 = (22.0, 269.0): cluster 3\n", + "Example n. 1114 = (22.0, 268.0): cluster 3\n", + "Example n. 1115 = (22.0, 261.0): cluster 3\n", + "Example n. 1116 = (22.0, 260.0): cluster 3\n", + "Example n. 1117 = (22.0, 259.0): cluster 3\n", + "Example n. 1118 = (22.0, 258.0): cluster 3\n", + "Example n. 1119 = (22.0, 257.0): cluster 3\n", + "Example n. 1120 = (22.0, 256.0): cluster 3\n", + "Example n. 1121 = (22.0, 255.0): cluster 3\n", + "Example n. 1122 = (22.0, 254.0): cluster 0\n", + "Example n. 1123 = (22.0, 253.0): cluster 0\n", + "Example n. 1124 = (22.0, 252.0): cluster 0\n", + "Example n. 1125 = (22.0, 251.0): cluster 0\n", + "Example n. 1126 = (22.0, 250.0): cluster 0\n", + "Example n. 1127 = (22.0, 249.0): cluster 0\n", + "Example n. 1128 = (22.0, 248.0): cluster 0\n", + "Example n. 1129 = (22.0, 247.0): cluster 0\n", + "Example n. 1130 = (22.0, 246.0): cluster 0\n", + "Example n. 1131 = (22.0, 245.0): cluster 0\n", + "Example n. 1132 = (22.0, 244.0): cluster 0\n", + "Example n. 1133 = (22.0, 243.0): cluster 0\n", + "Example n. 1134 = (22.0, 237.0): cluster 0\n", + "Example n. 1135 = (22.0, 236.0): cluster 0\n", + "Example n. 1136 = (22.0, 235.0): cluster 0\n", + "Example n. 1137 = (22.0, 234.0): cluster 0\n", + "Example n. 1138 = (22.0, 233.0): cluster 0\n", + "Example n. 1139 = (22.0, 232.0): cluster 0\n", + "Example n. 1140 = (22.0, 231.0): cluster 0\n", + "Example n. 1141 = (22.0, 230.0): cluster 0\n", + "Example n. 1142 = (22.0, 227.0): cluster 0\n", + "Example n. 1143 = (22.0, 226.0): cluster 0\n", + "Example n. 1144 = (22.0, 225.0): cluster 0\n", + "Example n. 1145 = (22.0, 224.0): cluster 0\n", + "Example n. 1146 = (22.0, 223.0): cluster 0\n", + "Example n. 1147 = (22.0, 222.0): cluster 0\n", + "Example n. 1148 = (22.0, 221.0): cluster 0\n", + "Example n. 1149 = (22.0, 220.0): cluster 0\n", + "Example n. 1150 = (22.0, 219.0): cluster 0\n", + "Example n. 1151 = (22.0, 218.0): cluster 0\n", + "Example n. 1152 = (22.0, 217.0): cluster 0\n", + "Example n. 1153 = (22.0, 216.0): cluster 0\n", + "Example n. 1154 = (22.0, 215.0): cluster 0\n", + "Example n. 1155 = (22.0, 214.0): cluster 0\n", + "Example n. 1156 = (22.0, 213.0): cluster 0\n", + "Example n. 1157 = (22.0, 212.0): cluster 0\n", + "Example n. 1158 = (22.0, 211.0): cluster 0\n", + "Example n. 1159 = (22.0, 210.0): cluster 0\n", + "Example n. 1160 = (22.0, 207.0): cluster 0\n", + "Example n. 1161 = (22.0, 206.0): cluster 0\n", + "Example n. 1162 = (22.0, 205.0): cluster 0\n", + "Example n. 1163 = (22.0, 203.0): cluster 0\n", + "Example n. 1164 = (22.0, 202.0): cluster 0\n", + "Example n. 1165 = (22.0, 201.0): cluster 1\n", + "Example n. 1166 = (22.0, 200.0): cluster 1\n", + "Example n. 1167 = (22.0, 198.0): cluster 1\n", + "Example n. 1168 = (22.0, 197.0): cluster 1\n", + "Example n. 1169 = (22.0, 196.0): cluster 1\n", + "Example n. 1170 = (22.0, 195.0): cluster 3\n", + "Example n. 1171 = (22.0, 193.0): cluster 3\n", + "Example n. 1172 = (22.0, 192.0): cluster 3\n", + "Example n. 1173 = (22.0, 191.0): cluster 3\n", + "Example n. 1174 = (22.0, 190.0): cluster 3\n", + "Example n. 1175 = (22.0, 189.0): cluster 3\n", + "Example n. 1176 = (22.0, 188.0): cluster 3\n", + "Example n. 1177 = (22.0, 149.0): cluster 3\n", + "Example n. 1178 = (22.0, 148.0): cluster 3\n", + "Example n. 1179 = (22.0, 147.0): cluster 3\n", + "Example n. 1180 = (22.0, 144.0): cluster 3\n", + "Example n. 1181 = (22.0, 143.0): cluster 3\n", + "Example n. 1182 = (22.0, 142.0): cluster 3\n", + "Example n. 1183 = (22.0, 141.0): cluster 3\n", + "Example n. 1184 = (22.0, 140.0): cluster 3\n", + "Example n. 1185 = (22.0, 131.0): cluster 3\n", + "Example n. 1186 = (22.0, 130.0): cluster 3\n", + "Example n. 1187 = (22.0, 129.0): cluster 3\n", + "Example n. 1188 = (22.0, 126.0): cluster 3\n", + "Example n. 1189 = (22.0, 125.0): cluster 3\n", + "Example n. 1190 = (22.0, 124.0): cluster 3\n", + "Example n. 1191 = (22.0, 123.0): cluster 3\n", + "Example n. 1192 = (22.0, 122.0): cluster 3\n", + "Example n. 1193 = (22.0, 121.0): cluster 3\n", + "Example n. 1194 = (22.0, 120.0): cluster 0\n", + "Example n. 1195 = (22.0, 119.0): cluster 0\n", + "Example n. 1196 = (22.0, 118.0): cluster 0\n", + "Example n. 1197 = (22.0, 93.0): cluster 0\n", + "Example n. 1198 = (22.0, 92.0): cluster 0\n", + "Example n. 1199 = (22.0, 91.0): cluster 0\n", + "Example n. 1200 = (22.0, 90.0): cluster 0\n", + "Example n. 1201 = (23.0, 456.0): cluster 0\n", + "Example n. 1202 = (23.0, 455.0): cluster 0\n", + "Example n. 1203 = (23.0, 454.0): cluster 0\n", + "Example n. 1204 = (23.0, 453.0): cluster 0\n", + "Example n. 1205 = (23.0, 383.0): cluster 0\n", + "Example n. 1206 = (23.0, 382.0): cluster 0\n", + "Example n. 1207 = (23.0, 381.0): cluster 0\n", + "Example n. 1208 = (23.0, 380.0): cluster 0\n", + "Example n. 1209 = (23.0, 379.0): cluster 0\n", + "Example n. 1210 = (23.0, 372.0): cluster 0\n", + "Example n. 1211 = (23.0, 371.0): cluster 0\n", + "Example n. 1212 = (23.0, 370.0): cluster 0\n", + "Example n. 1213 = (23.0, 369.0): cluster 0\n", + "Example n. 1214 = (23.0, 368.0): cluster 0\n", + "Example n. 1215 = (23.0, 367.0): cluster 0\n", + "Example n. 1216 = (23.0, 350.0): cluster 0\n", + "Example n. 1217 = (23.0, 349.0): cluster 0\n", + "Example n. 1218 = (23.0, 348.0): cluster 0\n", + "Example n. 1219 = (23.0, 347.0): cluster 0\n", + "Example n. 1220 = (23.0, 308.0): cluster 0\n", + "Example n. 1221 = (23.0, 307.0): cluster 0\n", + "Example n. 1222 = (23.0, 306.0): cluster 0\n", + "Example n. 1223 = (23.0, 305.0): cluster 0\n", + "Example n. 1224 = (23.0, 304.0): cluster 0\n", + "Example n. 1225 = (23.0, 303.0): cluster 0\n", + "Example n. 1226 = (23.0, 302.0): cluster 0\n", + "Example n. 1227 = (23.0, 301.0): cluster 0\n", + "Example n. 1228 = (23.0, 300.0): cluster 0\n", + "Example n. 1229 = (23.0, 299.0): cluster 0\n", + "Example n. 1230 = (23.0, 295.0): cluster 0\n", + "Example n. 1231 = (23.0, 294.0): cluster 0\n", + "Example n. 1232 = (23.0, 293.0): cluster 0\n", + "Example n. 1233 = (23.0, 292.0): cluster 0\n", + "Example n. 1234 = (23.0, 291.0): cluster 1\n", + "Example n. 1235 = (23.0, 288.0): cluster 1\n", + "Example n. 1236 = (23.0, 287.0): cluster 1\n", + "Example n. 1237 = (23.0, 286.0): cluster 1\n", + "Example n. 1238 = (23.0, 285.0): cluster 1\n", + "Example n. 1239 = (23.0, 284.0): cluster 1\n", + "Example n. 1240 = (23.0, 283.0): cluster 1\n", + "Example n. 1241 = (23.0, 282.0): cluster 1\n", + "Example n. 1242 = (23.0, 281.0): cluster 3\n", + "Example n. 1243 = (23.0, 280.0): cluster 3\n", + "Example n. 1244 = (23.0, 279.0): cluster 3\n", + "Example n. 1245 = (23.0, 278.0): cluster 3\n", + "Example n. 1246 = (23.0, 277.0): cluster 3\n", + "Example n. 1247 = (23.0, 276.0): cluster 3\n", + "Example n. 1248 = (23.0, 275.0): cluster 3\n", + "Example n. 1249 = (23.0, 267.0): cluster 3\n", + "Example n. 1250 = (23.0, 266.0): cluster 3\n", + "Example n. 1251 = (23.0, 265.0): cluster 3\n", + "Example n. 1252 = (23.0, 261.0): cluster 3\n", + "Example n. 1253 = (23.0, 260.0): cluster 3\n", + "Example n. 1254 = (23.0, 259.0): cluster 3\n", + "Example n. 1255 = (23.0, 258.0): cluster 3\n", + "Example n. 1256 = (23.0, 257.0): cluster 3\n", + "Example n. 1257 = (23.0, 256.0): cluster 3\n", + "Example n. 1258 = (23.0, 255.0): cluster 3\n", + "Example n. 1259 = (23.0, 254.0): cluster 3\n", + "Example n. 1260 = (23.0, 253.0): cluster 3\n", + "Example n. 1261 = (23.0, 252.0): cluster 3\n", + "Example n. 1262 = (23.0, 251.0): cluster 3\n", + "Example n. 1263 = (23.0, 250.0): cluster 3\n", + "Example n. 1264 = (23.0, 249.0): cluster 3\n", + "Example n. 1265 = (23.0, 248.0): cluster 3\n", + "Example n. 1266 = (23.0, 247.0): cluster 0\n", + "Example n. 1267 = (23.0, 246.0): cluster 0\n", + "Example n. 1268 = (23.0, 245.0): cluster 0\n", + "Example n. 1269 = (23.0, 244.0): cluster 0\n", + "Example n. 1270 = (23.0, 243.0): cluster 0\n", + "Example n. 1271 = (23.0, 242.0): cluster 0\n", + "Example n. 1272 = (23.0, 239.0): cluster 0\n", + "Example n. 1273 = (23.0, 238.0): cluster 0\n", + "Example n. 1274 = (23.0, 237.0): cluster 0\n", + "Example n. 1275 = (23.0, 236.0): cluster 0\n", + "Example n. 1276 = (23.0, 235.0): cluster 0\n", + "Example n. 1277 = (23.0, 234.0): cluster 0\n", + "Example n. 1278 = (23.0, 233.0): cluster 0\n", + "Example n. 1279 = (23.0, 232.0): cluster 0\n", + "Example n. 1280 = (23.0, 231.0): cluster 0\n", + "Example n. 1281 = (23.0, 230.0): cluster 0\n", + "Example n. 1282 = (23.0, 229.0): cluster 0\n", + "Example n. 1283 = (23.0, 228.0): cluster 0\n", + "Example n. 1284 = (23.0, 227.0): cluster 0\n", + "Example n. 1285 = (23.0, 226.0): cluster 0\n", + "Example n. 1286 = (23.0, 225.0): cluster 0\n", + "Example n. 1287 = (23.0, 224.0): cluster 0\n", + "Example n. 1288 = (23.0, 223.0): cluster 0\n", + "Example n. 1289 = (23.0, 222.0): cluster 0\n", + "Example n. 1290 = (23.0, 221.0): cluster 0\n", + "Example n. 1291 = (23.0, 220.0): cluster 0\n", + "Example n. 1292 = (23.0, 219.0): cluster 0\n", + "Example n. 1293 = (23.0, 218.0): cluster 0\n", + "Example n. 1294 = (23.0, 217.0): cluster 0\n", + "Example n. 1295 = (23.0, 215.0): cluster 0\n", + "Example n. 1296 = (23.0, 214.0): cluster 0\n", + "Example n. 1297 = (23.0, 213.0): cluster 0\n", + "Example n. 1298 = (23.0, 212.0): cluster 0\n", + "Example n. 1299 = (23.0, 211.0): cluster 0\n", + "Example n. 1300 = (23.0, 210.0): cluster 0\n", + "Example n. 1301 = (23.0, 209.0): cluster 0\n", + "Example n. 1302 = (23.0, 192.0): cluster 0\n", + "Example n. 1303 = (23.0, 191.0): cluster 0\n", + "Example n. 1304 = (23.0, 190.0): cluster 1\n", + "Example n. 1305 = (23.0, 189.0): cluster 1\n", + "Example n. 1306 = (23.0, 145.0): cluster 1\n", + "Example n. 1307 = (23.0, 144.0): cluster 1\n", + "Example n. 1308 = (23.0, 143.0): cluster 1\n", + "Example n. 1309 = (23.0, 142.0): cluster 1\n", + "Example n. 1310 = (23.0, 141.0): cluster 1\n", + "Example n. 1311 = (23.0, 140.0): cluster 1\n", + "Example n. 1312 = (23.0, 131.0): cluster 1\n", + "Example n. 1313 = (23.0, 130.0): cluster 1\n", + "Example n. 1314 = (23.0, 129.0): cluster 3\n", + "Example n. 1315 = (23.0, 128.0): cluster 3\n", + "Example n. 1316 = (23.0, 127.0): cluster 3\n", + "Example n. 1317 = (23.0, 126.0): cluster 3\n", + "Example n. 1318 = (23.0, 125.0): cluster 3\n", + "Example n. 1319 = (23.0, 124.0): cluster 3\n", + "Example n. 1320 = (23.0, 123.0): cluster 3\n", + "Example n. 1321 = (23.0, 122.0): cluster 3\n", + "Example n. 1322 = (24.0, 477.0): cluster 3\n", + "Example n. 1323 = (24.0, 476.0): cluster 3\n", + "Example n. 1324 = (24.0, 475.0): cluster 3\n", + "Example n. 1325 = (24.0, 474.0): cluster 3\n", + "Example n. 1326 = (24.0, 456.0): cluster 3\n", + "Example n. 1327 = (24.0, 455.0): cluster 3\n", + "Example n. 1328 = (24.0, 454.0): cluster 3\n", + "Example n. 1329 = (24.0, 453.0): cluster 3\n", + "Example n. 1330 = (24.0, 452.0): cluster 3\n", + "Example n. 1331 = (24.0, 388.0): cluster 3\n", + "Example n. 1332 = (24.0, 387.0): cluster 3\n", + "Example n. 1333 = (24.0, 386.0): cluster 3\n", + "Example n. 1334 = (24.0, 385.0): cluster 3\n", + "Example n. 1335 = (24.0, 381.0): cluster 3\n", + "Example n. 1336 = (24.0, 372.0): cluster 3\n", + "Example n. 1337 = (24.0, 371.0): cluster 3\n", + "Example n. 1338 = (24.0, 370.0): cluster 0\n", + "Example n. 1339 = (24.0, 369.0): cluster 0\n", + "Example n. 1340 = (24.0, 368.0): cluster 0\n", + "Example n. 1341 = (24.0, 367.0): cluster 0\n", + "Example n. 1342 = (24.0, 308.0): cluster 0\n", + "Example n. 1343 = (24.0, 307.0): cluster 0\n", + "Example n. 1344 = (24.0, 306.0): cluster 0\n", + "Example n. 1345 = (24.0, 305.0): cluster 0\n", + "Example n. 1346 = (24.0, 304.0): cluster 0\n", + "Example n. 1347 = (24.0, 303.0): cluster 0\n", + "Example n. 1348 = (24.0, 302.0): cluster 0\n", + "Example n. 1349 = (24.0, 301.0): cluster 0\n", + "Example n. 1350 = (24.0, 300.0): cluster 0\n", + "Example n. 1351 = (24.0, 299.0): cluster 0\n", + "Example n. 1352 = (24.0, 296.0): cluster 0\n", + "Example n. 1353 = (24.0, 295.0): cluster 0\n", + "Example n. 1354 = (24.0, 294.0): cluster 0\n", + "Example n. 1355 = (24.0, 293.0): cluster 0\n", + "Example n. 1356 = (24.0, 292.0): cluster 0\n", + "Example n. 1357 = (24.0, 291.0): cluster 0\n", + "Example n. 1358 = (24.0, 287.0): cluster 0\n", + "Example n. 1359 = (24.0, 286.0): cluster 0\n", + "Example n. 1360 = (24.0, 285.0): cluster 0\n", + "Example n. 1361 = (24.0, 284.0): cluster 0\n", + "Example n. 1362 = (24.0, 283.0): cluster 0\n", + "Example n. 1363 = (24.0, 282.0): cluster 0\n", + "Example n. 1364 = (24.0, 281.0): cluster 0\n", + "Example n. 1365 = (24.0, 280.0): cluster 0\n", + "Example n. 1366 = (24.0, 279.0): cluster 0\n", + "Example n. 1367 = (24.0, 278.0): cluster 0\n", + "Example n. 1368 = (24.0, 277.0): cluster 0\n", + "Example n. 1369 = (24.0, 276.0): cluster 0\n", + "Example n. 1370 = (24.0, 275.0): cluster 0\n", + "Example n. 1371 = (24.0, 274.0): cluster 0\n", + "Example n. 1372 = (24.0, 268.0): cluster 0\n", + "Example n. 1373 = (24.0, 267.0): cluster 0\n", + "Example n. 1374 = (24.0, 266.0): cluster 0\n", + "Example n. 1375 = (24.0, 265.0): cluster 0\n", + "Example n. 1376 = (24.0, 264.0): cluster 0\n", + "Example n. 1377 = (24.0, 262.0): cluster 1\n", + "Example n. 1378 = (24.0, 261.0): cluster 1\n", + "Example n. 1379 = (24.0, 260.0): cluster 1\n", + "Example n. 1380 = (24.0, 259.0): cluster 1\n", + "Example n. 1381 = (24.0, 258.0): cluster 1\n", + "Example n. 1382 = (24.0, 257.0): cluster 1\n", + "Example n. 1383 = (24.0, 254.0): cluster 1\n", + "Example n. 1384 = (24.0, 253.0): cluster 1\n", + "Example n. 1385 = (24.0, 252.0): cluster 1\n", + "Example n. 1386 = (24.0, 251.0): cluster 1\n", + "Example n. 1387 = (24.0, 250.0): cluster 1\n", + "Example n. 1388 = (24.0, 249.0): cluster 1\n", + "Example n. 1389 = (24.0, 248.0): cluster 1\n", + "Example n. 1390 = (24.0, 247.0): cluster 3\n", + "Example n. 1391 = (24.0, 246.0): cluster 3\n", + "Example n. 1392 = (24.0, 245.0): cluster 3\n", + "Example n. 1393 = (24.0, 244.0): cluster 3\n", + "Example n. 1394 = (24.0, 243.0): cluster 3\n", + "Example n. 1395 = (24.0, 242.0): cluster 3\n", + "Example n. 1396 = (24.0, 241.0): cluster 3\n", + "Example n. 1397 = (24.0, 240.0): cluster 3\n", + "Example n. 1398 = (24.0, 239.0): cluster 3\n", + "Example n. 1399 = (24.0, 238.0): cluster 3\n", + "Example n. 1400 = (24.0, 237.0): cluster 3\n", + "Example n. 1401 = (24.0, 236.0): cluster 3\n", + "Example n. 1402 = (24.0, 235.0): cluster 3\n", + "Example n. 1403 = (24.0, 234.0): cluster 3\n", + "Example n. 1404 = (24.0, 233.0): cluster 3\n", + "Example n. 1405 = (24.0, 232.0): cluster 3\n", + "Example n. 1406 = (24.0, 231.0): cluster 3\n", + "Example n. 1407 = (24.0, 230.0): cluster 3\n", + "Example n. 1408 = (24.0, 229.0): cluster 3\n", + "Example n. 1409 = (24.0, 228.0): cluster 3\n", + "Example n. 1410 = (24.0, 227.0): cluster 3\n", + "Example n. 1411 = (24.0, 226.0): cluster 3\n", + "Example n. 1412 = (24.0, 225.0): cluster 3\n", + "Example n. 1413 = (24.0, 224.0): cluster 3\n", + "Example n. 1414 = (24.0, 223.0): cluster 3\n", + "Example n. 1415 = (24.0, 222.0): cluster 3\n", + "Example n. 1416 = (24.0, 221.0): cluster 3\n", + "Example n. 1417 = (24.0, 220.0): cluster 3\n", + "Example n. 1418 = (24.0, 219.0): cluster 3\n", + "Example n. 1419 = (24.0, 218.0): cluster 3\n", + "Example n. 1420 = (24.0, 217.0): cluster 3\n", + "Example n. 1421 = (24.0, 216.0): cluster 3\n", + "Example n. 1422 = (24.0, 214.0): cluster 0\n", + "Example n. 1423 = (24.0, 213.0): cluster 0\n", + "Example n. 1424 = (24.0, 212.0): cluster 0\n", + "Example n. 1425 = (24.0, 211.0): cluster 0\n", + "Example n. 1426 = (24.0, 210.0): cluster 0\n", + "Example n. 1427 = (24.0, 209.0): cluster 0\n", + "Example n. 1428 = (24.0, 208.0): cluster 0\n", + "Example n. 1429 = (24.0, 205.0): cluster 0\n", + "Example n. 1430 = (24.0, 204.0): cluster 0\n", + "Example n. 1431 = (24.0, 203.0): cluster 0\n", + "Example n. 1432 = (24.0, 192.0): cluster 0\n", + "Example n. 1433 = (24.0, 191.0): cluster 0\n", + "Example n. 1434 = (24.0, 190.0): cluster 0\n", + "Example n. 1435 = (24.0, 189.0): cluster 0\n", + "Example n. 1436 = (24.0, 177.0): cluster 0\n", + "Example n. 1437 = (24.0, 176.0): cluster 0\n", + "Example n. 1438 = (24.0, 175.0): cluster 0\n", + "Example n. 1439 = (24.0, 174.0): cluster 0\n", + "Example n. 1440 = (24.0, 145.0): cluster 0\n", + "Example n. 1441 = (24.0, 144.0): cluster 0\n", + "Example n. 1442 = (24.0, 143.0): cluster 0\n", + "Example n. 1443 = (24.0, 142.0): cluster 0\n", + "Example n. 1444 = (24.0, 141.0): cluster 0\n", + "Example n. 1445 = (24.0, 140.0): cluster 0\n", + "Example n. 1446 = (24.0, 131.0): cluster 0\n", + "Example n. 1447 = (24.0, 130.0): cluster 0\n", + "Example n. 1448 = (24.0, 129.0): cluster 0\n", + "Example n. 1449 = (24.0, 128.0): cluster 0\n", + "Example n. 1450 = (24.0, 127.0): cluster 0\n", + "Example n. 1451 = (24.0, 126.0): cluster 0\n", + "Example n. 1452 = (24.0, 125.0): cluster 0\n", + "Example n. 1453 = (24.0, 124.0): cluster 0\n", + "Example n. 1454 = (24.0, 123.0): cluster 1\n", + "Example n. 1455 = (24.0, 122.0): cluster 1\n", + "Example n. 1456 = (24.0, 120.0): cluster 1\n", + "Example n. 1457 = (24.0, 117.0): cluster 1\n", + "Example n. 1458 = (24.0, 116.0): cluster 1\n", + "Example n. 1459 = (25.0, 478.0): cluster 1\n", + "Example n. 1460 = (25.0, 477.0): cluster 1\n", + "Example n. 1461 = (25.0, 476.0): cluster 1\n", + "Example n. 1462 = (25.0, 475.0): cluster 1\n", + "Example n. 1463 = (25.0, 474.0): cluster 1\n", + "Example n. 1464 = (25.0, 473.0): cluster 1\n", + "Example n. 1465 = (25.0, 472.0): cluster 1\n", + "Example n. 1466 = (25.0, 457.0): cluster 3\n", + "Example n. 1467 = (25.0, 456.0): cluster 3\n", + "Example n. 1468 = (25.0, 455.0): cluster 3\n", + "Example n. 1469 = (25.0, 454.0): cluster 3\n", + "Example n. 1470 = (25.0, 453.0): cluster 3\n", + "Example n. 1471 = (25.0, 452.0): cluster 3\n", + "Example n. 1472 = (25.0, 389.0): cluster 3\n", + "Example n. 1473 = (25.0, 388.0): cluster 3\n", + "Example n. 1474 = (25.0, 387.0): cluster 3\n", + "Example n. 1475 = (25.0, 386.0): cluster 3\n", + "Example n. 1476 = (25.0, 385.0): cluster 3\n", + "Example n. 1477 = (25.0, 384.0): cluster 3\n", + "Example n. 1478 = (25.0, 371.0): cluster 3\n", + "Example n. 1479 = (25.0, 370.0): cluster 3\n", + "Example n. 1480 = (25.0, 369.0): cluster 3\n", + "Example n. 1481 = (25.0, 368.0): cluster 3\n", + "Example n. 1482 = (25.0, 367.0): cluster 3\n", + "Example n. 1483 = (25.0, 308.0): cluster 3\n", + "Example n. 1484 = (25.0, 307.0): cluster 3\n", + "Example n. 1485 = (25.0, 306.0): cluster 3\n", + "Example n. 1486 = (25.0, 305.0): cluster 3\n", + "Example n. 1487 = (25.0, 304.0): cluster 3\n", + "Example n. 1488 = (25.0, 303.0): cluster 3\n", + "Example n. 1489 = (25.0, 302.0): cluster 3\n", + "Example n. 1490 = (25.0, 301.0): cluster 3\n", + "Example n. 1491 = (25.0, 300.0): cluster 3\n", + "Example n. 1492 = (25.0, 299.0): cluster 3\n", + "Example n. 1493 = (25.0, 296.0): cluster 3\n", + "Example n. 1494 = (25.0, 295.0): cluster 3\n", + "Example n. 1495 = (25.0, 294.0): cluster 3\n", + "Example n. 1496 = (25.0, 293.0): cluster 3\n", + "Example n. 1497 = (25.0, 292.0): cluster 3\n", + "Example n. 1498 = (25.0, 291.0): cluster 0\n", + "Example n. 1499 = (25.0, 287.0): cluster 0\n", + "Example n. 1500 = (25.0, 286.0): cluster 0\n", + "Example n. 1501 = (25.0, 285.0): cluster 0\n", + "Example n. 1502 = (25.0, 284.0): cluster 0\n", + "Example n. 1503 = (25.0, 283.0): cluster 0\n", + "Example n. 1504 = (25.0, 282.0): cluster 0\n", + "Example n. 1505 = (25.0, 281.0): cluster 0\n", + "Example n. 1506 = (25.0, 280.0): cluster 0\n", + "Example n. 1507 = (25.0, 279.0): cluster 0\n", + "Example n. 1508 = (25.0, 278.0): cluster 0\n", + "Example n. 1509 = (25.0, 277.0): cluster 0\n", + "Example n. 1510 = (25.0, 276.0): cluster 0\n", + "Example n. 1511 = (25.0, 275.0): cluster 0\n", + "Example n. 1512 = (25.0, 274.0): cluster 0\n", + "Example n. 1513 = (25.0, 273.0): cluster 0\n", + "Example n. 1514 = (25.0, 272.0): cluster 0\n", + "Example n. 1515 = (25.0, 271.0): cluster 0\n", + "Example n. 1516 = (25.0, 268.0): cluster 0\n", + "Example n. 1517 = (25.0, 267.0): cluster 0\n", + "Example n. 1518 = (25.0, 266.0): cluster 0\n", + "Example n. 1519 = (25.0, 265.0): cluster 0\n", + "Example n. 1520 = (25.0, 264.0): cluster 0\n", + "Example n. 1521 = (25.0, 262.0): cluster 0\n", + "Example n. 1522 = (25.0, 261.0): cluster 0\n", + "Example n. 1523 = (25.0, 260.0): cluster 0\n", + "Example n. 1524 = (25.0, 259.0): cluster 0\n", + "Example n. 1525 = (25.0, 258.0): cluster 0\n", + "Example n. 1526 = (25.0, 257.0): cluster 0\n", + "Example n. 1527 = (25.0, 253.0): cluster 0\n", + "Example n. 1528 = (25.0, 252.0): cluster 1\n", + "Example n. 1529 = (25.0, 251.0): cluster 1\n", + "Example n. 1530 = (25.0, 250.0): cluster 1\n", + "Example n. 1531 = (25.0, 249.0): cluster 1\n", + "Example n. 1532 = (25.0, 248.0): cluster 1\n", + "Example n. 1533 = (25.0, 247.0): cluster 1\n", + "Example n. 1534 = (25.0, 246.0): cluster 1\n", + "Example n. 1535 = (25.0, 245.0): cluster 1\n", + "Example n. 1536 = (25.0, 244.0): cluster 1\n", + "Example n. 1537 = (25.0, 243.0): cluster 1\n", + "Example n. 1538 = (25.0, 242.0): cluster 1\n", + "Example n. 1539 = (25.0, 241.0): cluster 1\n", + "Example n. 1540 = (25.0, 240.0): cluster 1\n", + "Example n. 1541 = (25.0, 239.0): cluster 1\n", + "Example n. 1542 = (25.0, 238.0): cluster 3\n", + "Example n. 1543 = (25.0, 237.0): cluster 3\n", + "Example n. 1544 = (25.0, 236.0): cluster 3\n", + "Example n. 1545 = (25.0, 234.0): cluster 3\n", + "Example n. 1546 = (25.0, 233.0): cluster 3\n", + "Example n. 1547 = (25.0, 232.0): cluster 3\n", + "Example n. 1548 = (25.0, 231.0): cluster 3\n", + "Example n. 1549 = (25.0, 230.0): cluster 3\n", + "Example n. 1550 = (25.0, 229.0): cluster 3\n", + "Example n. 1551 = (25.0, 228.0): cluster 3\n", + "Example n. 1552 = (25.0, 227.0): cluster 3\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Example n. 1553 = (25.0, 226.0): cluster 3\n", + "Example n. 1554 = (25.0, 225.0): cluster 3\n", + "Example n. 1555 = (25.0, 224.0): cluster 3\n", + "Example n. 1556 = (25.0, 220.0): cluster 3\n", + "Example n. 1557 = (25.0, 219.0): cluster 3\n", + "Example n. 1558 = (25.0, 218.0): cluster 3\n", + "Example n. 1559 = (25.0, 217.0): cluster 3\n", + "Example n. 1560 = (25.0, 216.0): cluster 3\n", + "Example n. 1561 = (25.0, 214.0): cluster 3\n", + "Example n. 1562 = (25.0, 213.0): cluster 3\n", + "Example n. 1563 = (25.0, 212.0): cluster 3\n", + "Example n. 1564 = (25.0, 211.0): cluster 3\n", + "Example n. 1565 = (25.0, 210.0): cluster 3\n", + "Example n. 1566 = (25.0, 209.0): cluster 3\n", + "Example n. 1567 = (25.0, 208.0): cluster 3\n", + "Example n. 1568 = (25.0, 206.0): cluster 3\n", + "Example n. 1569 = (25.0, 205.0): cluster 3\n", + "Example n. 1570 = (25.0, 204.0): cluster 3\n", + "Example n. 1571 = (25.0, 203.0): cluster 3\n", + "Example n. 1572 = (25.0, 202.0): cluster 3\n", + "Example n. 1573 = (25.0, 193.0): cluster 3\n", + "Example n. 1574 = (25.0, 192.0): cluster 0\n", + "Example n. 1575 = (25.0, 191.0): cluster 0\n", + "Example n. 1576 = (25.0, 190.0): cluster 0\n", + "Example n. 1577 = (25.0, 189.0): cluster 0\n", + "Example n. 1578 = (25.0, 188.0): cluster 0\n", + "Example n. 1579 = (25.0, 184.0): cluster 0\n", + "Example n. 1580 = (25.0, 183.0): cluster 0\n", + "Example n. 1581 = (25.0, 182.0): cluster 0\n", + "Example n. 1582 = (25.0, 178.0): cluster 0\n", + "Example n. 1583 = (25.0, 177.0): cluster 0\n", + "Example n. 1584 = (25.0, 176.0): cluster 0\n", + "Example n. 1585 = (25.0, 175.0): cluster 0\n", + "Example n. 1586 = (25.0, 174.0): cluster 0\n", + "Example n. 1587 = (25.0, 167.0): cluster 0\n", + "Example n. 1588 = (25.0, 166.0): cluster 0\n", + "Example n. 1589 = (25.0, 165.0): cluster 0\n", + "Example n. 1590 = (25.0, 164.0): cluster 0\n", + "Example n. 1591 = (25.0, 163.0): cluster 0\n", + "Example n. 1592 = (25.0, 145.0): cluster 0\n", + "Example n. 1593 = (25.0, 144.0): cluster 0\n", + "Example n. 1594 = (25.0, 143.0): cluster 0\n", + "Example n. 1595 = (25.0, 142.0): cluster 0\n", + "Example n. 1596 = (25.0, 141.0): cluster 0\n", + "Example n. 1597 = (25.0, 131.0): cluster 0\n", + "Example n. 1598 = (25.0, 130.0): cluster 0\n", + "Example n. 1599 = (25.0, 129.0): cluster 0\n", + "Example n. 1600 = (25.0, 128.0): cluster 0\n", + "Example n. 1601 = (25.0, 127.0): cluster 0\n", + "Example n. 1602 = (25.0, 124.0): cluster 1\n", + "Example n. 1603 = (25.0, 122.0): cluster 1\n", + "Example n. 1604 = (25.0, 121.0): cluster 1\n", + "Example n. 1605 = (25.0, 120.0): cluster 1\n", + "Example n. 1606 = (25.0, 119.0): cluster 1\n", + "Example n. 1607 = (25.0, 118.0): cluster 1\n", + "Example n. 1608 = (25.0, 117.0): cluster 1\n", + "Example n. 1609 = (25.0, 116.0): cluster 1\n", + "Example n. 1610 = (25.0, 115.0): cluster 1\n", + "Example n. 1611 = (26.0, 478.0): cluster 1\n", + "Example n. 1612 = (26.0, 477.0): cluster 1\n", + "Example n. 1613 = (26.0, 476.0): cluster 1\n", + "Example n. 1614 = (26.0, 475.0): cluster 1\n", + "Example n. 1615 = (26.0, 474.0): cluster 1\n", + "Example n. 1616 = (26.0, 473.0): cluster 1\n", + "Example n. 1617 = (26.0, 472.0): cluster 1\n", + "Example n. 1618 = (26.0, 471.0): cluster 3\n", + "Example n. 1619 = (26.0, 468.0): cluster 3\n", + "Example n. 1620 = (26.0, 467.0): cluster 3\n", + "Example n. 1621 = (26.0, 466.0): cluster 3\n", + "Example n. 1622 = (26.0, 465.0): cluster 3\n", + "Example n. 1623 = (26.0, 457.0): cluster 3\n", + "Example n. 1624 = (26.0, 456.0): cluster 3\n", + "Example n. 1625 = (26.0, 455.0): cluster 3\n", + "Example n. 1626 = (26.0, 454.0): cluster 3\n", + "Example n. 1627 = (26.0, 453.0): cluster 3\n", + "Example n. 1628 = (26.0, 452.0): cluster 3\n", + "Example n. 1629 = (26.0, 389.0): cluster 3\n", + "Example n. 1630 = (26.0, 388.0): cluster 3\n", + "Example n. 1631 = (26.0, 387.0): cluster 3\n", + "Example n. 1632 = (26.0, 386.0): cluster 3\n", + "Example n. 1633 = (26.0, 385.0): cluster 3\n", + "Example n. 1634 = (26.0, 384.0): cluster 3\n", + "Example n. 1635 = (26.0, 371.0): cluster 3\n", + "Example n. 1636 = (26.0, 370.0): cluster 3\n", + "Example n. 1637 = (26.0, 369.0): cluster 3\n", + "Example n. 1638 = (26.0, 368.0): cluster 3\n", + "Example n. 1639 = (26.0, 367.0): cluster 3\n", + "Example n. 1640 = (26.0, 366.0): cluster 3\n", + "Example n. 1641 = (26.0, 307.0): cluster 3\n", + "Example n. 1642 = (26.0, 306.0): cluster 3\n", + "Example n. 1643 = (26.0, 305.0): cluster 3\n", + "Example n. 1644 = (26.0, 303.0): cluster 3\n", + "Example n. 1645 = (26.0, 302.0): cluster 3\n", + "Example n. 1646 = (26.0, 301.0): cluster 3\n", + "Example n. 1647 = (26.0, 300.0): cluster 3\n", + "Example n. 1648 = (26.0, 299.0): cluster 3\n", + "Example n. 1649 = (26.0, 297.0): cluster 0\n", + "Example n. 1650 = (26.0, 296.0): cluster 0\n", + "Example n. 1651 = (26.0, 295.0): cluster 0\n", + "Example n. 1652 = (26.0, 294.0): cluster 0\n", + "Example n. 1653 = (26.0, 293.0): cluster 0\n", + "Example n. 1654 = (26.0, 292.0): cluster 0\n", + "Example n. 1655 = (26.0, 288.0): cluster 0\n", + "Example n. 1656 = (26.0, 287.0): cluster 0\n", + "Example n. 1657 = (26.0, 286.0): cluster 0\n", + "Example n. 1658 = (26.0, 285.0): cluster 0\n", + "Example n. 1659 = (26.0, 284.0): cluster 0\n", + "Example n. 1660 = (26.0, 283.0): cluster 0\n", + "Example n. 1661 = (26.0, 282.0): cluster 0\n", + "Example n. 1662 = (26.0, 281.0): cluster 0\n", + "Example n. 1663 = (26.0, 280.0): cluster 0\n", + "Example n. 1664 = (26.0, 279.0): cluster 0\n", + "Example n. 1665 = (26.0, 278.0): cluster 0\n", + "Example n. 1666 = (26.0, 277.0): cluster 0\n", + "Example n. 1667 = (26.0, 276.0): cluster 0\n", + "Example n. 1668 = (26.0, 275.0): cluster 0\n", + "Example n. 1669 = (26.0, 274.0): cluster 0\n", + "Example n. 1670 = (26.0, 273.0): cluster 0\n", + "Example n. 1671 = (26.0, 272.0): cluster 0\n", + "Example n. 1672 = (26.0, 271.0): cluster 0\n", + "Example n. 1673 = (26.0, 270.0): cluster 0\n", + "Example n. 1674 = (26.0, 269.0): cluster 1\n", + "Example n. 1675 = (26.0, 268.0): cluster 1\n", + "Example n. 1676 = (26.0, 267.0): cluster 1\n", + "Example n. 1677 = (26.0, 266.0): cluster 1\n", + "Example n. 1678 = (26.0, 265.0): cluster 1\n", + "Example n. 1679 = (26.0, 264.0): cluster 1\n", + "Example n. 1680 = (26.0, 261.0): cluster 1\n", + "Example n. 1681 = (26.0, 260.0): cluster 1\n", + "Example n. 1682 = (26.0, 259.0): cluster 1\n", + "Example n. 1683 = (26.0, 258.0): cluster 1\n", + "Example n. 1684 = (26.0, 254.0): cluster 1\n", + "Example n. 1685 = (26.0, 253.0): cluster 1\n", + "Example n. 1686 = (26.0, 252.0): cluster 1\n", + "Example n. 1687 = (26.0, 251.0): cluster 1\n", + "Example n. 1688 = (26.0, 250.0): cluster 1\n", + "Example n. 1689 = (26.0, 249.0): cluster 1\n", + "Example n. 1690 = (26.0, 248.0): cluster 1\n", + "Example n. 1691 = (26.0, 247.0): cluster 1\n", + "Example n. 1692 = (26.0, 246.0): cluster 1\n", + "Example n. 1693 = (26.0, 245.0): cluster 1\n", + "Example n. 1694 = (26.0, 244.0): cluster 1\n", + "Example n. 1695 = (26.0, 243.0): cluster 1\n", + "Example n. 1696 = (26.0, 242.0): cluster 1\n", + "Example n. 1697 = (26.0, 241.0): cluster 3\n", + "Example n. 1698 = (26.0, 240.0): cluster 3\n", + "Example n. 1699 = (26.0, 239.0): cluster 3\n", + "Example n. 1700 = (26.0, 238.0): cluster 3\n", + "Example n. 1701 = (26.0, 237.0): cluster 3\n", + "Example n. 1702 = (26.0, 236.0): cluster 3\n", + "Example n. 1703 = (26.0, 234.0): cluster 3\n", + "Example n. 1704 = (26.0, 233.0): cluster 3\n", + "Example n. 1705 = (26.0, 232.0): cluster 3\n", + "Example n. 1706 = (26.0, 231.0): cluster 3\n", + "Example n. 1707 = (26.0, 230.0): cluster 3\n", + "Example n. 1708 = (26.0, 229.0): cluster 3\n", + "Example n. 1709 = (26.0, 228.0): cluster 3\n", + "Example n. 1710 = (26.0, 227.0): cluster 3\n", + "Example n. 1711 = (26.0, 226.0): cluster 3\n", + "Example n. 1712 = (26.0, 225.0): cluster 3\n", + "Example n. 1713 = (26.0, 224.0): cluster 3\n", + "Example n. 1714 = (26.0, 220.0): cluster 3\n", + "Example n. 1715 = (26.0, 219.0): cluster 3\n", + "Example n. 1716 = (26.0, 218.0): cluster 3\n", + "Example n. 1717 = (26.0, 217.0): cluster 3\n", + "Example n. 1718 = (26.0, 216.0): cluster 3\n", + "Example n. 1719 = (26.0, 213.0): cluster 3\n", + "Example n. 1720 = (26.0, 212.0): cluster 3\n", + "Example n. 1721 = (26.0, 211.0): cluster 3\n", + "Example n. 1722 = (26.0, 210.0): cluster 3\n", + "Example n. 1723 = (26.0, 209.0): cluster 3\n", + "Example n. 1724 = (26.0, 208.0): cluster 3\n", + "Example n. 1725 = (26.0, 207.0): cluster 3\n", + "Example n. 1726 = (26.0, 206.0): cluster 3\n", + "Example n. 1727 = (26.0, 205.0): cluster 3\n", + "Example n. 1728 = (26.0, 204.0): cluster 3\n", + "Example n. 1729 = (26.0, 203.0): cluster 3\n", + "Example n. 1730 = (26.0, 202.0): cluster 3\n", + "Example n. 1731 = (26.0, 193.0): cluster 3\n", + "Example n. 1732 = (26.0, 192.0): cluster 0\n", + "Example n. 1733 = (26.0, 191.0): cluster 0\n", + "Example n. 1734 = (26.0, 190.0): cluster 0\n", + "Example n. 1735 = (26.0, 189.0): cluster 0\n", + "Example n. 1736 = (26.0, 188.0): cluster 0\n", + "Example n. 1737 = (26.0, 187.0): cluster 0\n", + "Example n. 1738 = (26.0, 185.0): cluster 0\n", + "Example n. 1739 = (26.0, 184.0): cluster 0\n", + "Example n. 1740 = (26.0, 183.0): cluster 0\n", + "Example n. 1741 = (26.0, 182.0): cluster 0\n", + "Example n. 1742 = (26.0, 181.0): cluster 0\n", + "Example n. 1743 = (26.0, 178.0): cluster 0\n", + "Example n. 1744 = (26.0, 177.0): cluster 0\n", + "Example n. 1745 = (26.0, 176.0): cluster 0\n", + "Example n. 1746 = (26.0, 175.0): cluster 0\n", + "Example n. 1747 = (26.0, 174.0): cluster 0\n", + "Example n. 1748 = (26.0, 173.0): cluster 0\n", + "Example n. 1749 = (26.0, 167.0): cluster 0\n", + "Example n. 1750 = (26.0, 166.0): cluster 0\n", + "Example n. 1751 = (26.0, 165.0): cluster 1\n", + "Example n. 1752 = (26.0, 164.0): cluster 1\n", + "Example n. 1753 = (26.0, 163.0): cluster 1\n", + "Example n. 1754 = (26.0, 162.0): cluster 1\n", + "Example n. 1755 = (26.0, 144.0): cluster 1\n", + "Example n. 1756 = (26.0, 143.0): cluster 1\n", + "Example n. 1757 = (26.0, 142.0): cluster 1\n", + "Example n. 1758 = (26.0, 141.0): cluster 1\n", + "Example n. 1759 = (26.0, 131.0): cluster 1\n", + "Example n. 1760 = (26.0, 130.0): cluster 1\n", + "Example n. 1761 = (26.0, 129.0): cluster 1\n", + "Example n. 1762 = (26.0, 128.0): cluster 1\n", + "Example n. 1763 = (26.0, 127.0): cluster 1\n", + "Example n. 1764 = (26.0, 122.0): cluster 1\n", + "Example n. 1765 = (26.0, 121.0): cluster 1\n", + "Example n. 1766 = (26.0, 120.0): cluster 1\n", + "Example n. 1767 = (26.0, 119.0): cluster 1\n", + "Example n. 1768 = (26.0, 118.0): cluster 1\n", + "Example n. 1769 = (26.0, 117.0): cluster 1\n", + "Example n. 1770 = (26.0, 116.0): cluster 1\n", + "Example n. 1771 = (26.0, 115.0): cluster 1\n", + "Example n. 1772 = (26.0, 114.0): cluster 3\n", + "Example n. 1773 = (27.0, 501.0): cluster 3\n", + "Example n. 1774 = (27.0, 500.0): cluster 3\n", + "Example n. 1775 = (27.0, 481.0): cluster 3\n", + "Example n. 1776 = (27.0, 480.0): cluster 3\n", + "Example n. 1777 = (27.0, 479.0): cluster 3\n", + "Example n. 1778 = (27.0, 478.0): cluster 3\n", + "Example n. 1779 = (27.0, 477.0): cluster 3\n", + "Example n. 1780 = (27.0, 476.0): cluster 3\n", + "Example n. 1781 = (27.0, 475.0): cluster 3\n", + "Example n. 1782 = (27.0, 474.0): cluster 3\n", + "Example n. 1783 = (27.0, 473.0): cluster 3\n", + "Example n. 1784 = (27.0, 472.0): cluster 3\n", + "Example n. 1785 = (27.0, 471.0): cluster 3\n", + "Example n. 1786 = (27.0, 470.0): cluster 3\n", + "Example n. 1787 = (27.0, 469.0): cluster 3\n", + "Example n. 1788 = (27.0, 468.0): cluster 3\n", + "Example n. 1789 = (27.0, 467.0): cluster 3\n", + "Example n. 1790 = (27.0, 466.0): cluster 3\n", + "Example n. 1791 = (27.0, 465.0): cluster 3\n", + "Example n. 1792 = (27.0, 464.0): cluster 3\n", + "Example n. 1793 = (27.0, 456.0): cluster 3\n", + "Example n. 1794 = (27.0, 455.0): cluster 3\n", + "Example n. 1795 = (27.0, 454.0): cluster 3\n", + "Example n. 1796 = (27.0, 453.0): cluster 3\n", + "Example n. 1797 = (27.0, 389.0): cluster 3\n", + "Example n. 1798 = (27.0, 388.0): cluster 3\n", + "Example n. 1799 = (27.0, 387.0): cluster 3\n", + "Example n. 1800 = (27.0, 386.0): cluster 3\n", + "Example n. 1801 = (27.0, 385.0): cluster 3\n", + "Example n. 1802 = (27.0, 384.0): cluster 3\n", + "Example n. 1803 = (27.0, 370.0): cluster 0\n", + "Example n. 1804 = (27.0, 369.0): cluster 0\n", + "Example n. 1805 = (27.0, 368.0): cluster 0\n", + "Example n. 1806 = (27.0, 367.0): cluster 0\n", + "Example n. 1807 = (27.0, 366.0): cluster 0\n", + "Example n. 1808 = (27.0, 302.0): cluster 0\n", + "Example n. 1809 = (27.0, 301.0): cluster 0\n", + "Example n. 1810 = (27.0, 297.0): cluster 0\n", + "Example n. 1811 = (27.0, 296.0): cluster 0\n", + "Example n. 1812 = (27.0, 295.0): cluster 0\n", + "Example n. 1813 = (27.0, 294.0): cluster 0\n", + "Example n. 1814 = (27.0, 293.0): cluster 0\n", + "Example n. 1815 = (27.0, 290.0): cluster 0\n", + "Example n. 1816 = (27.0, 289.0): cluster 0\n", + "Example n. 1817 = (27.0, 288.0): cluster 0\n", + "Example n. 1818 = (27.0, 287.0): cluster 0\n", + "Example n. 1819 = (27.0, 286.0): cluster 1\n", + "Example n. 1820 = (27.0, 285.0): cluster 1\n", + "Example n. 1821 = (27.0, 284.0): cluster 1\n", + "Example n. 1822 = (27.0, 283.0): cluster 1\n", + "Example n. 1823 = (27.0, 282.0): cluster 1\n", + "Example n. 1824 = (27.0, 281.0): cluster 1\n", + "Example n. 1825 = (27.0, 280.0): cluster 1\n", + "Example n. 1826 = (27.0, 279.0): cluster 1\n", + "Example n. 1827 = (27.0, 278.0): cluster 1\n", + "Example n. 1828 = (27.0, 277.0): cluster 1\n", + "Example n. 1829 = (27.0, 276.0): cluster 1\n", + "Example n. 1830 = (27.0, 275.0): cluster 1\n", + "Example n. 1831 = (27.0, 274.0): cluster 1\n", + "Example n. 1832 = (27.0, 273.0): cluster 1\n", + "Example n. 1833 = (27.0, 272.0): cluster 1\n", + "Example n. 1834 = (27.0, 271.0): cluster 1\n", + "Example n. 1835 = (27.0, 270.0): cluster 1\n", + "Example n. 1836 = (27.0, 269.0): cluster 1\n", + "Example n. 1837 = (27.0, 268.0): cluster 1\n", + "Example n. 1838 = (27.0, 267.0): cluster 1\n", + "Example n. 1839 = (27.0, 266.0): cluster 1\n", + "Example n. 1840 = (27.0, 265.0): cluster 1\n", + "Example n. 1841 = (27.0, 264.0): cluster 1\n", + "Example n. 1842 = (27.0, 260.0): cluster 1\n", + "Example n. 1843 = (27.0, 259.0): cluster 3\n", + "Example n. 1844 = (27.0, 254.0): cluster 3\n", + "Example n. 1845 = (27.0, 253.0): cluster 3\n", + "Example n. 1846 = (27.0, 252.0): cluster 3\n", + "Example n. 1847 = (27.0, 251.0): cluster 3\n", + "Example n. 1848 = (27.0, 250.0): cluster 3\n", + "Example n. 1849 = (27.0, 249.0): cluster 3\n", + "Example n. 1850 = (27.0, 248.0): cluster 3\n", + "Example n. 1851 = (27.0, 247.0): cluster 3\n", + "Example n. 1852 = (27.0, 246.0): cluster 3\n", + "Example n. 1853 = (27.0, 245.0): cluster 3\n", + "Example n. 1854 = (27.0, 244.0): cluster 3\n", + "Example n. 1855 = (27.0, 243.0): cluster 3\n", + "Example n. 1856 = (27.0, 242.0): cluster 3\n", + "Example n. 1857 = (27.0, 241.0): cluster 3\n", + "Example n. 1858 = (27.0, 240.0): cluster 3\n", + "Example n. 1859 = (27.0, 239.0): cluster 3\n", + "Example n. 1860 = (27.0, 238.0): cluster 3\n", + "Example n. 1861 = (27.0, 237.0): cluster 3\n", + "Example n. 1862 = (27.0, 236.0): cluster 3\n", + "Example n. 1863 = (27.0, 235.0): cluster 3\n", + "Example n. 1864 = (27.0, 234.0): cluster 3\n", + "Example n. 1865 = (27.0, 233.0): cluster 3\n", + "Example n. 1866 = (27.0, 232.0): cluster 3\n", + "Example n. 1867 = (27.0, 231.0): cluster 3\n", + "Example n. 1868 = (27.0, 230.0): cluster 3\n", + "Example n. 1869 = (27.0, 229.0): cluster 3\n", + "Example n. 1870 = (27.0, 226.0): cluster 3\n", + "Example n. 1871 = (27.0, 220.0): cluster 3\n", + "Example n. 1872 = (27.0, 219.0): cluster 3\n", + "Example n. 1873 = (27.0, 218.0): cluster 3\n", + "Example n. 1874 = (27.0, 217.0): cluster 0\n", + "Example n. 1875 = (27.0, 216.0): cluster 0\n", + "Example n. 1876 = (27.0, 215.0): cluster 0\n", + "Example n. 1877 = (27.0, 212.0): cluster 0\n", + "Example n. 1878 = (27.0, 211.0): cluster 0\n", + "Example n. 1879 = (27.0, 210.0): cluster 0\n", + "Example n. 1880 = (27.0, 209.0): cluster 0\n", + "Example n. 1881 = (27.0, 208.0): cluster 0\n", + "Example n. 1882 = (27.0, 206.0): cluster 0\n", + "Example n. 1883 = (27.0, 205.0): cluster 0\n", + "Example n. 1884 = (27.0, 204.0): cluster 0\n", + "Example n. 1885 = (27.0, 203.0): cluster 0\n", + "Example n. 1886 = (27.0, 202.0): cluster 0\n", + "Example n. 1887 = (27.0, 201.0): cluster 0\n", + "Example n. 1888 = (27.0, 192.0): cluster 0\n", + "Example n. 1889 = (27.0, 191.0): cluster 0\n", + "Example n. 1890 = (27.0, 190.0): cluster 1\n", + "Example n. 1891 = (27.0, 189.0): cluster 1\n", + "Example n. 1892 = (27.0, 188.0): cluster 1\n", + "Example n. 1893 = (27.0, 187.0): cluster 1\n", + "Example n. 1894 = (27.0, 186.0): cluster 1\n", + "Example n. 1895 = (27.0, 185.0): cluster 1\n", + "Example n. 1896 = (27.0, 184.0): cluster 1\n", + "Example n. 1897 = (27.0, 183.0): cluster 1\n", + "Example n. 1898 = (27.0, 182.0): cluster 1\n", + "Example n. 1899 = (27.0, 181.0): cluster 1\n", + "Example n. 1900 = (27.0, 178.0): cluster 1\n", + "Example n. 1901 = (27.0, 177.0): cluster 1\n", + "Example n. 1902 = (27.0, 176.0): cluster 1\n", + "Example n. 1903 = (27.0, 175.0): cluster 1\n", + "Example n. 1904 = (27.0, 174.0): cluster 1\n", + "Example n. 1905 = (27.0, 173.0): cluster 1\n", + "Example n. 1906 = (27.0, 168.0): cluster 1\n", + "Example n. 1907 = (27.0, 167.0): cluster 1\n", + "Example n. 1908 = (27.0, 166.0): cluster 1\n", + "Example n. 1909 = (27.0, 165.0): cluster 1\n", + "Example n. 1910 = (27.0, 164.0): cluster 1\n", + "Example n. 1911 = (27.0, 163.0): cluster 1\n", + "Example n. 1912 = (27.0, 162.0): cluster 1\n", + "Example n. 1913 = (27.0, 161.0): cluster 1\n", + "Example n. 1914 = (27.0, 130.0): cluster 3\n", + "Example n. 1915 = (27.0, 129.0): cluster 3\n", + "Example n. 1916 = (27.0, 128.0): cluster 3\n", + "Example n. 1917 = (27.0, 122.0): cluster 3\n", + "Example n. 1918 = (27.0, 121.0): cluster 3\n", + "Example n. 1919 = (27.0, 120.0): cluster 3\n", + "Example n. 1920 = (27.0, 119.0): cluster 3\n", + "Example n. 1921 = (27.0, 118.0): cluster 3\n", + "Example n. 1922 = (27.0, 117.0): cluster 3\n", + "Example n. 1923 = (27.0, 116.0): cluster 3\n", + "Example n. 1924 = (27.0, 115.0): cluster 3\n", + "Example n. 1925 = (27.0, 114.0): cluster 3\n", + "Example n. 1926 = (28.0, 502.0): cluster 3\n", + "Example n. 1927 = (28.0, 501.0): cluster 3\n", + "Example n. 1928 = (28.0, 500.0): cluster 3\n", + "Example n. 1929 = (28.0, 499.0): cluster 3\n", + "Example n. 1930 = (28.0, 498.0): cluster 3\n", + "Example n. 1931 = (28.0, 483.0): cluster 3\n", + "Example n. 1932 = (28.0, 482.0): cluster 3\n", + "Example n. 1933 = (28.0, 481.0): cluster 3\n", + "Example n. 1934 = (28.0, 480.0): cluster 3\n", + "Example n. 1935 = (28.0, 479.0): cluster 3\n", + "Example n. 1936 = (28.0, 478.0): cluster 3\n", + "Example n. 1937 = (28.0, 477.0): cluster 3\n", + "Example n. 1938 = (28.0, 476.0): cluster 3\n", + "Example n. 1939 = (28.0, 475.0): cluster 3\n", + "Example n. 1940 = (28.0, 474.0): cluster 3\n", + "Example n. 1941 = (28.0, 473.0): cluster 3\n", + "Example n. 1942 = (28.0, 472.0): cluster 3\n", + "Example n. 1943 = (28.0, 471.0): cluster 3\n", + "Example n. 1944 = (28.0, 470.0): cluster 3\n", + "Example n. 1945 = (28.0, 469.0): cluster 0\n", + "Example n. 1946 = (28.0, 468.0): cluster 0\n", + "Example n. 1947 = (28.0, 467.0): cluster 0\n", + "Example n. 1948 = (28.0, 466.0): cluster 0\n", + "Example n. 1949 = (28.0, 465.0): cluster 0\n", + "Example n. 1950 = (28.0, 464.0): cluster 0\n", + "Example n. 1951 = (28.0, 455.0): cluster 0\n", + "Example n. 1952 = (28.0, 454.0): cluster 0\n", + "Example n. 1953 = (28.0, 388.0): cluster 0\n", + "Example n. 1954 = (28.0, 387.0): cluster 0\n", + "Example n. 1955 = (28.0, 386.0): cluster 0\n", + "Example n. 1956 = (28.0, 385.0): cluster 0\n", + "Example n. 1957 = (28.0, 370.0): cluster 0\n", + "Example n. 1958 = (28.0, 369.0): cluster 0\n", + "Example n. 1959 = (28.0, 368.0): cluster 0\n", + "Example n. 1960 = (28.0, 367.0): cluster 0\n", + "Example n. 1961 = (28.0, 366.0): cluster 0\n", + "Example n. 1962 = (28.0, 297.0): cluster 0\n", + "Example n. 1963 = (28.0, 296.0): cluster 0\n", + "Example n. 1964 = (28.0, 295.0): cluster 0\n", + "Example n. 1965 = (28.0, 294.0): cluster 1\n", + "Example n. 1966 = (28.0, 293.0): cluster 1\n", + "Example n. 1967 = (28.0, 291.0): cluster 1\n", + "Example n. 1968 = (28.0, 290.0): cluster 1\n", + "Example n. 1969 = (28.0, 289.0): cluster 1\n", + "Example n. 1970 = (28.0, 288.0): cluster 1\n", + "Example n. 1971 = (28.0, 287.0): cluster 1\n", + "Example n. 1972 = (28.0, 286.0): cluster 1\n", + "Example n. 1973 = (28.0, 285.0): cluster 1\n", + "Example n. 1974 = (28.0, 284.0): cluster 1\n", + "Example n. 1975 = (28.0, 283.0): cluster 1\n", + "Example n. 1976 = (28.0, 282.0): cluster 1\n", + "Example n. 1977 = (28.0, 281.0): cluster 1\n", + "Example n. 1978 = (28.0, 280.0): cluster 1\n", + "Example n. 1979 = (28.0, 279.0): cluster 1\n", + "Example n. 1980 = (28.0, 278.0): cluster 1\n", + "Example n. 1981 = (28.0, 277.0): cluster 1\n", + "Example n. 1982 = (28.0, 276.0): cluster 1\n", + "Example n. 1983 = (28.0, 275.0): cluster 1\n", + "Example n. 1984 = (28.0, 274.0): cluster 1\n", + "Example n. 1985 = (28.0, 273.0): cluster 1\n", + "Example n. 1986 = (28.0, 272.0): cluster 1\n", + "Example n. 1987 = (28.0, 271.0): cluster 1\n", + "Example n. 1988 = (28.0, 270.0): cluster 1\n", + "Example n. 1989 = (28.0, 269.0): cluster 3\n", + "Example n. 1990 = (28.0, 268.0): cluster 3\n", + "Example n. 1991 = (28.0, 267.0): cluster 3\n", + "Example n. 1992 = (28.0, 266.0): cluster 3\n", + "Example n. 1993 = (28.0, 265.0): cluster 3\n", + "Example n. 1994 = (28.0, 264.0): cluster 3\n", + "Example n. 1995 = (28.0, 257.0): cluster 3\n", + "Example n. 1996 = (28.0, 256.0): cluster 3\n", + "Example n. 1997 = (28.0, 255.0): cluster 3\n", + "Example n. 1998 = (28.0, 254.0): cluster 3\n", + "Example n. 1999 = (28.0, 253.0): cluster 3\n", + "Example n. 2000 = (28.0, 252.0): cluster 3\n", + "Example n. 2001 = (28.0, 251.0): cluster 3\n", + "Example n. 2002 = (28.0, 250.0): cluster 3\n", + "Example n. 2003 = (28.0, 249.0): cluster 3\n", + "Example n. 2004 = (28.0, 248.0): cluster 3\n", + "Example n. 2005 = (28.0, 247.0): cluster 3\n", + "Example n. 2006 = (28.0, 246.0): cluster 3\n", + "Example n. 2007 = (28.0, 245.0): cluster 3\n", + "Example n. 2008 = (28.0, 244.0): cluster 3\n", + "Example n. 2009 = (28.0, 243.0): cluster 3\n", + "Example n. 2010 = (28.0, 242.0): cluster 3\n", + "Example n. 2011 = (28.0, 241.0): cluster 3\n", + "Example n. 2012 = (28.0, 240.0): cluster 3\n", + "Example n. 2013 = (28.0, 239.0): cluster 3\n", + "Example n. 2014 = (28.0, 238.0): cluster 3\n", + "Example n. 2015 = (28.0, 237.0): cluster 3\n", + "Example n. 2016 = (28.0, 236.0): cluster 3\n", + "Example n. 2017 = (28.0, 235.0): cluster 3\n", + "Example n. 2018 = (28.0, 234.0): cluster 3\n", + "Example n. 2019 = (28.0, 233.0): cluster 3\n", + "Example n. 2020 = (28.0, 232.0): cluster 3\n", + "Example n. 2021 = (28.0, 231.0): cluster 3\n", + "Example n. 2022 = (28.0, 230.0): cluster 3\n", + "Example n. 2023 = (28.0, 229.0): cluster 3\n", + "Example n. 2024 = (28.0, 228.0): cluster 0\n", + "Example n. 2025 = (28.0, 227.0): cluster 0\n", + "Example n. 2026 = (28.0, 226.0): cluster 0\n", + "Example n. 2027 = (28.0, 225.0): cluster 0\n", + "Example n. 2028 = (28.0, 224.0): cluster 0\n", + "Example n. 2029 = (28.0, 220.0): cluster 0\n", + "Example n. 2030 = (28.0, 219.0): cluster 0\n", + "Example n. 2031 = (28.0, 218.0): cluster 0\n", + "Example n. 2032 = (28.0, 217.0): cluster 0\n", + "Example n. 2033 = (28.0, 216.0): cluster 0\n", + "Example n. 2034 = (28.0, 215.0): cluster 0\n", + "Example n. 2035 = (28.0, 214.0): cluster 0\n", + "Example n. 2036 = (28.0, 211.0): cluster 0\n", + "Example n. 2037 = (28.0, 210.0): cluster 0\n", + "Example n. 2038 = (28.0, 209.0): cluster 0\n", + "Example n. 2039 = (28.0, 206.0): cluster 0\n", + "Example n. 2040 = (28.0, 205.0): cluster 0\n", + "Example n. 2041 = (28.0, 204.0): cluster 0\n", + "Example n. 2042 = (28.0, 203.0): cluster 0\n", + "Example n. 2043 = (28.0, 202.0): cluster 0\n", + "Example n. 2044 = (28.0, 201.0): cluster 1\n", + "Example n. 2045 = (28.0, 200.0): cluster 1\n", + "Example n. 2046 = (28.0, 199.0): cluster 1\n", + "Example n. 2047 = (28.0, 196.0): cluster 1\n", + "Example n. 2048 = (28.0, 193.0): cluster 1\n", + "Example n. 2049 = (28.0, 192.0): cluster 1\n", + "Example n. 2050 = (28.0, 191.0): cluster 1\n", + "Example n. 2051 = (28.0, 190.0): cluster 1\n", + "Example n. 2052 = (28.0, 189.0): cluster 1\n", + "Example n. 2053 = (28.0, 188.0): cluster 1\n", + "Example n. 2054 = (28.0, 187.0): cluster 1\n", + "Example n. 2055 = (28.0, 186.0): cluster 1\n", + "Example n. 2056 = (28.0, 185.0): cluster 1\n", + "Example n. 2057 = (28.0, 184.0): cluster 1\n", + "Example n. 2058 = (28.0, 183.0): cluster 1\n", + "Example n. 2059 = (28.0, 182.0): cluster 1\n", + "Example n. 2060 = (28.0, 181.0): cluster 3\n", + "Example n. 2061 = (28.0, 177.0): cluster 3\n", + "Example n. 2062 = (28.0, 176.0): cluster 3\n", + "Example n. 2063 = (28.0, 175.0): cluster 3\n", + "Example n. 2064 = (28.0, 174.0): cluster 3\n", + "Example n. 2065 = (28.0, 168.0): cluster 3\n", + "Example n. 2066 = (28.0, 167.0): cluster 3\n", + "Example n. 2067 = (28.0, 166.0): cluster 3\n", + "Example n. 2068 = (28.0, 165.0): cluster 3\n", + "Example n. 2069 = (28.0, 164.0): cluster 3\n", + "Example n. 2070 = (28.0, 163.0): cluster 3\n", + "Example n. 2071 = (28.0, 162.0): cluster 3\n", + "Example n. 2072 = (28.0, 161.0): cluster 3\n", + "Example n. 2073 = (28.0, 122.0): cluster 3\n", + "Example n. 2074 = (28.0, 121.0): cluster 3\n", + "Example n. 2075 = (28.0, 120.0): cluster 3\n", + "Example n. 2076 = (28.0, 119.0): cluster 3\n", + "Example n. 2077 = (28.0, 118.0): cluster 3\n", + "Example n. 2078 = (28.0, 117.0): cluster 3\n", + "Example n. 2079 = (28.0, 116.0): cluster 3\n", + "Example n. 2080 = (28.0, 115.0): cluster 3\n", + "Example n. 2081 = (28.0, 114.0): cluster 3\n", + "Example n. 2082 = (29.0, 503.0): cluster 3\n", + "Example n. 2083 = (29.0, 502.0): cluster 3\n", + "Example n. 2084 = (29.0, 501.0): cluster 3\n", + "Example n. 2085 = (29.0, 500.0): cluster 3\n", + "Example n. 2086 = (29.0, 499.0): cluster 3\n", + "Example n. 2087 = (29.0, 498.0): cluster 3\n", + "Example n. 2088 = (29.0, 484.0): cluster 3\n", + "Example n. 2089 = (29.0, 483.0): cluster 3\n", + "Example n. 2090 = (29.0, 482.0): cluster 3\n", + "Example n. 2091 = (29.0, 481.0): cluster 3\n", + "Example n. 2092 = (29.0, 480.0): cluster 0\n", + "Example n. 2093 = (29.0, 479.0): cluster 0\n", + "Example n. 2094 = (29.0, 478.0): cluster 0\n", + "Example n. 2095 = (29.0, 477.0): cluster 0\n", + "Example n. 2096 = (29.0, 476.0): cluster 0\n", + "Example n. 2097 = (29.0, 475.0): cluster 0\n", + "Example n. 2098 = (29.0, 474.0): cluster 0\n", + "Example n. 2099 = (29.0, 473.0): cluster 0\n", + "Example n. 2100 = (29.0, 472.0): cluster 0\n", + "Example n. 2101 = (29.0, 471.0): cluster 0\n", + "Example n. 2102 = (29.0, 470.0): cluster 0\n", + "Example n. 2103 = (29.0, 469.0): cluster 0\n", + "Example n. 2104 = (29.0, 468.0): cluster 0\n", + "Example n. 2105 = (29.0, 467.0): cluster 0\n", + "Example n. 2106 = (29.0, 466.0): cluster 0\n", + "Example n. 2107 = (29.0, 465.0): cluster 0\n", + "Example n. 2108 = (29.0, 464.0): cluster 0\n", + "Example n. 2109 = (29.0, 463.0): cluster 1\n", + "Example n. 2110 = (29.0, 370.0): cluster 1\n", + "Example n. 2111 = (29.0, 369.0): cluster 1\n", + "Example n. 2112 = (29.0, 368.0): cluster 1\n", + "Example n. 2113 = (29.0, 367.0): cluster 1\n", + "Example n. 2114 = (29.0, 366.0): cluster 1\n", + "Example n. 2115 = (29.0, 297.0): cluster 1\n", + "Example n. 2116 = (29.0, 296.0): cluster 1\n", + "Example n. 2117 = (29.0, 295.0): cluster 1\n", + "Example n. 2118 = (29.0, 294.0): cluster 1\n", + "Example n. 2119 = (29.0, 293.0): cluster 1\n", + "Example n. 2120 = (29.0, 292.0): cluster 1\n", + "Example n. 2121 = (29.0, 291.0): cluster 1\n", + "Example n. 2122 = (29.0, 290.0): cluster 1\n", + "Example n. 2123 = (29.0, 289.0): cluster 1\n", + "Example n. 2124 = (29.0, 288.0): cluster 1\n", + "Example n. 2125 = (29.0, 287.0): cluster 1\n", + "Example n. 2126 = (29.0, 286.0): cluster 1\n", + "Example n. 2127 = (29.0, 285.0): cluster 1\n", + "Example n. 2128 = (29.0, 284.0): cluster 3\n", + "Example n. 2129 = (29.0, 280.0): cluster 3\n", + "Example n. 2130 = (29.0, 279.0): cluster 3\n", + "Example n. 2131 = (29.0, 278.0): cluster 3\n", + "Example n. 2132 = (29.0, 277.0): cluster 3\n", + "Example n. 2133 = (29.0, 276.0): cluster 3\n", + "Example n. 2134 = (29.0, 274.0): cluster 3\n", + "Example n. 2135 = (29.0, 273.0): cluster 3\n", + "Example n. 2136 = (29.0, 272.0): cluster 3\n", + "Example n. 2137 = (29.0, 271.0): cluster 3\n", + "Example n. 2138 = (29.0, 270.0): cluster 3\n", + "Example n. 2139 = (29.0, 269.0): cluster 3\n", + "Example n. 2140 = (29.0, 268.0): cluster 3\n", + "Example n. 2141 = (29.0, 267.0): cluster 3\n", + "Example n. 2142 = (29.0, 266.0): cluster 3\n", + "Example n. 2143 = (29.0, 265.0): cluster 3\n", + "Example n. 2144 = (29.0, 264.0): cluster 3\n", + "Example n. 2145 = (29.0, 263.0): cluster 3\n", + "Example n. 2146 = (29.0, 262.0): cluster 3\n", + "Example n. 2147 = (29.0, 261.0): cluster 3\n", + "Example n. 2148 = (29.0, 260.0): cluster 3\n", + "Example n. 2149 = (29.0, 259.0): cluster 3\n", + "Example n. 2150 = (29.0, 258.0): cluster 3\n", + "Example n. 2151 = (29.0, 257.0): cluster 3\n", + "Example n. 2152 = (29.0, 256.0): cluster 3\n", + "Example n. 2153 = (29.0, 255.0): cluster 3\n", + "Example n. 2154 = (29.0, 254.0): cluster 3\n", + "Example n. 2155 = (29.0, 253.0): cluster 3\n", + "Example n. 2156 = (29.0, 252.0): cluster 3\n", + "Example n. 2157 = (29.0, 251.0): cluster 3\n", + "Example n. 2158 = (29.0, 250.0): cluster 3\n", + "Example n. 2159 = (29.0, 249.0): cluster 3\n", + "Example n. 2160 = (29.0, 248.0): cluster 0\n", + "Example n. 2161 = (29.0, 247.0): cluster 0\n", + "Example n. 2162 = (29.0, 246.0): cluster 0\n", + "Example n. 2163 = (29.0, 245.0): cluster 0\n", + "Example n. 2164 = (29.0, 244.0): cluster 0\n", + "Example n. 2165 = (29.0, 243.0): cluster 0\n", + "Example n. 2166 = (29.0, 242.0): cluster 0\n", + "Example n. 2167 = (29.0, 241.0): cluster 0\n", + "Example n. 2168 = (29.0, 240.0): cluster 0\n", + "Example n. 2169 = (29.0, 239.0): cluster 0\n", + "Example n. 2170 = (29.0, 238.0): cluster 0\n", + "Example n. 2171 = (29.0, 237.0): cluster 0\n", + "Example n. 2172 = (29.0, 236.0): cluster 0\n", + "Example n. 2173 = (29.0, 235.0): cluster 0\n", + "Example n. 2174 = (29.0, 234.0): cluster 1\n", + "Example n. 2175 = (29.0, 233.0): cluster 1\n", + "Example n. 2176 = (29.0, 232.0): cluster 1\n", + "Example n. 2177 = (29.0, 231.0): cluster 1\n", + "Example n. 2178 = (29.0, 230.0): cluster 1\n", + "Example n. 2179 = (29.0, 229.0): cluster 1\n", + "Example n. 2180 = (29.0, 228.0): cluster 1\n", + "Example n. 2181 = (29.0, 227.0): cluster 1\n", + "Example n. 2182 = (29.0, 226.0): cluster 1\n", + "Example n. 2183 = (29.0, 225.0): cluster 1\n", + "Example n. 2184 = (29.0, 224.0): cluster 1\n", + "Example n. 2185 = (29.0, 218.0): cluster 1\n", + "Example n. 2186 = (29.0, 217.0): cluster 1\n", + "Example n. 2187 = (29.0, 216.0): cluster 1\n", + "Example n. 2188 = (29.0, 215.0): cluster 1\n", + "Example n. 2189 = (29.0, 214.0): cluster 1\n", + "Example n. 2190 = (29.0, 205.0): cluster 1\n", + "Example n. 2191 = (29.0, 204.0): cluster 1\n", + "Example n. 2192 = (29.0, 203.0): cluster 1\n", + "Example n. 2193 = (29.0, 202.0): cluster 1\n", + "Example n. 2194 = (29.0, 201.0): cluster 1\n", + "Example n. 2195 = (29.0, 200.0): cluster 1\n", + "Example n. 2196 = (29.0, 199.0): cluster 3\n", + "Example n. 2197 = (29.0, 198.0): cluster 3\n", + "Example n. 2198 = (29.0, 197.0): cluster 3\n", + "Example n. 2199 = (29.0, 196.0): cluster 3\n", + "Example n. 2200 = (29.0, 195.0): cluster 3\n", + "Example n. 2201 = (29.0, 194.0): cluster 3\n", + "Example n. 2202 = (29.0, 193.0): cluster 3\n", + "Example n. 2203 = (29.0, 192.0): cluster 3\n", + "Example n. 2204 = (29.0, 191.0): cluster 3\n", + "Example n. 2205 = (29.0, 190.0): cluster 3\n", + "Example n. 2206 = (29.0, 189.0): cluster 3\n", + "Example n. 2207 = (29.0, 188.0): cluster 3\n", + "Example n. 2208 = (29.0, 187.0): cluster 3\n", + "Example n. 2209 = (29.0, 186.0): cluster 3\n", + "Example n. 2210 = (29.0, 185.0): cluster 3\n", + "Example n. 2211 = (29.0, 184.0): cluster 3\n", + "Example n. 2212 = (29.0, 183.0): cluster 3\n", + "Example n. 2213 = (29.0, 182.0): cluster 3\n", + "Example n. 2214 = (29.0, 181.0): cluster 3\n", + "Example n. 2215 = (29.0, 177.0): cluster 3\n", + "Example n. 2216 = (29.0, 176.0): cluster 3\n", + "Example n. 2217 = (29.0, 175.0): cluster 3\n", + "Example n. 2218 = (29.0, 167.0): cluster 3\n", + "Example n. 2219 = (29.0, 166.0): cluster 3\n", + "Example n. 2220 = (29.0, 165.0): cluster 3\n", + "Example n. 2221 = (29.0, 164.0): cluster 3\n", + "Example n. 2222 = (29.0, 163.0): cluster 3\n", + "Example n. 2223 = (29.0, 162.0): cluster 3\n", + "Example n. 2224 = (29.0, 161.0): cluster 3\n", + "Example n. 2225 = (29.0, 121.0): cluster 3\n", + "Example n. 2226 = (29.0, 120.0): cluster 3\n", + "Example n. 2227 = (29.0, 119.0): cluster 3\n", + "Example n. 2228 = (29.0, 118.0): cluster 0\n", + "Example n. 2229 = (29.0, 117.0): cluster 0\n", + "Example n. 2230 = (29.0, 116.0): cluster 0\n", + "Example n. 2231 = (29.0, 115.0): cluster 0\n", + "Example n. 2232 = (30.0, 503.0): cluster 0\n", + "Example n. 2233 = (30.0, 502.0): cluster 0\n", + "Example n. 2234 = (30.0, 501.0): cluster 0\n", + "Example n. 2235 = (30.0, 500.0): cluster 0\n", + "Example n. 2236 = (30.0, 499.0): cluster 0\n", + "Example n. 2237 = (30.0, 498.0): cluster 0\n", + "Example n. 2238 = (30.0, 486.0): cluster 0\n", + "Example n. 2239 = (30.0, 485.0): cluster 0\n", + "Example n. 2240 = (30.0, 484.0): cluster 1\n", + "Example n. 2241 = (30.0, 483.0): cluster 1\n", + "Example n. 2242 = (30.0, 482.0): cluster 1\n", + "Example n. 2243 = (30.0, 481.0): cluster 1\n", + "Example n. 2244 = (30.0, 480.0): cluster 1\n", + "Example n. 2245 = (30.0, 479.0): cluster 1\n", + "Example n. 2246 = (30.0, 478.0): cluster 1\n", + "Example n. 2247 = (30.0, 477.0): cluster 1\n", + "Example n. 2248 = (30.0, 476.0): cluster 1\n", + "Example n. 2249 = (30.0, 475.0): cluster 1\n", + "Example n. 2250 = (30.0, 474.0): cluster 1\n", + "Example n. 2251 = (30.0, 473.0): cluster 1\n", + "Example n. 2252 = (30.0, 472.0): cluster 1\n", + "Example n. 2253 = (30.0, 471.0): cluster 1\n", + "Example n. 2254 = (30.0, 470.0): cluster 1\n", + "Example n. 2255 = (30.0, 469.0): cluster 1\n", + "Example n. 2256 = (30.0, 468.0): cluster 1\n", + "Example n. 2257 = (30.0, 467.0): cluster 1\n", + "Example n. 2258 = (30.0, 466.0): cluster 1\n", + "Example n. 2259 = (30.0, 465.0): cluster 3\n", + "Example n. 2260 = (30.0, 464.0): cluster 3\n", + "Example n. 2261 = (30.0, 463.0): cluster 3\n", + "Example n. 2262 = (30.0, 370.0): cluster 3\n", + "Example n. 2263 = (30.0, 369.0): cluster 3\n", + "Example n. 2264 = (30.0, 368.0): cluster 3\n", + "Example n. 2265 = (30.0, 367.0): cluster 3\n", + "Example n. 2266 = (30.0, 302.0): cluster 3\n", + "Example n. 2267 = (30.0, 301.0): cluster 3\n", + "Example n. 2268 = (30.0, 300.0): cluster 3\n", + "Example n. 2269 = (30.0, 299.0): cluster 3\n", + "Example n. 2270 = (30.0, 296.0): cluster 3\n", + "Example n. 2271 = (30.0, 295.0): cluster 3\n", + "Example n. 2272 = (30.0, 294.0): cluster 3\n", + "Example n. 2273 = (30.0, 293.0): cluster 3\n", + "Example n. 2274 = (30.0, 292.0): cluster 3\n", + "Example n. 2275 = (30.0, 291.0): cluster 3\n", + "Example n. 2276 = (30.0, 290.0): cluster 3\n", + "Example n. 2277 = (30.0, 289.0): cluster 3\n", + "Example n. 2278 = (30.0, 288.0): cluster 3\n", + "Example n. 2279 = (30.0, 287.0): cluster 3\n", + "Example n. 2280 = (30.0, 286.0): cluster 3\n", + "Example n. 2281 = (30.0, 281.0): cluster 3\n", + "Example n. 2282 = (30.0, 280.0): cluster 3\n", + "Example n. 2283 = (30.0, 279.0): cluster 3\n", + "Example n. 2284 = (30.0, 278.0): cluster 3\n", + "Example n. 2285 = (30.0, 277.0): cluster 3\n", + "Example n. 2286 = (30.0, 276.0): cluster 3\n", + "Example n. 2287 = (30.0, 273.0): cluster 3\n", + "Example n. 2288 = (30.0, 272.0): cluster 3\n", + "Example n. 2289 = (30.0, 271.0): cluster 3\n", + "Example n. 2290 = (30.0, 270.0): cluster 3\n", + "Example n. 2291 = (30.0, 269.0): cluster 0\n", + "Example n. 2292 = (30.0, 268.0): cluster 0\n", + "Example n. 2293 = (30.0, 267.0): cluster 0\n", + "Example n. 2294 = (30.0, 266.0): cluster 0\n", + "Example n. 2295 = (30.0, 265.0): cluster 0\n", + "Example n. 2296 = (30.0, 264.0): cluster 0\n", + "Example n. 2297 = (30.0, 263.0): cluster 0\n", + "Example n. 2298 = (30.0, 262.0): cluster 0\n", + "Example n. 2299 = (30.0, 261.0): cluster 0\n", + "Example n. 2300 = (30.0, 260.0): cluster 1\n", + "Example n. 2301 = (30.0, 259.0): cluster 1\n", + "Example n. 2302 = (30.0, 258.0): cluster 1\n", + "Example n. 2303 = (30.0, 257.0): cluster 1\n", + "Example n. 2304 = (30.0, 256.0): cluster 1\n", + "Example n. 2305 = (30.0, 255.0): cluster 1\n", + "Example n. 2306 = (30.0, 254.0): cluster 1\n", + "Example n. 2307 = (30.0, 253.0): cluster 1\n", + "Example n. 2308 = (30.0, 252.0): cluster 1\n", + "Example n. 2309 = (30.0, 251.0): cluster 1\n", + "Example n. 2310 = (30.0, 250.0): cluster 1\n", + "Example n. 2311 = (30.0, 249.0): cluster 1\n", + "Example n. 2312 = (30.0, 248.0): cluster 1\n", + "Example n. 2313 = (30.0, 247.0): cluster 1\n", + "Example n. 2314 = (30.0, 246.0): cluster 1\n", + "Example n. 2315 = (30.0, 245.0): cluster 1\n", + "Example n. 2316 = (30.0, 244.0): cluster 1\n", + "Example n. 2317 = (30.0, 243.0): cluster 1\n", + "Example n. 2318 = (30.0, 242.0): cluster 1\n", + "Example n. 2319 = (30.0, 241.0): cluster 1\n", + "Example n. 2320 = (30.0, 240.0): cluster 1\n", + "Example n. 2321 = (30.0, 239.0): cluster 1\n", + "Example n. 2322 = (30.0, 238.0): cluster 3\n", + "Example n. 2323 = (30.0, 237.0): cluster 3\n", + "Example n. 2324 = (30.0, 236.0): cluster 3\n", + "Example n. 2325 = (30.0, 235.0): cluster 3\n", + "Example n. 2326 = (30.0, 234.0): cluster 3\n", + "Example n. 2327 = (30.0, 233.0): cluster 3\n", + "Example n. 2328 = (30.0, 232.0): cluster 3\n", + "Example n. 2329 = (30.0, 231.0): cluster 3\n", + "Example n. 2330 = (30.0, 230.0): cluster 3\n", + "Example n. 2331 = (30.0, 228.0): cluster 3\n", + "Example n. 2332 = (30.0, 227.0): cluster 3\n", + "Example n. 2333 = (30.0, 226.0): cluster 3\n", + "Example n. 2334 = (30.0, 225.0): cluster 3\n", + "Example n. 2335 = (30.0, 224.0): cluster 3\n", + "Example n. 2336 = (30.0, 223.0): cluster 3\n", + "Example n. 2337 = (30.0, 218.0): cluster 3\n", + "Example n. 2338 = (30.0, 217.0): cluster 3\n", + "Example n. 2339 = (30.0, 216.0): cluster 3\n", + "Example n. 2340 = (30.0, 215.0): cluster 3\n", + "Example n. 2341 = (30.0, 214.0): cluster 3\n", + "Example n. 2342 = (30.0, 213.0): cluster 3\n", + "Example n. 2343 = (30.0, 212.0): cluster 3\n", + "Example n. 2344 = (30.0, 211.0): cluster 3\n", + "Example n. 2345 = (30.0, 210.0): cluster 3\n", + "Example n. 2346 = (30.0, 204.0): cluster 3\n", + "Example n. 2347 = (30.0, 203.0): cluster 3\n", + "Example n. 2348 = (30.0, 202.0): cluster 3\n", + "Example n. 2349 = (30.0, 201.0): cluster 3\n", + "Example n. 2350 = (30.0, 200.0): cluster 3\n", + "Example n. 2351 = (30.0, 199.0): cluster 3\n", + "Example n. 2352 = (30.0, 198.0): cluster 3\n", + "Example n. 2353 = (30.0, 197.0): cluster 3\n", + "Example n. 2354 = (30.0, 196.0): cluster 0\n", + "Example n. 2355 = (30.0, 195.0): cluster 0\n", + "Example n. 2356 = (30.0, 194.0): cluster 0\n", + "Example n. 2357 = (30.0, 193.0): cluster 0\n", + "Example n. 2358 = (30.0, 192.0): cluster 0\n", + "Example n. 2359 = (30.0, 191.0): cluster 0\n", + "Example n. 2360 = (30.0, 190.0): cluster 0\n", + "Example n. 2361 = (30.0, 189.0): cluster 1\n", + "Example n. 2362 = (30.0, 188.0): cluster 1\n", + "Example n. 2363 = (30.0, 187.0): cluster 1\n", + "Example n. 2364 = (30.0, 186.0): cluster 1\n", + "Example n. 2365 = (30.0, 185.0): cluster 1\n", + "Example n. 2366 = (30.0, 184.0): cluster 1\n", + "Example n. 2367 = (30.0, 183.0): cluster 1\n", + "Example n. 2368 = (30.0, 182.0): cluster 1\n", + "Example n. 2369 = (30.0, 181.0): cluster 1\n", + "Example n. 2370 = (30.0, 180.0): cluster 1\n", + "Example n. 2371 = (30.0, 179.0): cluster 1\n", + "Example n. 2372 = (30.0, 178.0): cluster 1\n", + "Example n. 2373 = (30.0, 177.0): cluster 1\n", + "Example n. 2374 = (30.0, 176.0): cluster 1\n", + "Example n. 2375 = (30.0, 175.0): cluster 1\n", + "Example n. 2376 = (30.0, 166.0): cluster 1\n", + "Example n. 2377 = (30.0, 165.0): cluster 1\n", + "Example n. 2378 = (30.0, 164.0): cluster 1\n", + "Example n. 2379 = (30.0, 163.0): cluster 1\n", + "Example n. 2380 = (30.0, 162.0): cluster 1\n", + "Example n. 2381 = (30.0, 161.0): cluster 1\n", + "Example n. 2382 = (30.0, 160.0): cluster 1\n", + "Example n. 2383 = (30.0, 158.0): cluster 3\n", + "Example n. 2384 = (30.0, 157.0): cluster 3\n", + "Example n. 2385 = (30.0, 156.0): cluster 3\n", + "Example n. 2386 = (30.0, 155.0): cluster 3\n", + "Example n. 2387 = (30.0, 154.0): cluster 3\n", + "Example n. 2388 = (31.0, 502.0): cluster 3\n", + "Example n. 2389 = (31.0, 501.0): cluster 3\n", + "Example n. 2390 = (31.0, 500.0): cluster 3\n", + "Example n. 2391 = (31.0, 499.0): cluster 3\n", + "Example n. 2392 = (31.0, 498.0): cluster 3\n", + "Example n. 2393 = (31.0, 486.0): cluster 3\n", + "Example n. 2394 = (31.0, 485.0): cluster 3\n", + "Example n. 2395 = (31.0, 484.0): cluster 3\n", + "Example n. 2396 = (31.0, 483.0): cluster 3\n", + "Example n. 2397 = (31.0, 482.0): cluster 3\n", + "Example n. 2398 = (31.0, 481.0): cluster 3\n", + "Example n. 2399 = (31.0, 480.0): cluster 3\n", + "Example n. 2400 = (31.0, 479.0): cluster 3\n", + "Example n. 2401 = (31.0, 478.0): cluster 3\n", + "Example n. 2402 = (31.0, 477.0): cluster 3\n", + "Example n. 2403 = (31.0, 476.0): cluster 3\n", + "Example n. 2404 = (31.0, 475.0): cluster 3\n", + "Example n. 2405 = (31.0, 474.0): cluster 3\n", + "Example n. 2406 = (31.0, 473.0): cluster 3\n", + "Example n. 2407 = (31.0, 472.0): cluster 3\n", + "Example n. 2408 = (31.0, 471.0): cluster 3\n", + "Example n. 2409 = (31.0, 470.0): cluster 3\n", + "Example n. 2410 = (31.0, 469.0): cluster 3\n", + "Example n. 2411 = (31.0, 468.0): cluster 3\n", + "Example n. 2412 = (31.0, 467.0): cluster 3\n", + "Example n. 2413 = (31.0, 466.0): cluster 3\n", + "Example n. 2414 = (31.0, 465.0): cluster 3\n", + "Example n. 2415 = (31.0, 464.0): cluster 0\n", + "Example n. 2416 = (31.0, 463.0): cluster 0\n", + "Example n. 2417 = (31.0, 462.0): cluster 0\n", + "Example n. 2418 = (31.0, 461.0): cluster 0\n", + "Example n. 2419 = (31.0, 396.0): cluster 1\n", + "Example n. 2420 = (31.0, 395.0): cluster 1\n", + "Example n. 2421 = (31.0, 303.0): cluster 1\n", + "Example n. 2422 = (31.0, 302.0): cluster 1\n", + "Example n. 2423 = (31.0, 301.0): cluster 1\n", + "Example n. 2424 = (31.0, 300.0): cluster 1\n", + "Example n. 2425 = (31.0, 299.0): cluster 1\n", + "Example n. 2426 = (31.0, 295.0): cluster 1\n", + "Example n. 2427 = (31.0, 294.0): cluster 1\n", + "Example n. 2428 = (31.0, 293.0): cluster 1\n", + "Example n. 2429 = (31.0, 292.0): cluster 1\n", + "Example n. 2430 = (31.0, 291.0): cluster 1\n", + "Example n. 2431 = (31.0, 290.0): cluster 1\n", + "Example n. 2432 = (31.0, 289.0): cluster 1\n", + "Example n. 2433 = (31.0, 288.0): cluster 1\n", + "Example n. 2434 = (31.0, 287.0): cluster 1\n", + "Example n. 2435 = (31.0, 286.0): cluster 1\n", + "Example n. 2436 = (31.0, 281.0): cluster 1\n", + "Example n. 2437 = (31.0, 280.0): cluster 1\n", + "Example n. 2438 = (31.0, 279.0): cluster 1\n", + "Example n. 2439 = (31.0, 278.0): cluster 1\n", + "Example n. 2440 = (31.0, 277.0): cluster 1\n", + "Example n. 2441 = (31.0, 276.0): cluster 1\n", + "Example n. 2442 = (31.0, 275.0): cluster 1\n", + "Example n. 2443 = (31.0, 271.0): cluster 1\n", + "Example n. 2444 = (31.0, 270.0): cluster 1\n", + "Example n. 2445 = (31.0, 269.0): cluster 3\n", + "Example n. 2446 = (31.0, 268.0): cluster 3\n", + "Example n. 2447 = (31.0, 267.0): cluster 3\n", + "Example n. 2448 = (31.0, 266.0): cluster 3\n", + "Example n. 2449 = (31.0, 265.0): cluster 3\n", + "Example n. 2450 = (31.0, 264.0): cluster 3\n", + "Example n. 2451 = (31.0, 263.0): cluster 3\n", + "Example n. 2452 = (31.0, 262.0): cluster 3\n", + "Example n. 2453 = (31.0, 261.0): cluster 3\n", + "Example n. 2454 = (31.0, 260.0): cluster 3\n", + "Example n. 2455 = (31.0, 259.0): cluster 3\n", + "Example n. 2456 = (31.0, 258.0): cluster 3\n", + "Example n. 2457 = (31.0, 257.0): cluster 3\n", + "Example n. 2458 = (31.0, 256.0): cluster 3\n", + "Example n. 2459 = (31.0, 255.0): cluster 3\n", + "Example n. 2460 = (31.0, 254.0): cluster 3\n", + "Example n. 2461 = (31.0, 253.0): cluster 3\n", + "Example n. 2462 = (31.0, 252.0): cluster 3\n", + "Example n. 2463 = (31.0, 251.0): cluster 3\n", + "Example n. 2464 = (31.0, 250.0): cluster 3\n", + "Example n. 2465 = (31.0, 249.0): cluster 3\n", + "Example n. 2466 = (31.0, 248.0): cluster 3\n", + "Example n. 2467 = (31.0, 247.0): cluster 3\n", + "Example n. 2468 = (31.0, 246.0): cluster 3\n", + "Example n. 2469 = (31.0, 245.0): cluster 3\n", + "Example n. 2470 = (31.0, 244.0): cluster 3\n", + "Example n. 2471 = (31.0, 243.0): cluster 3\n", + "Example n. 2472 = (31.0, 242.0): cluster 3\n", + "Example n. 2473 = (31.0, 241.0): cluster 3\n", + "Example n. 2474 = (31.0, 240.0): cluster 3\n", + "Example n. 2475 = (31.0, 239.0): cluster 3\n", + "Example n. 2476 = (31.0, 238.0): cluster 3\n", + "Example n. 2477 = (31.0, 237.0): cluster 1\n", + "Example n. 2478 = (31.0, 236.0): cluster 1\n", + "Example n. 2479 = (31.0, 235.0): cluster 1\n", + "Example n. 2480 = (31.0, 234.0): cluster 1\n", + "Example n. 2481 = (31.0, 233.0): cluster 1\n", + "Example n. 2482 = (31.0, 232.0): cluster 1\n", + "Example n. 2483 = (31.0, 231.0): cluster 1\n", + "Example n. 2484 = (31.0, 230.0): cluster 1\n", + "Example n. 2485 = (31.0, 228.0): cluster 1\n", + "Example n. 2486 = (31.0, 227.0): cluster 1\n", + "Example n. 2487 = (31.0, 226.0): cluster 1\n", + "Example n. 2488 = (31.0, 225.0): cluster 1\n", + "Example n. 2489 = (31.0, 224.0): cluster 1\n", + "Example n. 2490 = (31.0, 223.0): cluster 1\n", + "Example n. 2491 = (31.0, 218.0): cluster 1\n", + "Example n. 2492 = (31.0, 217.0): cluster 1\n", + "Example n. 2493 = (31.0, 216.0): cluster 1\n", + "Example n. 2494 = (31.0, 215.0): cluster 1\n", + "Example n. 2495 = (31.0, 214.0): cluster 1\n", + "Example n. 2496 = (31.0, 213.0): cluster 1\n", + "Example n. 2497 = (31.0, 212.0): cluster 1\n", + "Example n. 2498 = (31.0, 211.0): cluster 1\n", + "Example n. 2499 = (31.0, 210.0): cluster 1\n", + "Example n. 2500 = (31.0, 209.0): cluster 1\n", + "Example n. 2501 = (31.0, 204.0): cluster 3\n", + "Example n. 2502 = (31.0, 203.0): cluster 3\n", + "Example n. 2503 = (31.0, 202.0): cluster 3\n", + "Example n. 2504 = (31.0, 201.0): cluster 3\n", + "Example n. 2505 = (31.0, 200.0): cluster 3\n", + "Example n. 2506 = (31.0, 199.0): cluster 3\n", + "Example n. 2507 = (31.0, 198.0): cluster 3\n", + "Example n. 2508 = (31.0, 197.0): cluster 3\n", + "Example n. 2509 = (31.0, 196.0): cluster 3\n", + "Example n. 2510 = (31.0, 195.0): cluster 3\n", + "Example n. 2511 = (31.0, 194.0): cluster 3\n", + "Example n. 2512 = (31.0, 193.0): cluster 3\n", + "Example n. 2513 = (31.0, 192.0): cluster 3\n", + "Example n. 2514 = (31.0, 191.0): cluster 3\n", + "Example n. 2515 = (31.0, 188.0): cluster 3\n", + "Example n. 2516 = (31.0, 186.0): cluster 3\n", + "Example n. 2517 = (31.0, 185.0): cluster 3\n", + "Example n. 2518 = (31.0, 184.0): cluster 3\n", + "Example n. 2519 = (31.0, 183.0): cluster 3\n", + "Example n. 2520 = (31.0, 182.0): cluster 3\n", + "Example n. 2521 = (31.0, 181.0): cluster 3\n", + "Example n. 2522 = (31.0, 180.0): cluster 3\n", + "Example n. 2523 = (31.0, 179.0): cluster 3\n", + "Example n. 2524 = (31.0, 178.0): cluster 3\n", + "Example n. 2525 = (31.0, 177.0): cluster 3\n", + "Example n. 2526 = (31.0, 176.0): cluster 3\n", + "Example n. 2527 = (31.0, 175.0): cluster 3\n", + "Example n. 2528 = (31.0, 174.0): cluster 3\n", + "Example n. 2529 = (31.0, 168.0): cluster 1\n", + "Example n. 2530 = (31.0, 167.0): cluster 1\n", + "Example n. 2531 = (31.0, 166.0): cluster 1\n", + "Example n. 2532 = (31.0, 165.0): cluster 1\n", + "Example n. 2533 = (31.0, 164.0): cluster 1\n", + "Example n. 2534 = (31.0, 163.0): cluster 1\n", + "Example n. 2535 = (31.0, 162.0): cluster 1\n", + "Example n. 2536 = (31.0, 161.0): cluster 1\n", + "Example n. 2537 = (31.0, 160.0): cluster 1\n", + "Example n. 2538 = (31.0, 159.0): cluster 1\n", + "Example n. 2539 = (31.0, 158.0): cluster 1\n", + "Example n. 2540 = (31.0, 157.0): cluster 1\n", + "Example n. 2541 = (31.0, 156.0): cluster 1\n", + "Example n. 2542 = (31.0, 155.0): cluster 1\n", + "Example n. 2543 = (31.0, 154.0): cluster 1\n", + "Example n. 2544 = (31.0, 153.0): cluster 1\n", + "Example n. 2545 = (31.0, 132.0): cluster 1\n", + "Example n. 2546 = (31.0, 131.0): cluster 1\n", + "Example n. 2547 = (31.0, 130.0): cluster 1\n", + "Example n. 2548 = (31.0, 129.0): cluster 1\n", + "Example n. 2549 = (32.0, 501.0): cluster 1\n", + "Example n. 2550 = (32.0, 500.0): cluster 1\n", + "Example n. 2551 = (32.0, 499.0): cluster 1\n", + "Example n. 2552 = (32.0, 486.0): cluster 1\n", + "Example n. 2553 = (32.0, 485.0): cluster 3\n", + "Example n. 2554 = (32.0, 484.0): cluster 3\n", + "Example n. 2555 = (32.0, 483.0): cluster 3\n", + "Example n. 2556 = (32.0, 482.0): cluster 3\n", + "Example n. 2557 = (32.0, 481.0): cluster 3\n", + "Example n. 2558 = (32.0, 480.0): cluster 3\n", + "Example n. 2559 = (32.0, 479.0): cluster 3\n", + "Example n. 2560 = (32.0, 478.0): cluster 3\n", + "Example n. 2561 = (32.0, 477.0): cluster 3\n", + "Example n. 2562 = (32.0, 476.0): cluster 3\n", + "Example n. 2563 = (32.0, 475.0): cluster 3\n", + "Example n. 2564 = (32.0, 474.0): cluster 3\n", + "Example n. 2565 = (32.0, 473.0): cluster 3\n", + "Example n. 2566 = (32.0, 472.0): cluster 3\n", + "Example n. 2567 = (32.0, 471.0): cluster 3\n", + "Example n. 2568 = (32.0, 470.0): cluster 3\n", + "Example n. 2569 = (32.0, 469.0): cluster 3\n", + "Example n. 2570 = (32.0, 468.0): cluster 3\n", + "Example n. 2571 = (32.0, 467.0): cluster 3\n", + "Example n. 2572 = (32.0, 466.0): cluster 3\n", + "Example n. 2573 = (32.0, 465.0): cluster 3\n", + "Example n. 2574 = (32.0, 464.0): cluster 3\n", + "Example n. 2575 = (32.0, 463.0): cluster 3\n", + "Example n. 2576 = (32.0, 462.0): cluster 3\n", + "Example n. 2577 = (32.0, 461.0): cluster 3\n", + "Example n. 2578 = (32.0, 460.0): cluster 3\n", + "Example n. 2579 = (32.0, 397.0): cluster 3\n", + "Example n. 2580 = (32.0, 396.0): cluster 3\n", + "Example n. 2581 = (32.0, 395.0): cluster 1\n", + "Example n. 2582 = (32.0, 394.0): cluster 1\n", + "Example n. 2583 = (32.0, 393.0): cluster 1\n", + "Example n. 2584 = (32.0, 303.0): cluster 1\n", + "Example n. 2585 = (32.0, 302.0): cluster 1\n", + "Example n. 2586 = (32.0, 301.0): cluster 1\n", + "Example n. 2587 = (32.0, 300.0): cluster 1\n", + "Example n. 2588 = (32.0, 299.0): cluster 1\n", + "Example n. 2589 = (32.0, 298.0): cluster 1\n", + "Example n. 2590 = (32.0, 295.0): cluster 1\n", + "Example n. 2591 = (32.0, 294.0): cluster 1\n", + "Example n. 2592 = (32.0, 293.0): cluster 1\n", + "Example n. 2593 = (32.0, 292.0): cluster 1\n", + "Example n. 2594 = (32.0, 291.0): cluster 1\n", + "Example n. 2595 = (32.0, 290.0): cluster 1\n", + "Example n. 2596 = (32.0, 289.0): cluster 1\n", + "Example n. 2597 = (32.0, 288.0): cluster 1\n", + "Example n. 2598 = (32.0, 287.0): cluster 1\n", + "Example n. 2599 = (32.0, 286.0): cluster 1\n", + "Example n. 2600 = (32.0, 285.0): cluster 1\n", + "Example n. 2601 = (32.0, 284.0): cluster 1\n", + "Example n. 2602 = (32.0, 281.0): cluster 1\n", + "Example n. 2603 = (32.0, 280.0): cluster 1\n", + "Example n. 2604 = (32.0, 279.0): cluster 1\n", + "Example n. 2605 = (32.0, 278.0): cluster 3\n", + "Example n. 2606 = (32.0, 277.0): cluster 3\n", + "Example n. 2607 = (32.0, 276.0): cluster 3\n", + "Example n. 2608 = (32.0, 275.0): cluster 3\n", + "Example n. 2609 = (32.0, 271.0): cluster 3\n", + "Example n. 2610 = (32.0, 270.0): cluster 3\n", + "Example n. 2611 = (32.0, 269.0): cluster 3\n", + "Example n. 2612 = (32.0, 268.0): cluster 3\n", + "Example n. 2613 = (32.0, 267.0): cluster 3\n", + "Example n. 2614 = (32.0, 266.0): cluster 3\n", + "Example n. 2615 = (32.0, 265.0): cluster 3\n", + "Example n. 2616 = (32.0, 264.0): cluster 3\n", + "Example n. 2617 = (32.0, 263.0): cluster 3\n", + "Example n. 2618 = (32.0, 262.0): cluster 3\n", + "Example n. 2619 = (32.0, 261.0): cluster 3\n", + "Example n. 2620 = (32.0, 260.0): cluster 3\n", + "Example n. 2621 = (32.0, 259.0): cluster 3\n", + "Example n. 2622 = (32.0, 258.0): cluster 3\n", + "Example n. 2623 = (32.0, 257.0): cluster 3\n", + "Example n. 2624 = (32.0, 256.0): cluster 3\n", + "Example n. 2625 = (32.0, 255.0): cluster 3\n", + "Example n. 2626 = (32.0, 254.0): cluster 3\n", + "Example n. 2627 = (32.0, 253.0): cluster 3\n", + "Example n. 2628 = (32.0, 252.0): cluster 3\n", + "Example n. 2629 = (32.0, 251.0): cluster 3\n", + "Example n. 2630 = (32.0, 250.0): cluster 3\n", + "Example n. 2631 = (32.0, 249.0): cluster 3\n", + "Example n. 2632 = (32.0, 248.0): cluster 3\n", + "Example n. 2633 = (32.0, 247.0): cluster 1\n", + "Example n. 2634 = (32.0, 246.0): cluster 1\n", + "Example n. 2635 = (32.0, 245.0): cluster 1\n", + "Example n. 2636 = (32.0, 244.0): cluster 1\n", + "Example n. 2637 = (32.0, 243.0): cluster 1\n", + "Example n. 2638 = (32.0, 242.0): cluster 1\n", + "Example n. 2639 = (32.0, 241.0): cluster 1\n", + "Example n. 2640 = (32.0, 240.0): cluster 1\n", + "Example n. 2641 = (32.0, 239.0): cluster 1\n", + "Example n. 2642 = (32.0, 238.0): cluster 1\n", + "Example n. 2643 = (32.0, 237.0): cluster 1\n", + "Example n. 2644 = (32.0, 236.0): cluster 1\n", + "Example n. 2645 = (32.0, 235.0): cluster 1\n", + "Example n. 2646 = (32.0, 234.0): cluster 1\n", + "Example n. 2647 = (32.0, 233.0): cluster 1\n", + "Example n. 2648 = (32.0, 232.0): cluster 1\n", + "Example n. 2649 = (32.0, 231.0): cluster 1\n", + "Example n. 2650 = (32.0, 230.0): cluster 1\n", + "Example n. 2651 = (32.0, 228.0): cluster 1\n", + "Example n. 2652 = (32.0, 227.0): cluster 1\n", + "Example n. 2653 = (32.0, 226.0): cluster 1\n", + "Example n. 2654 = (32.0, 225.0): cluster 1\n", + "Example n. 2655 = (32.0, 224.0): cluster 1\n", + "Example n. 2656 = (32.0, 223.0): cluster 1\n", + "Example n. 2657 = (32.0, 222.0): cluster 3\n", + "Example n. 2658 = (32.0, 217.0): cluster 3\n", + "Example n. 2659 = (32.0, 216.0): cluster 3\n", + "Example n. 2660 = (32.0, 215.0): cluster 3\n", + "Example n. 2661 = (32.0, 214.0): cluster 3\n", + "Example n. 2662 = (32.0, 213.0): cluster 3\n", + "Example n. 2663 = (32.0, 212.0): cluster 3\n", + "Example n. 2664 = (32.0, 211.0): cluster 3\n", + "Example n. 2665 = (32.0, 210.0): cluster 3\n", + "Example n. 2666 = (32.0, 209.0): cluster 3\n", + "Example n. 2667 = (32.0, 202.0): cluster 3\n", + "Example n. 2668 = (32.0, 201.0): cluster 3\n", + "Example n. 2669 = (32.0, 200.0): cluster 3\n", + "Example n. 2670 = (32.0, 199.0): cluster 3\n", + "Example n. 2671 = (32.0, 198.0): cluster 3\n", + "Example n. 2672 = (32.0, 197.0): cluster 3\n", + "Example n. 2673 = (32.0, 196.0): cluster 3\n", + "Example n. 2674 = (32.0, 195.0): cluster 3\n", + "Example n. 2675 = (32.0, 194.0): cluster 3\n", + "Example n. 2676 = (32.0, 193.0): cluster 3\n", + "Example n. 2677 = (32.0, 192.0): cluster 3\n", + "Example n. 2678 = (32.0, 191.0): cluster 3\n", + "Example n. 2679 = (32.0, 184.0): cluster 3\n", + "Example n. 2680 = (32.0, 183.0): cluster 3\n", + "Example n. 2681 = (32.0, 182.0): cluster 3\n", + "Example n. 2682 = (32.0, 181.0): cluster 3\n", + "Example n. 2683 = (32.0, 180.0): cluster 3\n", + "Example n. 2684 = (32.0, 179.0): cluster 3\n", + "Example n. 2685 = (32.0, 178.0): cluster 1\n", + "Example n. 2686 = (32.0, 177.0): cluster 1\n", + "Example n. 2687 = (32.0, 176.0): cluster 1\n", + "Example n. 2688 = (32.0, 175.0): cluster 1\n", + "Example n. 2689 = (32.0, 174.0): cluster 1\n", + "Example n. 2690 = (32.0, 170.0): cluster 1\n", + "Example n. 2691 = (32.0, 169.0): cluster 1\n", + "Example n. 2692 = (32.0, 168.0): cluster 1\n", + "Example n. 2693 = (32.0, 167.0): cluster 1\n", + "Example n. 2694 = (32.0, 166.0): cluster 1\n", + "Example n. 2695 = (32.0, 165.0): cluster 1\n", + "Example n. 2696 = (32.0, 164.0): cluster 1\n", + "Example n. 2697 = (32.0, 163.0): cluster 1\n", + "Example n. 2698 = (32.0, 162.0): cluster 1\n", + "Example n. 2699 = (32.0, 161.0): cluster 1\n", + "Example n. 2700 = (32.0, 160.0): cluster 1\n", + "Example n. 2701 = (32.0, 159.0): cluster 1\n", + "Example n. 2702 = (32.0, 158.0): cluster 1\n", + "Example n. 2703 = (32.0, 157.0): cluster 1\n", + "Example n. 2704 = (32.0, 156.0): cluster 1\n", + "Example n. 2705 = (32.0, 155.0): cluster 1\n", + "Example n. 2706 = (32.0, 154.0): cluster 1\n", + "Example n. 2707 = (32.0, 153.0): cluster 1\n", + "Example n. 2708 = (32.0, 136.0): cluster 1\n", + "Example n. 2709 = (32.0, 135.0): cluster 1\n", + "Example n. 2710 = (32.0, 134.0): cluster 1\n", + "Example n. 2711 = (32.0, 133.0): cluster 1\n", + "Example n. 2712 = (32.0, 132.0): cluster 1\n", + "Example n. 2713 = (32.0, 131.0): cluster 1\n", + "Example n. 2714 = (32.0, 130.0): cluster 1\n", + "Example n. 2715 = (32.0, 129.0): cluster 1\n", + "Example n. 2716 = (32.0, 128.0): cluster 1\n", + "Example n. 2717 = (33.0, 488.0): cluster 1\n", + "Example n. 2718 = (33.0, 487.0): cluster 1\n", + "Example n. 2719 = (33.0, 486.0): cluster 3\n", + "Example n. 2720 = (33.0, 485.0): cluster 3\n", + "Example n. 2721 = (33.0, 484.0): cluster 3\n", + "Example n. 2722 = (33.0, 483.0): cluster 3\n", + "Example n. 2723 = (33.0, 482.0): cluster 3\n", + "Example n. 2724 = (33.0, 481.0): cluster 3\n", + "Example n. 2725 = (33.0, 480.0): cluster 3\n", + "Example n. 2726 = (33.0, 479.0): cluster 3\n", + "Example n. 2727 = (33.0, 478.0): cluster 3\n", + "Example n. 2728 = (33.0, 477.0): cluster 3\n", + "Example n. 2729 = (33.0, 476.0): cluster 3\n", + "Example n. 2730 = (33.0, 475.0): cluster 3\n", + "Example n. 2731 = (33.0, 474.0): cluster 3\n", + "Example n. 2732 = (33.0, 473.0): cluster 3\n", + "Example n. 2733 = (33.0, 472.0): cluster 3\n", + "Example n. 2734 = (33.0, 471.0): cluster 3\n", + "Example n. 2735 = (33.0, 470.0): cluster 3\n", + "Example n. 2736 = (33.0, 469.0): cluster 3\n", + "Example n. 2737 = (33.0, 468.0): cluster 3\n", + "Example n. 2738 = (33.0, 467.0): cluster 3\n", + "Example n. 2739 = (33.0, 466.0): cluster 3\n", + "Example n. 2740 = (33.0, 465.0): cluster 3\n", + "Example n. 2741 = (33.0, 464.0): cluster 3\n", + "Example n. 2742 = (33.0, 463.0): cluster 3\n", + "Example n. 2743 = (33.0, 462.0): cluster 3\n", + "Example n. 2744 = (33.0, 461.0): cluster 3\n", + "Example n. 2745 = (33.0, 460.0): cluster 3\n", + "Example n. 2746 = (33.0, 459.0): cluster 3\n", + "Example n. 2747 = (33.0, 398.0): cluster 3\n", + "Example n. 2748 = (33.0, 397.0): cluster 3\n", + "Example n. 2749 = (33.0, 396.0): cluster 3\n", + "Example n. 2750 = (33.0, 395.0): cluster 3\n", + "Example n. 2751 = (33.0, 394.0): cluster 3\n", + "Example n. 2752 = (33.0, 393.0): cluster 3\n", + "Example n. 2753 = (33.0, 380.0): cluster 3\n", + "Example n. 2754 = (33.0, 379.0): cluster 3\n", + "Example n. 2755 = (33.0, 378.0): cluster 3\n", + "Example n. 2756 = (33.0, 377.0): cluster 3\n", + "Example n. 2757 = (33.0, 303.0): cluster 1\n", + "Example n. 2758 = (33.0, 302.0): cluster 1\n", + "Example n. 2759 = (33.0, 301.0): cluster 1\n", + "Example n. 2760 = (33.0, 300.0): cluster 1\n", + "Example n. 2761 = (33.0, 299.0): cluster 1\n", + "Example n. 2762 = (33.0, 298.0): cluster 1\n", + "Example n. 2763 = (33.0, 297.0): cluster 1\n", + "Example n. 2764 = (33.0, 295.0): cluster 1\n", + "Example n. 2765 = (33.0, 294.0): cluster 1\n", + "Example n. 2766 = (33.0, 293.0): cluster 1\n", + "Example n. 2767 = (33.0, 292.0): cluster 1\n", + "Example n. 2768 = (33.0, 291.0): cluster 1\n", + "Example n. 2769 = (33.0, 290.0): cluster 1\n", + "Example n. 2770 = (33.0, 289.0): cluster 1\n", + "Example n. 2771 = (33.0, 288.0): cluster 1\n", + "Example n. 2772 = (33.0, 287.0): cluster 1\n", + "Example n. 2773 = (33.0, 286.0): cluster 1\n", + "Example n. 2774 = (33.0, 285.0): cluster 1\n", + "Example n. 2775 = (33.0, 284.0): cluster 1\n", + "Example n. 2776 = (33.0, 283.0): cluster 1\n", + "Example n. 2777 = (33.0, 281.0): cluster 1\n", + "Example n. 2778 = (33.0, 280.0): cluster 1\n", + "Example n. 2779 = (33.0, 279.0): cluster 1\n", + "Example n. 2780 = (33.0, 278.0): cluster 1\n", + "Example n. 2781 = (33.0, 277.0): cluster 1\n", + "Example n. 2782 = (33.0, 276.0): cluster 1\n", + "Example n. 2783 = (33.0, 275.0): cluster 1\n", + "Example n. 2784 = (33.0, 270.0): cluster 1\n", + "Example n. 2785 = (33.0, 269.0): cluster 1\n", + "Example n. 2786 = (33.0, 268.0): cluster 1\n", + "Example n. 2787 = (33.0, 267.0): cluster 1\n", + "Example n. 2788 = (33.0, 266.0): cluster 1\n", + "Example n. 2789 = (33.0, 265.0): cluster 1\n", + "Example n. 2790 = (33.0, 264.0): cluster 1\n", + "Example n. 2791 = (33.0, 263.0): cluster 3\n", + "Example n. 2792 = (33.0, 262.0): cluster 3\n", + "Example n. 2793 = (33.0, 261.0): cluster 3\n", + "Example n. 2794 = (33.0, 260.0): cluster 3\n", + "Example n. 2795 = (33.0, 259.0): cluster 3\n", + "Example n. 2796 = (33.0, 258.0): cluster 3\n", + "Example n. 2797 = (33.0, 257.0): cluster 3\n", + "Example n. 2798 = (33.0, 256.0): cluster 3\n", + "Example n. 2799 = (33.0, 255.0): cluster 3\n", + "Example n. 2800 = (33.0, 254.0): cluster 3\n", + "Example n. 2801 = (33.0, 253.0): cluster 3\n", + "Example n. 2802 = (33.0, 252.0): cluster 3\n", + "Example n. 2803 = (33.0, 251.0): cluster 3\n", + "Example n. 2804 = (33.0, 250.0): cluster 3\n", + "Example n. 2805 = (33.0, 249.0): cluster 3\n", + "Example n. 2806 = (33.0, 248.0): cluster 3\n", + "Example n. 2807 = (33.0, 247.0): cluster 3\n", + "Example n. 2808 = (33.0, 246.0): cluster 3\n", + "Example n. 2809 = (33.0, 245.0): cluster 3\n", + "Example n. 2810 = (33.0, 244.0): cluster 3\n", + "Example n. 2811 = (33.0, 243.0): cluster 3\n", + "Example n. 2812 = (33.0, 242.0): cluster 3\n", + "Example n. 2813 = (33.0, 241.0): cluster 3\n", + "Example n. 2814 = (33.0, 240.0): cluster 3\n", + "Example n. 2815 = (33.0, 239.0): cluster 3\n", + "Example n. 2816 = (33.0, 238.0): cluster 3\n", + "Example n. 2817 = (33.0, 237.0): cluster 3\n", + "Example n. 2818 = (33.0, 236.0): cluster 3\n", + "Example n. 2819 = (33.0, 235.0): cluster 3\n", + "Example n. 2820 = (33.0, 234.0): cluster 3\n", + "Example n. 2821 = (33.0, 233.0): cluster 3\n", + "Example n. 2822 = (33.0, 232.0): cluster 3\n", + "Example n. 2823 = (33.0, 231.0): cluster 3\n", + "Example n. 2824 = (33.0, 230.0): cluster 3\n", + "Example n. 2825 = (33.0, 227.0): cluster 3\n", + "Example n. 2826 = (33.0, 226.0): cluster 3\n", + "Example n. 2827 = (33.0, 225.0): cluster 3\n", + "Example n. 2828 = (33.0, 224.0): cluster 3\n", + "Example n. 2829 = (33.0, 223.0): cluster 1\n", + "Example n. 2830 = (33.0, 222.0): cluster 1\n", + "Example n. 2831 = (33.0, 214.0): cluster 1\n", + "Example n. 2832 = (33.0, 213.0): cluster 1\n", + "Example n. 2833 = (33.0, 212.0): cluster 1\n", + "Example n. 2834 = (33.0, 211.0): cluster 1\n", + "Example n. 2835 = (33.0, 210.0): cluster 1\n", + "Example n. 2836 = (33.0, 209.0): cluster 1\n", + "Example n. 2837 = (33.0, 201.0): cluster 1\n", + "Example n. 2838 = (33.0, 200.0): cluster 1\n", + "Example n. 2839 = (33.0, 199.0): cluster 1\n", + "Example n. 2840 = (33.0, 198.0): cluster 1\n", + "Example n. 2841 = (33.0, 197.0): cluster 1\n", + "Example n. 2842 = (33.0, 196.0): cluster 1\n", + "Example n. 2843 = (33.0, 195.0): cluster 1\n", + "Example n. 2844 = (33.0, 194.0): cluster 1\n", + "Example n. 2845 = (33.0, 193.0): cluster 1\n", + "Example n. 2846 = (33.0, 192.0): cluster 1\n", + "Example n. 2847 = (33.0, 183.0): cluster 1\n", + "Example n. 2848 = (33.0, 182.0): cluster 1\n", + "Example n. 2849 = (33.0, 181.0): cluster 1\n", + "Example n. 2850 = (33.0, 180.0): cluster 1\n", + "Example n. 2851 = (33.0, 179.0): cluster 1\n", + "Example n. 2852 = (33.0, 178.0): cluster 1\n", + "Example n. 2853 = (33.0, 177.0): cluster 1\n", + "Example n. 2854 = (33.0, 176.0): cluster 1\n", + "Example n. 2855 = (33.0, 175.0): cluster 1\n", + "Example n. 2856 = (33.0, 171.0): cluster 1\n", + "Example n. 2857 = (33.0, 170.0): cluster 1\n", + "Example n. 2858 = (33.0, 169.0): cluster 1\n", + "Example n. 2859 = (33.0, 168.0): cluster 1\n", + "Example n. 2860 = (33.0, 167.0): cluster 1\n", + "Example n. 2861 = (33.0, 166.0): cluster 1\n", + "Example n. 2862 = (33.0, 165.0): cluster 1\n", + "Example n. 2863 = (33.0, 164.0): cluster 3\n", + "Example n. 2864 = (33.0, 163.0): cluster 3\n", + "Example n. 2865 = (33.0, 162.0): cluster 3\n", + "Example n. 2866 = (33.0, 161.0): cluster 3\n", + "Example n. 2867 = (33.0, 160.0): cluster 3\n", + "Example n. 2868 = (33.0, 159.0): cluster 3\n", + "Example n. 2869 = (33.0, 158.0): cluster 3\n", + "Example n. 2870 = (33.0, 157.0): cluster 3\n", + "Example n. 2871 = (33.0, 156.0): cluster 3\n", + "Example n. 2872 = (33.0, 155.0): cluster 3\n", + "Example n. 2873 = (33.0, 154.0): cluster 3\n", + "Example n. 2874 = (33.0, 153.0): cluster 3\n", + "Example n. 2875 = (33.0, 136.0): cluster 3\n", + "Example n. 2876 = (33.0, 135.0): cluster 3\n", + "Example n. 2877 = (33.0, 134.0): cluster 3\n", + "Example n. 2878 = (33.0, 133.0): cluster 3\n", + "Example n. 2879 = (33.0, 132.0): cluster 3\n", + "Example n. 2880 = (33.0, 131.0): cluster 3\n", + "Example n. 2881 = (33.0, 130.0): cluster 3\n", + "Example n. 2882 = (33.0, 129.0): cluster 3\n", + "Example n. 2883 = (33.0, 128.0): cluster 3\n", + "Example n. 2884 = (34.0, 488.0): cluster 3\n", + "Example n. 2885 = (34.0, 487.0): cluster 3\n", + "Example n. 2886 = (34.0, 486.0): cluster 3\n", + "Example n. 2887 = (34.0, 485.0): cluster 3\n", + "Example n. 2888 = (34.0, 484.0): cluster 3\n", + "Example n. 2889 = (34.0, 483.0): cluster 3\n", + "Example n. 2890 = (34.0, 482.0): cluster 3\n", + "Example n. 2891 = (34.0, 481.0): cluster 3\n", + "Example n. 2892 = (34.0, 480.0): cluster 3\n", + "Example n. 2893 = (34.0, 479.0): cluster 3\n", + "Example n. 2894 = (34.0, 478.0): cluster 3\n", + "Example n. 2895 = (34.0, 477.0): cluster 3\n", + "Example n. 2896 = (34.0, 476.0): cluster 3\n", + "Example n. 2897 = (34.0, 475.0): cluster 3\n", + "Example n. 2898 = (34.0, 474.0): cluster 3\n", + "Example n. 2899 = (34.0, 473.0): cluster 3\n", + "Example n. 2900 = (34.0, 472.0): cluster 3\n", + "Example n. 2901 = (34.0, 471.0): cluster 1\n", + "Example n. 2902 = (34.0, 470.0): cluster 1\n", + "Example n. 2903 = (34.0, 469.0): cluster 1\n", + "Example n. 2904 = (34.0, 468.0): cluster 1\n", + "Example n. 2905 = (34.0, 467.0): cluster 1\n", + "Example n. 2906 = (34.0, 466.0): cluster 1\n", + "Example n. 2907 = (34.0, 465.0): cluster 1\n", + "Example n. 2908 = (34.0, 464.0): cluster 1\n", + "Example n. 2909 = (34.0, 463.0): cluster 1\n", + "Example n. 2910 = (34.0, 462.0): cluster 1\n", + "Example n. 2911 = (34.0, 461.0): cluster 1\n", + "Example n. 2912 = (34.0, 460.0): cluster 1\n", + "Example n. 2913 = (34.0, 459.0): cluster 1\n", + "Example n. 2914 = (34.0, 398.0): cluster 1\n", + "Example n. 2915 = (34.0, 397.0): cluster 1\n", + "Example n. 2916 = (34.0, 396.0): cluster 1\n", + "Example n. 2917 = (34.0, 395.0): cluster 1\n", + "Example n. 2918 = (34.0, 394.0): cluster 1\n", + "Example n. 2919 = (34.0, 393.0): cluster 1\n", + "Example n. 2920 = (34.0, 382.0): cluster 1\n", + "Example n. 2921 = (34.0, 381.0): cluster 1\n", + "Example n. 2922 = (34.0, 380.0): cluster 1\n", + "Example n. 2923 = (34.0, 379.0): cluster 1\n", + "Example n. 2924 = (34.0, 378.0): cluster 1\n", + "Example n. 2925 = (34.0, 377.0): cluster 1\n", + "Example n. 2926 = (34.0, 376.0): cluster 1\n", + "Example n. 2927 = (34.0, 306.0): cluster 1\n", + "Example n. 2928 = (34.0, 305.0): cluster 1\n", + "Example n. 2929 = (34.0, 304.0): cluster 1\n", + "Example n. 2930 = (34.0, 303.0): cluster 1\n", + "Example n. 2931 = (34.0, 302.0): cluster 1\n", + "Example n. 2932 = (34.0, 301.0): cluster 1\n", + "Example n. 2933 = (34.0, 300.0): cluster 1\n", + "Example n. 2934 = (34.0, 299.0): cluster 1\n", + "Example n. 2935 = (34.0, 298.0): cluster 1\n", + "Example n. 2936 = (34.0, 297.0): cluster 1\n", + "Example n. 2937 = (34.0, 294.0): cluster 1\n", + "Example n. 2938 = (34.0, 293.0): cluster 1\n", + "Example n. 2939 = (34.0, 292.0): cluster 1\n", + "Example n. 2940 = (34.0, 291.0): cluster 1\n", + "Example n. 2941 = (34.0, 290.0): cluster 1\n", + "Example n. 2942 = (34.0, 289.0): cluster 1\n", + "Example n. 2943 = (34.0, 288.0): cluster 1\n", + "Example n. 2944 = (34.0, 287.0): cluster 1\n", + "Example n. 2945 = (34.0, 286.0): cluster 1\n", + "Example n. 2946 = (34.0, 285.0): cluster 1\n", + "Example n. 2947 = (34.0, 284.0): cluster 1\n", + "Example n. 2948 = (34.0, 283.0): cluster 1\n", + "Example n. 2949 = (34.0, 282.0): cluster 1\n", + "Example n. 2950 = (34.0, 281.0): cluster 1\n", + "Example n. 2951 = (34.0, 280.0): cluster 1\n", + "Example n. 2952 = (34.0, 279.0): cluster 1\n", + "Example n. 2953 = (34.0, 278.0): cluster 1\n", + "Example n. 2954 = (34.0, 277.0): cluster 3\n", + "Example n. 2955 = (34.0, 276.0): cluster 3\n", + "Example n. 2956 = (34.0, 269.0): cluster 3\n", + "Example n. 2957 = (34.0, 268.0): cluster 3\n", + "Example n. 2958 = (34.0, 267.0): cluster 3\n", + "Example n. 2959 = (34.0, 266.0): cluster 3\n", + "Example n. 2960 = (34.0, 265.0): cluster 3\n", + "Example n. 2961 = (34.0, 264.0): cluster 3\n", + "Example n. 2962 = (34.0, 263.0): cluster 3\n", + "Example n. 2963 = (34.0, 262.0): cluster 3\n", + "Example n. 2964 = (34.0, 261.0): cluster 3\n", + "Example n. 2965 = (34.0, 260.0): cluster 3\n", + "Example n. 2966 = (34.0, 252.0): cluster 3\n", + "Example n. 2967 = (34.0, 251.0): cluster 3\n", + "Example n. 2968 = (34.0, 248.0): cluster 3\n", + "Example n. 2969 = (34.0, 247.0): cluster 3\n", + "Example n. 2970 = (34.0, 246.0): cluster 3\n", + "Example n. 2971 = (34.0, 245.0): cluster 3\n", + "Example n. 2972 = (34.0, 244.0): cluster 3\n", + "Example n. 2973 = (34.0, 243.0): cluster 3\n", + "Example n. 2974 = (34.0, 242.0): cluster 3\n", + "Example n. 2975 = (34.0, 241.0): cluster 3\n", + "Example n. 2976 = (34.0, 240.0): cluster 3\n", + "Example n. 2977 = (34.0, 239.0): cluster 3\n", + "Example n. 2978 = (34.0, 238.0): cluster 3\n", + "Example n. 2979 = (34.0, 237.0): cluster 3\n", + "Example n. 2980 = (34.0, 236.0): cluster 3\n", + "Example n. 2981 = (34.0, 235.0): cluster 3\n", + "Example n. 2982 = (34.0, 234.0): cluster 3\n", + "Example n. 2983 = (34.0, 233.0): cluster 3\n", + "Example n. 2984 = (34.0, 232.0): cluster 3\n", + "Example n. 2985 = (34.0, 231.0): cluster 3\n", + "Example n. 2986 = (34.0, 227.0): cluster 3\n", + "Example n. 2987 = (34.0, 226.0): cluster 3\n", + "Example n. 2988 = (34.0, 225.0): cluster 3\n", + "Example n. 2989 = (34.0, 224.0): cluster 3\n", + "Example n. 2990 = (34.0, 223.0): cluster 3\n", + "Example n. 2991 = (34.0, 222.0): cluster 3\n", + "Example n. 2992 = (34.0, 221.0): cluster 3\n", + "Example n. 2993 = (34.0, 220.0): cluster 3\n", + "Example n. 2994 = (34.0, 219.0): cluster 3\n", + "Example n. 2995 = (34.0, 218.0): cluster 1\n", + "Example n. 2996 = (34.0, 214.0): cluster 1\n", + "Example n. 2997 = (34.0, 213.0): cluster 1\n", + "Example n. 2998 = (34.0, 212.0): cluster 1\n", + "Example n. 2999 = (34.0, 211.0): cluster 1\n", + "Example n. 3000 = (34.0, 210.0): cluster 1\n", + "Example n. 3001 = (34.0, 187.0): cluster 1\n", + "Example n. 3002 = (34.0, 186.0): cluster 1\n", + "Example n. 3003 = (34.0, 185.0): cluster 1\n", + "Example n. 3004 = (34.0, 184.0): cluster 1\n", + "Example n. 3005 = (34.0, 183.0): cluster 1\n", + "Example n. 3006 = (34.0, 182.0): cluster 1\n", + "Example n. 3007 = (34.0, 181.0): cluster 1\n", + "Example n. 3008 = (34.0, 180.0): cluster 1\n", + "Example n. 3009 = (34.0, 179.0): cluster 1\n", + "Example n. 3010 = (34.0, 178.0): cluster 1\n", + "Example n. 3011 = (34.0, 177.0): cluster 1\n", + "Example n. 3012 = (34.0, 176.0): cluster 1\n", + "Example n. 3013 = (34.0, 171.0): cluster 1\n", + "Example n. 3014 = (34.0, 170.0): cluster 1\n", + "Example n. 3015 = (34.0, 169.0): cluster 1\n", + "Example n. 3016 = (34.0, 168.0): cluster 1\n", + "Example n. 3017 = (34.0, 167.0): cluster 1\n", + "Example n. 3018 = (34.0, 166.0): cluster 1\n", + "Example n. 3019 = (34.0, 165.0): cluster 1\n", + "Example n. 3020 = (34.0, 164.0): cluster 1\n", + "Example n. 3021 = (34.0, 163.0): cluster 1\n", + "Example n. 3022 = (34.0, 162.0): cluster 1\n", + "Example n. 3023 = (34.0, 161.0): cluster 1\n", + "Example n. 3024 = (34.0, 160.0): cluster 1\n", + "Example n. 3025 = (34.0, 159.0): cluster 1\n", + "Example n. 3026 = (34.0, 158.0): cluster 1\n", + "Example n. 3027 = (34.0, 157.0): cluster 1\n", + "Example n. 3028 = (34.0, 156.0): cluster 1\n", + "Example n. 3029 = (34.0, 155.0): cluster 1\n", + "Example n. 3030 = (34.0, 154.0): cluster 1\n", + "Example n. 3031 = (34.0, 153.0): cluster 1\n", + "Example n. 3032 = (34.0, 152.0): cluster 1\n", + "Example n. 3033 = (34.0, 151.0): cluster 1\n", + "Example n. 3034 = (34.0, 136.0): cluster 1\n", + "Example n. 3035 = (34.0, 135.0): cluster 1\n", + "Example n. 3036 = (34.0, 134.0): cluster 1\n", + "Example n. 3037 = (34.0, 133.0): cluster 1\n", + "Example n. 3038 = (34.0, 132.0): cluster 1\n", + "Example n. 3039 = (34.0, 131.0): cluster 1\n", + "Example n. 3040 = (34.0, 130.0): cluster 1\n", + "Example n. 3041 = (34.0, 129.0): cluster 1\n", + "Example n. 3042 = (34.0, 128.0): cluster 1\n", + "Example n. 3043 = (34.0, 113.0): cluster 1\n", + "Example n. 3044 = (34.0, 112.0): cluster 1\n", + "Example n. 3045 = (34.0, 111.0): cluster 1\n", + "Example n. 3046 = (34.0, 108.0): cluster 1\n", + "Example n. 3047 = (34.0, 107.0): cluster 1\n", + "Example n. 3048 = (34.0, 106.0): cluster 3\n", + "Example n. 3049 = (34.0, 105.0): cluster 3\n", + "Example n. 3050 = (35.0, 489.0): cluster 3\n", + "Example n. 3051 = (35.0, 488.0): cluster 3\n", + "Example n. 3052 = (35.0, 487.0): cluster 3\n", + "Example n. 3053 = (35.0, 486.0): cluster 3\n", + "Example n. 3054 = (35.0, 485.0): cluster 3\n", + "Example n. 3055 = (35.0, 484.0): cluster 3\n", + "Example n. 3056 = (35.0, 483.0): cluster 3\n", + "Example n. 3057 = (35.0, 482.0): cluster 3\n", + "Example n. 3058 = (35.0, 481.0): cluster 3\n", + "Example n. 3059 = (35.0, 480.0): cluster 3\n", + "Example n. 3060 = (35.0, 479.0): cluster 3\n", + "Example n. 3061 = (35.0, 478.0): cluster 3\n", + "Example n. 3062 = (35.0, 477.0): cluster 3\n", + "Example n. 3063 = (35.0, 476.0): cluster 3\n", + "Example n. 3064 = (35.0, 475.0): cluster 3\n", + "Example n. 3065 = (35.0, 474.0): cluster 3\n", + "Example n. 3066 = (35.0, 473.0): cluster 3\n", + "Example n. 3067 = (35.0, 472.0): cluster 3\n", + "Example n. 3068 = (35.0, 471.0): cluster 3\n", + "Example n. 3069 = (35.0, 470.0): cluster 3\n", + "Example n. 3070 = (35.0, 469.0): cluster 3\n", + "Example n. 3071 = (35.0, 468.0): cluster 3\n", + "Example n. 3072 = (35.0, 467.0): cluster 3\n", + "Example n. 3073 = (35.0, 466.0): cluster 3\n", + "Example n. 3074 = (35.0, 465.0): cluster 3\n", + "Example n. 3075 = (35.0, 464.0): cluster 3\n", + "Example n. 3076 = (35.0, 463.0): cluster 3\n", + "Example n. 3077 = (35.0, 462.0): cluster 3\n", + "Example n. 3078 = (35.0, 461.0): cluster 3\n", + "Example n. 3079 = (35.0, 460.0): cluster 3\n", + "Example n. 3080 = (35.0, 459.0): cluster 3\n", + "Example n. 3081 = (35.0, 458.0): cluster 3\n", + "Example n. 3082 = (35.0, 397.0): cluster 3\n", + "Example n. 3083 = (35.0, 396.0): cluster 3\n", + "Example n. 3084 = (35.0, 395.0): cluster 3\n", + "Example n. 3085 = (35.0, 394.0): cluster 1\n", + "Example n. 3086 = (35.0, 393.0): cluster 1\n", + "Example n. 3087 = (35.0, 382.0): cluster 1\n", + "Example n. 3088 = (35.0, 381.0): cluster 1\n", + "Example n. 3089 = (35.0, 380.0): cluster 1\n", + "Example n. 3090 = (35.0, 379.0): cluster 1\n", + "Example n. 3091 = (35.0, 378.0): cluster 1\n", + "Example n. 3092 = (35.0, 377.0): cluster 1\n", + "Example n. 3093 = (35.0, 376.0): cluster 1\n", + "Example n. 3094 = (35.0, 375.0): cluster 1\n", + "Example n. 3095 = (35.0, 307.0): cluster 1\n", + "Example n. 3096 = (35.0, 306.0): cluster 1\n", + "Example n. 3097 = (35.0, 305.0): cluster 1\n", + "Example n. 3098 = (35.0, 304.0): cluster 1\n", + "Example n. 3099 = (35.0, 303.0): cluster 1\n", + "Example n. 3100 = (35.0, 302.0): cluster 1\n", + "Example n. 3101 = (35.0, 301.0): cluster 1\n", + "Example n. 3102 = (35.0, 300.0): cluster 1\n", + "Example n. 3103 = (35.0, 299.0): cluster 1\n", + "Example n. 3104 = (35.0, 298.0): cluster 1\n", + "Example n. 3105 = (35.0, 297.0): cluster 1\n", + "Example n. 3106 = (35.0, 292.0): cluster 1\n", + "Example n. 3107 = (35.0, 291.0): cluster 1\n", + "Example n. 3108 = (35.0, 290.0): cluster 1\n", + "Example n. 3109 = (35.0, 289.0): cluster 1\n", + "Example n. 3110 = (35.0, 288.0): cluster 1\n", + "Example n. 3111 = (35.0, 287.0): cluster 1\n", + "Example n. 3112 = (35.0, 286.0): cluster 1\n", + "Example n. 3113 = (35.0, 285.0): cluster 1\n", + "Example n. 3114 = (35.0, 284.0): cluster 1\n", + "Example n. 3115 = (35.0, 283.0): cluster 1\n", + "Example n. 3116 = (35.0, 282.0): cluster 1\n", + "Example n. 3117 = (35.0, 281.0): cluster 1\n", + "Example n. 3118 = (35.0, 280.0): cluster 1\n", + "Example n. 3119 = (35.0, 279.0): cluster 1\n", + "Example n. 3120 = (35.0, 278.0): cluster 1\n", + "Example n. 3121 = (35.0, 277.0): cluster 1\n", + "Example n. 3122 = (35.0, 269.0): cluster 1\n", + "Example n. 3123 = (35.0, 268.0): cluster 1\n", + "Example n. 3124 = (35.0, 267.0): cluster 1\n", + "Example n. 3125 = (35.0, 266.0): cluster 1\n", + "Example n. 3126 = (35.0, 265.0): cluster 1\n", + "Example n. 3127 = (35.0, 264.0): cluster 1\n", + "Example n. 3128 = (35.0, 263.0): cluster 1\n", + "Example n. 3129 = (35.0, 262.0): cluster 1\n", + "Example n. 3130 = (35.0, 261.0): cluster 1\n", + "Example n. 3131 = (35.0, 260.0): cluster 1\n", + "Example n. 3132 = (35.0, 259.0): cluster 1\n", + "Example n. 3133 = (35.0, 258.0): cluster 1\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Example n. 3134 = (35.0, 247.0): cluster 1\n", + "Example n. 3135 = (35.0, 246.0): cluster 1\n", + "Example n. 3136 = (35.0, 245.0): cluster 1\n", + "Example n. 3137 = (35.0, 244.0): cluster 1\n", + "Example n. 3138 = (35.0, 243.0): cluster 3\n", + "Example n. 3139 = (35.0, 242.0): cluster 3\n", + "Example n. 3140 = (35.0, 241.0): cluster 3\n", + "Example n. 3141 = (35.0, 240.0): cluster 3\n", + "Example n. 3142 = (35.0, 239.0): cluster 3\n", + "Example n. 3143 = (35.0, 238.0): cluster 3\n", + "Example n. 3144 = (35.0, 237.0): cluster 3\n", + "Example n. 3145 = (35.0, 236.0): cluster 3\n", + "Example n. 3146 = (35.0, 235.0): cluster 3\n", + "Example n. 3147 = (35.0, 234.0): cluster 3\n", + "Example n. 3148 = (35.0, 233.0): cluster 3\n", + "Example n. 3149 = (35.0, 232.0): cluster 3\n", + "Example n. 3150 = (35.0, 226.0): cluster 3\n", + "Example n. 3151 = (35.0, 225.0): cluster 3\n", + "Example n. 3152 = (35.0, 224.0): cluster 3\n", + "Example n. 3153 = (35.0, 223.0): cluster 3\n", + "Example n. 3154 = (35.0, 222.0): cluster 3\n", + "Example n. 3155 = (35.0, 221.0): cluster 3\n", + "Example n. 3156 = (35.0, 220.0): cluster 3\n", + "Example n. 3157 = (35.0, 219.0): cluster 3\n", + "Example n. 3158 = (35.0, 218.0): cluster 3\n", + "Example n. 3159 = (35.0, 214.0): cluster 3\n", + "Example n. 3160 = (35.0, 213.0): cluster 3\n", + "Example n. 3161 = (35.0, 212.0): cluster 3\n", + "Example n. 3162 = (35.0, 211.0): cluster 3\n", + "Example n. 3163 = (35.0, 210.0): cluster 3\n", + "Example n. 3164 = (35.0, 193.0): cluster 3\n", + "Example n. 3165 = (35.0, 192.0): cluster 3\n", + "Example n. 3166 = (35.0, 191.0): cluster 3\n", + "Example n. 3167 = (35.0, 190.0): cluster 3\n", + "Example n. 3168 = (35.0, 187.0): cluster 3\n", + "Example n. 3169 = (35.0, 186.0): cluster 3\n", + "Example n. 3170 = (35.0, 185.0): cluster 3\n", + "Example n. 3171 = (35.0, 184.0): cluster 3\n", + "Example n. 3172 = (35.0, 183.0): cluster 3\n", + "Example n. 3173 = (35.0, 182.0): cluster 3\n", + "Example n. 3174 = (35.0, 181.0): cluster 3\n", + "Example n. 3175 = (35.0, 180.0): cluster 1\n", + "Example n. 3176 = (35.0, 179.0): cluster 1\n", + "Example n. 3177 = (35.0, 171.0): cluster 1\n", + "Example n. 3178 = (35.0, 170.0): cluster 1\n", + "Example n. 3179 = (35.0, 169.0): cluster 1\n", + "Example n. 3180 = (35.0, 168.0): cluster 1\n", + "Example n. 3181 = (35.0, 167.0): cluster 1\n", + "Example n. 3182 = (35.0, 166.0): cluster 1\n", + "Example n. 3183 = (35.0, 165.0): cluster 1\n", + "Example n. 3184 = (35.0, 164.0): cluster 1\n", + "Example n. 3185 = (35.0, 163.0): cluster 1\n", + "Example n. 3186 = (35.0, 162.0): cluster 1\n", + "Example n. 3187 = (35.0, 161.0): cluster 1\n", + "Example n. 3188 = (35.0, 160.0): cluster 1\n", + "Example n. 3189 = (35.0, 159.0): cluster 1\n", + "Example n. 3190 = (35.0, 158.0): cluster 1\n", + "Example n. 3191 = (35.0, 157.0): cluster 1\n", + "Example n. 3192 = (35.0, 156.0): cluster 1\n", + "Example n. 3193 = (35.0, 155.0): cluster 1\n", + "Example n. 3194 = (35.0, 154.0): cluster 1\n", + "Example n. 3195 = (35.0, 153.0): cluster 1\n", + "Example n. 3196 = (35.0, 152.0): cluster 1\n", + "Example n. 3197 = (35.0, 151.0): cluster 1\n", + "Example n. 3198 = (35.0, 150.0): cluster 1\n", + "Example n. 3199 = (35.0, 136.0): cluster 1\n", + "Example n. 3200 = (35.0, 135.0): cluster 1\n", + "Example n. 3201 = (35.0, 134.0): cluster 1\n", + "Example n. 3202 = (35.0, 133.0): cluster 1\n", + "Example n. 3203 = (35.0, 132.0): cluster 1\n", + "Example n. 3204 = (35.0, 131.0): cluster 1\n", + "Example n. 3205 = (35.0, 130.0): cluster 1\n", + "Example n. 3206 = (35.0, 129.0): cluster 1\n", + "Example n. 3207 = (35.0, 128.0): cluster 1\n", + "Example n. 3208 = (35.0, 127.0): cluster 1\n", + "Example n. 3209 = (35.0, 114.0): cluster 1\n", + "Example n. 3210 = (35.0, 113.0): cluster 1\n", + "Example n. 3211 = (35.0, 112.0): cluster 1\n", + "Example n. 3212 = (35.0, 111.0): cluster 1\n", + "Example n. 3213 = (35.0, 110.0): cluster 1\n", + "Example n. 3214 = (35.0, 109.0): cluster 1\n", + "Example n. 3215 = (35.0, 108.0): cluster 1\n", + "Example n. 3216 = (35.0, 107.0): cluster 1\n", + "Example n. 3217 = (35.0, 106.0): cluster 1\n", + "Example n. 3218 = (35.0, 105.0): cluster 1\n", + "Example n. 3219 = (35.0, 104.0): cluster 1\n", + "Example n. 3220 = (36.0, 489.0): cluster 1\n", + "Example n. 3221 = (36.0, 488.0): cluster 1\n", + "Example n. 3222 = (36.0, 487.0): cluster 1\n", + "Example n. 3223 = (36.0, 486.0): cluster 1\n", + "Example n. 3224 = (36.0, 485.0): cluster 1\n", + "Example n. 3225 = (36.0, 484.0): cluster 1\n", + "Example n. 3226 = (36.0, 483.0): cluster 1\n", + "Example n. 3227 = (36.0, 482.0): cluster 1\n", + "Example n. 3228 = (36.0, 481.0): cluster 3\n", + "Example n. 3229 = (36.0, 480.0): cluster 3\n", + "Example n. 3230 = (36.0, 479.0): cluster 3\n", + "Example n. 3231 = (36.0, 478.0): cluster 3\n", + "Example n. 3232 = (36.0, 477.0): cluster 3\n", + "Example n. 3233 = (36.0, 476.0): cluster 3\n", + "Example n. 3234 = (36.0, 475.0): cluster 3\n", + "Example n. 3235 = (36.0, 474.0): cluster 3\n", + "Example n. 3236 = (36.0, 473.0): cluster 3\n", + "Example n. 3237 = (36.0, 472.0): cluster 3\n", + "Example n. 3238 = (36.0, 471.0): cluster 3\n", + "Example n. 3239 = (36.0, 470.0): cluster 3\n", + "Example n. 3240 = (36.0, 469.0): cluster 3\n", + "Example n. 3241 = (36.0, 468.0): cluster 3\n", + "Example n. 3242 = (36.0, 467.0): cluster 3\n", + "Example n. 3243 = (36.0, 466.0): cluster 3\n", + "Example n. 3244 = (36.0, 465.0): cluster 3\n", + "Example n. 3245 = (36.0, 464.0): cluster 3\n", + "Example n. 3246 = (36.0, 463.0): cluster 3\n", + "Example n. 3247 = (36.0, 462.0): cluster 3\n", + "Example n. 3248 = (36.0, 461.0): cluster 3\n", + "Example n. 3249 = (36.0, 460.0): cluster 3\n", + "Example n. 3250 = (36.0, 459.0): cluster 3\n", + "Example n. 3251 = (36.0, 458.0): cluster 3\n", + "Example n. 3252 = (36.0, 396.0): cluster 3\n", + "Example n. 3253 = (36.0, 395.0): cluster 3\n", + "Example n. 3254 = (36.0, 394.0): cluster 3\n", + "Example n. 3255 = (36.0, 382.0): cluster 3\n", + "Example n. 3256 = (36.0, 381.0): cluster 3\n", + "Example n. 3257 = (36.0, 380.0): cluster 3\n", + "Example n. 3258 = (36.0, 379.0): cluster 3\n", + "Example n. 3259 = (36.0, 378.0): cluster 3\n", + "Example n. 3260 = (36.0, 377.0): cluster 3\n", + "Example n. 3261 = (36.0, 376.0): cluster 3\n", + "Example n. 3262 = (36.0, 375.0): cluster 3\n", + "Example n. 3263 = (36.0, 312.0): cluster 3\n", + "Example n. 3264 = (36.0, 311.0): cluster 3\n", + "Example n. 3265 = (36.0, 310.0): cluster 1\n", + "Example n. 3266 = (36.0, 309.0): cluster 1\n", + "Example n. 3267 = (36.0, 308.0): cluster 1\n", + "Example n. 3268 = (36.0, 307.0): cluster 1\n", + "Example n. 3269 = (36.0, 306.0): cluster 1\n", + "Example n. 3270 = (36.0, 305.0): cluster 1\n", + "Example n. 3271 = (36.0, 304.0): cluster 1\n", + "Example n. 3272 = (36.0, 303.0): cluster 1\n", + "Example n. 3273 = (36.0, 302.0): cluster 1\n", + "Example n. 3274 = (36.0, 301.0): cluster 1\n", + "Example n. 3275 = (36.0, 300.0): cluster 1\n", + "Example n. 3276 = (36.0, 299.0): cluster 1\n", + "Example n. 3277 = (36.0, 298.0): cluster 1\n", + "Example n. 3278 = (36.0, 297.0): cluster 1\n", + "Example n. 3279 = (36.0, 291.0): cluster 1\n", + "Example n. 3280 = (36.0, 290.0): cluster 1\n", + "Example n. 3281 = (36.0, 289.0): cluster 1\n", + "Example n. 3282 = (36.0, 288.0): cluster 1\n", + "Example n. 3283 = (36.0, 287.0): cluster 1\n", + "Example n. 3284 = (36.0, 286.0): cluster 1\n", + "Example n. 3285 = (36.0, 285.0): cluster 1\n", + "Example n. 3286 = (36.0, 284.0): cluster 1\n", + "Example n. 3287 = (36.0, 283.0): cluster 1\n", + "Example n. 3288 = (36.0, 282.0): cluster 1\n", + "Example n. 3289 = (36.0, 281.0): cluster 1\n", + "Example n. 3290 = (36.0, 280.0): cluster 1\n", + "Example n. 3291 = (36.0, 279.0): cluster 1\n", + "Example n. 3292 = (36.0, 278.0): cluster 1\n", + "Example n. 3293 = (36.0, 277.0): cluster 1\n", + "Example n. 3294 = (36.0, 269.0): cluster 1\n", + "Example n. 3295 = (36.0, 268.0): cluster 1\n", + "Example n. 3296 = (36.0, 267.0): cluster 1\n", + "Example n. 3297 = (36.0, 266.0): cluster 1\n", + "Example n. 3298 = (36.0, 265.0): cluster 1\n", + "Example n. 3299 = (36.0, 264.0): cluster 1\n", + "Example n. 3300 = (36.0, 263.0): cluster 1\n", + "Example n. 3301 = (36.0, 262.0): cluster 1\n", + "Example n. 3302 = (36.0, 261.0): cluster 1\n", + "Example n. 3303 = (36.0, 260.0): cluster 1\n", + "Example n. 3304 = (36.0, 259.0): cluster 1\n", + "Example n. 3305 = (36.0, 258.0): cluster 1\n", + "Example n. 3306 = (36.0, 257.0): cluster 1\n", + "Example n. 3307 = (36.0, 256.0): cluster 1\n", + "Example n. 3308 = (36.0, 255.0): cluster 1\n", + "Example n. 3309 = (36.0, 254.0): cluster 1\n", + "Example n. 3310 = (36.0, 253.0): cluster 1\n", + "Example n. 3311 = (36.0, 245.0): cluster 1\n", + "Example n. 3312 = (36.0, 244.0): cluster 1\n", + "Example n. 3313 = (36.0, 243.0): cluster 1\n", + "Example n. 3314 = (36.0, 242.0): cluster 1\n", + "Example n. 3315 = (36.0, 241.0): cluster 1\n", + "Example n. 3316 = (36.0, 240.0): cluster 1\n", + "Example n. 3317 = (36.0, 239.0): cluster 1\n", + "Example n. 3318 = (36.0, 238.0): cluster 3\n", + "Example n. 3319 = (36.0, 237.0): cluster 3\n", + "Example n. 3320 = (36.0, 236.0): cluster 3\n", + "Example n. 3321 = (36.0, 235.0): cluster 3\n", + "Example n. 3322 = (36.0, 234.0): cluster 3\n", + "Example n. 3323 = (36.0, 226.0): cluster 3\n", + "Example n. 3324 = (36.0, 225.0): cluster 3\n", + "Example n. 3325 = (36.0, 224.0): cluster 3\n", + "Example n. 3326 = (36.0, 223.0): cluster 3\n", + "Example n. 3327 = (36.0, 222.0): cluster 3\n", + "Example n. 3328 = (36.0, 221.0): cluster 3\n", + "Example n. 3329 = (36.0, 220.0): cluster 3\n", + "Example n. 3330 = (36.0, 219.0): cluster 3\n", + "Example n. 3331 = (36.0, 218.0): cluster 3\n", + "Example n. 3332 = (36.0, 214.0): cluster 3\n", + "Example n. 3333 = (36.0, 213.0): cluster 3\n", + "Example n. 3334 = (36.0, 212.0): cluster 3\n", + "Example n. 3335 = (36.0, 211.0): cluster 3\n", + "Example n. 3336 = (36.0, 210.0): cluster 3\n", + "Example n. 3337 = (36.0, 195.0): cluster 3\n", + "Example n. 3338 = (36.0, 194.0): cluster 3\n", + "Example n. 3339 = (36.0, 193.0): cluster 3\n", + "Example n. 3340 = (36.0, 192.0): cluster 3\n", + "Example n. 3341 = (36.0, 191.0): cluster 3\n", + "Example n. 3342 = (36.0, 190.0): cluster 3\n", + "Example n. 3343 = (36.0, 189.0): cluster 3\n", + "Example n. 3344 = (36.0, 188.0): cluster 3\n", + "Example n. 3345 = (36.0, 187.0): cluster 3\n", + "Example n. 3346 = (36.0, 186.0): cluster 3\n", + "Example n. 3347 = (36.0, 185.0): cluster 3\n", + "Example n. 3348 = (36.0, 184.0): cluster 3\n", + "Example n. 3349 = (36.0, 183.0): cluster 3\n", + "Example n. 3350 = (36.0, 182.0): cluster 3\n", + "Example n. 3351 = (36.0, 181.0): cluster 3\n", + "Example n. 3352 = (36.0, 180.0): cluster 3\n", + "Example n. 3353 = (36.0, 179.0): cluster 3\n", + "Example n. 3354 = (36.0, 178.0): cluster 3\n", + "Example n. 3355 = (36.0, 170.0): cluster 3\n", + "Example n. 3356 = (36.0, 169.0): cluster 3\n", + "Example n. 3357 = (36.0, 168.0): cluster 3\n", + "Example n. 3358 = (36.0, 167.0): cluster 3\n", + "Example n. 3359 = (36.0, 166.0): cluster 1\n", + "Example n. 3360 = (36.0, 165.0): cluster 1\n", + "Example n. 3361 = (36.0, 163.0): cluster 1\n", + "Example n. 3362 = (36.0, 162.0): cluster 1\n", + "Example n. 3363 = (36.0, 161.0): cluster 1\n", + "Example n. 3364 = (36.0, 160.0): cluster 1\n", + "Example n. 3365 = (36.0, 159.0): cluster 1\n", + "Example n. 3366 = (36.0, 158.0): cluster 1\n", + "Example n. 3367 = (36.0, 157.0): cluster 1\n", + "Example n. 3368 = (36.0, 156.0): cluster 1\n", + "Example n. 3369 = (36.0, 155.0): cluster 1\n", + "Example n. 3370 = (36.0, 154.0): cluster 1\n", + "Example n. 3371 = (36.0, 153.0): cluster 1\n", + "Example n. 3372 = (36.0, 152.0): cluster 1\n", + "Example n. 3373 = (36.0, 151.0): cluster 1\n", + "Example n. 3374 = (36.0, 150.0): cluster 1\n", + "Example n. 3375 = (36.0, 149.0): cluster 1\n", + "Example n. 3376 = (36.0, 148.0): cluster 1\n", + "Example n. 3377 = (36.0, 147.0): cluster 1\n", + "Example n. 3378 = (36.0, 136.0): cluster 1\n", + "Example n. 3379 = (36.0, 135.0): cluster 1\n", + "Example n. 3380 = (36.0, 134.0): cluster 1\n", + "Example n. 3381 = (36.0, 133.0): cluster 1\n", + "Example n. 3382 = (36.0, 132.0): cluster 1\n", + "Example n. 3383 = (36.0, 131.0): cluster 1\n", + "Example n. 3384 = (36.0, 130.0): cluster 1\n", + "Example n. 3385 = (36.0, 129.0): cluster 1\n", + "Example n. 3386 = (36.0, 128.0): cluster 1\n", + "Example n. 3387 = (36.0, 127.0): cluster 1\n", + "Example n. 3388 = (36.0, 126.0): cluster 1\n", + "Example n. 3389 = (36.0, 114.0): cluster 1\n", + "Example n. 3390 = (36.0, 113.0): cluster 1\n", + "Example n. 3391 = (36.0, 112.0): cluster 1\n", + "Example n. 3392 = (36.0, 111.0): cluster 1\n", + "Example n. 3393 = (36.0, 110.0): cluster 1\n", + "Example n. 3394 = (36.0, 109.0): cluster 1\n", + "Example n. 3395 = (36.0, 108.0): cluster 1\n", + "Example n. 3396 = (36.0, 107.0): cluster 1\n", + "Example n. 3397 = (36.0, 106.0): cluster 1\n", + "Example n. 3398 = (36.0, 105.0): cluster 1\n", + "Example n. 3399 = (36.0, 104.0): cluster 1\n", + "Example n. 3400 = (37.0, 490.0): cluster 1\n", + "Example n. 3401 = (37.0, 489.0): cluster 1\n", + "Example n. 3402 = (37.0, 488.0): cluster 1\n", + "Example n. 3403 = (37.0, 487.0): cluster 1\n", + "Example n. 3404 = (37.0, 486.0): cluster 3\n", + "Example n. 3405 = (37.0, 485.0): cluster 3\n", + "Example n. 3406 = (37.0, 484.0): cluster 3\n", + "Example n. 3407 = (37.0, 483.0): cluster 3\n", + "Example n. 3408 = (37.0, 482.0): cluster 3\n", + "Example n. 3409 = (37.0, 481.0): cluster 3\n", + "Example n. 3410 = (37.0, 480.0): cluster 3\n", + "Example n. 3411 = (37.0, 479.0): cluster 3\n", + "Example n. 3412 = (37.0, 478.0): cluster 3\n", + "Example n. 3413 = (37.0, 477.0): cluster 3\n", + "Example n. 3414 = (37.0, 476.0): cluster 3\n", + "Example n. 3415 = (37.0, 475.0): cluster 3\n", + "Example n. 3416 = (37.0, 474.0): cluster 3\n", + "Example n. 3417 = (37.0, 473.0): cluster 3\n", + "Example n. 3418 = (37.0, 472.0): cluster 3\n", + "Example n. 3419 = (37.0, 471.0): cluster 3\n", + "Example n. 3420 = (37.0, 470.0): cluster 3\n", + "Example n. 3421 = (37.0, 469.0): cluster 3\n", + "Example n. 3422 = (37.0, 468.0): cluster 3\n", + "Example n. 3423 = (37.0, 467.0): cluster 3\n", + "Example n. 3424 = (37.0, 466.0): cluster 3\n", + "Example n. 3425 = (37.0, 464.0): cluster 3\n", + "Example n. 3426 = (37.0, 463.0): cluster 3\n", + "Example n. 3427 = (37.0, 462.0): cluster 3\n", + "Example n. 3428 = (37.0, 461.0): cluster 3\n", + "Example n. 3429 = (37.0, 460.0): cluster 3\n", + "Example n. 3430 = (37.0, 459.0): cluster 3\n", + "Example n. 3431 = (37.0, 458.0): cluster 3\n", + "Example n. 3432 = (37.0, 457.0): cluster 3\n", + "Example n. 3433 = (37.0, 382.0): cluster 3\n", + "Example n. 3434 = (37.0, 381.0): cluster 3\n", + "Example n. 3435 = (37.0, 380.0): cluster 3\n", + "Example n. 3436 = (37.0, 379.0): cluster 3\n", + "Example n. 3437 = (37.0, 378.0): cluster 3\n", + "Example n. 3438 = (37.0, 377.0): cluster 1\n", + "Example n. 3439 = (37.0, 376.0): cluster 1\n", + "Example n. 3440 = (37.0, 312.0): cluster 1\n", + "Example n. 3441 = (37.0, 311.0): cluster 1\n", + "Example n. 3442 = (37.0, 310.0): cluster 1\n", + "Example n. 3443 = (37.0, 309.0): cluster 1\n", + "Example n. 3444 = (37.0, 308.0): cluster 1\n", + "Example n. 3445 = (37.0, 307.0): cluster 1\n", + "Example n. 3446 = (37.0, 306.0): cluster 1\n", + "Example n. 3447 = (37.0, 305.0): cluster 1\n", + "Example n. 3448 = (37.0, 304.0): cluster 1\n", + "Example n. 3449 = (37.0, 303.0): cluster 1\n", + "Example n. 3450 = (37.0, 302.0): cluster 1\n", + "Example n. 3451 = (37.0, 301.0): cluster 1\n", + "Example n. 3452 = (37.0, 300.0): cluster 1\n", + "Example n. 3453 = (37.0, 299.0): cluster 1\n", + "Example n. 3454 = (37.0, 298.0): cluster 1\n", + "Example n. 3455 = (37.0, 297.0): cluster 1\n", + "Example n. 3456 = (37.0, 293.0): cluster 1\n", + "Example n. 3457 = (37.0, 292.0): cluster 1\n", + "Example n. 3458 = (37.0, 291.0): cluster 1\n", + "Example n. 3459 = (37.0, 290.0): cluster 1\n", + "Example n. 3460 = (37.0, 289.0): cluster 1\n", + "Example n. 3461 = (37.0, 288.0): cluster 1\n", + "Example n. 3462 = (37.0, 287.0): cluster 1\n", + "Example n. 3463 = (37.0, 286.0): cluster 1\n", + "Example n. 3464 = (37.0, 285.0): cluster 1\n", + "Example n. 3465 = (37.0, 284.0): cluster 1\n", + "Example n. 3466 = (37.0, 281.0): cluster 1\n", + "Example n. 3467 = (37.0, 280.0): cluster 1\n", + "Example n. 3468 = (37.0, 279.0): cluster 1\n", + "Example n. 3469 = (37.0, 278.0): cluster 1\n", + "Example n. 3470 = (37.0, 277.0): cluster 1\n", + "Example n. 3471 = (37.0, 270.0): cluster 1\n", + "Example n. 3472 = (37.0, 269.0): cluster 1\n", + "Example n. 3473 = (37.0, 268.0): cluster 1\n", + "Example n. 3474 = (37.0, 267.0): cluster 1\n", + "Example n. 3475 = (37.0, 266.0): cluster 1\n", + "Example n. 3476 = (37.0, 265.0): cluster 1\n", + "Example n. 3477 = (37.0, 264.0): cluster 1\n", + "Example n. 3478 = (37.0, 263.0): cluster 1\n", + "Example n. 3479 = (37.0, 262.0): cluster 1\n", + "Example n. 3480 = (37.0, 261.0): cluster 1\n", + "Example n. 3481 = (37.0, 260.0): cluster 1\n", + "Example n. 3482 = (37.0, 259.0): cluster 1\n", + "Example n. 3483 = (37.0, 258.0): cluster 3\n", + "Example n. 3484 = (37.0, 257.0): cluster 3\n", + "Example n. 3485 = (37.0, 256.0): cluster 3\n", + "Example n. 3486 = (37.0, 255.0): cluster 3\n", + "Example n. 3487 = (37.0, 254.0): cluster 3\n", + "Example n. 3488 = (37.0, 253.0): cluster 3\n", + "Example n. 3489 = (37.0, 252.0): cluster 3\n", + "Example n. 3490 = (37.0, 245.0): cluster 3\n", + "Example n. 3491 = (37.0, 244.0): cluster 3\n", + "Example n. 3492 = (37.0, 243.0): cluster 3\n", + "Example n. 3493 = (37.0, 242.0): cluster 3\n", + "Example n. 3494 = (37.0, 241.0): cluster 3\n", + "Example n. 3495 = (37.0, 240.0): cluster 3\n", + "Example n. 3496 = (37.0, 239.0): cluster 3\n", + "Example n. 3497 = (37.0, 238.0): cluster 3\n", + "Example n. 3498 = (37.0, 237.0): cluster 3\n", + "Example n. 3499 = (37.0, 236.0): cluster 3\n", + "Example n. 3500 = (37.0, 235.0): cluster 3\n", + "Example n. 3501 = (37.0, 230.0): cluster 3\n", + "Example n. 3502 = (37.0, 229.0): cluster 3\n", + "Example n. 3503 = (37.0, 228.0): cluster 3\n", + "Example n. 3504 = (37.0, 227.0): cluster 3\n", + "Example n. 3505 = (37.0, 226.0): cluster 3\n", + "Example n. 3506 = (37.0, 225.0): cluster 3\n", + "Example n. 3507 = (37.0, 224.0): cluster 3\n", + "Example n. 3508 = (37.0, 223.0): cluster 3\n", + "Example n. 3509 = (37.0, 222.0): cluster 3\n", + "Example n. 3510 = (37.0, 221.0): cluster 3\n", + "Example n. 3511 = (37.0, 220.0): cluster 3\n", + "Example n. 3512 = (37.0, 219.0): cluster 3\n", + "Example n. 3513 = (37.0, 218.0): cluster 3\n", + "Example n. 3514 = (37.0, 214.0): cluster 3\n", + "Example n. 3515 = (37.0, 213.0): cluster 3\n", + "Example n. 3516 = (37.0, 212.0): cluster 3\n", + "Example n. 3517 = (37.0, 211.0): cluster 1\n", + "Example n. 3518 = (37.0, 210.0): cluster 1\n", + "Example n. 3519 = (37.0, 209.0): cluster 1\n", + "Example n. 3520 = (37.0, 208.0): cluster 1\n", + "Example n. 3521 = (37.0, 196.0): cluster 1\n", + "Example n. 3522 = (37.0, 195.0): cluster 1\n", + "Example n. 3523 = (37.0, 194.0): cluster 1\n", + "Example n. 3524 = (37.0, 193.0): cluster 1\n", + "Example n. 3525 = (37.0, 192.0): cluster 1\n", + "Example n. 3526 = (37.0, 191.0): cluster 1\n", + "Example n. 3527 = (37.0, 190.0): cluster 1\n", + "Example n. 3528 = (37.0, 189.0): cluster 1\n", + "Example n. 3529 = (37.0, 188.0): cluster 1\n", + "Example n. 3530 = (37.0, 187.0): cluster 1\n", + "Example n. 3531 = (37.0, 186.0): cluster 1\n", + "Example n. 3532 = (37.0, 185.0): cluster 1\n", + "Example n. 3533 = (37.0, 184.0): cluster 1\n", + "Example n. 3534 = (37.0, 183.0): cluster 1\n", + "Example n. 3535 = (37.0, 182.0): cluster 1\n", + "Example n. 3536 = (37.0, 181.0): cluster 1\n", + "Example n. 3537 = (37.0, 180.0): cluster 1\n", + "Example n. 3538 = (37.0, 179.0): cluster 1\n", + "Example n. 3539 = (37.0, 178.0): cluster 1\n", + "Example n. 3540 = (37.0, 177.0): cluster 1\n", + "Example n. 3541 = (37.0, 163.0): cluster 1\n", + "Example n. 3542 = (37.0, 162.0): cluster 1\n", + "Example n. 3543 = (37.0, 161.0): cluster 1\n", + "Example n. 3544 = (37.0, 160.0): cluster 1\n", + "Example n. 3545 = (37.0, 159.0): cluster 1\n", + "Example n. 3546 = (37.0, 158.0): cluster 1\n", + "Example n. 3547 = (37.0, 157.0): cluster 1\n", + "Example n. 3548 = (37.0, 156.0): cluster 1\n", + "Example n. 3549 = (37.0, 155.0): cluster 1\n", + "Example n. 3550 = (37.0, 154.0): cluster 1\n", + "Example n. 3551 = (37.0, 153.0): cluster 1\n", + "Example n. 3552 = (37.0, 152.0): cluster 1\n", + "Example n. 3553 = (37.0, 151.0): cluster 1\n", + "Example n. 3554 = (37.0, 150.0): cluster 1\n", + "Example n. 3555 = (37.0, 149.0): cluster 1\n", + "Example n. 3556 = (37.0, 148.0): cluster 1\n", + "Example n. 3557 = (37.0, 147.0): cluster 1\n", + "Example n. 3558 = (37.0, 146.0): cluster 1\n", + "Example n. 3559 = (37.0, 132.0): cluster 1\n", + "Example n. 3560 = (37.0, 131.0): cluster 1\n", + "Example n. 3561 = (37.0, 130.0): cluster 1\n", + "Example n. 3562 = (37.0, 129.0): cluster 3\n", + "Example n. 3563 = (37.0, 128.0): cluster 3\n", + "Example n. 3564 = (37.0, 127.0): cluster 3\n", + "Example n. 3565 = (37.0, 126.0): cluster 3\n", + "Example n. 3566 = (37.0, 114.0): cluster 3\n", + "Example n. 3567 = (37.0, 113.0): cluster 3\n", + "Example n. 3568 = (37.0, 112.0): cluster 3\n", + "Example n. 3569 = (37.0, 111.0): cluster 3\n", + "Example n. 3570 = (37.0, 110.0): cluster 3\n", + "Example n. 3571 = (37.0, 109.0): cluster 3\n", + "Example n. 3572 = (37.0, 108.0): cluster 3\n", + "Example n. 3573 = (37.0, 107.0): cluster 3\n", + "Example n. 3574 = (37.0, 106.0): cluster 3\n", + "Example n. 3575 = (37.0, 105.0): cluster 3\n", + "Example n. 3576 = (37.0, 104.0): cluster 3\n", + "Example n. 3577 = (38.0, 491.0): cluster 3\n", + "Example n. 3578 = (38.0, 490.0): cluster 3\n", + "Example n. 3579 = (38.0, 489.0): cluster 3\n", + "Example n. 3580 = (38.0, 488.0): cluster 3\n", + "Example n. 3581 = (38.0, 487.0): cluster 3\n", + "Example n. 3582 = (38.0, 486.0): cluster 3\n", + "Example n. 3583 = (38.0, 485.0): cluster 3\n", + "Example n. 3584 = (38.0, 484.0): cluster 3\n", + "Example n. 3585 = (38.0, 483.0): cluster 3\n", + "Example n. 3586 = (38.0, 482.0): cluster 3\n", + "Example n. 3587 = (38.0, 481.0): cluster 3\n", + "Example n. 3588 = (38.0, 480.0): cluster 3\n", + "Example n. 3589 = (38.0, 479.0): cluster 3\n", + "Example n. 3590 = (38.0, 478.0): cluster 3\n", + "Example n. 3591 = (38.0, 477.0): cluster 3\n", + "Example n. 3592 = (38.0, 476.0): cluster 3\n", + "Example n. 3593 = (38.0, 475.0): cluster 3\n", + "Example n. 3594 = (38.0, 474.0): cluster 3\n", + "Example n. 3595 = (38.0, 473.0): cluster 3\n", + "Example n. 3596 = (38.0, 472.0): cluster 1\n", + "Example n. 3597 = (38.0, 471.0): cluster 1\n", + "Example n. 3598 = (38.0, 470.0): cluster 1\n", + "Example n. 3599 = (38.0, 469.0): cluster 1\n", + "Example n. 3600 = (38.0, 468.0): cluster 1\n", + "Example n. 3601 = (38.0, 467.0): cluster 1\n", + "Example n. 3602 = (38.0, 466.0): cluster 1\n", + "Example n. 3603 = (38.0, 465.0): cluster 1\n", + "Example n. 3604 = (38.0, 464.0): cluster 1\n", + "Example n. 3605 = (38.0, 463.0): cluster 1\n", + "Example n. 3606 = (38.0, 462.0): cluster 1\n", + "Example n. 3607 = (38.0, 461.0): cluster 1\n", + "Example n. 3608 = (38.0, 460.0): cluster 1\n", + "Example n. 3609 = (38.0, 459.0): cluster 1\n", + "Example n. 3610 = (38.0, 458.0): cluster 1\n", + "Example n. 3611 = (38.0, 457.0): cluster 1\n", + "Example n. 3612 = (38.0, 456.0): cluster 1\n", + "Example n. 3613 = (38.0, 381.0): cluster 1\n", + "Example n. 3614 = (38.0, 380.0): cluster 1\n", + "Example n. 3615 = (38.0, 379.0): cluster 1\n", + "Example n. 3616 = (38.0, 378.0): cluster 1\n", + "Example n. 3617 = (38.0, 377.0): cluster 1\n", + "Example n. 3618 = (38.0, 313.0): cluster 1\n", + "Example n. 3619 = (38.0, 312.0): cluster 1\n", + "Example n. 3620 = (38.0, 311.0): cluster 1\n", + "Example n. 3621 = (38.0, 310.0): cluster 1\n", + "Example n. 3622 = (38.0, 309.0): cluster 1\n", + "Example n. 3623 = (38.0, 308.0): cluster 1\n", + "Example n. 3624 = (38.0, 307.0): cluster 1\n", + "Example n. 3625 = (38.0, 306.0): cluster 1\n", + "Example n. 3626 = (38.0, 305.0): cluster 1\n", + "Example n. 3627 = (38.0, 304.0): cluster 1\n", + "Example n. 3628 = (38.0, 303.0): cluster 1\n", + "Example n. 3629 = (38.0, 302.0): cluster 1\n", + "Example n. 3630 = (38.0, 301.0): cluster 1\n", + "Example n. 3631 = (38.0, 300.0): cluster 1\n", + "Example n. 3632 = (38.0, 299.0): cluster 1\n", + "Example n. 3633 = (38.0, 298.0): cluster 1\n", + "Example n. 3634 = (38.0, 297.0): cluster 1\n", + "Example n. 3635 = (38.0, 296.0): cluster 1\n", + "Example n. 3636 = (38.0, 293.0): cluster 1\n", + "Example n. 3637 = (38.0, 292.0): cluster 1\n", + "Example n. 3638 = (38.0, 291.0): cluster 1\n", + "Example n. 3639 = (38.0, 290.0): cluster 1\n", + "Example n. 3640 = (38.0, 289.0): cluster 1\n", + "Example n. 3641 = (38.0, 288.0): cluster 3\n", + "Example n. 3642 = (38.0, 287.0): cluster 3\n", + "Example n. 3643 = (38.0, 286.0): cluster 3\n", + "Example n. 3644 = (38.0, 285.0): cluster 3\n", + "Example n. 3645 = (38.0, 284.0): cluster 3\n", + "Example n. 3646 = (38.0, 281.0): cluster 3\n", + "Example n. 3647 = (38.0, 280.0): cluster 3\n", + "Example n. 3648 = (38.0, 279.0): cluster 3\n", + "Example n. 3649 = (38.0, 278.0): cluster 3\n", + "Example n. 3650 = (38.0, 272.0): cluster 3\n", + "Example n. 3651 = (38.0, 271.0): cluster 3\n", + "Example n. 3652 = (38.0, 270.0): cluster 3\n", + "Example n. 3653 = (38.0, 269.0): cluster 3\n", + "Example n. 3654 = (38.0, 268.0): cluster 3\n", + "Example n. 3655 = (38.0, 267.0): cluster 3\n", + "Example n. 3656 = (38.0, 266.0): cluster 3\n", + "Example n. 3657 = (38.0, 265.0): cluster 3\n", + "Example n. 3658 = (38.0, 264.0): cluster 3\n", + "Example n. 3659 = (38.0, 263.0): cluster 3\n", + "Example n. 3660 = (38.0, 262.0): cluster 3\n", + "Example n. 3661 = (38.0, 261.0): cluster 3\n", + "Example n. 3662 = (38.0, 260.0): cluster 3\n", + "Example n. 3663 = (38.0, 259.0): cluster 3\n", + "Example n. 3664 = (38.0, 258.0): cluster 3\n", + "Example n. 3665 = (38.0, 257.0): cluster 3\n", + "Example n. 3666 = (38.0, 256.0): cluster 3\n", + "Example n. 3667 = (38.0, 255.0): cluster 3\n", + "Example n. 3668 = (38.0, 254.0): cluster 3\n", + "Example n. 3669 = (38.0, 253.0): cluster 3\n", + "Example n. 3670 = (38.0, 252.0): cluster 3\n", + "Example n. 3671 = (38.0, 245.0): cluster 3\n", + "Example n. 3672 = (38.0, 244.0): cluster 3\n", + "Example n. 3673 = (38.0, 243.0): cluster 3\n", + "Example n. 3674 = (38.0, 242.0): cluster 3\n", + "Example n. 3675 = (38.0, 241.0): cluster 3\n", + "Example n. 3676 = (38.0, 240.0): cluster 3\n", + "Example n. 3677 = (38.0, 239.0): cluster 3\n", + "Example n. 3678 = (38.0, 238.0): cluster 3\n", + "Example n. 3679 = (38.0, 237.0): cluster 3\n", + "Example n. 3680 = (38.0, 236.0): cluster 3\n", + "Example n. 3681 = (38.0, 235.0): cluster 3\n", + "Example n. 3682 = (38.0, 234.0): cluster 1\n", + "Example n. 3683 = (38.0, 233.0): cluster 1\n", + "Example n. 3684 = (38.0, 232.0): cluster 1\n", + "Example n. 3685 = (38.0, 231.0): cluster 1\n", + "Example n. 3686 = (38.0, 230.0): cluster 1\n", + "Example n. 3687 = (38.0, 229.0): cluster 1\n", + "Example n. 3688 = (38.0, 228.0): cluster 1\n", + "Example n. 3689 = (38.0, 227.0): cluster 1\n", + "Example n. 3690 = (38.0, 226.0): cluster 1\n", + "Example n. 3691 = (38.0, 224.0): cluster 1\n", + "Example n. 3692 = (38.0, 223.0): cluster 1\n", + "Example n. 3693 = (38.0, 222.0): cluster 1\n", + "Example n. 3694 = (38.0, 221.0): cluster 1\n", + "Example n. 3695 = (38.0, 220.0): cluster 1\n", + "Example n. 3696 = (38.0, 219.0): cluster 1\n", + "Example n. 3697 = (38.0, 218.0): cluster 1\n", + "Example n. 3698 = (38.0, 212.0): cluster 1\n", + "Example n. 3699 = (38.0, 211.0): cluster 1\n", + "Example n. 3700 = (38.0, 210.0): cluster 1\n", + "Example n. 3701 = (38.0, 209.0): cluster 1\n", + "Example n. 3702 = (38.0, 208.0): cluster 1\n", + "Example n. 3703 = (38.0, 207.0): cluster 1\n", + "Example n. 3704 = (38.0, 197.0): cluster 1\n", + "Example n. 3705 = (38.0, 196.0): cluster 1\n", + "Example n. 3706 = (38.0, 195.0): cluster 1\n", + "Example n. 3707 = (38.0, 194.0): cluster 1\n", + "Example n. 3708 = (38.0, 193.0): cluster 1\n", + "Example n. 3709 = (38.0, 192.0): cluster 1\n", + "Example n. 3710 = (38.0, 191.0): cluster 1\n", + "Example n. 3711 = (38.0, 190.0): cluster 1\n", + "Example n. 3712 = (38.0, 189.0): cluster 1\n", + "Example n. 3713 = (38.0, 188.0): cluster 1\n", + "Example n. 3714 = (38.0, 187.0): cluster 1\n", + "Example n. 3715 = (38.0, 186.0): cluster 1\n", + "Example n. 3716 = (38.0, 185.0): cluster 1\n", + "Example n. 3717 = (38.0, 184.0): cluster 1\n", + "Example n. 3718 = (38.0, 183.0): cluster 1\n", + "Example n. 3719 = (38.0, 182.0): cluster 1\n", + "Example n. 3720 = (38.0, 181.0): cluster 1\n", + "Example n. 3721 = (38.0, 180.0): cluster 1\n", + "Example n. 3722 = (38.0, 179.0): cluster 1\n", + "Example n. 3723 = (38.0, 178.0): cluster 3\n", + "Example n. 3724 = (38.0, 177.0): cluster 3\n", + "Example n. 3725 = (38.0, 174.0): cluster 3\n", + "Example n. 3726 = (38.0, 173.0): cluster 3\n", + "Example n. 3727 = (38.0, 172.0): cluster 3\n", + "Example n. 3728 = (38.0, 171.0): cluster 3\n", + "Example n. 3729 = (38.0, 163.0): cluster 3\n", + "Example n. 3730 = (38.0, 162.0): cluster 3\n", + "Example n. 3731 = (38.0, 161.0): cluster 3\n", + "Example n. 3732 = (38.0, 160.0): cluster 3\n", + "Example n. 3733 = (38.0, 159.0): cluster 3\n", + "Example n. 3734 = (38.0, 158.0): cluster 3\n", + "Example n. 3735 = (38.0, 157.0): cluster 3\n", + "Example n. 3736 = (38.0, 156.0): cluster 3\n", + "Example n. 3737 = (38.0, 155.0): cluster 3\n", + "Example n. 3738 = (38.0, 154.0): cluster 3\n", + "Example n. 3739 = (38.0, 153.0): cluster 3\n", + "Example n. 3740 = (38.0, 152.0): cluster 3\n", + "Example n. 3741 = (38.0, 151.0): cluster 3\n", + "Example n. 3742 = (38.0, 150.0): cluster 3\n", + "Example n. 3743 = (38.0, 149.0): cluster 3\n", + "Example n. 3744 = (38.0, 148.0): cluster 3\n", + "Example n. 3745 = (38.0, 147.0): cluster 3\n", + "Example n. 3746 = (38.0, 146.0): cluster 3\n", + "Example n. 3747 = (38.0, 132.0): cluster 3\n", + "Example n. 3748 = (38.0, 131.0): cluster 3\n", + "Example n. 3749 = (38.0, 130.0): cluster 3\n", + "Example n. 3750 = (38.0, 129.0): cluster 3\n", + "Example n. 3751 = (38.0, 128.0): cluster 3\n", + "Example n. 3752 = (38.0, 127.0): cluster 3\n", + "Example n. 3753 = (38.0, 126.0): cluster 3\n", + "Example n. 3754 = (38.0, 114.0): cluster 3\n", + "Example n. 3755 = (38.0, 113.0): cluster 3\n", + "Example n. 3756 = (38.0, 112.0): cluster 1\n", + "Example n. 3757 = (38.0, 111.0): cluster 1\n", + "Example n. 3758 = (38.0, 110.0): cluster 1\n", + "Example n. 3759 = (38.0, 109.0): cluster 1\n", + "Example n. 3760 = (38.0, 108.0): cluster 1\n", + "Example n. 3761 = (38.0, 107.0): cluster 1\n", + "Example n. 3762 = (38.0, 106.0): cluster 1\n", + "Example n. 3763 = (38.0, 105.0): cluster 1\n", + "Example n. 3764 = (39.0, 492.0): cluster 1\n", + "Example n. 3765 = (39.0, 491.0): cluster 1\n", + "Example n. 3766 = (39.0, 490.0): cluster 1\n", + "Example n. 3767 = (39.0, 489.0): cluster 1\n", + "Example n. 3768 = (39.0, 488.0): cluster 1\n", + "Example n. 3769 = (39.0, 487.0): cluster 1\n", + "Example n. 3770 = (39.0, 486.0): cluster 1\n", + "Example n. 3771 = (39.0, 485.0): cluster 1\n", + "Example n. 3772 = (39.0, 484.0): cluster 1\n", + "Example n. 3773 = (39.0, 483.0): cluster 1\n", + "Example n. 3774 = (39.0, 482.0): cluster 1\n", + "Example n. 3775 = (39.0, 481.0): cluster 1\n", + "Example n. 3776 = (39.0, 480.0): cluster 1\n", + "Example n. 3777 = (39.0, 479.0): cluster 1\n", + "Example n. 3778 = (39.0, 478.0): cluster 1\n", + "Example n. 3779 = (39.0, 477.0): cluster 1\n", + "Example n. 3780 = (39.0, 476.0): cluster 1\n", + "Example n. 3781 = (39.0, 475.0): cluster 1\n", + "Example n. 3782 = (39.0, 474.0): cluster 1\n", + "Example n. 3783 = (39.0, 473.0): cluster 1\n", + "Example n. 3784 = (39.0, 472.0): cluster 1\n", + "Example n. 3785 = (39.0, 471.0): cluster 1\n", + "Example n. 3786 = (39.0, 470.0): cluster 1\n", + "Example n. 3787 = (39.0, 469.0): cluster 1\n", + "Example n. 3788 = (39.0, 468.0): cluster 1\n", + "Example n. 3789 = (39.0, 467.0): cluster 1\n", + "Example n. 3790 = (39.0, 466.0): cluster 1\n", + "Example n. 3791 = (39.0, 465.0): cluster 1\n", + "Example n. 3792 = (39.0, 464.0): cluster 1\n", + "Example n. 3793 = (39.0, 463.0): cluster 1\n", + "Example n. 3794 = (39.0, 462.0): cluster 1\n", + "Example n. 3795 = (39.0, 461.0): cluster 1\n", + "Example n. 3796 = (39.0, 460.0): cluster 1\n", + "Example n. 3797 = (39.0, 459.0): cluster 3\n", + "Example n. 3798 = (39.0, 458.0): cluster 3\n", + "Example n. 3799 = (39.0, 457.0): cluster 3\n", + "Example n. 3800 = (39.0, 456.0): cluster 3\n", + "Example n. 3801 = (39.0, 455.0): cluster 3\n", + "Example n. 3802 = (39.0, 318.0): cluster 3\n", + "Example n. 3803 = (39.0, 317.0): cluster 3\n", + "Example n. 3804 = (39.0, 316.0): cluster 3\n", + "Example n. 3805 = (39.0, 315.0): cluster 3\n", + "Example n. 3806 = (39.0, 314.0): cluster 3\n", + "Example n. 3807 = (39.0, 313.0): cluster 3\n", + "Example n. 3808 = (39.0, 312.0): cluster 3\n", + "Example n. 3809 = (39.0, 311.0): cluster 3\n", + "Example n. 3810 = (39.0, 310.0): cluster 3\n", + "Example n. 3811 = (39.0, 309.0): cluster 3\n", + "Example n. 3812 = (39.0, 308.0): cluster 3\n", + "Example n. 3813 = (39.0, 307.0): cluster 3\n", + "Example n. 3814 = (39.0, 306.0): cluster 3\n", + "Example n. 3815 = (39.0, 305.0): cluster 3\n", + "Example n. 3816 = (39.0, 304.0): cluster 3\n", + "Example n. 3817 = (39.0, 303.0): cluster 3\n", + "Example n. 3818 = (39.0, 302.0): cluster 3\n", + "Example n. 3819 = (39.0, 301.0): cluster 3\n", + "Example n. 3820 = (39.0, 300.0): cluster 3\n", + "Example n. 3821 = (39.0, 299.0): cluster 3\n", + "Example n. 3822 = (39.0, 298.0): cluster 3\n", + "Example n. 3823 = (39.0, 297.0): cluster 3\n", + "Example n. 3824 = (39.0, 296.0): cluster 3\n", + "Example n. 3825 = (39.0, 293.0): cluster 3\n", + "Example n. 3826 = (39.0, 292.0): cluster 3\n", + "Example n. 3827 = (39.0, 291.0): cluster 3\n", + "Example n. 3828 = (39.0, 290.0): cluster 3\n", + "Example n. 3829 = (39.0, 289.0): cluster 3\n", + "Example n. 3830 = (39.0, 288.0): cluster 1\n", + "Example n. 3831 = (39.0, 287.0): cluster 1\n", + "Example n. 3832 = (39.0, 286.0): cluster 1\n", + "Example n. 3833 = (39.0, 285.0): cluster 1\n", + "Example n. 3834 = (39.0, 272.0): cluster 1\n", + "Example n. 3835 = (39.0, 271.0): cluster 1\n", + "Example n. 3836 = (39.0, 270.0): cluster 1\n", + "Example n. 3837 = (39.0, 269.0): cluster 1\n", + "Example n. 3838 = (39.0, 268.0): cluster 1\n", + "Example n. 3839 = (39.0, 267.0): cluster 1\n", + "Example n. 3840 = (39.0, 266.0): cluster 1\n", + "Example n. 3841 = (39.0, 265.0): cluster 1\n", + "Example n. 3842 = (39.0, 264.0): cluster 1\n", + "Example n. 3843 = (39.0, 263.0): cluster 1\n", + "Example n. 3844 = (39.0, 262.0): cluster 1\n", + "Example n. 3845 = (39.0, 261.0): cluster 1\n", + "Example n. 3846 = (39.0, 260.0): cluster 1\n", + "Example n. 3847 = (39.0, 259.0): cluster 1\n", + "Example n. 3848 = (39.0, 258.0): cluster 1\n", + "Example n. 3849 = (39.0, 257.0): cluster 1\n", + "Example n. 3850 = (39.0, 256.0): cluster 1\n", + "Example n. 3851 = (39.0, 255.0): cluster 1\n", + "Example n. 3852 = (39.0, 254.0): cluster 1\n", + "Example n. 3853 = (39.0, 253.0): cluster 1\n", + "Example n. 3854 = (39.0, 246.0): cluster 1\n", + "Example n. 3855 = (39.0, 245.0): cluster 1\n", + "Example n. 3856 = (39.0, 244.0): cluster 1\n", + "Example n. 3857 = (39.0, 243.0): cluster 1\n", + "Example n. 3858 = (39.0, 242.0): cluster 1\n", + "Example n. 3859 = (39.0, 241.0): cluster 1\n", + "Example n. 3860 = (39.0, 240.0): cluster 1\n", + "Example n. 3861 = (39.0, 239.0): cluster 1\n", + "Example n. 3862 = (39.0, 238.0): cluster 1\n", + "Example n. 3863 = (39.0, 237.0): cluster 1\n", + "Example n. 3864 = (39.0, 236.0): cluster 1\n", + "Example n. 3865 = (39.0, 235.0): cluster 1\n", + "Example n. 3866 = (39.0, 234.0): cluster 1\n", + "Example n. 3867 = (39.0, 233.0): cluster 1\n", + "Example n. 3868 = (39.0, 232.0): cluster 1\n", + "Example n. 3869 = (39.0, 231.0): cluster 1\n", + "Example n. 3870 = (39.0, 230.0): cluster 1\n", + "Example n. 3871 = (39.0, 229.0): cluster 3\n", + "Example n. 3872 = (39.0, 228.0): cluster 3\n", + "Example n. 3873 = (39.0, 227.0): cluster 3\n", + "Example n. 3874 = (39.0, 226.0): cluster 3\n", + "Example n. 3875 = (39.0, 220.0): cluster 3\n", + "Example n. 3876 = (39.0, 214.0): cluster 3\n", + "Example n. 3877 = (39.0, 213.0): cluster 3\n", + "Example n. 3878 = (39.0, 212.0): cluster 3\n", + "Example n. 3879 = (39.0, 211.0): cluster 3\n", + "Example n. 3880 = (39.0, 210.0): cluster 3\n", + "Example n. 3881 = (39.0, 209.0): cluster 3\n", + "Example n. 3882 = (39.0, 208.0): cluster 3\n", + "Example n. 3883 = (39.0, 207.0): cluster 3\n", + "Example n. 3884 = (39.0, 206.0): cluster 3\n", + "Example n. 3885 = (39.0, 204.0): cluster 3\n", + "Example n. 3886 = (39.0, 203.0): cluster 3\n", + "Example n. 3887 = (39.0, 202.0): cluster 3\n", + "Example n. 3888 = (39.0, 201.0): cluster 3\n", + "Example n. 3889 = (39.0, 200.0): cluster 3\n", + "Example n. 3890 = (39.0, 197.0): cluster 3\n", + "Example n. 3891 = (39.0, 196.0): cluster 3\n", + "Example n. 3892 = (39.0, 195.0): cluster 3\n", + "Example n. 3893 = (39.0, 194.0): cluster 3\n", + "Example n. 3894 = (39.0, 193.0): cluster 3\n", + "Example n. 3895 = (39.0, 192.0): cluster 3\n", + "Example n. 3896 = (39.0, 191.0): cluster 3\n", + "Example n. 3897 = (39.0, 190.0): cluster 3\n", + "Example n. 3898 = (39.0, 187.0): cluster 3\n", + "Example n. 3899 = (39.0, 186.0): cluster 3\n", + "Example n. 3900 = (39.0, 185.0): cluster 3\n", + "Example n. 3901 = (39.0, 184.0): cluster 3\n", + "Example n. 3902 = (39.0, 183.0): cluster 3\n", + "Example n. 3903 = (39.0, 182.0): cluster 3\n", + "Example n. 3904 = (39.0, 181.0): cluster 1\n", + "Example n. 3905 = (39.0, 180.0): cluster 1\n", + "Example n. 3906 = (39.0, 179.0): cluster 1\n", + "Example n. 3907 = (39.0, 178.0): cluster 1\n", + "Example n. 3908 = (39.0, 177.0): cluster 1\n", + "Example n. 3909 = (39.0, 175.0): cluster 1\n", + "Example n. 3910 = (39.0, 174.0): cluster 1\n", + "Example n. 3911 = (39.0, 173.0): cluster 1\n", + "Example n. 3912 = (39.0, 172.0): cluster 1\n", + "Example n. 3913 = (39.0, 171.0): cluster 1\n", + "Example n. 3914 = (39.0, 170.0): cluster 1\n", + "Example n. 3915 = (39.0, 163.0): cluster 1\n", + "Example n. 3916 = (39.0, 162.0): cluster 1\n", + "Example n. 3917 = (39.0, 161.0): cluster 1\n", + "Example n. 3918 = (39.0, 160.0): cluster 1\n", + "Example n. 3919 = (39.0, 159.0): cluster 1\n", + "Example n. 3920 = (39.0, 158.0): cluster 1\n", + "Example n. 3921 = (39.0, 157.0): cluster 1\n", + "Example n. 3922 = (39.0, 156.0): cluster 1\n", + "Example n. 3923 = (39.0, 155.0): cluster 1\n", + "Example n. 3924 = (39.0, 154.0): cluster 1\n", + "Example n. 3925 = (39.0, 153.0): cluster 1\n", + "Example n. 3926 = (39.0, 152.0): cluster 1\n", + "Example n. 3927 = (39.0, 151.0): cluster 1\n", + "Example n. 3928 = (39.0, 150.0): cluster 1\n", + "Example n. 3929 = (39.0, 149.0): cluster 1\n", + "Example n. 3930 = (39.0, 148.0): cluster 1\n", + "Example n. 3931 = (39.0, 147.0): cluster 1\n", + "Example n. 3932 = (39.0, 146.0): cluster 1\n", + "Example n. 3933 = (39.0, 144.0): cluster 1\n", + "Example n. 3934 = (39.0, 143.0): cluster 1\n", + "Example n. 3935 = (39.0, 132.0): cluster 1\n", + "Example n. 3936 = (39.0, 131.0): cluster 1\n", + "Example n. 3937 = (39.0, 130.0): cluster 1\n", + "Example n. 3938 = (39.0, 129.0): cluster 1\n", + "Example n. 3939 = (39.0, 128.0): cluster 1\n", + "Example n. 3940 = (39.0, 127.0): cluster 1\n", + "Example n. 3941 = (39.0, 126.0): cluster 1\n", + "Example n. 3942 = (39.0, 114.0): cluster 1\n", + "Example n. 3943 = (39.0, 113.0): cluster 1\n", + "Example n. 3944 = (39.0, 112.0): cluster 1\n", + "Example n. 3945 = (39.0, 111.0): cluster 3\n", + "Example n. 3946 = (39.0, 110.0): cluster 3\n", + "Example n. 3947 = (39.0, 109.0): cluster 3\n", + "Example n. 3948 = (40.0, 492.0): cluster 3\n", + "Example n. 3949 = (40.0, 491.0): cluster 3\n", + "Example n. 3950 = (40.0, 490.0): cluster 3\n", + "Example n. 3951 = (40.0, 489.0): cluster 3\n", + "Example n. 3952 = (40.0, 488.0): cluster 3\n", + "Example n. 3953 = (40.0, 487.0): cluster 3\n", + "Example n. 3954 = (40.0, 486.0): cluster 3\n", + "Example n. 3955 = (40.0, 485.0): cluster 3\n", + "Example n. 3956 = (40.0, 484.0): cluster 3\n", + "Example n. 3957 = (40.0, 483.0): cluster 3\n", + "Example n. 3958 = (40.0, 482.0): cluster 3\n", + "Example n. 3959 = (40.0, 481.0): cluster 3\n", + "Example n. 3960 = (40.0, 480.0): cluster 3\n", + "Example n. 3961 = (40.0, 479.0): cluster 3\n", + "Example n. 3962 = (40.0, 478.0): cluster 3\n", + "Example n. 3963 = (40.0, 477.0): cluster 3\n", + "Example n. 3964 = (40.0, 476.0): cluster 3\n", + "Example n. 3965 = (40.0, 475.0): cluster 3\n", + "Example n. 3966 = (40.0, 474.0): cluster 3\n", + "Example n. 3967 = (40.0, 473.0): cluster 3\n", + "Example n. 3968 = (40.0, 472.0): cluster 3\n", + "Example n. 3969 = (40.0, 471.0): cluster 3\n", + "Example n. 3970 = (40.0, 470.0): cluster 3\n", + "Example n. 3971 = (40.0, 469.0): cluster 3\n", + "Example n. 3972 = (40.0, 468.0): cluster 3\n", + "Example n. 3973 = (40.0, 467.0): cluster 3\n", + "Example n. 3974 = (40.0, 466.0): cluster 3\n", + "Example n. 3975 = (40.0, 465.0): cluster 3\n", + "Example n. 3976 = (40.0, 464.0): cluster 3\n", + "Example n. 3977 = (40.0, 463.0): cluster 3\n", + "Example n. 3978 = (40.0, 462.0): cluster 3\n", + "Example n. 3979 = (40.0, 461.0): cluster 3\n", + "Example n. 3980 = (40.0, 460.0): cluster 3\n", + "Example n. 3981 = (40.0, 459.0): cluster 1\n", + "Example n. 3982 = (40.0, 458.0): cluster 1\n", + "Example n. 3983 = (40.0, 457.0): cluster 1\n", + "Example n. 3984 = (40.0, 456.0): cluster 1\n", + "Example n. 3985 = (40.0, 455.0): cluster 1\n", + "Example n. 3986 = (40.0, 410.0): cluster 1\n", + "Example n. 3987 = (40.0, 409.0): cluster 1\n", + "Example n. 3988 = (40.0, 408.0): cluster 1\n", + "Example n. 3989 = (40.0, 407.0): cluster 1\n", + "Example n. 3990 = (40.0, 395.0): cluster 1\n", + "Example n. 3991 = (40.0, 394.0): cluster 1\n", + "Example n. 3992 = (40.0, 393.0): cluster 1\n", + "Example n. 3993 = (40.0, 392.0): cluster 1\n", + "Example n. 3994 = (40.0, 345.0): cluster 1\n", + "Example n. 3995 = (40.0, 344.0): cluster 1\n", + "Example n. 3996 = (40.0, 343.0): cluster 1\n", + "Example n. 3997 = (40.0, 342.0): cluster 1\n", + "Example n. 3998 = (40.0, 319.0): cluster 1\n", + "Example n. 3999 = (40.0, 318.0): cluster 1\n", + "Example n. 4000 = (40.0, 317.0): cluster 1\n", + "Example n. 4001 = (40.0, 316.0): cluster 1\n", + "Example n. 4002 = (40.0, 315.0): cluster 1\n", + "Example n. 4003 = (40.0, 314.0): cluster 1\n", + "Example n. 4004 = (40.0, 313.0): cluster 1\n", + "Example n. 4005 = (40.0, 312.0): cluster 1\n", + "Example n. 4006 = (40.0, 311.0): cluster 1\n", + "Example n. 4007 = (40.0, 310.0): cluster 1\n", + "Example n. 4008 = (40.0, 309.0): cluster 1\n", + "Example n. 4009 = (40.0, 308.0): cluster 3\n", + "Example n. 4010 = (40.0, 307.0): cluster 3\n", + "Example n. 4011 = (40.0, 306.0): cluster 3\n", + "Example n. 4012 = (40.0, 305.0): cluster 3\n", + "Example n. 4013 = (40.0, 304.0): cluster 3\n", + "Example n. 4014 = (40.0, 303.0): cluster 3\n", + "Example n. 4015 = (40.0, 302.0): cluster 3\n", + "Example n. 4016 = (40.0, 300.0): cluster 3\n", + "Example n. 4017 = (40.0, 299.0): cluster 3\n", + "Example n. 4018 = (40.0, 298.0): cluster 3\n", + "Example n. 4019 = (40.0, 297.0): cluster 3\n", + "Example n. 4020 = (40.0, 296.0): cluster 3\n", + "Example n. 4021 = (40.0, 293.0): cluster 3\n", + "Example n. 4022 = (40.0, 292.0): cluster 3\n", + "Example n. 4023 = (40.0, 291.0): cluster 3\n", + "Example n. 4024 = (40.0, 290.0): cluster 3\n", + "Example n. 4025 = (40.0, 289.0): cluster 3\n", + "Example n. 4026 = (40.0, 288.0): cluster 3\n", + "Example n. 4027 = (40.0, 287.0): cluster 3\n", + "Example n. 4028 = (40.0, 286.0): cluster 3\n", + "Example n. 4029 = (40.0, 285.0): cluster 3\n", + "Example n. 4030 = (40.0, 284.0): cluster 3\n", + "Example n. 4031 = (40.0, 283.0): cluster 3\n", + "Example n. 4032 = (40.0, 272.0): cluster 3\n", + "Example n. 4033 = (40.0, 271.0): cluster 3\n", + "Example n. 4034 = (40.0, 270.0): cluster 3\n", + "Example n. 4035 = (40.0, 269.0): cluster 3\n", + "Example n. 4036 = (40.0, 268.0): cluster 3\n", + "Example n. 4037 = (40.0, 267.0): cluster 3\n", + "Example n. 4038 = (40.0, 266.0): cluster 3\n", + "Example n. 4039 = (40.0, 265.0): cluster 3\n", + "Example n. 4040 = (40.0, 264.0): cluster 3\n", + "Example n. 4041 = (40.0, 263.0): cluster 1\n", + "Example n. 4042 = (40.0, 260.0): cluster 1\n", + "Example n. 4043 = (40.0, 259.0): cluster 1\n", + "Example n. 4044 = (40.0, 258.0): cluster 1\n", + "Example n. 4045 = (40.0, 257.0): cluster 1\n", + "Example n. 4046 = (40.0, 256.0): cluster 1\n", + "Example n. 4047 = (40.0, 255.0): cluster 1\n", + "Example n. 4048 = (40.0, 254.0): cluster 1\n", + "Example n. 4049 = (40.0, 253.0): cluster 1\n", + "Example n. 4050 = (40.0, 249.0): cluster 1\n", + "Example n. 4051 = (40.0, 248.0): cluster 1\n", + "Example n. 4052 = (40.0, 247.0): cluster 1\n", + "Example n. 4053 = (40.0, 246.0): cluster 1\n", + "Example n. 4054 = (40.0, 245.0): cluster 1\n", + "Example n. 4055 = (40.0, 244.0): cluster 1\n", + "Example n. 4056 = (40.0, 243.0): cluster 1\n", + "Example n. 4057 = (40.0, 242.0): cluster 1\n", + "Example n. 4058 = (40.0, 241.0): cluster 1\n", + "Example n. 4059 = (40.0, 240.0): cluster 1\n", + "Example n. 4060 = (40.0, 239.0): cluster 1\n", + "Example n. 4061 = (40.0, 238.0): cluster 1\n", + "Example n. 4062 = (40.0, 237.0): cluster 1\n", + "Example n. 4063 = (40.0, 236.0): cluster 1\n", + "Example n. 4064 = (40.0, 235.0): cluster 1\n", + "Example n. 4065 = (40.0, 234.0): cluster 1\n", + "Example n. 4066 = (40.0, 233.0): cluster 1\n", + "Example n. 4067 = (40.0, 232.0): cluster 1\n", + "Example n. 4068 = (40.0, 231.0): cluster 1\n", + "Example n. 4069 = (40.0, 230.0): cluster 3\n", + "Example n. 4070 = (40.0, 229.0): cluster 3\n", + "Example n. 4071 = (40.0, 228.0): cluster 3\n", + "Example n. 4072 = (40.0, 227.0): cluster 3\n", + "Example n. 4073 = (40.0, 226.0): cluster 3\n", + "Example n. 4074 = (40.0, 215.0): cluster 3\n", + "Example n. 4075 = (40.0, 214.0): cluster 3\n", + "Example n. 4076 = (40.0, 213.0): cluster 3\n", + "Example n. 4077 = (40.0, 212.0): cluster 3\n", + "Example n. 4078 = (40.0, 211.0): cluster 3\n", + "Example n. 4079 = (40.0, 210.0): cluster 3\n", + "Example n. 4080 = (40.0, 209.0): cluster 3\n", + "Example n. 4081 = (40.0, 208.0): cluster 3\n", + "Example n. 4082 = (40.0, 207.0): cluster 3\n", + "Example n. 4083 = (40.0, 206.0): cluster 3\n", + "Example n. 4084 = (40.0, 205.0): cluster 3\n", + "Example n. 4085 = (40.0, 204.0): cluster 3\n", + "Example n. 4086 = (40.0, 203.0): cluster 3\n", + "Example n. 4087 = (40.0, 202.0): cluster 3\n", + "Example n. 4088 = (40.0, 201.0): cluster 3\n", + "Example n. 4089 = (40.0, 200.0): cluster 3\n", + "Example n. 4090 = (40.0, 199.0): cluster 3\n", + "Example n. 4091 = (40.0, 197.0): cluster 3\n", + "Example n. 4092 = (40.0, 196.0): cluster 3\n", + "Example n. 4093 = (40.0, 195.0): cluster 3\n", + "Example n. 4094 = (40.0, 194.0): cluster 3\n", + "Example n. 4095 = (40.0, 193.0): cluster 3\n", + "Example n. 4096 = (40.0, 192.0): cluster 3\n", + "Example n. 4097 = (40.0, 187.0): cluster 3\n", + "Example n. 4098 = (40.0, 186.0): cluster 3\n", + "Example n. 4099 = (40.0, 185.0): cluster 3\n", + "Example n. 4100 = (40.0, 184.0): cluster 3\n", + "Example n. 4101 = (40.0, 183.0): cluster 1\n", + "Example n. 4102 = (40.0, 182.0): cluster 1\n", + "Example n. 4103 = (40.0, 181.0): cluster 1\n", + "Example n. 4104 = (40.0, 180.0): cluster 1\n", + "Example n. 4105 = (40.0, 179.0): cluster 1\n", + "Example n. 4106 = (40.0, 178.0): cluster 1\n", + "Example n. 4107 = (40.0, 175.0): cluster 1\n", + "Example n. 4108 = (40.0, 174.0): cluster 1\n", + "Example n. 4109 = (40.0, 173.0): cluster 1\n", + "Example n. 4110 = (40.0, 172.0): cluster 1\n", + "Example n. 4111 = (40.0, 171.0): cluster 1\n", + "Example n. 4112 = (40.0, 170.0): cluster 1\n", + "Example n. 4113 = (40.0, 161.0): cluster 1\n", + "Example n. 4114 = (40.0, 160.0): cluster 1\n", + "Example n. 4115 = (40.0, 159.0): cluster 1\n", + "Example n. 4116 = (40.0, 158.0): cluster 1\n", + "Example n. 4117 = (40.0, 157.0): cluster 1\n", + "Example n. 4118 = (40.0, 156.0): cluster 1\n", + "Example n. 4119 = (40.0, 155.0): cluster 1\n", + "Example n. 4120 = (40.0, 154.0): cluster 1\n", + "Example n. 4121 = (40.0, 153.0): cluster 1\n", + "Example n. 4122 = (40.0, 152.0): cluster 1\n", + "Example n. 4123 = (40.0, 151.0): cluster 1\n", + "Example n. 4124 = (40.0, 150.0): cluster 1\n", + "Example n. 4125 = (40.0, 149.0): cluster 1\n", + "Example n. 4126 = (40.0, 148.0): cluster 1\n", + "Example n. 4127 = (40.0, 147.0): cluster 1\n", + "Example n. 4128 = (40.0, 146.0): cluster 1\n", + "Example n. 4129 = (40.0, 145.0): cluster 3\n", + "Example n. 4130 = (40.0, 144.0): cluster 3\n", + "Example n. 4131 = (40.0, 143.0): cluster 3\n", + "Example n. 4132 = (40.0, 142.0): cluster 3\n", + "Example n. 4133 = (40.0, 130.0): cluster 3\n", + "Example n. 4134 = (40.0, 129.0): cluster 3\n", + "Example n. 4135 = (40.0, 128.0): cluster 3\n", + "Example n. 4136 = (40.0, 113.0): cluster 3\n", + "Example n. 4137 = (40.0, 112.0): cluster 3\n", + "Example n. 4138 = (40.0, 111.0): cluster 3\n", + "Example n. 4139 = (40.0, 110.0): cluster 3\n", + "Example n. 4140 = (40.0, 109.0): cluster 3\n", + "Example n. 4141 = (41.0, 491.0): cluster 3\n", + "Example n. 4142 = (41.0, 490.0): cluster 3\n", + "Example n. 4143 = (41.0, 489.0): cluster 3\n", + "Example n. 4144 = (41.0, 488.0): cluster 3\n", + "Example n. 4145 = (41.0, 487.0): cluster 3\n", + "Example n. 4146 = (41.0, 486.0): cluster 3\n", + "Example n. 4147 = (41.0, 485.0): cluster 3\n", + "Example n. 4148 = (41.0, 484.0): cluster 3\n", + "Example n. 4149 = (41.0, 483.0): cluster 3\n", + "Example n. 4150 = (41.0, 482.0): cluster 3\n", + "Example n. 4151 = (41.0, 481.0): cluster 3\n", + "Example n. 4152 = (41.0, 480.0): cluster 3\n", + "Example n. 4153 = (41.0, 479.0): cluster 3\n", + "Example n. 4154 = (41.0, 478.0): cluster 3\n", + "Example n. 4155 = (41.0, 477.0): cluster 3\n", + "Example n. 4156 = (41.0, 476.0): cluster 3\n", + "Example n. 4157 = (41.0, 475.0): cluster 3\n", + "Example n. 4158 = (41.0, 474.0): cluster 3\n", + "Example n. 4159 = (41.0, 473.0): cluster 3\n", + "Example n. 4160 = (41.0, 472.0): cluster 3\n", + "Example n. 4161 = (41.0, 471.0): cluster 3\n", + "Example n. 4162 = (41.0, 470.0): cluster 3\n", + "Example n. 4163 = (41.0, 469.0): cluster 3\n", + "Example n. 4164 = (41.0, 468.0): cluster 3\n", + "Example n. 4165 = (41.0, 467.0): cluster 3\n", + "Example n. 4166 = (41.0, 466.0): cluster 3\n", + "Example n. 4167 = (41.0, 465.0): cluster 3\n", + "Example n. 4168 = (41.0, 464.0): cluster 3\n", + "Example n. 4169 = (41.0, 463.0): cluster 3\n", + "Example n. 4170 = (41.0, 462.0): cluster 3\n", + "Example n. 4171 = (41.0, 461.0): cluster 3\n", + "Example n. 4172 = (41.0, 460.0): cluster 3\n", + "Example n. 4173 = (41.0, 459.0): cluster 3\n", + "Example n. 4174 = (41.0, 458.0): cluster 3\n", + "Example n. 4175 = (41.0, 457.0): cluster 3\n", + "Example n. 4176 = (41.0, 456.0): cluster 3\n", + "Example n. 4177 = (41.0, 455.0): cluster 3\n", + "Example n. 4178 = (41.0, 454.0): cluster 3\n", + "Example n. 4179 = (41.0, 433.0): cluster 3\n", + "Example n. 4180 = (41.0, 432.0): cluster 3\n", + "Example n. 4181 = (41.0, 431.0): cluster 3\n", + "Example n. 4182 = (41.0, 430.0): cluster 3\n", + "Example n. 4183 = (41.0, 411.0): cluster 3\n", + "Example n. 4184 = (41.0, 410.0): cluster 3\n", + "Example n. 4185 = (41.0, 409.0): cluster 3\n", + "Example n. 4186 = (41.0, 408.0): cluster 3\n", + "Example n. 4187 = (41.0, 407.0): cluster 3\n", + "Example n. 4188 = (41.0, 396.0): cluster 3\n", + "Example n. 4189 = (41.0, 395.0): cluster 3\n", + "Example n. 4190 = (41.0, 394.0): cluster 3\n", + "Example n. 4191 = (41.0, 393.0): cluster 3\n", + "Example n. 4192 = (41.0, 392.0): cluster 3\n", + "Example n. 4193 = (41.0, 346.0): cluster 3\n", + "Example n. 4194 = (41.0, 345.0): cluster 3\n", + "Example n. 4195 = (41.0, 344.0): cluster 3\n", + "Example n. 4196 = (41.0, 343.0): cluster 3\n", + "Example n. 4197 = (41.0, 342.0): cluster 3\n", + "Example n. 4198 = (41.0, 341.0): cluster 3\n", + "Example n. 4199 = (41.0, 319.0): cluster 3\n", + "Example n. 4200 = (41.0, 318.0): cluster 3\n", + "Example n. 4201 = (41.0, 317.0): cluster 3\n", + "Example n. 4202 = (41.0, 316.0): cluster 3\n", + "Example n. 4203 = (41.0, 315.0): cluster 3\n", + "Example n. 4204 = (41.0, 314.0): cluster 3\n", + "Example n. 4205 = (41.0, 313.0): cluster 3\n", + "Example n. 4206 = (41.0, 312.0): cluster 3\n", + "Example n. 4207 = (41.0, 311.0): cluster 3\n", + "Example n. 4208 = (41.0, 310.0): cluster 3\n", + "Example n. 4209 = (41.0, 309.0): cluster 3\n", + "Example n. 4210 = (41.0, 308.0): cluster 3\n", + "Example n. 4211 = (41.0, 307.0): cluster 3\n", + "Example n. 4212 = (41.0, 306.0): cluster 3\n", + "Example n. 4213 = (41.0, 305.0): cluster 3\n", + "Example n. 4214 = (41.0, 304.0): cluster 3\n", + "Example n. 4215 = (41.0, 303.0): cluster 3\n", + "Example n. 4216 = (41.0, 302.0): cluster 3\n", + "Example n. 4217 = (41.0, 300.0): cluster 3\n", + "Example n. 4218 = (41.0, 299.0): cluster 3\n", + "Example n. 4219 = (41.0, 298.0): cluster 3\n", + "Example n. 4220 = (41.0, 297.0): cluster 3\n", + "Example n. 4221 = (41.0, 296.0): cluster 3\n", + "Example n. 4222 = (41.0, 292.0): cluster 3\n", + "Example n. 4223 = (41.0, 291.0): cluster 3\n", + "Example n. 4224 = (41.0, 290.0): cluster 3\n", + "Example n. 4225 = (41.0, 289.0): cluster 3\n", + "Example n. 4226 = (41.0, 288.0): cluster 3\n", + "Example n. 4227 = (41.0, 287.0): cluster 3\n", + "Example n. 4228 = (41.0, 286.0): cluster 3\n", + "Example n. 4229 = (41.0, 285.0): cluster 3\n", + "Example n. 4230 = (41.0, 284.0): cluster 3\n", + "Example n. 4231 = (41.0, 283.0): cluster 3\n", + "Example n. 4232 = (41.0, 282.0): cluster 3\n", + "Example n. 4233 = (41.0, 278.0): cluster 3\n", + "Example n. 4234 = (41.0, 277.0): cluster 3\n", + "Example n. 4235 = (41.0, 276.0): cluster 3\n", + "Example n. 4236 = (41.0, 275.0): cluster 3\n", + "Example n. 4237 = (41.0, 272.0): cluster 3\n", + "Example n. 4238 = (41.0, 271.0): cluster 3\n", + "Example n. 4239 = (41.0, 270.0): cluster 3\n", + "Example n. 4240 = (41.0, 269.0): cluster 3\n", + "Example n. 4241 = (41.0, 268.0): cluster 3\n", + "Example n. 4242 = (41.0, 267.0): cluster 3\n", + "Example n. 4243 = (41.0, 266.0): cluster 3\n", + "Example n. 4244 = (41.0, 265.0): cluster 3\n", + "Example n. 4245 = (41.0, 264.0): cluster 3\n", + "Example n. 4246 = (41.0, 263.0): cluster 3\n", + "Example n. 4247 = (41.0, 260.0): cluster 3\n", + "Example n. 4248 = (41.0, 259.0): cluster 3\n", + "Example n. 4249 = (41.0, 258.0): cluster 3\n", + "Example n. 4250 = (41.0, 257.0): cluster 3\n", + "Example n. 4251 = (41.0, 256.0): cluster 3\n", + "Example n. 4252 = (41.0, 255.0): cluster 3\n", + "Example n. 4253 = (41.0, 254.0): cluster 3\n", + "Example n. 4254 = (41.0, 253.0): cluster 3\n", + "Example n. 4255 = (41.0, 250.0): cluster 3\n", + "Example n. 4256 = (41.0, 249.0): cluster 3\n", + "Example n. 4257 = (41.0, 248.0): cluster 3\n", + "Example n. 4258 = (41.0, 247.0): cluster 3\n", + "Example n. 4259 = (41.0, 246.0): cluster 3\n", + "Example n. 4260 = (41.0, 245.0): cluster 3\n", + "Example n. 4261 = (41.0, 244.0): cluster 3\n", + "Example n. 4262 = (41.0, 243.0): cluster 3\n", + "Example n. 4263 = (41.0, 242.0): cluster 3\n", + "Example n. 4264 = (41.0, 241.0): cluster 3\n", + "Example n. 4265 = (41.0, 240.0): cluster 3\n", + "Example n. 4266 = (41.0, 239.0): cluster 3\n", + "Example n. 4267 = (41.0, 238.0): cluster 3\n", + "Example n. 4268 = (41.0, 237.0): cluster 3\n", + "Example n. 4269 = (41.0, 236.0): cluster 3\n", + "Example n. 4270 = (41.0, 235.0): cluster 3\n", + "Example n. 4271 = (41.0, 234.0): cluster 3\n", + "Example n. 4272 = (41.0, 233.0): cluster 3\n", + "Example n. 4273 = (41.0, 232.0): cluster 3\n", + "Example n. 4274 = (41.0, 231.0): cluster 3\n", + "Example n. 4275 = (41.0, 230.0): cluster 3\n", + "Example n. 4276 = (41.0, 229.0): cluster 3\n", + "Example n. 4277 = (41.0, 228.0): cluster 3\n", + "Example n. 4278 = (41.0, 227.0): cluster 3\n", + "Example n. 4279 = (41.0, 226.0): cluster 3\n", + "Example n. 4280 = (41.0, 215.0): cluster 3\n", + "Example n. 4281 = (41.0, 214.0): cluster 3\n", + "Example n. 4282 = (41.0, 213.0): cluster 3\n", + "Example n. 4283 = (41.0, 212.0): cluster 3\n", + "Example n. 4284 = (41.0, 211.0): cluster 3\n", + "Example n. 4285 = (41.0, 210.0): cluster 3\n", + "Example n. 4286 = (41.0, 209.0): cluster 3\n", + "Example n. 4287 = (41.0, 208.0): cluster 3\n", + "Example n. 4288 = (41.0, 207.0): cluster 3\n", + "Example n. 4289 = (41.0, 206.0): cluster 3\n", + "Example n. 4290 = (41.0, 205.0): cluster 3\n", + "Example n. 4291 = (41.0, 204.0): cluster 3\n", + "Example n. 4292 = (41.0, 203.0): cluster 3\n", + "Example n. 4293 = (41.0, 202.0): cluster 3\n", + "Example n. 4294 = (41.0, 201.0): cluster 3\n", + "Example n. 4295 = (41.0, 200.0): cluster 3\n", + "Example n. 4296 = (41.0, 199.0): cluster 3\n", + "Example n. 4297 = (41.0, 196.0): cluster 3\n", + "Example n. 4298 = (41.0, 195.0): cluster 3\n", + "Example n. 4299 = (41.0, 194.0): cluster 3\n", + "Example n. 4300 = (41.0, 193.0): cluster 3\n", + "Example n. 4301 = (41.0, 188.0): cluster 3\n", + "Example n. 4302 = (41.0, 187.0): cluster 3\n", + "Example n. 4303 = (41.0, 186.0): cluster 3\n", + "Example n. 4304 = (41.0, 185.0): cluster 3\n", + "Example n. 4305 = (41.0, 184.0): cluster 3\n", + "Example n. 4306 = (41.0, 183.0): cluster 3\n", + "Example n. 4307 = (41.0, 182.0): cluster 3\n", + "Example n. 4308 = (41.0, 181.0): cluster 3\n", + "Example n. 4309 = (41.0, 180.0): cluster 3\n", + "Example n. 4310 = (41.0, 175.0): cluster 3\n", + "Example n. 4311 = (41.0, 174.0): cluster 3\n", + "Example n. 4312 = (41.0, 173.0): cluster 3\n", + "Example n. 4313 = (41.0, 172.0): cluster 3\n", + "Example n. 4314 = (41.0, 171.0): cluster 3\n", + "Example n. 4315 = (41.0, 161.0): cluster 3\n", + "Example n. 4316 = (41.0, 160.0): cluster 3\n", + "Example n. 4317 = (41.0, 159.0): cluster 3\n", + "Example n. 4318 = (41.0, 158.0): cluster 3\n", + "Example n. 4319 = (41.0, 157.0): cluster 3\n", + "Example n. 4320 = (41.0, 156.0): cluster 3\n", + "Example n. 4321 = (41.0, 155.0): cluster 3\n", + "Example n. 4322 = (41.0, 154.0): cluster 3\n", + "Example n. 4323 = (41.0, 153.0): cluster 3\n", + "Example n. 4324 = (41.0, 152.0): cluster 3\n", + "Example n. 4325 = (41.0, 151.0): cluster 3\n", + "Example n. 4326 = (41.0, 150.0): cluster 3\n", + "Example n. 4327 = (41.0, 149.0): cluster 3\n", + "Example n. 4328 = (41.0, 148.0): cluster 3\n", + "Example n. 4329 = (41.0, 147.0): cluster 3\n", + "Example n. 4330 = (41.0, 146.0): cluster 3\n", + "Example n. 4331 = (41.0, 145.0): cluster 3\n", + "Example n. 4332 = (41.0, 144.0): cluster 3\n", + "Example n. 4333 = (41.0, 143.0): cluster 3\n", + "Example n. 4334 = (41.0, 142.0): cluster 3\n", + "Example n. 4335 = (41.0, 141.0): cluster 3\n", + "Example n. 4336 = (41.0, 113.0): cluster 3\n", + "Example n. 4337 = (41.0, 112.0): cluster 3\n", + "Example n. 4338 = (41.0, 111.0): cluster 3\n", + "Example n. 4339 = (41.0, 110.0): cluster 3\n", + "Example n. 4340 = (41.0, 109.0): cluster 3\n", + "Example n. 4341 = (42.0, 491.0): cluster 2\n", + "Example n. 4342 = (42.0, 490.0): cluster 3\n", + "Example n. 4343 = (42.0, 489.0): cluster 3\n", + "Example n. 4344 = (42.0, 488.0): cluster 3\n", + "Example n. 4345 = (42.0, 487.0): cluster 3\n", + "Example n. 4346 = (42.0, 486.0): cluster 3\n", + "Example n. 4347 = (42.0, 485.0): cluster 3\n", + "Example n. 4348 = (42.0, 484.0): cluster 3\n", + "Example n. 4349 = (42.0, 483.0): cluster 3\n", + "Example n. 4350 = (42.0, 482.0): cluster 3\n", + "Example n. 4351 = (42.0, 481.0): cluster 3\n", + "Example n. 4352 = (42.0, 480.0): cluster 3\n", + "Example n. 4353 = (42.0, 479.0): cluster 3\n", + "Example n. 4354 = (42.0, 478.0): cluster 3\n", + "Example n. 4355 = (42.0, 477.0): cluster 3\n", + "Example n. 4356 = (42.0, 476.0): cluster 3\n", + "Example n. 4357 = (42.0, 475.0): cluster 3\n", + "Example n. 4358 = (42.0, 474.0): cluster 3\n", + "Example n. 4359 = (42.0, 473.0): cluster 3\n", + "Example n. 4360 = (42.0, 472.0): cluster 3\n", + "Example n. 4361 = (42.0, 471.0): cluster 3\n", + "Example n. 4362 = (42.0, 470.0): cluster 3\n", + "Example n. 4363 = (42.0, 469.0): cluster 3\n", + "Example n. 4364 = (42.0, 468.0): cluster 3\n", + "Example n. 4365 = (42.0, 467.0): cluster 3\n", + "Example n. 4366 = (42.0, 466.0): cluster 3\n", + "Example n. 4367 = (42.0, 465.0): cluster 3\n", + "Example n. 4368 = (42.0, 464.0): cluster 3\n", + "Example n. 4369 = (42.0, 463.0): cluster 3\n", + "Example n. 4370 = (42.0, 462.0): cluster 3\n", + "Example n. 4371 = (42.0, 461.0): cluster 3\n", + "Example n. 4372 = (42.0, 460.0): cluster 3\n", + "Example n. 4373 = (42.0, 459.0): cluster 3\n", + "Example n. 4374 = (42.0, 458.0): cluster 3\n", + "Example n. 4375 = (42.0, 457.0): cluster 3\n", + "Example n. 4376 = (42.0, 456.0): cluster 3\n", + "Example n. 4377 = (42.0, 455.0): cluster 2\n", + "Example n. 4378 = (42.0, 454.0): cluster 2\n", + "Example n. 4379 = (42.0, 433.0): cluster 3\n", + "Example n. 4380 = (42.0, 432.0): cluster 3\n", + "Example n. 4381 = (42.0, 431.0): cluster 3\n", + "Example n. 4382 = (42.0, 430.0): cluster 3\n", + "Example n. 4383 = (42.0, 429.0): cluster 3\n", + "Example n. 4384 = (42.0, 411.0): cluster 3\n", + "Example n. 4385 = (42.0, 410.0): cluster 3\n", + "Example n. 4386 = (42.0, 409.0): cluster 3\n", + "Example n. 4387 = (42.0, 408.0): cluster 3\n", + "Example n. 4388 = (42.0, 407.0): cluster 3\n", + "Example n. 4389 = (42.0, 406.0): cluster 3\n", + "Example n. 4390 = (42.0, 396.0): cluster 3\n", + "Example n. 4391 = (42.0, 395.0): cluster 3\n", + "Example n. 4392 = (42.0, 394.0): cluster 3\n", + "Example n. 4393 = (42.0, 393.0): cluster 3\n", + "Example n. 4394 = (42.0, 392.0): cluster 3\n", + "Example n. 4395 = (42.0, 391.0): cluster 3\n", + "Example n. 4396 = (42.0, 346.0): cluster 3\n", + "Example n. 4397 = (42.0, 345.0): cluster 3\n", + "Example n. 4398 = (42.0, 344.0): cluster 3\n", + "Example n. 4399 = (42.0, 343.0): cluster 3\n", + "Example n. 4400 = (42.0, 342.0): cluster 3\n", + "Example n. 4401 = (42.0, 341.0): cluster 3\n", + "Example n. 4402 = (42.0, 319.0): cluster 3\n", + "Example n. 4403 = (42.0, 318.0): cluster 3\n", + "Example n. 4404 = (42.0, 317.0): cluster 3\n", + "Example n. 4405 = (42.0, 316.0): cluster 3\n", + "Example n. 4406 = (42.0, 315.0): cluster 3\n", + "Example n. 4407 = (42.0, 314.0): cluster 3\n", + "Example n. 4408 = (42.0, 313.0): cluster 3\n", + "Example n. 4409 = (42.0, 312.0): cluster 3\n", + "Example n. 4410 = (42.0, 311.0): cluster 3\n", + "Example n. 4411 = (42.0, 310.0): cluster 3\n", + "Example n. 4412 = (42.0, 309.0): cluster 3\n", + "Example n. 4413 = (42.0, 308.0): cluster 2\n", + "Example n. 4414 = (42.0, 307.0): cluster 2\n", + "Example n. 4415 = (42.0, 306.0): cluster 2\n", + "Example n. 4416 = (42.0, 305.0): cluster 2\n", + "Example n. 4417 = (42.0, 304.0): cluster 3\n", + "Example n. 4418 = (42.0, 303.0): cluster 3\n", + "Example n. 4419 = (42.0, 299.0): cluster 3\n", + "Example n. 4420 = (42.0, 298.0): cluster 3\n", + "Example n. 4421 = (42.0, 297.0): cluster 3\n", + "Example n. 4422 = (42.0, 296.0): cluster 3\n", + "Example n. 4423 = (42.0, 291.0): cluster 3\n", + "Example n. 4424 = (42.0, 290.0): cluster 3\n", + "Example n. 4425 = (42.0, 289.0): cluster 3\n", + "Example n. 4426 = (42.0, 288.0): cluster 3\n", + "Example n. 4427 = (42.0, 287.0): cluster 3\n", + "Example n. 4428 = (42.0, 286.0): cluster 3\n", + "Example n. 4429 = (42.0, 285.0): cluster 3\n", + "Example n. 4430 = (42.0, 284.0): cluster 3\n", + "Example n. 4431 = (42.0, 283.0): cluster 3\n", + "Example n. 4432 = (42.0, 282.0): cluster 3\n", + "Example n. 4433 = (42.0, 279.0): cluster 3\n", + "Example n. 4434 = (42.0, 278.0): cluster 3\n", + "Example n. 4435 = (42.0, 277.0): cluster 3\n", + "Example n. 4436 = (42.0, 276.0): cluster 3\n", + "Example n. 4437 = (42.0, 275.0): cluster 3\n", + "Example n. 4438 = (42.0, 274.0): cluster 3\n", + "Example n. 4439 = (42.0, 271.0): cluster 3\n", + "Example n. 4440 = (42.0, 270.0): cluster 3\n", + "Example n. 4441 = (42.0, 269.0): cluster 3\n", + "Example n. 4442 = (42.0, 268.0): cluster 3\n", + "Example n. 4443 = (42.0, 267.0): cluster 3\n", + "Example n. 4444 = (42.0, 266.0): cluster 3\n", + "Example n. 4445 = (42.0, 265.0): cluster 3\n", + "Example n. 4446 = (42.0, 264.0): cluster 3\n", + "Example n. 4447 = (42.0, 263.0): cluster 3\n", + "Example n. 4448 = (42.0, 260.0): cluster 3\n", + "Example n. 4449 = (42.0, 259.0): cluster 2\n", + "Example n. 4450 = (42.0, 258.0): cluster 2\n", + "Example n. 4451 = (42.0, 257.0): cluster 2\n", + "Example n. 4452 = (42.0, 256.0): cluster 2\n", + "Example n. 4453 = (42.0, 255.0): cluster 2\n", + "Example n. 4454 = (42.0, 254.0): cluster 2\n", + "Example n. 4455 = (42.0, 250.0): cluster 2\n", + "Example n. 4456 = (42.0, 249.0): cluster 2\n", + "Example n. 4457 = (42.0, 248.0): cluster 2\n", + "Example n. 4458 = (42.0, 247.0): cluster 3\n", + "Example n. 4459 = (42.0, 246.0): cluster 3\n", + "Example n. 4460 = (42.0, 245.0): cluster 3\n", + "Example n. 4461 = (42.0, 244.0): cluster 3\n", + "Example n. 4462 = (42.0, 243.0): cluster 3\n", + "Example n. 4463 = (42.0, 242.0): cluster 3\n", + "Example n. 4464 = (42.0, 240.0): cluster 3\n", + "Example n. 4465 = (42.0, 238.0): cluster 3\n", + "Example n. 4466 = (42.0, 237.0): cluster 3\n", + "Example n. 4467 = (42.0, 236.0): cluster 3\n", + "Example n. 4468 = (42.0, 235.0): cluster 3\n", + "Example n. 4469 = (42.0, 234.0): cluster 3\n", + "Example n. 4470 = (42.0, 233.0): cluster 3\n", + "Example n. 4471 = (42.0, 232.0): cluster 3\n", + "Example n. 4472 = (42.0, 231.0): cluster 3\n", + "Example n. 4473 = (42.0, 230.0): cluster 3\n", + "Example n. 4474 = (42.0, 229.0): cluster 3\n", + "Example n. 4475 = (42.0, 228.0): cluster 3\n", + "Example n. 4476 = (42.0, 227.0): cluster 3\n", + "Example n. 4477 = (42.0, 226.0): cluster 3\n", + "Example n. 4478 = (42.0, 225.0): cluster 3\n", + "Example n. 4479 = (42.0, 215.0): cluster 3\n", + "Example n. 4480 = (42.0, 214.0): cluster 3\n", + "Example n. 4481 = (42.0, 213.0): cluster 3\n", + "Example n. 4482 = (42.0, 212.0): cluster 3\n", + "Example n. 4483 = (42.0, 211.0): cluster 3\n", + "Example n. 4484 = (42.0, 210.0): cluster 3\n", + "Example n. 4485 = (42.0, 209.0): cluster 2\n", + "Example n. 4486 = (42.0, 208.0): cluster 2\n", + "Example n. 4487 = (42.0, 207.0): cluster 2\n", + "Example n. 4488 = (42.0, 206.0): cluster 2\n", + "Example n. 4489 = (42.0, 205.0): cluster 2\n", + "Example n. 4490 = (42.0, 204.0): cluster 2\n", + "Example n. 4491 = (42.0, 203.0): cluster 2\n", + "Example n. 4492 = (42.0, 202.0): cluster 2\n", + "Example n. 4493 = (42.0, 201.0): cluster 2\n", + "Example n. 4494 = (42.0, 200.0): cluster 2\n", + "Example n. 4495 = (42.0, 199.0): cluster 3\n", + "Example n. 4496 = (42.0, 189.0): cluster 3\n", + "Example n. 4497 = (42.0, 188.0): cluster 3\n", + "Example n. 4498 = (42.0, 187.0): cluster 3\n", + "Example n. 4499 = (42.0, 186.0): cluster 3\n", + "Example n. 4500 = (42.0, 185.0): cluster 3\n", + "Example n. 4501 = (42.0, 184.0): cluster 3\n", + "Example n. 4502 = (42.0, 183.0): cluster 3\n", + "Example n. 4503 = (42.0, 182.0): cluster 3\n", + "Example n. 4504 = (42.0, 181.0): cluster 3\n", + "Example n. 4505 = (42.0, 174.0): cluster 3\n", + "Example n. 4506 = (42.0, 173.0): cluster 3\n", + "Example n. 4507 = (42.0, 172.0): cluster 3\n", + "Example n. 4508 = (42.0, 171.0): cluster 3\n", + "Example n. 4509 = (42.0, 170.0): cluster 3\n", + "Example n. 4510 = (42.0, 161.0): cluster 3\n", + "Example n. 4511 = (42.0, 160.0): cluster 3\n", + "Example n. 4512 = (42.0, 159.0): cluster 3\n", + "Example n. 4513 = (42.0, 158.0): cluster 3\n", + "Example n. 4514 = (42.0, 157.0): cluster 3\n", + "Example n. 4515 = (42.0, 156.0): cluster 3\n", + "Example n. 4516 = (42.0, 155.0): cluster 3\n", + "Example n. 4517 = (42.0, 154.0): cluster 3\n", + "Example n. 4518 = (42.0, 153.0): cluster 3\n", + "Example n. 4519 = (42.0, 152.0): cluster 3\n", + "Example n. 4520 = (42.0, 151.0): cluster 3\n", + "Example n. 4521 = (42.0, 150.0): cluster 2\n", + "Example n. 4522 = (42.0, 149.0): cluster 2\n", + "Example n. 4523 = (42.0, 146.0): cluster 2\n", + "Example n. 4524 = (42.0, 145.0): cluster 2\n", + "Example n. 4525 = (42.0, 144.0): cluster 2\n", + "Example n. 4526 = (42.0, 143.0): cluster 2\n", + "Example n. 4527 = (42.0, 142.0): cluster 2\n", + "Example n. 4528 = (42.0, 141.0): cluster 2\n", + "Example n. 4529 = (42.0, 130.0): cluster 2\n", + "Example n. 4530 = (42.0, 129.0): cluster 2\n", + "Example n. 4531 = (42.0, 128.0): cluster 2\n", + "Example n. 4532 = (42.0, 127.0): cluster 2\n", + "Example n. 4533 = (42.0, 111.0): cluster 3\n", + "Example n. 4534 = (43.0, 494.0): cluster 3\n", + "Example n. 4535 = (43.0, 493.0): cluster 3\n", + "Example n. 4536 = (43.0, 492.0): cluster 3\n", + "Example n. 4537 = (43.0, 491.0): cluster 3\n", + "Example n. 4538 = (43.0, 490.0): cluster 3\n", + "Example n. 4539 = (43.0, 489.0): cluster 3\n", + "Example n. 4540 = (43.0, 488.0): cluster 3\n", + "Example n. 4541 = (43.0, 487.0): cluster 3\n", + "Example n. 4542 = (43.0, 486.0): cluster 3\n", + "Example n. 4543 = (43.0, 485.0): cluster 3\n", + "Example n. 4544 = (43.0, 484.0): cluster 3\n", + "Example n. 4545 = (43.0, 483.0): cluster 3\n", + "Example n. 4546 = (43.0, 482.0): cluster 3\n", + "Example n. 4547 = (43.0, 481.0): cluster 3\n", + "Example n. 4548 = (43.0, 480.0): cluster 3\n", + "Example n. 4549 = (43.0, 479.0): cluster 3\n", + "Example n. 4550 = (43.0, 478.0): cluster 3\n", + "Example n. 4551 = (43.0, 477.0): cluster 3\n", + "Example n. 4552 = (43.0, 476.0): cluster 3\n", + "Example n. 4553 = (43.0, 475.0): cluster 3\n", + "Example n. 4554 = (43.0, 474.0): cluster 3\n", + "Example n. 4555 = (43.0, 473.0): cluster 3\n", + "Example n. 4556 = (43.0, 472.0): cluster 3\n", + "Example n. 4557 = (43.0, 471.0): cluster 2\n", + "Example n. 4558 = (43.0, 470.0): cluster 2\n", + "Example n. 4559 = (43.0, 469.0): cluster 2\n", + "Example n. 4560 = (43.0, 468.0): cluster 2\n", + "Example n. 4561 = (43.0, 467.0): cluster 2\n", + "Example n. 4562 = (43.0, 466.0): cluster 2\n", + "Example n. 4563 = (43.0, 465.0): cluster 2\n", + "Example n. 4564 = (43.0, 464.0): cluster 2\n", + "Example n. 4565 = (43.0, 463.0): cluster 2\n", + "Example n. 4566 = (43.0, 462.0): cluster 2\n", + "Example n. 4567 = (43.0, 461.0): cluster 2\n", + "Example n. 4568 = (43.0, 460.0): cluster 2\n", + "Example n. 4569 = (43.0, 459.0): cluster 3\n", + "Example n. 4570 = (43.0, 458.0): cluster 3\n", + "Example n. 4571 = (43.0, 457.0): cluster 3\n", + "Example n. 4572 = (43.0, 456.0): cluster 3\n", + "Example n. 4573 = (43.0, 455.0): cluster 3\n", + "Example n. 4574 = (43.0, 454.0): cluster 3\n", + "Example n. 4575 = (43.0, 434.0): cluster 3\n", + "Example n. 4576 = (43.0, 433.0): cluster 3\n", + "Example n. 4577 = (43.0, 432.0): cluster 3\n", + "Example n. 4578 = (43.0, 431.0): cluster 3\n", + "Example n. 4579 = (43.0, 430.0): cluster 3\n", + "Example n. 4580 = (43.0, 429.0): cluster 3\n", + "Example n. 4581 = (43.0, 411.0): cluster 3\n", + "Example n. 4582 = (43.0, 410.0): cluster 3\n", + "Example n. 4583 = (43.0, 409.0): cluster 3\n", + "Example n. 4584 = (43.0, 408.0): cluster 3\n", + "Example n. 4585 = (43.0, 407.0): cluster 3\n", + "Example n. 4586 = (43.0, 396.0): cluster 3\n", + "Example n. 4587 = (43.0, 395.0): cluster 3\n", + "Example n. 4588 = (43.0, 394.0): cluster 3\n", + "Example n. 4589 = (43.0, 393.0): cluster 3\n", + "Example n. 4590 = (43.0, 392.0): cluster 3\n", + "Example n. 4591 = (43.0, 346.0): cluster 3\n", + "Example n. 4592 = (43.0, 345.0): cluster 3\n", + "Example n. 4593 = (43.0, 344.0): cluster 2\n", + "Example n. 4594 = (43.0, 343.0): cluster 2\n", + "Example n. 4595 = (43.0, 342.0): cluster 2\n", + "Example n. 4596 = (43.0, 341.0): cluster 2\n", + "Example n. 4597 = (43.0, 340.0): cluster 2\n", + "Example n. 4598 = (43.0, 339.0): cluster 2\n", + "Example n. 4599 = (43.0, 338.0): cluster 2\n", + "Example n. 4600 = (43.0, 319.0): cluster 2\n", + "Example n. 4601 = (43.0, 318.0): cluster 2\n", + "Example n. 4602 = (43.0, 317.0): cluster 2\n", + "Example n. 4603 = (43.0, 316.0): cluster 2\n", + "Example n. 4604 = (43.0, 315.0): cluster 2\n", + "Example n. 4605 = (43.0, 314.0): cluster 2\n", + "Example n. 4606 = (43.0, 313.0): cluster 2\n", + "Example n. 4607 = (43.0, 312.0): cluster 2\n", + "Example n. 4608 = (43.0, 311.0): cluster 2\n", + "Example n. 4609 = (43.0, 310.0): cluster 2\n", + "Example n. 4610 = (43.0, 309.0): cluster 2\n", + "Example n. 4611 = (43.0, 308.0): cluster 2\n", + "Example n. 4612 = (43.0, 307.0): cluster 2\n", + "Example n. 4613 = (43.0, 306.0): cluster 2\n", + "Example n. 4614 = (43.0, 305.0): cluster 2\n", + "Example n. 4615 = (43.0, 304.0): cluster 2\n", + "Example n. 4616 = (43.0, 303.0): cluster 2\n", + "Example n. 4617 = (43.0, 302.0): cluster 3\n", + "Example n. 4618 = (43.0, 301.0): cluster 3\n", + "Example n. 4619 = (43.0, 299.0): cluster 3\n", + "Example n. 4620 = (43.0, 298.0): cluster 3\n", + "Example n. 4621 = (43.0, 297.0): cluster 3\n", + "Example n. 4622 = (43.0, 296.0): cluster 3\n", + "Example n. 4623 = (43.0, 295.0): cluster 3\n", + "Example n. 4624 = (43.0, 291.0): cluster 3\n", + "Example n. 4625 = (43.0, 290.0): cluster 3\n", + "Example n. 4626 = (43.0, 289.0): cluster 3\n", + "Example n. 4627 = (43.0, 288.0): cluster 3\n", + "Example n. 4628 = (43.0, 287.0): cluster 3\n", + "Example n. 4629 = (43.0, 286.0): cluster 3\n", + "Example n. 4630 = (43.0, 285.0): cluster 3\n", + "Example n. 4631 = (43.0, 284.0): cluster 3\n", + "Example n. 4632 = (43.0, 283.0): cluster 3\n", + "Example n. 4633 = (43.0, 282.0): cluster 3\n", + "Example n. 4634 = (43.0, 279.0): cluster 3\n", + "Example n. 4635 = (43.0, 278.0): cluster 3\n", + "Example n. 4636 = (43.0, 277.0): cluster 3\n", + "Example n. 4637 = (43.0, 276.0): cluster 2\n", + "Example n. 4638 = (43.0, 275.0): cluster 2\n", + "Example n. 4639 = (43.0, 274.0): cluster 2\n", + "Example n. 4640 = (43.0, 270.0): cluster 2\n", + "Example n. 4641 = (43.0, 269.0): cluster 2\n", + "Example n. 4642 = (43.0, 268.0): cluster 2\n", + "Example n. 4643 = (43.0, 267.0): cluster 2\n", + "Example n. 4644 = (43.0, 266.0): cluster 2\n", + "Example n. 4645 = (43.0, 265.0): cluster 2\n", + "Example n. 4646 = (43.0, 264.0): cluster 2\n", + "Example n. 4647 = (43.0, 263.0): cluster 2\n", + "Example n. 4648 = (43.0, 262.0): cluster 2\n", + "Example n. 4649 = (43.0, 261.0): cluster 2\n", + "Example n. 4650 = (43.0, 260.0): cluster 2\n", + "Example n. 4651 = (43.0, 259.0): cluster 2\n", + "Example n. 4652 = (43.0, 258.0): cluster 2\n", + "Example n. 4653 = (43.0, 257.0): cluster 2\n", + "Example n. 4654 = (43.0, 256.0): cluster 2\n", + "Example n. 4655 = (43.0, 255.0): cluster 2\n", + "Example n. 4656 = (43.0, 254.0): cluster 2\n", + "Example n. 4657 = (43.0, 253.0): cluster 2\n", + "Example n. 4658 = (43.0, 250.0): cluster 2\n", + "Example n. 4659 = (43.0, 249.0): cluster 2\n", + "Example n. 4660 = (43.0, 248.0): cluster 3\n", + "Example n. 4661 = (43.0, 247.0): cluster 3\n", + "Example n. 4662 = (43.0, 246.0): cluster 3\n", + "Example n. 4663 = (43.0, 245.0): cluster 3\n", + "Example n. 4664 = (43.0, 242.0): cluster 3\n", + "Example n. 4665 = (43.0, 241.0): cluster 3\n", + "Example n. 4666 = (43.0, 240.0): cluster 3\n", + "Example n. 4667 = (43.0, 239.0): cluster 3\n", + "Example n. 4668 = (43.0, 238.0): cluster 3\n", + "Example n. 4669 = (43.0, 237.0): cluster 3\n", + "Example n. 4670 = (43.0, 236.0): cluster 3\n", + "Example n. 4671 = (43.0, 235.0): cluster 3\n", + "Example n. 4672 = (43.0, 234.0): cluster 3\n", + "Example n. 4673 = (43.0, 233.0): cluster 3\n", + "Example n. 4674 = (43.0, 232.0): cluster 3\n", + "Example n. 4675 = (43.0, 229.0): cluster 3\n", + "Example n. 4676 = (43.0, 228.0): cluster 3\n", + "Example n. 4677 = (43.0, 227.0): cluster 3\n", + "Example n. 4678 = (43.0, 226.0): cluster 3\n", + "Example n. 4679 = (43.0, 225.0): cluster 3\n", + "Example n. 4680 = (43.0, 218.0): cluster 2\n", + "Example n. 4681 = (43.0, 215.0): cluster 2\n", + "Example n. 4682 = (43.0, 214.0): cluster 2\n", + "Example n. 4683 = (43.0, 213.0): cluster 2\n", + "Example n. 4684 = (43.0, 212.0): cluster 2\n", + "Example n. 4685 = (43.0, 211.0): cluster 2\n", + "Example n. 4686 = (43.0, 209.0): cluster 2\n", + "Example n. 4687 = (43.0, 208.0): cluster 2\n", + "Example n. 4688 = (43.0, 207.0): cluster 2\n", + "Example n. 4689 = (43.0, 206.0): cluster 2\n", + "Example n. 4690 = (43.0, 205.0): cluster 2\n", + "Example n. 4691 = (43.0, 204.0): cluster 2\n", + "Example n. 4692 = (43.0, 203.0): cluster 2\n", + "Example n. 4693 = (43.0, 202.0): cluster 2\n", + "Example n. 4694 = (43.0, 201.0): cluster 2\n", + "Example n. 4695 = (43.0, 200.0): cluster 2\n", + "Example n. 4696 = (43.0, 199.0): cluster 2\n", + "Example n. 4697 = (43.0, 198.0): cluster 2\n", + "Example n. 4698 = (43.0, 189.0): cluster 2\n", + "Example n. 4699 = (43.0, 188.0): cluster 2\n", + "Example n. 4700 = (43.0, 187.0): cluster 2\n", + "Example n. 4701 = (43.0, 186.0): cluster 2\n", + "Example n. 4702 = (43.0, 185.0): cluster 2\n", + "Example n. 4703 = (43.0, 184.0): cluster 2\n", + "Example n. 4704 = (43.0, 183.0): cluster 3\n", + "Example n. 4705 = (43.0, 182.0): cluster 3\n", + "Example n. 4706 = (43.0, 181.0): cluster 3\n", + "Example n. 4707 = (43.0, 178.0): cluster 3\n", + "Example n. 4708 = (43.0, 177.0): cluster 3\n", + "Example n. 4709 = (43.0, 176.0): cluster 3\n", + "Example n. 4710 = (43.0, 175.0): cluster 3\n", + "Example n. 4711 = (43.0, 173.0): cluster 3\n", + "Example n. 4712 = (43.0, 172.0): cluster 3\n", + "Example n. 4713 = (43.0, 171.0): cluster 3\n", + "Example n. 4714 = (43.0, 170.0): cluster 3\n", + "Example n. 4715 = (43.0, 169.0): cluster 3\n", + "Example n. 4716 = (43.0, 168.0): cluster 3\n", + "Example n. 4717 = (43.0, 161.0): cluster 3\n", + "Example n. 4718 = (43.0, 160.0): cluster 3\n", + "Example n. 4719 = (43.0, 159.0): cluster 3\n", + "Example n. 4720 = (43.0, 158.0): cluster 3\n", + "Example n. 4721 = (43.0, 157.0): cluster 3\n", + "Example n. 4722 = (43.0, 156.0): cluster 3\n", + "Example n. 4723 = (43.0, 155.0): cluster 3\n", + "Example n. 4724 = (43.0, 154.0): cluster 2\n", + "Example n. 4725 = (43.0, 153.0): cluster 2\n", + "Example n. 4726 = (43.0, 152.0): cluster 2\n", + "Example n. 4727 = (43.0, 151.0): cluster 2\n", + "Example n. 4728 = (43.0, 150.0): cluster 2\n", + "Example n. 4729 = (43.0, 149.0): cluster 2\n", + "Example n. 4730 = (43.0, 148.0): cluster 2\n", + "Example n. 4731 = (43.0, 146.0): cluster 2\n", + "Example n. 4732 = (43.0, 145.0): cluster 2\n", + "Example n. 4733 = (43.0, 144.0): cluster 2\n", + "Example n. 4734 = (43.0, 143.0): cluster 2\n", + "Example n. 4735 = (43.0, 142.0): cluster 2\n", + "Example n. 4736 = (43.0, 130.0): cluster 2\n", + "Example n. 4737 = (43.0, 129.0): cluster 2\n", + "Example n. 4738 = (43.0, 128.0): cluster 2\n", + "Example n. 4739 = (43.0, 127.0): cluster 2\n", + "Example n. 4740 = (43.0, 126.0): cluster 2\n", + "Example n. 4741 = (43.0, 120.0): cluster 2\n", + "Example n. 4742 = (43.0, 119.0): cluster 2\n", + "Example n. 4743 = (43.0, 118.0): cluster 2\n", + "Example n. 4744 = (43.0, 117.0): cluster 2\n", + "Example n. 4745 = (43.0, 116.0): cluster 2\n", + "Example n. 4746 = (44.0, 495.0): cluster 2\n", + "Example n. 4747 = (44.0, 494.0): cluster 2\n", + "Example n. 4748 = (44.0, 493.0): cluster 2\n", + "Example n. 4749 = (44.0, 492.0): cluster 2\n", + "Example n. 4750 = (44.0, 491.0): cluster 2\n", + "Example n. 4751 = (44.0, 490.0): cluster 2\n", + "Example n. 4752 = (44.0, 489.0): cluster 2\n", + "Example n. 4753 = (44.0, 488.0): cluster 2\n", + "Example n. 4754 = (44.0, 487.0): cluster 2\n", + "Example n. 4755 = (44.0, 486.0): cluster 2\n", + "Example n. 4756 = (44.0, 485.0): cluster 3\n", + "Example n. 4757 = (44.0, 484.0): cluster 3\n", + "Example n. 4758 = (44.0, 483.0): cluster 3\n", + "Example n. 4759 = (44.0, 482.0): cluster 3\n", + "Example n. 4760 = (44.0, 481.0): cluster 3\n", + "Example n. 4761 = (44.0, 480.0): cluster 3\n", + "Example n. 4762 = (44.0, 479.0): cluster 3\n", + "Example n. 4763 = (44.0, 478.0): cluster 3\n", + "Example n. 4764 = (44.0, 477.0): cluster 3\n", + "Example n. 4765 = (44.0, 476.0): cluster 3\n", + "Example n. 4766 = (44.0, 475.0): cluster 3\n", + "Example n. 4767 = (44.0, 474.0): cluster 3\n", + "Example n. 4768 = (44.0, 473.0): cluster 3\n", + "Example n. 4769 = (44.0, 472.0): cluster 3\n", + "Example n. 4770 = (44.0, 471.0): cluster 3\n", + "Example n. 4771 = (44.0, 470.0): cluster 3\n", + "Example n. 4772 = (44.0, 469.0): cluster 3\n", + "Example n. 4773 = (44.0, 468.0): cluster 3\n", + "Example n. 4774 = (44.0, 467.0): cluster 3\n", + "Example n. 4775 = (44.0, 466.0): cluster 3\n", + "Example n. 4776 = (44.0, 465.0): cluster 2\n", + "Example n. 4777 = (44.0, 464.0): cluster 2\n", + "Example n. 4778 = (44.0, 463.0): cluster 2\n", + "Example n. 4779 = (44.0, 462.0): cluster 2\n", + "Example n. 4780 = (44.0, 461.0): cluster 2\n", + "Example n. 4781 = (44.0, 460.0): cluster 2\n", + "Example n. 4782 = (44.0, 459.0): cluster 2\n", + "Example n. 4783 = (44.0, 458.0): cluster 2\n", + "Example n. 4784 = (44.0, 457.0): cluster 2\n", + "Example n. 4785 = (44.0, 456.0): cluster 2\n", + "Example n. 4786 = (44.0, 455.0): cluster 2\n", + "Example n. 4787 = (44.0, 454.0): cluster 2\n", + "Example n. 4788 = (44.0, 433.0): cluster 2\n", + "Example n. 4789 = (44.0, 432.0): cluster 2\n", + "Example n. 4790 = (44.0, 431.0): cluster 2\n", + "Example n. 4791 = (44.0, 430.0): cluster 2\n", + "Example n. 4792 = (44.0, 429.0): cluster 2\n", + "Example n. 4793 = (44.0, 410.0): cluster 2\n", + "Example n. 4794 = (44.0, 409.0): cluster 2\n", + "Example n. 4795 = (44.0, 408.0): cluster 2\n", + "Example n. 4796 = (44.0, 407.0): cluster 2\n", + "Example n. 4797 = (44.0, 396.0): cluster 2\n", + "Example n. 4798 = (44.0, 395.0): cluster 2\n", + "Example n. 4799 = (44.0, 394.0): cluster 2\n", + "Example n. 4800 = (44.0, 393.0): cluster 2\n", + "Example n. 4801 = (44.0, 392.0): cluster 2\n", + "Example n. 4802 = (44.0, 345.0): cluster 2\n", + "Example n. 4803 = (44.0, 344.0): cluster 2\n", + "Example n. 4804 = (44.0, 343.0): cluster 2\n", + "Example n. 4805 = (44.0, 342.0): cluster 2\n", + "Example n. 4806 = (44.0, 341.0): cluster 2\n", + "Example n. 4807 = (44.0, 340.0): cluster 2\n", + "Example n. 4808 = (44.0, 339.0): cluster 3\n", + "Example n. 4809 = (44.0, 338.0): cluster 3\n", + "Example n. 4810 = (44.0, 337.0): cluster 3\n", + "Example n. 4811 = (44.0, 327.0): cluster 3\n", + "Example n. 4812 = (44.0, 326.0): cluster 3\n", + "Example n. 4813 = (44.0, 325.0): cluster 3\n", + "Example n. 4814 = (44.0, 324.0): cluster 3\n", + "Example n. 4815 = (44.0, 323.0): cluster 3\n", + "Example n. 4816 = (44.0, 322.0): cluster 3\n", + "Example n. 4817 = (44.0, 321.0): cluster 3\n", + "Example n. 4818 = (44.0, 320.0): cluster 3\n", + "Example n. 4819 = (44.0, 319.0): cluster 2\n", + "Example n. 4820 = (44.0, 318.0): cluster 2\n", + "Example n. 4821 = (44.0, 317.0): cluster 2\n", + "Example n. 4822 = (44.0, 316.0): cluster 2\n", + "Example n. 4823 = (44.0, 315.0): cluster 2\n", + "Example n. 4824 = (44.0, 314.0): cluster 2\n", + "Example n. 4825 = (44.0, 313.0): cluster 2\n", + "Example n. 4826 = (44.0, 312.0): cluster 2\n", + "Example n. 4827 = (44.0, 311.0): cluster 2\n", + "Example n. 4828 = (44.0, 310.0): cluster 2\n", + "Example n. 4829 = (44.0, 309.0): cluster 2\n", + "Example n. 4830 = (44.0, 308.0): cluster 2\n", + "Example n. 4831 = (44.0, 307.0): cluster 2\n", + "Example n. 4832 = (44.0, 306.0): cluster 2\n", + "Example n. 4833 = (44.0, 305.0): cluster 2\n", + "Example n. 4834 = (44.0, 304.0): cluster 2\n", + "Example n. 4835 = (44.0, 303.0): cluster 2\n", + "Example n. 4836 = (44.0, 302.0): cluster 2\n", + "Example n. 4837 = (44.0, 301.0): cluster 2\n", + "Example n. 4838 = (44.0, 300.0): cluster 2\n", + "Example n. 4839 = (44.0, 299.0): cluster 2\n", + "Example n. 4840 = (44.0, 298.0): cluster 2\n", + "Example n. 4841 = (44.0, 297.0): cluster 2\n", + "Example n. 4842 = (44.0, 296.0): cluster 2\n", + "Example n. 4843 = (44.0, 295.0): cluster 2\n", + "Example n. 4844 = (44.0, 294.0): cluster 2\n", + "Example n. 4845 = (44.0, 289.0): cluster 2\n", + "Example n. 4846 = (44.0, 288.0): cluster 2\n", + "Example n. 4847 = (44.0, 287.0): cluster 2\n", + "Example n. 4848 = (44.0, 286.0): cluster 2\n", + "Example n. 4849 = (44.0, 285.0): cluster 2\n", + "Example n. 4850 = (44.0, 284.0): cluster 2\n", + "Example n. 4851 = (44.0, 283.0): cluster 2\n", + "Example n. 4852 = (44.0, 282.0): cluster 2\n", + "Example n. 4853 = (44.0, 279.0): cluster 3\n", + "Example n. 4854 = (44.0, 278.0): cluster 3\n", + "Example n. 4855 = (44.0, 277.0): cluster 3\n", + "Example n. 4856 = (44.0, 276.0): cluster 3\n", + "Example n. 4857 = (44.0, 275.0): cluster 3\n", + "Example n. 4858 = (44.0, 274.0): cluster 3\n", + "Example n. 4859 = (44.0, 267.0): cluster 3\n", + "Example n. 4860 = (44.0, 266.0): cluster 3\n", + "Example n. 4861 = (44.0, 265.0): cluster 3\n", + "Example n. 4862 = (44.0, 264.0): cluster 3\n", + "Example n. 4863 = (44.0, 263.0): cluster 2\n", + "Example n. 4864 = (44.0, 262.0): cluster 2\n", + "Example n. 4865 = (44.0, 261.0): cluster 2\n", + "Example n. 4866 = (44.0, 260.0): cluster 2\n", + "Example n. 4867 = (44.0, 259.0): cluster 2\n", + "Example n. 4868 = (44.0, 258.0): cluster 2\n", + "Example n. 4869 = (44.0, 257.0): cluster 2\n", + "Example n. 4870 = (44.0, 256.0): cluster 2\n", + "Example n. 4871 = (44.0, 255.0): cluster 2\n", + "Example n. 4872 = (44.0, 254.0): cluster 2\n", + "Example n. 4873 = (44.0, 253.0): cluster 2\n", + "Example n. 4874 = (44.0, 252.0): cluster 2\n", + "Example n. 4875 = (44.0, 251.0): cluster 2\n", + "Example n. 4876 = (44.0, 250.0): cluster 2\n", + "Example n. 4877 = (44.0, 249.0): cluster 2\n", + "Example n. 4878 = (44.0, 248.0): cluster 2\n", + "Example n. 4879 = (44.0, 247.0): cluster 2\n", + "Example n. 4880 = (44.0, 246.0): cluster 2\n", + "Example n. 4881 = (44.0, 243.0): cluster 2\n", + "Example n. 4882 = (44.0, 242.0): cluster 2\n", + "Example n. 4883 = (44.0, 241.0): cluster 2\n", + "Example n. 4884 = (44.0, 240.0): cluster 2\n", + "Example n. 4885 = (44.0, 239.0): cluster 2\n", + "Example n. 4886 = (44.0, 238.0): cluster 2\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Example n. 4887 = (44.0, 237.0): cluster 2\n", + "Example n. 4888 = (44.0, 236.0): cluster 2\n", + "Example n. 4889 = (44.0, 235.0): cluster 2\n", + "Example n. 4890 = (44.0, 234.0): cluster 2\n", + "Example n. 4891 = (44.0, 229.0): cluster 2\n", + "Example n. 4892 = (44.0, 228.0): cluster 2\n", + "Example n. 4893 = (44.0, 227.0): cluster 2\n", + "Example n. 4894 = (44.0, 226.0): cluster 2\n", + "Example n. 4895 = (44.0, 225.0): cluster 2\n", + "Example n. 4896 = (44.0, 220.0): cluster 2\n", + "Example n. 4897 = (44.0, 219.0): cluster 2\n", + "Example n. 4898 = (44.0, 218.0): cluster 3\n", + "Example n. 4899 = (44.0, 217.0): cluster 3\n", + "Example n. 4900 = (44.0, 216.0): cluster 3\n", + "Example n. 4901 = (44.0, 213.0): cluster 3\n", + "Example n. 4902 = (44.0, 209.0): cluster 3\n", + "Example n. 4903 = (44.0, 208.0): cluster 3\n", + "Example n. 4904 = (44.0, 207.0): cluster 3\n", + "Example n. 4905 = (44.0, 206.0): cluster 3\n", + "Example n. 4906 = (44.0, 205.0): cluster 3\n", + "Example n. 4907 = (44.0, 204.0): cluster 2\n", + "Example n. 4908 = (44.0, 203.0): cluster 2\n", + "Example n. 4909 = (44.0, 202.0): cluster 2\n", + "Example n. 4910 = (44.0, 201.0): cluster 2\n", + "Example n. 4911 = (44.0, 200.0): cluster 2\n", + "Example n. 4912 = (44.0, 199.0): cluster 2\n", + "Example n. 4913 = (44.0, 198.0): cluster 2\n", + "Example n. 4914 = (44.0, 189.0): cluster 2\n", + "Example n. 4915 = (44.0, 188.0): cluster 2\n", + "Example n. 4916 = (44.0, 187.0): cluster 2\n", + "Example n. 4917 = (44.0, 186.0): cluster 2\n", + "Example n. 4918 = (44.0, 185.0): cluster 2\n", + "Example n. 4919 = (44.0, 184.0): cluster 2\n", + "Example n. 4920 = (44.0, 183.0): cluster 2\n", + "Example n. 4921 = (44.0, 182.0): cluster 2\n", + "Example n. 4922 = (44.0, 181.0): cluster 2\n", + "Example n. 4923 = (44.0, 179.0): cluster 2\n", + "Example n. 4924 = (44.0, 178.0): cluster 2\n", + "Example n. 4925 = (44.0, 177.0): cluster 2\n", + "Example n. 4926 = (44.0, 176.0): cluster 2\n", + "Example n. 4927 = (44.0, 175.0): cluster 2\n", + "Example n. 4928 = (44.0, 174.0): cluster 2\n", + "Example n. 4929 = (44.0, 173.0): cluster 2\n", + "Example n. 4930 = (44.0, 172.0): cluster 2\n", + "Example n. 4931 = (44.0, 171.0): cluster 2\n", + "Example n. 4932 = (44.0, 170.0): cluster 2\n", + "Example n. 4933 = (44.0, 169.0): cluster 2\n", + "Example n. 4934 = (44.0, 168.0): cluster 2\n", + "Example n. 4935 = (44.0, 167.0): cluster 2\n", + "Example n. 4936 = (44.0, 166.0): cluster 2\n", + "Example n. 4937 = (44.0, 161.0): cluster 2\n", + "Example n. 4938 = (44.0, 160.0): cluster 2\n", + "Example n. 4939 = (44.0, 159.0): cluster 2\n", + "Example n. 4940 = (44.0, 158.0): cluster 2\n", + "Example n. 4941 = (44.0, 157.0): cluster 2\n", + "Example n. 4942 = (44.0, 156.0): cluster 2\n", + "Example n. 4943 = (44.0, 155.0): cluster 2\n", + "Example n. 4944 = (44.0, 154.0): cluster 2\n", + "Example n. 4945 = (44.0, 153.0): cluster 2\n", + "Example n. 4946 = (44.0, 152.0): cluster 2\n", + "Example n. 4947 = (44.0, 151.0): cluster 2\n", + "Example n. 4948 = (44.0, 150.0): cluster 2\n", + "Example n. 4949 = (44.0, 149.0): cluster 3\n", + "Example n. 4950 = (44.0, 148.0): cluster 3\n", + "Example n. 4951 = (44.0, 145.0): cluster 3\n", + "Example n. 4952 = (44.0, 144.0): cluster 3\n", + "Example n. 4953 = (44.0, 143.0): cluster 3\n", + "Example n. 4954 = (44.0, 142.0): cluster 3\n", + "Example n. 4955 = (44.0, 131.0): cluster 3\n", + "Example n. 4956 = (44.0, 130.0): cluster 2\n", + "Example n. 4957 = (44.0, 129.0): cluster 2\n", + "Example n. 4958 = (44.0, 128.0): cluster 2\n", + "Example n. 4959 = (44.0, 127.0): cluster 2\n", + "Example n. 4960 = (44.0, 126.0): cluster 2\n", + "Example n. 4961 = (44.0, 120.0): cluster 2\n", + "Example n. 4962 = (44.0, 119.0): cluster 2\n", + "Example n. 4963 = (44.0, 118.0): cluster 2\n", + "Example n. 4964 = (44.0, 117.0): cluster 2\n", + "Example n. 4965 = (44.0, 116.0): cluster 2\n", + "Example n. 4966 = (45.0, 496.0): cluster 2\n", + "Example n. 4967 = (45.0, 495.0): cluster 2\n", + "Example n. 4968 = (45.0, 494.0): cluster 2\n", + "Example n. 4969 = (45.0, 493.0): cluster 2\n", + "Example n. 4970 = (45.0, 492.0): cluster 2\n", + "Example n. 4971 = (45.0, 491.0): cluster 2\n", + "Example n. 4972 = (45.0, 490.0): cluster 2\n", + "Example n. 4973 = (45.0, 489.0): cluster 2\n", + "Example n. 4974 = (45.0, 488.0): cluster 2\n", + "Example n. 4975 = (45.0, 487.0): cluster 2\n", + "Example n. 4976 = (45.0, 486.0): cluster 2\n", + "Example n. 4977 = (45.0, 485.0): cluster 2\n", + "Example n. 4978 = (45.0, 484.0): cluster 2\n", + "Example n. 4979 = (45.0, 483.0): cluster 2\n", + "Example n. 4980 = (45.0, 482.0): cluster 2\n", + "Example n. 4981 = (45.0, 481.0): cluster 2\n", + "Example n. 4982 = (45.0, 480.0): cluster 2\n", + "Example n. 4983 = (45.0, 479.0): cluster 2\n", + "Example n. 4984 = (45.0, 478.0): cluster 2\n", + "Example n. 4985 = (45.0, 477.0): cluster 2\n", + "Example n. 4986 = (45.0, 476.0): cluster 2\n", + "Example n. 4987 = (45.0, 475.0): cluster 2\n", + "Example n. 4988 = (45.0, 474.0): cluster 2\n", + "Example n. 4989 = (45.0, 473.0): cluster 2\n", + "Example n. 4990 = (45.0, 472.0): cluster 2\n", + "Example n. 4991 = (45.0, 471.0): cluster 2\n", + "Example n. 4992 = (45.0, 470.0): cluster 2\n", + "Example n. 4993 = (45.0, 469.0): cluster 2\n", + "Example n. 4994 = (45.0, 468.0): cluster 2\n", + "Example n. 4995 = (45.0, 467.0): cluster 2\n", + "Example n. 4996 = (45.0, 466.0): cluster 2\n", + "Example n. 4997 = (45.0, 465.0): cluster 2\n", + "Example n. 4998 = (45.0, 464.0): cluster 2\n", + "Example n. 4999 = (45.0, 463.0): cluster 2\n", + "Example n. 5000 = (45.0, 462.0): cluster 2\n", + "Example n. 5001 = (45.0, 461.0): cluster 2\n", + "Example n. 5002 = (45.0, 460.0): cluster 2\n", + "Example n. 5003 = (45.0, 459.0): cluster 2\n", + "Example n. 5004 = (45.0, 458.0): cluster 2\n", + "Example n. 5005 = (45.0, 457.0): cluster 2\n", + "Example n. 5006 = (45.0, 456.0): cluster 2\n", + "Example n. 5007 = (45.0, 455.0): cluster 2\n", + "Example n. 5008 = (45.0, 454.0): cluster 2\n", + "Example n. 5009 = (45.0, 453.0): cluster 2\n", + "Example n. 5010 = (45.0, 433.0): cluster 2\n", + "Example n. 5011 = (45.0, 432.0): cluster 2\n", + "Example n. 5012 = (45.0, 431.0): cluster 2\n", + "Example n. 5013 = (45.0, 430.0): cluster 2\n", + "Example n. 5014 = (45.0, 394.0): cluster 2\n", + "Example n. 5015 = (45.0, 341.0): cluster 2\n", + "Example n. 5016 = (45.0, 340.0): cluster 2\n", + "Example n. 5017 = (45.0, 339.0): cluster 2\n", + "Example n. 5018 = (45.0, 338.0): cluster 2\n", + "Example n. 5019 = (45.0, 337.0): cluster 2\n", + "Example n. 5020 = (45.0, 336.0): cluster 2\n", + "Example n. 5021 = (45.0, 327.0): cluster 2\n", + "Example n. 5022 = (45.0, 326.0): cluster 2\n", + "Example n. 5023 = (45.0, 325.0): cluster 2\n", + "Example n. 5024 = (45.0, 324.0): cluster 2\n", + "Example n. 5025 = (45.0, 323.0): cluster 2\n", + "Example n. 5026 = (45.0, 322.0): cluster 2\n", + "Example n. 5027 = (45.0, 321.0): cluster 2\n", + "Example n. 5028 = (45.0, 320.0): cluster 2\n", + "Example n. 5029 = (45.0, 319.0): cluster 2\n", + "Example n. 5030 = (45.0, 318.0): cluster 2\n", + "Example n. 5031 = (45.0, 317.0): cluster 2\n", + "Example n. 5032 = (45.0, 316.0): cluster 2\n", + "Example n. 5033 = (45.0, 315.0): cluster 2\n", + "Example n. 5034 = (45.0, 314.0): cluster 2\n", + "Example n. 5035 = (45.0, 313.0): cluster 2\n", + "Example n. 5036 = (45.0, 312.0): cluster 2\n", + "Example n. 5037 = (45.0, 311.0): cluster 2\n", + "Example n. 5038 = (45.0, 310.0): cluster 2\n", + "Example n. 5039 = (45.0, 309.0): cluster 2\n", + "Example n. 5040 = (45.0, 308.0): cluster 2\n", + "Example n. 5041 = (45.0, 307.0): cluster 2\n", + "Example n. 5042 = (45.0, 306.0): cluster 2\n", + "Example n. 5043 = (45.0, 305.0): cluster 2\n", + "Example n. 5044 = (45.0, 304.0): cluster 2\n", + "Example n. 5045 = (45.0, 303.0): cluster 2\n", + "Example n. 5046 = (45.0, 302.0): cluster 2\n", + "Example n. 5047 = (45.0, 301.0): cluster 2\n", + "Example n. 5048 = (45.0, 300.0): cluster 2\n", + "Example n. 5049 = (45.0, 299.0): cluster 2\n", + "Example n. 5050 = (45.0, 298.0): cluster 2\n", + "Example n. 5051 = (45.0, 297.0): cluster 2\n", + "Example n. 5052 = (45.0, 296.0): cluster 2\n", + "Example n. 5053 = (45.0, 295.0): cluster 2\n", + "Example n. 5054 = (45.0, 294.0): cluster 2\n", + "Example n. 5055 = (45.0, 288.0): cluster 2\n", + "Example n. 5056 = (45.0, 287.0): cluster 2\n", + "Example n. 5057 = (45.0, 286.0): cluster 2\n", + "Example n. 5058 = (45.0, 285.0): cluster 2\n", + "Example n. 5059 = (45.0, 284.0): cluster 2\n", + "Example n. 5060 = (45.0, 283.0): cluster 2\n", + "Example n. 5061 = (45.0, 282.0): cluster 2\n", + "Example n. 5062 = (45.0, 281.0): cluster 2\n", + "Example n. 5063 = (45.0, 279.0): cluster 2\n", + "Example n. 5064 = (45.0, 278.0): cluster 2\n", + "Example n. 5065 = (45.0, 277.0): cluster 2\n", + "Example n. 5066 = (45.0, 276.0): cluster 2\n", + "Example n. 5067 = (45.0, 275.0): cluster 2\n", + "Example n. 5068 = (45.0, 266.0): cluster 2\n", + "Example n. 5069 = (45.0, 265.0): cluster 2\n", + "Example n. 5070 = (45.0, 262.0): cluster 2\n", + "Example n. 5071 = (45.0, 261.0): cluster 2\n", + "Example n. 5072 = (45.0, 260.0): cluster 2\n", + "Example n. 5073 = (45.0, 259.0): cluster 2\n", + "Example n. 5074 = (45.0, 258.0): cluster 2\n", + "Example n. 5075 = (45.0, 257.0): cluster 2\n", + "Example n. 5076 = (45.0, 256.0): cluster 2\n", + "Example n. 5077 = (45.0, 255.0): cluster 2\n", + "Example n. 5078 = (45.0, 254.0): cluster 2\n", + "Example n. 5079 = (45.0, 253.0): cluster 2\n", + "Example n. 5080 = (45.0, 252.0): cluster 2\n", + "Example n. 5081 = (45.0, 251.0): cluster 2\n", + "Example n. 5082 = (45.0, 250.0): cluster 2\n", + "Example n. 5083 = (45.0, 249.0): cluster 2\n", + "Example n. 5084 = (45.0, 244.0): cluster 2\n", + "Example n. 5085 = (45.0, 243.0): cluster 2\n", + "Example n. 5086 = (45.0, 242.0): cluster 2\n", + "Example n. 5087 = (45.0, 241.0): cluster 2\n", + "Example n. 5088 = (45.0, 240.0): cluster 2\n", + "Example n. 5089 = (45.0, 239.0): cluster 2\n", + "Example n. 5090 = (45.0, 238.0): cluster 2\n", + "Example n. 5091 = (45.0, 237.0): cluster 2\n", + "Example n. 5092 = (45.0, 236.0): cluster 2\n", + "Example n. 5093 = (45.0, 235.0): cluster 2\n", + "Example n. 5094 = (45.0, 229.0): cluster 2\n", + "Example n. 5095 = (45.0, 228.0): cluster 2\n", + "Example n. 5096 = (45.0, 227.0): cluster 2\n", + "Example n. 5097 = (45.0, 226.0): cluster 2\n", + "Example n. 5098 = (45.0, 225.0): cluster 2\n", + "Example n. 5099 = (45.0, 224.0): cluster 2\n", + "Example n. 5100 = (45.0, 220.0): cluster 2\n", + "Example n. 5101 = (45.0, 219.0): cluster 2\n", + "Example n. 5102 = (45.0, 218.0): cluster 2\n", + "Example n. 5103 = (45.0, 217.0): cluster 2\n", + "Example n. 5104 = (45.0, 216.0): cluster 2\n", + "Example n. 5105 = (45.0, 209.0): cluster 2\n", + "Example n. 5106 = (45.0, 208.0): cluster 2\n", + "Example n. 5107 = (45.0, 207.0): cluster 2\n", + "Example n. 5108 = (45.0, 206.0): cluster 2\n", + "Example n. 5109 = (45.0, 205.0): cluster 2\n", + "Example n. 5110 = (45.0, 202.0): cluster 2\n", + "Example n. 5111 = (45.0, 201.0): cluster 2\n", + "Example n. 5112 = (45.0, 200.0): cluster 2\n", + "Example n. 5113 = (45.0, 199.0): cluster 2\n", + "Example n. 5114 = (45.0, 198.0): cluster 2\n", + "Example n. 5115 = (45.0, 197.0): cluster 2\n", + "Example n. 5116 = (45.0, 189.0): cluster 2\n", + "Example n. 5117 = (45.0, 188.0): cluster 2\n", + "Example n. 5118 = (45.0, 187.0): cluster 2\n", + "Example n. 5119 = (45.0, 186.0): cluster 2\n", + "Example n. 5120 = (45.0, 185.0): cluster 2\n", + "Example n. 5121 = (45.0, 184.0): cluster 2\n", + "Example n. 5122 = (45.0, 183.0): cluster 2\n", + "Example n. 5123 = (45.0, 182.0): cluster 2\n", + "Example n. 5124 = (45.0, 181.0): cluster 2\n", + "Example n. 5125 = (45.0, 179.0): cluster 2\n", + "Example n. 5126 = (45.0, 178.0): cluster 2\n", + "Example n. 5127 = (45.0, 177.0): cluster 2\n", + "Example n. 5128 = (45.0, 176.0): cluster 2\n", + "Example n. 5129 = (45.0, 175.0): cluster 2\n", + "Example n. 5130 = (45.0, 174.0): cluster 2\n", + "Example n. 5131 = (45.0, 173.0): cluster 2\n", + "Example n. 5132 = (45.0, 172.0): cluster 2\n", + "Example n. 5133 = (45.0, 171.0): cluster 2\n", + "Example n. 5134 = (45.0, 170.0): cluster 2\n", + "Example n. 5135 = (45.0, 169.0): cluster 2\n", + "Example n. 5136 = (45.0, 168.0): cluster 2\n", + "Example n. 5137 = (45.0, 167.0): cluster 2\n", + "Example n. 5138 = (45.0, 166.0): cluster 2\n", + "Example n. 5139 = (45.0, 165.0): cluster 2\n", + "Example n. 5140 = (45.0, 161.0): cluster 2\n", + "Example n. 5141 = (45.0, 160.0): cluster 2\n", + "Example n. 5142 = (45.0, 159.0): cluster 2\n", + "Example n. 5143 = (45.0, 158.0): cluster 2\n", + "Example n. 5144 = (45.0, 157.0): cluster 2\n", + "Example n. 5145 = (45.0, 156.0): cluster 2\n", + "Example n. 5146 = (45.0, 155.0): cluster 2\n", + "Example n. 5147 = (45.0, 154.0): cluster 2\n", + "Example n. 5148 = (45.0, 153.0): cluster 2\n", + "Example n. 5149 = (45.0, 152.0): cluster 2\n", + "Example n. 5150 = (45.0, 151.0): cluster 2\n", + "Example n. 5151 = (45.0, 150.0): cluster 2\n", + "Example n. 5152 = (45.0, 149.0): cluster 2\n", + "Example n. 5153 = (45.0, 136.0): cluster 2\n", + "Example n. 5154 = (45.0, 135.0): cluster 2\n", + "Example n. 5155 = (45.0, 134.0): cluster 2\n", + "Example n. 5156 = (45.0, 131.0): cluster 2\n", + "Example n. 5157 = (45.0, 130.0): cluster 2\n", + "Example n. 5158 = (45.0, 129.0): cluster 2\n", + "Example n. 5159 = (45.0, 128.0): cluster 2\n", + "Example n. 5160 = (45.0, 127.0): cluster 2\n", + "Example n. 5161 = (45.0, 126.0): cluster 2\n", + "Example n. 5162 = (45.0, 120.0): cluster 2\n", + "Example n. 5163 = (45.0, 119.0): cluster 2\n", + "Example n. 5164 = (45.0, 118.0): cluster 2\n", + "Example n. 5165 = (45.0, 117.0): cluster 2\n", + "Example n. 5166 = (45.0, 116.0): cluster 2\n", + "Example n. 5167 = (46.0, 496.0): cluster 2\n", + "Example n. 5168 = (46.0, 495.0): cluster 2\n", + "Example n. 5169 = (46.0, 494.0): cluster 2\n", + "Example n. 5170 = (46.0, 493.0): cluster 2\n", + "Example n. 5171 = (46.0, 492.0): cluster 2\n", + "Example n. 5172 = (46.0, 491.0): cluster 2\n", + "Example n. 5173 = (46.0, 490.0): cluster 2\n", + "Example n. 5174 = (46.0, 489.0): cluster 2\n", + "Example n. 5175 = (46.0, 488.0): cluster 2\n", + "Example n. 5176 = (46.0, 487.0): cluster 2\n", + "Example n. 5177 = (46.0, 486.0): cluster 2\n", + "Example n. 5178 = (46.0, 485.0): cluster 2\n", + "Example n. 5179 = (46.0, 484.0): cluster 2\n", + "Example n. 5180 = (46.0, 483.0): cluster 2\n", + "Example n. 5181 = (46.0, 482.0): cluster 2\n", + "Example n. 5182 = (46.0, 481.0): cluster 2\n", + "Example n. 5183 = (46.0, 480.0): cluster 2\n", + "Example n. 5184 = (46.0, 479.0): cluster 2\n", + "Example n. 5185 = (46.0, 478.0): cluster 2\n", + "Example n. 5186 = (46.0, 477.0): cluster 2\n", + "Example n. 5187 = (46.0, 476.0): cluster 2\n", + "Example n. 5188 = (46.0, 475.0): cluster 2\n", + "Example n. 5189 = (46.0, 474.0): cluster 2\n", + "Example n. 5190 = (46.0, 473.0): cluster 2\n", + "Example n. 5191 = (46.0, 472.0): cluster 2\n", + "Example n. 5192 = (46.0, 471.0): cluster 2\n", + "Example n. 5193 = (46.0, 470.0): cluster 2\n", + "Example n. 5194 = (46.0, 469.0): cluster 2\n", + "Example n. 5195 = (46.0, 468.0): cluster 2\n", + "Example n. 5196 = (46.0, 467.0): cluster 2\n", + "Example n. 5197 = (46.0, 466.0): cluster 2\n", + "Example n. 5198 = (46.0, 465.0): cluster 2\n", + "Example n. 5199 = (46.0, 464.0): cluster 2\n", + "Example n. 5200 = (46.0, 463.0): cluster 2\n", + "Example n. 5201 = (46.0, 462.0): cluster 2\n", + "Example n. 5202 = (46.0, 461.0): cluster 2\n", + "Example n. 5203 = (46.0, 460.0): cluster 2\n", + "Example n. 5204 = (46.0, 459.0): cluster 2\n", + "Example n. 5205 = (46.0, 458.0): cluster 2\n", + "Example n. 5206 = (46.0, 457.0): cluster 2\n", + "Example n. 5207 = (46.0, 456.0): cluster 2\n", + "Example n. 5208 = (46.0, 455.0): cluster 2\n", + "Example n. 5209 = (46.0, 454.0): cluster 2\n", + "Example n. 5210 = (46.0, 453.0): cluster 2\n", + "Example n. 5211 = (46.0, 452.0): cluster 2\n", + "Example n. 5212 = (46.0, 341.0): cluster 2\n", + "Example n. 5213 = (46.0, 340.0): cluster 2\n", + "Example n. 5214 = (46.0, 339.0): cluster 2\n", + "Example n. 5215 = (46.0, 338.0): cluster 2\n", + "Example n. 5216 = (46.0, 337.0): cluster 2\n", + "Example n. 5217 = (46.0, 336.0): cluster 2\n", + "Example n. 5218 = (46.0, 327.0): cluster 2\n", + "Example n. 5219 = (46.0, 326.0): cluster 2\n", + "Example n. 5220 = (46.0, 325.0): cluster 2\n", + "Example n. 5221 = (46.0, 324.0): cluster 2\n", + "Example n. 5222 = (46.0, 323.0): cluster 2\n", + "Example n. 5223 = (46.0, 322.0): cluster 2\n", + "Example n. 5224 = (46.0, 321.0): cluster 2\n", + "Example n. 5225 = (46.0, 320.0): cluster 2\n", + "Example n. 5226 = (46.0, 319.0): cluster 2\n", + "Example n. 5227 = (46.0, 318.0): cluster 2\n", + "Example n. 5228 = (46.0, 317.0): cluster 2\n", + "Example n. 5229 = (46.0, 316.0): cluster 2\n", + "Example n. 5230 = (46.0, 315.0): cluster 2\n", + "Example n. 5231 = (46.0, 314.0): cluster 2\n", + "Example n. 5232 = (46.0, 313.0): cluster 2\n", + "Example n. 5233 = (46.0, 312.0): cluster 2\n", + "Example n. 5234 = (46.0, 311.0): cluster 2\n", + "Example n. 5235 = (46.0, 310.0): cluster 2\n", + "Example n. 5236 = (46.0, 309.0): cluster 2\n", + "Example n. 5237 = (46.0, 308.0): cluster 2\n", + "Example n. 5238 = (46.0, 307.0): cluster 2\n", + "Example n. 5239 = (46.0, 306.0): cluster 2\n", + "Example n. 5240 = (46.0, 305.0): cluster 2\n", + "Example n. 5241 = (46.0, 304.0): cluster 2\n", + "Example n. 5242 = (46.0, 303.0): cluster 2\n", + "Example n. 5243 = (46.0, 302.0): cluster 2\n", + "Example n. 5244 = (46.0, 301.0): cluster 2\n", + "Example n. 5245 = (46.0, 300.0): cluster 2\n", + "Example n. 5246 = (46.0, 299.0): cluster 2\n", + "Example n. 5247 = (46.0, 298.0): cluster 2\n", + "Example n. 5248 = (46.0, 297.0): cluster 2\n", + "Example n. 5249 = (46.0, 296.0): cluster 2\n", + "Example n. 5250 = (46.0, 295.0): cluster 2\n", + "Example n. 5251 = (46.0, 294.0): cluster 2\n", + "Example n. 5252 = (46.0, 288.0): cluster 2\n", + "Example n. 5253 = (46.0, 287.0): cluster 2\n", + "Example n. 5254 = (46.0, 286.0): cluster 2\n", + "Example n. 5255 = (46.0, 285.0): cluster 2\n", + "Example n. 5256 = (46.0, 284.0): cluster 2\n", + "Example n. 5257 = (46.0, 283.0): cluster 2\n", + "Example n. 5258 = (46.0, 282.0): cluster 2\n", + "Example n. 5259 = (46.0, 281.0): cluster 2\n", + "Example n. 5260 = (46.0, 280.0): cluster 2\n", + "Example n. 5261 = (46.0, 279.0): cluster 2\n", + "Example n. 5262 = (46.0, 278.0): cluster 2\n", + "Example n. 5263 = (46.0, 277.0): cluster 2\n", + "Example n. 5264 = (46.0, 276.0): cluster 2\n", + "Example n. 5265 = (46.0, 275.0): cluster 2\n", + "Example n. 5266 = (46.0, 274.0): cluster 2\n", + "Example n. 5267 = (46.0, 273.0): cluster 2\n", + "Example n. 5268 = (46.0, 272.0): cluster 2\n", + "Example n. 5269 = (46.0, 271.0): cluster 2\n", + "Example n. 5270 = (46.0, 261.0): cluster 2\n", + "Example n. 5271 = (46.0, 260.0): cluster 2\n", + "Example n. 5272 = (46.0, 259.0): cluster 2\n", + "Example n. 5273 = (46.0, 258.0): cluster 2\n", + "Example n. 5274 = (46.0, 257.0): cluster 2\n", + "Example n. 5275 = (46.0, 256.0): cluster 2\n", + "Example n. 5276 = (46.0, 255.0): cluster 2\n", + "Example n. 5277 = (46.0, 254.0): cluster 2\n", + "Example n. 5278 = (46.0, 253.0): cluster 2\n", + "Example n. 5279 = (46.0, 252.0): cluster 2\n", + "Example n. 5280 = (46.0, 251.0): cluster 2\n", + "Example n. 5281 = (46.0, 250.0): cluster 2\n", + "Example n. 5282 = (46.0, 249.0): cluster 2\n", + "Example n. 5283 = (46.0, 248.0): cluster 2\n", + "Example n. 5284 = (46.0, 245.0): cluster 2\n", + "Example n. 5285 = (46.0, 244.0): cluster 2\n", + "Example n. 5286 = (46.0, 243.0): cluster 2\n", + "Example n. 5287 = (46.0, 242.0): cluster 2\n", + "Example n. 5288 = (46.0, 241.0): cluster 2\n", + "Example n. 5289 = (46.0, 240.0): cluster 2\n", + "Example n. 5290 = (46.0, 239.0): cluster 2\n", + "Example n. 5291 = (46.0, 238.0): cluster 2\n", + "Example n. 5292 = (46.0, 237.0): cluster 2\n", + "Example n. 5293 = (46.0, 236.0): cluster 2\n", + "Example n. 5294 = (46.0, 235.0): cluster 2\n", + "Example n. 5295 = (46.0, 228.0): cluster 2\n", + "Example n. 5296 = (46.0, 227.0): cluster 2\n", + "Example n. 5297 = (46.0, 226.0): cluster 2\n", + "Example n. 5298 = (46.0, 225.0): cluster 2\n", + "Example n. 5299 = (46.0, 224.0): cluster 2\n", + "Example n. 5300 = (46.0, 223.0): cluster 2\n", + "Example n. 5301 = (46.0, 220.0): cluster 2\n", + "Example n. 5302 = (46.0, 219.0): cluster 2\n", + "Example n. 5303 = (46.0, 218.0): cluster 2\n", + "Example n. 5304 = (46.0, 217.0): cluster 2\n", + "Example n. 5305 = (46.0, 216.0): cluster 2\n", + "Example n. 5306 = (46.0, 208.0): cluster 2\n", + "Example n. 5307 = (46.0, 207.0): cluster 2\n", + "Example n. 5308 = (46.0, 206.0): cluster 2\n", + "Example n. 5309 = (46.0, 202.0): cluster 2\n", + "Example n. 5310 = (46.0, 201.0): cluster 2\n", + "Example n. 5311 = (46.0, 200.0): cluster 2\n", + "Example n. 5312 = (46.0, 199.0): cluster 2\n", + "Example n. 5313 = (46.0, 198.0): cluster 2\n", + "Example n. 5314 = (46.0, 188.0): cluster 2\n", + "Example n. 5315 = (46.0, 187.0): cluster 2\n", + "Example n. 5316 = (46.0, 186.0): cluster 2\n", + "Example n. 5317 = (46.0, 185.0): cluster 2\n", + "Example n. 5318 = (46.0, 184.0): cluster 2\n", + "Example n. 5319 = (46.0, 183.0): cluster 2\n", + "Example n. 5320 = (46.0, 182.0): cluster 2\n", + "Example n. 5321 = (46.0, 181.0): cluster 2\n", + "Example n. 5322 = (46.0, 179.0): cluster 2\n", + "Example n. 5323 = (46.0, 178.0): cluster 2\n", + "Example n. 5324 = (46.0, 177.0): cluster 2\n", + "Example n. 5325 = (46.0, 176.0): cluster 2\n", + "Example n. 5326 = (46.0, 175.0): cluster 2\n", + "Example n. 5327 = (46.0, 174.0): cluster 2\n", + "Example n. 5328 = (46.0, 173.0): cluster 2\n", + "Example n. 5329 = (46.0, 172.0): cluster 2\n", + "Example n. 5330 = (46.0, 171.0): cluster 2\n", + "Example n. 5331 = (46.0, 170.0): cluster 2\n", + "Example n. 5332 = (46.0, 169.0): cluster 2\n", + "Example n. 5333 = (46.0, 168.0): cluster 2\n", + "Example n. 5334 = (46.0, 167.0): cluster 2\n", + "Example n. 5335 = (46.0, 166.0): cluster 2\n", + "Example n. 5336 = (46.0, 165.0): cluster 2\n", + "Example n. 5337 = (46.0, 161.0): cluster 2\n", + "Example n. 5338 = (46.0, 160.0): cluster 2\n", + "Example n. 5339 = (46.0, 159.0): cluster 2\n", + "Example n. 5340 = (46.0, 158.0): cluster 2\n", + "Example n. 5341 = (46.0, 157.0): cluster 2\n", + "Example n. 5342 = (46.0, 156.0): cluster 2\n", + "Example n. 5343 = (46.0, 155.0): cluster 2\n", + "Example n. 5344 = (46.0, 151.0): cluster 2\n", + "Example n. 5345 = (46.0, 150.0): cluster 2\n", + "Example n. 5346 = (46.0, 148.0): cluster 2\n", + "Example n. 5347 = (46.0, 147.0): cluster 2\n", + "Example n. 5348 = (46.0, 146.0): cluster 2\n", + "Example n. 5349 = (46.0, 145.0): cluster 2\n", + "Example n. 5350 = (46.0, 140.0): cluster 2\n", + "Example n. 5351 = (46.0, 139.0): cluster 2\n", + "Example n. 5352 = (46.0, 138.0): cluster 2\n", + "Example n. 5353 = (46.0, 137.0): cluster 2\n", + "Example n. 5354 = (46.0, 136.0): cluster 2\n", + "Example n. 5355 = (46.0, 135.0): cluster 2\n", + "Example n. 5356 = (46.0, 134.0): cluster 2\n", + "Example n. 5357 = (46.0, 133.0): cluster 2\n", + "Example n. 5358 = (46.0, 132.0): cluster 2\n", + "Example n. 5359 = (46.0, 131.0): cluster 2\n", + "Example n. 5360 = (46.0, 130.0): cluster 2\n", + "Example n. 5361 = (46.0, 129.0): cluster 2\n", + "Example n. 5362 = (46.0, 128.0): cluster 2\n", + "Example n. 5363 = (46.0, 127.0): cluster 2\n", + "Example n. 5364 = (46.0, 120.0): cluster 2\n", + "Example n. 5365 = (46.0, 119.0): cluster 2\n", + "Example n. 5366 = (46.0, 118.0): cluster 2\n", + "Example n. 5367 = (46.0, 117.0): cluster 2\n", + "Example n. 5368 = (46.0, 116.0): cluster 2\n", + "Example n. 5369 = (47.0, 495.0): cluster 2\n", + "Example n. 5370 = (47.0, 494.0): cluster 2\n", + "Example n. 5371 = (47.0, 493.0): cluster 2\n", + "Example n. 5372 = (47.0, 492.0): cluster 2\n", + "Example n. 5373 = (47.0, 491.0): cluster 2\n", + "Example n. 5374 = (47.0, 490.0): cluster 2\n", + "Example n. 5375 = (47.0, 489.0): cluster 2\n", + "Example n. 5376 = (47.0, 488.0): cluster 2\n", + "Example n. 5377 = (47.0, 487.0): cluster 2\n", + "Example n. 5378 = (47.0, 486.0): cluster 2\n", + "Example n. 5379 = (47.0, 485.0): cluster 2\n", + "Example n. 5380 = (47.0, 484.0): cluster 2\n", + "Example n. 5381 = (47.0, 483.0): cluster 2\n", + "Example n. 5382 = (47.0, 482.0): cluster 2\n", + "Example n. 5383 = (47.0, 481.0): cluster 2\n", + "Example n. 5384 = (47.0, 480.0): cluster 2\n", + "Example n. 5385 = (47.0, 479.0): cluster 2\n", + "Example n. 5386 = (47.0, 478.0): cluster 2\n", + "Example n. 5387 = (47.0, 477.0): cluster 2\n", + "Example n. 5388 = (47.0, 476.0): cluster 2\n", + "Example n. 5389 = (47.0, 475.0): cluster 2\n", + "Example n. 5390 = (47.0, 474.0): cluster 2\n", + "Example n. 5391 = (47.0, 473.0): cluster 2\n", + "Example n. 5392 = (47.0, 472.0): cluster 2\n", + "Example n. 5393 = (47.0, 471.0): cluster 2\n", + "Example n. 5394 = (47.0, 470.0): cluster 2\n", + "Example n. 5395 = (47.0, 469.0): cluster 2\n", + "Example n. 5396 = (47.0, 468.0): cluster 2\n", + "Example n. 5397 = (47.0, 467.0): cluster 2\n", + "Example n. 5398 = (47.0, 466.0): cluster 2\n", + "Example n. 5399 = (47.0, 465.0): cluster 2\n", + "Example n. 5400 = (47.0, 464.0): cluster 2\n", + "Example n. 5401 = (47.0, 463.0): cluster 2\n", + "Example n. 5402 = (47.0, 462.0): cluster 2\n", + "Example n. 5403 = (47.0, 461.0): cluster 2\n", + "Example n. 5404 = (47.0, 460.0): cluster 2\n", + "Example n. 5405 = (47.0, 459.0): cluster 2\n", + "Example n. 5406 = (47.0, 458.0): cluster 2\n", + "Example n. 5407 = (47.0, 457.0): cluster 2\n", + "Example n. 5408 = (47.0, 456.0): cluster 2\n", + "Example n. 5409 = (47.0, 455.0): cluster 2\n", + "Example n. 5410 = (47.0, 454.0): cluster 2\n", + "Example n. 5411 = (47.0, 453.0): cluster 2\n", + "Example n. 5412 = (47.0, 452.0): cluster 2\n", + "Example n. 5413 = (47.0, 341.0): cluster 2\n", + "Example n. 5414 = (47.0, 340.0): cluster 2\n", + "Example n. 5415 = (47.0, 339.0): cluster 2\n", + "Example n. 5416 = (47.0, 338.0): cluster 2\n", + "Example n. 5417 = (47.0, 337.0): cluster 2\n", + "Example n. 5418 = (47.0, 327.0): cluster 2\n", + "Example n. 5419 = (47.0, 326.0): cluster 2\n", + "Example n. 5420 = (47.0, 325.0): cluster 2\n", + "Example n. 5421 = (47.0, 324.0): cluster 2\n", + "Example n. 5422 = (47.0, 323.0): cluster 2\n", + "Example n. 5423 = (47.0, 322.0): cluster 2\n", + "Example n. 5424 = (47.0, 321.0): cluster 2\n", + "Example n. 5425 = (47.0, 320.0): cluster 2\n", + "Example n. 5426 = (47.0, 319.0): cluster 2\n", + "Example n. 5427 = (47.0, 318.0): cluster 2\n", + "Example n. 5428 = (47.0, 317.0): cluster 2\n", + "Example n. 5429 = (47.0, 316.0): cluster 2\n", + "Example n. 5430 = (47.0, 315.0): cluster 2\n", + "Example n. 5431 = (47.0, 314.0): cluster 2\n", + "Example n. 5432 = (47.0, 313.0): cluster 2\n", + "Example n. 5433 = (47.0, 312.0): cluster 2\n", + "Example n. 5434 = (47.0, 311.0): cluster 2\n", + "Example n. 5435 = (47.0, 310.0): cluster 2\n", + "Example n. 5436 = (47.0, 308.0): cluster 2\n", + "Example n. 5437 = (47.0, 307.0): cluster 2\n", + "Example n. 5438 = (47.0, 306.0): cluster 2\n", + "Example n. 5439 = (47.0, 305.0): cluster 2\n", + "Example n. 5440 = (47.0, 304.0): cluster 2\n", + "Example n. 5441 = (47.0, 303.0): cluster 2\n", + "Example n. 5442 = (47.0, 302.0): cluster 2\n", + "Example n. 5443 = (47.0, 301.0): cluster 2\n", + "Example n. 5444 = (47.0, 300.0): cluster 2\n", + "Example n. 5445 = (47.0, 299.0): cluster 2\n", + "Example n. 5446 = (47.0, 298.0): cluster 2\n", + "Example n. 5447 = (47.0, 297.0): cluster 2\n", + "Example n. 5448 = (47.0, 296.0): cluster 2\n", + "Example n. 5449 = (47.0, 295.0): cluster 2\n", + "Example n. 5450 = (47.0, 294.0): cluster 2\n", + "Example n. 5451 = (47.0, 287.0): cluster 2\n", + "Example n. 5452 = (47.0, 286.0): cluster 2\n", + "Example n. 5453 = (47.0, 285.0): cluster 2\n", + "Example n. 5454 = (47.0, 284.0): cluster 2\n", + "Example n. 5455 = (47.0, 283.0): cluster 2\n", + "Example n. 5456 = (47.0, 282.0): cluster 2\n", + "Example n. 5457 = (47.0, 281.0): cluster 2\n", + "Example n. 5458 = (47.0, 280.0): cluster 2\n", + "Example n. 5459 = (47.0, 279.0): cluster 2\n", + "Example n. 5460 = (47.0, 278.0): cluster 2\n", + "Example n. 5461 = (47.0, 277.0): cluster 2\n", + "Example n. 5462 = (47.0, 276.0): cluster 2\n", + "Example n. 5463 = (47.0, 275.0): cluster 2\n", + "Example n. 5464 = (47.0, 274.0): cluster 2\n", + "Example n. 5465 = (47.0, 273.0): cluster 2\n", + "Example n. 5466 = (47.0, 272.0): cluster 2\n", + "Example n. 5467 = (47.0, 271.0): cluster 2\n", + "Example n. 5468 = (47.0, 264.0): cluster 2\n", + "Example n. 5469 = (47.0, 263.0): cluster 2\n", + "Example n. 5470 = (47.0, 262.0): cluster 2\n", + "Example n. 5471 = (47.0, 261.0): cluster 2\n", + "Example n. 5472 = (47.0, 260.0): cluster 2\n", + "Example n. 5473 = (47.0, 259.0): cluster 2\n", + "Example n. 5474 = (47.0, 258.0): cluster 2\n", + "Example n. 5475 = (47.0, 257.0): cluster 2\n", + "Example n. 5476 = (47.0, 256.0): cluster 2\n", + "Example n. 5477 = (47.0, 255.0): cluster 2\n", + "Example n. 5478 = (47.0, 254.0): cluster 2\n", + "Example n. 5479 = (47.0, 253.0): cluster 2\n", + "Example n. 5480 = (47.0, 252.0): cluster 2\n", + "Example n. 5481 = (47.0, 251.0): cluster 2\n", + "Example n. 5482 = (47.0, 250.0): cluster 2\n", + "Example n. 5483 = (47.0, 249.0): cluster 2\n", + "Example n. 5484 = (47.0, 248.0): cluster 2\n", + "Example n. 5485 = (47.0, 247.0): cluster 2\n", + "Example n. 5486 = (47.0, 246.0): cluster 2\n", + "Example n. 5487 = (47.0, 245.0): cluster 2\n", + "Example n. 5488 = (47.0, 244.0): cluster 2\n", + "Example n. 5489 = (47.0, 243.0): cluster 2\n", + "Example n. 5490 = (47.0, 242.0): cluster 2\n", + "Example n. 5491 = (47.0, 241.0): cluster 2\n", + "Example n. 5492 = (47.0, 240.0): cluster 2\n", + "Example n. 5493 = (47.0, 239.0): cluster 2\n", + "Example n. 5494 = (47.0, 238.0): cluster 2\n", + "Example n. 5495 = (47.0, 237.0): cluster 2\n", + "Example n. 5496 = (47.0, 236.0): cluster 2\n", + "Example n. 5497 = (47.0, 227.0): cluster 2\n", + "Example n. 5498 = (47.0, 226.0): cluster 2\n", + "Example n. 5499 = (47.0, 225.0): cluster 2\n", + "Example n. 5500 = (47.0, 224.0): cluster 2\n", + "Example n. 5501 = (47.0, 223.0): cluster 2\n", + "Example n. 5502 = (47.0, 220.0): cluster 2\n", + "Example n. 5503 = (47.0, 219.0): cluster 2\n", + "Example n. 5504 = (47.0, 218.0): cluster 2\n", + "Example n. 5505 = (47.0, 217.0): cluster 2\n", + "Example n. 5506 = (47.0, 216.0): cluster 2\n", + "Example n. 5507 = (47.0, 215.0): cluster 2\n", + "Example n. 5508 = (47.0, 214.0): cluster 2\n", + "Example n. 5509 = (47.0, 213.0): cluster 2\n", + "Example n. 5510 = (47.0, 212.0): cluster 2\n", + "Example n. 5511 = (47.0, 210.0): cluster 2\n", + "Example n. 5512 = (47.0, 209.0): cluster 2\n", + "Example n. 5513 = (47.0, 201.0): cluster 2\n", + "Example n. 5514 = (47.0, 200.0): cluster 2\n", + "Example n. 5515 = (47.0, 199.0): cluster 2\n", + "Example n. 5516 = (47.0, 194.0): cluster 2\n", + "Example n. 5517 = (47.0, 193.0): cluster 2\n", + "Example n. 5518 = (47.0, 192.0): cluster 2\n", + "Example n. 5519 = (47.0, 188.0): cluster 2\n", + "Example n. 5520 = (47.0, 187.0): cluster 2\n", + "Example n. 5521 = (47.0, 186.0): cluster 2\n", + "Example n. 5522 = (47.0, 185.0): cluster 2\n", + "Example n. 5523 = (47.0, 184.0): cluster 2\n", + "Example n. 5524 = (47.0, 183.0): cluster 2\n", + "Example n. 5525 = (47.0, 182.0): cluster 2\n", + "Example n. 5526 = (47.0, 181.0): cluster 2\n", + "Example n. 5527 = (47.0, 179.0): cluster 2\n", + "Example n. 5528 = (47.0, 178.0): cluster 2\n", + "Example n. 5529 = (47.0, 177.0): cluster 2\n", + "Example n. 5530 = (47.0, 176.0): cluster 2\n", + "Example n. 5531 = (47.0, 175.0): cluster 2\n", + "Example n. 5532 = (47.0, 174.0): cluster 2\n", + "Example n. 5533 = (47.0, 173.0): cluster 2\n", + "Example n. 5534 = (47.0, 172.0): cluster 2\n", + "Example n. 5535 = (47.0, 171.0): cluster 2\n", + "Example n. 5536 = (47.0, 170.0): cluster 2\n", + "Example n. 5537 = (47.0, 169.0): cluster 2\n", + "Example n. 5538 = (47.0, 168.0): cluster 2\n", + "Example n. 5539 = (47.0, 167.0): cluster 2\n", + "Example n. 5540 = (47.0, 166.0): cluster 2\n", + "Example n. 5541 = (47.0, 165.0): cluster 2\n", + "Example n. 5542 = (47.0, 164.0): cluster 2\n", + "Example n. 5543 = (47.0, 163.0): cluster 2\n", + "Example n. 5544 = (47.0, 161.0): cluster 2\n", + "Example n. 5545 = (47.0, 160.0): cluster 2\n", + "Example n. 5546 = (47.0, 159.0): cluster 2\n", + "Example n. 5547 = (47.0, 158.0): cluster 2\n", + "Example n. 5548 = (47.0, 157.0): cluster 2\n", + "Example n. 5549 = (47.0, 156.0): cluster 2\n", + "Example n. 5550 = (47.0, 155.0): cluster 2\n", + "Example n. 5551 = (47.0, 149.0): cluster 2\n", + "Example n. 5552 = (47.0, 148.0): cluster 2\n", + "Example n. 5553 = (47.0, 147.0): cluster 2\n", + "Example n. 5554 = (47.0, 146.0): cluster 2\n", + "Example n. 5555 = (47.0, 145.0): cluster 2\n", + "Example n. 5556 = (47.0, 144.0): cluster 2\n", + "Example n. 5557 = (47.0, 143.0): cluster 2\n", + "Example n. 5558 = (47.0, 142.0): cluster 2\n", + "Example n. 5559 = (47.0, 141.0): cluster 2\n", + "Example n. 5560 = (47.0, 140.0): cluster 2\n", + "Example n. 5561 = (47.0, 139.0): cluster 2\n", + "Example n. 5562 = (47.0, 138.0): cluster 2\n", + "Example n. 5563 = (47.0, 137.0): cluster 2\n", + "Example n. 5564 = (47.0, 136.0): cluster 2\n", + "Example n. 5565 = (47.0, 135.0): cluster 2\n", + "Example n. 5566 = (47.0, 134.0): cluster 2\n", + "Example n. 5567 = (47.0, 133.0): cluster 2\n", + "Example n. 5568 = (47.0, 132.0): cluster 2\n", + "Example n. 5569 = (47.0, 131.0): cluster 2\n", + "Example n. 5570 = (47.0, 130.0): cluster 2\n", + "Example n. 5571 = (47.0, 129.0): cluster 2\n", + "Example n. 5572 = (47.0, 128.0): cluster 2\n", + "Example n. 5573 = (47.0, 127.0): cluster 2\n", + "Example n. 5574 = (47.0, 119.0): cluster 2\n", + "Example n. 5575 = (47.0, 118.0): cluster 2\n", + "Example n. 5576 = (47.0, 117.0): cluster 2\n", + "Example n. 5577 = (47.0, 116.0): cluster 2\n", + "Example n. 5578 = (48.0, 494.0): cluster 2\n", + "Example n. 5579 = (48.0, 493.0): cluster 2\n", + "Example n. 5580 = (48.0, 492.0): cluster 2\n", + "Example n. 5581 = (48.0, 491.0): cluster 2\n", + "Example n. 5582 = (48.0, 490.0): cluster 2\n", + "Example n. 5583 = (48.0, 489.0): cluster 2\n", + "Example n. 5584 = (48.0, 488.0): cluster 2\n", + "Example n. 5585 = (48.0, 487.0): cluster 2\n", + "Example n. 5586 = (48.0, 486.0): cluster 2\n", + "Example n. 5587 = (48.0, 485.0): cluster 2\n", + "Example n. 5588 = (48.0, 484.0): cluster 2\n", + "Example n. 5589 = (48.0, 483.0): cluster 2\n", + "Example n. 5590 = (48.0, 482.0): cluster 2\n", + "Example n. 5591 = (48.0, 481.0): cluster 2\n", + "Example n. 5592 = (48.0, 480.0): cluster 2\n", + "Example n. 5593 = (48.0, 479.0): cluster 2\n", + "Example n. 5594 = (48.0, 478.0): cluster 2\n", + "Example n. 5595 = (48.0, 477.0): cluster 2\n", + "Example n. 5596 = (48.0, 476.0): cluster 2\n", + "Example n. 5597 = (48.0, 475.0): cluster 2\n", + "Example n. 5598 = (48.0, 474.0): cluster 2\n", + "Example n. 5599 = (48.0, 473.0): cluster 2\n", + "Example n. 5600 = (48.0, 472.0): cluster 2\n", + "Example n. 5601 = (48.0, 471.0): cluster 2\n", + "Example n. 5602 = (48.0, 470.0): cluster 2\n", + "Example n. 5603 = (48.0, 469.0): cluster 2\n", + "Example n. 5604 = (48.0, 468.0): cluster 2\n", + "Example n. 5605 = (48.0, 467.0): cluster 2\n", + "Example n. 5606 = (48.0, 466.0): cluster 2\n", + "Example n. 5607 = (48.0, 465.0): cluster 2\n", + "Example n. 5608 = (48.0, 464.0): cluster 2\n", + "Example n. 5609 = (48.0, 463.0): cluster 2\n", + "Example n. 5610 = (48.0, 462.0): cluster 2\n", + "Example n. 5611 = (48.0, 461.0): cluster 2\n", + "Example n. 5612 = (48.0, 460.0): cluster 2\n", + "Example n. 5613 = (48.0, 459.0): cluster 2\n", + "Example n. 5614 = (48.0, 458.0): cluster 2\n", + "Example n. 5615 = (48.0, 457.0): cluster 2\n", + "Example n. 5616 = (48.0, 456.0): cluster 2\n", + "Example n. 5617 = (48.0, 455.0): cluster 2\n", + "Example n. 5618 = (48.0, 454.0): cluster 2\n", + "Example n. 5619 = (48.0, 453.0): cluster 2\n", + "Example n. 5620 = (48.0, 452.0): cluster 2\n", + "Example n. 5621 = (48.0, 339.0): cluster 2\n", + "Example n. 5622 = (48.0, 338.0): cluster 2\n", + "Example n. 5623 = (48.0, 327.0): cluster 2\n", + "Example n. 5624 = (48.0, 326.0): cluster 2\n", + "Example n. 5625 = (48.0, 325.0): cluster 2\n", + "Example n. 5626 = (48.0, 324.0): cluster 2\n", + "Example n. 5627 = (48.0, 323.0): cluster 2\n", + "Example n. 5628 = (48.0, 322.0): cluster 2\n", + "Example n. 5629 = (48.0, 321.0): cluster 2\n", + "Example n. 5630 = (48.0, 320.0): cluster 2\n", + "Example n. 5631 = (48.0, 319.0): cluster 2\n", + "Example n. 5632 = (48.0, 318.0): cluster 2\n", + "Example n. 5633 = (48.0, 316.0): cluster 2\n", + "Example n. 5634 = (48.0, 315.0): cluster 2\n", + "Example n. 5635 = (48.0, 314.0): cluster 2\n", + "Example n. 5636 = (48.0, 313.0): cluster 2\n", + "Example n. 5637 = (48.0, 308.0): cluster 2\n", + "Example n. 5638 = (48.0, 307.0): cluster 2\n", + "Example n. 5639 = (48.0, 306.0): cluster 2\n", + "Example n. 5640 = (48.0, 305.0): cluster 2\n", + "Example n. 5641 = (48.0, 304.0): cluster 2\n", + "Example n. 5642 = (48.0, 303.0): cluster 2\n", + "Example n. 5643 = (48.0, 302.0): cluster 2\n", + "Example n. 5644 = (48.0, 301.0): cluster 2\n", + "Example n. 5645 = (48.0, 300.0): cluster 2\n", + "Example n. 5646 = (48.0, 299.0): cluster 2\n", + "Example n. 5647 = (48.0, 298.0): cluster 2\n", + "Example n. 5648 = (48.0, 297.0): cluster 2\n", + "Example n. 5649 = (48.0, 296.0): cluster 2\n", + "Example n. 5650 = (48.0, 295.0): cluster 2\n", + "Example n. 5651 = (48.0, 294.0): cluster 2\n", + "Example n. 5652 = (48.0, 287.0): cluster 2\n", + "Example n. 5653 = (48.0, 286.0): cluster 2\n", + "Example n. 5654 = (48.0, 285.0): cluster 2\n", + "Example n. 5655 = (48.0, 284.0): cluster 2\n", + "Example n. 5656 = (48.0, 283.0): cluster 2\n", + "Example n. 5657 = (48.0, 282.0): cluster 2\n", + "Example n. 5658 = (48.0, 281.0): cluster 2\n", + "Example n. 5659 = (48.0, 280.0): cluster 2\n", + "Example n. 5660 = (48.0, 279.0): cluster 2\n", + "Example n. 5661 = (48.0, 278.0): cluster 2\n", + "Example n. 5662 = (48.0, 277.0): cluster 2\n", + "Example n. 5663 = (48.0, 276.0): cluster 2\n", + "Example n. 5664 = (48.0, 275.0): cluster 2\n", + "Example n. 5665 = (48.0, 274.0): cluster 2\n", + "Example n. 5666 = (48.0, 273.0): cluster 2\n", + "Example n. 5667 = (48.0, 272.0): cluster 2\n", + "Example n. 5668 = (48.0, 271.0): cluster 2\n", + "Example n. 5669 = (48.0, 270.0): cluster 2\n", + "Example n. 5670 = (48.0, 265.0): cluster 2\n", + "Example n. 5671 = (48.0, 264.0): cluster 2\n", + "Example n. 5672 = (48.0, 263.0): cluster 2\n", + "Example n. 5673 = (48.0, 262.0): cluster 2\n", + "Example n. 5674 = (48.0, 261.0): cluster 2\n", + "Example n. 5675 = (48.0, 260.0): cluster 2\n", + "Example n. 5676 = (48.0, 259.0): cluster 2\n", + "Example n. 5677 = (48.0, 258.0): cluster 2\n", + "Example n. 5678 = (48.0, 257.0): cluster 2\n", + "Example n. 5679 = (48.0, 256.0): cluster 2\n", + "Example n. 5680 = (48.0, 255.0): cluster 2\n", + "Example n. 5681 = (48.0, 254.0): cluster 2\n", + "Example n. 5682 = (48.0, 253.0): cluster 2\n", + "Example n. 5683 = (48.0, 252.0): cluster 2\n", + "Example n. 5684 = (48.0, 251.0): cluster 2\n", + "Example n. 5685 = (48.0, 250.0): cluster 2\n", + "Example n. 5686 = (48.0, 249.0): cluster 2\n", + "Example n. 5687 = (48.0, 248.0): cluster 2\n", + "Example n. 5688 = (48.0, 247.0): cluster 2\n", + "Example n. 5689 = (48.0, 246.0): cluster 2\n", + "Example n. 5690 = (48.0, 245.0): cluster 2\n", + "Example n. 5691 = (48.0, 244.0): cluster 2\n", + "Example n. 5692 = (48.0, 243.0): cluster 2\n", + "Example n. 5693 = (48.0, 242.0): cluster 2\n", + "Example n. 5694 = (48.0, 241.0): cluster 2\n", + "Example n. 5695 = (48.0, 240.0): cluster 2\n", + "Example n. 5696 = (48.0, 239.0): cluster 2\n", + "Example n. 5697 = (48.0, 238.0): cluster 2\n", + "Example n. 5698 = (48.0, 237.0): cluster 2\n", + "Example n. 5699 = (48.0, 236.0): cluster 2\n", + "Example n. 5700 = (48.0, 229.0): cluster 2\n", + "Example n. 5701 = (48.0, 228.0): cluster 2\n", + "Example n. 5702 = (48.0, 227.0): cluster 2\n", + "Example n. 5703 = (48.0, 226.0): cluster 2\n", + "Example n. 5704 = (48.0, 225.0): cluster 2\n", + "Example n. 5705 = (48.0, 224.0): cluster 2\n", + "Example n. 5706 = (48.0, 223.0): cluster 2\n", + "Example n. 5707 = (48.0, 219.0): cluster 2\n", + "Example n. 5708 = (48.0, 218.0): cluster 2\n", + "Example n. 5709 = (48.0, 217.0): cluster 2\n", + "Example n. 5710 = (48.0, 216.0): cluster 2\n", + "Example n. 5711 = (48.0, 215.0): cluster 2\n", + "Example n. 5712 = (48.0, 214.0): cluster 2\n", + "Example n. 5713 = (48.0, 213.0): cluster 2\n", + "Example n. 5714 = (48.0, 212.0): cluster 2\n", + "Example n. 5715 = (48.0, 211.0): cluster 2\n", + "Example n. 5716 = (48.0, 210.0): cluster 2\n", + "Example n. 5717 = (48.0, 209.0): cluster 2\n", + "Example n. 5718 = (48.0, 208.0): cluster 2\n", + "Example n. 5719 = (48.0, 195.0): cluster 2\n", + "Example n. 5720 = (48.0, 194.0): cluster 2\n", + "Example n. 5721 = (48.0, 193.0): cluster 2\n", + "Example n. 5722 = (48.0, 192.0): cluster 2\n", + "Example n. 5723 = (48.0, 191.0): cluster 2\n", + "Example n. 5724 = (48.0, 188.0): cluster 2\n", + "Example n. 5725 = (48.0, 187.0): cluster 2\n", + "Example n. 5726 = (48.0, 186.0): cluster 2\n", + "Example n. 5727 = (48.0, 185.0): cluster 2\n", + "Example n. 5728 = (48.0, 184.0): cluster 2\n", + "Example n. 5729 = (48.0, 183.0): cluster 2\n", + "Example n. 5730 = (48.0, 182.0): cluster 2\n", + "Example n. 5731 = (48.0, 181.0): cluster 2\n", + "Example n. 5732 = (48.0, 180.0): cluster 2\n", + "Example n. 5733 = (48.0, 179.0): cluster 2\n", + "Example n. 5734 = (48.0, 178.0): cluster 2\n", + "Example n. 5735 = (48.0, 177.0): cluster 2\n", + "Example n. 5736 = (48.0, 176.0): cluster 2\n", + "Example n. 5737 = (48.0, 175.0): cluster 2\n", + "Example n. 5738 = (48.0, 174.0): cluster 2\n", + "Example n. 5739 = (48.0, 173.0): cluster 2\n", + "Example n. 5740 = (48.0, 172.0): cluster 2\n", + "Example n. 5741 = (48.0, 171.0): cluster 2\n", + "Example n. 5742 = (48.0, 170.0): cluster 2\n", + "Example n. 5743 = (48.0, 169.0): cluster 2\n", + "Example n. 5744 = (48.0, 168.0): cluster 2\n", + "Example n. 5745 = (48.0, 167.0): cluster 2\n", + "Example n. 5746 = (48.0, 166.0): cluster 2\n", + "Example n. 5747 = (48.0, 165.0): cluster 2\n", + "Example n. 5748 = (48.0, 164.0): cluster 2\n", + "Example n. 5749 = (48.0, 163.0): cluster 2\n", + "Example n. 5750 = (48.0, 162.0): cluster 2\n", + "Example n. 5751 = (48.0, 161.0): cluster 2\n", + "Example n. 5752 = (48.0, 160.0): cluster 2\n", + "Example n. 5753 = (48.0, 159.0): cluster 2\n", + "Example n. 5754 = (48.0, 158.0): cluster 2\n", + "Example n. 5755 = (48.0, 157.0): cluster 2\n", + "Example n. 5756 = (48.0, 156.0): cluster 2\n", + "Example n. 5757 = (48.0, 149.0): cluster 2\n", + "Example n. 5758 = (48.0, 148.0): cluster 2\n", + "Example n. 5759 = (48.0, 147.0): cluster 2\n", + "Example n. 5760 = (48.0, 146.0): cluster 2\n", + "Example n. 5761 = (48.0, 145.0): cluster 2\n", + "Example n. 5762 = (48.0, 144.0): cluster 2\n", + "Example n. 5763 = (48.0, 143.0): cluster 2\n", + "Example n. 5764 = (48.0, 142.0): cluster 2\n", + "Example n. 5765 = (48.0, 141.0): cluster 2\n", + "Example n. 5766 = (48.0, 140.0): cluster 2\n", + "Example n. 5767 = (48.0, 139.0): cluster 2\n", + "Example n. 5768 = (48.0, 138.0): cluster 2\n", + "Example n. 5769 = (48.0, 137.0): cluster 2\n", + "Example n. 5770 = (48.0, 136.0): cluster 2\n", + "Example n. 5771 = (48.0, 135.0): cluster 2\n", + "Example n. 5772 = (48.0, 134.0): cluster 2\n", + "Example n. 5773 = (48.0, 133.0): cluster 2\n", + "Example n. 5774 = (48.0, 132.0): cluster 2\n", + "Example n. 5775 = (48.0, 131.0): cluster 2\n", + "Example n. 5776 = (48.0, 130.0): cluster 2\n", + "Example n. 5777 = (48.0, 129.0): cluster 2\n", + "Example n. 5778 = (48.0, 128.0): cluster 2\n", + "Example n. 5779 = (48.0, 127.0): cluster 2\n", + "Example n. 5780 = (49.0, 494.0): cluster 2\n", + "Example n. 5781 = (49.0, 493.0): cluster 2\n", + "Example n. 5782 = (49.0, 492.0): cluster 2\n", + "Example n. 5783 = (49.0, 491.0): cluster 2\n", + "Example n. 5784 = (49.0, 490.0): cluster 2\n", + "Example n. 5785 = (49.0, 489.0): cluster 2\n", + "Example n. 5786 = (49.0, 488.0): cluster 2\n", + "Example n. 5787 = (49.0, 487.0): cluster 2\n", + "Example n. 5788 = (49.0, 486.0): cluster 2\n", + "Example n. 5789 = (49.0, 485.0): cluster 2\n", + "Example n. 5790 = (49.0, 484.0): cluster 2\n", + "Example n. 5791 = (49.0, 483.0): cluster 2\n", + "Example n. 5792 = (49.0, 482.0): cluster 2\n", + "Example n. 5793 = (49.0, 481.0): cluster 2\n", + "Example n. 5794 = (49.0, 480.0): cluster 2\n", + "Example n. 5795 = (49.0, 479.0): cluster 2\n", + "Example n. 5796 = (49.0, 478.0): cluster 2\n", + "Example n. 5797 = (49.0, 477.0): cluster 2\n", + "Example n. 5798 = (49.0, 476.0): cluster 2\n", + "Example n. 5799 = (49.0, 475.0): cluster 2\n", + "Example n. 5800 = (49.0, 474.0): cluster 2\n", + "Example n. 5801 = (49.0, 473.0): cluster 2\n", + "Example n. 5802 = (49.0, 472.0): cluster 2\n", + "Example n. 5803 = (49.0, 471.0): cluster 2\n", + "Example n. 5804 = (49.0, 470.0): cluster 2\n", + "Example n. 5805 = (49.0, 469.0): cluster 2\n", + "Example n. 5806 = (49.0, 468.0): cluster 2\n", + "Example n. 5807 = (49.0, 467.0): cluster 2\n", + "Example n. 5808 = (49.0, 466.0): cluster 2\n", + "Example n. 5809 = (49.0, 465.0): cluster 2\n", + "Example n. 5810 = (49.0, 464.0): cluster 2\n", + "Example n. 5811 = (49.0, 463.0): cluster 2\n", + "Example n. 5812 = (49.0, 462.0): cluster 2\n", + "Example n. 5813 = (49.0, 461.0): cluster 2\n", + "Example n. 5814 = (49.0, 460.0): cluster 2\n", + "Example n. 5815 = (49.0, 459.0): cluster 2\n", + "Example n. 5816 = (49.0, 458.0): cluster 2\n", + "Example n. 5817 = (49.0, 457.0): cluster 2\n", + "Example n. 5818 = (49.0, 456.0): cluster 2\n", + "Example n. 5819 = (49.0, 455.0): cluster 2\n", + "Example n. 5820 = (49.0, 454.0): cluster 2\n", + "Example n. 5821 = (49.0, 453.0): cluster 2\n", + "Example n. 5822 = (49.0, 452.0): cluster 2\n", + "Example n. 5823 = (49.0, 427.0): cluster 2\n", + "Example n. 5824 = (49.0, 426.0): cluster 2\n", + "Example n. 5825 = (49.0, 425.0): cluster 2\n", + "Example n. 5826 = (49.0, 424.0): cluster 2\n", + "Example n. 5827 = (49.0, 326.0): cluster 2\n", + "Example n. 5828 = (49.0, 325.0): cluster 2\n", + "Example n. 5829 = (49.0, 324.0): cluster 2\n", + "Example n. 5830 = (49.0, 323.0): cluster 2\n", + "Example n. 5831 = (49.0, 322.0): cluster 2\n", + "Example n. 5832 = (49.0, 321.0): cluster 2\n", + "Example n. 5833 = (49.0, 320.0): cluster 2\n", + "Example n. 5834 = (49.0, 319.0): cluster 2\n", + "Example n. 5835 = (49.0, 318.0): cluster 2\n", + "Example n. 5836 = (49.0, 308.0): cluster 2\n", + "Example n. 5837 = (49.0, 307.0): cluster 2\n", + "Example n. 5838 = (49.0, 306.0): cluster 2\n", + "Example n. 5839 = (49.0, 305.0): cluster 2\n", + "Example n. 5840 = (49.0, 304.0): cluster 2\n", + "Example n. 5841 = (49.0, 303.0): cluster 2\n", + "Example n. 5842 = (49.0, 302.0): cluster 2\n", + "Example n. 5843 = (49.0, 301.0): cluster 2\n", + "Example n. 5844 = (49.0, 300.0): cluster 2\n", + "Example n. 5845 = (49.0, 299.0): cluster 2\n", + "Example n. 5846 = (49.0, 298.0): cluster 2\n", + "Example n. 5847 = (49.0, 297.0): cluster 2\n", + "Example n. 5848 = (49.0, 296.0): cluster 2\n", + "Example n. 5849 = (49.0, 295.0): cluster 2\n", + "Example n. 5850 = (49.0, 294.0): cluster 2\n", + "Example n. 5851 = (49.0, 290.0): cluster 2\n", + "Example n. 5852 = (49.0, 289.0): cluster 2\n", + "Example n. 5853 = (49.0, 288.0): cluster 2\n", + "Example n. 5854 = (49.0, 287.0): cluster 2\n", + "Example n. 5855 = (49.0, 286.0): cluster 2\n", + "Example n. 5856 = (49.0, 285.0): cluster 2\n", + "Example n. 5857 = (49.0, 284.0): cluster 2\n", + "Example n. 5858 = (49.0, 283.0): cluster 2\n", + "Example n. 5859 = (49.0, 282.0): cluster 2\n", + "Example n. 5860 = (49.0, 281.0): cluster 2\n", + "Example n. 5861 = (49.0, 280.0): cluster 2\n", + "Example n. 5862 = (49.0, 279.0): cluster 2\n", + "Example n. 5863 = (49.0, 278.0): cluster 2\n", + "Example n. 5864 = (49.0, 277.0): cluster 2\n", + "Example n. 5865 = (49.0, 276.0): cluster 2\n", + "Example n. 5866 = (49.0, 275.0): cluster 2\n", + "Example n. 5867 = (49.0, 274.0): cluster 2\n", + "Example n. 5868 = (49.0, 273.0): cluster 2\n", + "Example n. 5869 = (49.0, 272.0): cluster 2\n", + "Example n. 5870 = (49.0, 271.0): cluster 2\n", + "Example n. 5871 = (49.0, 267.0): cluster 2\n", + "Example n. 5872 = (49.0, 266.0): cluster 2\n", + "Example n. 5873 = (49.0, 265.0): cluster 2\n", + "Example n. 5874 = (49.0, 264.0): cluster 2\n", + "Example n. 5875 = (49.0, 263.0): cluster 2\n", + "Example n. 5876 = (49.0, 262.0): cluster 2\n", + "Example n. 5877 = (49.0, 261.0): cluster 2\n", + "Example n. 5878 = (49.0, 260.0): cluster 2\n", + "Example n. 5879 = (49.0, 259.0): cluster 2\n", + "Example n. 5880 = (49.0, 258.0): cluster 2\n", + "Example n. 5881 = (49.0, 257.0): cluster 2\n", + "Example n. 5882 = (49.0, 256.0): cluster 2\n", + "Example n. 5883 = (49.0, 255.0): cluster 2\n", + "Example n. 5884 = (49.0, 254.0): cluster 2\n", + "Example n. 5885 = (49.0, 253.0): cluster 2\n", + "Example n. 5886 = (49.0, 252.0): cluster 2\n", + "Example n. 5887 = (49.0, 251.0): cluster 2\n", + "Example n. 5888 = (49.0, 250.0): cluster 2\n", + "Example n. 5889 = (49.0, 249.0): cluster 2\n", + "Example n. 5890 = (49.0, 248.0): cluster 2\n", + "Example n. 5891 = (49.0, 247.0): cluster 2\n", + "Example n. 5892 = (49.0, 246.0): cluster 2\n", + "Example n. 5893 = (49.0, 245.0): cluster 2\n", + "Example n. 5894 = (49.0, 244.0): cluster 2\n", + "Example n. 5895 = (49.0, 243.0): cluster 2\n", + "Example n. 5896 = (49.0, 242.0): cluster 2\n", + "Example n. 5897 = (49.0, 241.0): cluster 2\n", + "Example n. 5898 = (49.0, 240.0): cluster 2\n", + "Example n. 5899 = (49.0, 239.0): cluster 2\n", + "Example n. 5900 = (49.0, 238.0): cluster 2\n", + "Example n. 5901 = (49.0, 237.0): cluster 2\n", + "Example n. 5902 = (49.0, 236.0): cluster 2\n", + "Example n. 5903 = (49.0, 230.0): cluster 2\n", + "Example n. 5904 = (49.0, 229.0): cluster 2\n", + "Example n. 5905 = (49.0, 228.0): cluster 2\n", + "Example n. 5906 = (49.0, 227.0): cluster 2\n", + "Example n. 5907 = (49.0, 226.0): cluster 2\n", + "Example n. 5908 = (49.0, 225.0): cluster 2\n", + "Example n. 5909 = (49.0, 224.0): cluster 2\n", + "Example n. 5910 = (49.0, 223.0): cluster 2\n", + "Example n. 5911 = (49.0, 215.0): cluster 2\n", + "Example n. 5912 = (49.0, 214.0): cluster 2\n", + "Example n. 5913 = (49.0, 213.0): cluster 2\n", + "Example n. 5914 = (49.0, 212.0): cluster 2\n", + "Example n. 5915 = (49.0, 211.0): cluster 2\n", + "Example n. 5916 = (49.0, 210.0): cluster 2\n", + "Example n. 5917 = (49.0, 209.0): cluster 2\n", + "Example n. 5918 = (49.0, 208.0): cluster 2\n", + "Example n. 5919 = (49.0, 207.0): cluster 2\n", + "Example n. 5920 = (49.0, 196.0): cluster 2\n", + "Example n. 5921 = (49.0, 195.0): cluster 2\n", + "Example n. 5922 = (49.0, 194.0): cluster 2\n", + "Example n. 5923 = (49.0, 193.0): cluster 2\n", + "Example n. 5924 = (49.0, 192.0): cluster 2\n", + "Example n. 5925 = (49.0, 191.0): cluster 2\n", + "Example n. 5926 = (49.0, 190.0): cluster 2\n", + "Example n. 5927 = (49.0, 188.0): cluster 2\n", + "Example n. 5928 = (49.0, 187.0): cluster 2\n", + "Example n. 5929 = (49.0, 186.0): cluster 2\n", + "Example n. 5930 = (49.0, 185.0): cluster 2\n", + "Example n. 5931 = (49.0, 184.0): cluster 2\n", + "Example n. 5932 = (49.0, 183.0): cluster 2\n", + "Example n. 5933 = (49.0, 182.0): cluster 2\n", + "Example n. 5934 = (49.0, 181.0): cluster 2\n", + "Example n. 5935 = (49.0, 180.0): cluster 2\n", + "Example n. 5936 = (49.0, 179.0): cluster 2\n", + "Example n. 5937 = (49.0, 178.0): cluster 2\n", + "Example n. 5938 = (49.0, 177.0): cluster 2\n", + "Example n. 5939 = (49.0, 176.0): cluster 2\n", + "Example n. 5940 = (49.0, 175.0): cluster 2\n", + "Example n. 5941 = (49.0, 174.0): cluster 2\n", + "Example n. 5942 = (49.0, 173.0): cluster 2\n", + "Example n. 5943 = (49.0, 172.0): cluster 2\n", + "Example n. 5944 = (49.0, 171.0): cluster 2\n", + "Example n. 5945 = (49.0, 170.0): cluster 2\n", + "Example n. 5946 = (49.0, 169.0): cluster 2\n", + "Example n. 5947 = (49.0, 168.0): cluster 2\n", + "Example n. 5948 = (49.0, 167.0): cluster 2\n", + "Example n. 5949 = (49.0, 166.0): cluster 2\n", + "Example n. 5950 = (49.0, 165.0): cluster 2\n", + "Example n. 5951 = (49.0, 164.0): cluster 2\n", + "Example n. 5952 = (49.0, 163.0): cluster 2\n", + "Example n. 5953 = (49.0, 162.0): cluster 2\n", + "Example n. 5954 = (49.0, 160.0): cluster 2\n", + "Example n. 5955 = (49.0, 159.0): cluster 2\n", + "Example n. 5956 = (49.0, 158.0): cluster 2\n", + "Example n. 5957 = (49.0, 157.0): cluster 2\n", + "Example n. 5958 = (49.0, 156.0): cluster 2\n", + "Example n. 5959 = (49.0, 149.0): cluster 2\n", + "Example n. 5960 = (49.0, 148.0): cluster 2\n", + "Example n. 5961 = (49.0, 147.0): cluster 2\n", + "Example n. 5962 = (49.0, 146.0): cluster 2\n", + "Example n. 5963 = (49.0, 145.0): cluster 2\n", + "Example n. 5964 = (49.0, 144.0): cluster 2\n", + "Example n. 5965 = (49.0, 143.0): cluster 2\n", + "Example n. 5966 = (49.0, 142.0): cluster 2\n", + "Example n. 5967 = (49.0, 141.0): cluster 2\n", + "Example n. 5968 = (49.0, 140.0): cluster 2\n", + "Example n. 5969 = (49.0, 139.0): cluster 2\n", + "Example n. 5970 = (49.0, 138.0): cluster 2\n", + "Example n. 5971 = (49.0, 137.0): cluster 2\n", + "Example n. 5972 = (49.0, 136.0): cluster 2\n", + "Example n. 5973 = (49.0, 135.0): cluster 2\n", + "Example n. 5974 = (49.0, 134.0): cluster 2\n", + "Example n. 5975 = (49.0, 133.0): cluster 2\n", + "Example n. 5976 = (49.0, 132.0): cluster 2\n", + "Example n. 5977 = (49.0, 131.0): cluster 2\n", + "Example n. 5978 = (49.0, 130.0): cluster 2\n", + "Example n. 5979 = (49.0, 129.0): cluster 2\n", + "Example n. 5980 = (49.0, 128.0): cluster 2\n", + "Example n. 5981 = (50.0, 495.0): cluster 2\n", + "Example n. 5982 = (50.0, 494.0): cluster 2\n", + "Example n. 5983 = (50.0, 493.0): cluster 2\n", + "Example n. 5984 = (50.0, 492.0): cluster 2\n", + "Example n. 5985 = (50.0, 491.0): cluster 2\n", + "Example n. 5986 = (50.0, 490.0): cluster 2\n", + "Example n. 5987 = (50.0, 489.0): cluster 2\n", + "Example n. 5988 = (50.0, 488.0): cluster 2\n", + "Example n. 5989 = (50.0, 487.0): cluster 2\n", + "Example n. 5990 = (50.0, 486.0): cluster 2\n", + "Example n. 5991 = (50.0, 485.0): cluster 2\n", + "Example n. 5992 = (50.0, 484.0): cluster 2\n", + "Example n. 5993 = (50.0, 483.0): cluster 2\n", + "Example n. 5994 = (50.0, 482.0): cluster 2\n", + "Example n. 5995 = (50.0, 481.0): cluster 2\n", + "Example n. 5996 = (50.0, 480.0): cluster 2\n", + "Example n. 5997 = (50.0, 479.0): cluster 2\n", + "Example n. 5998 = (50.0, 478.0): cluster 2\n", + "Example n. 5999 = (50.0, 477.0): cluster 2\n", + "Example n. 6000 = (50.0, 476.0): cluster 2\n", + "Example n. 6001 = (50.0, 475.0): cluster 2\n", + "Example n. 6002 = (50.0, 474.0): cluster 2\n", + "Example n. 6003 = (50.0, 473.0): cluster 2\n", + "Example n. 6004 = (50.0, 472.0): cluster 2\n", + "Example n. 6005 = (50.0, 471.0): cluster 2\n", + "Example n. 6006 = (50.0, 470.0): cluster 2\n", + "Example n. 6007 = (50.0, 469.0): cluster 2\n", + "Example n. 6008 = (50.0, 468.0): cluster 2\n", + "Example n. 6009 = (50.0, 467.0): cluster 2\n", + "Example n. 6010 = (50.0, 466.0): cluster 2\n", + "Example n. 6011 = (50.0, 465.0): cluster 2\n", + "Example n. 6012 = (50.0, 464.0): cluster 2\n", + "Example n. 6013 = (50.0, 463.0): cluster 2\n", + "Example n. 6014 = (50.0, 462.0): cluster 2\n", + "Example n. 6015 = (50.0, 461.0): cluster 2\n", + "Example n. 6016 = (50.0, 460.0): cluster 2\n", + "Example n. 6017 = (50.0, 459.0): cluster 2\n", + "Example n. 6018 = (50.0, 458.0): cluster 2\n", + "Example n. 6019 = (50.0, 457.0): cluster 2\n", + "Example n. 6020 = (50.0, 456.0): cluster 2\n", + "Example n. 6021 = (50.0, 455.0): cluster 2\n", + "Example n. 6022 = (50.0, 454.0): cluster 2\n", + "Example n. 6023 = (50.0, 453.0): cluster 2\n", + "Example n. 6024 = (50.0, 452.0): cluster 2\n", + "Example n. 6025 = (50.0, 428.0): cluster 2\n", + "Example n. 6026 = (50.0, 427.0): cluster 2\n", + "Example n. 6027 = (50.0, 426.0): cluster 2\n", + "Example n. 6028 = (50.0, 425.0): cluster 2\n", + "Example n. 6029 = (50.0, 424.0): cluster 2\n", + "Example n. 6030 = (50.0, 423.0): cluster 2\n", + "Example n. 6031 = (50.0, 329.0): cluster 2\n", + "Example n. 6032 = (50.0, 328.0): cluster 2\n", + "Example n. 6033 = (50.0, 327.0): cluster 2\n", + "Example n. 6034 = (50.0, 326.0): cluster 2\n", + "Example n. 6035 = (50.0, 325.0): cluster 2\n", + "Example n. 6036 = (50.0, 324.0): cluster 2\n", + "Example n. 6037 = (50.0, 323.0): cluster 2\n", + "Example n. 6038 = (50.0, 322.0): cluster 2\n", + "Example n. 6039 = (50.0, 321.0): cluster 2\n", + "Example n. 6040 = (50.0, 320.0): cluster 2\n", + "Example n. 6041 = (50.0, 319.0): cluster 2\n", + "Example n. 6042 = (50.0, 318.0): cluster 2\n", + "Example n. 6043 = (50.0, 307.0): cluster 2\n", + "Example n. 6044 = (50.0, 306.0): cluster 2\n", + "Example n. 6045 = (50.0, 305.0): cluster 2\n", + "Example n. 6046 = (50.0, 304.0): cluster 2\n", + "Example n. 6047 = (50.0, 303.0): cluster 2\n", + "Example n. 6048 = (50.0, 302.0): cluster 2\n", + "Example n. 6049 = (50.0, 301.0): cluster 2\n", + "Example n. 6050 = (50.0, 300.0): cluster 2\n", + "Example n. 6051 = (50.0, 299.0): cluster 2\n", + "Example n. 6052 = (50.0, 296.0): cluster 2\n", + "Example n. 6053 = (50.0, 293.0): cluster 2\n", + "Example n. 6054 = (50.0, 292.0): cluster 2\n", + "Example n. 6055 = (50.0, 291.0): cluster 2\n", + "Example n. 6056 = (50.0, 290.0): cluster 2\n", + "Example n. 6057 = (50.0, 289.0): cluster 2\n", + "Example n. 6058 = (50.0, 288.0): cluster 2\n", + "Example n. 6059 = (50.0, 287.0): cluster 2\n", + "Example n. 6060 = (50.0, 286.0): cluster 2\n", + "Example n. 6061 = (50.0, 285.0): cluster 2\n", + "Example n. 6062 = (50.0, 284.0): cluster 2\n", + "Example n. 6063 = (50.0, 283.0): cluster 2\n", + "Example n. 6064 = (50.0, 282.0): cluster 2\n", + "Example n. 6065 = (50.0, 281.0): cluster 2\n", + "Example n. 6066 = (50.0, 280.0): cluster 2\n", + "Example n. 6067 = (50.0, 279.0): cluster 2\n", + "Example n. 6068 = (50.0, 278.0): cluster 2\n", + "Example n. 6069 = (50.0, 277.0): cluster 2\n", + "Example n. 6070 = (50.0, 275.0): cluster 2\n", + "Example n. 6071 = (50.0, 274.0): cluster 2\n", + "Example n. 6072 = (50.0, 273.0): cluster 2\n", + "Example n. 6073 = (50.0, 272.0): cluster 2\n", + "Example n. 6074 = (50.0, 271.0): cluster 2\n", + "Example n. 6075 = (50.0, 268.0): cluster 2\n", + "Example n. 6076 = (50.0, 267.0): cluster 2\n", + "Example n. 6077 = (50.0, 266.0): cluster 2\n", + "Example n. 6078 = (50.0, 265.0): cluster 2\n", + "Example n. 6079 = (50.0, 264.0): cluster 2\n", + "Example n. 6080 = (50.0, 263.0): cluster 2\n", + "Example n. 6081 = (50.0, 262.0): cluster 2\n", + "Example n. 6082 = (50.0, 261.0): cluster 2\n", + "Example n. 6083 = (50.0, 260.0): cluster 2\n", + "Example n. 6084 = (50.0, 259.0): cluster 2\n", + "Example n. 6085 = (50.0, 258.0): cluster 2\n", + "Example n. 6086 = (50.0, 257.0): cluster 2\n", + "Example n. 6087 = (50.0, 256.0): cluster 2\n", + "Example n. 6088 = (50.0, 255.0): cluster 2\n", + "Example n. 6089 = (50.0, 254.0): cluster 2\n", + "Example n. 6090 = (50.0, 253.0): cluster 2\n", + "Example n. 6091 = (50.0, 252.0): cluster 2\n", + "Example n. 6092 = (50.0, 251.0): cluster 2\n", + "Example n. 6093 = (50.0, 250.0): cluster 2\n", + "Example n. 6094 = (50.0, 249.0): cluster 2\n", + "Example n. 6095 = (50.0, 248.0): cluster 2\n", + "Example n. 6096 = (50.0, 247.0): cluster 2\n", + "Example n. 6097 = (50.0, 246.0): cluster 2\n", + "Example n. 6098 = (50.0, 245.0): cluster 2\n", + "Example n. 6099 = (50.0, 244.0): cluster 2\n", + "Example n. 6100 = (50.0, 243.0): cluster 2\n", + "Example n. 6101 = (50.0, 242.0): cluster 2\n", + "Example n. 6102 = (50.0, 241.0): cluster 2\n", + "Example n. 6103 = (50.0, 240.0): cluster 2\n", + "Example n. 6104 = (50.0, 239.0): cluster 2\n", + "Example n. 6105 = (50.0, 238.0): cluster 2\n", + "Example n. 6106 = (50.0, 237.0): cluster 2\n", + "Example n. 6107 = (50.0, 236.0): cluster 2\n", + "Example n. 6108 = (50.0, 230.0): cluster 2\n", + "Example n. 6109 = (50.0, 229.0): cluster 2\n", + "Example n. 6110 = (50.0, 228.0): cluster 2\n", + "Example n. 6111 = (50.0, 227.0): cluster 2\n", + "Example n. 6112 = (50.0, 226.0): cluster 2\n", + "Example n. 6113 = (50.0, 225.0): cluster 2\n", + "Example n. 6114 = (50.0, 224.0): cluster 2\n", + "Example n. 6115 = (50.0, 215.0): cluster 2\n", + "Example n. 6116 = (50.0, 214.0): cluster 2\n", + "Example n. 6117 = (50.0, 213.0): cluster 2\n" + ] + }, + { + "ename": "IndexError", + "evalue": "index 6118 is out of bounds for axis 0 with size 6118", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mIndexError\u001b[0m Traceback (most recent call last)", + "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m\u001b[0m\n\u001b[1;32m 7\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 8\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mn_samples3\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 9\u001b[0;31m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'Example n. {i} = ({(data3[i,0])}, {data3[i,1]}): cluster {km2.labels_[i]}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", + "\u001b[0;31mIndexError\u001b[0m: index 6118 is out of bounds for axis 0 with size 6118" + ] + } + ], + "source": [ + "data3, feature_names3, n_samples3, n_features2 = load_data(file_path, file_name3)\n", + "\n", + "np.random.seed(5)\n", + "\n", + "k=4\n", + "km3 = KMeans(n_clusters=k, random_state=0).fit(data3)\n", + "\n", + "for i in range(n_samples3):\n", + " print(f'Example n. {i} = ({(data3[i,0])}, {data3[i,1]}): cluster {km2.labels_[i]}')" + ] }, { "cell_type": "markdown", @@ -689,10 +13035,25 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 10, "metadata": {}, - "outputs": [], - "source": [] + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfQAAAI1CAYAAAAgiggGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOy9eZydRZU+/pz3dneWTmdPIAmEAGETZAuRJYMKArKp6IigjA7gMt8ZZ1PHZRhH56ejONtvYMQZt+/MODooKKuiLAoiRLawJpAEsi8k6WydpNOdXu57vn/Ue6pOnVv3dickZPF98unPvfddTp2qevM+55w6VUXMjBIlSpQoUaLE/o1sbytQokSJEiVKlHjtKAm9RIkSJUqUOABQEnqJEiVKlChxAKAk9BIlSpQoUeIAQEnoJUqUKFGixAGAktBLlChRokSJAwAloZc4oEFEf0dEP9jbeuwsiOi/iejv94Dc64jou7tb7p4sh4iuJqJHd4esEiUOZJSEXmK/BxF9gIjmEFEnEa0hol8Q0e/tRvnTiIiJqGl3ydxbYOavMvNHBnPtazGGdqac3YnXy4Db0+UQ0aeJaB4RbSOipUT06T1VVokDByWhl9ivQUSfBHADgK8COAjAVAD/DuBde1MvjQPBECjxuoMAfAjAGAAXAvhTIrpy76pUYl9HSegl9lsQ0SgAXwLwcWa+nZm3M3MfM/+UmWs8GiJ6KxGtMseWEdF5xfc3FZ7+ViJaR0T/f3HZb4rPjiIKcGZx/bVENJ+INhPRfUR0mJLLRPRxInoFwCvFsUuJ6Dki6iCi3xLRier6U4jomcIjuwXA0Ab1vpqIZhPRTUS0hYgWENHb1PnJRHQ3EW0iokVE9FF1znuWKvLwh0S0gog2ENHfFOcuBHAdgCuKOj+vyl6iPMer6ug4qHLq3Duu0H8rET0J4Ehz/kYiWlmcf5qIzh5A52uKftpW6P5HStZ4IvpZ0SebiOgRIspUO95GROuLuv55o3IS9VhGRH9FRC8U/XQLEdXtVw1m/kdmfoaZ+5l5IYC7AMwazL0lfndREnqJ/RlnwhHfHbtJ3o0AbmTmkXAkcmtx/M3F52hmHsHMjxHRu+Be6u8BMAHAIwB+aORdBuB0AG8golMA/CeAPwIwDsC3ANxNREOIqAXAnQC+D2AsgB8D+P0BdD0dwGIA4wF8EcDtRDS2OPcjAKsATAbwXgBfJaJzG8j6PQDHAHgbgC8Q0XHMfC9c1OOWos4nEVErgH8DcBEztwE4C8BzA+jZsJw6130DwA4AkwBcW/xpPAXgZLi2uhnAj4loaErn4vp2AJcCGAngGgD/SkSnFuc+BddWE+AiPNcB4ILUfwrgeQBTCp3/koje3qCcFN4H52EfDuBEAFc3uDYJIiIAZwN4cWfvLfG7hZLQS+zPGAdgAzP37yZ5fQCmE9F4Zu5k5scbXPt/AFzPzPOL8r8K4GTtpRfnNzFzN4CPAfgWMz/BzFVm/h6AHgBnFH/NAG4oIgw/gSOtRmhX198CYCGAS4joUDhP7rPMvIOZnwPwXbjwbT38f8zczczPwxFYI4LKAZxARMOYeQ0z7wzJDFgOEVXgjJkvFBGXeQC+p69h5h8w88bCe/0XAEPgDIUkmPkeZl7MDg8DuB+OIAHX55MAHFa05SPsNriYCWACM3+JmXuZeQmA7wDY2bD3vzHzq8y8Cc5AOHkn7weAv4N7V//XLtxb4ncIJaGX2J+xEcD43ThG/WEARwNYQERPEdGlDa49DMCNRai2A8AmuHHPKeqaleb6T8n1xT2HwnnRkwGs5ninpOUD6Jq6XmRtYuZt5pzWy2Kt+t4FYETqImbeDuAKOGNmDRHdQ0THDqDnzpYzAUAT4raL2qIIY88vwtgdAEbBRSqSIKKLiOjxIqTeAeBidf0/AVgE4P4iHP+54vhhACab/roOzovfGQyqbRvo/qdwxtglzNyzk2WX+B1DSegl9mc8BuflXjbI67cDGC4/Cm9wgvxm5leY+f0AJgL4BwA/KcLMqS0JVwL4I2Yerf6GMfNv1TVsrv+KuX44M/8QwBoAU4rQqmDqAHVJXf9q8TeWiNrMudUDyEuhpt7MfB8znw/n1S6A81p3J9YD6IczdgS+LYrx8s/AhbLHMPNoAFvgjKkanYloCIDbAPwzgIOK638u1zPzNmb+FDMfAeCdAD5Z5COsBLDU9FcbM1+cKmdPgIiuBfA5AG9j5lUDXV+iREnoJfZbMPMWAF8A8A0iuoyIhhNRc+GR/WPilpcBDCWiS4ioGcDn4cK1AAAi+gMimsDMOYCO4nAORzI5gCOUrG8C+GsiOr64dxQRXd5A3e8A+D9EdDo5tBZ6tMEZJv0A/rzQ/z0A3jRA9Seq6y8HcByAnzPzSgC/BXA9EQ0ll3j3YQC7MsVqHYBpKknsICJ6V2Hk9ADohGuX3QZmrgK4HcDfFf35BgB/qC5pg2ur9QCaiOgLcGPjSZ0BtMD18XoA/UR0EYAL5GJyiYrTC+NoC4BqUacnAWwjos8S0TAiqhDRCUQ0s045uxXkkg2/CuD8ItxfosSAKAm9xH6NYgz1k3DkvB7Os/pTuCQze+0WAH8CN6a8Gs5j157PhQBeJKJOuAS5K4sx3y4AXwEwuwi/nsHMd8B58T8ioq0A5gG4qIGecwB8FMBNADbDhXmvLs71wiXXXQ0Xur8CjtQa4QkARwHYUOj2XmbeWJx7P4BpcN76HQC+yMy/HEBeCj8uPjcS0TNw74tPFnI3AXgLgD/eBbkD4U/hQtNrAfw34rHj+wDcC2ecLYdLntPh+UjnYujhz+ESHDcD+ACAu9X1RwH4JZxx8hiAf2fmhwrD4lK4Me+lcO38Xbjwfk05O1tBIppaZMjXi8T8PVyOyFPFdZ1E9M2dLafE7xYoHoYrUaLEvg4iuhrAR5h5ty2eU6JEif0fpYdeokSJEiVKHAAoCb1EiRIlSpQ4AFCG3EuUKFGiRIkDAKWHXqJEiRIlShwAKAm9RIkSJUqUOACwX+8CNX78eJ42bdreVqNEiRIlSpR4XfD0009vYOYJqXP7NaFPmzYNc+bM2dtqlChRokSJEq8LiKjustBlyL1EiRIlSpQ4AFASeokSJUqUKHEAoCT0EiVKlChR4gBASeglSpQoUaLEAYCS0EuUKFGiRIkDACWhlyhRokSJEgcASkIvUaJEiRIlDgCUhF6iRIkSJUocACgJvUSJEiVKlDgAUBJ6iRIlSpQocQCgJPQSJUqUKFHiAEBJ6CVKlChRosQBgJLQS5QoUaJEiQMAJaGXKFGiRIkSBwBKQi9RokSJEiUOAJSEXqJEiRIlShwAKAm9RIkSJUqUOABQEnqJEiVKlChxAKAk9BIlSpQoUeIAQEnoJUqUKFGixAGApr2tQInfcaxfD9xwA9DdDUyYAFx8MXDzzQARcOyxwNVX720NS5QYNB5c+iA2dG3A+45/395WpcTvIIiZ97YOu4zTTjuN58yZs7fVKJHC/fcDjz/uiJnZEfVpp4XzXV3ATTcBmzcD7e3uWHMzMGqUO5ZlwLBhwNix7n4AOPts4JxzgB/8AFi8GMhzJ/+aa4DDDquvyzPPAD/7mbsWcPLk+9vfDrzpTbu//iV+p/Dsmmdx98K70d3XjX7ux6gho3DWoWfhbUe8bZfkPbn6Sdy76F4QKDqecw4AYDByzpFRhg+88QO4Y/4d2Na7DQAwduhYXHjUhbhl3i3IKAMR4Zhxx+CKE66IZP1o3o/w8saXkXOOKlcBBporzfjYjI/h4BEH75LeJfY8iOhpZj4tea4k9BKDRnc38N3vAtvciwOjRwMf/SiwbBlwxx2BJIkcKW/eHO6dMAFobQXe/W5g2jTgW98C5s4FqtVwn9xbqYTv+nPsWGDkSGDdOmD7dncvs5P3kY8ABx3krluzBvj+9935SsVdu359XEbuXoyYOBEYMQJ45zuBY47ZjY21m9HbC3znO+7zmmtc25fYZ/DI8kfww7k/BAAQEZgZY4ePxaiho3DutHMxY/KMQct6sf1F3PrirdjQtQEAwMwAAWBH5ID7ZHZ/B7cdjPbt7ajmVQBAc9aMCa0TsKZzDTJyo6ojW0Zi4oiJABf6gdG+vR3bereB2RkHnDMoIxw55kh8bMbH0Jf34X+e/x/0V/vBYFSyCg4ffTguP/7y3dhyJXYWJaG/3li/Hrj1VmDMGOD9748Ja39GZyfwhS84QicCWlqAo45yhLlypTtmSVieL/mcOhV43/scoW/aFMuX+4mch65l5bk7puXluftrbgY+8xnX3jffDGzd6jx4ZkfoVicNkXfIIc7gkGNEwPDhwB/8gYsULF8O/PSn7vjhh7uIw57CAw8AL78cdLv4YuA3vwHmzAH6+oCTTnJDEcOH7zkd9iGs3roaj6x4BFeecOXeVsVj/fb1uPXFWx2JEtDZ04nV21YDcIQJBsS5Prj1YIweNtqRMQPvOOYdeGjpQ9jSswWtza246sSrsGXHFtz20m1gMLb1bMOqbasAABkydx8BxE5gjhxgoJpXkZP7DsB58/I6J6dHhSrhXip0yt0ng73xUa1WweQMhApV8MaD3ojzjjgP//b4v6GfHaE3URPahrRh6qipUR1nTp6Jjd0bsXjTYhc5yF3k4L3Hv7f09PcAGhH6Hh1DJ6JlALYBqALoZ+bTiGgsgFsATAOwDMD7mHkzERGAGwFcDKALwNXM/Mye1G+X8PDDwNKlLvx75JG157dsAX78Y2DePEcEI0cCl176+us5WLz8MjB7dkx6zMDppwPHHeeO3XmnM1L6+4GennBtXx/w0ku1BCzEq0PbguXLnTff3Z0m2BSZi0z9Gwjy8xy46y5H3i++GDx3fY82MLIseOhS35Ur3THRO8uc597X5wyEO+5wbUXkIgAyTEAEnHgicMopO9/2KTz2GPDoo8DGjUGXnh5gxQqnC7Or409+Alx5pTOqAGDJEkf6zMDRRwOzZu0effYyNnZtxO0v3Y6lW5aiu68bU0ZOwQVHXoC56+bi6Vef9t7mzMkzcfzE43d7+cyMOxfciY3dG8HMGNo0FOcdcR7uXHAn5q2b50gRhCzLQCBP2gwGsfu9tnMt1nSu8WTXV+3D0o6l6OnvQXOlGdW8ip5qD+avn+88annkKfMeNQBHwsyhHCFqBpjYtwXIkTuBwMTIkHmCBwOcsTcOwIiMD9H5pfUvobe/1xkPCPI6ezvxUvtLyLLMRw627NiC7v5ubOza6OpYzYEM6J/Xj9bm1qAvgDHDxuA9x73HRw5K7F68Hklx5zDzBvX7cwB+xcxfI6LPFb8/C+AiAEcVf6cD+I/ic9/B0087Qm9vd17q7/8+MHlyON/dDdx9twsl57kbJ549G2hrA97ylr2nN+C86F/8wpHC6NHAhRcCq1a5Yy+/7K7RZLl5s6sHADz1lDNULFKELQSp5WmvmhlYuLD2nhTqedT2eJ47XbURUC9SIPpZo0NHquSe3t5gfCxYEAyEzZtdm8h969Y5b3l3hOxfftmRucYrr8Rt2NcHvPACcHkR+lyzBvj5z52OzMDq1S7acPLJr12fvYhtPdtw14K78NL6l8BgPLX6KSzevBgdOzqwcstKLO1Y6r3Nia0Tdyuhz14xG6u3rXblrnoKW3u2gkBoqjRha89WvLj+ReTI/Ri1gBybes9aiFOIHwws3LjQe9L9eT+eWfOMlyHXEQicK9I2IXcgkK8QuHwvFHHGgCJ2Oe51U8cpL8onBoPRl/dh/ob5Ud28bkSRHq9ue9WP7Uv9OWe81P6Svy+Da6fhLcMxomUETph4AmavmI1DRx2Ksw49a7f0WYm9k+X+LgBvLb5/D8Cv4Qj9XQD+h90YwONENJqIJjHzmr2gYy3mz3eh0PZ299KcP9+9RN/9bmDcOHfN9u3Ak0+GcWFmoKPDHdubhN7b64j7wQedlz1ypDMyiAKZAzFZLlrk/iSsnSI+IThNNnKN/tTQ5+x9QG0Z+pgmZ+2dp/RPlal/p77be6pVYO1aF5HR+trkuuXLXRLgokXOU9dG3q7ADlPoOuryBatXO69d+mPtWhchGojQe3udwTlrVvD09yFs6dmCZ9Y8g5xz711u6t6Eh5c9DKAYCyZHMos3L8bKLStx6KhDd7qcpZuX4qX1L4GIcPS4o9HZ24lfLvkl1nWu8+PLgCO7vmofnl/7vCsf5BLTsuBVA45oc+SeyIT8cs59uFy8bfiuduPXGkRUc50mbqYiClCEyjWBezC8l10IqCmfwYHwuTiu7o8IXBkVGbIoSU/qLm3lZZHTgdh5+A8ufRArt6zEs2ufxcEjDkbHjo4o8nD8xOMxbfS0nevEEgD2PKEzgPuJiAF8i5m/DeAgRdJrARSZTJgCYKW6d1VxLCJ0IvoYgI8BwNSpU/eg6gbPPOPCskB4uc+ZA8ycGQi9udklaC1cGId0Xy+8+GIIBYuegIsUPPSQe4ETOW/7nnucR2k9U0semgztOcky12WlrrH6aINHk3s97z5F3vqclq2/S2g91QdWLwnTSzmSnDdligtn19NRrl2wwA0/SNJeR0d62GH4cDecUe+5mDzZGVsSEclzd2z9+jDcQeSy+m0yoW6L9naXrDhtWm0ZzC60v2kT8OtfO0N03DjgjDNe3+d1EMiLfxlnyCnHiKYRGNEyAuu71ntCyinH+u3rsbF746AJfWPXRsxtnwswMH/DfMxdNxcg4A3j34BtvduwtnMtAPhscvE6Zcw58m7hjkloW+4jokB4kngmIXkOIXLx3PM8GC6ecLMQKhcwczASkEdk6L1oDuF2X4ZTLHjTHN9TZZdUJ2Ts7QLRR64l8tf79ijqKtdIXXyUQpH6lu4tePrVp5FlGdZtW4d7Xr4nDFEQY+WWlbjsuMuwrnMdNndvximTTsGooaN26fn5XcOeJvTfY+bVRDQRwANEtECfZGYuyH7QKIyCbwMuKW73qToAxFN1SoRP/QIcNcolwX35y448s8yRvGRf70ksW+bGuleujAkqpSfgCGLTptrxZQ0hJC1LnwPCeLdNWrPyLOFrUrdGgyXilCcuSXGa1FN1EJ1S4Xi5V7eTlCNDJr/+NaIQvTXU7HDCihWO2NetS0ca2trc8abEf71Jk4Dzz3fDOY8+6nIwhg4F3vteFwGQJMLWVpcUN3RouFfrReSiBffc4+61z19PD3D77aH/b7/dEfqMGfuUp96cNWPCsAlRgthBIw7Cmw97M+5ccCe27CiMHgZOn3I6Tj54cEMMnb2deGDJA3ho6UMRCROTD+/7f+Jhch68U0W4GWUuCSxz3qoQpxghWoZkpYsMbagIqVdRRQWVQJwcjAKpq3jk1bzqvekcoWyJWhAIVVTdeHeuvPBCjvauZdqaREN8PeHG8kUPXU+JUIg8oIhCCLEXQwbyXXv6GTJU82oyovD8uucxrHkYlm5eijWda7Bxx0YcOvJQnHzwyWipvPbnc+66uTh63NEY0jTkNcva17BHCZ2ZVxef7UR0B4A3AVgnoXQimgRAXMrVALR5fUhxbN/AqFHuBdrdHV7ebW21L8Ascy/Hnh73e+pU4EMf2jM6bdzoiANw4f/ly2Pi1KRok8AqlXC+nlcm5zX55nl8b72Qt/UYBdVqOKZlVCq10QL51Prp8nQd9XFtKIhOKcNFk7k1gES+JvIsCwaMraf8PfxwkGWNjSxznvt3vhMiALpub3wjcO65bkGdBQuAM8908+6BkKCYwpAhjuRl9oEYRC+8EAyCUcrDWbLEGZy6ntIv+whyzrGlZwuuOeUafOvpb6Gf+9FETRgzbAxOP+R0bO/bjoeWPuQJorXFzVBY3rEcE1onYHhzPAOgmlexcMNCgIAlm5fgl4t/6ULcjECEhSw/vlx45xImZ8fCbsw6D6Funbymw9USdpf78jz3x8AAGOhHfxjnhpPlSU6Rt0ehBzP7ULp4wz60zojk5Hk4rhPUAMTGRiG3ylUXTi9yBLyhksHrLUaDELYOxXtZYsz4gJYzUKpUDboUwwaRXgTMXjnbRy3uWXgPmrImXHHCFSFrngFkwLCmYThizBGDeqaWdSzD1h1b8f2538fF0y/Gmw97MypZZVD37i/YY9PWiKgVQMbM24rvDwD4EoC3AdiokuLGMvNniOgSAH8Kl+V+OoB/Y+aGK3687tPWfvhDF3qXF/VVVwGnnrr7y1m1yr1gp05Nk21XlwupPvgg8MQT7pgmD/2CtqQGxASXyijX5F+tDhyZ0HJsNrr1frWXq89rHbIszEUX71hDe+ap6AEQiLhSqa2jNTZEjq23vUcTutZZ10XqJzqKvqn218aBlZXnLnHx0ksdYQ+EJ55wU/b07AEp/5xzgMsucx5/tQp85SsuL0DXd+xY4PrrB1fWHsbKLSuxvnM9bp53Mz5y6kdw7IRjB7xnQ9cGdPZ24nvPfQ+nTT4NJ0w8AVNGTkFT1oQVHSuwZtsafPeZ76KP+5xXDqCSVRyZZI6YhKzFewYcoVQh/5dC2NgbA0X4WbLcJWlMCAe5kaXC1OJVCzHKd0m0i2Qp0iU48u7P+51a6vm3SXlCWJKUpkP92quW6Xc+IpHnqGQV/5uYUMkq3ngRcvfJfOwiAJyz99KjoYPikZTs9go5WRKJECNEZAi5V/NqiAIUwx4AfL7BxOET8Scz/wTDmofhoBH1I6HrOtfhxsdvxKudrwIAhlSG4COnfgSnTT7N12F/wV6Zh05ERwC4o/jZBOBmZv4KEY0DcCuAqQCWw01b21RMW7sJwIVw09auYeaGbP26E7pODgNqiWZ3YNUq4J/+yXlQf/EXzlvL8zjz+dFHgXvvjb1DgSU6ra/+bokOiL9rD1MbBxZCRtbbTIXxraFhidWSpiXA1Li37RM9FCD11F6rJXQtS3SUMm19pc1sGN/Wob+/tl/0eVtPbbTYe66+GjhrkFnA99/vpkymDIRLLgEuusgNAXV1AdddF8b5idw8/L/9271O6Cs6VuCfH/tnbO3ZijzPMXbYWHx61qcxZeSUhvfd9MRNmNs+15NJRhmuPeVaTBg+Af/6+L9ie+92d2FBnBlljoQ5kGnkZRbE2c/9UZhczgHwhFShipfFYJckp2TJ92peDR47hzIERE6vSlbx5z2BFrpr79fLYzWVDSHikGUZKlQJZRTErMf5xUiRRWm83oVho8muKWuK9VWRBT2mn+d5jV4SyhcjpSlrCp575jx3MTR0+8iwQjSkoNpNZE0bPQ0fPfWjGNY8DG1D2mqej+sfuR6vbHol6svhzcPxpXO/hPHDxzd8tvY17JV56My8BMBJieMb4bx0e5wBfHxP6bNbsCcIXGPLFuDrX3efeQ78x38An/60C6t/+9vumnrhZ02Qmkw0MdkwsU7+0vekyMqWqYlRyFaIypadCoFb/XV5Vl+rl/XyrWGTCvfbBD59TyoCkQr/azm2XXU5FvWMBB3GtwaG/tTo7HSfI0bUnmtqcoTd3x/Xg8iR/cSJLvFNcjtkHH/UKGc8vkYy7+zt9N8J5MPgg0VvtRdff/Lr2Lpjq/dqO3Z04KYnb8KXz/2yI4EEevp7UOWqD49zzuhHP7777HcBdlPDomQxFRoWXT0ZqrB57tzrAArjzhmyyDOW8WJ9XJLlNOHBdGnOuSc0L5uLEH5GIWSdhXuFLLXHLsck1C71Y3CYiw5E1wtpWgPDZ+IX9fThcSkvI1fPnOFOK1kc/iQcryMR/lr/f0ctnuMbM9wn10ZkL90hSX05Y1nHMnzx4S/ixAkn4tpTr60ZH2+uNLvrJVpALl9hf15YLYVyc5Z9BczAv/5rvGBJRwfwpS+57z09IQxtl0YVD09/6rC09bytJ20JXBOMnq9dT1bqfu15alk6FG3roH9r+ToiomXZbHmLVMTAXqe991R9hIxtNCNl9Giit4ZOPaTaL6WrLHBz/fXums9+1q2M19wcrjn7bDdH/oEHatvzoovcjAzAjav/zd/EerzGpWTXda7DPzz6D9je5zzhia0T8Xdv/budHqOUMLIkqzG7OdEA0NvfWxMeZWbcPO9mLNiwwHneyJBnjgjyPI8SuCQsXKFCpzx4oDJFzP2An37lp5JxSGITYtcep2S5+7Fl5el6WSBwFsgPHBaP8WRM5Fda80SaBfKTe3KEBDUGew8XgE9e8yF2djIyOLKPkuwAP4QgoW6dqCZetvSFzrjPMjWfXRlEXq/iGdR9KW0mcuVe3/6So8DOyEEGl6+Q0IuJfZQEuSPoue1zceuLt+KDJ30wquMfzfgj/OOj/4hV21b5MpupGXsCfVX3vDZX9oz8RigJfV9CT0/ti7i3N/6tx6h1qFaO1SNMCyEfG9rWnylv38rQsjTppe6xXrJdsU3/WQK1ZWrCTOmjjQEbTUh51qm20d64XKuNqZQxkYoE1KtDaiU7Dbm/v9+Ne69f78LlgDP0xo4FPv/5kJjZ3OyIe/t24Le/jes4dGjQHdita8F39nbixsdvxKbuTZAwcU9/z64JI0CP8cqY6/Nrn8f3X/h+SPZS7dXT3+O8OTmUB89TiDFH7slZCERC4zpEXjNtDLEnL56yrOgmZJ1Vsui4lO1D9aKb5+ZgCDCCp5vBDQUA8OQn38GIkuQivSjIlvvBQS8/px3KW1detGShe11Flgr5S56ADtnbZW69dy7dScErzuAy+jNkIcwu3j7C9LYwQkB+zFwTukQJdBnemEGO3mpvzWPVNqQttAuAlkqLG5JpnVBz7WvB8o7luOHxGzBu+Dhcd/Z1r/uKeOVa7nsTP/iBW+HsPe9xO3799V+7RUFSIWeiMB6cImG5Tv5S09YsyelENnuNlaXlWyJLjXtb/RvpJaFnq1eKNLWsRlnu8im6ye8UsWtZgpRn3dRUa8RoWL1sPTVkPL6RrPPOc9MQV6920RobRahU3Fr6n/1sbNh1dDjyv/deNxNj1iw3l71150LgA+FffvsvuPToS3HIyEPw2Qc+i87eTkecRGiiJhw7/lh8etand0rmi+0v4t+f/HcXvidg1NBRuPioi3HvonvR2dvpCU5elN5DFJJDyEwXL9BnaVPwhCPiFM9P/1cqwsfRuLc7geIm/13G0IGCXNjIYjPmbULcekxextG10aLHrX1YG/G0NS1L7tWyvHHh1Y+HAyJZKsStE3L0/o0AACAASURBVNZ8sp56VvW4tsiTyIjUU8uStqlklaAn4mffJ9gVxoHMj9dt5kmfCBnHfTmsaRiGtwwHg3H+Eef7He9WblmJ7v5uENwCN7fNvw0EwqdnfRojh4ys90gOCpu6N+GfZ/8ztvdtR8eODjRnzXjDxDfgk2d+8jXJTaHcnGVfxJ13urW3u7qct9Tc7F7Csl63Jjr5ncqKth6xfNfjqPq4Jk2bsZ3yNoEgq573KbJsjoH1gDWhW8NBk1tqt7UUCVtDQ98j36Wu9pzWSeqmjYVU9MNGDlLGlHy3Qx7SlwLb/lqe3DdmjJuK1tNTOw4vcpub3U52lQowfTrwQRVqXL3ajYuP371JP9977ntYsnkJ2re3o62lDU1ZE9q3t/spWRL+ndw2GV8+98s7JZuZMa99Hvqr/QABG7s34sElD2Ld9nU1GdsAIpLyY98cssp1KFjgs8kraoqYEDQFApYQtJCVEIrIEFLRZOfPQckSvfIcTEWCGBv9C2+fMkXoRRkpQgeQzHIH4MP+QprikUuo3pIwAJ8UV7PyG4VhBWn3KMlOyZJISJ7nvg0J5JeTleEKAL4vvbGhvHgxygBE8+N9/xUzDCTaIH3pIxGqn0cOHYn3HPsenH5IWEV8R/8O3PDYDXhl0yvIKMOE4RMwrnUc/uL0v6ibpzEQNnZtxBcf+iK6+rp8PYY0DcGpk07Fh0/98C7JrIe9tjlLiTr41a/cXuE7driXsiTB2elNesxcCCZF5vKpPUybia3RyDjQpGnH5UWPlJeq5dlrgFoCF30t6Vq5qU8tSxs9Kd0aTTMTvQQyJm/bQbeZ1dO2q66XPmZD9fYvVTd5LlLtI5957vIusswlzLW1uSlqgFvlbg9gY9dGv13n5u7NjriK+cTy0h81bBTec9x7dlo2EeGNB73R/168aTE2dm90XrAicvEco5CvIlA/hxqBPN2N7kWvs7/15ifug3wY3Gd/A1Fymb5PE50mYdHBk3ZWjG8jcyF2Diu5yfh7hsyP8evxaj0MIWHsSlZxc9DVOL/WTcjSe/uFp62XZxXd/E5rWSVMzdP1Ud5+1GZSXrHKWwUV/zxoUtcbxMjvqC9Nn2aU+aETCdP7sDzDGwoynU7rqcf6t+3Yhp+9/DM8tOwh91wUY/srt670UQoxRl8rxKirkFsYqK/ah41dGwe+cTeiJPTXE08/7ZbclM1dLAHql7uNnNTzNu39csx6dJao6xFdivTsWHfqfEqvlC5nneU8xl/+0o312nIt0VmSTpGs1UV76/UMlnqyRN9GxklqtoM1XESezXrXsuq1ma2PNfRS9SNyc9Aff9yF6U88cY/tH6DDubJISMZSF2BEywhcevSlOGXSru9At2XHFvzv3P/Flh1bPPHpcK8fg3Zvd+gQeLTQCzjsYMbxdeJJ1VYQYQpasbBKRplbda0gXu0Nai9YE51kiHtPHkGWGB9C/pp8/b2iS2FgyBi0lCPj3vJd66/D90KAPvSdhWQ5b3ggQzWr7UspS+tohx2EXH29C0MKCO3o9eTwPWo/CtvEWgNHIMM5kRwoOcpoAeDrubF7IzZ2bfRTGrWxJf2gjaddhW/vok7jho3z4f7XCyWhvx5YtMitILZ+vZtnLtAvcPHgtJcoaLR8qfy23rjOArdyUuXXkyffNdlp2PvqGQkiZ8MGF5nQJKWXexU9tc5AXL+B6mDHpi1RNtIxdT7V7ilyTdVXH08Zbiljwx7XpG6NN30N4AzFBQtctntrK3BaMjJXi/Xr3U6Bee7G2y+5ZMBbZDzar2lOhJZKC849/Fy8+bA3D67cBLr7uvGTl36CuevmOg9UeV+pFy85Fgjkro55z1B76YBPxhK9XYVCpMEvnyqrmBF7otNZ6/5TEbMmYW1AEAgVVFBF1Q9LRNEB78hnPqwsBGfL1FnieszeGznqWi9D9DMk7OuHuC91Jj+gDKWMYp0oRA/AiiALQ8uXyUGWNcykrtGqcaI+B4NGX5MyzCRaIVPU7PCA71sO10mbvxa0trTizEPPxG+W/QZVVNHW0oYLpl+AGZNnvCa5O4uS0F8PrF3rVpgD6r/g7bitfrEPFMKWF7oOZwO1GfH1CMYmxaVgj6f0rKejgCjeClQbC7Yt6pGw/i3HrJFhx79T96fqp8uWMfB6iWupvpPf9eaVN/LIbVvIfSJXD3foftb9oOvc3u6GdlpbGy8bC7jpcHfd5TYbIsLmNUsx94S2uqSsPTa9e1hT1oQzDjkDb5/+9l3O7u2r9uGnC3/qdlkrxpw1GafC5zWEx/F5SdDSBK49Rx361h4fuJiaVWX/m4kjA0Gu1+PceltTnawmZOmnqoEjIpRseyH3iISBmsiAvtbnFDDHYX7fFIH0/W/Az5/3dRMjopBjrxfClHXqvSGiDCkiikLYOsQftRuUISB9SSEHQ9pFGx82s9+TdNGXvv0pTpaU9vL9VNSXiTG0MhSnTjo1yoLfWQxtGopLjroE1byK/rwfY4aNwVunvXWX5e0qSkLfU+jocF55lrnkJCDtzdmXtyYzm0ilCdd67anNS1IkruU0IkRLSFZnnXEtBNjIg00ZKHpBF6tjyrjR11q5tm1TOuvzKUi9rPyUQWQNGd1mOiFOYCMSti/tMZuAmBrysP1oIxcrVwL33ed2eDvssPr17ugAnn3W3593dKDpqWeAOoR+9LijMaxpmM/aFgxrHobLjr1slzbQeGT5IzhuwnF4ctWTeHTFo+it9kbeFRh+gxEfOpZV3ooXuPfS4a6XEG2UAc5KpjQVAgGJASCocjUYLVkhV+a2F7J8+LaAJ16zRrkQk+zeJtf4czn7ud1eFgeiEkKyG8XkyN1+5v6/cx7C6kSxEVQgp2CI+eI46OHJHcpYQEii0561l4WYtLXR4g2Iog9kfF1P7fOGlu5LQhhO0P+d5R5F7KJDtEWtbstiXF8W85FERzHqfrPsN3jr4W+taavBom1IG6468apdvn93oCT0PYW1a90e5ECa5CxRaIKoF47W5K6na1m5+rwcF6S8/ZRnbo81Wn+93m99fYp46nn99RLdNOnZdqtXTsrjt8aFbZtU/9SLXqTaP9UP1kjQu62lVpdL1cnWV99fz/BbtMiF4BsRui4HwLiuHGetDOUs3bwUSzuWFtcBMybNwMVHXdxY3iDxwroXsGH7BvxyyS+xZNMSPLfuOXT3dQciKBYXYbiEK52l7dQuCCY3m44Y4nAXKcLhwvtmjhaD0XK991hAiFJnqDMYFa5EhKZ3WdPEKAlZBAqkW5CnnhteYwSo735zF9FRVmvLOMwlJ7iEsiw2YKJQvDIIInlFBEJ73X6+Oqu54AU5RrKKayXT3Wevq/4QIhWZUdZ90eaUh/b1nxxk+6hFkWQofSl66L7U5UdGUEHqYKCrrwv3Lb4Po4eORn/ej/Gt4we9c9++hpLQ9wR6eoD58+OXcSoMq1/wqWlTAiEFGRO3yVup6+WYXbhFkPKqNbQeA60LLyFhPQaeqqvoY71RgdSv3lK0tm3qkaee116tppPY6rV/PQ9d5Ntd4VLGiZZlz9m62mERLVfaSuTUW3ZXztl8A5G3erXbKnXs2Fo9dFvodlO6PLn6STy8/GH/wj1z25l41zHvwuihu744zbKOZWjf3o77Ft2H1dtWI89zrO9e7wiKEbw0E3L3CVtA8NzB0ctfQuySra09Pr83t0wdYydDhlC992xlFW0fJZ+hIFlyRC2kXEU18khdE4es7xpZHIgtkkXKMNHGipCmbPpSkJqEmPXYvs4098f04jTFP1kGVQ876O1g/b2F1yz9YGVJe8r9OvFPzzjwCYNSJ8R7ukt9Zeqh7cuccmS5my3gz3EwAHVfSjvY5yKvqq1iAWzu3oxbX7wVk9omobW5FUeNO6rmud3XURL67ka1CjzyiPPOdYKbvGTrhdstCaeSxOzLtpGRkJoHrcvOspicrNFhjQtNdgI97qv1tPXW+qQMF20Y2D+tv+inidsaNXahmaamuJyUYZJqf9HZ6qP1lGs0UrIs+cqn7WNrZKTk2b7UBp7uF123xx8Hpk0Dzj0XSTQ3uy1/u7vdzINx49wceLgpapt2bPLj2ciB2StmY/qY6Zg1dVZa3gBY17kOt82/DS9veNl7sjnn8Z7ZDE8UellT2Y1LL1XqTsATtQ9JU5jPrPcnj5LJ4MhH74qmvbsqVz0p2VXLAPhP2W7Ufy+I2ddD6cbEQBWgjELmtuiSU+Rt6qz6yHApyJeJAzHp/c4RSNeTqehcEGO0X3lhcMhWrLqOzhkusvyL8X/ZgU4MJL0GgA9rE8Xr4VPoyypXvQevSVWO6Z3bNMn7KYnFPvLI4acB+miOWsJWL5aj92pH3FTe0AEDr257FbfOuxVXn3L1gJsC7WsoCX13Y+1a4JZb3AIxQEyc+qWuYT0xwC3moglMkAq7Cyw5WWLShoAlK7nOepr2eIqs5ZgNL6fC9HbfbU1iqSln+pglStltTetiIxZaX9t2ts20LEvEzGGlvnr9aGWlykrpKpEEOabbyxpAKb10n9p2lPs2bXKzC4YOrdV78mTg7//eRZXuuQf4+Mfdc9zejl9vfATPvfqcf2lrQtkVdOzowE9e+okjc3lxQ225mTuCjZYPlRBq0X0VVCBzlKPlUzm8uJ2SgdSJyU/5Aof52DYxy4dlFdFpUvKkTnFoXDxBYkdWMt4u4DyseCZhds7V2D6pMek8RAm0V6tD39Jecq1erEYImKuMalb1dfXtBeVVC5ELcRJH7SHEmOe5b389bs2IDQMh9CpXG/YlZxwMFg55ArKQjNbLk38R3fDT99jJ81MnkUXj7b69lNGoN57xbUbqGSO38t+yLctw+/zb8Wen/9lOPuF7FyWhW/T3u6zfeuHJgbB+vfvUXpj+tN5hygPUXq/1DPWnhT6vy0l58nrRGKuLQBM/UDckGy1JW09Winjs9aJPyuu0dUzprq/ROjSaNy5kattZhhB0G+q+1KSsZabmjKf6pV6fSr103oQme9uXun1tPXU/33+/I+5GW7Eed5xbbe7++4G770bfcceg+5yJ/qUo3mJLpWWXN564d9G9eGHdC646aqzTZ37LtKmMPIn6ub0U1v3WmeKSZa5XG/OkyIjGckWOXdRESMyPEUt0gMgnnKXmK+ssd58oRrFeEvoVvXTZfi65EG4WZAGISbqYE08I87ajufYodCy8zZrFYBRp6YiCX9BFhdfF0/WL52RZkKXrklFE2hllbs93PZWuiKbotpKFfRhusxw9Rc5HBwh+sSIvP8uCxw/3vMgmK9IOWhcZy8+pGHtXfenLRShXzyO3i+fsD3htk+8ONGzd6sKT3/iG+y4bYQwW7e3Ad74T9sMGYo/ZemZyXmCPay9NX2vv0denyFIfq1eehvX0rd42zCweqdVPy7B10tcRhQ1GUqF5XY6tv75Ge7L6fKP7bJRDn7PHUzpZWYJUX6aiCLY/bBtbPWzUw3rrqf6V/unqcs/11q1Igtlt6nLXXUCeY+H6hZi9cjYqVHEvP+SoUAXvOPodeNOUN6VlDAbigSlvMAqfI5wHOY9VvFEhIfGi/RaaULKUZyblgRAIBiExy4dgEQgt0kt5j14mhWtBhe66PkBN4hzY6aHDy6InKIT79bCDbg8tS4fH5X4fspapb3lo0yiRTn2KXtGyqmLgcLEBSmEUyVrvVa56T9wvwSr1KPS37S31tn3pIxaIN17RXnWN3uq5AMNvOxuN46OYVsfsEyrr9aU3umRGhDwT5Np3R/+O2ud3H0ZJ6ILFi4HPfQ74/veB5cuB664D/vu/d06GeHvaC7MvZEsWmrTrkaslFi3bkpQ9bgnfes72mkblpo5pffTvVBvUI8K2NuCKK9x4d2rddUuE9dq2Xp1SsqzeKbJPGU71YMtIZa6nDBvbP6l7Uu1Wr+zUVD+55rbbgM98xm0C9NJLYSc/wYsvAjff7CMCXIR1Tz74ZMyYNANNWRPOP/J8nHvEubvkveScO2KQFzJQ8+KPVHZM5z1NAD6Rzb/wKR7r1U6mJm7tdbui2BOC532ZY67lAVEI35etjAUfVcgDwdnxbB+SlpCvdD8Cwftyi/M68SyC/y9Cvg4+/C11zOLzkmRndYrqC0Tjz54EVcQgQwbK44hItNyqNaSgfufw696LLN2eti+jcL3SBYiNIW0w2b7U32v6jYMhqLP15bmbv2E+fjj3h9Hyv/s6ypC7gNm94CTM2dvrstV35n77W4d85QVox9MlTFupxOPLlrzsH3NYXc6SnE6wqpdRrmUBtQuU6PFY6/XXMwoGkmXD2KLXtm2ObPr7wy5kuo4p8rP1te0u1+txbzsGruulQ9s2+U7kat103+nwuM42t8MVej13u/SuNQj0kIg2MKRMkZG63xo8cq6/P0SPbroJGDbMEfxBBwUdJXeDCFSQQktTC37/uN/HJUdfgtFDR+/yBhb3LboPT6x+woeLdUjXLunpCY/c2HeFKtGWpZII5ZrLvdBl1bpokRWZmoZ4q88KKt4786uywZ3zYdoiDCxLtvrruNCrWMbV7kIWee8c5lrrTUiEOOxqZr6ORcgZQE04HUC0op3vdpDf+9wvdkOIw8iZWhNeh8WVdyyJa9qj1Wuxg+EMBiF6DnUmivdDt4af7h8/V176AIj6KecclaziQvhQOqmyZLgiq2S+Pj6xDuG5qHLVh9hlaEMn68mQEuCiHDJU8syaZ9A2pA3vfcN7B/mU712UhC5IZVAPFl1dwNe+5jbT2LEjfvFqEpNj1hvUn/q4fclrckh54vrFrwlAl2HHxfW51LF6eqUS8gQpIyL1XZOfDG9Yuakwtr7f1sXWV5OnlZWqf72kM1u3lIdqE+3099TY90C/9XHbhzujr5Yh57u63N+NNzpSHz0aOPZYt5XvHXcAAI7eQLjyhRxLpzDahrShbUhbWr9Boruv24UwC1L11ZOwcKGbn4ZFYTGYaP4xQrKYHm8V7zcKARsPTOQLqdhFWUQ3G+qVEK/29n0EgILeoov3GkUsB+9Zh49Fhiwo471slYBWc5/aBc7ry0ZP5fFKhEFn00chbKWbbz/ERkdOYZZA0XChXP9fKR4SAFyoXsb85Xoh6po54qySHAu52jCI+lL6ls1YOJPPwJf2yBESB3291LMj/ZDD6aXH1XurvW4HNQNmxtce/RrWdq4Fkduy9VNnfQrjh+/enQ13FiWhC5jDYh/aCxsIfX3O23n11XjOs/VYrcdly079tqSuE8b0MX2/DeE2NblwNhFw660u8iD7elsvXe63HrGWBwQi1nWybWZJ2XrTmoBsfRq1v9ZPPOZUe+m2T9VTX5PyrsWblusG05c2WU4MRLuqn5y3XrrUwX6mFtzRejZqf2vY6Wl8QDye3tLivPVCXgszZrQdjWOmX1jbDzuJ2Stm47FVj4UV1RThEhU7VOk5zeLJ5fCLuXjvVYhJPFnAE3ZOeZwwJoRD8LKAOPSrk7Fck4XEMvkniVl+m9Tov1yc5GWXvfUGhMp8z/Pc5yXoNcozylz0QFk82sv109yKsLCOKHiyy4KXLXu/ixy/OUzRxqKLJ1VyxkTOTj/tjaf2MJd2jgifYmPM1yEvvF/Vl5Jop9vNj32z08l774XxlFEWJ+ypCImPLqi+1GvUe0+cEfWljhRJm4lOqWWMvznnm1i2ZRn6q/1gZvQ299ZcszdQErpgyhTg7W93Gb7MwKRJwMWDWA2LyG04AtRmQctLW4eQ9UtXv2hTBKFJRD7tS9t6bCJXjlUqLrv57ruDwaEJzi5sosvUZGDL0zo1InOgdqEYTaC2TbQ8LcvWT9dxZwnderN6URxLqNbYkDazeqfkynl7rX4uBLaNU31pny/dd43azLabLqe313npIqOnJ2qzNdvX4Y7n/wczp8zE9t7tmNc+z7+A+7nfkx8QvD5NspId3dnTic7ezph4CJ7EdThae+OchZCsDkf7FdF82yOsuCZ2GquFQxQxSBhd9PNOOAVisR54lGhXkIFOqoqMAwllyyOkiCrLMr/EK+BC/55kKm7c3Ge6+36DL9uPO1PQK8syb7xE4WiErTxFnmxv6uUCPnNcjAtZylaI0PcjElueFsIqqIAz9uPk0k56elndvjR19X1GQZb20kVGNKwg4XI1vFEoWNOX9rnQfZkhZPPLsMaL7S/iHx79B5x92Nk461A3S2RD14YoItNX7cN/Pftf+OOZf4wRLSOwt1ASumDoUOC881yCFpFbXOPoowe+T16s+uUNxC9b/VnvmA6Fpzxaud7+rvcS17Juu82t6y3lpMhOfltPUuuWqkc94rOZ9ZZYgNrhCNs29e7Vho8ux47Pp/S1si35WkK19w0meqPrnyLklC4a1tiQaXVa30aGmS1D9LB1l0/Za0CgpiF29nVi+ZZObO/djn7uR8eODn+Zz1QuyEteuppUKlTx52vGhxk+nO4XY0GxxKt4tYzo5azDt56IEdZiF0L3+44LCfvmDt6+JgQxUvynEGShAxAMBr/JSiFfL+CiCSna+QuOePXCLa5buGafce2R63F/vcmJlKeXUvWevpIleurs7iA81Nu3ExAW4KF4RzpvLDGidvCkKWPYyPxc/ApXfFQBDB+JqdeXmoD9UIoiTj1c4o0JFR3RwwoAor70W+GqdtIb5uiNavRud1t7tmJr71Y8sPgBzFk9B+cdeV5Yx6Doy5ZKC46bcByGNiXWeXgdURK6xsiRwPnn7/x99qW8M4SpX9DaI0695FPEp2HD2JJxv2JFKENgvU2ru9bDJnppOVo3XX49crTXp9okNaSg623JScu1YW9rBNm2s/Ia6dvIgBLUk2XLTBlntkz9bNhFY1LhfZskqXXSfSn31OtLJXvq1gxnryLMPmRT8KwBHw7VmcpEwasBFYQiahDiMW8ORJshQz/1R4RoPTchuoyzmDTFQyteyH5jFPWyjsLXHOTqBC8vk0NyVE5qnD6LicJ7glD7jktZQr5SR3UPKIz1a2PCe6sUoge++yjUMxxEtHqbnjfv64owD133nYTefRkc5Em9o21nlY7Sl97IQCA+KbPK1XjuuIpeeI9f7dPunx9pB4qXrNXJf1F9DLFHBo3oWpThV+4D4iRIBB38kIIMy2QhUgMG2re3o317O06ZdIo3CMXLr2QVnHHIGbucMLq7UDs4UGLnUKkAM2e68Uf94kz9NSJr+bReoP1rNB3KvsRtWNeGk1M61TMYtLGRCpWn5KTINAVLTvp+e8zqbj3Wet61llmPnG1fpe5JheStIab7r1Ff2nrq+ti2122Zag87/GDrrnXWMnWSZsJIGd8NHL5ZidRkSAjTg4qXvN9XuiAJuccvz6kI2hJUlIzF8XXaY5PxVT1u6sduxSgQ8gpMEwglU/O8M9SEyDU5+VBz8U9nYVtPWEcetBfqIwGmPXSUINrjuyA4P1Qg90MZCARvUOTsVnHzCXXakDBkCQQvN9WXPneh0EPrZPtS6l4zxq+MqahdofqxuFePuUu5euhGr51f05eqLSRHQS9Fq5cMbtSXNYaj1hEUPYOCGZNn4KypZ6GtpQ0zJs3ArENnYXjzcOxtlB76a0WWARdd5Bbk6OlJZ7bbF2k9j3OgJUJFdspzk/s0UmFv+W3L0cdSn5ZArfwUGmWg19M5RWj6nNZDt682MupFCOqRpdTBLhyjP1MhbdsG+nr714hcgSC/EamLHrpMkau97YEMhHo62LYqjARHtOElKOTrvRdNYu7S6CUv5OM95+K7hfa6ZWUvITR/vRCNLGoihMBqD25Vnl2f3b/sE+Shycenw3EYq/ZrqrOZ117IlPs1YdS2eZCts8a9PArl6x3YfBk1XRjmTet53H69dpsMBsS6iT5Sp2JJWkk+82vAS9tkqs04XiNeyujnfm88yXOhyVD3pe5byUYXuT7qI8+cMnLkedGREP1s6bF3aSdG/b6MhmhUX0q9bJTjgiMvQDWv4vam23HO4efs9ex2Qemh7w5kGXDkkXFYWghCk4QlC01EdnqRvk97UZrAZd50Sm49kk4ZD6m/1PXWMEmVnUouszrZOtXTybZfSod6OmtdGq1kp2XVi35YTzZFlKl+rqeX/hsxAjjmGLfU8EB1s+WkDMDUM2XraGXW0624/6BO4PDN8csUBL+5iN7wgh2LuszpgvQ557CaWkHAdvUyeXFHm6sg3iFL368XBWHmsNJacVyvfhaqybVbqvqT8C9/v/IYw69s5o2Y4p/oz2CfJa6NCLnf6qSNjkiGfGf2iWR6R7WIzEUXDu3n56VrQ4TDdzBqSVHpJ30pZCbthNyUVfSlj4oYmX5TFlb9JdewqqvuP7m/qJs2wuT58vVUddLt5vtNfff3qH6q15dC2P651c+zejYmjpiIccPHAXBh9suPv3yfIXOg9NB3D4YMAT74QbdAyiuv1CZnDeQpWuKy1zAHzyw17Ut+1yMPgRCTzY6uN7VKCEEbDlqWLqveNqVWH+1J1/Nu9cIp+ri+RxO+LcfqodvKRkSsZ27b0UZKdD1t2dZYqWdASJtXKi4Jc+ZM4Lnnwj4A9XTTMnRbyG9bR1tP/VvLSbWVuv6IzcCM1cDiMeqlyuElKQuJAKh50cvLUmc+R/OKNdkV47p+jXEZv87Je+Q6wcqOpdrlY2VpVCnHryleIT/3PAplJ7xgTZD+OobL6AZ8VniFKxEp6f3C/daoHEhb2kfKzyn3+71nXGSaF9uDakPKkw+zL8O3CULI2euvNivRG5PodpF2tfPdpY3B8Iuz6HH2cFl8n8gSYk/1pejvh1ZUOF33pTeyVF/KGvJZ7mYEVPNqNAUtMvrYzFsvnrUo2iTZ90VfSVJmhgwjmkdg7PCx/vrzjjgPb5jwBuyrKAl9d2HECODaa4G//VsXerekLS/g1JipJkUNISchCU2+llzsy1p7p1oHOabJwpJDPcNAHxcdRIae+tVoqEGTsByzww8y9i/Z3fpaKytFdnqzFb2xifzWWLdlJgAAIABJREFU9dXt1cig0saQ1lu3qVxrDQRdL+lnyV5ftQr43vfi/rFtpfs2pZOul33OUvW0srIMmDAh9KNN5uzqArgTeZEFPHLISHT1dqG7vxsMdktjEjB++PiaqU16rBgAunq7sK13W0Sc2svXnhWAsPJZ7nbVknFVCQ/7tbwVqfitXhVBMxhcLZLbcvJZ2j47OwNkmVlZmUz0iUid4QyB3NVHktLsWL5eeEUMG+256+1FicjXTzLEhVzsWDA4bF3q91UvyBEMXz8m9oZBFNaWeei5IjjkUUTFDktQ5vYOb8qa/I51PkGQEYwTIu8F6yhLeGaDfnr+t+7LyDumsK66NhIJTh95LqK+LIYHxGDTz5PtS/+M5SrpDqEvq1zFMeOPwYdP/TD2F5SErtHX59ZxJwKGD3dz0XcGK1a4F6UmIiAmQEv0KSLW9/b3x6QhY+hyTBsB1su1npkmIe2l1yNg+UwZGnoqlZYj3wXWq5alR61eVoaWoz1vTZaSxa/l6Xu0PjoCoclQGwECG6q2eqTaS5dp+1Ku030pcrWeKVmpvrTRjcFEgaSOsuWsjmJMnuz2MWhtrb0HAO6+G80rZmPUUPdivOzYyzDn1Tl4Ye0LfoONicMn4q/O+itMbJ2YllHgh3N/iIeWPhS8SA6E7l++he56HFZI12e6EwJhFC9/eUnLFpwiS64RD117p376le5yDuFwSwI+UYsQiJLg5pYjC0uKCqkD0XitbDAjHrt8CgnllKOJmzxByth4xIscwvdC0Hp3Ob8+vTz+TJFesja53mddb3AiZfos9WJaXrTdqNoFTpBXc0/C3iiSqWBFe2kDyifYpfqyaO8qV6PpZD6qIcaa6vuoL4vr5R7dl74uKtIhszWQFX1ZGGnyfO8vKAldkOfAY4+5DVmyDJg2ze0LPWbM4O7fsgX4z/90S78C4cWamjdej9AEluBSsuR3SpY1JrSHmCIirV89PbT3aq/TdbKkrM/ba205oqNFivTqlZOqUz3itORv62f1Si3eYutn61jP4KlnvGgP29bffqaOWd2sXtYYGTEC+MQn6pM5ALzznZiFd2IW4D23Mw85E9c/ej1Wb12N5kozPv6mjw9I5kENKlQLL2oAcQhdXsbiKZI776cwFd6dJ1Y/7FobYvVjy7JVaTEE4L1/8YCLMmyCm/fkhJB8XFiRkquAJxvAGRB5Hod3/fi0Tsoi5VUj3h9cL3qSK0PTet3e49Uhd2lbz72GhBWJaogeejzcLlATJbnpvkRoE0aYiw8gav8o09xstSrRA62zN/LU3vDSZ74vqU5fqufCPxPqmfHJl4m+tPXc11ESumDpUrfTGuBedsuXu99//ueDl5EiLTvubT0vSyopb7AetNemy9ZhZq2XfNf62DH5VD2AOARejzxS3rE+p2WlDIQUkWod5YUmutg62ax7af/U/Gxdfgop40XLSvVrSq6OKliirWe8pHSwBoLVx17XSK7OTWhpafyMGehw+l+d9VfY2rMVzG5hjcGgpdKCClXCDlZCQggvcQDx4ikFkcta25qktAfGYD+mrQlBj7MLoqVNoWSx2hENgag0sdisbS8LimjkXv0ZLJdYLypkFGO3TdTkr5d/1mDRc7y1Z+vblJT3SfF86mCPxPXQe6TrNvMkJ82k6ugjFAjyrVdbYwiQkl/oraeb+cx5rpURlYV4ERxTaCQrQ4Y+7ov60mfVp/qSVX/tRygJXUNCmxLWtqHmRkgRhH5p6weu3nSzlCeWIvZ6L+yUd2fvEUNCe+v1Vnazmd2aKFM6SdvVq1sjnVKEr6fo6XbQutoERHtdo+lm9nqtZ+qcbq+BZKXu0+dSq8dppIwq3W6pdm1kJKQMhdbWnYtCGfRX+3HD4zegvbMdQ5qG4FNnfgrTxkxreM87jnkHNnVtwpNrnqwhU+/BKg9bPhnxGumaUPT+2N6TU6Sn5yz7aV1CqMYLi8LbQn6KBKP5z4Vu3hs1Y92yYI3/LbozRXUT6OQ2sFooRcb4izaKxvQVonqoujGz34lOPFEhM845JnErozAMwPCGgV3IxmelKxk1HjbUSm0clqqVMWt9n51SKG2mnWVpb0kmjOorslRfRoYWap8L+fQREd2X+xHKaWsaOiS5s51JFHuCckyfF5xyCnDiieGYJtfUi7recqOahORFnpIl16US5FI62Cx4rU8jr9bqIN+t/o1kyHWpMe96Ie9Gdaux3JXRYK9pZHRY/fT19coT4ylFyPWgZTYyzOS79Jdcb40T3Z763iFDgA9/GJg+vb4uA+CGx2/Aq1tfRW+1F9t6tuEbT30Dm7o3NbynpdKC5kqzS5ZDTNDyAo32Hy/OyaYl8ttVj2tJmQJZi8cnsjRZaY/Rj78zhwVelCzvOarpbNpw8PpnhiDyRFniceuuLZLxQGpVtYyiSIAkAsq/aDlXBJKXJDJpM92uel57hsxHRrShFOnHod38UEXx/8aTrshSRKrbXfIDpFzRx+tPvuDQnqrffVum+pLg9dJ9KXpHK+6JwYfQ1yJLohc6UU92v9vfSL0kdI1Gmc4DYdQo4Jpr3JrwAv1C1YQzbx6wYEF8XT0yGcwDlSIRfSxFKinPW2TVC/vaetlrdPKbvqeeN9noXD39rMdu9R6M0aBRr+3r6Ze6vlE9bdsO1J+2rjZ6YaM21njT5afuI3I7rl11FXDyyY11GQCdfZ1RotH2vu2RJ1q/joVKsuiJ2tXMvrx1wpZdOEY2JXHVC9dHnl0We9T+miz2zPSKadYT9LL0eQp6eCOCE9cqfYWgI5kIMv3cb6trsbqdL5PYZ5trsvTlqnaw/2RTE38vhTp6vYTIKRggPklQeeNeluivoxeqbTgvjCBJtGNE12m9bf6Dbk/py2j4w/Slbk9ffsqIUuf99RyuyTnH0eOPxruPfTf2J5Qhd8HUqcDll7uNTLLMZf5efvnOyTj5ZDfm/uqrwI9/HLKardffW2y1p8cyZUMMIWQdbgbCvdYbs8um6nvFC5UtMzVZaM/Zhodt+D3l7ct3653aY0CtjJRMfY9esjZlbEgb2WVtU+u+22Q2LUd/t1O9UrkFdhlXDa2bDDvYvrRtZSMOui90JEF74aKTbptUwp58t30pdZs8uX64f5CoUMWNh3PVe12gge+7cPqF2NS9CQs3LIQkXdXsjkWFh6rmi4tXnxr3FQ9NL2sqY8sSbpaw7bjh4/Chkz6El9a/hAeXPhiRiPa4JUwspOrDzYiXpvXeNZT3ClcPWQ8cVGxUQiGjXUjFRwWKzG+/Apr81lPEVL2lrRgcNsFR5UvYWPeNrlNE2oVxVeVq8MIRwvUSPfDhcQqb1+jIg+zcpiMfXi81dBBtSKPbXLWZPBd6tbioL7MgV9ba1/908mRE7nX6UmZREBEOG3UY3n/C+/0iMvsLSkIXNDcDb3mLG08kch73oYfunIwsA44/3q0a19bm5hg/8ED84k15nCmvC6glbiEBfU5fL2SiPzUxaKKyoXBNEvXuTXmXmpgajQtrotTT+lJkmzIMUoRYLylQXyeGkpav62R11G2n297W1cqT9tekqZPWUkaL6KHlWNJOrTWg2yNltFjDTZDn7jm/8ELgkEPwWiEvW0kI0y/pRjh4xMF497Hvxvqu9WBm3PPKPVi7ba0ndyFwYkJWCVnffm9qgntRF96YhOJlPLuKKvwYs5AYZ0DFvfy3923HXQvvQndfd7wDHAeCtR5etI84h+xt0UsIPaq/dCWFxV8YjEpegSwYI2QkRAvARR6UHL/MLYVxaO9Vc/BQpQ1EprSXDhtLH+Wc+wVePPmKYQAEIlRb0goBVqgSxrSVASZyhXSjvgSFTVEK4tQGoDWkvDy1gp7tSzFwKlnFE7Guq89VyFAzXTGKbGhDigiT2ybjfce/D1NGThnwWRZs2bEF/zv3f/Ghkz60V7dPJRvi2Z9w2mmn8Zw5c/a2GvXxwgvAf/yH+5562QIxmesV2exLW+5tNI4MBJLq74/LkOtTpJLyOPVzIXPrrSwrx5JdSp6WJfdb/TQJNgq56/nZ9QyOpqbGBJySlYpI6La3BpqVJfK0kabL10Sc0kvrxxz3pS7LtnvquZDyTzkFOPxwd2zmTLez4GvEU6ufwqbuTXhgyQM4ffLpmDBiAmYdOgvNleZBy7hj/h3IOcec1XOwoXtD5IW6usiHMxx0uDiVPBWNdSN+gfsxUpAP12svVi82Ih+y6Ip8rxmH17K0p692KdPDBjJH3i+kQmr8Fghz2YG6iVmSY1Azd18ME9FNGRvea1VRCGk70S9aLpdD28v8bm24SRl6ip/ORdBrAei+9HqZBYhCsfFNslCNkLvoJfUgkJv5oPTU0wvlOqlfzYwD05fjW8fj8uMvx8kHD244qrfai/969r+wuXszVmxdgakjp2Lc8HG45pRr9tjOa0T0NDOfljpXeuh7EpMnA6eeCjz7rPvdyOO02doC68mmPFf9XXvJlpQ0SQ80p1p7sNpL1aul2Tql5NhPKbsR2aXkWp1sfVNGSb3Et1Q/pEL2+n7rqWtZ9l676IuG7kPrSdvEPpFnIyS6La1MrZ/guONir7yvD7jlFuCCC3Y5wx0AZk6ZiZxzNFeacfqU09Ha0mAuu8HDyx7Gwo0LsWTzEowdOhY9eU8Ib0PNRQfgw73KixMyqYkIENwKacV2mRLKlsVRtPfnF6jxt0Y/XPlFmDfaR1ttAxtltQOhrEwZEwihYq4y8iwPZM2B0HS42JOTIiUhIdk0BTlqw8YUPGZNml6ehPcpkFoGtzCNbGmaI/dhd50d7o0WRuxtmwx0X1/dl9JGyriIvGnxzG09EIwtPyyhZgn4rU4lioCwra02XvzUPVJbx6q+HNEyAmdPPRsTRkwYNJkDzuBYvGkxOno6wDljyeYl2NKzpdaYeZ1QEvqexPjxwMUXA93dwMKF7li9l7IlOXmB253bUgShPzW5aXKyhKTJNSWzHhnb7/KZMkLq6arrab3NFClbXYDaRL+BykzVr1F99NxxawRY3aRO2hCwfWnlp4wfK9+SuiDVl6lnAwCmTQPe/vZA5j/7mVsz/oUXgE2b3KYw73pXnMzZCI884gzVI490VaUM5x5+7uDuVVi0eRGeX/c8wMDWHVujzTyENDwJFm2jvWB3QJGwDihJqJcoLA2rPGFPJgVpR2PuIrPwOIXsorFivauZEIMiCKdObca3Jmt/DdUSmzVSIi9dHhNNdGqBHIlARDoYfULzxeckuuGnqvki1UwC016RcVCcizZmKXQWMtaGD+CWCx4xZASWdyx3bV3sQW/Llr6qiUSoIQu5zpO5MiS0UZTqy2HNw/COY9+BNdvW4FdLfoW3HfE2DArqedEzC6yB83qhJPQ9jSlTHKm3tgLt7cDq1eGcJTj7HajNHAdqPWR7jw0D6+MpkrCesr1e/06RXL1IgJWZ8sptPe30uVQdrGxNiCmSr1cfK0++azK37Zs6btvIruVuUU/HVAhfoI06rYe0U8prnzTJeeFHHRWOz5kDrFnjrn32WfdcXnjhwIT+wgvA4sXA3LmYd+w4vNI7GWcdehYOGnFQ4/vqofB+naruhXj0+KOxvGM5evp7AIJfD7xm/rDxroqDwSCQ65n9fGuR4zObOfY+tWy9Jrhf85w5mtcssmQzFZ3AB4I/FlfZ/fYGh1n1TK92JnXyxGDbQUUvRK8cKrlQvGHtPWvdNeFKWar+OplPyowyxYUkE/PDfbtqHZl9REEjowzN1ByRrjaClGBfvj9H4dmxYfuaOvr/Ngz9j0AY0jQEJ0w8ARu2b8A9r9yDVVtXYWvPVkwdNRUzJs9AQxSGnyC5O97riJLQdwfyHHj4YTc2/OY3u9W3NKZPd39PP+2Wl21vBzZujO8XIhCkiEyvUgbUen+aQOrJsl6u9ebrTXlLkZMOv9twdcoL1zpbIrWkmPJsbX30NSmiTxGxrrc9b9deB0JSmiXvlAFl2ytlbMn1es0Cab+U8SYyrXwgNshEz5EjnREJADNmACedVFu2rnNfHzB7tvPi9apzFs89BzzxBMCM+WPX4aHFL2LqqKm7TuiijvLsxg8dj1XZKnesmJvdXGnGEWOOwI7+HVi1dZW/x45ly3QoeeHLi1XCqzKOKmOxOgFMCFzGaAlhDe9o17Q8JLbJzmM6guB1MzvOhaYvkuOUF68jEgz2i6TolfL0sqeaWL1HjVCe1FuHvL2c4ns0jszKYNGyc2XA1PHqtQzblzpxz8+zh1tHvpJVfHnrOtehfXt7/ExwrFeqL0VunuXegCqUcMYgch8JsVupyv3Tx01HU9aEtiFtOP/I8/HzV36OF9a+AAB4YMkDOHTkoRjePBzHTTgO9ZBlGaaPmY5ne55Ff96P5qwZ08dOj6MqryNKQn+teO45R8533RWSvc45J/1ynDHD/f3qV8D997vtVq1nXs+rBmpD75rw9LKsKRJuRHg2i1onmslvLUuPp1vdBJYYtWxNXFYfkd/IE24ky3quqelt1tDQbZYylvRnvaiGrauWab1q6S+dF5EyXoSo6/WlbodKxXnZZ5wBXHYZGkK3WV+fGw465xxg2LDaa1evdssgb9jg7zm0gzG6e+c8kHWd67Bo0yL/uzlrxrDmYdjeu71QKcejKx91RSgPqzlrxqxDZ2FI8xB886lv1nig0Tg6OTmemDiQv2y5qb276DvCymMCvwAJOSLyZJDDk4bo61XQ67QjjhZU82o8F1uIlYJHS3mxeIsQYx4yxauo+sVqcsp9wl7knSryB8Fvc6rXKLfXCrnqLUr9cEURgZCISSqc7eUpg8En7hVRAz8VESGqQRTL17J8FAXxDnzeMCvKFx0J5MfTvQFWpy+ZGdPGTMO1p1zr9zJfsGEBZq+Y7ds9Q4aVW1bisVWPNST0lkoLrjrxKnT1daGzrxMjh4zEB974gT2WEDcQSkJ/LWAGbr0VWLs2vFzvvNNNeZs5s/59b3ubI/OHH3abucjL33pomgCAWmKQY9YbFyKwpKqvkft1wpU1HjRh6E/rwdpQsECTsZZVz9Cw92p9bf2sXtYAEN10DoKVYz18q5tuk3rEmzLAbF9qI0K3e6oN7FQ6XU9riOi2J3Lj5AOR+fjx7nnt63NyWlrcgkiWzDs6gGXLXIh+zpyo3DNWMhaMGTyhd+zowC8W/QKPr3rcH7vgyAtw2uTTsGDDAuScY/329eiv9of9swuvdduObfjxSz/GOdPOweiho9GxoyOEZWUqUkEWQuY6CzwKFVfhk6pkepueaiaJZtrLFyJgsFsmNVPhcgoeq2wJSkR+fXHxML2jqseWufCa1RawRAVpy+PA8Ti5kK33oIvvUm89Nlylqh9T93WlzG+QEkUQGH4tAWkzPY4t49oyBKKngQGFfIloQC1CY7LTKXdj/chDP/h58IU8MRga9SXBbZ9a05d5kX1fGAp61zU//S1nHDTiIPzhSX/oybxQvGbv9uEtwzF22NgBn+/Wllb85Zl/OfB/hNcBJaHXw/btboGYkSOBgxqEFWU7SnnZbtsG/OAHwAknpD0ewWWXuWvnzQO2bg1ygFpi1USlF10RCIloAk0ljWlCsAvXyDVSpsgV40CXZa+X4zZz3hI3czwFS+ucmmttDRlrUNhpa7audq63/m7JPeUJWwNK2k3LShkAWpaG9vDFq7ZT16xeUkdrSIisLANGj3ay2towIP7kT4CvfS143PV2WZszxxmrdYYpWptbB70hyzOvPoPfrvhtob4jqI7uDrz3+PfiyhOuBAB8/sHPY9WWVd6LArs/yggbtm/A7FWzccUJV+BH836E3movunq7MKx5GIY1D0NXXxd29O2IPURFxtFYchH+lfB2FVVPLFFCmNhMeTynXfTS4WTO2O/45b18RvCGlafp90RHMFpydlnlXGVUKmoKFuKhBW9cFH/93B+8ZvXcSJi+klV86Fw8Wz1Gr6e8+c1yit96TF3qUuFizfQ86ETs9nH33j3HU+DEUJC+5Co7Uucg2+dEFHvC6+lzUk8/9NCgL/vRH5ahVUaeGE7Dm4ejudKMa065pmaOeVOlCa3Nrdjau9Vn+x85+khcduwABvI+hpLQU+jtdQvC/PznbsrP+97niL3RC1N7XiMGubDABz/oPr/+dWDRovDS7uysTYIS2ZaY9Tmg1nvX0CFoonj8W6DvT3nyur6aOFNlWi/WevKpVdBS9+t7Ux6sNTK0rJRHrHWQY7LgTardrD5aliV1uS4VGdF9Zxf1qRflqLeoT6UCDB/uEt+uu67WyKuHSgX4m78Z3LVab9Fz2DCgUsHlJ10BHPTGQd3uQ7TFy5uI8OTqJzFxxERcevSlrgjxdCmM9UrYN6tkaGtpw6mTTsWpk07Fg0sfxD0v34OZU2biyhOuxLfnfBtz1szxJBztOibTq0h5xFSb6Q64le88mYnnr8ae001TeMGpjWOk/mpTFDuFy9/L8AvK6AxyCeH7RDOlFxCma+n2lbrqjPqikUM7g/x0PJGrs/vlGukPvfa+Tnjz96pFajRkLF+GEfTQhHjk/ljRVzrL3s8oUH0onznlYTojxVMb/Vz5or2JCe8/4f0449Azkv04fex0fPjUD+P/Pvt/fb1HDNl7C8TsKkpCt8hz4De/AX7xC/cCW7AA+MpXgFmz3PrXFkOHxi/icePc/tKNvHOLP/7j8NLu7ga++tU4jG+9MiE2S+IpEpH7ddhX7hEZdpxZzte734aR7fdG4ehU+Flfq2XY6ITVxxJ+Pb2k7VJtVO8+rZc1AOrJEnkpyL26z0SW1UPLstPnBBMmOCJvbR08me8qpNymJpcRf9557vsg0ZQ1oSlrQn+135NjU6UpGmcc0jQEQ5uG+rBqb3+xPDIBU9qm4BNnfMJfe860c/B7U38vrGoGeDKT8VRPxhIapiKpSmXG5xSSrDSRR/ytf0ePhRm7V0vO+rA9GDW2AKP2GIXjsmKbn1YlUYdU8p1cw2HeNbv4sifinNS4t1rdLc9DFEEIO5pVUOjkIwAUVsvzc/ILPXREQxtvsryqtFeobugnnSSor9PX+GgNmfbm0G/RLnnSpqrPGhllm7s3AwAOGXkIPn/25zFq6Cgnot7/5X0YJaFbPP888JOfxOHd/v76L+9PfAK4/noXwhw+3G1HOX587bWNoF+Ozc3An/0Z8E//5ELy1tusRySaDCwhWfLT+utPe589pn8Pxru099hyU2P3KV3rGSpax5Q3L+fsXPt6hsb/Y++74+OozrWfM9uk1ao3q8uW5Sr3hiumuGATAw4GgyGYkgIhDRKSG8i9N/nlJiQfyU24Nx98phlCMyYYB8dgm25sbFyQmyy525Js9bq70rY53x/vnpnZ1aqvtCt7H//2py0zZ87Mruc9b3ueQIsT7TaBogj+Y/mnOrTb9CSK0tm8tCkSi4V+IwkJne/bH3Cu5tjFHDkH5s0DFi3qlTEHgHm585Q8urjJL8hdgOtHXK9s85OrfqLcnBvbGvHU7qfQ6mxFUnQSfnTVjxBtUBfIjDGfcL9ep1cMDU2/Y1GUD1+81tvTeJr+N30lZK/xLH164pma31a8VqZ5+P8MlINBMXj0UvW8IQNMp6nk7vBfgfvMRVmceKvllL527jXUzNdD96lK17Rbaeem3aZTOVGm2cd7LsoiQlbnol2UiHkJzXGxuBLKZkqtg2YBpf0utdX82iiBxCQYJGIn1DFKC4hUifb8HG6HGunwoqKlAr/f+Xu4ZBe4zJEbn4vH5j0Gk96EoYiIQfcH54FDr50ZlYQE4LHHVOPUW2MeCFlZwBNP+B5X/N23jyrqtR5fIAOoNUj+xlLrKQp0VsHtb8DFc62Hrz1uV3zugeaifU+Eov3H6m7BEGhhol0Aaees3cf/+mjf938P8CVzCdQXrr1O/hEL/+N1ta/2HESefO1aYNQodbtg/M46w969lG7Szl+SSHK1l8YcIK7tpSOXwua0YeeFnZiZNRMrRq/w8dC1LHMWowW/uvpXys3Xp3gpAFaNW4VmRzNKa0oVQ6j1YAGopCuMqqG14WNhtP1bsnx6wxEgpCy+No2CmbKNxqNUwueMvHetofH3QCXlNwafeSmtdwG0y32eayISPhEM7dwCLRI05689P5+5a6+r/7VQ1h++CwrxnnIM74JDSX9ADa8r24pjBjg/7VhgdO0z4zLx8MyHlfE5ODYf34wDlw74zGHj8Y1INCdiYvpEn3N3eVxweVwAAy60XMAz+5/Bj68KjyK33iJi0P1RVETMWe++2/Hm3BmSB0CRp7MbtsXStRcbyOBpQ/X++wQqHAs0rn8vtPa5v7pboIWQNlStFWfxXzxoi73EWP6Gzz80rT1WICMZ6LV47j+WGKOzyvjOqGY7G6urSEeghYd2W1mm4smiIqpg74Mx7RNcLuq+0C6wpk0j77yPMOlN+Mbob2Bu7lwkRCX4eNyB0J0R1yLWFIu7J96NdQfW4WzjWeV9Ef71yGrPuPcDyExWirx8RFe8xsaH19w7lmJkAdVr5JpWLEgqZSpUT9AnVK6VUPUacjBQ5ThXK8N98vzaOWj+ar1ZH8/Ve46KIWRQjajmPWVdwdAhZO2vgCf+aiMHAYv2/Oeoua6KB+43lkL7qpmXItOq/LelbUWRmzgnLhOfQFpMms/3ZTKYfI/HgaUFSzE2RW1Bq7PX4X/2/g+cHqdPt4HdZcdQRcSg+8NopJ5cqxX4+GO6qY0bR8Qb4QDtTbYzg+LPeQ4E9jr9DXwgg+8v/ak14Np9/AvbAnnGOh2lL7SkKv7H9Dfm/n/9vexAVev+xYOBjKZ2n0Dz1W6rLWDrjKJWe47aBUug8xDfkb+oCkAiKvPnq6/z8nwr0Rsbgb/8hWotOAfS0oBHHw1uHj3QeVosPaui7wKxpljEmvo3RmdIMafApDNB0G8KIwsORbxDK8QiisIAjRFkUDxyJUcsqfznCt+5FxIkcIkqrJVxRZ6aq96mUhgm+rkhU785k5TCLs4vMsuhAAAgAElEQVS5EjlQcvHCaGmNNYOSI/enaeWSxpvmUGhItWIq4ny1xljiks95iZC4jwOuuW7KucPX4xdetyKZLhZATBNl8M5Nu1gJxPYmxvKX1VUiGBpDX2WtwnMHnsO3p31b2X/pyKWot9ejtL5UWTSlWlI7CAe1OlsBkGIfl7zFfyyI/5cGGRGDHggxMWTACwropjlsGBUhhQO0hiVQCFn81UqgasOm2jC99qE1YmI7sSjQjqXVbRd/hXHyZ4vz345z8jJFntm/8tx/7mLO2ipv4T1rDVigXLz23AK104nz9CeS8Z+Xdh7+hr+zsQIVK2rPU2vMc3KInlWMm51NleuBYLMB69YR2YsY05+VMFjwT334L2DCDG+XvI2KlgrFgMmQYYQR3xjzDSSbkyExCTW2Gmwp29KBiESp5tZ434oB0eiO65leJY3xetqisprBVw5UVHVzcKVvWxSoKf3RjIyp1kiK+fuM5V04KIsEiXL7CuOaVvpU3BIk3/EAmotWsEWbf1cY6sB9QvXaqnIxL2WRwn350QGNZ6/5uSjRCO+CCYwWWcp8vNEOf0EagBYlOkmHZSOXYVjsMDS1N2Fz6WY4PA7lc1mWcbTmKDYc3YDbi24HAKTFpGHl2JUKCx3nHGNSxvj8ZhKjErFyzEr8o/QfcHvcAKP9bhlzSx9/haFHxKB3hsTErslhQoUxY4DVq6n6vqUFmDnT93OtkXc6KRdqs/l6kEDnBkoLrSfu3y6mDdVrPTr/vmqtV6oNu/sbf+0x/b1hrUHxTxNoX/sz6fkbczEn7V//ELo/zau/p97V+QWapz/5S2wscMMNavg8NZVC6j2BLKs87OJYLS3Aa6+pLZDBwJgxwO23q685B4YPD974A4BaWy0cbgf1lntJVXQ6HSakT0BeQh4AwO6yw2K0oLKlErvKdymeoeJBej3gDjlumfLRinCI11j65NC1HqgWkjIQRP5XeObasbTeqpBQ1eaqhWHVhuT9DaUwrCK64O/t+7SweeekePLefnIlby3C5JwrrG7KuJpWMQamUtkytVpezFmJLoi2NsZ8r5f3XCRIqhIc6DuYmTVT+e5mZM2AxWiB0+OEw+3A+yffh1t2K3NxyS5U26p9Ti8vIU/ZPxB0kg4L8hfAoDcoFf4J0QkYnTK6y99aOCNi0IcaUlOBhQuB/Hyqgp/QRS+wLJNB+eADlZEO6Gj0tNC+9vd4O9sOCOyda8cR2/gX3wUaS2wb6LnYXoyjZcTzz7sHmov/e/7RC/GZ/2LB/1r5e/zi3PwjC5JEXvT8+dQFERMDXHstbXPiBKV1vvhCHW/KlI6LNO0xdTrfhZbTSdSswURKCv3GhhCuyr4K1TbiBVdC6xpPDwDMBjOuGX4N6u31SIhOUEPQYCirL8PphtOK1yjyr8K4Ts+ajsPVh4nAhqs94cIz9yk28xpKsb8wtIqxYhw67hvSBlSDqw3d++wn8v0iP+5dEKgpdXVb7SJFm+f2Mf6aOWv7vZXQtjbMrwlxi8WHwp/uvwhiqrH2OXfm+1lqTCpmZs3EqYZTOFF/AkxiGJk4EoXJheDgmJYxDRmxvtEqo86IaZnTsO30Nt9FlDd60VuY9CYszF8Y8LPiqmLsrdirvJ6ZNRNTMqb0+hiDiYhBH6rIz+9+G0minmGnE7DbOxokYawqK0lJK1B4WPwNlIMHfMPKgRYI2u21hr2oiDjERfGV/7b++wQ6pjCY2jkEGkN7PXp6Hv5jBFogBJr3mDGU19a2PZpMpLjnz01QUwMcOeI7VkpK5wbdaKTPPv+cCtcYo9x2Z9sPUbS727Hp+Ca0udsAUGj0pjE3dZnbnJIxBTsv7FRCrFoP0h/J5mSF0EYgryYPx+KO0QuvMQagFHpdk38NEkwJ+PTcp3B6nKqHDHTIUSvwGkKtIReKbCK87Z/X9g9Xd/D4NXNTjql5TxkvgEa52FYYQX/VNzG+ErkQnrXGKGsjE9pz1B5bjCuiEYJuFqAOhmkZ06CX9MiIzcCCvAUorStVDHdRWhHGpY5DV4gzxWFa5jTsq9ynzC0pOglTM6Z2uV9vUdlSiSPV9P+TM46m9iaYDeaw9uAjBv1yh9HYPb/3sWPAP/8JVFSohquz6mzxXmfGTftZZ8aTMfJYz55Vi7sCRQwCGdtAcwl0HK137b9tRgblqv3PUXjYbjdR8opebO2xxRyTkohU6OJF0gfPzKTPFyxQtMK7xKVLxHngf16BzlHAaAS+8Q3A4aCHmEc/qs/DDW7ZjX+V/Qu7yncRuQwH0ixpWDF6RQcDFQjafLgsy/j0/Ke41XyrT1tcIIxPG4/xaeO73ObG0Tei1dmKfZX74Jbd4oCkvOYN/Yqwc4d+b6+BVMLmTPWGlX52pobkFWMvvHFhZMV5ahXbNMxuykJEu79Gvx2AL5mLmINGspXBS+oi0gCaaIHPfJj6nHGGvIQ8JJuTcaz6GKwuq3IN9Ho9itKKoJf0SIpOwsqxK30WZ2NSxnTIb3eFWFMsbhp9ExiYonqXFZeFublzezxGr8GBC80XUN5SHjHoEYQ5xo8nD37fPnodyNAJY3fihEq0oy0gk2WVHvfSJXVs7Vhaw6vTkaTsoUMqv7u/YfMXfvHfxn+O2vdEyNufsz4piboY5nbxn99uB156ibxnwcsuDG9sLJCbS38bG+m9mBgy6OPH00KhJ7DbgebmjgV9XRl0gML23/pWz44xBOF0O/FlxZekiQ4yIm3uNuwu3435efO73FeIn4i/btmNfRX7sHjE4m4Nek8QpY/CreNuhcPtgMOjkpTYXDZUtFQoueOE6AQwxlBvJ4lk0Q6VE5+DS62X0O5u9zHO2p55f152/15tnxYwqBEEsY02HSBa6RQFOC/EsYZZhqHV2Qqr06qIlwAUcRgePxwXmi74tO+ZDWbkJeShsa0RVdYqxdtPs6QhOToZC/MXYnzaeLx59E1Ut1YrC5koQxTumngXLMbgUakmm5Nx75R7gzZeZwjUmhfOiBj0CAgzZnRfBNjWBjz/PBki/+IwnQ4YPpxCwO+9R+8HMlLCQ05IoOK+khLyhLXQesPaFIAWqalkUEUqwX8bf9EWsfAYMaJrYw6Q0VyzBnjxRUoJiLkwRqmO1avpHPd41cNOnKBHRYVK/iLOffr0wDTABQUUpXj9dd8UQlWV6vVfidCEo8GpX7zN3YYT9Se6Neg++3mrrbPjs3ssItMTxJpi8b0Z3/N5r6yuDP8s+6cy/wlpE2DUGbHj9A40tzdDhoz0mHTcPOZm7C7fTcVb3tC20+NERUuFYoiF96xw0Hs9YuE5i4p2IUCjeMxer1p49+DUi50TlwOb04ZaW61iYBOiE5AYnYir867G5+c/R4ujha6fdw56SY9vjvsmtpzYAofHoXjoufG5WDNxDfZV7sNn5z5TogGzs2f7eMdCcGeoQ1HZU0ItoZ1PTxAx6BH0HNHRRDnaFXbvJqNutdJrYWgTEiiXzDlw553EhtfSQvnmc+c6Erloi9QsFvL+m5vJgCckACtXArt2AYcPU42AgFhcpHmJJtxuMvwmE3Uu9JQEKCEBeOSRrrfRLjxkmSIc+/apEQ29njz24cMD72+xkNSu1lMvKaHe8xUrejbPywyiWKrJ0UQGQwbiouOwdvLabvdNNafirHQWdg8RgzDGsLpoNZLNA0D8pMHolNH4WcrPOrxvdVpRXFUMDo5vjPoGxqaO7aCtXWurxbP7n6WIBCPv2OVxobG9UfW2AVWkRfZ65l6jrlCtej1yBgadpENqTCoyYzPx0IyHsK9yH94/+b4y1lXZV2FRAaVpjtcdR6uzFdW2arjdbqRGpyLGFIMUcwoenfNowPOdkTUDM7J63wHkcDtwrPaY8jreFI+CpB6kp0KE+Kh4DLMMQ4ujBTqmg8VoQZwpLtTT6hIRgx5BcDFnDoXct23zNdDXX0868FrExQHf/S4RpdhspMGtDa+bTGT0pk8HvvlNYMMGoLiY8siTJlGF/x//SMVlAoyRkXzsMcpxf/017Td6NOl+BwtaSdNA7wfSjvfHlCmkAbBxY/d1C1cIog3ReHD6g3hq91Owu+zg4IiLivOpou4Md028C62OVqVaXa/Td6A/HUysGL2Ccv9dIDUmFb+6+lc+7+2/uB//KPkHADU8rmWdE8bcwz3KawEGhsToRDw29zElMtGVAf7WJErfPLX7KdTYavDdGd9FfkJ+n863K3hkD3Zd2IXXj7yuyMXmxuXigWkPIDM2PKNR8/PmY37efGwu3Qy7y64U3dmctqCkcAYCLNxzAl1h+vTpfP/+/aGeRgT+eOcdYPt2XwN1221k1DvDqVOkMse52os+fjzw4zDlVN6+HdiyRY0OmEyUOnC51HZBvZ4U0UaM6Hqct9/2XcgsW0b0wxFcFqi31yPGGIMofVSopxIylDeX4zef/QYu2UVRBFDbXlF6Udjzprc4WvDKoVdwpPoIOOe4ofAGLB+1PKipnN6AMXaAcz490GdDl+MugvBGINWxrtDTorBwwaJF1KttMtFj0SJagGh57aN6cAPX66l6faicdwQ9hs1pQ729HuuL1+OjMx+h3l7fgSP+SkKUPkqpqocUoCUvTPFe2Xs4Wn1UiYZ8cOoDHLh4INTTCohIyH2owuOh1iWDgR7hBJ2OjJTHo1aa94RrXNsyF+7/2Rmj0P/ixfQ6Kory9U8/TT39AHDffV175wBw9dVUS7BjR0finZ6AcypWFM/1elpgRBByvF3yttLidqrhFLac2ILvTPtO2JOTDARy4nPw7anfxl/3/hUej0cpCtQq7oUrRNeET599mCL8r2YEHcE55ai3bydSkdtvDy++7SVLqJJ80ybg4EFiRps3r/v9/Ilcgik4MhDwX0zp9VQTIKr2k5K6H0OnIyrYuXPV87f0or3n0CHghRfUfYuKgAceCP9rdwXALbtJyUtTeT8QKU6P7OnwnhCkCSdEG6Kp/1xHxnxE4gglhx/u0LYWiq4D/+seDtc8YtCHInbuJIPucFDr1MGDFPINF0W4qCh6rFpFc0pO7l5EZMQI4I47iJccoF7vO+8c+LkGG/Hxvd/HZOqb+M+pU2TMReseY1T1/8Yb1HYXQUjh027GfalWg4V2dzt+9/nv0NDeAICq5BOjEvH4gsdDluPtDCMSR+BXV/9KuR7RhuiwrxoXULoNRL+/zPHkF0+iyloFDg6zwYyfzfkZUmNCK+IVMehDEQ6H2h8tWMO2bQM++4yKqWbNCu38BBIS6NET6HTkxWdl0euYmM414SMgeDxquF0YdKdTbRmMIOTQ0r5qXwcTLc4WWB30nTPGYJDCLAXnhU7SITc+N9TT6DVuKLwBM7Jm4L2y93Cy4SQWFyzG7vLdONd0Di7ZBc45rE4rnt77NH4292chXaREDHq4gHPg2Wcp/Jqf3/l2xcXARx+pIWkRwrPb6eYuDP1QhNFI7WXBQlkZ8Oab1Ad+ubKr6XRq3l2E3fsTbj9+HHjrLXWBUFQE3NJ7OckzjWfwyqFXAAAFiQW4e1IQ1eCGCJYULEFjWyPONJwBAFw74lqMSh4V1GMIfnWhEy5kTjvD9tPbsbdiLxgYrhtxHWbnzA7qfC5HpJhTkGJOQZQ+CvX2ehQkFeDpPU8rzH6CEKjKWoU/7vojkqKT8IOZP+igvT4YiBj0cMCGDaSYdeEC9VRbLMA99wT2ULOziTL14EHfnHOkSppQVwesX08Fam1tQHU1FZ099RQJp9x4Y7dDDBlkZxON7SefqAu87Gyipf3Tn2gbzmmbadN6NqbdTmx14rdls9F4XbUc+uFi60W8duQ1XGq9BIlJaHW2IrY0FjeP6UZTIMyw4egGnGs6B8YY7ii6AznxOb3aPyc+BytGr0CVjWhSJ6RPGJD+ZR3TKfrpikEPcDvYXb4bn537DI1tjQAD3j/5PswGMyYNmxT0OV2OyE/IV3r0JSZBx3REw8tIYpaDo9pajQZ7A14ufhn3T71/0Cv5IwY9HFBeTkIlHg8pnxmNvuxnWqSkAEuXkideWjp0qsIHAy0tRKV68iS9Fj3tdjux0cUNjXxdjxETQzUKcXHq95+ZSb+lM2fU7drbic527NjA4whUVFChJaCOZ7X6cvP3AG2uNlxqUfdpc7ahsqWyV2OEGltPbsXBSwfR3N4MDo4Nxzbgnkn39DpHOjpl9ICKeRh0Blw74lpsPbkV7e52aLXT/VFjq0FTexPRwzKO+rZ6bDmxBWaDGYXJhT0+5qGqQ9hVvgtFaUVYkLcgiGczdHDdiOvwzvF30Nze7COcw0EaAkdqjmBjyUbcNv62QZ1XxKCHA7RKYz3px87Ophu5w0E3b3HzLSzsnGb0SoDDQS1j/spo/tfyq68otAzQZwsWdN9eFq5ITCQiGoGSEjo37QLv0iWK/HRn0KOiqOahvNz3/b4sFhnRlSqa3ENowfn5+c+x+8JuWJ1W8sBkjjONZ2Bz2kJe9OQPiUm4fsT1cHqcsLvtkCAh1hQbsB3MXz8cDLjYchF19roeG/ST9Sex7dQ2XGi5QIQ5hhhMy+xh9Ocywuyc2Wh1tqLWVou9FXvRxtuU6AgHR7u7HSfrTw76vCIGPVzgb3y6uwEWFlIY9NAhdREwcyZVh1+JcDiADz+kULv/tRO1BtXVRAV78iSwf78q/GK1Ejf85SKIok3FCI74niAlhQoqtRrtKSmq4Iw/duwAGhro+i5dSqF5wJemdejYcQXHa4+jsb0RABRvdiAq1IMFg86Am8b0jFlQkWr1irmMSBqhaJF3h4qWCnxw6gNcaLkAAKi2VuOjsx8hxhjTK/nTywWLCxbDLbvBwLC7fDfJ/YKusdlgxqT0wU9lRJpVwwGFhdS6JG6+BQWBFbr8MWkSFXvNmEE3XrN54OcarvB4SIRFK+2qfTBGBn37dvJYtTzsx45RquNywLhxamGhOMeMDFWspjskJVEh3Pjx9FiwgBaK/ti7lwz6xx/TY+tWpSAz1hiLEYkjFM5x8TqUsLvseP/k+/j03Kfd9oKPSByBWFMsOIgvXUiLDmWabADIisvCuNRxSIpOUpTaLAYLSmpK8K8T/6LcugY2pw2fnftMIVIpby5HaW2pItPKwXG+6TxK60pDcTphAb2kxy1jb8GUjCkYlToKkACjzogZmTOwfNTywZ/PoB8xgo5YsYLyvxUVRE6yejWFUgWsVvIoGaP3J05UPzt7lm6mZ86QURo1ikRLekJqcjnBbKa+95IStdJfW+0tjNv586rH2lt62qGC9HRKvdTVEfHN0qXdh9sF8vOBhx7qfrtPPlFV4mSZWibj4oBFi5BuSceqcauw4egGMMYwPHE4lo5c2q9T6g9cHhc+OvMRtp7YCrPRDLPBjJlZARYpXiwqWIRWZyu+uPAFtYMxIDs2G9GGHiyywxjTM6djeuZ07Di9A4erD6PaWo1DNYdQXF0MGTKsTituHHUjTHoT9pTvQY29BrvLd8PlcSHJnISEqASkWlJRYyMxJMZJ+z0tpoeLxcsU0YZo3D/1fjS0NWB98XrEGGJwW9FtIUkzRcRZwh1OJ/WYC43xjAzyygsKqBr57bfJwxSQZVIVi4ujCmURyp88mQrELmc0NADPPEMGnTFibGtsVD30zgoIExOBu+4i7/ZywpYt9DtYMACFS7/7HS0iGaPoCABER8N1260oHZ2MCcMmdr3/IEHmMnae34k3j7wJp+wE4wyJ5kTcN+W+DlKm/ni5+GWcbTwLxhjumXzPgKiQhRJP73kaJXUl4DKHBx7omR6PznmUFOA+/hXsTlK7kyQJmbGZ+M01v8GBiwew8dhGNLU3Ic4Uh0UFixQp1ggGB12Js0Q89HDHxYvAP/9JoWRJotdvvEE84bt3A0ePdpTdvHiRFM9qa+m96Gjg17/uedh1qCIpCXj8cfV1SQkxqbW0qEbcbKbqcABobSUPduXKy8+YAwPboufx0EMskjgH7HY4XnsZ++cmYdSa38OkDz2nfKujFRuPbYTdY6ciPcZQ11aHt469hf9Y+B9d7nvP5HsGaZahgQwZsiwrmuoeiRZmnHPIXIaHe8AYg0f2KDSn0zKnweayYdupbZiWOS1izMMMEYMe7uDcV1ubcwobv/MOhVWNRioI0xr1f/2LbrbaivmLFwMbdEFIk5w8eOc0WBg3juhkX3mFrofZTP3Yq1fT5+vXU3X79ICL3Qi6QlwcFcFZrepvT5ZhcTLcc9CDquUXkZkSHh0XMpcBGeAScarrmA4WYy/48i9XcLWAURhxAdl7zxERXG2h44K8BVdsu1q4I1IUF+7QpkRE2FinoyK6ZcuILlXvXZcZjWS0hJ64MObt7dSf3ehb9AKXi4rEXnmFvHmbbfDOq79oaqKoRXeYPh345jepwOvJJ0mXXeBb3wLmzx+4OV7OePhhIuvJ8ZKtaBaPTY5mPPPVM6hurUZTe1NIpyl4thljYDK1baXGpOJHs34U0nmFA6KN0dAxnWK0o/VUI8DBYTaaychDho7pEGMIPiFOBMFHxKCHOxgjQ62tyh4zhowRY0TLOW0abbNkCfD73/v2ojNG/cUPPeRbaCcKmd5/nyhS/+M/gI0bVaWwcEV7O3mFzz5LKYee1IDMnw/cf39HGddwUqgLB1ittKizWlWO+M4gJHJ/8hMqwtOCATX2Gjzx8RN4Zt8zAzffHiAhKgE/mPUDxJpiYdAbYNAZYNKZwkIZK9RYO2ktRqWMomtiMGHtlLUoTC5EUnQSvj/z+4g1xcKkMyEjNiOyAAIJ4dicNlgdVlid1rDseogUxYU7OCcilOefp9f5+cAPf6j0/AKgsHl7O1HGGo1UffzHP6oMX1FRZLC1N94DB2hM0bfNGHn6115LHm04wuOhOR87RmkGnQ749reBKVeevnTQceIE8Ne/qp72qFHkhfekkLK8nLz1lhYAQIOJ49+vZbCbGPLj8/HLBb8Mue51Y1sj9V97SW4SonooGnSZo9XRCpfsAjiQEJ3gwwMvrpkkSZHrBeCFgy/gUPUhpWXv3sn3hoRUJ1IUN5TBGIWNhQqZyeRrzAEKs2t70OPjyehrQ9L++XOPh7xx4aVyrr4XrnjjDZIHdblovm438NJLdD1Gjgz17IY2RGpG0OWWlVGNwf33d79vTg7pwP/P/wB2OxIcDI98CfzXfBkXrRex7sA6PDSjB61wA4jE6MTuN7oCEWuK7fSzyDXzRbu7HQ63g+hdPW6sP7QeFqNlQKl9e4uIQR8K0OmI7rU38A+D+kPLJiZQVAQsXty74wwmbDZ1wSEWITab2jYVQd9QWQk895yq2gbQde6NDOvYscDatcALL0ByuRDrZJDA4JbdaHW0Dsi0I4hgsLDx2EaU1pUq3rnEJNiddp9CwnBAJId+pWLsWFVBS5Jo0RAXF96ENMuWUf+9iCrodGoBYAR9R1oapVoMBrq24vfQGxlWEUlatYrSPjoJEpOQYcnAqvGrBm7uXrx17C38+tNfo7iqWHmvzdWGP3zxBzz+0eN48osnYXVGdOIj6Btm58xGbnwuJIl+1+IRbhoFEQ/9SoXFAixapIbygd5HAQYbOTkUXteK2VxzTfjPO9xhMFDOXJJU+uGsrN5Ha3Q6IrGxWhH34TY8ejIBjntuG1Da1w/PfIjiqmLU2GpgdVrx7vF3Ea2PRkFSAV78+kWcbjwNWZZRZ6/DS8Uv4TtTvxMW/fERDC1kx2Uj1hSriNtwxrEwfyFy4nonqTvQiBj0KxnJyUOvbWvePPLSRTHnrFlXNod9sJCeDtx8s5rKSEtTOeF7A5MJuP56mKKjMSo+HkifEPy5alBlrcK5xnPgIMGRGnsN/nH8H7h9/O0obymHzGVFkKS8udxXOCaCCHqBuTlzMTxhOMCoP39m1swB0bfvDyIG/UpBSwuR0YgcdFoacFPPFJrCChMn+nLZRxAcxMYS53u4jdUNhAoa80qhcc5R0VKB5vZmSJB85EIZVN3qCIKPHad3oCCpIORCPAOFCekTMGGAF6j9RcSgXwlobwc2b6b2N1FAlp09NA16BBFoIVKYQnUYDKOSRym65QyqXKjP9hEEDfsq9+Fs41kcqTmC0vpSpJpTcd3w68JOO36gYXVasfXEVri5G4lRiVg6cumg59gjBv1KQFsbqbUF0gqPANizh5TJRJX3woVUYxBB2CM/Ph9ldWWot9cDoFznitErkBmbifFp41Fnr8OpxlMYkTACyeZk6FiEUCbYOFx9GAcvHQQA1NnrAAYUpRX5GHSb04ZPzn5C/PBgSIlJwZycOaGactDhcDvw/sn38cm5TyDLMmKjYhFrjMW8vHmDOo+IQb+ccOwYUFOjvp4+ncKfgg9ecMEDvsxzVzKOHCH52Zoail7o9aRwt3w55YMjCGvMzZ0Lq9OKktoSAMCSkUswPJGYEtdMXAOb04YNxzbg1nG3Is4UF8qpXrbgRAqv1idwdEhtNLY3YuvJrXDJLuigQ2psKixGCyamXx7pM6vTis/Pfw6X7IIECc3tzfjk3CcRgx5BH3H2LIXVL1yg17JMjHEil6kVeRGFTxEAH35IxlwseFwuem/2bJKqDRdwDhQXU5RlyhSVvz8CLBm5BEtGLgn4WbQhGhPTJ+Jk/UlMyZjiw4QWKhyqOoQ2dxvAAb1Oj6kZU8NiXn2FEHbhTGPYNYFAt+zGsZpj8HAP9W0zoKa1BttObQsrg64sDDkABqSYU3pVD8A5p4f3OoSiRz1yV7hc8P77pMImvHCPh1TXJkyg3vKUFKKIFQjnfvOe4Nw5IpUpKCBq274iMZGMY3u7uuARSnXhhGPHiBXP5QLWrKHvc8yYUM8q7HHw4kG89PVLAAO+xb+FpKgkFCYXhmw+ZXVlePnQy2i0N4KBwWwyozCpEPFR8SGbU3+REJUAg96Adnc7OOeIN8XDqDMqnze0NWBT6Sa4PFSQ64Eq0xpOON1wGuv2r6PFFQMKkwqxrHAZACDdko4Uc0qX+3u4Bx7uUchnIgb9SjP1L+EAACAASURBVEV1NeW5Oade6756X/4yq8LrTEwE/vM/gzbdkEKowq1fT9GI1auJFEWvpwUNAOTl9Xy8tWtJWvbUKfX6iV73cIHbTefb0kLf53PPEb3vk08Gr2XP4yFOdgGzObDc7hCC3WXHK4degc1lA+ccf/vqb0g1p+IPi/4QEm55zjnWF69Hg70BAOmRe2QPKloqQm7QL7ZehNPtRE58Tq+Fa1aNXwWr04pjtcfAOcfdk+7GqORRPtvEGeNQ7aqmrgMwMB2DxRScOhWH24FLraRbYTFZujW8geCW3aiyVgGg74XJDCfrT+LphqcBALOyZ2HVuFVdUuXGm+KhZ3rY3DbEGeO63HagEDHooUZzM92gz52j1/feC8yZ03uDEh1NRk3L3242h5dhCgY2bCA+d2F833qLDHBKCvCXv1BR2yOP+CrOdQfBhS/LRLLy4IPhFW4H1EWalrI3WB6OEAB64QV6LkkU+XjwQVo4XAZgjKktbCFCQ1uDbx88Iza71468hsfnPx6ynubGtkb871f/ixpbDb499dsYnjAcaRZ1MeeRPahvq4dBMnTK737P5HuU5/7pg7SYNNw+/nY8e+BZuDwu6JgOo5NH43vTv9fvuXtkDz499yn+UfIPSJAwLm0c7pt6X6/17k/Un8BbJW/R/Lmk/F6EoM9XFV/BbDDjljG3BCQmSjYn43fX/w7nms7h1cOv4udzfw6DztDv8+stIgY9EGTZVxvcaBy4AqmXX1ZD5R4PaZMPG0Y31N5gzRqacwkVB0GWgfvu6/044Q5Z9i3oc7tVJTqxmPnb34Df/rbnofgHH/Q1jv0J4Q8UtPr2QHBTAufP0+9Q2wVx5gzw6qvA978fvOOEAIwxcKj826HEM/ueQbW1Wn2DA2ajGT+c9cOQEpQ8s/8Z8nA58NyB52DUG/GrBb9CRiwtavdf3I8XDr6AvIQ8/GT2T2A2dIwKdXVtTzecxovFL5KqGyg3fe/ke4PyfRyuPoxNxzdBhgwOjpLaEmw8thH3Trm3V+MwMOihh5u7ITNZnZu3wE9mMj499ymSopKweGRgBkWJSRiROAKPzX0sZGyEEYMeCF9+Scpe4iY6YwZw990DcyxhhIT31dfqc6ORwscul3pTvky8qw4QFKUAnavorRfX0O2m69BTw9zfxZrHQw9hcA2G4EdG/HnVeyJr2lOIayaOI85DG+0JQ3hkDzyyRynAMkiGDh64xCSA018uh9aou2W3QoIj8qs6pkOUPrQLSFlW2fTcshs6Wadcx5LaErxc/DI8sgfnG89j3f51eHjWwz1OWQjv3+YgB4kxhlp7LZ4/+Dwem/dYv+fOweHyuBSCIRlyn3LXY1LG4K5Jd+Hvh/8OxlWSIsa8xEWc9ZhlMJTfZ8Sg++PwYQrrOp30WpLU5wOBYIZQr4Teae310l4zxlSj19wM/PnPwOOPD041+IYNpC8vwuJr1lDLYLCg1wMPPUTn1NpK6ZXvfQ+ICaJXpzXkWq78MMa/Tv4Ln5//HADdfFeMXoGr869WPjcbzHhwxoP4656/wua0ISE6AQ9Nfyik2uwMjHK0jCFaH43vTvtu6LXGvVSmwnAJZr3y5nKsO7AOba42gNGi6ET9CawvXo8Hpj7Q4+EdHgeYRGNyRsViTjmI91SmRgik3ggKaYdgDLNzZqMgkSKauyt248MzH5JR52TU5+fNx/y88KbKjhh0f3g8ajW4uMEdOgS8+y5xXQcb/iFUbZ40go647TYy2KLmYOVK4KOPgKYm9dpZLFQsNxjGfNMmIqYRWuLAwGjK5+cDP/mJGgkY0Ud6zT/+kYoAGaMIxiOPEGvg6tXAa6+pi5K8PLrWYYodp3fg47Mfw+6yU64THJtKNyHaEI2ZWTOV7UYlj8LayWvx/IHn0eJowatHXsUv5v0iZEZdVHaLv+mWbmSOBwHCoxUeqDDunHPYnDbIkKGDDuCAi7t6rVon2rnAAHAg1ZKKtZPXBmXuoshOhqzkvPvK16+X9MiMywQALB25FHanHbvKdyl99haDBdGG6KDMe6AQMeiBILwVSaKbW0EBcNVVA3OslSvJQG3cSKH9/PyIelhXSE8H7rqLDDhjpBKWmUlV4DYbFbfdc0/fhEX6AqHRrtOpaYB336W2wGDPoTeFfp2hpYW8fMGK98wzwI9/DMydS2I9AnFxVMsRprC77HC4HT4h9HZ3Oxxuh892rY5WfHDqA7i5G4wxXGi+gL/t+xt+OPOHg1Yg9+rhV3Gy/iTq7HXQSTpITMLKMSuRGZeJpOjQt4+umbAG6w6so/npdFg1bhVSzamobKmEjumgk3RKQWF2XDZuGXNLj8eOj4rHHUV34I2jb8DhJk/dqDMiOy4497jRKaPxg1k/AABlYdeXKnd/WIwWLB+1HJOGTVLSEcGa80AiYtD9UVBA1J+fUygPkjSwN7f8fPU4eXl0rAi6hrhmAlOmkOfqcFAtwdSpgzcXbehfhPsmTqRFRqiwcyfVgQBkqOfOpeevvAI0Nqpa8pyrlLcmEzBp0qBMb/vp7TAbzJiX23cWrZlZM1HeXI6SOpUIRMl5etHubsfzB59HeUu5wuUuyzLqbHX9PYUOOFJ9BKV1pQG13+vt9aix1cAje8AYg8QkFCYXoiApPApWC5IKsGbiGrS0t0CSJEzLmAadpEO6JR03FN6ArSe3UlEhJMQYYpCX0PO2UIlJmJ0zGwadAW7ZDQlS0NrVAEqrTBo2ML/bZHMyks3J3W8YRogYdH/ExZFOeEqKGvrOGQTN2wnhreIT1pAkYObM7rcbCMycCVRWqp0KAFBYSK10g4ELFyjsr60tqK8nQw0Qv0FMDDB5MoXatUWT2rqDQcC+yn3YU7EH1bZq6CQdDl46iGuHX4uitKJej5URm4H02HQcrz2uVLJPGjYJhUkqaQznHBdbL4JzKobj4IpB7QofnPoAJ+pPKKxnN4+5uVMjdr7pPN4tfRetzla0OlpRZa3CuNRxuG7Edco2EpOU48pchiRJIW2fC4RAjG0mvQnXjbgO0YZoXGq9hL0Ve1FlrcLTe57G2NSxWFSwqEdjG3QGzM6ZHewpRxAAEYMeCKmpwOLArQkRROCDwkJgyRKi3hU5/NzcwTt+SwtQVuZbd6GNGtTUEIug2Uyeen09pXgAqjGYM2fQ9OQvtl5EWX2Z4lHX2eswIa3vC9kJaRNgYGqv75SMKT45aYPOgDnZc/DRuY/gclNdgyj66sygfnHhC+y6sMunZ7zF0dLpHFocLSirK6OxGUNZXRlijb6LOdHTLCruAVXuNdxhMVowIW0CjlQfARiJrJTVl6GurQ4XWi5gYtpEzMiaEeppRuDFgBt0xpgOwH4AlZzzGxljwwG8CSAZwAEAd3POnYwxE4BXAEwDUA/gds75uYGeXwRhgsZG4IMPVC83Kwu4+uqu9+kpzp4lL3reAAklTJo0aOFqH9TXU3gd6FhMqa3+NxrJeM+fT8b8ww+pc2P6dBKhMRo7jj0AUAqYmKxWU/cDY1LGYExK5/S3ekmPZaOWweq04suKL6HX6TFl2JQuZT1PNZxCY1ujMl8Ojp0XdiIjNqPT3KzIL3PVWnfcxpuHZRLDhLQJoa9s7wUa2hpwuuG0TwtXQ1sDGtsaUWurRYwxBuNSx4V6mhFgcDz0HwE4DkAkh/8A4L85528yxp4FcD+AZ7x/GznnIxljq73b3T4I84sg1LDZiHd+1y41F52aSqHi/rR/1dYCu3dTqLm2FmhooMKyUKY3mpqAzz6j54wRI92MPno49fWkFieMubh2WmMOAEVFalX8smWA1UqPW24JbutbNxAELwxMEbAYaJj0Jtw89ma4ZBcMegNWj1/dNYOXVlyEU9i+zdVGPeQBN+dUYc1Vo+7PUT4qeRSMOqPy/orRKzplXAtXKN8XA7Wggc69orkC205tQ7Q+WlG5iyB0GFCDzhjLBrAcwH8BeITRkvxaAHd6N3kZwH+CDPpN3ucA8DaA/2WMMR5uDP4RBB8NDVTEpSXVqa2ldrD+GPTKSvL6ATJuH3wAzJoVOoNut9McPv2UXksSVcL31aBbLNQRIRT2RMteRgbl0BsaaGGk5WSXJGpRCwGGWYYhKSoJ9W2kXZ5qTg1KRXJ3iDXF4r6p9/Vo26y4LJTVl6GlvUUxzosKFmGYpfOiWFGMp/2rxeKC4KXvjtYcVdjmpmZMHZSFQawpFlmWLFS0VkDikqIoJtIGpxpO4Xjt8X4Z9LONZ3Gm8YzyelzqOIWpLpwhcxm7LuyCS3ZhXu48H1GaUGCgPfS/AHgMgEgqJQNo4pyL5W4FgCzv8ywA5QDAOXczxpq92/uUpDLGvgPgOwCQO5i5yggGDiYTtaNduODLAtfSQhSkfe25FgI1wmvtDxNfMFBTA3zyCc1BeM8tLcDp032j6M3MJOP8+utqcduYMdQK+eGHwL59FGafMiV459APzMqeBZvLhr0VewEGXJN/DSakh08xaHlzOeJMcUiISoDEqKJbhhyQ6lRA9FiLNAIHR2N7Iy40X0BufHDvT2caz2Bz6WaUt5SDc476tnosL1w+4LSxufG5WDV+Fd4ueRvt7nbUtdUpkQwGhoToBCSZ+95+V2WtwtaTW3Gs9hgAuqZXZV+F0SmjMTZlbMiFa7rC/sr9eLvkbThlJ1weF1LMKZiaMTVkRY8DZtAZYzcCqOGcH2CMLQzWuJzzdQDWAcD06dOHhvfe0kJV0ImJwe0xr6sDqkghCBkZvn3EQwlpacCdd5J6mJAulSS6Zu+8Q8xrfRVLERz54j9YKA26VtpWzKe2loRR+sq5P2IE8MQTHd+//np6BBOVleT15+b2mVb42uHX4trh1wIgA3qk+ggAyjGPTBoZMtrMWlstNpduVhTD5uXOw01jbuqRYpbMZUhc5f4+1XAKn537DHdPCi5d9PbT23Gh+YLCuLb91HYUpRZhXFrg/LXdZcfphtMASOI0J77v3TqjU0bj8QWP46vKr7ClbAva3G2wOq0wG8y4bvh1uCq77zwde8r34EjNEcXj55xjd/lu7CrfhaUjl6IwqRCjU0aH3Pv1R2ldKd48+iZa2lsABrx59E0kmBIwLnVcyAhoBtJDnwtgBWNsGYAoUA79rwASGGN6r5eeDaDSu30lgBwAFYwxPYB4UHHc0EZ7O7B9O7BtG4VX7703OIa3tRXYvBnYu5dez5kD3Hpr8OlfL16k1ifGqE8+mBziWuh0NHdRgS2M+okTwNatwP33935Mg4FoUm02GstoDC09rjjHhga1Ej0vD7jjjuCMb7ORFK9AUhKQEKTiq4YGUrY7fpzaOm+8ka5tP/DO8XdwpPqIkp/90awfYeKwju1Tg4FPzn6CwzWHlTzxZxc+Q3Z8NhbmL+xyP72kR7QxGm3ONoWnPUoXNSBes8xlyKAFqexdmHbGW+7yuPDJ2U+w6fgmAEBhciHun3p/v1McM7NmYmbWTGw7tQ0fn/0Yc3Lm+LTo9QUcXOGTF56/COtvKdsCMODnc3+O0SmDRBbVAwgp3Pq2eir09M7fA0+fmeqCgQFrQuWc/xvnPJtzng9gNYCPOedrAHwC4FbvZvcA2Ox9/k/va3g//3jI58/dbsqXbttGr0+eBN58kwqStGhoIE+7qopuyN2dtsNBRWT79tFrxijf/MUXwZ1/YyNJu/72t8DvfkfHG6ivZPhwUocThk7bH93X8FVREXn3FgsVf02bFlo609xc4DvfoR712FiaV7BaxpxOWjj+4Q9E7/qHP6j1A/2FzUbGvLSUvosdO4CPP+6XeEuLo4WIRiQJkvc21NDW0CdhjaCAqdXqjCuq3d3uNj5tPO6eeDdijDGwGC2wmCyYmT2zV2xqvYGIBHQX0j1ee1wx5hKTcLrxNN4peSdo81gycgmevP5JrBi9osvtamw1qLZWo9pajSprFWpttR22idJHwaQ3KcWIQmhFfB8SpJAayc4QY4hROAbE7ybU0wxFH/rPAbzJGPstgK8BvOB9/wUAf2eMnQLQAFoEDG0cOUKkH7KsMnMdOUKe9Zo1tI3dTkbz1Cn6XKcDHnus6xDszp20UNDmYmU5uMZWzKuiQg0Tr19PeduBql3Q6ciL1grjiPf6ihkz1JYyQekbSowZAzz1lPo6WMQue/eSARc1AwBdR6ez/21pW7YAxcW+egObN1PdQx+LFt869hZO1J8Al1X+8NePvo6M2IzQeGLeinbRqiZkMzuD6E2PM8VheuZ0H2IWnaTrdw7VLbthc6oSzlH6KJh0JhpXJmNn0Bk6FSNRivO8CmS9UQvrKbo7R5vTht/v/D2a25uV7dNi0vDba38LnaRG+hYXLEaroxW7yqm4TKinca5W1ofaUPqDMYZHZj+Cf//431HfXk/Xmcshk00VGBSDzjn/FMCn3udnAHSg9eKctwPoyJs41KHtDxZFWULuEwBeeomKokTuGCAPvDtoi8fEcYKJl1+mefkvGgYS48ZRLv211+g1Y1SRfuutXe/nD391vIHSsu8rBmI+2gJA8XrfPmI8XLYseMcQCNYCUgIZdQ54PB44Pf1T4XJ6nAFlVLudBpMURTCf1wHGr2ypxJ92/wlxUXF4fP7jPkVzEpP6LfzCOcfHZz7GeyfeUxYYC/MX4s4Jd6LV2YqyujLIkHFH0R0YmzK283G8dK2Md8+ON1CQuVfOlFO7m9BE10In6XDTmJuwZOQSvHn0TRy4dEAJu8tcDkru3OlxKl0IXY3HOff5DeokXaffp0lvgl6nh47pwDmHJdqCH8z6QZdFlAONCFOcP/xvUv1ZaU+YQBXHb7+tji1Jvl6Zw6G+73bT47nngJ/+lMhVAmH+fCqIE1668GSD6X06nerY4uY90BkQxqgWYNQoSiGcOkWRjJ7qmgMUUXjqKZXiND8fePTR0HvmgwF/TvmpU0mXoL+48UZKvxw82P+xvLht/G1obm9GWV2ZEmr1cA+eO/AcHp3zaK/4wgFVvez/fvV/MS93HqZlTuuVUb+h8AY0OZqwr3IfGBgWjVjUodCrub0Z//X5f6HV0Qq72w6ry4o/f/lnLClYgtePvA7GiDTm3in39stD31OxB++VvQeH7FDCzS6PC9GGaKydvBZtrjYwMKTEpHR6HM45ee8cSn98KCqvFXU5xsFl3mkaw6Q3waQ34Y4Jd8DmsqG0rhRMolD26qLVGJUyql9zeGbfMzjVcAoA8P0Z38eY1MCERAcuHcCrh15V2P2mZkzFmolrAl47vaTHT+f8FDKXlcVCqNXzIgZdixMngGefVfO4Y8f2rRhLQK/vWJQ0dmxgj0l4wpJEi4CuqrpNJmL4stmoQnraNOC663z7jfuLO+4A1q0jljVg8DSy9XoSwlm8mIx7X6qphQIaQHULzz4LPPhgcOcZjtASzAC0EApGjj4mhn5z2kjN8uUkQtNHxJnicPeku7HuwDqcazqnhFbb3G0d+rh7glcOvYKjNUdhdVpxpvEMXj/6Oh6a8RBGJo3s0f4xxhisHLsS1+Zfq4SG/T05xhisTitsbht5vmA413QOrx5+FXa3HRKTcLDqIMzHzFhd1PeMoUt2oV1uV7xzDo4vK79EkjkJiwsW94hlbmzqWNwy+hZsKtsEDo6RSSOxcuzKPs+pP1C+Twndhv3jTHG4a+JdsDqsSltcdnx2vyIMzx98HqV1pXC4HZAhY93BdfjxVT8O2Fbolt2wuqxK/v6ryq8QY4zBLWMD10R0xToYCkQMukB5OfDii8TkJXS0Dx8GNmwAbu8HYd2ECcAPf6jeDJOTqfpYQOuxi1x7dnb3udWvv6ZtHn6YDGBqP35YO3dS37KY49VXA9dcQzKlL74IXLpEWvA5OX1vH+stzObeG6P6elqEeDxqxIJz6gi43DFlCrVF7t9P+fSiouC2rd1wA9UjiJvz8OH9Th2kxaQhSh9FRXHeIiiZy3jh6xfww1k/7NXNstXZihYHkcG4XC4wN+uU3a0zJEUndS1nygEd84ZgNXapzd0GHUhi1C27e60X7g8GRsVvTOWBH5s8FtMypvV4DIPOgIXDFyI3gYxWQlTCoJD4aGE2mHHXxLuwvng92jxtYJxBx7rvkkmLSUNaTPCcE6vTSq2F3gI2m9MGj+wJuC0DFeFJjK6/U3b2+/scTEQMuoDLRWFFYQgYo7BzY2P/xo2J6dqTWbaM2Mv+9S/arjujefgwbWu10pyrq2n/a67p2/wOHqRiKu15itax/HwiLmluprkNIk1onxAfD8yeDfzzn/TdBaqYv1wRG0uLx/JyOt+KCuCFF4i/Phgc9sOGDYiEsMQkCg3LlG/VS3pclX1Vr7nOhfiJUikt9axKvVfHYAySTgJzM5XX3Htc4dEpVdr9QFFaEWbnzMbeyr1KcVuSOanXUp4xxpiQEvcwxjAtcxrAALfHDQ6OGEOMT0HcYM1DqN1x8C69fQZGaniaLodQ1R/0BRGDroVOp+aNByvEPMqbGzIYqLI9sRsqx5YWumkL2Gx9JyUByJA3NfmeZ3Ex9UdPmUKFakMFej0ZtS1b1O8wNTX4BCvhisOHqRCOMYpKtLZS5CUmJmzY4vxx3fDr0NjeiFortTNJkoSitKKu+dYDQGKSj0EHgq9oFm2IxuLhi7H11Fa4PF71Nq+hUFqreP8NQGJ0IpaMXOJDBDMisXO2xDZXG9YXr4fNZUOsMRb3TL4nZAQ9/tBJOszM6l7aeNeFXThw8QBmZs/sF0lNIAjjLORru1LbA6PWQCbRQmp44nDMyp4V1PkMJCIGXSAlhUKKe/eqxi0trWc6204nheatVgoT33577wq5gJ61/5w7RwImgaQy+wMRmhbj1NURTelQRGIiyZm63WTUU1JIC/xKQG0tPbTfZ5h/l5OGTYLNZUONtQYyZOiZvteeKECEJ9U26neWJAkzMmYgzRLEmhJQEdR1BdeBM64adDDU2GpwqPoQGBiGxQ4jr7SfyIjN6BGXucvjwsZjG3Gk+gg88EDP9Nh4bCPumHBHv6vtBwvFVcX46OxHqLHVoNXZCrPBHFCfvafYf3E/vr70tSIClG5JR729HjX2GoVjoDPkxudiaeFSZWE4PGF4j+swwgFD4xsfDMTFUTWvzQaUlFAx27XXdu/ZyDJ5hHv3UsW6yUSPW29Vc/HBgmA+E4bXX1WrL8jPp8e5c+p7eXmUIx2KiIkBVnRNdnFFQGvUw7zCf07OnH6PMS1zGmwuGypbKsHBcd3w6wZEojRKH4UbR93o815FSwUsJmIgHJ4wvF/GqLeQuYyS2hK4uRvggAsuHKg8gFhjLKZnTsdn5z8DQJGEEYkjgu799hcn6k9gx+kdqLZVQ+ISLrVewvbT22E2mPtkSEtqS/DRmY9Q3lKu1GQUJBXAzd2QIEFm1HrbmVEfZhmGFWOG7v0jYtC1SE2l4i+TiQrT5s/vfh+PhwqRRO+400l56ZtvDr5Bz8qiVqSyMt8oQl7vWnx8UFAALF0KvPsunUt2NkULRvW9TeSKxu7dKv3qwoXdp1CCiawselRWqou9zMzO2x8vMyzIWxCS42bHZePOCXd2v+EAQeYyIENRQLO77Thw6QAyYzOxq3yXEnI+23gWMQbfvPqX5V+i2laNOTlzglqI1lOcbTyL883nwTgjAhwwnG04i/NN5/tk0M80nlHEa0Q9g2hXEx0DMpfx+fnPsdK8csCFbQYbEYPuj+xs4Nvf7t0+/m02A9mvnZxMnPACkyb1T2IUoIpom40424PRt3yl4tAhovkVgjnt7RQtGKxiwjFjKN2wa5f63lVXAePHD87xIwgJOFQ5Uw4Oo86IjNgMH/Uyzjgutl5EcVWxYtCPVB/BB6c/QK21Fs2OZmRaMjE7ZzYsxsHVO9C2KXJwpMem92txIYw5AGWRoPDDc9Kvr7HVwOFxIAYRgx6BFoL9TfSRDzQBy9ix9Ag2Zg2dwo+wxaefqlz8nAOff04GVZu+sNup6FBUpQcbU6fS4wpEaV0p0mLSum49CyO4ZTcOVR3C5GGT+1z5zRhDdlw2mh3NkGUZkiRhdMpozM6ajWf2P0MV21ypElS48k83nMZ7J95DjbUGHBy7z+8GkxiGJw5uzjgpOgkJUQlobG9UyFmmZUzrc3W+IHkRinSccySbk9HqbIVdtoODQ8d0WDl25ZD5nfQGEYPeX3BOhVf19WqFfMrg9ntGMMhobKROAy2RC2NUFCmMuVjkaRd3LhcZ/U2bKL0TE9N3rfcrFM3tzTjfdF7xwNJi0uCRPaiz12HLiS0YmTQSy0ctH3Qvs7coqSlBXVsdNpduhsPtQEJUQqcyqF3BqDPigakP4K97/opWZyviTHF4YOoDqGypREJ0AprampRe9mhdtGLEPjzzIS40XyCvFRwe7oFOHtx2MgCYkTUDNpcNn577VHkvzhTX5/HiTfFIs6TB6rSCMYYYQwyWjlyKYzXHcLbxLLVG6vRhJ8UaLLChLGg2ffp0vn///sE/cFUVhajFtUtNBf78Z+rXtlhIXCWu7z/KCAYJ1dVkVDuTVJVlYsrjnAROcnPJw373XeCzz3y31UZmBC+/Tgf88pdktGWZ1PBeeUWtwM/JIQa7nvZ3nztHY48YEfaFbgOBNlcb3jvxHnac3gGAvNMpw6bA4XbgWN0xhexlxegVWDJySdjetNvd7fjFjl+gsb1RESFJjknGk9c/GdQ5H64+jBcPvgi7yw6jzoi5uXNxxwSS6n3t8Gv4svxLOD1OyJDhkT2QmIRfzPsFCpMLe3yMens9mtqblNeZsZkh0wLXYtPxTYjSR+GGwhtCPZWggzF2gHMeMM8a8dB7i6YmUh07fVq9if/0p8Cvf931fu3tJJMqEB8f/kQtgwlZVovJtItMszl4mt4Cokf7zTfJSM+ZQ4syg1/vc3Ex0cY6nVR8+LOfAceOkZcdqHpcpF+E9rqWLvXSJeDVV8mYi30vXQJefx145BF1jJYW8vSF55+SQmOdOwf893/TXB56iOY7AEQv4YyvL32NSmbezQAAIABJREFUD09/SD3f3pzxoepDxOAGHWRG4dbNZZuRE5+DycPCs11ROFESkxQdbc45amw1yI7LDtpxJqZPxOoJq/HWsbcwPnW8YswBYM3ENWhzt+HgpYOKulmMkeRAq6xVyrwkJnXKT251WvF2yds4eIk4/jk4VoyixVRveQSCjc6oWgcKVdYqJWUwzBK6/5cRg95bvP46cOaMGlYVJB7d4fPPKdQqbvBz5wKrVvVf2vJygTCewiAKYpjJk4Hvfjd4x3E4gI0biYBFlqljYPt2YsRbuFBllaupIaY1odxWW0vzmzdPXchpGehE2N1goL/33ksFi/6LBEA11v4h+fZ2WmR8/TW9lmX6jUyeDPztbyoB0J//TIucn/2MqtivEIjwMAenfmKvIVQEQMS/IRB1FPMU7HKNbY148esX8cSCJ4LKTDYraxamDpsaUGZ17eS1sDqsONV4CjKXcd/k+5BuSccTHz2BFmcLGBgSohLw87k/D9jTv+XEFhRXFftc7/fK3kO6JR0zsmYE7RzCHeXN5fg/u/8PrA4r4qPi8djcx3rEITAQiBj03kIrcyo8sldfJU+qs3yoLJNhELKpjFH4NT6eet+vRLS3q8+1xlNbWMgYebQuV2DD2Bds2UJthm4NzzfnZORTUnxpesV3LRZhYi5aHgDxGxCvb76ZjL7F0tGD1+nU7bQ3WI+Hxn3nHWp51Mrr/uMfNDfRFimuTUMD8PTTFNIX3AeXOSRJgp7piaPdq/MtPHNZlhXdbJ2kCzpLXLChl/Tq4gQMSVFJeHjmw0GnGWWMwagP7DToJT2+M/07cHvoesYaY2Fz2eCW3XDLbjAw1LfV4+mvnsavF/66Q+GekDcF80rgMk1R2hWCNlcbnt77tFKr0NDegKf3Po1fzv8lYk2xgz6foUNSGy7Q0sKKG/Pq1V0TsXz5JbUzCQgj4O6deMRlg7Y24De/oVDzE0+QcdJyrwswRiFuIT8bDLjdZDAZ8+Xtd7k615cXc7p4kXjiA30uOOOjo6mC3d+YZ2WRcp9YmIjtGQN27AD+7d+oh12ruifmq9V3116j6mrgF78gj707iCK9gda0H0BMz5yOGwpvUHTKGSeu8InpE6HT6ShELEm4ddytKEorCvFsA0PmMkx6Ex6c/iBijbHQMZ3CHx6KQj6zwYy4qDjEmeIUOlQhUAIOgANOd+ca9QxMiZYweMVXwnstFXQ43U7l/MEp/P6XPX9ROgoGExEPvbe4/XYqfjtzRr2xJiZ2XaTkdpNHKm7G4qZ9BRY2weMB/vQnyh9zTkVmL7xABlWrFy88UY+HFgDBgn90RRxH64k3NwN/+Qt9Z9piN1lWPWUxRzEW5ySQ01X735QpJEv72mtAejrwwAPEv/7+++o5+kcoxOJDHEu7kBDXz27v/rz/3/8jKVnOge99z5fLYIjAoDMojGx07+SINkTjlrG3YJlzmVIUlxmbOegCID0B5xxP7nwSVTbiKXDLbjw842EkxyRDx3ThU8TnXUsKA9+Z5OmywmVoaG/A4arDCivbDYU3YFL6pMGaaXjA7zYeHxWPuyfeHRJRl4hB7y1SUoC776biJYGeiKMIbwzo+PdKg9UrRyjOXxSBacPQ4noVFgbWj+8rrr+e8uElJb4tZsuX+xq51lbVixdz1UZnxHvi+ezZJDEaranwbWtTux8ASrE8/DAZ8+hoikzs3EkLB+25a426JPkuALXG3Z+DvzP8/e8U6XA6aXH54otUWNcfhsEQoLiqGFtPblWVzThw4NIBpMWkYVHBolBPr0ewuqxodbQqXu17J9/DI7Mf6VerVjAhvExANeidGab4qHjcOvZWLMxbqKQO8hLyYNJf/ukfASH0ooNOKdQ0SAZkxYWGnTFi0PuCnJzut9FCa6z8869XGsS1EB6uJFEl+/z5pAwmjFd2NlHSBqrmfuMNYj/rTJa2shJ46SVVFz0/n7TdAVqQ3XwzEb589BFw/jwdb/9+4OhR4J57qOBMp/P9zrRhda1BF99jcjIZbAG3m7TZz59Xc+LNzdQh8f3vk7f8zjvU/qjTdexp10YOtMfS/m7EfHQBvNEdO6jwDyCBFm2aobmZIiJhBpfHhb/t+xua2qgNymKy4KEZD8FsMAMA2l3tsDqtiq41B0ebqw2tjqGhd//i1y+iqb1JufmDAQ1tDd3vOIgwG8xYO3ktXLILpE/CEG2I7jTikW5J77QK/kqASWfCPZPuQZ2tDptKN0Gv02PV+FUhq/KPGPTBwJgxxN5VXKy+N3Zs2EpaDjj8vXG9HliwgIyigH+BGkB1CCUllMs+eZIWAiO9rFY2Gxlxu52M1YUL9L5OR9XhtbXkgS9bRguynBzgwAHaTqiSMaYW6wXyxrvy0A8cAE6d8iWVOX9erdoH6LlINVitRFDj75l3tsgT72tD/ZJEfAfLl6vb7d9PXn99vapxLxYBYt8wTfdwznGp9RLq7fVgYNBLerxc/DLun3o/jDojCpMLMTNrJvZXEvfEUNKpBoDJwyajtK4ULrcLYFQDIDEpZEVk75W9h5MNJ5WK+9vG34bc+NygqMUNFdTb6/H3Q3+H00N1AllxWVgzcU2P92eMYUrGFLg8Lhj1RkiShOmZ/aTi7gciBn0wkJZGHNttbdQmlZ9P3md28HpOhwwYo7B3fT218rW302PDBtXTlCS6Vv4GvaKC+v9lmQz4O+8Ad95JXvwbb5CxF4WGgtiFMTLyp07RMWNjfUV3tMWNnFPRm8FAufJAxjZQ2oRzVboU8GWJ828X0nr4Wm9bbKf1yLVeuYDWgwdo8bJvHxlyxmgOly75zs2/0DBMjfnGko0KwxcDg4d7cK7pnGLwks3JGJk4EgcuHlCq2PMS8zBx2OCpm/UH0zKnYXPZZjQ7miFBgl6nx4K8BUoEYjDxydlP8GXFl2hubybBEo+Mjcc24u5Jd4dEpCUUsDqt2HB0A0rrShX+9xpbDeJMcfjG6G/0aiyDzoCr868eoJn2HBGDPljIzSWjnp5OnmJhz9mYLiswRsVjoqr8888pt3vsmGp8dDoy3sLDHDWK3q+oUMcAiMXtnXfIUy8uVhnYAhkxgDz1M2fIoO/Z42v4hAE+ccLXYIrPtYYwJYVC/ozRAqOiwtdIa71o8R5AefOZM+nzzEyK0pw44VvwlpdHCxxBsqMdQ3tOYgHgcFBhXaDFhniuXSRMnEiRkKTw4rHmnON43XE4PVQxDFnN4WqRm5CLkUkjcbrxtFKBvadiD1weF8amDoDGQZAxO3s2GtoawBjlWpeOXBqS8OzpxtNobvfWdnAqfDvTeAatjtYrxqA73A6U1ZeRgIs3F25z2/BlxZdoam/C8MThmJc7L9TT7BUiBn0wMXr0kKwuHhAYDMDixdSqJdqytEapqYmMLkC86Xl5Hb1VWfYtbhPvBSDR6OCVHj9OYXbt5/7ebyBPljGKCNx+O71+4w3K2QOBK9HHjaMFB+fUm758Oc0vkBRuejpFL6xWIrtpauo4J52OFjgnTqh58Z6Cc2LFC8dUzyefwNLSjlqdNwQtAXqdHhPTJvrkb/MT8jE+dTwZdACXrJdwyXoJSdFJQ8KghwsV6ejk0TjXdA6NbY0QMqNDhZQnWIjSR2F82ngcvHQQXFY5AZram7CnYg/sLnvEoF9x2LWLbsDz5kWoXPsCf7Y0oKNhrKwEhg+nKEdNTcdt/YvHPB7yhtPSVG1wxqjYLTeXjKE/zaz/ONrPtAZVkihysG0bvX/xYuAxOKfagBUroHAU2GzAxx+riw6R5xeFgBYLFdZNnUpV9p9+Sr8t0Tsuy0Qgs3gxRQY8HnUs//C6/0JAOz+A9v3sM7Wd0mymOgax/8WLVCToX6wHkNyuYKjbuVNlSjQagauv7hsJ0O5dMGe2gqeS1CeXOUxGE24eezP0UsfblMxlhQedcdZpa1UEgTE/bz4OVR9Si/IYkBefp7YFXgGIMcbgxsIbcfjSYTiZE+AqWZFQahtqiBj0/uDQIRLqaG2lPG16OoVU9ZHL2iv4Gx9/D9n/fe1ff6IUYUjHjqXvYt06NQSel0fh/g8/JKOsNXDasUT+fdgwtYhNvM85ifNs3ux7XP/5yzKQkQFERanb1NbS78XtVvP7Wpw6RUxxBQXkyZeUUHuk/7h6PXn4Z87Q+2Le/vUA/tdKe5137SIq4rY2eh0XR3MVffSlpb5UxdrvqaKCOgUSEoCtWynSIcu0v0ipBIqSdIXsbDj11ZDhUMLtOXE5AcPuwniLvCdnHBF7/v/Ze/P4qqqrfXydc+/NHJKQhCFMCUEgzLMgFlBAURHneQCt1Urt+Ku1b0e/nd72rbZWbbVatYqtQxFQUARRKSAyIyhDGAJhzjzn5o7n98fDyl5n33OTAIEEzcrnfnKHc/bZZ+999rPmderUI7kH1fnr6ETtCfKGvZSTmkP7yvdRWlzaVyb0LMYdQ33S+lCtv5aO1RwDc2iASTwfE+ScX26i7Y0WLQIw+HxwpnrlFUgsW7fitXt3W/ewfZNpQn1tWUralC8GsPh4gIf+O0u28vguXZCC9957FbDxb1VVkNgHDQLzxefxtYnUe4+H6Oab0YbeN04Jy+0mJuK6sbH2fg0fDvDfvBmv3bsB5k736aQhyMwEeMuxCIdh/549W92D7ItlAVi7dcO4yXZTU/Gd34+1W1en1PbV1fguHMZ7ZmT4JcdnwwaYLPg3DsvzevEcnE4ioDlzyH3BBY21u3ul9KJ7Rt4T1WEsZIWI87ifj5JUe6Dr8q6jRy5+hDISMigQDtCy/cto3mfzaHXhagqFQxQIBWjr8a2Nr33l+9q6y61OGQkZ9PDEh+m2IbehjCyFKBwOk8flofT49OYbaGfUIUqeLh05AnUlb+6Whc8vv6yO6daN6H//t116FZ8Vqq1VjmZEzTtfdepEdP/9RL/4hcrAJtW7rE7Oy0NUwOuvA2g5tEyCJwPZ9dcTjRkD0Dp2TP1OBDX1/PkAyfz8SMCSkm6nTqrmuWxDp3CYaOxYgP/jj6NdBtcFCyLNCbrznPQbiI+3l3KdMwfgevCgulZCAo7PygLT8tRTYFS4jdhY9OfWW3Gvmzere7z5ZoRQ+nyKMeG+GAbG7NAhrG0uDysd7fh4qRVhZkeq8k8DYE3DpLlj59If1vyB6gP19MDoByg1zrnKXrwnnjrHnVxXBhFZ1C5Kdp6vFAqHKBTG3IYoRP/Z+R8a0nUI+YN++uvGvzbmeu/VqRd9c8w32yxpytkkt8tNaXFpZFkWuUwX9U/vTzcNvqmtu3XK1AHop0sLF0Y6Vcn3ulfyV4HWrSN6800FUgwsTr4Fx48DCNgmrjujSaDz+QAyF18Mjci+fWAeJCBZFiTDl16CHToUQkpVngeW1nftUn1wCueyLDANt99O9PTTzglY5Pw6ZXiTfWLSbdHy5fEAzMePR5x8RQWkZyIwKF27Otul+/ZF+tjnn1d9GTwYY05EdOONeJWXwyTE6nKdeSLCXJSWIg3vzJnoD6eUlUmAnDQJTmF4p0Exrhj6+eSfN3vcpD6TaFKfSad9ndOhE7UnGkHvywZopmk2MkYcDmiFLTpRd6Lxs0EGnag9Qf/6/F/0o4k/asPenh3q17kf/fGyP7Z1N86YOgD9dOnBB4keewwOVkT2DU5Kjec7VVer2O6mctYHg/ac4uEw4qMTEwFKPp8CR5+P6I9/VF7cus2ciQFi+3a8TBNOV7/+NWqD798feV5dHdEzzxD97ndEd9+NHOZSypbSdjQHuLg4JQ0z+Dkd52Sn1hkTmbZVBzw+LjcXhVuSkjCGr76KMD4exzvusDusSRo8GOtQOu1JqqtDdro9e+wFgVjVLiVw08S9jxsHhmnhQnvRGr5GbKzKGSDV+vHx9tS3oRC0Bx4P4v/PUyqpK6HHPnmMyrxllOBJoO+N/x5dkP7lCTtl724iagzhKvGW0Gufv0b+kB8JfIyTldQ6nBXaNXXY0E+X3G6ib38bqk/TtDvC8UZ5vjvH+Xyow/3jH6Mq2hdfRD92/XolEUupb9UqpFj9xz+IfvpTtPXLX0Jq1NWz0g4sQZfV4cEgVMGrVkWvjCZtuh4P5iAmBi9ZvtQJzIkAVvffD/v3gw8qyVheTwdlLiDj5KSnMw9sE5e0bx9szy4XSqXu2IE22d7+2mt2zYJObjf66fFEpoF97TUV4sbjLcuzSurcmei730Wfp0whmj5djR+37/HAYe/CC3GtBx4AoxcXh+fhRz8CY+L1QmPz05/CMbG+3l417jwhb8BLf1n3FyrzlhERUZ2/jv664a8qhvtLQG7TTR6Xh1wmCsS4DBdlJmTS/aPvpxhXDHwUTjoqesy2SWnaQS2j8xxx2pgSE1ECNBSCp/uf/6wKj6SnY4M7F2p3acN0yut9uvT3vyN5SyCAzfi553C/vXvbj+Nc5HpcNNtd+fxAwA4mTglcnEKw+DcigNySJXZw0CV8vu6wYUS/+Y1qb8UKoo8/dpbQiQBes2dD6iWChM790e3dTKaJ8rgbNtileV16JwLwzZpF9O9/2+/vgguIrrsO73l8+LdwWFWia454/KWU7vdHagmc7p3vPyFB9WH6dBSd0e85JUU5Cz73nIoYeOop+B7ccQcYOK4fv28fysMOHw6/gFP1gG9Demr9U3Ss9hgkUwvSbCDc/vLgnw5xec/7R6EmusvlaowWyEjMgHROZmMegD4pfejrI7/eVt3toBZQB6CfKbHTV2Ym0cMPq+9dLtg+zwW9+aZyfrr+eiQPaQ3yeu2OYjU1cPySG3yXLkQ/+Qkc0U6cQEIUXYL98ENnAGaSQOxEOgD7fM5qbn5fVYV+/uQnmBcmVgfrGgC+xk03Qd3Mv/XoAQnTsiDlvv56ZF9HjMC9yZhtXcXOnvD334+kMKEQ2mK7tMejnOGuuw6mCJbIW2q6Wb8e0v3gwUT33BN97Lh6m05paZC2t29HtAaD7ogRqrCNE7GZxTCwXlijwWuHgb++HiF5SUlwzjtP6K7hd9Hjax+nUm8pGcbJuPcviW/MS1tfop2lO8myLApbYXpg9AO25Dy9UnrRTyf9FKGBlkVx7jhKiUtposUOamvqAPTWJF1yPRe0ZAlUmxwq9NZbkLJGjDjztlnak6DOzlr8ubCQ6Mknib7/fWVj152kuK64LAUqf+e2nJzi9O+JEOecnw8vdNZIsJaCjz98GDXNf/hDSM9vv62SqMi2+PgrroCd2uWC38Djj+NeDQPalmnTIp31iJS3vA7i/OL75exwbjdSz8o8/jt2gHEgguNfXJy9oEs4jOiJhx5yXmOffw6mrqYG4ZKJiXbQlPeqZ9KLi4NWIj0d0vS8eap0rGXBD0IyF9KeHgyqAjPSSdBJs2KaaF+un/OAuiV1a0xsY1mWzUmsLejlz16mgRkD6cKeF55xW7X+WqppQFKgYDhIz21+juI98fTQuIcoKzmLTMOk3qltsKdFoWpfNT2+9nGqC2ANZSRk0A8v+qFj4qGvKnWMxPlO1dX2QiL19Sqs60yJN2dWqTuptokgUfJnvUzs8OFEo0fDvn7kiD2uWiZAkW127w7V9OefA4wlQIwZg5z4o0fDo52TmsgSpER4X1UFyZnzBfA9MUkA3rgR0umtt8KefeSI6ltNDRglLvsq+yqrs0li8ON++XzwRn/oIYBo3klJiAurcAKZlStxb336qLA5zk737LNwLnvgAaUZ2rcP0n5tLa4VCGDMkpPBpFxxBc49dAhtcX94jFwupCNOTcVY1daq+ySC9oFV6ky6pzvPO/eVSCXO0cfrPJRuXaaLTDIbJXNfwEfPbXmOHhr3EMW545o5u3VoyZ4ltO3ENirzllF+aT4t27+Mbhl8Cw3IOL1U0ot2L6LCykIyTZMssshFLqrx1VCNv4ae3vA0dYrtRHPHzm03ddq9AS89u+lZOlpzlKywRWQQ1fnq6O+b/05zx8xtnJs1h9bQyoMrG1PZXpJ9CU3sPbGNe3/uqAPQz3eSmylR63rWX345AHTpUjixOUnLus1W37C7dkUbcXEA32XLkNhl61YlCfJ5lgVJ8frr0ebBg5HFVrp3B5h17gwVeVUVQOfdd+1e9twvLlPqJDWytGpZqMTGAH3iBH7n+wqF0Pemyqc6mQBcLpxrmnC2GzUKDmZMO3ZAw1JdrdqpqYFpIDnZDojhMNLV1tTY/RC8XvRd1pivr1chldnZyOpWXAwfgqIida2YGAC+jH13Mn/oJhSZsEfXRni98I7n43QG7zyynzMZhkEuQ6UDtciiEzUn6FwmtCmtL6WjNagZ4A16iRqI6gP1zZwVnQZlDqIdxTvIW+NtzIoWpjCRhYpjdf66xjC99kAWWVRUW2QzeYQoRMW1KhX0luNbaPm+5TCPnNSgVDRUtEl/24o6AP3LQmdDCho6FP9XrrRfQ75PToYKXF5bVz0TIf83EWys/fqpsCw+LyEBIWkZGQCY//xHgZIEgS1boOZnCfGaawDuH3ygzA6GgX7l5aHv8nwnqVL2Xf+Nz01Jgef3kSOQ5Pk4aZt2eu9yQSsxbRpU7bIv5eVIB6vHecv+sA1aAqM+vzpD1bMnrvnss/Z7Zk2OZD4OHgRDxOlkx46FP4bTNXRHRb4et8WgLqvPyT736AHmTqdFi6DJ4Zz37YwMVC9B7XKCLfmSnEsoxhXT/Mmt2AfDMIhjxc80fKx/en9KjUulY9XHbG3xNdobGSf/TANrnUvs8nsiMD1l3rLG+TLIoG0ntlFOag4N7jK4zfp+LqkD0M93GjwYm/Hu3ZC+Bgxovs768eOQlLt3hxTeEtIlN45X7tzZXl9cHm8Y6FdVlQKRSy9F2Fl9vR0kWEqurIQaWgcFphMnVDY6y4LN+ZNP7PZxIki5kyahTUlOqnH53uNBgpeVKwF0zGxceimk2UOHcM9cm90JuFjy57zrEyZgnNlh7PhxOA+yVkD39Od7k23qfWXq1g3M0o4dqi2fD3neZdY+2SZTKITzFiyAzT0tDYltWDtAhPs9cEBdP5rfg5xLnblMT0fEQXY2tDNM//0vGIpdu3CNzp0REpeREXmfbUSGYdCEXhNo+b7l5A14ye1y00W9LqLpudNtVeDORT84LS6+cLbjVzZU0jv571AoFKLMxEyaOWBm1DZHdR9FmYlwGt14ZCNV+auIwmBcTMNsMz8BJ/K4PDS+53haeWAl+cNwsE30JNKEXhMaj4nwb7CI4t3xX6ksgh2Afr7TsGF4rV6NTXziREhC0aiigmjxYtQPT02FNDzxFGxMvInHxkI1nuCca5uIlLTGVcUsCxv69u0KgFnS9vkA5FLN7CQRMklwko5a8viuXQEQXi8AQ7YpnfT43Lw8eMXPnIn+MNPRqZNifHr3xmvJEpXYJlq/TBNzc801ysO+vBw2+m3bogP13r1ggmSbug2bKTOT6Oqr0T6D6vHjqqSr07jwNS0LNvdt2zCXMTFg8qRD3fbtWC/NMQfyO6c569UL0j/T5s1gmjhTYGUlzguFkN2uU/uw3RIRzeg3g+r8dVTpq6Q4VxzNGjDrnErnRJCo95fvp9J6qJNzO+c2gjFTnb+O3t79Nq07so5CVohSY1MpOTaZJmdPdmxzQq8JNIEAiLFmLH108CPyBrzkMl00pMuQdlWgxW26aeaAmeQL+WBysIjS4tPostzLGo/pldKLeqf0pkNVh8gkk8JGmAZmDKS+aX3bsOfnljoA/ctCX/uas6Qsqa4O1bG2bMHmX1kJVXVSElSeOm3aBJUw26alM5TfT7RmDVTJLPV27QqNQWmpKk+qh0ht26bak57pMrueDgwShPkz/zdNAObmzSqcLSYG4WFEkArHjlX50GU7erKYSy5RYW533OE8hseP4x4OHHDuJ9uW+/QBKN1wgx2cSkrATEnSAVcyCtzH3FxoRWJj8ZLUq5c9VG3RIjBSsn3ZP25XH28nGjYM62bpUqXt0DPH8fxF0zSUlWF+xo9Xv3/2GcCcr89tbN5MNHVqqwK6L+ijlQdXUjCMLHkpsSl0cZ+W17k2DbPN83qP7zmeav21tKdsD1mWRZf3uzwiBW2tv5Y2HdtEIQu272pfNa3Yv4KSYpJodNboJtu/euDVVO2vpnJvOcW4YujGQTdGSLZFtUW09fjWRhV9dmr2Oa1BH+eOozuGRXkuiWhgxkCqz62ndUfWNX7XPbn7uehau6EOQP8qUVkZQFh6mRcXQ2U9fDhAj6VpIiRhOXEC4UkSVNnx6sMPYVdmR69Bg/BauxZOatIZjc/buFElQJEMglTZsso6MRFMgmGg7yy16oBsmgDuykoUZMnLg5S9Zg3OLSiwh94x6UDXnJNTeTmAbePGptvq3p3olluQa12S1wuJV/ZbtqOPsZTgZ81SnvE6VVSgXT6ea8Drbeugy+Pcu7ezs9rx4/CiJ4I2p6wscpzYk79XL3zm6ADpf8DHSdLHXFaeay0fEELhkZUHV9Lbu98mf9hPpmVS907daWLviXS+xZNP6zuNpvWdFvV3i0467Z30ArfIotL6UlpVuKpZQDcNk+4aflfU3ysbKmnZ/mX06eFPG9X9uWm5lOBJoD6pfU77npqi7UXbqdJbSWEKk0kmje0xtln1+ajuo2hU91FnpT/nA3UA+leJeFOXGd3i42GzLCoCCMsEKVLiZHUok2EgqYwTXXQRpLrly/FfSnSy3Kf0Bk9PV05whgEpdPRoVWjklVeQkY3I3t7+/dAW3H8/PNpffRWFVdasgapYqu+lQxifr9+rZQEc/X47KBkGrrV+fSTg6JoDLs+qU1ERNCLSyU2P83cae0mhEDQE8pz9+4k++iiyDXmPcvx5LjmpzbBhUPOzEyQRmISlS+1hg8xscVucStYwkAEuNpbo0UftSYT4PmtrIX0TIV2yHG95rxkZzoVoTpNq/bW0ZO8Sagg1kEEGBSlI3qCX8kvzaWDmwFa7TnuhsBWGtzoB0GNcMRGqeSKiwspCKq3H82YYBvVP709JMUlZwA07AAAgAElEQVQRxzHtL99Pnxz6hIiITAu53feV76O1h9eeNUB/b897VFBR0Ajo3qCXpvad2hF33gR1jMxXiVwueGuzZ3UoBHv7zTcjzpq9t4nsedV1FatlQU3/zW9COvd64XnOlJqKtKFHjiDpjQQVLg4igTUhAc5tixahT263vWoY90f2iQjHcg50LmDyne8A4BYvxrVkPLQEx8REqLCrqwFAyckApJ07ES9eWxspJTvFoTPJ+1m+HPc0fTo0Hjk5Kv5bH1cdfPX2LAvqZwa5qir0T3r0y34115a8biikirekpCB3/cCTILd5M+ZOtpeYCJs7mzacyqnKMrQsvRsGTBTPPYf3w4ej/7JaXigEp7h77oGGoxUpHApTyAo1OpWV1JfQ6ztep0enPNqq12lrMg2T0uLTqC5QR1UNVRTjiqG8zDy6c5g9019xXTHN3zmf8svyicPwrhlwDV2We1lUu7lFFoXCIWrMlmcajd+fDTpafZS8QS+FrBCFrTCFKEQLdi2gYV2HfeXU6KdCHYD+VaKePSHJPv20koxlFSzd41p+Jz24DQPn8bFffIHNmo8fPBgpTBl0JOkqVcsCeM6fj8+mCQnuG9+I7IfeJ75+IID86J06IdabKLKGeTgMwGbntClTYNN97jmo5KdOBWg++2wkmDvdA0uebG6QeeotC57jHg8k8ltugcR+4kSkqlkfCz2szLLgkBcbi3s4dixyjnQVfUwM3rOWgY/hcZDncR+qqohefJHo97+PVL9z21y+dtMmuxkmMVFdOznZHm7oxPRs2xbZZyJEEfTuDZNBOEyWYdDxFJOykrOc56GFlBybTHWBukbwcZGLkmPatvrb0eqjAFPLItM0W6Uka5fELvTbqb+lfeX76OkNT1NWchbNHTs34rh397xL+WX5tu8W7V5E2anZNLTr0IjjiYg8pofiPfHUEGwgmQo2wdOEU+wZ0JI9S6ioDn44HO/PJoUOik4dgP5Vo379kNaUCJtyZSU2YK4W5gQWvOkySCYkIF1oUhK+q6mxn7NzJ8LVnHKGO3lb69cLhZQDlmUBhLlaGgMJn8vX8PtRy7vPSfVffLw9zajLRTRyJPrNn//2NxWStWCB3cbs1C/5ncuFPl14IRgAtqvzsVwlzbJQsU7+JqVplmKZdG0IETLBud0wJbz1lj2VLvdVtjV4MOZm1Sp733VHOFnfXEr3cm50hmDOHIzrvn1gWEwTmevYhv6tb6FIEZfKDQQiQT0as/T66+j/woVE1dVUE2vQM1d0ormXPnLaUllKXAp9d/x36f8++T+q9FZSnCeOuid3p+9e+N3Taq816Fj1MfrjJ3+kKh98QlJiU+i3U39LiTGJrdJ+v8796PHLHo/6OxeaIYIt3LIsinPFNcZ4O9GQLkPoin5X0MLdC4mIyO1y07ie42jWgFmt0med7ht1H/1+ze+poLyg8bt4dzy1o0i6dkkdgP5VJFb/rlkDgNC9lJ0kdbmx19WhEtsjj0DV/tZbkefqVcOiqYB1MgxIaD/7mQKsCy+E+r2iwlm6k4BlWfAIv+ceqKaJ0M7AgSgywiVt/X7FHDi95BjowBcTA9CcPRvv//UvxWw43Z+0d0tg5e+caqbr4+X3wz+AgVIfP55TLncq29cZtWj9kuV+WY0vK7i5XHjdf7/dsTFRAFFODqR8y0I2wJdeitTyyHGV6y4QwD3yvISIfDWV9NT6p+hXl/7qtG2n3ZO704NjH6RXtr1C3x//fYr3xJPb1TZbXzAcpCfXP0mVDZXKiY0sqg/UtxqgE1GTMfIe00Nu003BcBDgbhDdNvQ2GpQ5yPF4y7Lo0yOf0jv570BCNogGdB5A1w68tkkm4Ez773F5yO1ykxk2ye1y0wOjHzhjbc2XnToAXSfelA2j6VKkoVDrlio917RlCyQhlqD0zTYa8TGlpUS/+AXCvGbNQmY36ejFjldOEpkOakR2idOy7PPg8wFsZC1xvQ0JGMEgHLz+3/9TIBUTYw/3evNNVVjF6f74PRPfU9++MAe43WjPNJGCtqYGueejMSncR3kN3T7vpNGQxzOY633s3h3mgy++ILrvPvRr0SLntiTJ+vBpaSj3y3Pwta/Bq//DD/F5/HgVix/XTP5yzk0QFxeZY1+Ohx55wHN38n2nANH31hE9c+WZlyvtn96ffjDhB9Q5vvMZt3WmJIHUJJOqfFX0xLon6JdTfnlO4ttvGHQDVfmq6IviL4gspHyNc8dF9frfWbKT/r393xQMB8kkrI/dZbvpnfx36Paht5+1fj4w+oHGUEODDEqLTztr1/qyUAegS8rPJ3rmGWws/fpBfRiN/vY3JGQZdZ6GSIRCAEqnjVYHDCcVqWFA9Xv0KBzQ5O95edjUN22yn9OcelevBMbHbN+OamnyO10K5n4eP070u99Bos3KQh1uJ7rqKnidM6hLMJVV4VJSVEEVIvz/4gtEBFgWssDdcAOc8tger0vaOoPA/gs6IyUZGnlstPHndtl7vlcveIm/+y5MAE7V7bgtjweSdrdu+M3lUu+J8PuVV0I7QoRxaA7IdRo+HElv3nnH2Qci2lyK9/HB1tOxtgcwZ+JsZhZZZFgGbNPniBI8CXTH0Duo1l9LnBu9W1K3qMeHrTD5gj4yTKScNQyj8buzSR2lWk+dOgCd6dAhqGirqrC57dgBdaFM2EEEleCOHTiuoAB2vwcegJr3fKH8fNh3iZzBg8gORjrQSzDhghwyjrygAIzOkCEAP2nrjuZoRuScIc4wiC64AB7sbPPn2GjZnscDTcGrr0KytCz046mnIHmuXo0wLHmebvs3TTjzvf8+zArx8UT33ovrM23YAIBi+/zq1WBerrgCfdBV9rqqWdcoOKnYmZzU5Hp/2bxRX488AvPmod/sia4zFaxKv/VWqMdfeAHrNzU1sv2EBCVtL1gABk2aJaQGJi4OJXRl/XmPB86G9fUIq+vfH/kBPvgA2gxJkskT49QpLoXuHXnvlytUyaBGUDTIoOTYZLp35L3nNPtcWnxaiyVew0Aeef5jJuBsebjP2zaPdpbupLAVJsuy6N6R99LAjC9fiOHZoC/RU3KGFAgghIlthaEQPutUVaXKSfKmzmrC9kh1dQC12lp4gX/725Be6+qi166Wn3VQ0YGF25BlM30nOffrr4cWwzCQKpXTkertJCTAEzwhAd/v24ekNnz95GTEqfOxehIajq3/739VSB4Rvs/Ph+rd67UnpuF71YuO9OypnL0CAeS8HzBASeYLFtjz0Pt8SGGalIR7l9XRWOrWwVsfs2hMDt+bZHRkhTg+j6X9YBB262HDYHLYscPeBznmH3wAkC0pIfrjHxEy9u1v26vBMS1fjmRBtbX2tSC9+qurUX/eMJT6nhleLgpz7BhMHVVVkQAu2zUMMCU330yuzp0pp7s9oc66I+vog/0fNH6+uPfFdEnOJRHd9ga89NSGp6jaV00GGZQSl3JOS57qZFkWPbPxGapuqG4sx2oYBrlNN+Wk5rRJn1pCuWm5NKPfDPqgAGPOdvOzZT+v8lVRubecyMKYzds2j+4fff9Zi3f/MlEHoDOxzVxKVE4brSxTKZ2Z2iP5/dAyHDyoSoC+9BKksylT4AXNwBJNYnQCEiYpReuq4/x8gAURqqhxaJM8hwjHezzwQGfnqpwcqMuZODf90qXQpEjJVM4bFzuR8xEMIlxMfi/t1U5zKQG0oABx2vfdByBnpk/ed69eiH0/etSu5mZg168tx8vJzMHj4pS9TTqu6WuVzSChEFTdfj/6pTvd8VhJlX1TTGl5uWJi5HV1bQSXZi0vJ3rsMQA5f2cYYKqc6sdzO3Idut1gSjrb1eTbi7bTe3veoxJvSaPEuKJgBSXGJNK4HuNsx3LJzSpvFZGJalz/3PpP+sbob5zTwiqSSupLiOO4TUL1NtNAopb2SvGeeMpJy0HBlsbn9uwA+tK9S+lw1WFyEcrVhilMZd6ys67e/7JQB6Az6RtjNKB2ArH2SpZFdPiwAr5wGICYnAz155o1OM5JSte/0793+i+Pq67Gy7KwkQ8YgI2+tNSuto2LQyUzaZ+NVsHt+HFoFpyAWTIgzfWdz+XzJNi9954qHMNjtmMHQs/q653nu6wM0if3TV5TOp5JqVxPBCP7IImZFXm/TmPP7WZlwYHt6FGYJ2QbTsya3taBA7DBGwbMSDNmOGtyovlYGAY0G5xoiLUH0UwPcj743sJhMANvvokwObE2KrwVVOotbXTOIgNpSUvqSiLnxSJAvglJOBQO0aHqQ9RWscyy/KlpmAB2URa1vZNBqPhmGScTy2h9Xpy/mAqrCikcDpNpmHTj4BubtM07Ue+U3pQck0y1PmiDGsuktuNttj3R2dGZnI/UuTPsvrx5de5srw7FNHq0shOaJs4503KPy5ZBcn7xRVXUpLVI30i9XjiY9eiB0CsdRJzAjz87SZX6b0SRYFlUhDGTBTdkn8rKWm620LUBTTFiUpKWoOTkeCcBze+3H+P3o8Tn4cPO911cDOZFl/BdLjiVTZwILYTTeBkGks5kZ0eaAlrKXMnf0tMxr4cPg4Fi0udYZ3zq64lefhmRDzt34vXJJ4rpa64fTuPrdLycM/5+8GCsDfmdZcHO/tZbtrVhVOWQcWAqGZXwWTEJ0u2O4h2UX2pPlmIYAE6TTAXuDKBtRBN7T6SEmIRGMIx1x9JFvS6iGHekqWNH8Q7659Z/0sufvUxbjm85xz21U49OPVBT3CAyLIOykrJoeDdV0GnlwZX06ZFPaWfJTtpZvJN2lu6kGl9NEy060+AugykjMYPYz8AkE2VeEyLT13ZQJHVI6EwpKXCqcruxofToYa8OxTR2LGyJXMTksssi1IKnRGvXwrGqokLZgm+6ydlJ6XRJSlMNDbhmIOAssen/nbyRZbtOoKK3RwRPdb24ChH6sXkzNvRZs5oOBczLg+Sn1zh3Iv1eZJ+6dcM4c7UvnalxAnwiaDYGD8Z5Bw4gl312NkwaxcXOgOVywWHO7Va+BRL4iYhycnD/S5ZAC6GPaV6eYoZkfyVwVlRAvV5cjBBCWV1OJpSJNpcNDXB6k/0qL4fpZORIVWO9pMQu7XNbTmtIB3QijH12NtqKj8dnj0f1V/Y5EACo33hjo6mhR1Jvyvb2psLyL8jothXSomXQISI6kmTQAMlbh91kHR1D1FBLZBHFxhk0MjelTQF9eu50qvZV06rCVRQKh2hM1hi68oIrI5z+DlQcoGX7lzUmVimqK6JETyINyBjQFt2mbknd6Kr+VyHDnoHkNRLQd5XsoqqGqkZp2gpbtKpwFXVN6kqdYk+tct6gzEHIQGfBbHJZ7mUdIWstpA5Al9SlC1R8zdEll7TeNTdvVl7ZoRCcmqZNO3NAX7MGau6BA7FRy4xddXVwIJNqXyewlu/5sxPwyd8kcMj/e/ZEqn35PK8XecOvvBIAeOKEKuRhGMj+NngwQsRqa+HMxY5VeipTp81a/z43F9fkWtzR7l1XK3fqRHT33XCOW7MGQDt5MrQrshQov0IhjP306dHLsUrauVPVHec2+vSBc2Hv3s7nlJeDQeM0ryUlmNtoJoiWziWfSwQt1KhRGPcVK5wZM6c1IKX2cBiM75QpMKd8/DHGLCkJRXdqaiL75XZjDQsGKzf3ZGHAt4eQUTzENhRWX/vQuAwPDTFvpuqTAn4nF9E1/YnMFuD55s0I789q5TwmpmHStQOvpfpAPflDfro+7/oIJ73jNcfp/X3vA8wNOIYVVhZSfll+mwE6EcqlZo/IdvzNOvnH+fItw6JqXzUFQqeeQ2BSn0k0qc+kM+nqV5Y6AL2tiTdYWbTjTGx827djk1+2DHbUWbMAgjt22LO3yY3eycuYCH3q1g2SVGGhkjCZ9Bhn/k6245Q7XGYZY9V0nz52tfe776rz+vaFd3tODsCRwwZ1yTArC7Zs6fDmJJ2eOAEGx6lQij4G0tbPaWWHDMGLqUcPpLrlqAi9QMmpSIT6GujWDZI3h+pZFjzRJ0wA4B09qmzecowl08Tv9bnhMS8sdJ5LfS1eeimYjspK+7jpzJu8vsz9P3Gi8o245BLkcmDmQ2/L7QYzeuutEdXXMjJgWZDZgWU3t2xRyyM7G4o2vYR8NNq7F0to1Sr4OmZnQ0GRcgoh0YcOYQlbFtJZ9Oxp/93j8tDsEbOjnl9QUUCfF30OtbwJ1TObDNor9ezUkwoqCqjOX9cI7FcPuJrSE9LbumtfKeoA9LYmHcyjZQtrCe3fDxvosWMAzQ8/xEZ6++1I1VpQYD/eKQRKl+D694fK+Pe/Vx7K/BuRPUxLesHrql4d8BnUDQNAPWcOgKq2FupXCTAFBZDgc3JUGzIGmkFuyBDY67l8p5QS5f3t2YMdmvO9SzTQK4V166bs3NGk7Msug4S5axckZM6LHw7bvdKbo/R0MC5cHIYIduzVq+1zkpCAPk2caO+z7hWfnIxxqq1V88Dj5vFA6na7URWPx4yZINPEdTI126XOGBDZnS6ZYmMxZi4XxiMpCUi8cSMYoKwsewU92WbXruj77NlE8fFUXw8ejgiXSUuDK8vSpfapPXIEzS9YAN/LcBjLon//lhdwW70abVgW2vv0U1xv+PDmzyWC0uG994g++wxDOHIkUhuciptNUkwSpbqyqKLipPqaeWNvK5rhWplmDZhFNb4aKqgooJAVItMwKdYVyUXVB+ppR8mORgYlLT6NcjurHB6FlYVUXFfc+HlgxkBKjm3bQjrnE3UAeltTSgo2VS64YRgAMK8Xm+GAAS2X8N55B7sQETbZYBAOcMOGIV3pE08A7Imw4XbujI1cAoikcBiq0dhYHMcbsMtl90hmkoyBlNZk2zJ2OTUVG/cDD6gEJvn5UO3qtlgddBmc+PpeL2zQ8tp6eJlsa/RoSOpcMpaBjts3DEjlDz9sd+aTVFamnBjz8vBasQJjXF6Oe+O4+ZYQMwZbt6IvlZWYQ1mHngjz9Z//AJAtC8dIxoxzsI8cieM/+UQxYwy+DQ1oQ86hjBrweIhGjCC65hr7taX2Qc6lXA8uF8bs5z/H+48+Qjz6P/6BY4YPR94B2Ra3kZFBNHduY6hiIAAhfv58O//HAC15xbVr8ZJTGQpBudG1a8umQS4p7tKRI5hapxB9ndatw/TxUG3aBEl/xozmz2Ua3m041acOp/krFL+ZmEiU1KXlbbQF3TGsabNSIBSg/x78L83fOb8xDn9Ql0H0vfHfIyKUdV24ayHtKt1FHIlw36j7aFzPcU0120GCOgC9rYmrV+3di82cC1SEwwCE3/2u5fq+5GQwB7JsJoNfly6oX/6XvwAAevYEWK1ejd3S67Wr2onwORRCalci+y7Jn10ugBaf71R6k/9L6dfjIfrud5UaW5KsQKZL/qWlirngdnW7b0ICvLb1GHA+Li4O0nlSEsaLHQT1+5bMSiCgEuNweytWIHOcvNcZM5BS9YknYPvu31/9dviwko6zsgCqRUWqX6mpSCN7ww34/L//C62LrgKXDI7bjd2eEwXxmMXGYhxmzcJv7EfhpC2R886g3q0bNDM6JSdjbjivvJOpIhzGGjt+HABdUWEf/61bsR719WoYiPc/CebhMABy/nxVOsE0oc4+eDCSz3Vaan4/aufk5UVKyV4vhl/mLWKSy2vZMkyjTBjYEnLiZ1tKEyZg2pYvx/mTJiHp3vlMhZWFNH/nfGTIMxBx0BBooLL6MkpPSKele5fSrjKAOWehK6kvoUAoQB6Xp5nWO4ioA9DbBz34IJ7an/1MOUXxblVa2nJAv/deHL9vn9r0uf43ERyr/vhH+zlf+xo2XJZuWbKV4CFJSs7x8diYr7wSu6ZhYCOXDIU8Lz5eAb/sl066t7jbrWKRX39d7eYScLlfSUlEX/86QgBZ1czEYHrxxeizZWG88k+GO8nqXxw/TYTvPvkE90gUWXBE7tzLlgFg77kH7ZWUQG29YwfRk08CCLt0Ifr//j/odhcvVmN+0UVEd93lPNbSxEGEa1gWNDgzZyppm48bPFghwA03ALkKCuz3qNvZ+XqmCebjtdeQ8U3SnDlglvbssecE0OevvBw28ksugYgtidfXddfB76CgQGkOSJVTP3KE6JVXnPM3OQ2/Pky8PGQxOKZgEBapJUsA9rffDv4nLi4y3H/27FMHc9mv06WpUzF8+r2fv2SQEUogIxwmw3QTkUEHSk/Qf7YvoW+On01GKI4okEBkNpBhhMgiixbuXkg9knvQiO4j2rrzEVTuLadQWGnPMhPbPrSuA9DbA+kZ2IjwuaYGm+Kvf61U0kzBoFKjWhZ2Io8HekFOXcqlLrlWNRE2Yz6HHdJ0D2cnBykm3iVTUlBtLTkZ0uCQIThn+XLskrwLSTv6jBnQiRYXYwdNTwfoJiVF3juR6sPw4Ur16+RrIIHJ44Er9Jw5GDtdeh80CGDO/ZN6VF0S5jk4cgRgrpcblZ+l9uKttyBWcn/uvpvo3/9WToXFxVBH8zVYRR4IgBniPuk6Xr4Op1ZNTITJgK8lx+yzzzCus2bh2ocO2e9L3mc0x0Ynfw7ThPkmFCL6zW8UAyrnmf0WiopQ7vbKK1WZXr6O34+88336EN12G+7F5aKiqjj6/W9VTiLdoqO7C+h8iW7ej4kBvyylc8tCkMLixTh+506iRx8Fn3frrViSX3yh+LpTKarIxweD6vE63aKMfP6Xhipyyf3R/5G3gcg8WfbA5bbIPdpNNJ7Itfd6MtdeTaGh/yTK3EFEYXK7TjoFngF5A14KhoOUGJPYatntLMuiJ9Y9Qceqj5FFFsW54+iHF/2QctJyWqX906UOQG9PJJ9eXf2t0wcfoIgI09SpSPf5jW/Yk7QkJir16MGDSuXOm+6AAZDc5WYrGQwpresSe3y8kpyZ4XC77alTZVtLl0JCD4chQXP/fvYzgHsgEJlgxuXCrszOZbJfurd8cjIqo3XqFFkZjPu9cyd8DW65BW3cey/U4+yWLNurq8N43XhjpLd6NBFRV9cHAghrkxECbPfm++O2Nm+G2v366/HbN76BFKqc88CyANLf+paKpwoEImvPM7Oxdi3R+vUIg6yqgvpe90mQmet0Rk4HeJmvX86xvHf5XWYmGMpNm+zV5VjjEQ4T5edTaMVH5LpnNtUNHE1PPKE09HKYo6UG0N87VaHVeeHt24neeAPDxsszGMSL+S/2ESQ6NQ/36dMx1GvWoN0pU1o3yrUlFBCRYqzgag+Um0s0+454evFFovDJ3E0D+hPddjN+v/4aD9VUemjrttlkmX4yDYtuuoVoaNfTcwYMhfB6Ycs/Kb9sN90+5E6a0MchWdhpUCAUoEAoQEELz3Gtv5ae3vA0/Xzyzyk1ru2cF9vJVHcQGQZEiccfh7rSMAB23/xm5I5EhN1HVjlbsQKAJnePhgZI99yeZUH84Hz0hoFwq/37VR/kf12ii7aT6vfB5+rHycQqfj/+cwnX/ftRRIbj5U0Tes7bb7dL8BI0pJQcGwtVu6x6J++HzwmF4LqcnAxVdVJSpN1fAk9xMXK58/0w06X3g0kHRXmver90pz2W0pnYXs0UG2uv/rZ3L/om25D6WXao83jsKVj1eUlOJrr2WqjYuf+8RojAABHhGCKUid27V2XH0800Evg7dYIpYcAANSZLlkCDcHI+llhXU8/gMHr3cZV234mflAELulJJnsPfO1mMiDAkfr8aWn36ExOd1fQtoZgYKEUmT8bntLSIqLuzSocOEf35z0pDkJsLV5X2QIZBNG4ctqA33oByZs4cNdbx8dCQ1L2QSHv2JNJVVxFNGtuyvAFO9MYb4JG9xs0UsBro3x9kUvJ99ojT06Un1z9pSzdsGAb5Qr42Dy3sAPT2RD16QKL5298AdHPmqI1Q0po18BwmUruYz2eXAoNBSHdHj0ZKsnyeYSgJb8AASPjvv6/0jU7E59XUEP3pT0SPPGIP8r34YqhaP/3UrgvVpT1+7/WiGhqrt3nH7tUL9uRu3SKv78Ro+HywI/fpg3b+8Q8lgumq/4YGXDcaSUaBSJkp3O7o4zJxIjyZWL3tNGbys9N7J7rnHrvDomRYePz068h7lb/JzyyKcmhZnz5YM2++id979cJ6/PGPlWln3TqYTWprMcbSIS4cBuM5cyYiI2bPVqJtXJwqsMOfT/ZnhXUprQhPpTvJYwu20C0l0W5FKlWcyO8Hn/jwwyqh4+DBAF12GznVVAHNUVJSpBXpVGjjRlhRBg/GFBFhaW/eDD9FZhZ0Ki4mevppaDj4MdIrBLc1uVxw28nOxnLRk2ympiI6tKYGS+ZMmKG6Og7gSSeyiOpCqBb8rW8hP8CZ0M2Db6En/+alklIsHMMgCsebRBNTiNqmmB8RdQB6+6OBAwHqoVB0VtLrVdK53IlWroSac/RofOanmYHIScxh8SQ5GZJfQgLAkat0Edk3bemUdeWVAPNdu+CsJsF47lw4Q+3erdpw2q0tS0l6Mgd6XBycpf7+d/u9V1ba1dQcQkeEHe2xxzB2NTXKj0BKrZKxefNNqODLy+3AT+RcVET2MTcXYWZ8Lz17wlB7550qJ79lYexvvhmiWkkJbMkpKRBVVqywS/pDhkBHK+m995R3vGVBjPnWtwCy8+bhGKlxcdIarFypwiDluGVn2/0rJk0CwFsWdtZdu1SGFsPANZcuxX+pOZBjNnYsUU4OfBWi0clxXBceRx+6ZpA/nEALFmAZyFo2loWhuv12bOx79qiIRh2AdYmcpzI2lujyy+2JF2NjkSOnvh5L1DDAz06bFr3L54p27MASqapCMEByMqZ8/XoA1HvvYQxuuAHRqJLCYbXsiTCWRUVwLGTGoD1QXFzTTobdukXy8US4tyefVFWr09OhfYhmUpAmGF4zdXWRUaCnQ71SelFcPZH75OMRF0d027XRI1zPFXUAenuk5nRCw4cD7LZts4snVVWIO09IQJyNBGx9F9TVs/y/Rw9VKIP1nRpylaIAACAASURBVLqdlQi6xbw82OVff91eJrOuDq7KrOqX19SldV2HynTiBMCD04FF08Pq6l5OwcrI4OSXkJeH9GHvvKPiyKVNXr8Wn8vj2bMn1M9Ou1JuLuzznDWOCOAdEwMgTEzE/PTtq1Le8j0cPQoGqGtXdW5pqeojH/fUU+pe5fjKeZKiK6dV5Xvg/PJdtMDm+HjEnhNBJb52beQa4bb4WhxPFg5DHH72Wfy2cCGO790bTA7TypVE+fm03RhGy11XUKWVSqYJqVJ3uUhMhDvB2LFoKjsbQyYjBeXUSn5R3q6WPZaI4Oe4cyf4mSlTENfe0uQzZ4sKCsBjcpE8nw9AbhhKSVNTg6XFoKaT5O1MEwo4jho4n6mhAdL1wYOR20c0koyNZeH4Sy+NzNx3OmQYcK2prVX5o3idtiV1APr5SF26YEP2+SC2SCotxWZ6882RErxuU24K7OV/yeaaJkScMWOg1q6tBeBICbGuTqmppbGTd2xW1Urgkdc0TVU7WwcsfbfWgd3JTsz/e/WC9iInB0+103W5Qhrnkp82DbsoVx3LzCS66qqmRQxddGKKi4NJgiknB4DOY1RTA/ErMRHju2hRJFNEpJIHSSZDn1v5XqriL7oIDnUTJzadD7VLFxzH2gsipUUoLrYzcERqLmRFOtMEQ5acDJX+mjXQUlRVUZExjoqoW6PFUU4lUzCIiMIJEyChp6VhyDZsiOS7Bg3CpU6csPM1kyfjPEmFhXhEuGu1teCv2prq6uzDahj2nE98v/pjLUlX1kie/nymcFilgeC1UlsLtw/JL0qaMAHrgc+78EIo1U7XP0KnEe0vkq4D0M9b6t0bALVnT6Sa9cgRSJ+DBgEk2AbqtOkTQSIcNUp9lmApv+NzLAvZ0Pbutavh5bHyvbS1jhgBUPT7YWeXfeNjOneGuLR7t50J0I8zDLSXn28XWXRGxTBwzcsuU+YIPk721eWCND1zpso0MmMGxpAdE7t2bd0nWZ+LsjJEMGzfjrltaHD26NeZFf09kz6HI0aoDHJEuLdly2BUlmFyWVlAz5071XX79YP++sQJvDgF76efRoZDMjU0gGnZvRu+BaZJB6k37TbyyBDdk1PKlwuFMAwvvojpu/56AO/AgYqP5Uv17YvjWZnB33MeIg56KCpCatjDhzGstbVwuE9KUpl024okCMtlzr81R506AbQ4W96XkST/GBNjV2bplJeHbYEtVhMmRDJ3XzbqAPTziY4eRdWInj3hWdKvHzbSggK7WtYwALa8QUtVt74zdO4Mz3i5yeuikvyeCDvk3r3233RpUB4vjwmFYKtNSEAVDfYFIALrPGIEwDw3F2DAOjYn4GJHvs6dkfGOJXqZaY4IBtRLLrGDORG8jo4cUVLw4MFoLyMD/+UYcSgZ0+7dUEvLfo0ZY3daa47kOEltxdGjeMnxk+YP3SHQqU0nBJDfBQIIxC4vB2jX1alwtG7doJvMyVElhA8exNxt3Kj6PHkyGJ+tW7EOpSpeInV1NV6mSSesrvSReRntDg9w7JYOYg0NaD4pSYWSSbDi47lmjDw3HIZ/56JFqEjs8+G9zgOXlWE5tzWgZ2QAhNjthAiKkr59EVV54gS+69fPuQpcQgK2hU8/Vd916tTyPPTtiVavVo+AYSjFlrQkWZY9KMSJxozB66tCHYB+vlBZGTxiOG1mVRWk9DvuIHr5ZfW0EymD0fbtkWAYG6uMioYBVShXwGLSxQP+jtsmsmcb48/NgbplYYdNSICkJ9tiNf5ttylQmDEDPgGcf757d9w7n3f55bD5X3stGIPaWmWG4D4kJkLN7OQafDKZie1zU0G7RUUwapomnAYZGfje0tPtgM4FcurrcX/TptnddnUzhBx3+ZnnUyd9rOV46wyQZWGuWfPg90M0ZYfETz5R85eeDgQdNw7jsWsXmMMDB/Di69TVgfHJy4MenFPy6lqHk30qC6fRSmMKbbFG2m5VbtLyFmTXa2vtFX91RQXXHdLt6MEgFAQxMVgaHC3nNDxtTd26QVHClhDDwKM6eTIAbtcufD95cvRqusnJsPiwQqdHj3MfB3+mtGULHpviYqUAHDQoco7q6rDsrryybfrZHqkD0M8Hqq1FONmWLXhKuaRTdjZ0ihzvyx7JRNGzvcXHw4u+qXgQfYfl9uTvRJHtR5PMpXbAsgCKMoc6ET736WNX3w8Zgqd2yRJIkqmpsF3n5dndYGNiECZFBMZn3jyokktKIJVfdVXkPebnw+lO1kTfsAFimlMt+spKzMG6dXYpWI5zQQHaY6Zj9WqUNmVAd7uRAEg3UURLHuQ0dk7Io5s9dLMEM19cxYzblaYQSWVlAPhx48CArVjhfO1NmyDB33gjmEefz26qEahbY3SitXQRrbYudgRw7o5cQjpfKGsCyeHRb1le2rLQLQ4o4GOJIl0y2gNlZ6NWkU5f+1ok3+1EnTs7n38+0fr1eHTlVrZtmzO/qMsV54K2bcO2axhwhONHqj1QB6CfD3TsmMqHLY1sXKXikktgPNu5E8ewbfPECXsyF7fbWVenk9zs2ZjJT1dWFt6fOGHfTWX4mG5/53a43VBI1cFmqq0F+E2YYJeaL7wQ9/jf/yKmZ8cOgMisWRgX6TVkmgDwb34TYL1+PdTG69dD6uTg08OHoW5mBy6JJrt2AbxHjrQzPYWFME463Rv/LyxEKtTSUrhtv/MO+kcEDcLSpdAdyjz2TSESVyPTU7E6aUIkqFsWGDzLUqGLloX+Dx2qDIlyXrg9KQIXFytbuTyeKSMD98K7La8VeS+mSfVWHG0wRtNSQ5Uc0yMi5eatfy83a2md8XgUDyuP56WrKz2cFEr6NHyVaN8+FURCBP65tRzGzoR4i+D3lgX+Oy1NWXZ4uTc0YNsbPPjc9G3vXlWh2jCg4JoxIzIxZVtRB6C3N9qzx54gxuNRUhZRxGbZuLqvvVZl8iKCrvEnP7EnJenVC5IsFyMxDLCXuv6OQURu+KmpUJXPnQvG4C9/wapmEOD+EDmHf8ldVrYrd1mfDxJhp07UWPu8rAzgKPN0fvKJssGzW7Bl4anKzQUgL12KMXjhBfyel0f0PZRppI8/xm5GZGc0LAsqfo8Hqv8uXXAety/FAYkMTJMnI3D6+ecVwyNRJRAAQxIXh/4nJEC1fdK+HIFM/F+Oma4VkAweU0ICNA2lpWBmZFuHD2P3MwyIc1VV9rmX87VpE3TU+m/cFzZ1vP66YhxkP2JiiNLT6YSRS+/7bqJQtZ0P0fk8uRQ6d1ZDEgop+zgfn5uLYzi/kr505dDx93pQhRzCqiqlXPkq0LFjSFbDGY+JENPdGlnUzpTkcuelftddsLg9/jh8K/kxKC1FLqf773cu3NjatGQJxo77+O67MHG0hygJog5Ab18UDCKOt7xc7UBJSfC65hXOQZVE6n9NDaTYzEw7i52UBN0VZ/G6+mrY4Lk8q8sF4Lz/fpwrz9PLsN5zj3raLQtq7BdeUOlFJdBIAGTidnT3WwliFRXIq96zJ3Kyd+0KUGEvIfmkL1umcqZLcD10SGUxY1ZfMkTymjqicFs+H1yrOW/64MF2sUAXG+PjoVKPjUUczfHj9vvia1dXI5UXETQd3/kO5mTBAogeTiKjzgCxfo+zsCQnq3A7IrwfMQIaDM6XL9fLhx/CTXzAAIwxM31E9nkLh8F4xMbaiwBJUTYYROpZjrnXReH0dKLf/pb6EtEl74FX0nkCffiJcNkf/ADTYFmY0kWL7OFYSUmqOlpDQ2Rbchh5WuX3eh6enTsxDbfddmq5288Hqq+3871paZgLBnN+HT8O/rWtC8IkJ4Onl2uCrYrf+hZqQsngmGPHAOr/8z9nv29yrerbQHugDkBvbyQNfEQq8TF/JyViXlGbNuE1bZoKP3K7ke/yf/4H4ofXiyTPTNxGYSEkLFkm8447IBnv2KGuUV6udoXiYjjiyQIjuiOUvBcm7rvUNjj9fvw40ls9/DAc38rK7CYH/R4YJRoakEpKXp8lWBnTxO7S/LvUKPD3Lhfs988/j7rknL3Of7KqhNutvJcmTUI1OMNQanmndLvyc1ER4vgfegji57vv2vvN48T/uRDOD36Avjz6KJi3738fyKYX9CbC8W435kmKOjKdMAcq6+r8UAjmi4oKVZRbj/uPNtemqbQP2iFS0cClShmQ5VTm52PZe73gNR95xDlEiR2jiHAL3JYO3kQq8pAjAWW3LQu8bkYGXAK+LBQIYPree0/JAuPHK+9wvv9QCBainJxTS4taW4t2m6qGfKrE7jCSdP5Zpqhwu9XjHR9/9rO16VtFezLXdAB6eyVeJSxayLSdTiDB0ldiIuy0lmXf3XgV6g5UTqtRl7BDIQCsPFZW+JK2cwlE3L6uX2XyeJQqmkk3fBIBOF0uHMegwt7ofI86WnDfuM19+wCCMrGN3j/ZhjyfCHq1W29VzNXQoWrnYbMItydV405jzNcpLARjlJ1t/16OPQPpHXdAss7IUF7/5eVEv/oVXJl/9KNIA+jNNwPxtm1TfZDiF9co1dcaEcS3V1+FtiQ2FmMvtQVybnRDdVoa8vNriWvkMnS5kBZg2jSoUbk+EBEk83nzVIGRkhIoNh55RF3G48HmfeON0PxzF/70J0y1LpF37gze9tgxomeeUUvuy25LX78e1ie5pNavV79LxdM997QczAMBMFvz52PZXXEFGKbWqOzWlNTbpQuk9D/8QdVxsiwoKB99FI9ltEQz3LZlnT7ox8QopRXXPGpP1AHo7Z0k0PBn1hfqXsrBIJy9uNBzOKykUSeQ0tvWSf4m1dv6eSyhyzabugciPBUPPYT/TzzhnFnO50PbM2dCUmRRzDAgEV94IdHvf28P2ePryHakMZYlZzkWTip42XcuzTV6NKRbw8Auxp4wwSCOcbsjmSGW1p3GgZkZ6XToRMxQxcRAIpfjHQohlv5vf4OPgHTkc7vBCNx0k/qOd7KaGqSQlap+y7LnsA8GVTWQN99E9jzZT7cbhtc//AGoy/frcjWaIHj5MS/GUzFpEpQvTsNvGArMeZqOHweg89IfMQLF9dxuO4jIXPD8iMTHY6llZoIfuuMOMAz6knRSPHxZSE4xA7jLpVxTQiF7XqHm6P33Ifnz47RmDfLLT5lyVrpvI+bvdQtfMEj0+eeoxswk55S3MLcbCkknu3dzY3DPPVjLTz4JX+ShQ52DYtqKOgC9vZGT2lq+JFDpuxfvhAxecod02qmkIbGlfZHXkhKwBDNp55fqXN453G5It0OG4Jx770URFlk69eBBVPoaORKVJaQUTARAZWOn3he9xKk0Y/AuL8dEquXlDhEOQ4r9yU+UWGmakJRlbNCbb8LkcfXV9jkyTVQWeestAKjsF/eloKAxg5ptrqQ3vWlibPx+iLPM7Mh72L8fKvwHH7TPGTvfORGXdZVjJBlGIuxwMTFq/OUchMNAyblzwZRVV2Ne5s4lio2lzz6DiZ01+IMGQaI2DPAVHg+k5aNHnS0x0kJjGPbuynrlku67D5YlmSaUr8c0YQIk0Q0boIqWt9SUdHg+0rhxsJAtW4bPvKwuugi5g155BUtHD95ojng5SqUSW+DONvXsSfTLX6rPn32mSgeEwyoaVm5/zFOHQnh0nnoqst3u3fGoN5Uql5Vgc+cCyNtLrXmmdtadDrLpCZ2kafneSaSQK1m6+EqgYNDLygJYOtV6dJJW9X5GMyQ5qc1le8EgQK53bzydo0ZB/DpwAPZ8Pqa2FpnJduxQjltOErT8Xg8uzsoCW71+PQqD6GMrz9fHlneI2loF6IYBFfYrr8BuvWqVyjv/9tvKk4vvYdEiqL2ddkyW0Fkcdbo3fS65nr2cSzleLSWp4ZFrS87bnj2YFyLcox6+ZppKP15dTfVWPP2bvk63dulLR/MB5rL61969CFCQCoPaWmXi58vqAR2SB+Pvdu4EH3XzzfbbSk+3PzYeD5Q5jz2mHKm6dcNtpaREqtzXrIHFoD1UXmsNiolRDmZ8n6NHw9VmwQJV6TcUgg/ld75DjQEmzVFbaTQMw15I5+BB+/rQ/W71qE8iKKb0rfbAAaIf/lA95kOGRK9Sl5HRKrfS6tQB6O2N3O5I+zMTixuShXRK5CHf6//5GoYB3VW0VKU33AAWfv587AA9ezZu3I2bOV87MxN6zM8/B8BJcNAlYpZSKyshlX/3u3g6+vVTiUmkRB0IqNAquaMvXgwWmaVmvmd5XHo6wDw3F2gSLee8bvMnijSOSak7HAZSxcfb+ybt+aylYETT51JK6fyb01xy3/Rsctwfvv8ePexI2RKSYyeZBH4Fg/awNj5myhR4/ufkwLh9+DD5zDh63bqNNjQMp/2/xbSxwxRfgr9jeu01lVM9Wni/k3LKMMALcjrQBQuwhCZNwufZs5UFp64O0innDyJSUpacAh5Wnw/HJyaCaSgoUAqoBx5onUpd55rGjIHtme+xWzcwM7JwnmWdWmlRVtnry/RcUFER0V//qvwggsHoykZd2SWZED3ak0hFkJomomKTk4muu+7s3k9rUgegtzeSq1ICGxGAMy8PntT8u1yxuq5Sl750dXlxMZy8brklsh+9e+PldiN+PTkZgPXGG3jqp0xBEHAohN3z3XcBoGPHwt7Nu+2aNfb4GCIFTOXl9h1E32HlGOj3MXSoqpjmxAAZhopL37YN/dCPGzEC7bz/vrLD6xK7zjzx9Q4cULpiJl1zIpFCvyfeNZzmPNpcOoEvEdDsllswnlyXngj6Vpn385VXIJKxOOzEQMnx11XxUjwuLMROeuwYBchD75ozaYM1gUKWaTOn64Aph6usTFlZGBzcbpjs16yx80c6oHfqBAXCvHlox+PBY3HVVZhSpqoqNSR8K5WV0B707RtZXZeBrbYW7cqMZf/8J3IWnUvpbNUqe272Sy/FI3YqlJrqbOfV5YNTAeQLL1QlaInQp2hFBluLamshA1RXw6eCKHKLdOKd9cdSVyDKc+Sa9fmQGHLPHkQGOGWPbm/UAehtTYsXKzEgHMZuEm0HTEiAVMRPuA4EQ4dipXPsua7G1cHS51NPRjQaPBjezkVF2P2mTsVump2NULJQCMa0w4cBoDfdBNY/Lw/np6ZCvc6plYhUf1jPFxOD4665BglRZB1uHQx5TPr2hQG0piZSE8DjUVkJg2pNDXZnHUB79IBBNTYWWTbKyiIBlY/n/zyGXq89aY8kaRB2u7EbrF+vjIwpKRhXLnattyF3G/ZcknMZF0c0fbqKFUpNhYH6o49U9jsirKWkJOy2CxeCsZGZ47ivhgGRdPp05VR34ACO52NlP8vLG3MlhMJEK81LaaV1KQXJQ6YD38LUpw+GWw6TDIrgR2DfPrtTnGniFqdPV+1mZcESI5OMlJdH5slxYgYSE6FO7d0bhUu2b49UZn36qUpzz0Nw/Hh02/3ZoM2bEbgi65lXVLRe+7pSyGkpRqPu3ZFDfeBAnDNgQCSj09AAHrKmBsvw7rtPP7yNLTt799oDbIjsfpz6Pcg5lWuNz9dBnYnbqKkBI1Ffr6oat2fqAPS2pI8/xs7BabDYkc3pyerUCWppCXY69e2LHaqoCO1wAQ69fCoRVnB6un2H1WnpUsWG19fjnJgYAMTbb9udyLhtqZNcuxbn6nXNJQCz9J6ejnscMgRssexnNBsvh1TJ9LbyyeRCLVIaluOwfbtKAjN8OBzb5C5OhB1o4kQgBefSl204iQOyn5YFTUhMDHTCMTEA9CFDwBB88UXk3Oj3OXYszlm3DlqF9HSgm3TJ3bcP/gZyZysvx/2tWAEPMDkPPP48LnFxMBzHxMBLiuvBO0npJ9+HLYM2GmNpuXU5NRjxZBqRU8XUtSu82vv3V9+xw1ZJid2N4PBhu02UCLwsAzoTV3Yliu7UxSXoP/xQLdeEBCXlXn45poFD3Xjj58IgcjhNE4/BSy9hKFNTYcM/W6FLxcUqNSvf55Yt4KXlOJ4uXXghHqENG7BUxo49Ne1D377RM6QFg7DWbdsGLYzHA575ttuaLiOh08KF2M5CIWRldsrtpEvpTo+S9NFw+p0oMp2GvE5ZGbz6ExLAO7dX6gD0tqS9ewEUulTI/2Ni4DDm8cBTZ+RIPBU68blffAGDD4N0QwN2gIYGtDVyJEBu/34AQ1YWnmqd1q2D1mDHDjAb0nP+iy+UnT+a/Z4IT/LHH9tzgUcj3infe0+lodW1D/yf29q0CSVN4+OV4Uumb3J6evX2jh4Fw0IE8WHnTiXBSmnf48Fut3VrpE+Ck3u2/M6yVCnb+nrsbmysHDoUYyznUe/zwIEI28vMxL1OnuwcROvx4BpOYsbu3cr3QV5L12oQQZOyZIld0pckjt1pDqIloZlUbaQQOVgo5CnhsD2jMRGkHbZxl5XZl1JMDJZobKxKhPivf6nLs+MWf46WqygmBpJkTQ2AS7cRZ2cD1H2+yFuWkj0v/6VLASyhEKYjKcleabc1SZecLQvTfCqA2BRxadHERIDm9OmtkyXPsrCENmxQdm7OeuzEdEWjDz+ETMBLV3+UnbZMXZEnj9fdZHSmkY/Ttx6+1rFjsM599hkK5fTq1fJ7OVfUAehtSfrqZolOrsb0dGzoLAbMnAnW9/DhSODavx87zvXXQyT6+GMlnRPhyTVN/L/zTmcp//PP8SRx8hImvlYggB3t61/HsUxJSQAfpvx8u5pd3p8u8fH3MmVYtLHi4/PzwQCw95MT0+B0HTlmst3PPoNOuGtXADvrVr1eSL59+kTuKk79kvcp75cNcswEde4M5iWazo/fjx+vXHqbQo4+faBlYFGTCJ5QvXsDeY4ejdRHyzFqaECqsPJyjK3cyeTu1qMHUXU1HahKo3esWVRidGlcrnl5mAYuFSBvp7QUFoHERLuNe/JkgO1HHwHc+RyPB0Uvtm7FVNTUqAqvRFDs5OZCoiwuxvc9e0ZmkwsEENwgpei6OhTC4+pleXmYcv2R4nvgVyAApQ4R7rehAXzlzJktV1WfCjk9OkOHUou90FtKre3Rb1lQ8HDAh5MCKxpt2wbmjQiPHfPXRNEDgLhdXYXeXLZnp/P1Puq+qPv341VTA7/h9ubt3gHobU2SXZSrjUWaNWuQUouNT/36QVqUekk+nwgS9KhRWHUrVijA8/vx2TQh4W3Y4Cydb9miwFyufPl0+P04TuqxYmPtsSTy/nQp3imRSkICGJEjR6BPrK6GdJ+Sgt2dRTh53tatkaF5kvRYp2hjb1nQ+951F0DwF7+wJ+Tx+VRteSbDQN9YU+Hz2XOay7bl9bi/usFXJryR58pxD4XAFIwd65wBIzMTY8ftjB4NMZcIhsA1a1R1ut69MdY8h3V10ClKs4U+dpmZRDNm0IGVhfROVR4dot6Np7PvZDAIK4dTkryjR7FsJKATARA3b0bX5HlVVZDeWcqTPM/RowDwyZPV1FxyiT3TWTgMnnbpUru1oaoKzmayHGn37lCClZc7A7pOTvnhW5u6dMF0cqnO9HR7yYX2TDk5eGSlRiTaWEratElZe+RW6AS2RM7e9XJemptLKT/p13JqR/L/iYkI/zvbqWZPhToAvS1Jj+vVATTajiFXnpQC5fnLlinAkIancBjfv/++M6BLEHFiNkwTO9/q1XZ2uLgYdvWEBNQs10HN6R7l08XX6NsXGc+2bkUfBw+Gh/3779tLpXI/5ThJ7299TCQ4cWCuZP8vvxxSLsdbW5YC2aoqgKEU87iiGau6KyqguSgqgrYiJgb2a+4jkwRMp1h+7iv/VlCAMdi3D2izdCnGoXNn6EslEzN0qB0tDx5U/gjduwPtSktxHzfeiOxyPKY6eOsag5QUiHIjR9KCj0bSnpB9Kn0+KEwkSemGmysvBy+qqyuzstRw8Wd9mJwSz0yZEj072fr16BNPqR6QIOmii8DTfPBBpFSoP25M0fjE1qLRo/FasQJTJgMWzhV5vfbMwWlpdkWcE5kmeOOaGmVRyspSFfSi0aFDeIz0bYffMzmBrv4YMTltMU5zycdyOJ5uq9cl+HAYW+CYMR2A3kFEkIJrapS0KkGTSZfQmJKSAHJVVeoJ0bMncJ5D/enQJXqd0tLgSeT1qnNl6lKPB6v4/fftUqzLBbFpwQIYLblinLwX/T7lvfr9UOGnpIAhGDkSr7Iy5UXP4Xrcf/mEMgDLe9Njpvjehw5Vdby5H9L9Nj1dMUM6yPJ49+2L+5R07bWoQDduHPS4f/6z0iyUlKg2wmH7rqHvSnIOP/4Y87FmjXJxfuMNzP+QIdHdhouKwGDt2qXGefp0MCH9+6tsLvq60A2KcXGYk/HjG0MReUjkstA3R9NUaUXl+yNHoERiQC8sxFCPHas0/QkJSB7o86lsvTLEKhTCMXooVmGhsrcGAkg8w0oTHk5eDk7x1kOHYglWV9sfJ7lc09OVoxr7lZ5tasskN0eOYEnzku3Zk2jOnMiKyzp5PETf+AYegVAICWuaS5G6ciX4VrmedAmfSbd/O3mw83H8X86jzhzKQBJ9K5XX1LUGZ5OhOx3qAPSWUnk5HMQsC4Y6p+xqLaHKSuwIS5ao1UtklziJ1GpNTo5cNePHKxUpU309njiPB9fw++2rlK/R1I5GhNCx2lqAJ5dPlZt+QwNcT7mPzNZyu/v22fMqMlglJqJvnKxEMhs8Lk8/DT3j/fer9latUpnaiOxPlVN6WwlIcix1dvv66/HSKT4e1R/+7/+U9GpZKgsH3280hujrX1fvH34Y/0tLiX7+88gSqU59ZpJ9feedyHt0QlFJK1bYHe6IIN2/9x5cjceOhWghPY6YMZO72oABCL4WxMs02mYbjSwLyoYrrsDn4mJERB44oDbIO+6AGt3lAs+bnIzh4+64XACGYcPAPzEVF8NprqAg0qLD73lJderkLFUtX64q9cqNOzER142JQdKaefPA72ZkIEf82bCftwcKh8GjSwA7fBg8+/e+1/z5CQlEP/1py65VWWl/xHT+33otigAAIABJREFUn0mXlPm/02MpHxO9Lan14XO5Lcln66Rf8+hRrIucnHOXWKcp6gD0llBdHRzRtmzB5+nT4aB0KtUMmD74AJstkbNhSK7CpCRVHlOnqVORZYKPf/lliD0jRgBQ2fAmN2tepdFAhOn223HPLN1JTYLsN79v6jN/d9VVEDVCIZS9ksG18ikpLUWxD9mOU3+jXUcnp6QzXi/0u6mpzhJuSgrRb36jPu/ejdRUbIh1uSIrmzGVl9vD6Lp0ibw+Mxh6v9lLnZNk6787gXpT5KQjDoch4WdnY2397nd2kVNeNyamce1xPC4ReLL4ePAnekywVGa4XAoIOb+2bP6NNyBVM4XDSALTty+6l5UFXuLJJ1V7/foBRCVVV4MxYGcqqZTRJanevQEyTjm4ExIUz8lkmmAcBg0CH2+aqKYrh/TLSkePgncPBBSfFw3ozoTq63EddjiU86VvkU5bgr6tNffoNKWQkvcX7T4lU/Dqq1jfv/hF+8gi2AHozZHfj9XGIUvhMEA5M1PlmjwVctIb8fdOElhTT49ceXfeidX4pz/Z42+cjErNsZJeL9pzu3H8r34FAHSq8e3UrtPvUoqPi7NLmPoTFApFFnjhdvmzLh46hY+xJkIvYrNlC+Zz6lSlz0xOdgZZIqjP58xBEhzThNR6113OY/ePfyBMLRzG+M2dC2TSiRFK9mv0aKy3TZvU907MnhwLplAIDAczGjExaJ/RSW+nthbMxj33gAGUOeiJcJ8XXEB0++3k8wF8mZ8NBgFyGzfa7d6ceE4uhe7dcVuLF+Oyfr+yZjgxA/pt9euHPOz8uxOg/Oc/UNfLZSCXFg9v167gYaKFfV13nQpv42ELh4n+/W+c8+CDCCTg61dXty/7aWtTr15Kba5Lpq1FgQDWBhdSJHL2b9UlcwnmTW2T0odDrwGvbxf8nf5YOvVL56mrq7GuWyPs70yoA9Cbo2XLoH6W6nBdD3QqJNk7/Tsd1E+llA8fy7k0dZIr0ck4xPZww1B1sC+9FN7rTmytvtqlWOZ0b36/Ag0uuckuxbppgJ9SCeqyD/JJbM6xzCnbBM/lhx8iXooIWomJE53H1ueDvfrRR9GOx+M8N36/3Tbt9xM9/zx2Rb4HqS3RmSsnfaKU5hkBXS779Rsa4Pz22WcwF3g8cL+trgYKc5+YAgE4xPH15NxrfQsZblq4EJ7HkjdYvBhDxgBXXg7gZbcO7v6xYyqsjAhuC5xLhy06kpyStDSlCAsGlfSvD5t8n5oKO25TG67LBWtEXZ09qCEcxhJ4/nkUAMzMhGr/9ddhVfF4Ikq/tysKh+1jdCp1vPVSpfrSO9N+ffCBSjipbys6DypJ1wRJ6dxJgRXt0dLXjVOAR1OSOj8+f/4zFKo//nFk6OS5pA5Ab46ksTBarDBTS3RROmsp38tNNy4OtuSUlOYlddkXqX/SdVDSk0S2t2IFnNyY/H7susuXY4eT7TfVFwYeKVHz+/fewzUsC/f1ox+hPyUlSkLk8WEQ0xkcfukMgLw3Ocb6fcr+m6YdoV57DVK6npC6oAClQVnSN01VPlVv+6WXVPkqbpfLnEoNhdMcsObASWy47TYEXT/xBNrj4t6JifDiev55tU5ffhl2/NhYpDGbOVPlzNRB3Wkutb4tXQp/PHZM4yEPBrFEWUJNTsaQ8FTq08VNB4P2tK7ysh4PbNR9+lCL6a23sFR1AJeq9oQE9K1Hj+bbi4uD8uXZZ+3Z43gqy8qwedfV4fMjj6C/P/pRy/t8rmnNGhT9I8J9fO1rzu4jTpSbS/TrX2OZrV2LRzcurnX69ckn2BL0cg7R1O1OqnL9cZJbgpOri5OGQd8m5fYjSb+2/N3vh8/qE0+gHns0i9zZpg5Ab44uvRSgs3Gj+o5n8k9/Uht4TAzRD37gHIvdHDmtsoYGVCIwDMR7/OQnTaeIevVViBUczCtXtwS71FSogdmpb80aFduj359lQd/ITmnRQF1eR2eX+XdOGhMKQS39ne/Y80bqLLQO5kwXXwxj6cGDzupoJ71cU8TX8nqhUn/oIRXMfPw4JFl2lebckJ9/DpDUayteey1i56VhmOePHeRk3zg074ILEEa2bBm+k8fNnIkdOBzGGrv1Vhh0DQO5Bl54QZVUNU1I6a+9BvE5ORmvu+/GMS+9pLycoo27+G5PvkUf7HcGYKfh5RhpbiYzE0qPJUsihX/9ckTgP8aObdm0MTU0RErofJ0774Qtnkj9bwmlpCgpVF9ipgkw5zT+dXV4tWfy+ex+pf/9LwDn8subP9c0IXGmpsKhsVu31unT5s1gxtgtxUnJ5iShN7Fcm1xf+vYqf3f67KT214/TFYREePwfe6z57fpsUQegN0fJyfD8bmhQukLOZrFvn11v+MwzcP/s3Nm5rZUrYSzSVy+RvcIAr1AuPFJfDy/pnj0jPYKIqFEn2tBgNwIxsS08ORnxQAMGqN8aGuxFRniV8nsGehlUzCSfxKwsSK6rViEsKiFBebnoT2hNDaTK2bOht2SvG6m9cFKVjxkD57oXX7SrornPponr3nWXYpHlDrBhgxp/Xe8WDqNfL7wAtXXPnvheFvRmKTsYhDS9b5/q26WXIiBa6ih5hzFNckzvlZGBua2sRFt790YyQl26wCHvxAmIAPPnQ3Pyve+hH1xrnfum1yglQhtdumDun3lGVbmLNpeGQYeMXvRm6AbyCeUUUdPhOhkZRN/+tpqK+HhEz+lTKaUpy4Irw6BB4KNOVZ0rFR88ZDzly5fjcWmJZO7Urrxn5kVffhmPPD9iTkqg9kRbt2IcpNLM71f5plpK0fJGnS75/dh6dIc2Ccq6t7keG84kQxqd5AxdFonGz3JbTluP3o40Q+jX0x+/c0kdgN4SyswEqHNFh9xcbPy6HrKiIno408aNMBjxBswkwc5JwuRVVFYWXY9TVaU8o1lClk8IkcpVuXAh4pOJILb06BHpBcKf5e4t1eeS+EmKjwfLn5sL8JCMCx/HfQmFMFZ+P8BFfxINAwzCjBn2a/XujTh5mf1B7gJEaHPtWmghdMNmRoYq/sJeVDoyce56Hg/pEyDRw+9XseVEmNvEREjplZUY43HjcB/R8kMWFCCSwOeD5Oz1OjsWlperePaKCiWS9u6NMf/gAzuyRHN6zMlR+tJophnTpCKzG71l3EJHrJ4RTTXl9+hyYXglpacjacu6ddGXdo8ekee1lCRfJ5evZSFg4sUX4cx2qik6pRpXxjhXVkZKlGerOEtrUHa2vd6RzgC1JelAyv/lkpRRqtLTXpJuzZLtR5PW9aXPx8utyIl5aMl23Zbhax2A3lLq1cue3opZORnQ2NSTUlmJzViuPB1MnFajbJMNePJYTu3Jx+m2ZH1VHz2qVl1VlT3WRz5NfI9Mut1aXoMI4LZqFVyFiaJrIgwDuqipU8Eo6XEivEMmJztnsiPCuZWVqgyTnAfLgqQ7bx4k5nfeiZwXDpnTn0Z2Zps/H9Ksfp/6riGZp4oK9ImZPtOEd3xysvM97N2LvnGlPadSrKNGQR0vry/7FB8PbcuKFer3Hj3sOU11uvJKuIVz9hRN01FupNE7xrW01+jfWApV3/QsC7xKnz4AzLIy8E533w3V7PHjULzw1Eh1r75hjxx5ZpXDxo+H493Bg859PXyY6LnnIPkbBpRnc+bYQXjjRvCAvEQnTsQSq6jAEpO+nvo0tPUG3hylpWHb4kq9RBjvUaPObT9KS5EngOUOXu76kpfbBP/G7iu8nvg4ffzlOfrWye3KbSLaXOqP2als10SQ2V59FevsXDNOHYB+unTxxdjJKiuxSjwe7ATREs44rV65YqIZbSSI+HxKzcvHsuFS382k5kBf5fydU01vpz5JtljqveR1vV5lO96yBQm9ZXuy/QsvRH76aE+Vk/gnadgwPDVvv61yRcp2gkHYuSsqkLWEx0OOgf5US11fYaHy7CKKDMLV+6fPrWXhups3A0luuikyTVZFBdToTDozxRlYMjKi66yLilT+dX7V16NdaVaRNHYsXNQZZcVOVG10oveNK2kLjaIumQg3KyhQ+Y/krQ8ciCR227eryrr//Cdul5epHDJ9iPiWcnOxhC0LoXFVVXAnaGkGtn79oMgpLMRwl5RESmUHDqjrxcXB4Y39G1mSP34cv4fDWFrXXYfhf/ttFYwhp0FfcqdCgQAYHr8ffgPReL7WogsuwBhJQG8u01trUk0N5laWP5WPUjStT7TH02lr0x9NHciZdMuhExPo9J+J29YZA0mBgJKTzjV1APrp0oQJWKnLl6vkzzLTBhF2iQ8/xPuiIvzXVyyTk66JWUZ99fK5/HnIEDjnFRXZV5muE9J3JL4uf3YCO111r7+XbTAdPqz6Iu8zLw879bRp8DzizHvy/lu6S06YAMlUArrsU0MDnMZk3/R7cPpeEo+Z7hWmX0uO3YcfYt63bgVocix4QgKYvVmzwPw5sfdZWcpRcMwYZfwdOxb36vUCMceMwfuFCyHpy75UVcFXIzERx+lkmmjv449t6X3rjQRaTZNoNU2i1FRIqJMmoQqZz2d33ieCAubwYdVEOIxj5WWceB592LduhdIoHMZ7tqvedhu+W7xYFcGIJgkPG4ZXaiq88svLI6eTr9vQAF7PSTHEx3CZzC5d8Ght3WrP7y4f21MF9cWLwXRs3YqtwuMBqLeW57gT6crF/5+9Lw+3pKru/e2qM95z7jx19+2ZqRlaJkWDMohCQ0OEICBg0CgmaqJRQ16MyfOZl5f41C9xNggJokYCDxkkKCjIICIgQzM1oz3Qw53vPfeee+ahar8/1l5V6+xbt7sRuiGm1/ed70xVu3btvWv/1rz2NVWroYXLJpvJi4qojTp+Iek56rs8N4ppsP+PYtxsWmi7XmiL3le01wBdKZUCcB+ApLnODVrrzymlVgG4DkAvgMcAXKK1riulkgC+D+BYANMA3qO1fmlv9e/VoF+86U2YrFSAQgG+UnDicaRmZrAulULR83CPCcodzOfxVgYXIJqNXL2aEifLGpLAfDs2MD+mggtLb94c1ljndiXZq7ejg9j1ycmwLviiRcRiTk7Ov2HbAGWvXm5/xQpqR0qg3E+u1d5sUgy4zOAhx2ZPKGo8op7QhRgS+c5PZdR9Rh1rHzc0RK+f/ITmgY1/jUbouZ5K0f3GYiQW2mO5ciV5sdu0fj3N7+ws6bZPPZU84jkXvc20TExQdrsoQFeKPOcLBTKs1uuoqRQe0W/CbWo9NBSUIuXTI48Q3yX9NQGa3my21XfSlpyihjuKtmyh4ZJ54R9/nG7T90lVzHXHd5fT/IQT6JY45b1cklF9WWiKgbBM5tq1xIc++eR831Hg5aekePTRsJihUjTG7e2UePLViu9+vVHUo8O0K/mD5Zmo9K/yfPkuz7Xnn/+Tv9vb8ELMgTzOTrMRtaZKJbJAnnzyHg3Rq0Z7cwnVAJyitS4qpeIA7ldK3Q7gLwB8RWt9nVLq2wAuBXC5eZ/RWh+olLoQwBcBvGcv9u8V0y/zeWw/+GAoAB4ApRTay2Wc6PuYqNdxp1LAMcegf24OpWQSSiloAAoAr4HDq1UMNZvA6afjgccfx1w+DygF7ft0vNZQZiU5vMIBQCkoACuWLsXBAHnen3QSBZxu3046T/ZQt1d1Mkm6zmXLqKjv449TUChAxrVSiSRNzvNtPwVMUv0uxaejjqI2br89lKC53BL35+67SUrkKAH7icznSZ9rx4VLitqNF3qy5X89PSR+jYyENmy+P/ZI59ymsq2F2HutWxOE28fycdVqqLGxUbK7e+HckY4DvEc8CqOjhFwSBaWBcXcMUSxGgciPP45Gw8dGfThuwLvRVDFA05D8/Ofk1dzZSaArb+W442ij+ru/my+5Ril2JB9rk5wayQP98pfhNQsF8vvr7AxdFGx6+mkaFs4HLnNA8WZs5++2ndnsvgAE5G9+M0nS8nHi++JHaU/JXhaVCj1669btO0B/4YXWyMqjjgozFO8t4rGVka0SKKMA2952oqRg6a8qSeaU2hWfLo9dCPztx5h/j8qfxTQ3RxHBvzOArrXWANiBP25eGsApAC42v38PwN+BAP1s8xkAbgDwTaWUMu28Lkmb2fa0htYajlIYSqXgGuD2TdcnOjrwo7e+NQBzSTva23FoJoND02lszGTwxNwc6paRR5nrOADaYjEsMgGOSikgm0WLT9E55+DpYhFzN98MnctBA3ABHFkq4YlMhvrU3Q3nXe+ia8zOAqtWYdGaNTjA5DXfVC5jvFwOUshq4yymAbha400vvIA4e6oMDBBoDw21pjh961tJ5fzkk/T91FNDu26jQTu0XaxEStz5PDEluwJ0rrkpmQIbSeSTzdfo7SXGpVAIGQ5+OoeGSEp+4QXSUnDVOqC1LRvcN24kbYfchRg5bBFAqdYSYkoRE/FyUwnbu6TcmXb32CgFb2g5Nr+o8QP/vSjrFFxLAhkZIXW4tIbIppcuDYuoDA7S/6w+51u0N285DPYmbitp5PvsLFm3FgL0++6jWjTSK1q2FRVux5+zWQJTrhkk/9c6TBRpT7nvk9Ll0EMXHuZNm0hhwsdzZCm3zxXM9pVj3fbtpESSdaGy2b0L6PE4PVZbtoTzzEvffjxtiloXtg3cBnp5nGxHPi42ODNFWTrt9STbiFpPNiOxL2mv8oRKKRekVj8QwLcAbAYwq7Xm8gc7AXCU6BCAHQCgtW4qpfIgtfyU1eafAPgTAFi+Lz07IqgvHsfOeh2+AfSU4+CPBgeRMay/D0BpDS0lbfOd6dG5OTxSKOAdXV04JpvF86USqmYlaK2hQExADMBAIoGV6TQutQJCh2s1TNXrxDAohRsmJjB87LFQjgPt+0g6DiZ6e3F7Loem8VZXY2NwYKR+AAe3teHCwUEsSSbx85kZPLFmDbQB4KZ5QnzfR7LZRM1x0FksItnRgdWnnIIXHnoI+oIL0BWLYZXs2Lp1LdkrNpXLKHoe0GhAr1gB3d0djE/wTJunqWPpUqyW6axGRmhnYCdAgPKQb9pE9voodp8Rxd7Nn3uOXraaHSAHtuFh4PvfD5PhcLvyeJtt58+9veSFJUPfbHEEmL9jvNynv9lsrQfKu1AyScbk7u5dn9/WhuKHPokbvgEkCgCmWkPTHSfkOfh33ycpmf0+P/Qh4qUuv5zswK5LRSrY5UBKw3LImNrbyT4vS9Dbw8vvzebueRRZiRhoBeH+/jAxIQ8V319/PymrXniBJKtyuZUfs3k02a/paXL40pqGfeXKsD9c3GTz5tZ7kBv/8uWU2G9v2NBLJbJ6SQ3C/feTVU4uPbb07a3sZt3d5PH9rW/RmDDJx1WOiRx7ey5tsLbnkn+LssXbfDYfC7SufVvNLtcdf7eZAvtaQOjG83I0OK+U9iqga609AEcppboA3AxgzavQ5pUArgSAN77xja+p9P6RoSF8cft2jBswbRPJUWJKoTMWQ0NrlJpNKJa0gWAFsMSutcbt09PgNajFsb7WcJXCqnQan16+PFC/M+UaDdwwMYGNpVKgytcgZkIbICv7Pn40OdlyPVepFsbi+XIZN01O4r2Dg/DNdbktTzxxlVgM3z31VABAVyyGd3Z34+bjj4caHsbKVAp/vHgxFkUkth6p1fD98XFsZ73lKafAN0yOBuCYJ0gpBVcpHNTWhk8DyDebmKjVyAsrlQJOPBGLEwlkjX5yW18f6q7baqJwnJCBYlFAmAYUgL56HTXHQWZ2FtlKJXyiKxUKeSuVokVKqae1pWGtKdwtnydxUZbtAubrAdnVOpF4eVU+HIeOn5ykz11doQr/sMMoYc8eUGcnZbS6/37iXxiIozZO5olOPLFVSk4kKDU/QN3p6Aj9IWMxugbnv5HDmMmQF/mjjxKQcsSgvTnaypaFSCpSbDBOpYC//EuS8j//+VaVu+cRwHV3A5/7HIHz/ffTZsztRik85PexMar8u2YNXYfp9ttDn0Xuk+/TGDkO+Ul+5COvvpd7s0k85QsvkA/C+95H11i0aD5IAsCDD5KWYe3aV7cfkhYtomX5+c/PL8oj3+U8AtRfuS7l8fyZ57LZnO/tLnn5qLmMmlcZE2//JwE9ijGQjMDYGEUz7GkJ2VeD9onVRms9q5S6B8DvAehSSsWMlL4UAPNswwCWAdiplIoB6AQ5x71uSSmFvxaJp0drNdQ8DyOeh/ZYDP904IF4rFDAd0dHUTMriSVulqY1APh+oIqXUrwyx3fH47hs2TIopVDzfUw3Gkg7Drrjcdw+PY1nS6VA0tZa03nWzpxxHJQ9j8DTAKdj+sHr9uliET+Lx5F1XcQcBw3fDxgLRylS15t2HaXQ9H3cODUFx9zT9moV109O4s8jbME/GB/HaLVK92mAt51zI2qNkucFYwMADd/HcLWKu2dncf/sLNSaNaQl2LED5/f349SeHvhK4V/OOQfjHGJmwLxNKZQ8jzQQjkP9Y7AHkHQcnNbTg8lGA6vvuAOHP/ssOms1xAHAcTDW1wefmRLxxGrHIWZB+jGYuepuNNDG53zyk5j40pfQnJwMdpQluRzt4Gxm4J0nHqf4sPcs4C4yORmWiZJ03nkUYN3bC1x22Z5XhJiYwHizF/2LXDgOuSlcc01ruQIp9JvbDDayYpGkrP7++UVT+vpIwXHllbS59vURwD3/PHD99WEkYDJJDm5vfzsBur1BR23ejrNwRGjUeUx8npTOpUKDl3StRhaY97yHpuiRR2jYpYYiSkMg2+J8SawcyWRoemWVL9elvPKHH777qfptacsWqn3E9/+lLxFY//mfz1cS+T4tpb0J5pJs6deee/ku1wCfI4+Vqnu7PTs2nYF6d3Mpz4nqj+yTrWGw74fTaexL2pte7v0AGgbM0wBOBTm63QPgPJCn+/sBmLRl+E/z/UHz/92vZ/u5pHyzidlGA/+8YweKvg+tNfricXx6+XIck81irq8PN01OQplVppRCw/fRlMYfllbNLSuQpA5N6vx8s4nueBz3zMzg5qkpHJ7J4JLBQdSMbVuyohpAQinUzXvadfGHg4P419FR1IzUqEFSuKMUmQaUChiNiwcGMN1o4NlisUWq11oj5brQWqMzFsMFfX349shIaFIAAXHZ80hbIUjej681EkrhU0uXYkUqhe3VKr6wbRsapg1fa2yrVPAP27YFfgja9EE+iXPNJvULIGAFkHUcvH9wEN8eGYGvNXwzLsyYKKXwxvZ2rO/tJe3HxRfja8PDeFtHB97Y3o6y4+ALqRTyRquifR+OUkgqhYYZL2Y8WFsSUwrru7txujFE5ptNfPXcczFer8MH0Far4ZO3347+004DrrmGNCcA4LqIH3ggspdcgpoZN+kECa1JW7BtWzAHin9XCjqdBioVuN//PjpkrvgoKhSAZhPlr/8brp04Byd8+DB0d1NafS6CJzdJuUkxaU1+jPfcQ76U73hHqxQzPk7p4ut1+n1igsD9M5+hy99+O7V3/PHkvzkzE0pN9iYupSGlyLP+4x9f+PaSydbCgNxuMknpX2s1SoIjIxClmnXjRsrz8973kr/g5CSlUoiSzm1iiW3zZkoocumlxLtdcAGp5GXltmRy79vL5VzaNmquDCfH+dXuz9xcyCBKy48EPbn12WtMzn2UVG4fAyycxkPepy1R22vNPsc+LirgKKo9bmvxYspSuC9pb0roiwF8z9jRHQDXa61/rJR6FsB1Sql/APA4gKvM8VcB+Hel1CYAOQAR8TuvL6p6Huq+jy9s24apZhMN3yfJGsBUo4Gv7tiBv1mxAm/q6MDabBZJIxlrANdNTOBx4yKsFTnW+WJlscOdUgozzSa+NTyMU7q6cMvkJHyt8VyxiM+WywHggaV9peCCQPnO2Vmc39+PA9JpzHkeYkrBM4DsmOvSKSGYKwAxo5aG+Z1XfJvr4uNLl2K5kURH6vVQ8jVt/aZcxrUTE/Ps/BJ0lXmKPa2DV6CZMKAJAE2jRofoAwDUfR++1vj6zp2YkOKPGYuY4yCmVOBcGICvDs0IvtaImQwXf7JsGVLM6nsetONAKwXfiFQD8TjO7OnB98bGyN5viQzHd3biHf39wfcrRkYw1mzCNztlMZXCl0z8ufOhD8H3fTgAXKVwcCaDDyuFH09N4b7ZWYi9BFpr+KedRk6XYhwh7scBsDSZxGd3sU7rvg/ve9+D2rQJ18ydjye8Q7HxcvqPq+3a6Q5sKYY/80Z9000ULHDssa3XsrN5NZsE8KeeSrmYAPr/2mupjgxfX6ZvkNdUipQQn/jErotdXHJJCOJMvHSy2cC/s4VsAOB+X3cdSbmyD1qode0xkp+feYa0EZyNLpEISyn4PhWMWfOKDY+7Jgk6doTB2WfPz6j8atrOazWqWTUyQvf9iU+EvrDcL98PM/dxHxcCbyYbwG1pXUrTUW3ZbUYdE6UFsBnLXfVP9oXvc187xu1NL/enABwd8fsWAMdF/F4FcP7e6s/eoK/u2IHttRqqvh9IgnIGR2o1fHrLFigAa9ra8JGhoQCsLhoYwNm9veHq8H18eXiY7PFSqtQaXfE4TuzsxHUTE4GU6Pk+mmZDV2KlK8fBH/T24m3d3TiivR09sRiUAeP/xTZ4pfBUsYjr2P1Wa/ggrkv7PjyjZWBpGaCF8keLFuHQtrYAZHWt1gpAitTy9YjdWcl3rdHwfXyVd1mtUTfXc0RbDoRUasZDA/jp9DS6YjHScAiplSXn1akU/nBwEN8VcfCB9kNrPDI3h07Xxdn9/cHY2H3lMe2KxfDxpUsx2WgE962AYLyUUki6LuJCzLl08WJ8cft2TDYawVw2YzGKFDBqe9aMvNBo4K83b4bn+2iw5C12D991A0CfJ70bxquxm3in74yO4rmTToJ+29tQvm0V8KIKgEnG+TKwc/PiMvMA3i6xHo5zK8jt3ElS+p/9GakftSbA5DLtUefJ7/zZBGAsSLv7n9vhlx3h94Y3UPIarQmUbIlXVbttAAAgAElEQVQvFiMwvuGGsPiGvYkDdJ68rwsvpPT+fFxPTzQgvJpkz6WkRGLX9eVfCfk++RLs3Bkyct/+NpWWXby41YNc9nMhidgGV/t//r7QmuFr2OfLa9sk14i8jvThsJk5+xqsFRkfp2KNn/rU7sfu1aJ9FCzxu0kV30fF88iZzUjISpGzme848BU5pJV9HxvLZfw7Z4sDkI3FMJBMYiCRoFcqhT8dGkJnLBZIzixxK8dBu+uialaOBgADClquQKVwWnc33t7TA1cp9MbjUEoh12jgM5s34xsjI8jEYhhIJNBhxIYALM35DxWL+MzWrdhUrYbSNABfKfSY9pgceW0gaCvqSXn/okVYYlx5ffNUV4yauSLuy5dtgWzgynGC/kEpVAH8cHISo1I6h3EEBLC9VsMNU1OBKQEwYGiQqaE17s3ncRdnIBHU5rq4bNkyfG7FCnxu5Upctnw5lpp+azNOvrn3IzIZfGb5cqyzquv1xOMBYGulkHIcfHxoCB9asgQxM2fa/NfUGhXfR0OOXZT4YV4M7MF389uuqOr7qMTjKKdS8ETbySQBTio1fwPkSzsO2Vft6MGoDTFKCvJ9Cii4+mr67dZbyQlL8nw64gakg9v0NEl9r4SGhkiKl2F03P4hh5DtPJMhZoOzFtv3tHw5gdOf/imBouw3M0WHHNJaa7yjIyx219+/bwq5HHjg/PxEe5uJAIhxkx70QOgyMjxM4C49/e3lbgOzvS5YXpIgah8X9fjwcfYaX6gNu3+7OnZXjpye9/Kr2r1S2q2ErpT6OIAfaK3n737/zSmQ5IIfWmPGNch7WylyICsa1v3WqSk8XCiA7eVn9/XhTR0dWJ5K4WNDQ6h6HmQSmrF6HddNTATObtw2q8rl9R+Ym8MB6TSOam9H1fPwzzt2INdoYLrRgOs4+PrOnbhs2TIckclgXW8v7jR5Mvlear5PYXMG7Nf19ODQtjZAKay0YmtWplI4t68PN5tiJwrAqnQa55jSVj+anMRjhULgdT9r9Ku8p2kg8CtgcgAsSSZx0eAgtNZ4aG4Oj3A9cvF0lTwPZ/T04IF8HrlmM3D2K3serh4dRUHEOSnHCRwMYcas6vuoLKAPWxYRQ3RoWxvW9/TgDh4vRb4JKyJEQ6UU/njJEtSM5sZ1HBzS1obt1So42sHlMRfnSGJ/g6DfAH12nICh0AB643H80R4WqXYA+ApQIhfQG95AIVu7Umk+/jg5ibW05bQeAxBgXXgh8B//ETqUKUWbOEu15TJJwPIajkPJ8B54YH7edKYtW8gWL697zDGt4BlFc3PEDChF+dkliPA10ukw51Gp1OqDqBSB8Hnn0ThxTHU2S/3cupXyvXObqdTC1ZP3FbkumTcWLw7HqrNz71/3zDNbcxcArVJrsdjqQb4QoEs1vCQ2ezDZc8lrWPoN2FqUhZhQEQQTyRDI7xLs7RwNct1GPSN7m/ZE5T4I4BGl1AYA3wHws/8qzmp7mxwJrgDiSuG8RYtQ8TzcOjkZOpuZ47ZWq7h1agr5ZhPT9ToAUmn/aHISba6LwzMZHNjWNu868VIJBc8LHbIMKMpJYAag5Hn4oWlveTKJ6UYDs40Gnas1co0GFEgSPbW7G2XPw0Nzc1idSiHtunimXA7uyQewoVjEEdksDonql+Pg5O5uDCQSgTq4Jx7H4mQSP8/l8EA+j4LxPGKnujhICmdbuRb9V1qjJ5HAewcHscYY9jZVqwEIw4yzb8ZzWSqF96ZS+N74OAqNRjAP+WaT1OFGJR2YJcxnvubLUU+lXBen9vRgeSqFjcUiHi4U8GKlgp/lcvMkdABBkh6myXqdbPCaQgZ5HgNQZwA3n1mC90GJgbTWAcjzOuiIxfDewUEcEDE3kjiqAQCckyeBYhx6NI1Gg1SCchNavJhCqZh+8QvKwsaObkAopdx4I3DbbWQvXraM/j/+eNq4b701tKc7DmUm+9GPWoGUyffJ6W3VKgqfKxRaJTHeNHO51k2ZMxwDpAp//vlwA7/oIkrYcuWVpAJWiqRvBgVuZ+lSCp9jkv1jDQNAfWMbvutSdCDQGnPvOMR43HorpXJ9LSmV2rue9FG0bBkxR7LisIjkDdYC/y4BVDpjSsZS/h8FjnabQOizsJCkL4HfBmI5lzaDy23JY23JXTInSpHa/dpraT3uC9otoGut/6dS6rMATgPwAVAGt+sBXKW13ry3O7gvaUe1iv9n7MorUimcv5v0SWf29uK68XHMNpuIOQ7W9/Xh8La2wHkKCKUwBaDseXgwnw82c5YqZ5pN3DgxgdtdFxcODAQqXklBHLsBM7bxyjWuQSCVazRQNmFgjrkG98dVoSd9VzyO03t7sSKVwpJkEhsKBYoHN+RyW3Y8taCM6+JNETHU041GEDoGLZzRzH04KkxvC3MMAKQdJwBzADgmm8W2SgUvGN0V3+ObOzpwYDqNrlgMec/DzRMTKJrwv+DFY2WNI7TGYZkMjrJiSpq+j++MjiLXbKLNdfFHixaRacJQRyyGtOPgJRN+V/I83Dc7i6zr4q1CBNJa47tjY5gwTBsURTUM12otIG4zG8yQ8Dk8v+xHwA6XvAYavo8NhQKOymbnSfiS3tHdjVyjgfF6HaqrgZPO9NA5QwVIRkdbQXJmhoDvwgsppIwzAPMxcqPkEu0/+AElmeFQtkMPpWxkfBtKkWNbrUbe5BLQtab0/itXUhIWLq8ZJcnZUt9vfkMMx0knkVc650j3vPAeRkfDTXdiglTgZ5wRplkdGCCwZtqdhBZFUvqrVknTwEUR3/pWSpX7SugXvyCfA4Du5dRT912Y2culM86gsZ+dBU45haIpu7rmhwzaLznPQAiQMmcAf5egyefZa8NmBiXYSsbBZjCj+sa/SwZArgvbY1+e09m5b+dqj5zitNZaKTUGYAxAE0A3gBuUUndqrf9qb3ZwX9FkvY7rJyawqVyGUgqT9TraXRen76KO47EdHSh6HmabTcQVhUP9cHISW03yFAZT3rg1gLznhbHd5jcNimGHUpRJTdBUvY6f8M6pwoxqynpnoLPJEeewBH/F6CgWJxI4b2AAg4kEBo2XjAawrVrFDhPHpJXC0ZlMiwq65Hm4dnwcMaVw0eAgkgvEvLBmgtsFCJz743HcbyT3mOvibZ2dlCTGjEm35cq8LJXC6b29OKCtLQA3ADimvT049riODvx4aoqYEQN+kgFqcRoEsCqdxuk9PRgSCXC01vjhxASeLhZRM1L0tRMTeN/gINJCz5drNslBznzPNxq4K5dDxnECBuGmyUk8MTeHsvCyNxcJ+uVYqCYjDQAEqvaWvovPGkBDazxVLOLGyUmctwvm8/BMBqXeXkyYBEhvXpHEj69tlS6Z6nUCj44O+n16ev4xvFGyFLZzZ5iIJZej8C+ZxA6gjS2VIoaBf/N9qma2bh2pqbdta43ZtqUdux9zcxRC9/zzYW0hPn7HDtIg1Gqtm30qRfHvu/IjlOpe16VEOjLMf3aWEtA0GqSit5mAuTl6AWRi2LCBrikr2larZJrg8zs7iQGx+/Xoo1Q8T5aFtSMLXk901FGkoZmdpSREzOt3dxNz86tfzU/VCswHc/mbDeD82T7PJtv5zj7fBuYoxsK+lmxbMhs2MQPS3k5rfF/RntjQPwHgfaAUrP8G4H9orRtKKQfAbwD8TgB6sdnEZgPECkDR87CJs5rtgk4SgZZT9TpeLJeDlK+BjVsJ1bL5DvndkK817pqZwUAigR4DViXfx6ZKJcymZuzBgfe3Iam+1lrj3pkZPFYooMKhdEYqrnsenikWsd110RGL4TShLj6krQ2n9vTgx1NTJM1pjbLv4ye5HN7Z3Y0HZmcx3WjgyWIRMcdBynFw/sAA3Ign6tC2NmyqVDBimAMFApa3dHTg0UIBc4ZxyRtTwrn9/ZhpNvHT6WlsqVTgKIXlqRRO6urCwW1tOHgXamUlgBJmfFsYHsuvYWkyGWnaeLZcRs301fN9PFUo4Eeui/P7+ynEDaHkL+dyvNHAz3I5PFUq4W2dnXixUkHNxK/LOXFUawpg13FgW6/kOcH9IQT6IOGQea/5Pp4vlRYcm1/l89hsmFQfwDu7u9EbiwfJ5ezKsABtyC++SDbRVasIaGU3BV8x73u5HEqncoMcHw8dpOSGvmZN6AEdjIGlet2VlDw1Ra+ofm3eHC2VLUSPPhqGt0lwOfbYsJx9uUzq/SeeCJkWBuGotkdH6VWpkDp6+XIa85tuovrtzMD09s5XDQPUH7vw4aOPkkZD5LR6XRGHJ0rKZEgT8+CDrXO9ELACrWr3qPVmg7o8nklK6bweWb0flWTGnkPZl6j/ZN/kMb5Pa2ah2gN7i/ZEQu8BcK7Wepv8UWvtK6XO2jvdeg2IN2kzK3pXT76he2dmMNVo4JTubmRdF3fPzAQhRkqsOoXWTRlAIJX5wm4K38fGUgk3T06iJx7H6T09CDyapZGJ2wcwEI9jeSqFLdUqpk1cuALwnEldqs09Bepcc17R93Hv7Cw5j4n2AAQ2Xq0UXjSJrcuNBp6vVlFpNsm26/v49dwcOlwXZ/T2QimFbdUqHhXlt7jF5akUBhIJLDYSMY+HpzU2FArojMXwts5O/HR6Go8VCsE9bK1UkHVdHLubdEuu42BtNouH8nk0NTmSpYwPAdvgYcZ/V+4fzBBoM04NrclEojV6Ewmc2tMTqsGtudxeq2F7tYpVqVTLDiHV6Nwue9z7WqMnHscB6XRwzPZqFRONxvx+CiYi0MaoMPdAFD1RKODemRmM1mqB/b3SbOL8wUGcdlocDzwQOqjJZtrbyf560EGk3v3xj0mdHQW0UvoxQxEpbeVyJLVFbcT29cWwtrRpD0cU6Nubu3xPpcgJMEqp9NxzVGFOBKJAKcrDbXw8Ua/TWHAtcyYrP1SkQ9emTXTueeeR1Mox+Nx3LsRnx4hH3T+rkP+rUWcnaSmeeab196i5lFK6DKuUv8t3ue7kMfZ5UpUvz5XSfFSYpn2dqPVlU3v767Damtb6c7v477lXtzuvPflKwWVQX+CYxwsFjNfr+MXMDGaaTVQ9D22xGH45O0sSnlFXwwB24PEOtAArb9qcd50zrv16bg5xZbLJaR0Zh8xAf0A6jfcuWoQrh4cxXa9Di2MlkHFfOB7bBXmd3zUz0+qUxf0S757nYUOx2AL6rL5/IJ/H6Uad+9PpaTxVLM7bzbOuiyWJBNpcFw/OzZHWwPSTj801GnisWAydwJTCWL2ODYXCbgE94Tg4r78fVc+jtkGhY0dkMtg0PByAmey7pEfm5rAimZwP+Eqh4nm4e3YWixMJHJnJ4HmTzMeeS6Uplv/5chlFw/TA/K/FLsDhdzyXA4kEPiCS8HxvdBTjRqUvc/tLSV/OJTR5/D8yN9fiy/Cbchl35HKkIRHnPVks4m2d3Rh+IB4UqZObYSpF5UJNun4ceSSphX/6U1K/87RKUFm5kjKjAfS+ejXlEbcB2XZs0pqOO+wwsr8fdBBdg10PzNC2XNPeaO22Zd/s45JJSqwSBejPPtuafIbPZ1MBQMzPgw+Gdn6mdJrueXY2tNlHAdTGjaR27uho7Z/WBO6sVpdtDw3R2LCUrjVJfStXzr+H1zslk60RAHKcmOwxW2g+ZdoPVnkvxOTIEEh7TUTNE7+inPS4vd2tR63puXnkkX0rpe+jCryvf0o5DhYnk9hRrSLhOFiSSrXYWCXdl8/jeQM+nu/jF7OzrQCqQvutBEYfALSmXOAMvGZV+kDwv681PKXws1yOpFWxE3JbmVgMg/F4YP/m6waMgbiGOZHejSOaB1LhBw5z5n8GqSDlqtaU8UyTmp/7yEDmA5htNnHH9DSetDNumLaeKZXwTKmEiUYDeeMsp821tGFYGBC1uTarlW1J9YVyOXQ20xpHtbejPRZDm+viT4aGWo59olAg84e5Z763iXodv5ydDaTdn8/M4MBUClUjdkkAZrCu+D5+NjODDYVCy1z6bAIx479hbi64D/bs5/UQhKIBgfp9rtnEi+VyYFLweczNuhhIJCjdrrmepzV21Gp0fXNf040G/nNqChnXxWGZDHZWq7htehov1WqtjIHpQ6NJ0ihXFWM1ZDxOEqwdCnb88eSANj3dCpyLFhGAX3ghqYwBer/wQuCqq8iuHQXikp56ipyGBgcJ0DdsaNUasGKKr8lq0vZ2UnVz9VvZNmfoitpsF6KozR6gDfmNbyQg4uvLY5Ui+/DHPkYVgW++eX57DCiDgxTuphR5hM/NhSaPTIbGzdZEHHccAcNjj4Xt7q6Q3uuVJiaozr3W4ZjYKmtJDJy7m0tZiTjqGPkekf+rBYRlfyQjYLcpQd02IfA7m5/2A/prQIuTySC72NJkEn8sa3tb5AvwgaIUpuDvWgdpPfsTCTR8H7PNZuitLDb1YNMX7z5CQFXiGgxuba6LvkQCa9racIFxhNpZrSLveVQhzbTPm7cWCauVonSmSik4DBpmVQaAp1vTjPoRbWmE3vI+gM3lMu7P5wPwA0JbsGRm7s/n8Y6uLmyt1TBTrwdq/YbWgZ8A37/mPopx31Gt4ubJSWyrVgOtRVcshrUREnzF87C1Uglitvl+FIgpsE0Sw8ary7eeXgXyAp1uNHDfzEzgZCglZCBkxmDu2RPXY58KiHvjcdlZq+HOXA4Ht7VhtFbDrAm/4//P6O3FW4QHfa5ex2e3bqVkPGIux42GZCAex23T03heRAXwHbGKP+U4QVUruZl1dFAIWhT19pKEJUud/v7vE9jZtGQJMQVf+1owjIEGwJbUtaYa3ckkObGZZdQiuUlwZOajr49Czria2exsmMSDj5VqccfZdc3v7u6wdo692cs2BgfDtLCyb/k8mSVssOe2urpI3c6lND/4QeAb3yB/BaV2Hbv+9rfT67868djIQkC28yQfB7QmbZFzaR8rPdn5fylty2tzW3JuokwkNmMlj+e2JINpMx3M6O2rcDWm/YAuaFU6jf8tY1gWIJamfa0DENWCVdNaoyuRwOdWrsRjhQKuGhlpDZ0ys94Wi6Hu+6ibdnyx0WtfVGAzbcYdB4dmMviokESnGw3cNDmJ35TL4cYNwKOOAOx4pVsTlGhFpgUPmM9kAMHq1EDIsLDq14A8J6IZqddJ4jfilDb3EzAkfA4obauvqLSsoxRm6nUUfB+3GH0uS/7McMjn7CfGYY7H2DeAuCaTQdzYpJkxeKlSwa3T0+ACK6zqVvLJNeMgNQPBXIox0wijFbisbNBX+QQbBkQyfBoArLmE1kH1OgUyXUzU67hrZoaA2PRBg/LlVz0PKeFKG+T9t+dSazw4N4dHi8WW++R7ao/F8IeDg1gdT6Ori6Rb6am7qwquZ51F8eFcaERrAmGmTZtaHex27AjLWcqkHVGZtR55BLj77jBlqjxGbqzsOczXSyQopWxvL4Hj00+3AoaUyJYupUIZUXXHZ2fp/1Wr6P6YWNJnymYpRv+f/olU665LQN3RQU6E0lGOKZOhcTrnnFZv52yWEuX8dyIGPgmMksljkiAs8/PL+YxaFxLIJfF65WtLcJf9sPsq32U7wHxNjd2GUtH1CfY27Qf035bMbMrNXTsOlPFubjeur3HHQZrTtmpSJScdBwmHSnhuLJXwXKkEabsOgNj3Q1UtgP54HOf19wddKJr49WdM+VQGm6zJJhYc5/vwTUx4xvSLU5PWxKpjMA0kfAkKfIxkOkC271tMCdXAvIBWfwG7HVcp/NnSpWhqjW/s2IEK3ycQGDh1BNvMCVIk03PD1BRWp9M4JJPBZL2Of96+PbinQDI37zLGnr2+JbvN/VWKcuU7sDQWzBSYMUooSsvL15O9ZaZIaWNiMTsJ/65Ne0opbKpUcMvUFNIRxt3bpqawIpnEsQZtR0zomTIMTbCHKEojWzRhkUpchxmTUzo7cUhbG+IO5Zf+/OcJpFlC/Iu/iLYvM51+OnkwSwm62QReegn46ldJhWx7FLsugVqzSY5fdvtaUwa3O+5oDT0z3Q9IqlT592efJal+/fr5dnffJ4nbccjG/YlPRJeyrFTIWe2Xv4yWEjOZ1rCk3l6q+vbP/0zt/e3fhglnZmbIA54pmaSQvHXrFh7T/06kjGmB8xrI7U4eY3+PAnH5PzD/nKi25HHy+nZfosBZ/saf7USTUefta9oP6C+Tis1mWPYUrRs+g3pHLIbLli1D3HFwbHs7io0Grp+cBGeUPLa9HRcPDiJmNnMHpNZltbdsi8PTFCgF7L+OjuLDxonqp7kcNhSLAfhq08a63l68o7sbLqjE6Oe2bsWc4yDjuvjL5csxZOzuD+bz+OHkJDyt0dShfVyqugOJFSFAB6DLvzPA6tBpz1GKHPu0RpPt+ghBr9Rs4rvj46gIkAvakkHOgpKOA8doFfiaKQZYADPGtBG0Zfohn7Tg+WZtijk+SN5i2nase+Q5YVB2AKxJp5FwHDw6N9cy/lxYxhfnMYPCc+lYbQVkzaUv7m+u2cQVIyOoifvhPUP7PrZWq9hmCuYwkEtG5ZbpaWRcF2/IZtGTjeP//J/WHWdXsdnlMsVNP/00fWc+8OKLKTOaTOspbgWLFlHu81/8gry45bQmEiHocz55/j9qE+Xf5W8bNrR6nfOG7DgkkXNls4Xu7b77gPvvb5W0uG8AmRWWLCEvfYAYnyVLwnrjst1YLCwSoxQxP6edtvCY/nej1auptOzll4dV9iTZkrNcK1I9bp8DRDNj9jufb68xSTbA2/2yJXKb2ZBMwmtB+wH9ZVDd9/H9sTFsMWpd22GLK2tprRFTKsiwdmx7O6ZMjDUUOaQ1fB/KcfDBRYvwdRMDz+u3LqTjYMM2318ql/HZrVvhgNTqXHKVj/MB3DI1hUXxOI7p6CB1PjWAQrOJf9m5E3+zYgU6YjEcbVK63jM7i7tFoZIATMx3JX5nhzXOWibj4QOnL6WwNpvFRf39uH5yEhsKBTRNH/jYqu+THdwyqCozjgqA41AZVKYLBwZQaDbxlIm7TiiF9y1ahIPb2jDTaODfRkdR5nMFMHI2tsDubdqWtm6bZL8cc5/MsHBbG4WduiWJjgRk+R2hxkL6U/Au1RTjGrQLKtNbMYglM/2xCUCCuqfUPJ8BiLm8dnwc109N4aOLF+MNlsgqC0k4Tqt6+sYbyUtbqkw9j9JaynrqfDs8bSMjwN///Xx7vVKUSWzdOgLBN7wB+L//l8LGogp4yO8StKNUtXztWGzXTIrvU9/ZqUqef9ppVO/9y1+mcqhak8T9yU+S815UuyedRBEC3E4y+dpt7K9XWruWfAmuvZa+RwGqPdeSJBhH2b5lGzYzaDMMC2kGxCO5S4YBmG9G4uOiTEv7gvYD+sug68bHsbFYJK9s+08tHMsAPFsq4eqxMVIxI3S2UgA2zM3hqUIB63p6cEZfHz48NISGWakVz8OXd+4Mcq6zQ5prdjYNBMdyWlO5bhTIxtoEAcHXdu5EyYCBrzUm6nV8ZedOnNHTg2tM0K0XsdvOY1ZE/xnQAJEIRYmUplpjY6GAn8fjuHhwEAXPIyc001ff9/HdsTHUhPd5S5lU8zogncYFfX0UJqYUko6D9y9ahMuHh7G5WsV5/f1BqFZ3PI5LFy/GV3fsQNV40XOVNgZZ7huUwls7OzHdaGBjqdRqVxdzyZ9h9Yv7ww6DDMw+QmAHQgc/2a7PgMsMEai0bofj4D6TFliq5qE1vj82hmsnJvDJZcvwZ0ND+MrOnZQPABSDH2gGtG6R/LnfUvtQ1xq62cS/jo7iMjeOlW2E2s8/T6UvtSaJ+cADSbJmQG005oOy1tGZ3XjY+DiZrU0OdTwe1uLOZoG//EuyZ3/5y61qWdm+DfJyA+XjbM96SSzJuy6laL3jjvltAcTMpFLEXHCp2GaTct//zd9QKJlNbF7YTwvTc89RJIA93kC0+twmuY6i2pBtRYE6f7YZRXmsXHd2trkoad0+z2Zq9yXtwmK2n2yqaeFExqpQlqh4YwVlP7t6dBRFz0PNp6pedSHdNYDgNwDojMXQl0igL5HAsnQaH16yBO2u2yIRsv1W2nGV4wQlQe3Xv4+N4cVKhSqKKZE/XSlsr1bx7+PjKPtUWY1LdyogqH5mlywN7tOsWgYlBpPgWNOfJgg8srEY4qYtJe6hJtxClTWmQ6kU1vX0YGu1in/Yvh0/FKmy2mMxfGDxYvzt8uU4sasrkEYBoNeITUr0Lbgv0zetFI5pb8fZfX34w0WLKBGMNZdKzKUc88AWb9+z+S/IIGd+c8T5SpyjrXPijgPHcRBkzJfoqBRqmsLmoDUOamvDZ5Ytw4eHhtCfSOC8/v6AqZo3XzyXVl+hFAo1H9/4JoUS7dgBXHEFhUdVq2RXfu454K//ml433NC6acnuLQSwfKx8j7i1FspkgO9+N6zMZl9nT9u1j2d66SViGr74xZAZ4Zhyee4JJ5C6/PLLw/zwvLHvQfLI/bQLOvhgqjtvz+GuwJB/syXyhYDVJvtaUSlho5iLqLW60NqV6/+1AHKm/RL6HtItU1PYaKRMOz5aAy1igdYaReNUBSNhsy1XSk5y3r++Ywe212qBmrgkw834FK2xNJXCwek0xb6b9mX4FFPZ8wLvdKki5/5WTMrVeWS1o8z5UTuro+ZL8vy/Etc7r78f+WYT22s1UoWDnAXf1deHn0xNoWqSwTDwTNbr+JXnoakpTvvBfB5PFYuhOtv05eKBAazNZoNL9yUS+NSyZcH3TZUKfjQ11YI4CpTohp0Wk64bahYQzmWLPwMDMo8Dx+WLe7RHktsKzDDiGNmWsuZY9hNojclnWpZOY3Eyif5EAkuTSTS0xq1TU+ExSgUmB/7OjnKO6HduirzD3/OeEESZms1QSn7wwWiAjEp7KZdJXx+poH/2s/DW5P93301S8rvfDRx9NPCVr4SOcfI4aTeN2pTl75KuuCKs/KUUvefzdK+f/nToVS+GHQB5rbe10f1HxSLvp9+eYrEw8zuAjRIAACAASURBVB6TdKSU88BzZgMwn8PH8v/2NmibUWwGwKaoNRd1vrxulLpdKXKeXCgEdG/SfkDfA7prZga/yudR98P83AEAAS32Vi1WpSuOYWlNEm/A3xsbw4vlMqq+H9hEFRCUGmWP8cFkEu8dHERvPI6K71OtcSEVcr80gNO6u7EymQzqbkvwZmnRLhwj//cBvK2rC4e2teH6iQlMNxrQSiGpKO96ZyyG8Xodt+ZyoSZBh2YHNgcAVN+8zXGC33xFfgT35/NocN/4fE1JdcrNZtC/mu+TRG+IVfTXTkwg7ThBXnZHqZZKbRX2VBfjf3BbG94uMnME//P8GQc+Zc2lgzB0jccPck4FA8BmEt/3EZNg7jhh2Jpx8FuaTOKs3l6kHAdj9TqeKZUCpzpmYHgNyPUTc5ygROuSZBKOaTtgTKIYBfE7lIJ2SMUNtHpy20oftpFLYLVjcoO5ETxPKkV52uV5cvOuVOh1ww1U0IXTrvL5AHmSv+c9VGP7zjvD/+3Nn9v0/bB8JpdXlRswf56ZoXuW98ptPfAA8OSTlCOey3/yIx6LvbYS2O8CHXwwZSK85x76HjWfUuLVOvRzAMK5lHPK69d2muPzoxgGJv4uy+rakvZCa42vyYyH71N2wUsuISfAfU37AX0PaKbRQIk9qBFumFqsFI47ZmBQQCCFMxAE5wI4LJPBMcYpaaJeJyc5GKcnFWaa47rnvbEY3t3Xh4Pa2vB4oYBtpoQnzHUAtCQvOaCtDW2uGyS5gWotDMIq+HkSpvn/2GwWp/X0YCCRgFIKc8Yt1QHwls5OJBwHJc9Dyfdxj3Gok8zLAW1t+D2REGVdTw/yk5MYr9UCaZFrwrMmg9uQ4BWMo9BuMNOQazRIFb0AcV+4TQ1KRLNEBFDz/PhiTOy5dMx8R80lE4M2MzO+1vPmMjhPUfa/wVQKf9Dfj5UGmM/s7UXN97G5Wg3GwzGM0Du6urBYZAWUdEA6jd/r6MDDpsSXvA63w/3SYh34SsNxSGJ65zspe5z02rV5Fkm29APMl5Ych3J3H3ccFRSRm7Sk2dnW7G5y447HgWOOofSwAwNUcOWRR1r7ZktuLDXZkr1sO+oe+XM+T9K5BAGOda7XgW9/m5KGfOhDrbH4+2nPKJulwjJyzqLMNVpHe7fLuZSgbp8PtDIFEsilP4i9DqTKf6E1K9eF7J/rUoTEYYf99uPzSmg/oO+GNhQKeMYk+2DpCmiVaPk/fvEKUJqKcPBaGEomAxA/KJ0OUsuy9KftjR8EDO2xGNb39uIY4wCWazQwbbyR+NhD2tpwUDod9ItLnjri+i2Awn222GIF4OBMBut6ezFgAOQY4Q1d8jz8cHIS5/T1IeO6ODyTwb0mjSpfQ4PSwY7Ualhh+nFENouf5HLhGKEVNCHOl++BfZjPYYbEOi6KlqVSeGN7O+WhBzCUSuHNVvaUE7q6MN1oYNJkruMnl/cPZc2ljK+XjBSbYWSudh7PAExFW47jIGvGj2m0XkfeMI7sIa+0xps6OnBqTw/aouo0mrEeNbXWbZOHHC/N9wNAQSMRVzj5ZFIPHnUUqcCBaIce+/sxx1BImueRpGUS77VsgLOzFBa2ejUBOrdjS2I2E8Hnp1LkaR6PE3CeeCI56w0OUmKZF14I2zj0UJKox8dbN2i5GdsbflTMvf27vdlrTZntpDr+5dC111I/fZ+8+y+6KDo+/nedVq2iNffkk/PnR64B/mzbz+Wx/Fjsas3aUvlCTEQUIxu1TqLU+UrROn3nO3cdXbE3aT+g74ZGajVM8mYpyAFwWDaLrZUK5jyvJSQpHYvhuGwWc56Hp0slKJB999SeHhy3QDou5ThwPC9M/So2fl9rUnnr0JbKmzUzEKtTKZxh1W6v+X6g/mXbacZ1cVx7O6UwtaRGmP6vzWax3Eqp1fR93DQ5iVyjgefLZRSaTfTG4zixqwtHZrN4yoAmQCCcdhy0uy5+lc9jk8liN9NoBNI5mywkAAUSOL/ErswAJffQIzIZDCYSqJu+sTd/eyyGt3Z04I6ZGSRcF2/v7oYCsCKVwmGWG/Kx7e0oeR7G63U0fR8Pz82hxJndxE4iGaP+RAKdrktZ6yyE0jAhcWIug3sw4qNyKP97rtHA3TMzOMWYALZXq0F0A2tDfKVwdDaLLqtOvKTpRgM7mCEx5wYV9pQKEtAoTbnve+JxwFeI/V4M604iML7jjvne4f39BDZbt4a/8SbGSWQ8rzXsS26IlQrw0EMUknbwwa1lVWVbPHyuS7Z0fkTSaQJ0uaEuWULXfuqp1r6uXk3n3n47qfm3bAmT5tigYDMsURoI+R9L/fK4SoXMBOedR/3esoWc7k45ZeG2bruNtAtc7TaZJHOGDehbtlCim5UrKRTud5EGBylHAGfmi2LCogDT/p2Pl3MaxcTJObTPt4E7igG0+8HHS/V7MknV1d7+9l2vqb1J+wF9N7TSFGkZ4VRURnI6IJXCWb29eLRQwH2zs62qX63RHo/jpO5upA37uDyVWhDM12azVLjEZHNb09aG31QqaBg1cFVrPFkq4dz+fuyoVrGxVAqlxCiNgSEXwFHZLH6Vz6OhNRKOg7d0dOC8BWqYS3oon8fiZDKQsD2t8VihgFkTLrWhUEBPLIaz+/pwVm8vyp6HzZVKAMJLk0k0tcY9uRxGjSsx535nOz8DlwY5a0ktiDJx4jCMBx8vi5wc19GB3ngct05N4Vf5fGBnTzsOphoNPF0ooCsex/reXpzABa0NPVYoYKvJJ6AAvLWzE/3xOJ4ulVDipDMSqGWfJAng5O8w9yV3Ia2Fc5u5h7lmE/eZGvMKwEsmn7y2d449IOm4GDAEZrzZEfHAdBrv6u8Piw6ZIm9TU1TSMuqS9ubKtHlza7x4PE6g/cILrfbOQoEqha1bRyDI+c6j2gQIHNev33Ua2u3b52eVe+EFSt+6Zs3Cw8bX8/0w1e2aNcDYWFhpTfbNntIo4oIjExN0n5yAxr7um99M0qjMN980RXLOPTdMZDM6SsD/3HNUhz6bpXrsv4s0NERMy0svRQP4rihqjqNU6LaqPYrknEdJ43bbUZqZWIzmaf361046B/YD+m7piGwWJd/HbdPTmKzX4SiFJYkEzuzrw4p0GktTKRQ9Dw/PzVF8uKKSm/fOzKDDdfFHojTmQnRaTw/mjNrUdRyc3t2Nbw4Po6F1AGqVZhMP5PPwQcVFFlr1j8zNYbLRwJsN2J3b34+K1ig1m8jGYjinv38emG+rVvGsLHsK4NFCAQPxOJalUjiuvR3thjFhiV+yu0tTKfx+Xx/uNNXhAGKEni6VMGZMA9L+DU226tVtbXC1xm8qFcQcBwel03AcByXjEQ9gnlSuRVtPFotYnkziwXw+qJTG4/+Eqcs+22zi57kcsq6Lo40o9HyphDtyOQybrGo+SHofiMdbsqxJ+7cGxc+7SmGqXseUdT98DhSljfVFP/n3FpOKJme3iXodk1wY2yZhWpBU8Tz8em4OJ3R1BXPJ12fVf18igUWyEh8odfDjc3PYAHKqO6mrK1Dj8+Xl5jY5GZbtNEM7r5tSWnnnOwnoufALHz8zQ1L+6tWU+3x0NCzFKq/p+6Saz2apzGlU3vWRkVDSl33ZtIleRx5J17IL/9nqWt8nJmTtWtIg3HorgbJ9zrJlBPgy6Q5AfXvXu6g4zL33hvdwzz2tDAG3lcsRcyPBoNEAHn4YOOMMAvRcjsD82Wfp/7ExcgTMZMKMd78tbdsWtnvggZQc57WmAw8kRu8//5PWhKQocJfvthRuM6O2NC5zuNsSetR1bbKlf17j/FsqRfUOotbsvqT9gL4H9OaODpQ8D08Uixiv1TCYSAQxz65SuGBgAE8Vi6gbFbdSCoVmE3fNzOCkrq556noA2FQuU15us4JO6e4mVSiAQrMZFH5hwKhojc3VKlaLFcPSpQYB7bOlEn4yPY2xeh2zjQZ+v68P7bEYLt0NU/FMsYgfm0ImcrWP1Gp4oljEongch2cyQdGRFoAydFBbGw7iotig4igPc0pUAXpaayxJpZB1XVyyaBHiSuG7o6NIuS4uXbwYCcfBg/k8vj86CqUUOoxj36znoaXWO4jpOCqTCRPTmD7x9dg+P9lo4P7Z2QDQHy0UsLNabWEOYNrQWmSvE7syJ4SZ97wLCT5oAwja8LQO7OfSzi4d9RbaRVo0GSAtyUNzc5iu1XBfPg8HCDQPQfvm+APTaVyyaFHQ1sZiEf/POCVqrRFzHFSbPrpe7MXMpNMSfsMeuxHKiHmfJUDF4wSAv/lN6wY4OUn50leupBC1Bx6YX4pV61B9f999pGoW3Q9oeprs5PK68lqPP95ahMPuv5zWqSkC0PPPBw4/PPSy5/4MDlKe+V/9iv7bvj0E+Z6ecIzsgixR4/TEE/MjA1yX2mKJbniYAF6qj7dvJzX9KwH08XEyRbCZ4pBDyJyxdOlv197oaFjlju/voIPIHPJy6Q1vIAl9IUCPAmKbbHOIPf5RjICcC7nt2REe8n+7HckYLFsW5vR/LWk/oO8hndLdjQPTaVw3Po5HCwW0OU4AmDBA4AGBHR2g9KaPFgroisVawA4A7pudxSOFQrChjzUaOKu3FxmTUCZQTRuA6HRdXDI4iPtN/DlXEOO1O16vY2OphNFaDVAK9+bzyLou1vX2IqYUniwW4QA4MpttYTByjQZG6/WAedAGCGV++C2VClan01iUSGC22Qzus6Y1ni2XcVSEV89Dc3PYJDQJAdAphXf39+MIET/+qeXLg8/FZhMvVatBeFV3LIaM6yJnfBHkmHBilsFEArlGI0xVK4BfK4WE46Dfklal1O+LJ5PbaqLVpMHSMpecjbLpQ/bNceAbW7ynVMu6UAzyLF3zmIeNEMNkQug2VSo4KJ3GU8Uirh0fD8w7N09NocOgQQD+ZnfLNRoYrdWw2KjXb8/lMGHqqGutUfM8/Hh8BvrmTqh8ODZKhZ66qVRrOVO5uUmvdAbiri7gfe+jbGojI6H9mc/dupWAvaeHAKVUCjdsu61nnyUbvu0HuHYtAd8tt8gxb92oo5gNW5rjY3buJGfA/v7W47u7qULamjX0qlaBb32LvNw/8pHWmuTy2vKereUT3Cf/FotReBO7ddhAofXCKuCXQxs3krqf6fnniUlYupS0Gv39FGq1p/TEEzT+MhqBc97/NtTVRVoZo1TbZZlSCaLyODv80Fa5y7Hn86Js6vYa5zYWAnW2nV988csbw71Fr8Jy+e9Bk/U6bpqcxOZqFb7WuHd2Fj/P5UjVq1TgtNTUOpBkJxsNfHtkBNeMj2PYqJC5rTnjNMWvO3M53Dszg7oB6p54nDZrpdAbjwftayCoBc6bs681Hi0W8RIX5tBUEOXW6Wlsr1SwYW4O/zo8jH8bGcHjhUJQghQg9f2z5XIQNw3Tvsd90xq3z8zgwXwely5ZgpTrUs5wrZFvNvEfExNBAhsAyDebeK5YxKwBWGZOeEyysRgSu9ilXqhUcHcuB89cY1OlgqeLRfiayoPKMfO1xnijgUsXL8bSdBrd8Tg6XTfov2fmY3EigQsHB4NrdJg++AhLkfqaytOe1tuLpLGVy7n0IXLnm3d5Ls+Lx++GmeBjZFueuC4zAMEYmXM9MW6353J4oVzGdRMTKHkemmZsZsX4e1Zft1areFxki5Ftc6EcDxq+rwP1oQz3WruWMqb5fqt6UdZRZ/DljW3TJnJYu+SS8Fh+cfsvvURgeNBBrb9zalWtCTxvvJFU0Kxel8TXlhu0/K3ZbO2X1uFv9jW1JvV8Lhd6nvf0AGeeSV7YTKkUcNllVPJUgnkySU5tUpLkMYu6f37x75s3hyCRTBKwyXFNJF65F7xkwHgcZmbITn/ttWTHfzkZ8Oz75LZ/WzrxRNLIJBKt8ybHSc6v54X3IeeR/5PnL/Ti4zitrxybhY6X/8n119k5n/F8rWi/hL6HdHsuh+fK5WADhtb48fQ0VqZSOLqjA59Ytgyf2bQJJSNVcbYxABiuVvHvY2O4aHAQymzQzxnHNsBIjFrjxslJrEqncVgmg79btQqjtRquGh3FXyxbFtg6E46DtOOgalZTixZKh5IpS9xbq1XcMj2NsvHE/9rOnRhMJPCFAw6AoxSOymbxUrWKe2ZmwlAs80SwBOn7Pm6cmgqkSD+4XGuJ1Yrn4bbpadwzMxPEfweSvlJoc128b3AQB1vaihaS7DKTYJ0DCdj06/bpaRyUTuNzK1cCAHZUq/j7l15Cw/PCJCqgzHlsqz65qwslz8PjRkOiQZnrGr6Pq0ZGUDCmEwAtcxlcVwnnM0OyTjqvD23ENW12O9+co0y7PEdJx4ELoCrvX8wlX7MjHkdeFGL2QQ5vMaWQjsUCPwJfaxydzWK9iHpoj8WogpznUf9Zc9JXA2aSLVLIwAAl/uAqZHbWrHSaNrBmk8BQa/LWvvJKAp93vcuMid+60WlNKtazzgKuuYZUlPX6fMmUz338cbIh//EfU7tsOUqlqA8iXH/e+VqT5Mv3xP2Uy4lf0kTwlrcAF1wQtjM2FpZmVYqkWrk016wB3vte4Hvfa7Wz21Y2yVywZFitUqrb9naKWz7kEGKGfvCD8PxjjiFNwSslnl9+PfwwRSBoTWr9TGbPw63kOEsJ95XQ+vU0fvfdF86rZByiJG6+F9knKSvY6ypqjclzozRREdbS4Ji2NmJCPvKRaPPQa0H7AX03VPU8FD2P7ONmBfCGzACYazQw22gEttF5+jalsKVSwee3bSNQZMBQqjUtq/V0LE4m8ZkVK6BAUj1A3vXv7O7GT3O5oOSpkivVrHKlNeJK4drx8ZYQMUeHjlYN38fPjGaA1dkxRbnFa74fVP+Ckey4LcdInkESGEMP5PO41ySZ8QECQvOfVgrn9vVFquclMWiyinpe8RYdepUrx8EHFi0KMsUBVEJVVoADSBq+fXoad5r7fGN7Oy4cGMB7TIWNvOfB830aYy0KrpjrcV9cM77yfwZvey75XVZ440RESlH4IM/J2kwG/YkE7piebtF2yB0rpRQ+tXQpPrtlC4qmHRdAynWxNpvFafU6bpmcXNAU8OElS/CVHTvwjImZ8rWGimuo9WPwr0rDKcWDzWtkBPiHfwg3R+m1DpD0+o530O9/+7dkI+fNr1Agqc9WPcdiJIGy5eOii0jl/swztIFLCU9rkhivvZZ+/8IXaMP8q78i9ezJJ5OE+dOftt6jfHwchzbaQw8lZ7cvfYnO4WOiJKooYLriCnIo05r6/7GPAUcc0XrMMccQw3D99SGDwv2x+2V/r1SIEfrHfyRQPfpoYnrksa+U4nF6cSEd+161pmx9Q0OkmdkTss9/Nejcc2lN/PrXobZGXsMG8ShFX5T/hzSH2FvlQmSDvGyLmdz3v5/m/vUinQP7AX2X5GmN23I53JXLEXAaVSyDTgxA0fPwxW3bMN1ohHXSLZupBCMPCFKASmmPj7Ezn7lKYUOhgCuGhwmQlcKB6TRO6OrC3QagJLhwuVHXdXFBfz9umZrCrEhWEkiRWuO+2VncPj0dMBcA8Kb2dpzb14erRkdJFY/ghKBPXG2MrxkcglCiVNaOHleqpRTqQuSAwu08n9LsOuaePMchsFNUZ52vVxf92lKp4EoTHaBA9mTtOHipUiEnOMMAPVoooJsjADwP/zoygs3mXhtirqDCwipxpRAzOwgDO6u6A6ZDArc5hucZYp6WJJP46+XLUfE8uK6LjOvCBXnkP2gqrgVaEPNe8amQTtxxEDN97I3H8amlS/FkoYBbjTs6z6WvNRrmeIDW0ceGhvC/t27FiIlZJ8bDh4qQaoBWFTwvkViMwGFXm5i9wcfjBKyXXho6DvGG6PvA178eJonh//javBmPjlI1uMsuI+lcpnG1JW6mep36uXgx5dX+5jcJKOR5QlkxjyqVVmCp1QjgP/tZ0mJIetvbCNRvvTWs6MYUpY6Wfbal4lcbIE44gXwh7rwzvG9WWwMvH5Bdl5gbCXqvRp8dhxi92Vmy+9tkz5OcQ16nNlMnfRYWAnnZnp2rwG7L98P5isVeX2AO7Af0BUlrjZ/ncrhLSE0+RAYxpXDewADuzOUwYZK+yNXFjk5AKHVqmDSsvqlbLZgDlnq/MzqK9lgsUEtvrVRw1chIYFtXvo8XKxVsqVSCrF8MoFprrO/txfGdnWgC+MbOnZR5zAIax3GwoVjEDRMTYe54898jc3PojsUC6TGQsMV1IH6T4+WZnYtTubJ06gBY39MzL0tbFB3Z3o53Dw7ixokJKKVwRCaDiwcHcdPkJB4rFACtcUZPD97S0QEfwBUjI2iPxdAfi+Gbw8MoCi0KP5UaZEtnEPN9P+jrVaOj2FKpBPZxLo8aVF/TGr3xOP7H8uWIW3Hot+VyeGB2lpg08burFGk3eK3QgoH2ffQnEvjEsmXYWC7jB2NjWNPWho8uXQpXKcTEHEnytca/jYwg4Tj4wOLFWJlKBVqCzngc2lwvSDusNTYYZ8x3C+RJGa1AwGQoBQfOPOlHbtR2KteTT6aYai6nKrN08UvwtRgaojKsyWQI5gyS7Hz30Y+SBD08HIKf3Gj587ZtlHb1z/88WoKSQO37wHe+Q+rsTAa4+upQdc59XkAZEtCVVxIjwe1x36P4UschrUU+T2Fs8hy+Fpsp5DU7OkjqfzllV2UbwO7V5I4DnH46MR18z//xH8DTT4f9eDmagFNOoXS+ctyFj+srokQiBEo7ekD21Xacs9cBMy1yTXNb8hz+Lte6ZA4Z2OVcnnkmcPzxpC16vdF+QF+AHpibw23T02iKjZ0rpikA6/v6cEJnJ+40KuZ5xhxFGbpifI75nwttOADgOEHyD9+sqpLv49vDw7hs+fIgAUjd2LM5u5sGgpKnjlitCkDGddGbSKDptxY04X51x+P46JIlGKvXg8IovME7ioqm3D07S5KqkOiVaKMFcsx9PVwo4LZcLtBiKGXy0GuyWZ/a0xNIuLsiVymc1NWFw9raoEGV0brjcZw/MIDTe3oAkGR65egotpTLKPo+/mV4GAB5yGugxYNcFo2Rc8TzUdM68Fzn/2Vu985YDH+6dCkGI5J2n9PXh4rnEaOhQlX3Ob29uDefx5QpaMPX7IzH8dGlSzFer+Oa8XGKEqhU8P2xMXxg8WJqQ6w3Lca9pjWqzSauGR/Hp5cvb/HaB8IUv7z/eSBtQxRxxjqAtSqtG2dU6tfjjgNOO4281D/72VDyOesskkrz+WgwHhujcqULScNnn00SJJ8TJTHJ7y++SEAt65HbGzP3o1QCvvY1+p3TLNjMi7UsWq5/8cV0/vDwfMCIooceoldUW0qRb8EvfxmGx6VSlA9+5cro9haiK64gZ0HPI+D7+Md3XwgkkSBnPyZmrvi+zjuPtCh7QnZbrzZddBGN1dVXkxMlMF8LY4O0PZfA/OyFUutjM3NAK4BHMRL8nsns3ft/JbQf0Beghu+jYlZPINGI927LW5vDyPhYKRU7QLCKgiIj5ru95SqQGl9rjbFaDZePjITgaoBdVvqad75YiQwILIVz+4sSCfTH4zi9pwd35HJBfzR1mMqZyr6JJ4cznmlrxTd9H2U/rG6m+VgQ2K9Ip1uKteyKko6DnbUabpmaCn47vrMTvy/qLhabTRRMpEBJxKiLgYACMBiP46j2dmK8xFPccm/WuTIkUCm1YFGU9lgMKVs3B6AnHg9z1JvfM66LDxrPme+MjqLUbJIk7/vYUChQ7XojegWmGtkn8z7VaOAft20LNEWsOQlMBOLefz03h954HKea3ecbO3digsVUc522tMYHPwZ0OBRW9sMfht3mWzjiCHLMuuMOsnlz9TXPo6Qg7HBmb7IAHSNTsMphB4CbbyYV+gc/SCpxmXAmCmjZEe/888l2zznibd6F+8KhUEx2ek/Z/pFHkmMYkx02l0iQ01pvLzmSfetbtLH/1V/R+bUaqemjAF9r4K67SLpbtiw8xk7w8uSTJD3zeB11FFWbY7r6apqDej301L78cuBTn3p5YWN/8AetaWqXL39tM5xJ6umh1wc/SGNsx6gzRckHUYC9AF8LIFp6599sZ1CAvPGPO+7l3c++pNfJFL4+yTEzyZKPBkl+x3d2BvHcfzgwgO+MjmKWpT6p+jSvoGqXChN/0AGiFKu5VsxxcE5fH1KOg38ZGcG0yU7ngCQrpRSWJ5NYkkziEVNdi1fgWzo7cbTRfX17ZAQFYzt3jcTXFovh4oGBgBFZlEi09E+2FVSTA+C4LtZ1deG+fB4FqQfTVFGMz3ZkW6LNsu+jbBsWd0FPFYv40dQU8o1GoDkoWuc7RgPAyVkCJknMVV88jvcvXoyBeBwHptN4rFBAodnEyd3dgfYjKCNrNCWuUjh7cBCDiQSgFBJKBel7bbprZgZPl0oBw8bagVumppA3UQWsLaj5Pn44OYmm76PQbMJ1nEA70tAas8ZjybHGnyMEgogBAIVGIyypas6RueZ5DA5Op3GUWQ9colcrBdfca9J1ceHAAI7timHnDvIwZv5EbpYdHQRihUIoFbKkxKkGuGSplHIA+izbsjfJcpmc4latImm1UqH/8nliLtjJzHHC8paOQ+rOd72Ljn/uubBNKaXbJgF+lypWufQ7O1tD0gDyYC+VwnPWriVG4qqriPnI50li/shH6HgeG5thYC3BHXeQd/T/Z+/NwySpyrzR34nIPbOy9rX3bqDZpVmURUcQkEVEBBlRQFwQZ7zjyKjfN99sjzo6ftd753Nm3GbcF0BlBwEBG2h22WnoBqFpeqs9s6py3yPi3D/e80aciMrqhRGkvfU+Tz1VlZkRceKcyPN71997+eXtiV2aTYohM6A/8QS5s9/zHno/uAb82r6WjQ0O0s+bWZYupdyHSoXW+dprPQZA3dvDz1jQWud5B/zPhf46ML8DH//PFv7hhwMnnUTvLV++b+GRN1oWAX0BYTDWm6EAtEFvrVYxWq+jM5XC4akULh4acl7TxgAAIABJREFUQtm2XUKW2VYLv52bg60+LwC3OUo7p7OA3xW/NpGAADDTbELPJDfUeXKWhbrjYE08juO1uPSBiYRbr55pNuGoxDKAlIGIYeBQtcFvqVaxfm6OQKWNacVjYlkTi+GxYtHvbdDGdkgyiePTaTxVLntWPI9b9yrshZRtG3mtH7oAsKlcxvJYDCd1duL6TMal4XWZ2ALjgpSImaabi7AuHMZgJIK642C1alcKAGcx7W69DkMIvLu3F+/q7kZsL7Jdcq0WKgq4XVe/lDgsmcSmSgU5KX0JhNOqOQ14rVlhFAKr43GEhXBBV4AUATdvQwWmBQAYBsXb1dwYgBtW0NnoeiMR1zWfbTbRktJtx2sIAVMIHJJIwBACzSbVYvOGxhvkihVUJ8zTq2+aoZAXL9ezi9NpsmI4lhx0fQYBb/16SjJjl2+5TCDJ8erghjs6Sq7/c87xM7bxuVnn1EGPN/6gcqCmc0Er7qCD/P8XClRqxsQ5jkMKxde/7u841w7QAcq0z+eBn/8cuOIKal3LwsQ7+j3XavP54fncpunN+5+qrF1Lvx2HPCA33UT3f9JJ9Hzx/QfXUv9ff0Z5bYKlcEGGOJ7TNWvoOVu58o2869cui4C+gByUSODIVMrtluZa1kJgttXCrTMziJsmVsZieKFaRdmy8OHBQXSHw9heq+GeXI4AlWObygqE+ps36sMTCayMxyGlxH25HIq2jVuzWby3rw96/Nc9B6jem2uOJTCv8Qi0sboxdsC3axUtC1nVFMTX2ENKvxdBXcM0DJze04PbZmepBh5AMhTCaV1dMAQR4ZzR24ualFQapbn5fX/vhXCmO19bgAhr7p2bw7OlEkYbDdRVSZxrHWtzBCmRUmPTZSQQB797bg6vVKuo2DYMw8AJ6TRO30swB7z1YIBmpeLAeByvVKtUIqgBrJ4o6R4Laqt7dm8vNpbL2FKreQ1qAByXTuP3lQryUs5bS1aSDCHQFw7j7Z2d2FqroWTbWNfRgTUaTbBLNavW19DG7n3GD+ZSEmCuWuW9rwOV/rcOupEI1Wc/8MB8gAvqdUJ4VKoAbdpXX+2vDdetaSHIqp+YoHj1li3zz6mlCPis8eD1g3/vjc5pWV5jGD5ns0kJe/o59aUKgo2UNH6NawoAKTLT0/MVGH1c73wnATy3iQ2F6LU3a0z3DyWGQYolA/Hy5fR8Ae2fyyCoB58D/bP6MYZBoZYTT/Sezf0FzIFFQF9QhqNRnNnbi7rjYGutBlN4iVIAtVW9JZNBKhTCi5UKGlLi+mwWFw8OznO18wbt/g3PVXxAIoHT1LfxsWIRecvClloNv56ZQdO23VabAPwdyQDkWy3sUh26dLljdtZ1t9NFyXV8UmcnIoEnXwTBlq1DqSXFSQnTMHBKdzclZ6kYe4dp4p2aj7Js2951hRde2E0Iq60si8VwSDKJ31erXlUBgEyr5VYU8JiPSafRHQqh6Th4rFhE03GQCIVwak9PW0VHl+21Gl5igh/N87HnXHwlPA7p1esflkxiRTyOk7q6cOfcnGvBG1LC0dZSSIneSATrUiksU21d9Ra0PHd5y4KlPTtuf3NtHdOmiXer+11TraJo274e9gBwfDqNuVYLM42GL58CUro13UHLWQgCr8ceI8KVo4/2gOPpp4kLPWjdRqNkPYXDuwdJHfAOOYRc+jfcQO7sF17wAFk/vw5w4+N0fbZedWuLrfO3vIUoZPkrom/eQtA1163zxhS0xttJMklz8fDD/pKodgpL0AJka7Jd9nYmQzH2PSkVb3kLAf/UlHevp51Gbvw/hjz2GFHJSknrvjdz+FolFPJyHAoF4NhjicIWmA/M7L1oB/T6/7q13tlJFRxDQ1QRsD/KIqDvRlbF43h3by/6SyXsqNcxpZqp8Ga6tVbzkrEE8aV3GAaO0TbT4PdTt36FlNhcqWBWJUOVLQsmKIntJW6arB2nu+wX+t4/kM/joXzejVkLUFz4uHQaZ/f2upnmQ5EIDozHqVd5G5DwWewKnMOG4SamFS0Ld87O4s65OZzd04OxRgN3zs5iVxAwAKyMxbCsTZb4QrIkGsVZvb3o0rJ0tqv5DwKeBHBaTw8ShoGQYaBm2+gIhXBGdzd25+Z/qljEdKvl3qd0HLxcrSLXamFggSS4ecLHavN3ZCqF4WgU/eEwtdW1bT9ZDs+LEOgMhXB+sKAZpEyxQrS1VoOtEfTwjsQenoRp4h1dXa7yckCbnf3RQgFjirJY7w5nSYl7cjkc1xjAiy8a88AJINDcsIGA7Oij6QegLGstZxEAgfjb3kYlUty8o810+SwoKYFnniEL98knvWz04OcCt+9Lngteg4HTsuj3unV03KZN/kz8ri4iM9kX4a5a5TKNO/iI6fcVHPdCXgqAAGrLlvnvc99wXU46ad/G/HrJpk3Eg88tcYtFSnBctsz7TKNBYQT2RqTTlA+wN96QDRvI5d4u2a+zk/jjFTu2b87HxvyVCfq1Dj2UnuXnnqMxhcOkJIXDpKwyy+H+KouAvgc5LJnEYckkfjE9jalGA1ynzOI2NQHQsm08WSrh8GTS11wFQqArFEIyFMIEmwvqHK9UqxQ3BZGp6Nas3jtbtwQZECAlJppNbKlWcVAigadLJddtD3iAGhICZ/f2+lzJS1U/93sNA7OtFib4GyclBqNR1G2bOpwBWB2LoVsD15pt487ZWdyfyyFpmrAcB5lWC1sU96XksSs5PJlsCzS7kzXxONZose6rp6YwrRQqBkUHlEB3anc3ehIJH1/7nqTiOGg5jleHz6Cs3m84Du7N5dyWuEJSS9IT9Ux9zTJnsH6xUsFBiQT6wmEcnEzisUIBLY0OmIG6wzCwNjAnK2IxvKgUPLeSQPPIAOT1WZtIIKxCOT3hMN6jUbzuqNVcNrgDEwnUHAf35HLu3DGYA1TJ8WixiJptQoo+AMKXWMaP+fg4xYl1FrGgpeM4tCm+5z3kcu/u9hjPhCBLOrjJ8jleesnL2G5nRUWjtLGXy55rWx+Hfgz/b9vUXSwapQx9x/Fb/gvFzPdG0mlSBByHgHjHDr+lroMIv75sGYHQli0EUOm036rWx6MfHwr98azv3cmrr1KCH68pQPe2a5cf0Gs18mawfZJK0c/JJy987mefpfM+/jhdZ2iIShuDRTIDA0ROFJQnn6Rz6LkLAP19zjmkJP3yl5RMmEwCF11EisifgiwC+h5krF7H1loNU42Gr1RLggCFrVep4uUrYzEij5GKllUdMxiJYHU8jgnNUmJXrxQCjgJPAb+SIAUR0ATbd7KisL1ex3OlEjXxmJ11y5Ic7fwOfz4g3PL0qWIRD+Xz7jVP7OzEdLPpeiDO6+9348+W4+DeuTk8mM/DkhJFy8Idc3MwtXAAgy0D3V4o4z6ZaTaxSe0Aw5EIDk4mMRKNosM0UVS15qwsSTG/dG9v5J1dXXixXMacyqTnuYagRLv7cjncns1Svb9aj75wGEnTxFuYRUN48XEb5M2YbrVQsCwMRiL4QH8/niuVXDY77q8eVb3Z9TI8ADihsxNl28Zmde9SSow2Gihr14qZJi4ZGppXhw4AU40G7pqbc+duba2Gqm3Ts6vuUX8uJIja+KlWAYPLujGxzdsOdICUkkqH7r/fuxZv0DpAWhZlZZ92GoHWRRfRxioEbfZMHMMKQPBa7cDWNEmR+PjHaaP++c/9G3XwXHw8X8e2qbTtiCP8rwevs6/S10eZ7c89R2VjPBYG4+B1TjiBYsBXXQWccoof9Fh0Zj4+3+gohUMSiTdXLDebJWUmCJq6New4tGY6X3+5TJb37gD9scc8b0omQ+c86KC972Z23HH0sztppwj8KcgioC8guVYLr1Sr2Fip4NlSyU0c442RxXWnCoEVsRg+NDCAJ0olAgfVIIRBpyccJlBSddOAx3nulilBARXHTYVwP2NrpUmm8MrKspaFX8/MYFzfuNkaEwJDkYhLAdpOjk2ncexesLhJKfFIoYDfzM6iocV1hVYfrzPP6cDbThxJjGbM4sb3+2q1igcKBQCUXZ8wTbyruxtl28YDuRwqKgjpCIGBcJhqwV+DDESjGFHhjqlGA00p8ftKBeP1Om6fmUHD8RrswHEw22zi5kwGScPAAYmEu066x+Dkzk43s14AGIpGUahUPNY8QZnlFwe8CZONBsYaDRyWTLp14wDwle3bUVItoaSgmviFKHQ3Vyp4rlRyx/x7rVsIK1jc4Y0VBAiBWFziwxcB117lZRNz2T67M196iX7c8zn+ZiMAJavdcQe5pWMxsqb1uDAvUxDMdVDQwV0IsvbXrfMy1vVj9eQ3/dhgx62776Z67yVLaIy5HFl9r6Vsy7bJ1c7XHRvzxq+X5DG4S0neis5Oeu0jH2l/3ljMPz597rduJUXpzQToxx9P6/vAA/6109ej0aBWq1wzzwrWnkrseN34b8OgZ2/FCvK4LMrCsgjobaRi27h7bg4bFAscu1MBeDFzwG1X2RUKIWGauHxkBE3Hwa3ZLCyANk3hMYid1NmJiUYD93J9kDqHfi4GQD0TXgCud4Chi+loGQC5PacL6OqYoUgEnxgeRjdTQ/035+WGTIY6t2kgymVTUMQy3EkMQiBlGEguwFjxTLGIH09OoqbK61gx4uxwAWBrvY57czl8bHgY5/b1oWTbeFFrMHLp0BCWa9ncuxPuES4ADEajuKC/36Uc++K2bRhrNHDLzAw46c4BvEx1QXXv480mfj07i88lEugIhdAXiaDsOKjZNpKm6XbFA4hqlZuiFCwLc5aFrlAIV4yM+EISuVYLt8/M4OlyGetSKby9sxNCCCyLRtETDhPRkCAO+z2tpa28ANB++55ZgF5j979DTXjqfXX80z+R5+GZZ4jcRDHtuqKDlJREq2qa5E5nmoBi0bOihfBnpwd7WgPeufQEMz3OXSxSZ7ZolLwEuuUbtOiDXgV+v7ubwPsf/oFcuLfdBnz2s3sXx9Xl5ZfJMr36an+GOpfA6a7yRIJcy1ICxxxD4YgXXqD3DjhgPjAtX04MfE88Afzwh+2t/DebdHTQfVSr3t96eCAep055f//3Ht1vcP3biQ7oUhLN8COP0DwuWfL63c+fgiwCeht5vlzGvbmcV84VeAql9gMh8MmRERyi2AbG6nU4Qrhc4bxn6Faq1M4n9XNKL27KbnK+jikEOkwThiB2sZqWKntcOo2WlHisUEBTHeNIykz/xPAwhvdBrZ1rtVC0LPRHIkgGyrfGVNhBAr7yNkdodeaa5Q4ARyaTOLoN0XPFtvEz1d9bCOFmcuv1+oY6n/79Z8t2qtFATSWLWY6zR1rZim3j9pkZPFwsQkgi4bmwvx+pUAhTjQZaUjXOUcIKk865byhQ7VDzckZPD87o6cGvpqfxdKmEc/r65nk6OkIh/NPKlXiyWMT3JydhOQ7GGw2s1IJ2D+fzeErRqT1TLuPpUglSSpzV24tPDA8jZpoYq9dhSYmabWNnrYYVbYJ+EnAteTeZMWCRO+r1NCsKUqJkWfjxxCT+MrwGHVEDRx9Nm/Stt5J1xWkfOuhKSdSoRxwBfO1rFDtlAOL6cf2rYxhkbScSdD69OEMH9KB1KyVZrN/4Br0ebMcaBFLH8ZQLKeman/60RxizZg1w5ZW7fVTmyfg4KTff+Q65mVlYYQkCFLuIzzqL/v/Vr4Df/MYDqc9/npKz2kk4TEDIhD18b6US/QSbFY6NedzuQpAV+0bJ2WdT6OXJJ4EzziAu+3aSTnvEQ4axZ973jg76PFv2hgF87GOLYL43sgjouxOOkfL/nLSmdg8GtJwqpZpptTDbbFIpmO6S1kwBqe1UuvXunlfb1fS64/5IBP+yejWEELg/n8dtMzPUAERKxAwDf97Xh4ptY2OpBEtdN6GY5fZWKraN6zIZbCyXcWpXF97X3++yyjUcB1dPT6Osmr3omdiuha7NjbpZPFosIm6aOLm72xf4rCi/qE7cwz+G9nc7d32+1cIPJiawo16HEAKfGBrC8cqqzSgLXAiBPhVnbjoO7pqdxSMqTwBK+UkYBs7r78dPpqYw1Wx6ddpSuiRBbsmgOm55NIpPcGNuJR8cGMCF/f0wF1AqXq5W8dPpabRsGwXbxvcmJvA1tZYNxyHqWp5DzfNy19wcEqaJI1Mp/NvoKGYVb0AyFMKVS5fOSzSMGAYipkmJfBry6uslJPVPP6+nBzfPzaGs7jeXB/7vHwJLBqj5yUEHAZ/7HLnMH3qo7dJCSrLKPv954B//kaxpYH6tL/9/yCHUiOVnP/NznvO5pSQwE8LrLc5KAZ9H1xsNgwBQSn9XtKBV+9+xbqUkSlq9TEy/L/2aut7/3HP0A8yviZ+bm38ulnXr6N5/8QuyTPkzTz5JcXu9N3omA/z7v1O1gRCkLF155Xw62ddTPvAB4IILFia36eoihW9f5CMfIUVhyxbPq/OnTJ7zh5RFQG8jTGxiS68XdsgwYBoGWXJsNToOHAA/n5oCAFybybj84vztdi15Ja7rUwMyAIiAGoWwe50/p5cZ1RwHqVAIJ3d14dhUCrfMzKBu23hnZydsKfHx4WF8y7bxaq0GxzTx8eFhX6b47qTlOLg5m8XGUgmOELgvn0cqFMIZPT0whEDUMPDZJUvwzzt2oKisajcUISlpj2PnUPfASXYb8nlsyOchtbmTgGuVB+8T8MIOEdOcFzP+ydQUdqp2qLaU+Nn0NDpCISQMA98YHUVNSvSHQvjb5csRMgw8UijgHuasZyVCCGzI59EbDiPMFq0aF4c3wlp3MkiiuY0oL4kuQhDrGgBUbdu17lOmiYlGA/81Po4KK0Ig70nFtpEKhfBooYCHFRKyt8P1fAC4OZvFDZmMO1cOgJJl4bvj4/jK6tU+L8o7OjtRsCzcpToE8r24IRuFwpbj4BfZrFffrt3PxAS5ftt1mdJd4uHwfIpM/gny8jD4xmIeCHM9tg52sRhZ0/39xLxWLs/viw14PdWFoFhuubxnTnfAow3dl85glUr7mK9uPYdC/pap/F7QZc7zee211NJ1zZr21zzxRPIE3Hyzf/zBe/vWt7zSQceh+/vP/wS+8pU3jp5UV7gWkn0FY8MgVz3rpULQs7Eoe5ZFQG8jx3R0YKrRwB2q2FUAODaVwvsHBnDV9DReUN0obPUkN6TET6em3CQq3pBF4NtoOY7Lfc4MZ4ayBi/o78f9+TwmVGKba1WpvzOtFr45Noa/X7kShhB4pFjEY4UChBB4YccOvKu7G+/t68MVIyOwJfXC7txNt4WWIofhcf5mdhaPFgquB8EWArdns0gZBt6h/JUx06RkPHUO/T7ZoubXDc3TwO1FHcCL8/MuLzVyFniUpALAwckkPjo0hGhgR7C1cwspUbdtfHNsDBKqP7qkbPO/277dpVd1cw60a/FafWpkBP/Pzp0YU2WJUkrETRNfWLYMvaqVLN/z7nq6W46D/xwfx9ZqFcIw8NHBQdwwM4Ocyszn62daLXxnfBx/u2IF3t7ZiZlWC/fOzblKDD8vAtQ1zcdir54pK7i7q7k7s6cHJcvCAyqzys1N0J4lV5nSPVASkIYDQxg+oNUtT/33+9/vJ2Vxx6BRbfKymSbVUn/oQwR8730vudE3bfJb3xddRC58IYAvfYncyd/+NsWr+dorVhBlKqcRxGJk8euufY7B8hhaLaJV/T//h5LT/vEfvTIlKT1LWEoCZ/1x+9a3KG4eFFZcBgfJ1Xz99f6kOJ0KV1s2GAZwySW7744mhHd/7RxfLGy96gpHsLXq/ip7SwWxKH5ZBPQ2srFcxr2qQwJviBHVxvPSwUF837axtV6n7G4QgLU0YOLvnhBe04xXGw1cPT2ND/T3I9dq4alSCX/W1YWTlau4LxzGYakU/nXXLsypRh2AB5ICcMufoK7H1pzlOLg3n0dHKIR3KfD9lx07kG218NdLl/q4ywHghUoFP5iYcIH48GQSnaEQAQffs5RoAbg+m0VHKISjOjrQEQrhUyMj+ObYmBu/dncaBnXhJWMBHiMeg78uBpQXRNv5BiIR/PWSJYAQiBuGy00fFGEYbnY9pERLuY51QGwqSldDG5tupf9ZVxfe0dmJmLK63UREdY6uUAjd+7Cz/GhqCltrNTSlhGPb+Nn0NKqK41/veDcQDuNjqvNa2DDwnt5eVB0Hv8vn59XwS8B3nwbgK4cMSsQw6H4Mwz8f8NYBgMsQyFhgpG3Ii8Ygr1oBAaEvq+9vlkTCD3xBVzRAS7p0KQFwPE7H3HorufAZpPlzAPF0L1lCLuOuLgLfyy6jRigM0OPjwC23kAWnXy94LgbzCy8EvvtdsrQrFXJn/9u/AX/3d/SZ++/3+NMBolF973u9c9fr893qAMV5//qvKT7c3e3FxKUE7rkH+N3v2rvVhSAPwZ6sWv2+dgfqfA39c4vy/19ZBPQ20pISVQVYemIaANw2O4udjYZrybJ7Vs8kZqBzXbggi7is3KznDwzg5O5uDEYiSGtW9EA4PC+Jzj0/vIz4h/N5bMjl3LgoQPXE3P/8P8bGsK1Wgw3g30dHkQyF8PmlS9EXiWBbrYYfap3YhBB4tlRyXcaAt8lLkAv5x5OTMFRYASDXv+R7ZNDQLD51A2oyVGKWEPPm0pGSqgOGh9EVDkMCiBkGRvbgX5OBH19rWd1tL7wuar4EO/X6E8UiNpbLkABO7erCpfG4T/Ho3YfKgKumpvBcuUx5Deo1zhPwKTNCYMay8PXRUSKJicdx+cgI3tfbi5NUQt2tMzN4WZEN6SEA17Ohfi+ECad2dWGm2cQzpRLelk7jHZ2d+EUmg121mvsZfrY5ec6BBKKq0YxmjQeXk1+74QYCXLao22WcGwbFnm+7DfjUp4iIZMMGLz6unw+gGPy3v03HxePAF75A9cTlMhGBsDW9cSNlml9yif8cmuMGF15ItfDXX+/xnrPl/OqrFHs+8URq/8pNVaQkIhRdLr+ceqKza1tKcmdffjllq7PobGbvex/d48aN/vk491wC/pUrF1g4TU44gebu4Yfp/2OOofp1XT75SeCb3/TK3LgN7Zu5G1g7Wb+ecjVYpCSlan+lX/1jyiKgtxEDlFXO1iW36QQofmlLiqc62kbLvwUAKM5udkMbAJbHYnifIhLpDYd9YPHjyUlsr9XgACir0i/f+QDXfQqQRf1KtYonuNmz5ubnMQrQ4vJ4GaQtKd3ENiGolaYlJVbFYogbBp7nkjDhZUpXuBGM5rZ1jTNl0RsqfADtOB3E+PrsRuYdNGQYWJNI7FNZ3YX9/chbFq7NZJBttdyOZjrxj97KVgjhdhkz4Vm4dcchXnoAG3I5rI7H3RryfZWSip277U/5vhmAeUwApOOg2GrBEAKbbBvXZjL44MCA643484EB5Fot3Dw7i7F63c1VYMVEgpSq74yN4cply3wcAw/l87gvl3O7/71UqWBnvY65VgumYVC5miaOYXjd/AxAmN5zpBOc6CAPEPA1GpSJ/J3vkCWrW8psNQ4MAO9+N5Ud3XMPHRPMVJfSawnKvdNLJeC//osSru67z980hnus//a3wKOPescYBoHe2rX0I6X/PX1c5TKNuVbzv/fkk1Sjzh3mli2jDOsf/5iO+eAH6Z4OO2zhZ6Gri0IS9TrVT/O8DQ3tfcJaKkVZ5G95Cx07PDyfWGXVKlKUuGLAMKjV5xsld99NngiWs8/et17hTz1F3pFq1V8iKSUpgclk+7DOoiwsi4C+gLibGv+vdiodzCS8DVYHVd6A+bNCCKRCISxrY3nekMlgU7nsZjr74p7QXNiG4XYg6wqHMaB6mfNYj0ylcHRHB345PY2ZVsuNBZvCny0vQJ3THCnd+D0ApEMhvLunB1XHwau1GgGg8GLHLg+4urdkKITzensx02zi3kLB7QoHwNdQhsXtigYQqxwo6ey9vb0+L8XeyMoAJWzBslwwF2rchpToiURwRDKJhwsFfxY9f1YD27xqzAIQp/73JyZQ0UoPB8JhfHJkBJsqFdw5OwsIgSOTSZylaFc5kZItaraADW3O9LXk56Ipqd2uLstiMSyLxXB3LkcudrV+nFvA6zJjWZ5HREnJsjCt2AIFSNHQqYC5yZAeEnGVH/XQGcIDvlWriDv88ceJ4ESPrd96K5VjTU7Sa3rNOVvo8Tid46WX/DztLGzdB8u/HIeIS66+mnjbGdD5c4ZBiWOZDH2eiWdeeAE46ii6LoO1/ngFFQ4Ge752pUKKArvMzzuPwPviiwk4jz127+K7Q0ME7Hp/9DvuoIYwumW/O+ntpZ/dSZDn/Y2UfN6bf4BAOJHYe6WCu8vpa8FSLHpshK+X5PNU8x+JkGL0p0BaswjobWR1PI5jOjqwUaXFrohGcZwqAP0zlcQ01Wj4YqOOZqHq1jSDyESjgbtmZ3Fm4Bs62Wyi4ThuVr0pBI7v7MRTxSJqyiWqKxMsR6RSbkkZQO1eByIRrE0k8EKlQlnV6r2m4+AnU1MIC4G6FmtmoF8SjeLEzk4sj8VwTl8fbspmMaosQ1/2ufpbCOre9rbOTmyv1bBBJefx/arBeiELIbAumcTKeBz3zc0RgKm5e7lWw0kqFLGv8vtKxU1EZNc6/90RDuO9vb0IC4FHtPFBm1OdD0BXehwpsavR8JXWcVJiybIw2WoBkuq3U6aJppTYqZhG2CPDzwW7/dlzESzTY08Qy325nMvFnlE936Xy2ui2tZDSA2NdGKh175E2FkfPfVDjcXnjIQg0NdDL58kK7+0lQNfZ3rJZLzGrHVAKQa7qb36TzqMPlYFct5515wG/PzXVXgnYuZMUCV2BAOh6N9wAXHopZcvrfc/18c3N0f0ceaRH+MKSy3ld3G68kVz73JRmX0SfBykJ/K6/nsqy9vea6iefpE522tcKc3Ne6eLeSDsg19fy9ZJmkxIpczlg2zZ6/q65hvI19rJz8ptWFgG9jfRHIjizpwf9ygW6Kh7HgcoVe3gqhYrj4LaZGao5h2fxgBxCAAAgAElEQVSp+3yTgM8qLNk2HikWkTJNvD3Q1pOBky26E9NpJE0TG3I5WIHPsSyPxdoypB3d0YGKbePX2Szytu2C6qvVqg9MOPtbgChpD1WBt7WJBM7o6cHtMzNeQw9oAMFAAAKV4WgUp3d3I6MavBzd0YHttRpeVdlEAgCkRN6ycEA8jlB3N+6Ym0NZlfe9VKng+mwWHx4cnJfNvju5Y3YWT5fLaGomI9evA0BcCJyQTuOpUskF6yD4sSLm3o963fWOMBADaNg2XlLtXLkkL99q4f58HrYCdxE41948F0HZVa9T/FxbHwNwKyp4XEJQKORHk5MYikRwnmK8488H75NuTM5LZHRBHQASFuTbs8AjA+5H8nnKNj/qKK+LFeDvZsaiZ7jz7dZqxLCmX9YdlwZ2QZd+u81ed40XCvMVCT52bIzeZ6r8doBeq9GGfsEFFAbYurXNYoDuVyeT2VcJjn90lEBvfwf0iQnynPB8tuPhX0hGR8ldrxFmBr8ibRMK/1DiONQNMJ/3/n/uOUrK/MAHXr/rvhGyCOgLCLs9AWBTuYxfTU/73j80mcRGFQvVCWLcjZo3SvXECymRa7WwtVZzAf13hQI1UxEaPacQCBsGTunqchugMDAULQu/VONYGYvhBC2o9lihgO0a/dYRqRRsKfFsuUyuZHUezxpT0ubpPbqjA48VCsgoS5ST//iTIcPAulQKYcNAwjBw/sAAdtbr2Far4ZTubtyazRKgQykq6hhTCDSkVscPytB/plRC2jRxXn+/Lzlvd/JipYKyTpGlRAKImiaO6ehwr62/5zu7BpprEwm34YnuleBkOrfuXrO0BYBpxZ6nE+vo609/EpDG1LyVbRsvVqvoVorUSk0xY0UgyLwXEgJHpFJIGgbqUuK5Ugl128azpRI6TRNpVeGwSj23o5rHgJ8vvrfDUymkTRNNx8HGchktFX6REQdiZQV4xD+tu3aR21ovidLjnfprOujy7+AmPTIyn9GMS+WeeYZc23wMu+Pd+RH+30FhYHjoIY8dTlcS9M/NzRG4HHAAJcrpioUuDz9MY9Z76eTzHvubaZKLPZi01s7ylJISA59+ml6PxSgBbH9z9x5wAFnoY2PeGq9ZQ1UNu5NslkI1unXfbk3XrHl9lR59rYWgZ27zZkqm3J9lEdD3IK/Walg/N4dttZoXGwVw0cAAtleryCvQY8tVd73r9Kj8tE41m/i1Spl9vlzGnGLO4Iz1A+JxRIXAA7kctVNl9zCAgmXhIRUPLliWC+ibymXck8thUsVOISXWdXTgPT09eLFaRV2jiY2aJpZGItihFfdmtTEJAIckEvMyrA2QW98UAtvqdRgA7piZce/PULvur2dmXMVCj8GvSyaxIhbDTyYmUGPPgfpMw3HwUKGADtPE6T09893IbYSvyy5lCC/vgBPfHsznyZMRjWJUEdHooK5XKJzQ2YkhtauGhcChySSeVl3sIKiEbiQaxU6FNr4GJwHTQk+SZGEl6OKhIWSaTURnZzESjeKcQMc1vh89sx0gID6zuxtLYjHcms36FIaSbeOpUgnv6u7Gwckkqo6Dp1XCpACwrVZDQcXRD4jH8YH+fgxHo8i1WthcraJl2/Tc6uPQQHBiwrPM9ff4fX6Np0EH4XafXb2aas6D0mySNd9ozHfH69duZ8XpCoaUVOMeDtO5QiFg1Spyr+pkL7btZbXzccExS0ltWJNJapna0eE1oXn4Ye9e+/ooJqwfy722t2/3rHwpqQMdz0cySUmD+xugH3YYxbh/8xsC6SVLiOp2d9SzxSLN26ZN/twCwO+hGRmhBLuVK1+fsetrzesXjVIFwuvpGXgjZBHQdyMc995Wr3sbLWiT3VKtouxo/cvZAtJA3U3U0gB/tF7HqGZJS+1neSyGC/r6sLlaxYZ8npjjhEf+IdRuKQ3DBxbPlMsE5tLLZn+6VMJxqRTWxGLYqDLdIQQSponTu7vx/akpd3ebajZx1+ysa5nuqteJDEXdkxQCK2MxcosLgS/u2IH7VOMaCbh13vzbtTAl9RDvVk1MAGBNIoFsq+W6yvkaFcvCb+fm0BkK4W170SdxVSyG8VoNNQZQbW0qto3fzs6iKxzGBwcHcU5fH27MZFB1HOK1V9ecbDaRMAykQyF0aTH8uGnig4ODqEvpejcGolGc0NGBazMZTCoKVh8/vw7kaq6Xx2KIqrVKqXNGDANLYzFcsYD5oVv1fD4HgLBtPFksYlOlggcKBbTgeYL0XACAPCxHa6Tft83M4NVaDRLA+/r6MByNwpYSjxeLRDCjjuUs+oUAWf+ff2u6IqScz44WBMegG12XJ57wwDxoQenH83t6MlvwOrZNcV4pKUHu/PPJpbp1qzfOvj4C09tvb+8q5s85DlHVvvOdtPGvX09grjuIZmeBO+/0u47XrqVs92ef9ZSJ7dv9zUdaLaqFZ8rbri5KvNsf5K1vJVB//nkqMdtTgt7MDM1ju+dCX8ujjiKa4NdLDIOUyuee89YimaTwy/4M5sAioO9WNpfLboIS4LlbJYAd9TpKWm9unTpUCOF2vTKF1/6UXfB6fNRtyALgPb29mGo2cXs2i4pt+xKh+G8pBAzHQcGysL1Ww6p43O3UBnjKAaRE2DBw6fAwKuPjKCsLrTscdruhucpCwLrcrN8zKKfgosFBDEQiFAZQlLccZgDgJgi6Pdsdr93n+f39GFCAfvHgIF6sVJDl+Lzmb8tZFm6emUFHKOTG9BeSCwcG8EKlgqriBPBl4XOIw7JwcyaDc/r6cFgyCVMIN9YMADdls1gejbZtHZsOhfCZpUtRt203ObIF4Nz+ftwxO4uyZSEfqDNna57v6wP9/Th4L4uCJxoN7KrXMdNseiWDANXXA7AArGf62oDi5Gjr0E6CfdcBoGbbuGt2FnXlLeHzRSLA8DIPSItFr++1LjroAv7fDGpshbGwRdpmOHAcsvaKxfaUqWz589/8vy48nqCyYVkUNvjwh4Ef/IA8DkIQGL3yir/pjD4e/ZyOQy7ZSISy4NnBpY9DB3PHAX7/e7Iyzz2Xsvzn5iirW292U6tRLTwfv2TJmwvQazWPkz6dnt9U5pRT5ocaFhL9GdHXOLiW7ZSrP6REo5SY+F//5VVeaB2L92tZBHRNGo6DVxShR4dp+shc+NvK/5/e04NN5TKeUzFIKTXGM/W0MrCaQusVrgE4P7kM8NtqNTxcKKCoso1sKV1L3wYlsnFS1o56HTfPzODiwUF0hUKIAp61KgQ6QyFETRNJ08Tnly/33eemctnlqXc9mgwgbVzH7+/rw6p4HGXLwk3ZLKoq61oHdNeFDk+5AYBny2Usj0ZxtraL94bDcCQ1s+FxsAs522rhxmx2j4AOtTYu3757G9JNZnNANKs/npzEIckk/mbZMt/x52vg3k4sx8HDhQKuy2RgAxgMh3HFyAj+fsUK3J/L4bbZWVdR0u/ZBtATCvmqEPYkjxYKWK+162WKXG7cI+FZ/lJ6Ge5SSkQMAz2voT2uA6qs4HM5AHr6qM0oy003kUUatKik9Lqq6RSkQhCQ9/fT+8Wit3EnElQCx125Rke9xKRVq+YrCPyzkHsWIAUhlSKlo1bzvn78G6DXb74Z+PKXifv9ppvo/YkJyjo/8EAaG2/u+jj4XM0mAa9t+7u56duDLvz6zAy5mG+6ieLN7RShYL3/m0UsC3jwQZojKUnZ+MQnXntHt3CYPBDZrDdf0Si9znOfSs3vKPd6SCpFxEV/arII6JpsrVbxjdFRCJD7+8B43G/5SGI2CxsGooaBTy9Zgv+9axe2VqsEutq30S1RUsBuaDzvfC7A26wThoE7ZmdhqeMY/B1tp2AAlepcv69UcEMmg7N6epBttVxlRAL4yNCQrzGLIyVGGw2siMXcFpoAgY8RNIl4iPAAhaXDNJHXfY0a8ECBg/seCIT0OTSEwBeWL0fFtvG3W7eSl0N9ltuU7u2exgqXayGreXF4/iVx0jOJTLbZdBPf9kamm038anqaaH1VaOKnU1P4yyVLcHJ3NyqOgwcVXSsAlG0bTeWZOLW7G8vbBEYtx8Guet1XzgZQTgQrdvp8SeG14nVzErT1CwmBNfE4rtCoygoqxLInkLel106VlbuOQN1OIkFWqU6Bqrui3XNpvc47Owk8X36ZyoO4Pv3II4lFjeWmmwjspKR672bTn3jH5ws2PtF/DjyQOOJvvpnKz0olOkfQ4uvoIEC94QZ/Jv6OHWR5nnIKgVe1Sta3bpnroM7KgB4KME1ymTPjnA7Mjz3mka9w4l/Qm6HfX9Dr8McUVnhYcZuYoF73n/oUkevsqyxfTiQ93/uevwPfsmUeU9zJJ3ukPvuL7NzpPbeh0BvbwjYoi4CuRErKQgcIHMYaDUw0GggLgZhpupvtGT09eLeWuPWFpUvxd9u2Ya7ZdEk7XAIRwEuQExpzmg78oA38rO5u3M9Z7/AsZ6msdH2n4GsboMS6TZUKzuzuxie19k2zyvo1hUCm2cRUo4GrpqfxlyMjdJ88PnVetpJ1d64EkDQMtyFJKhTC55Ytw//etQszKotavw+fkaJ2zZhhINbGUs00m14IQpsHU0rYjoMpdf6EyuBeYNF81+S/g/zxBoDttRqumZrCRYODSJnmvLr32VYLXaGQL8ueXdpC1W4bAMYbDfxschIfGRrCMakU1iWTSIdCSJom/m10FC9Vq7ClxPWZDFKmiQOUUiWEwEAkgldqNfzrrl3u3PvK2gLzyOAdN01KbIQXeoGUMAEsi0bx4YEBd74AYH0uh5rj4FxF2pNoU1wrQXH9quqJzqGVzwXSlM88k4Dq0UcJ1Hnj4hI1NRTXitat5yOPBP71X+ddGgBZ7tzrWkriZ9f1ShbdLSvE/Cz6Z58lF/5ll1EZ2ne+QyCtd4BLJoG/+Ru/t4CVjFCI3n/Peyip60c/IgYz3eMQJL3Rx8hu8pNPppanwff4dzBcwPPE59Wde3Nzbw4XcPCehSCvyq9+RTz2r0XWrqUmOboIAZx+ut9bsb9IrUa9ATgslU5TO+E/FqgvArqSbfU6fj49DV9yE4BDkkl8cmTE3dzDCoRY2OXqWk+aVa67gnUXNseZ9fdun5ujzVW7tlsOx5avZtUL7X2Ow3NtsiUlvjk2hnN6ezESjeL/3bULRbVxf3XHDt+4OBHKMAyEDQMWiHeez3tmdzdWRqMoKjKY3kgEHx0awrfGx1HXeqNzTN1QfxtCICQE3tnd7TaMYWk4Dr47Po4S9wFX88bkMGONBr60fTuEEDgimcTHh4cRa8f4oM2jALyWtOpcPHdQ9/j7ahVf2bEDb+3owIeGhhAxDFQUZet3xsZwWnc3Dk8mkdYtW80UY4DdVqvhyzt2uErXSZ2duFCZLFJKCMeBJQR+PDnpHpMKhfCFZctQYURUyY2uN0F4VMPsVmd3+gV9fbhTEfI4PCYAnaEQ3t7ZiS+pNXV7uKtxbSyVcHJXF85gdBACHaoRTUcohM8uW4av79qFqoqjRwzD92yznHcecM45wFVXEdi1qzkOupFLJXKntjkdAOC668iCD7rYWfgahuExs3Htu24dS0kWdbFI9Kyjo353PZ+LW6eGw56lbBhUHsWc8HuTCCgljUcISmiLRLz/dfe7ft/BeWLhLHwd8CcnKc7/t3/bft7eaIlEPI8HKzftHD/VKs1HKrVncpZ2z0RwzvYXYcpa9rwUCqRU/su/tJ+n11sWAV0TvSOWu7GC2LxCbVRHKSX+fXycSs+AedYig6IBuLsFb9S6SxpC4OzubjxcLLosZPo5fIACj65TquP1Tfj7ExN4pVpF3XHwQ8XJ2dT8eDqbmnsNIXB4MokPDQzgumwWTxeLrrJw29wcblUZ8CEh8MnhYfxkagoN1RNdv299vFJKHJNO490BMGexA/foI7CR0s0ZeL5SwQ3ZLC5R3cl0MdWYHPjXTmg/3oKoDnJS4olSCelwGO/v78dPJifx+0oFddvGTxsNhA0DX1i2DKuUZR0SxNnP49OTHG11jYcLBcqSV2vCveG5bawEucG/unOnq2S4SoemIOqJirzW5/b04MFCAbOaL5rvK2dZuD6bhcUeBOF1B2Rgvz+fx0OqLa4EcNnQEN6qqghGolF8ZdUqAN5zFOz1DtAGbZpEf1oqUQ1x0MJUUwwhaFP79repr3pwc7dtAgjtMW8bN+Zz9fUB/+N/UL32k09SL3H9GCHIe/DYY16GfFDqdRrPF79ILt8f/tDzKoRC9JsdHMFkuODY4nHgf/5PmofrriPLPxolT4GeBKjT2QZd7KwUXHYZMZRVq/733iwtUFesoCY03/sehRuECnFcdpn/c5YF/PSn9FxceCG5zPdHcN5XsSx6rrgkkZ9ZPYTyRssioCvxJRxpYPtStYrrMhl8WAMUveNVQ1m+ALyNlM8Frz6bv7GcIOcmg6lr3jk3h6rKaNZd8+wZ4N9u8w8ep+Jl512j4TjU69xx0BCa+1wDTN68hfZ+2DBwf6GAzeWy+xkJivkyuU1Y7VBNx+ulzp4JVgykdl/PlEp4oVLBmT09OL0NKTXPFQOjC8js5VD3xdeD/j6Av1qyBJaUqNo2/m101K06gOYl4Pl27xmkTLBC0bBtNNX9NR3Ha6YiJUYiEXxx+XJsqdXw86kpDCsK4NtnZtzQhKHW0lLXdasR2ph5TdumKgU1P755ENRql9eS1/wO5bnRhZUXIaiG3wispe7Gt9QasifjZ1NT6AiFcIhKOtwXHv1YbGGXKAOSbZO7+PLL21tqv/41kb4wZUJQKdDPlUwCn/409RwHKJ7PwtMnBG2sOsAHzwV4meXHHEMxW/5cNEqsZevX02t69jpbXvp5AIrHr1pFDVM41/P44ynZ67e/pf9N05/9zhaurnCkUp5lx9fTG9y8GeSoo6g64KqriDTm8sv96yAlhRo2byYl7frraX5eC1Xu/iahECUJfuMbXnJnO+X0DR3TH/fybx5ZFY/jov5+/CKTca1TAa+Vqi4/nJzElloNtiofc7+BjseTzQAcdGH6Ep74DyEIzIUWZ9d3A3iA6UiKi7uAp967v1BAj/JNOgCkapCiU7XqoQAaAL1/YCKB8/v6cPfcHGXsa9d1tM+2HAc/mppCVbtn112u7kdXPpoKjO+Ym0MiFMJJyjKMGgY+OTKCb42NoWJZPt5xV0HQlKKNlQr+17ZtntXa24u3d3W5te0V5S3QiX9cZYPXUhuXHu7QyW9YkShaFv5h+3a3YU6Pcpd3hMN4pVp1xwWQFe6eT1nzkFTx4K6vltTmU1oCoj8P/LsupXtO31oGA7vSC7sEW+/qCFHXlJl9lV/8AtiyZeH3dUBqV0Bw993AAw+QRcpD2p17O3geHbT5/d3Vveuf5b9Nk4CY5cEHaVycIQ9QJv473kH//+hH5AbXnGou4OuJYZEIcMYZlG/wyCP+xLvgmKUkxeGHP/QUjXZz8GYQw6A2s0uXknciGNu/8UbynLBXoV6nRMhE4o/bOOaNkoEBvwLW0QFcccUfx90OLAK6K6YQGInF3I5mAmTprI7H8R7NuvzV9DQ2VSpo2DbYbjKh4ty8yWpgZ0gvPuoT4Y9/M7CwhQntXAxw3B2NjxPajnFkKoUjk0k8XijA1CxSn3tfCHSFQrh4cBARtSsJAF2K+MUdpxC+mLilXjOFwKldXbhrbg516TU56Y9E8OHBQUAS1ezvikVfqKDhOG6vdpbl0SjCyiqFuj8YhnePGvg1HYfCBupeb5udRdI0sU7Vt7DyxYoB3wM3MHE53jXAf6JYxKC6ZxOAUNaxlBJXTU15xDpCoG7buDOXw5VLl6LDNDHWaODhfN7LRdDW1xTCO5daJ34uUqEQLh0cxGSrhdtmZnxlg/MUIn4upHQ73/HzEDdNXDw4iI5QCNvrddyWzbrJjWzdu2V02rMASa1yfzo1hZCghMXPLF2K3nAYO+t1/GBiAoORCN7e2YmbN5Vx8I5hfPjD3pqVy/5ENn5UdSDlYbSTapU2fN01HQQ9xyGrmduUplLe8dxtjSV4LT6XDpwsjQbw1a96n1++nKz/Wo1+9HNt3gwcdBD1JP/EJ7z+7Xzenh7KmP/e97yY/urVxJ62aVP7cbVLsKtUfF9zAOSN0Of8zSDhMOUatJNKhSxzPbxQrb55wgavt8TjlPXPIYlwmBL//liyCOiarIjFcHpPD+5Vm/WyWAwXDAwQu5iSnGVR0pgQbpmVKQTO7utDRAj8emYGTXgbMgOw61Jm64n/V99kpitlUI8aBt7X14e0aUIKgYlGA+tzOS/uGrC6+0MhdIfDOLevD1dPT2NaNY45u7fXJXWRoKzxdYFCz8cKBVw9PY0iJ7lBy7JXYMWJb0+WSrDUeKUC4Jhh4Ei18443m/OVF8x3QcdNE5cMDWG22cQ9+TzO7ulBXJl322s1PKR1xNCBUwAotlq4KZtF3DCIuCWgNLkkN/Cz2OmMexXbRsm28d7eXlxjWW6WuASQV+1n2eqXUuKVahU/mZrCx4eHMRiJuOczQP3pZ1otjKsOdXzdU7u68EihgLLjIB4K4cKBARybTmOTCmsIDWh1BUa/Vw7N6O+1pMTvikX8X4ptzlDeGP7Mmngcb1fekAfzeR8Vr6Huj5/b/xgdRUiQN2W62UTespBpNjFTD6P6DBGyAMQKxkClg2XQEgUWTorSj29nVTuOF68fHib3/I030vvDw8ARR/iP5d9HHEFxzJ07/YoGKwfnnEN/33GH556vVoGvfY1+B8d74IFei9MHHiAL/dJLSQkAqLb8Jz+hTnA0/3SeHTtI6WlHV8vX0O+Xf3d0EKNcLEYKjEpr2G+Ewwu6QvVm8jK83vJ6strtqywCuiZJxSXeG4kAUmIgEvHVcq+fmyMub3gWPIPw5kqFgESoMie1ebKF7nZTMwysisVgS+mei05IG3xICJysOr29vavLJSd5vlTCvXNzHnmN5gVYm0jgKAXSByeTeH9/P+ZUWdgJnZ1I7ibt9PlyGXfncshozTxcRUFKvDWdxsvVKo5MJpEyTaxXfSV1AM23Wrguk8GfDwzgyFQKW2s1vKTMGv5eB0HeEALHpdNoOA46wmEcn067JWMHxOOuEiLU3L6skW4LITDTaqGkYssx08SZPT24Y3bWpWp1gRxwwU73aLD1fFAigfP6+jDTbOLuuTmULAuONscMuraUmFRzdFgy6YtTTzabeKFSQU2FXIQQOKmzE2f09mIoFkPNshA1TZyQTmOi2cTts7PkPQggk+uOV78liAO+paV8cyXARLPphQqkR2okhEB3KITjFaB3hkK4IZt1x85VCBy+Yf5/oc7TchzMWBYMEUatRlnjQnhZzLsr4YrFgNNOo1arkQjRneosaLncwsfrqSaWRfXio6Ne5MlxiBc9aKFLScDabPrPbds0hlNPJea1a67xkpUMgyz2sbH59yAEcYn395MysWkT3fsvf0kJdQMDdC1u68pWd6vldQ/Tu84ddhjRwk5M0Os6VS6Pv9UitzVfe2/7ib8Z5MQTaS527qT/haB8ggCX1aK8QbII6AHpDodx6gKZ2QORCFKm6SND0algAc31K/yZ5Bz7XBqN4qzeXjxRLGJno+GrKzYE9Rg/S9UP6zISjWKd6tGux18FgCXRqK+V6jF7oFoabzRwu3L5zrVayCp/ke5B4N+HJBJYHY/jiGQSUSHwYD5PyV3qXIYQqDmOawUuiUZxZk8PGraN7cpiPTiZxAF6Jo0mUcNwY+ss/ZEITtOCdXOtFrao9q9sMbtudDWGE9Jp3D035wKt3gFNaGul76abKhWsicexrqMDN2cylDzG8yC07mq6R0StxYjmtfnp5CSKKhfAARHcZJpN1BwHfxZolVuxLOo1r11HF/6PE9+4rNDRPy8pGfPGbBZHJJPe+NRndjUaeLRQwImdnTgkmcRZto2JRgOPF4uYVT3WwfcGT+ni44X0m1dCECjp/7MbmcErHCau87PPJiCdmCASlm3b/MdwQlokQm5cKQnMVJ4hjUt65Wf8WrFIVKpHHEEucR0UVV8h91ghyGo0DEqCMwziUNete9ZxdY/DypXkLj30UEpue/JJr3nL9u2U1X7JJe09DfpS6u8tX04gzr3b21mvlkUUtEIQOKbT5FV4o+S22zzlJhSiFqILbIHz5MADKXeAPTlSUqhiL9oxLMrrIIuArklWWWkAbdrB+um3pFL4XaGAyUZjPlgL4ZYpcSzYRyYDAp6hSASHJpN4guPM8IBGADg+nW6bedwXiWBdRweeK5d9FtxINIq1C4BlO5lttXBLNovNlYoXCtDBnDd79fcTxSK6QyHsrNVggZKqpDZeKSj0wGN6uVrF08Wiy64GUAy9+d/xwQnhy87nMANLy3Hwm9lZtFTFgQiagYBvHVi5Gms0MNFsYqLZxKP5POUFqM+ykgXN8i3YNu7L5eY9F4BfaTAFxadDQsz7nNQ+y7FyBJ4T93zaeILkQk0p8UKlgsMSiXllerOtFrbXajhR7arHptPYkMt5rH88N4YBqUxJXVlw19b/kv8+NDc5QEDwznd6NeP5vB/MeSqXLCGrNRql3wDVo8/MeGCnLx8fW6sRH/qRR1IjECHIKmQPgr5cQWa5RAJYt46avwQ5xFmGhsi7sG4d/X/HHVSaxgqLlJQQyPH0IKjzHOkOtGXLSGnp7CQOd73/ty76MdUq3ecbBegbNlDZ39ycFxowDOqGt5dtCHDUUfSzKH98WQR0JUXLwu2zs3i8UIAhBHrDYaRME28NNO5Ym0ggJAS21mooNJtuMpsLAIKapwQTmoSghLQDNBc+VJydrfk18biv61dQBiIRrIjFsFMpFAK0gOF2u8QCMtdq4QVFnOxu2jwWaFas2qVertVgcbtT5aYNgmPCNHFoIoFd9Trunp3Fy8qa5l15e72OXfW6L3yxN7K9VsPz5TJ21evudRmQdC9FS0o8XSqR0qBd1wU5KSEZmAHPxa2Uk43lMvBuYuIAACAASURBVMqqpIxzGDhZ0XQcOOq+S5aFZ1Wb0nmirs3Jayel03iuUsFxqpsbS1c4jAMSCbyiQjSuJ0d4IRmmsOXXALhJikJDpLrj4Ili0bOw1dj7QiGsDsz1xlLJbaEKNQfScRAWAqsTCWyt1bw8Byl9CpM2pfOsS85qP/hgeu/XvyaLM5fzLHj+HIN/sKQpqDQswETsWunnngscd5znlm+zDC5l64YNFAI4+2z6P58nazjo9l+2zANz/VzBn3bv6aVmPPaBAVIQOL76xBMEmnysaXp92Dmxrt119lVKJaJRTaUo3LCnrWHjRppXwFujjRupR/veAvqivHlkEdCVzLRaeKJQcF2cs60WHi8W5wH6Kd3dOKW7Gzdms3gklyOmOH3HA1xSFL1GO6Vcy2/v6sLWahUZFQNlgHIAnNrd7fbkZqnaNh5RxCDd4TBO6OzEjulpQB27s9HApkplr7t6AXCVDWYd8xHqaPeig6ebtQ2K666Ox924+FA0irem07g1m8XLik+ej+Na8q3VKrXqVHJCOj2PfjUoW6pVrM/lPFY7ITAYDqNbHdet1Yboc+kKl47poKk+w68DwKp4HOP1OipqLfVyQH0tHSlRtCysn5vDSDSKw5JJvKKtJdS5HQDPlkp4sVYj/vhwGMel03iiWETMMHB+Xx9ugeYZATDbbGLGsuBIicFIBFXHoSY9UIlxurKhjqnYNp4ul32NWjpCIbyjqwsnBHye+tzwOQwhcEgqhXP7+nDrzAyKKhwAKSETFsSqMuT2lG9KBwcJNJi3nGPVZ5wBPP44cNdd/raiOsB2dVH5ky4MKPwZBteJCQ/oeOhSkuv/rrsIbEZGyDXMQKktAaSksT3yCHkDDIMIUaanvWYjlQrdy5IldE2WzZtJIdHPJQS55A2DSGz02H6AagJdXUQFe8wx9Przz3td5LhKYM0acm3fcsvCRD37Iq0W1ffPzJASw01rjj9+z8fyNds1nXkj5ZVXvFi8lJQz8Vo441+LFIu0rvy8DQzQ9fc3WQR0TXijY2t7HjmIJhf096PYauGpUgkNRUjiErIIoogd4VIwIXBAPI4zVfnbE8WiG3N3WcEkkdiUbRuGEG4c/L5cDrfPzMCWEv3RKA6Jx33Wk5QSmUYDk42GLxt/IYkbBgbCYUy2WvNLptT5AM8iZOBnshRDCKyOxfAXIyPo0AD5kXwez5RKruXPc8jnfrpcxtOlksuGVrFtnNXbi+geyJtdAh4lx6XTODtIUiOl20KWLXBWJFwiHu1eocbUqyoDTuvpwQvlMsrKt8ohBUffZdW6TjebuCmTwcp4HEXLaruWjuPgEcUJec/sLKKmiZlWCxvyeSQMAxcPDuLKQOe3B/N5/E5l9r+jqwsTKg5etizyLgAYVs/TRL1OndL4fgwDccPAYCSCtYmEL//g1VoNmWYTJdt2G7FweV7IMHBALIbReh1Hp1LIW5ab7W50NSGPzCOdTaGnx5uGU04hV/pjj3nu56EhAoP16z3CGDUsaseq6r4PO2x+q8377iNCFraow2HqS3377Z4lzVYwg+fEBL3/539O5WW/+50fEPUGKI5DpXLr1xOD2ZIlwJVX0nvPPUdu9Asv9MazfTude2rK35jFMMii3rixfftUttITCUoU0xuMbNgAZDJ+5eD882ksa9eS14GVoETCXye/t1IqUYMaXhPLImWho8MLbbST4WHKzufsfyFIUXotddQvvURA+Fp46MfGqH2uTgccj79+gO44lCMBkLdncpKSIAFPqUwmvWqH/UUWAV2JlNSZy2FyGBBP+2i9jqhheM0vhMDyaBRd4TA+OjyMV2o1TGvWdm84jGyrhZhp4qyeHoQMAx0BFygDJF+XweO3ytQIC+oOdmA8jttnZlymtmyjgZlGw7XsGIR3NBp4uVr1AfqWahXLo9F5HOhLYzFcMjSEn01NQQBoKKuT3e1Ci6s6IIvdVparIyWihoHLhod9YA5QhUA6HEbesjy3vQakXGvPYHLH7CwSQmBAjTkiBA7Vi46hMa5pVnNbJUtQLXyB3cZqB9YtbZ3FLhkKIWWaOLW721Wc+sJh1NTundXauro8+XRx102+rVbD9loN3N2Nx8aWtMs4B+o9fvfsLASAguPgV9PTSIdCODCRQF5l6/9ZV9e8BLqibeOpYhEN28ZANIpPjowgaRj4p+3bUWNzSgg46v2/GBnBrkYDz5fL6A+HIQHclMmQO12NW1/Lhm3jukzGnaPucBgndnbi7npd9WQni/xDH/KDzFvfSgCyfTu5jj/yEUrkYgBUw4JhEJhcccX8JQMokapaJWWAgdiyCLA/+lGi1Rwf94BQL4/avh249Va/Kz+d9uqig+7rdpSwb3nLfCvs3nsJ4NRyu8e1WgSQQRe2lJQ/0N9P7x16KLn3g59h65zHtGULgdgtt3hKUDhMoQvmlt+dvPqqx08P0Proioxtk6J0yy27B/SLLqJz5XJ0/UiEsvn3FZR37aIQyEEHkZJy0EEEyHsrjz5Kio0u4+O0nn9o1z/3pr/mGs/LlM9782eadD93370I6PuthAwDKcNAXn2DhZTYUa/j2ulpdIZCeLJUcgH0gv5+nNHbi7F6HQ1JzFsSlLF9yeAgvj02hjnLwrcmJmAoBeBSRQQyEImQ1asBkwRZW+zWtqXEr6an8b7eXjhSEbsAbh0x7ynczOOYjg6crMV1x+p1/GhyEqd1dWFVPE7ucW0nOjCRwFdXrwZApXjXTU+Td0EpNAB1WatLKmNiYJWgePX2ahVly8JqLRnvqI4OVB0H12UyBDZSImaaiBsGyraNhqIftdmSlhJXT0+7sfmecBhfX7PGZ7HHDQNx0yRXuBpTPKCg7KrX0XIcfGpkBP+4bRtq2g7O1rqeoS4AHJVM4tKAGfRp5Quu2jb+7tVXkbOseUmLbghF0a0ycLNyxT+8lvr66jiQtSz8ZGoKnxgawhPFIkabTZzf14fucBi9mmn08eFhlGwbu+p1fGx4GMtiMcy1WkiHQihrXPpcaXDn7Cw2KA6FI1IpOFLilVrNVdZ0TwugeaLUczWj8khMwdwJEi+/TC7uj32MxjQ9TUByyikE5KxDcsyc3eSGQaCwEJgDZFHu2uW5x1khuO46TxH48pcJTPnxZasY8Ge7S0mJWS+/7NWj6xZ7O0BvJ0ELn0GYRY/vGwYpEcuXA5/7nP88c3Oe275W8wAdII/BL3/p3Q+PNZWi2PWeZGKCuNPHxujYcJgsfsvyl8WFwzS+hYTXstGg9Tz9dC+pcV8km6UObLt2kTJ0993AZz9LCYx7klKJxlEo+OcdIM/LwQf/YV3frRbw/e97SYAA/c+ir+0fI+zw35VFQFeyIhbD5cPDbj90tsS21GpuPTOXqpUsCxONBr4/MeHWe7N7mUvAGFAMABONBr4+OopD4nF8cHAQdU56UsItUh14BCgNKXFtNusDIndcUPFXSd24dJCbbbXwg8lJZJtNXDM9jbBh4C+Gh7EkFsNgJIIp5U0IK6sWAATTxKp7iJomzunrw1OlEraqbHiWlpT47vg40uEw/mbpUh+on9jZiapt4y7laTipsxPv7+/HjyYn8VSpBItbgKp7NFUCmEtiIyUsx0HestAXieDk7m7UHAf3qjj6u1TntrlWCw3Hge04+Nb4OPKWhQ/197vW8UJzBpDSlVDehXyrhZhhoOI4aKixcQIbK0vukdIjqnEz3/k6GqrwWnKMXmfsYytfAJhuNPD1XbtcEp9/3bUL6zo6cPHgoC+34LOBoHNPOIzPLFmCL23f7lYOCCEw2WxiUpWkCSmxSVVDuBn1UvrK0TiGzjkjektelmDufaFAnN5bttBm94EPUOIXP36plJ91bXfVk4WCP9auTeu8mHTwM3qsWf/shg1+9zwfp5O6sDSbXqlcPE5xb/06uitdBxodgDs7yX0ftCArFXLfPv20/5p6fbo+Tr7H2Vlin/urvyJFaSEr+aqr/GVwzSYpQfG4dy5OuvvMZxaef15LxyHlKpmkUMEeomDz5KabyMpnEYLuRfWGwtBQ++S8RgO4807gnnv8689zf/75r08cW3+mAO+aPAbD8Nrq7m+yCOhtRKrMZt2y40QyKSXW53JYn8+jxa1T1be2atu4Ppv1lWixpSYkxci/umMHJZaBmLp4468pIHB5ttuoh1K9z4AYMgwcn07jHC2m/JPJSYw3Gq5F3bRtfHdyEgnDwAcHBvDzqSnYUmJpNIq/WbYMLcdBCIAFzzX9jnQap3R1YXO5jIhpwnYcNKQk8FVPfsmy8J3xcfyvFSvQH4nAlpQFfnQqhSMSCUAIF5guGxxE2bKwuVJxQY9d2fq3Kt9q4eVaDQ8WCviMYkF7W0cHTkynETdNNB0HuVYLP56cxCvVKnkvJDV3uSaTgSX9dLgQAiF4nd1MIXBsRwfO7OlBrtXCVVNTWBmLYVOlgh3KfX7JwIBniWuufgNaRzc1twaAkGnCBCk6nEPOioDutm+3lhzzZ0VuY7mMlGHgwsFBl1CorQhFL2tZ1A1OevkCbic3NY6FMtfZMmfhzHpOimR2QGigeu21wNat7uG48UaKtx5xBCWnfeYz1Ou6XqfYJ8eqg1KrkbvzlVe8c/Gy6T/ForeUwfd0AGj3Gf1/NWXavRKH+w030P+HH04Ur/E4WbVc+iYlWawM6M2m/2uZywH//M8Ub73ySgKBVIo8Ggzm+hi0qXfHweNyHDp+YgL40pcIjD/96d2Din7eaBT4+MeBn/+cgHJwcOH5B2gtef4But9f/ILW86CDFj5uoXEEvRjXXEO/YzGy1oeG5temP/oogXnwPKZJ67CnNqz/XdGfDfY0cdne2rXEDri/ySKgB4QtcXbXQlJzDLbYOP7odhvjz4FcxB8ZHMRPp6eJfEVKn3UopXSB863pNM7r63O7gf376Ch21uswhKAGKdKfJS6E8BQAENgckUzigoEBl2ENAGWSS3+XNktKFG0bP5uaItYxITDWaODvt23DAYkETu/uxj3KChagvtiGEPjkyAh1EpMSX9u5E5lm00uWA3kD/mNsDF9auRIPFwq4PpPxeTJO6uzE+1U7qksHB/Gt8XHsYBNOEwkgb9v44o4drhv4H7ZtcxWXY9JpfGhgAL/KZLCxXEbLcch1z/cJuGEJnZJ2KBLBu3p6cH0mAwGiaP3gwABumpnBo4UCWo6Dl1T2vaN+fp7JoKW+6by2QlsDCcA0DOJ/FwJv7ejA+/v7cdXUFDZXKn5QF15WvWut8/ppv93SRynxcLGIdDiM93IbrzZS5z7yGgC7XgLdW6B5CHSzlteHnyU+F4+RwwkwJESIbXh/khhAIMCpJQC5nr/6VW8YQsxvPiIlZZk//7y/XjwICmNjZEGyu72NTtSWdjZ4Lf2z9Tq5oJ97jsCcr795M1m4l11GVuHs7P/H3puHSVKV6eLvicg9s9astauqV3aaHRqRHWRHNhFRRkV81HEc73h15o6z3JnfzG+4zowyIzrjgsoIKK6DIAgoawPNDs3WQEPTdHftVbnvW8S5f3znizgRldXdLE2jt7/nqacqMyNPnBMRdd5vfymea5rUU/6II8jq/spXXFYtfdzxceBv/obcw5//vDcnQLeYda+BH9TZKuXfmzbR+v/4jxeuSzH1eq6ZaVIFwZVXuufYXo6s7tlg0d31b0SCQXfeLHqL3a9+la77P/yD12ujJxyykmaa5PV53/t2nYXsL65JJokSl52NUtIxb4CI8F0jv4dT3nWyfzyOjwwN4cbZWXIJ84aonjZu7sGbKD+/7Mo0hUBCPZ3c3Yv/Q4T2HygAPF0sYigUwllq4/7c6ChaUqKqqECzrRa54dWGPBAK4dSeHvxift6JowcNY0GWuIfPWrMoASI5ceYhqH+3Zds4K5nEsV1dzsbeoVTjmPrNrUeFtlaWpqQ67l/MzaHBlp0KN6zL5fC4ytyWQriJXHKh+xeSmqWwl6AhletaSjxVKOD5YhE123bYzdidzvOCpHr4/zE6il71n2gaBjpNk5qvCGr2EjFNtGwbDVVN0JLeagZLzV0ATj4BK3es6O0VjeJjitMzEQggbpr46NAQqraNa1hp0deplAKdKEcPuTgueihqV61qwi/pZhPfmppCnfvuS5d9D+r+ONdFut4hfad2rr3vHujfhRAwRqrA+2Yh88NtLV3LolhuZ6ebPMQb9gsvAN/5juvC1L/Hlq6eua6PX68D117ruu/5WG1qjoWul43p4KiPLwQB8tVXA3/7t/QeKwp8DCfS3Xyz6z4+5xxqYsMAyufWvQQs9Tr96O/rQHXOOZRV7a+b93Vy9qxFrxjQ5YorgG98w80VCIWI1lRjeN6hXHIJKSfMnqcD6xuVSy4hF/7Gje7cAXdNrRZdX7/4wycAVQicfTZZ9rrwmG80HOCXYBD43OeI8jSfX0jR+/suewBdE0NQD+6iZeHWVMrjvgTg7BYSLqhI9b1PDA1hNBzGWDiMv1q2DFtqNdygGj7zcTyWLQQagMc136slQ31qyRL858QEymo36g4G8dmREQyHw9hbC5S16yj3scFB/OfkJCbqdSf5jDdrHZRsAAH1XsQ0F2TD63LV+DjSzaZrTSrFoCsYxKeXLEFWxbQdi1TNryklmoDj/ndW61c61Gt2PbM3RAIQhoEmXKWC33doTzUwlAD6g0Hqxa/JoM9U4Xtha+M4Y2nKmlN/r+47Jzh+fGjI6TXP0hkIoBOUXOm0BdbWzDX/zvPAIK97cOBaytdOT+PAeHxBPXlvIIAP9ffje9PTzvPjgLEgV7zOSa9b6rZwyypZObU1FPRY7ABgSiBkY/16smJ0i5A32FKJwMUwaHP8X/+LXNHf+x6BKD9WzLCm6R7Oa93q5/fYsufX/t/77kvW9B13UCnZ9saSklzhl11GgHrddfS5Ph5AceTHH3fB/p57yOo96CACAH8IgMdg2byZXN7tGjd2dCwORvoYfDv8SpAuPT2UMFiruce80d7pnZ0EbuxJCAQImBdjVduedHSQe5qz7q+5xo2fs7RTFHTFiNcRjS4E80aDiHTKZeBLX3pjiks74SRG7gmwbNlbG+/dJHsA3Schw8CpPT2oWBbuy2Y9lpTjypZusxUBcvOOhMNYruo0VkSjGAmHMRAKYWO5jDszGU+CEluUi3m3DojH8fmxMSf2GzYMd+wdtHkdCodxxfAwCq0WrpudRVaVuTGQ8Dk5EardHH42N4fn1X/nxf39KDLfuHQpUwHySIyFw8hq7GR6yZqn57wGKjrQ6SaKoe2wDhsZ4AAUg7BDDeoODk4k/I/JSXxpbGy7TWu4N4BUY7NTWQd39mSYahdmKtaQYWBTtYpvTEw4xxzX3Y0zVAbTRwYHcc3kpEN6wq1o9VbB7OkwtOuvX6uH8nk0pMTGSgW3ptP46OAg9lf+RyEERiMR6Gx/nnvpU0IdZUkImNLlTNfj7jwv/s30tfyMNhrEOqa7ZBmgDYM2WoA29G9+E7jgAnK1Mij5gZPfW8zq8m/0+vcMA1ixgsqtliwhUOea75tv9jYmYbAyDLL2f/hDshaZEc0/Hz8VaLVKcWXDcDPWea5+sOW4e6lEbVsPOMCdhxAEIkuXktdifn6hy1u30IWguPxFF2FReTO16n75wAfItc2y115vPm7d3+9y13/yk7ROPZO8nTJz5JFkGT/2GGW7n3rqwnXZNimMExP099e+RsrIF7/opdZ9o/KHSh6zB9DbSNw0cXpvL1ZGInimXMYzpZJj6ejWI4OkKRYCY8gwcGA8jtFwGKORCDaUSni8WHQ2St5Y24kQwtnA34ysUOD/YdvGj2ZmUNRd5nA3fwNEd/p/tm7F6ngc5/X14fZ0Gk8WCtQKFcAv5uawprMTD+RyyGmgFDNNXNLXh5BhYN9YDCd1d+PBXM4FArb6FGiwxRuQboc6xxJUngO9/72h3gfgcsDz+1xjrsDSSbCTknjMtWuxoVzGr5gvXEoc0tGBk7u7Md9o4DVVn80eCwZ4wL2/iUAAF/f3I6zm2x0MYlu1irQy44SUuD+bRcI0cWxXF5ZFIrhsaAi5ZhO/mJ9HUVGVQriMe8xUx/fCn1FfUdeeW9n6ueT7gkF8cmgIU40G7shk4JDV6AoUgNWJBMqtFjbXag49L18Hw3cseI4gBdVRDIQAFCD63bKBgGvl8O2amKASJnWbPBzmDOz+bHa9MYvuBdAeAed8QlB2+g9/SH8feSSVWwHU8EUHSv9802n6jmkutIL5b1ZU+DtHHAGsX0+vdR53fX68xpER4KyzyJvh730EEACFw95z8Pd5rn19pBD19tJ4u1JGRxd27ns7ZK+9iNP9+usJqCMRqojwb2mdnVSe2NdH4YrFQFZXprJZNz6/RxbKHkBfRHqDQazp6sJss4lnAced7CQZgcDFkpTd/KPZWYQUuHQFAvjE8DDChoGuQABrOjuRb7XwZLEIwHW3PlUsYiwcdqhPLSnxX9PTyCiwiBkGPjE8vF3607W5HNarcU/s7vZwnR+USCBoGE5veUObO4NJ3bIwadsYUC7/+UYDFctyQCbTbOK5UgnHdHbiwUIBTdvG6T09SIZCWKPcwR2BAM7o7cVwOIxXKhU8p9qR6omDumdAD1fwtXAoTuECuP+7zns8rmZF8lh1y8KPZmfxyeFhTNTr+O+5Ocxowcgl4TDGkkmc29eHSfbrKqC9M51Gnpn01HUKCoGjOjo8IYlttZqHL75gWbgrk0HcMHBoRwcOiMed56RsWShbFn6byaBpWWStq+vgAVXNwuY5OZ9p97vYauGHMzOoWhbqkuLyyWAQ+0SjeEJ16uNrMl2vE0kOX3shcHxXF4ZDIdRsG3em06j4vC6sYOmKn/CVV+nA699YpSQLlC+h30rXwV1/z58gJSW5ro86itq36hZ7tUpNRwCyqhMJYvhiYG0H5u519oIwz5nZznRvgpTkDXjxxYXz18czDLI0L7iAlAamjAUoJrxmjXvs2WdTUmCh4I3x87ixGHD00fi9l8MOo5BAtUoKDN+fdrK9+LV+r7Zn7e8Rkj2Avh15oVTCc6qel2OsTrY7XNeslBLbtHTerkDA01gEgMfCZIuwNxDwxM6llHitUkFaqaCmYeBns7P4o6GhtmVMTxYKWJvLYa5ehwRQsW3ETNNhX2Nw1MFHf1/ftLfWaliXzzvv8ewFiHv7wr4+dAeDqEuJU7u7F8Tc+0IhnBIKYWUkgpFw2AGR50olTHF2vFJ+9HM4lrjaRRns2HLWe7DrAM9Wqe2bqwVgg0rS2z8ed2qzeTd4vVrFo/k83tPV1Zal7rZUCmXVBMfg+flEB1iDbhzmlbUcM03so5LwjunqQtWycMPMzAJF0L9+XVGBGpOfLV5fzbLw09lZbCiXnZI4AUp2nFVhD8B9JjPqPS5nOyyRwBm9vegPhdCybQSEQKrRwLpCgUr+9DXx9R+sAavzMF7o8gCdmqID8vrffEsB1+o88kj3vQ0bFtKj6pYqQJbsmjXUMpU/1/Qs571cjtq06u/pYy02J/9583n6e9UqUhKmp6kxyooV1Jf9N7/xdmbTxzQMUipMk7wE09OuUuHvNPae9xCY3367C3YAWa/HH//O9S7f1SIEKTNv95j8W7+v77Tcdhs10TEM4Pzz357wx9slewB9OzJdrzuxUCdrHfBme0u3rI1LrupS4sbZWZiC6FK5hztvwBx7XxWLeXjMb0ulUFVZ1lIIWLaNl6tVWHoqrybb6nXM1utOPHiyVsNsve4ClXKZcszUUCaBk+mu7XrZZtNTUsbrk0JgdSyGwXAYq3fAsw4Ay6NRLI9Gsb5YxIZymUrAdIsaPsub3fjaZzbgEKLwXHR6UsdNrsZ1ygrV66YQeKlSwb6KWpTHhRAImeaCbnN3ZTJOf3Qnnm7bCJkmjursbEuD6qxDCxNM1uuYazSwj6YotKTEywzABrXVZa+EP4t9NBzGymgUG5nwRQd4UJLhS5UKLNv23MuKZWEbU/pqSoOeOCgAlG0bv9GIzYOGgfd2d+OxUolawmrXyRmrowk5VAWe7/KA+YJr4QNO3Zrt6fF2QBsZoU1xdpYAb2iI6rY56SwUIuA75hhKRtve+dp5CPQ56L/17+nz1T9ftoxi81u30hwGByk2XCpRkhxnsvvFMAjIp6a8XogXX6S6bq7tFoJixdUqKQ587s5OyoTfY30ulDVrqF69VqOwy1FHvbmOdm+HrF1L9fMcBrBtao28s/zxu1r2APp2ZFRZmxOq2JYBiGOiDmMZ3BisIQTqirQDkhjSEqaJ0XAYR2p9GCWApb7sa26LqnsA9FKnduLvjqaLCWB1LIZHCwU01FiGvtsyaEuJ/mAQqxTxy6ZqFZlm04lTm8rK80uu2cS92axTYjUUDuO9XV14pVLB3dkstioFoV1GuuP18I3L15AtcM7QBuBJhuM8BuEbi88DwFnTZp4HgNFQCIdo2TSP5vNYm80i3Wy6Ge2ghjGHd3bi/ckkAr5ddjQcxlgkgglOOOQ5SV8JnJS4J5tFg+etWd3MBKd7IZaqToJrczmK76ux+pQXh9fKz54DvLppy/dUu87cO+BVpQywZykgBErNplOqx9dUH4vi7mj/gG1HdMD1A+BhhxGYvfoqWXH77kvZzUzSEgqR5cMsbH5PgD6+EJRdzg1v/I+pH+AjEQLtTZsWB/X3vMfLUnbPPQTA+nj6mO06yfF4W7fSj96sJRAg9/we2bFwmKJcpuchHCblcHcB+rPPkjeHwyXPPUfPVEcHze3MM3ff3IA9gL5d2T8ex2vVKiZUfYgDrmrjlqCWsdONBiUuCW9JG4RArtXCo4UC/nzp0h1SnJ7f34+nSiWnccjO+JUc8hDNNc0SMAxcPDCAkm3j2VKJ6puFm5zFjWoEgIhpIm9ZWBWJ4PiuLtyTzTqkLc8UCogJgcFwGMd3dyNmmqhYFu5Ip7E2l6NmK4Jq5ROmic3VqgvmbcwkB5RtG1LLUuvnVQAAIABJREFUAIdaM1u9zHZmQOuZvthYmss+IIjdbjQSwXFdXdhUrUJIie5g0OMReb5Uwm8zGWTUOh0FCoBhGOgPBBzrPNNs4vFCAf2hEI7o6MDqahXjeuGxeh6cpi1S4v5MBvdls86zAemWwOlrYMCdrdfx20wG+8ViONFH1MJr1dcrhIAtBOIG0dkWVEti5xpqxznz056Thm3jSeVHZsXCUb7UM9UucVO/XX6ACwQo0WrrVtroVq5sXxZ03HH0w3L++QuPaTbJ5c2tZnXQ1ZvKbN5M/cO7ukgxKJUWsqRxZCsSccuieO6sr7VLRJOSusrpbGm8Vl2HKhRISfB/Pjj4h1PjvLvENKmq4d0guhLHz9ijj9LvcJjA/LTTdn2Xu8VkD6DvQJLBIJYryzXTbKLAYAsAQmBFNEoAZ9toSYmpeh2WUCVC0Nyei8jL5TJSqutCU0oMhEKImyYsKTFeq6Fh21iXy1GCXTCIgzXrsl/NLdVsIigEOk3TwxH+VKGAsmVheSSCF4pF6qYm3S5yIdV4JdVsYluthm21Gg6IxXDRwABerlQoQQzk0n4gn0fIMHBAPI6Y+s4DuRxa0u0sNtto4IFcDqPhsBN+gHZOVj5sLaygz0dnVIuZJmKGgXSz6TkOUqIvGETRslAWbvkVn8cQAvvF4/jI4CDKloVXVXq2FAJdpollajd/rVrFbem0kzAnoTHgAahbFu7O5XBiTw9MSUx492WzGA6HEVcIoCsrnjACgCeLRfwmk0FJawAjhHDc5J5uber9TdUqNtVqeD0ex4X9/Rj0qfqmEBgNh/FSuexci5Bh4KBEAlcMD2N9sYi7MhkIIZBttZz751x//Xn03RsnAY6vAb8vJR1qe13Y+o+aPkwT2HtvsqB++UtyVV94IbmdH3qIXOtvhL0qmSRL9l//1Xsu/TcvZcMGKvOanXW5x/nS7r8/ubSFIGC++253vjz3zk7gxBMXJqQ99xy5enkDF4IUgLExVxGo18m7MD/vnlNKcsOedtrOkZTsEa+89BIlGK5e7fbZfzeIDuYsDOzVKvVF6Oggz9NO2GNvu+wB9B3IMV1dTnOP709NEac31EZs27g/l8N5ySS550MhfGtqCttqNWrxqjbyimXhGc5wVxv7cDiMum3jv+fnsa1edyhLvzg2hv3icaQbDfzd5s0oWRZuVK1LxyIRB9BTjQY6TBOXDQ7iyWIRXYGAx6p7sVzGjbOzyDJZDM9HfS6kRGcggH2iUczlco51u6FSQSSdprI15XK11A5q2TZeKpUw32wipRLdGAiFBmr6j+MOV2u3pYQ0DLK4NRe3x4UsKPfgoHgcN6dSnuY4UgjsHY1iU7WKogJLThDrD4XQHQjgk8PDCBkGNpTLeECxjwkAm6tVrM3lsDIaxX3ZLLbWagssc4sm46z3+WIRWcvCPdksICWm63X8cn4ew6GQWxOvudLH63UUmk3cnEohp9hHWLnxJ/s58X3d4yAlni2VEDUMXNDfjy6tnj5mmrhieBj/MTnpdN0bCIVwueq0cVhHh1PlcGc6jccLBeQsC1XtXjoNc1RRuRACg6EQLABpLW7v7EWsTCkw41p0KV1yFSkpmSsepwYj4+PAGWfQZ3fdRZznQhCYf/jDO1+O1Wy6/Nhak0EHhPVISGcnbaQc7+ZNN5Ui8hB/Nzp+bds072OOodi2X26+2WWSA+h3NEp90jml5LHHqLMd4D3PoYfSuO1kwwYKNfw+thfdGXnpJVKE2Ety8ME7nx+wdSvxBIyPk0K0ahXV9W+vle07JckkWeFcxqnbEwC9/8gjFPd/M5zyb1X+QB+nXSMdgQC6AwHYAAqtFuogt/EvUykEBXWL+/SSJbh6YgJVy0Kh1YJhGBiv1fAtVWfD7s9TenuRajQcUGGX8nithr1UHbkDpqANtm5ZeKVcBoTAI/k81hUKOCQex4cHBz3Z8gBww/Q0Wbdqh/G3bDWEwFS97vC8M2hKKSn+r4UPuHc9k6BwLNijgAo3US1umggbBqqtlnN+VmiZataxuBXhS0cggLyiLA0IgU51nfUuc3ztHlDtZLn/uhACQcPAn42Oejjh+Rry2jnbm9+3tbH5tR6sLbVa+N7UlLM2Q61la62GcRWGcRLz1Dj3Z7NYEQ6j0zQxLYTrqZBuhj9byob/OmpK0bp8HssiEQ8tLkB18H+7fDl2JGcmkzgzmcSNs7N4rlRy1plpNNDSwkNxw8BfjI0hZ1n41uQkmlKiaFlgj6GQEohYQKIJu0DPmA5uUtLGddZZ5FresAH40Y+8rVU5q/zVV4m45JOf3P7cX3uNlIXZWereptOnGgZt7MEgNa8BqNTrjDPIutYt7EoF+PKXXWpOwJ2TRiePVavau/wBl35U9xB0dy8MO3A9vj6ur30AAAKrfJ5qtDlDetWq3WPNvRXJ5cjbIQQpUboVvWUL8P3vk4UtJd2fK6/cOY71uTmXitW2qRpACOpAuO++u2w5Oy2XXUYeoPXrvfz2/GxEo9RKdneAObAH0N+QfGhgAB9SdSVf2boVr3KLLFBr0htmZ/G5kRH886pVeKlcxtcU2bPTqU3deQGg2mqhroE1b7i3ZTLYKxZDZyDgSWwCgLlmE18dH4fTkQ0UB44ZBi4ZHPTWq3N8XD1pEr5yMB8g8+dCJWsBcKw1wAVTf6tSCbibP6gZyoGxGLbVaniyUPDEY/2lfAyUw5EIPjwwgO9NTaFkWRgMh3FOby/WKeBmi1G34vWYuVPa5RvfOZ9heLLg3dO7/mPdw+CYoIaBhEF87uqCOPeS+9X7m8Kwtf7FsTFcuXUrKWzCzVXg3wKUZR40DNQsa8H5o+qztyofGRzER7Qg7l+/9hqmVDN1h7RFhY6+utdeeKJQwHenpmgtPNcVJcg1ARj3Di2wSAAC32uvdTc19/ouBKodLYl7rnMZWbuxli4lIFy3jl4feiht9hMT7jG8ZP7e9ubWaFBdu5QETHoHsi98gcBo82b6vK+P2MP8POP+DnKh0MIWsFJSBzVui/qDH5B34Mord+xWzuepSQuvZWBg9yVfVSrArbdSGEUI4Nhjqescr/cHP6AucbrMzOwcoOu99Pla6uGVd4N89rPA3/+9W3oJuHN9K93r3g7ZA+hvUsIGMZJZ2kZctW1cMz2NL42NeawtzhbXn9BHGezUMYaKq35yaAgrolGkGg1ETdMpY2Nh1zwDsgTwWKGAZDCIU3p6HDY1wM0Wp1O6AM/jcYmWPha3dw0YBgzTRN22HSB0gEuzYg1BpCf8+aZqFVdu3UrWMwChsqo5GUxqiCAAxAIB/NnoKAaCQVzc34/rZmYwWa/j/9+yxVkzJ43p63euh+YiZmUjq+LiJdUgh7vThU3TqeePGIZLV6tfJ8Cx+vuCQXx4cBDXTk2hokCdQwV633opJYR2XwAC6y+NjuJvX3+dPA9CLFAohkMhHBiL4R7VYa+lrlUQwNm9vXivDzVy3HVOSc92zICqcrXHAwEPgY9zrbTXOdX3QIJ6GUSUZ6Glrgk9F4DUrFy2zjWdaAGPuf7Y8+9Ggzb7eLy9C5XZzBYDXynJ0t+82Y1pP/II/fDntk2c4v393vayPIdIxM1SZrf+P/0TfXbKKdTVTD/3l75EnxeLFFLwJzyZpkuzalluI5VzzvEeVygsXO/OWuY33UR95vmaXnwxhQh2R5nb/fe7zX6kpL+jUfI4MPMaixDkhr7uOiLG2VHlK3fSY+sXeHeBOUssRnOt1dznOJEA/uf/bN/L/52SPYD+JuWzS5bg/2zZggnV1IU37EKrhW9OTOCywUGqFwYWmAkOyPvc2gHlOgaoUcsVQ0P45uQkGgpU2Rr0d1tj1rBrp6exsVKBDUWxqf6zPBakOpfJ7/ksbv79vp4enNbbi6u2bcMWrWmObvUDQDIQwN8pF/C92Sx+k057rXhNvfZ3eJNSIgBysW+qVnHj7CyVwEk321q0md+CcdUYAIUsvrJ1K7Go2TaGw2Gs6erC5moVVwwPO9f3QwMDKLZa2KC8LA5xiTpnh2niT0dGKCFyeBjfnZykZD7DcCx7QykyljZfncpWL3fTLX9e40Sthpl6HfvG4xgKhfBgPg8pJU7p6cH7ens9a880GvjKtm3IqiTBbtPEXy1bhj6fmWZJibpt41fz83ikUMDJ3d24oL/fmVdAKaLcSKbcauHKLVuc52MkFMLfLV+Oh/N5qlkXXOOvK1FeoF7sfX7sAwG3XeezzwLPP08JaBdd5AXHRgP49rcJOHk8HovL2QyDQFjPctfnwIpGsQj85CfuWAzsySTwv/83zemhhyhWq4/VaLigzBKNAn/91xQC+MY36L0vf9klCTniCIrx3n8/uYgPPZQAVwfbWo363OtNcNp5O9pJs0nXj9cmJc07mQQOP3zH33+7RZ83X7d77yUL/JRTFuYrRKPEcLYTbSxw6aWUs/Dii+57Otvdu0U+/3l6Vq66ikhtVq6k93cV5evOyh5Af5MSNU0PP7mt7QpNBcAOsQcIJHhT1XuP0wFk2V0yMIC9olFYUiLbbOK/ZmaIV10IhzDDAWFuUCIEju7sxFyjgZcUHzfHqXVrmF+bQjhAxFzsUlmGTlgA1Is+EQi43Niay1yo9QohEDAMxAMBPFEo4F7VaU63mp2Mbt3KVxZuyDDwmSVLkAwGkWk23TalwmWF81wjdq2rMbiZT8Aw8PHBQQyGQhiv1dC0bTTV/KabTWysVPCpJUs8hC1hw8DHhoZQUzv9N8fHMdloQAqBEIDPLFniEOIckkjgw4ODuHF2FuoGQgiBQ+JxBA0DTxWLzj09N5nEkVorX0BTPtQxBnsrQJUNhpQ4N5nEKSpe3qGeLZaWbePfx8cx12g41zdt2/j3iQn8/fLlni6Cd2cyuCOdRks9h/dns+gKBPA+5e/8s9FRlC0LV23bhqzKamNaXQPAeKOBf9q61VUY+VkV0gPUHC/2gyr/5iktXUp159wOtdWi76xdS2Vmp5/uTB2hELGIffWrXmYzIcjF/Od/TiVCt9220NWpv7ZtiuNzUpbuugXIkgoECGh0YWs/mXST+lhsm1jEuKHI179OHOgdHXSOeJwSuNasoXH97vDvfpdiy/pc9TltT265hRjl9OvPdK+7S/zntixXafr856kyYX7ene/OAl04TPdGD2F85CNUPfFukliMfr7wBVJk3i0Kxx5AfwvymSVLcNXEBFLNpkNAAgAQAnvHYrhkYAA/U7Usq6JRfHhwEJASd2YyeFolKgGu5XlLKoVfp9MICYHLBwfJzcugCwKvpZEIDojFcLdigjssHocN4IVKxeUK11zoPB9h2+gMBPC50VFM1Gq4L5/Hp5csweP5PO7KZl1XsOpHz9YhAxC75534sRDoDgTwmSVLYAqBQxMJTNTr+F0m47ZLFQLv6ezEfLOJV1QGE7v4Pz40hOWRCMY0v6sTFhDCMx/Hu8CWvVrbhf39uDebRVoxy92geOyZ1IRr2VtStqWa7QgEwEbDZ0dHXb54wJnXlmoV32AviTavfeNxXDo4CAHgtN5eZ4cbDIUQMAw0bRtXjY+jwKngal0M5PzbALCxVsOv02l6PhaRitotpbqXEALVNllXDSlR1XzgDb4eSnqDQcT0ButqzXytBIgchpUOr7Ne3QcNtP0ubd0qHRwkgNatLd74ms32fN86qYm/f3d/P7GDlUpkDfNc/KAOUCkZv88Z8X4Q0r0I/PrwwykmrEuzSR3ruA0tQNb6v/0bcN55ZJ1+6UsE4sw45hfOitavQzhMCYI7ip/Xam5Fgf793Sl+78wJJ7itXpNJyjPQ59yOrGYxufRSl8EPoM5979YOem9kXe+E7AH0NyC/mJsj5jWQ5XLp4CA+NTyMb09NIddoQAiBmGni8sFBdAQCOKG7m4BBuXBHVQ00J685LnMQWJdVjbuUEjfOzeHcvj7cmkqhqUDMBFnOZyaTqNo2HikUsLFWg9QsUidOK6UTQ2c3d0VK3JRK4fMjI1gVi2FZJIJXKxVv/TGA93R24lhVqsf/Rw6YSzeZqmzb+Pn8PP58bAwR00RPIOBptSoVkOYty7HWbWVZD4dCniYvyyMRfKC/H79UJXoQGkELewbULnZSdzcOTSSwNBLBA8orUFY11/wdPy3ojsSfHZ9uNvH18XGULQs5lX3PaxAAwkI4MezuNrFsASJS8fRwZyUJ8GS4N23bTbxrI1+fmKBrqN3LBVUGSgyQ90LPO7g3m0UyGMTRnZ2wpMTVk5MotVoOJzpfa+eaa/N01615RzQA1MlQuGyNKdyjUXJLx+OUdPbgg75r1GYB8ThZePPzwA03EEhecYWbhBaNegFQB2Sekx/gGQw6O2ks1u385+ee7O0Sm7i2Xf/O+Dhlq9dq5JLfe2/gE59Y+F2ArMzvfpcS4lgRMs3F2c4eeYQ8EUJQEpoOoFJSVQFTtL7TctxxwIoV1HBn/XpqxXrOOd7Y8ZIlb378vj76ebvlG9+gexYIUCb62Njbf47dLXsAfRFZXyw6fa95s8s2myirjVeCeMMvHxrCFUNDyDSbuDWdxnn9/ThEuVwjpon92+wOJ/f0YK7RoKYnaizdNQ4A+VYLRyQSCAC4aX4encEgPtTfj65gEBsrFbxUqcAAUFM+TEMDDRtuLNfpP69c1tlmExHDcChW/eAnQVbcnZkMXqlUMN9swgS8LVi1HTPXbDrAcVhHB7bWasT6BQKXxwsFNGwbpppXQF1LvwQNAyd2d6NsWeQxYCtbA0R1cgyGQjggkcB3JiepNFBzbTvApMBvMBzGRYuZTW2kZln4z8lJZJtNTCsXvICWPKiu9Wu1Gm5NpfD+7ew8hhCesIx/PQtIWtqIlBJ5JuvRvSb8fXXcs6USbk2lUFbc9QG4cfsD43Hso/mXsyq5Tgdvni+/x3kMjlK1dwkikwOe7XYAPBDwxs1NE9hvv4V9rbk+nOPSQgCHHOKykPlln30oJsnlaQcf7H1kjjqKxnv8cXL1NhrUpY0tcXa765Z5LEagqoPgwQeTAvHMMwTYJ5zQHmCDQbIaf/QjNyZP94ZeGwYl+j33HMW2P/CBhWMsXeom4wG0tgsvbJ/5/eyzlEXO7n2+ZryWY48lT4U/ZPBOSVcX/XR3E+nO2Ni7q/lLO7nuOuo4yGGYa68lUH8DW8PvhewB9DaysVLBLakU5huNBbFlJw4MsuIqto394nF8b2oKFdvGQ7kcXiyXcYWiT20nI+Ew3t/XR01lVMKZ0EDIVolXQcPAiT09iJgm4oooZEO5jNtSKRSZTINd8ppC4BDJSCIE0YlO9Mz3F8tlPJjPg5O8GAyfLhZRkxJF5TNzPufvatZhsdXC9TMzuHx4GL3BIIbDYQht19Ndzjwv3SOgS8w0sSoSwT18zaG54dUxByUSWK2UpCM7OrCpUqG6amghBrWW3lAIF/f3Y+82aae/zWTwAntb1M8FfX24O5PBpkoFDbagteviALuUqFmW0+GP5Vfz83itWsX7+/qwTzSKc5NJ/PfcHMqcAQ9X6TA0/7Qh2lvbfO3f39eHn83OImdZMKR0lLSyZeHfVBljpdXCvEqY070JB8ZiOLO31/EmXD8zg2Kr5T4H2rlZ2dA9EeBrH7Ow+pgmsmm3xAtwrVYhCFhvuAH41KcWgs0hh9B7zz1H3zn7bO9met11lDDG40ajZE23S6Tq6yMAZ1az1asJWB54gMhRjj+ezsOAyP2/16zxKgY9PfS94WFyiY+PAzfeSBaoX9lYvdrN4OYf3WUvBLn5dbc8QCxtL79MMflzz3VzA4SgfvHtWoQWi5Tt7z8P/+a1sydkd8mSJW/NEn8nZXaWQicA/dvNzbmv/5BkD6C3kVyr5SQgORaMDrjasVJK/HR2Fi+Vy2hJiderVUzX6/ipoj3Vk5t02TsWw1nJJG5JpTBTr3toWcNC4ISuLnSokiO9uUi22USq1fI2imGrTwNdtlodoIbrPuVdLdtsYl6FCvTPmb7VyaZXLn/9uzxm2DSxUrnOXyyX8VSx6Kl3Z++Dowz4LW6fjEUiOCKRwNPlsqchDMvScNhpiXpERwdumZ9HQQd9td6OYBCHJRJ4uFDAI6peaO9o1EkOm6rVsLladcAcAEqtFrbV67DgJYhxwFxXRoTA69Uqvs0NgwBsqdVQaLVw09wcuoJBnNnTg5BhoMKlh1IiEQhgdTyOJ0sld33aPWonhycSuGV+HoZy/Zug564FooNtB8LLIxGsTiSwbyzmCSdsqVSc8MwCYh/2QGj3aCQcxqEKVfeLxXBHN4Gm9hWopSEYBJYvb99UY3jY/ZHS6+68+WaySpl9TEoCuu9/n3739pJ1zfrx889T+1audX7lFQJAbjQzOUmb9QknEGgHg0SBulh/7ZERsvbXrqXOcpUKWfSrV7vHhELk5i6XKYb/0ENugh97BQCqhb//fjrf2rXkOs9kyOK+5BJKENwZ0badBQ6t4eF3D7vX74vwPdJzPbazDf3eyh5A3474OblZTWZLa3Ushv5gEC9rNKEGKHP5qWIRlpQYCodx9iKZE4VWCw3VKc0WAsKyYBgG1nR24uxkclELn2UBMApvUhmTtujkLW2t4zbKgZ4/pH9HVxxipomTu7txvPK3zTYamFYMZPp3pBDYNxrFcChEdftCILlIDXVdShQ4XtzmP87/3rHd3ci1WmhYFh4rFtGUElE1r5FQCN+bnnYAeKbRcMhLtqnfzr2V3uY0upXrea3GEqDe/hm9/6lyx483Ghiv13FMR4eTb8BMd8xMd2JXlyfUsXIR/+kd6TRm6nXsH4uhaFnUl0BT4DhJTpimJ99gSTiMs9o9d/r1YyVV9xBoQej+YBCn9/Z6WAJ1gGGrEaANMhQi63ixdqbPP08xV4Cs2YMPJiazRx8loNTHtW1qHwoQuCYSLkPZ9DRZ83z83JxLnmIYbjZ5qUSW+Y7A74UXaB6pFL2emaGOdrGYW45kmm5r2FKJ5s914fo1yeeB11+n+a5d61raExOU6f+hD+3Yql22jMIOr77aHnT2228P4cubEb9ypG97fyiyB9DbyFAohJXRKF6v1byJa1J6yrgMfQeCC3YSQN228Xg+j2Wx2KKA/mKlgiz7fTSX+End3Z4SK5bpeh3Pam5itvoAF7w58cyxPLUn2L83DIfDWBGNenjQ/am47TwBAUUIkgwGcVYyuajF7Vwv9b1jurqcxMB2kmo0cGsqRexogJtQpmc6aXJfLocjOjrQHwqhppShmm2jIxDAUR0d+BXv0Oq7adWD3vFiKEWK75khBA7p6MBD+TwaPje+03FP8zy0A0Gd7rbdVZEAOk0TF/X3b9dTAQAP5nJ4MJdDptn0UL46Y2vPn55851fIdDk0kUC+1UJVlVY68/atpzsYxEk9PR4w96xDO4FpAgceSHHUxW7va68Bv/sdgZ2UlPQWixHI6x3Q/O5lgNzh69bR+CedtJ0Lps1LSmpDWyh4AT2dpqx0gBL2Rkao3zx3mePzbtlClj4Dui6JBDVRKZXoHO1k06aF7vdXX6V17wjQx8bIG9DV5V6DrVvJ0zA8vPO98P9fkm3bSMEC2oM2NywCSOk78MDd39VtV8guA3QhxBiA6wEMgvaLa6SUVwshegH8DMByAFsAXCKlzAraja4GcDaACoDLpZRP76r5bU+WRSJY09GB15kEGcrNLoSnfenzpRLe09mJfWMxPKGSvzxWnxAotlp4olDAUe02Ri3ubauyKNu2sS6fxyuVCk7s6XGs9EyzidvTaacRit7KlDfjBcl1uvUptHpzdczKaBRnJZNYm8sh1Wxitl73jie1enYFZlLF5cfCYZyZTHpCCoOhEIZDIUxrSgr3R3+pXMaRHR1tAb1iWXggl8Nso4FnikXHqhX6Ds/zkRJPF4uYrdfxYD6PbdUqBsNhnNDdjT/iTh8AJut1SiZT6zG0+8aKjr8VrSkELurvR8Gy8HyxiOWxGCwp8ZpSeGwp0RUIYFkkgnSj4TC16coQ30sIgadLJaJO5esnJRKmifP7+iCEwEy9jqlGA4e3CRSvLxZxdzaLXKsFW0qs11CP76XQ25/B7XDXo+bI8kyxiGk11+O6u1GybWSY4U+xvHHffAMADAP9odCCPvL86OigqQ7H+edvP2uYu7vx5dq8md7bay8CzmLRG5fn8fnSFosEvPomzMezRe/fyC2LrP+BAZdW9Y47yF0OkJU7NORyqfvXtT0LrreXauhffNFLFKMrJvr8Ocyws8lj++zj5VBfu5bWfsQRO/f9/5dkbo7oc/Vafb/470lHBz0bJ53UvoXuxAQpm0uXEvj/vsiutNBbAL4kpXxaCNEB4CkhxF0ALgdwj5Tyn4UQXwbwZQB/CeAsAHurn6MBfFv93i0itR8AHncm1N+cbfxhRdWZaTYxUa97rL50q4XbUinETRMH+LorSJBlDW6PquKi92WzDpCf2tsLUwhM1Gp4vFh0So30c+jz8bwHOGVikNToZpkqo2M5MB7HgfE4fjU/j9l63SEzcRjlQEl63BJWgjLB78xk0GGaOL6727EMD4jHUUkmcXMq5XQ0YwVBAni1WsX+sZinzKth27gnk8Fv0mnHHW8CTotXXTFhq3RtNotXqlXYUmJdo4GAYaBq20gGAggKgaO7ujASDuOM3l58hzO4fJ4K/V4OBIOIBQJIqNawlw0O4kYAlwwMoGpZuGF2Fi0VGlkejeJDAwN4rFDAbek0Ms2mU7+tbgQkAMuy8Ijq/Ma7TMAwMBaJoNBq4YVyGZsVyUtJhRhGIxGn+uChfB5z6n4490G7l1JDsaAal9d2YDzuhEFeqVRwWyqFSQXouWYTF/b1OR6gGdVm11IoagvqIz+6CLXVyAiBMcerpaRprF9PALm9/uJ8KfTy+XPPdS1pXVnwW+xCkHV9220U2+7poddCEEh2d1PiU6nkfs+yyCNQqRCgz80RmDMA65Sn+jn11zsSHSgAd22Dg1Qml8+7n518cntu+J3ytqfeAAAgAElEQVSRE098c9/7Q5Nmk9jtOH8BIOv86ae9989/L/ke8f257z56VhsNUpR8zla88gqNuXIleZJWrHhn1vdWZZcBupRyGsC0+rsohHgJwAiA8wGcpA67DsD9IEA/H8D1knbaR4UQ3UKIYTXOOyqFVgsT9Tos5ZYVcIk+dOWP96WwYeBzo6PINpv45sQEtioGMwAwpcRMo4HfZTILAV263dccN7l6XbUs3JbJ4OjOTnQpAGQKTAF4TBKOk3PSmYSb5T4QDDod6zoDAVyutT/VxZYqe1p5ChhI9XI1S/tvqVoWbkql0BUIOGV6AHBkZyfGazX8LpdzCEz4mq3L5XBALIYjNUDPtlq4PZOhLnFCS9LS3P0cexbqWpYty2FakyCl4Lb5eUBQXD8gBNZ0dSFqmhhSNLU120ZvMIhiq+VUCABAVyCAC/r7cWhHByX15fOQQuCwRAKvVCowpMRZvb1OZj3L0Z2d2Far4T6VSq0rfz2BAHKWhabW/5173l/a34/783nckU47MfobVUD4gFgMx3V1YSwS8XT78zC2CbdfvK3+7gkG8RdLly64pxO1Gm5JpTChyFikELg3l0M8EMAZvb0IGwYCQqBPddjjWHp/KISLFQmRXy64gCzaV191S8PqdYo5H3309mO7fsvXz9rGv1kYyBmAbZvi26EQWU6pFG3Gp55KYPnjH9Nmz01cTJPKzTjipQP9YsDt90Bs3EhKAh938MFujkA4TJY6KwU8R9umFqilEiX78Zh/iC7ed1JsG3j4YWrpy62EAbrupun1lOj3kJ8hft74379Wox75+v1ndkAeZ9MmUiI/+EG31e+7Wd6RGLoQYjmAwwA8BmBQA+kZkEseILAf1742od7zALoQ4tMAPg0AS9tsYm+HbK5WqZ0nXGIQh9FL86ElAgFP282eYBCfHB7GP2zZgiZct7eABoaadAaDCACoq/PYgONOBeBYTSy8sQtodcu+cXnOrIyclkzixO7uRbPtWToCAfQGAihZlqcPPHdaEzrIK5DJN5u4bmYGnwsEsEpL6pJq7rp1zu7yBVdBupzrTg93Sb3meUdncBMAHlUMbEznqnsoAKDUbOInc3M4WGV4/38rVuCxQgEvlcu4fHgYd6TTVKqnvntOMolDOzogpcR1MzOYbzQ8bXyhAPNPRkac8rdCq4Wpeh25Vssh53E8JlLiqM5OPJzPI23bzr2EUkC21mq4LZ121ijgxq43lMt4sVzGe7q6ULFtWHB7zNs6uulmq22jblkYr9UcKx2gfISb5uexqVJxFDQpiYzm16kU9o9GsVcshlyrhU8ND+PqiQlkm030BIMeDvZss4m5RsMJGSATRjgcgGmSdePfLBeTWIysam6SEo267UAty9s61A+2DMI8ra1b6Qcgy2l4mCxhprZ8/XUaKxDwJugJ4T2P3yrTvQI8jxtvpHI27iX/sY9RuRm76Y89lihe+baEw24TnPPOo5898vbIM88QtapiLvY43vj3Yj3++TPdSm/3fRbd0t+wgRTVSy7ZNet6O2WXA7oQIgHgvwF8QUpZ8JBtSCmFENuJVC0UKeU1AK4BgCOPPPINfXdn5dCODmRbLerd7bPMdXB/X3e3w13OIoRAwjCQaTbBfcy5gYhfLlOu+o2VCixJdd8WHyuII5yfNEuBBVt17ALXRacRZcD76ews9olGMRqJwJYSk8p7EDIMp/wLAE7v7cXpvb24dnoajyk/oc4OFjNNVC0LlubqlaDY/rVTU7hy1SoAFA8v2T6GODVPJ8HNJ4LdvaysqOumvuzJLo8YBoJCoGxZaKnWuFJK4l3XuuXpcnRnJ45Wu+xZyWTb7O/ZRgOWbZPSZBiwbdtxb2caDVwzOYkvLV2KoXAYD+fz+NX8/IJKAM6zuD2Vcv7mZ4Dj53OKWURvc8u7DystD2uNzIV0GeEgpWsGquMNEK3u9TMz+JvlywFQb4Bfzc/jxUrFkzPB9zQeCCDVaqFWLuOH09O4fGgIF/T14YF8Hn+5dKmT7Fm2LNyaSmGd6lVgSwk83I8zliSRTrvZ5HyKqanFLfQTTqCfW26hYzljnUXfZPXNVLfC+DjWZYQgC3j9ehr7Ax8gasvFRKjYaaNBrttEggCec0L53Px3JuNaglLS9370Izr+F79wFRohCOwTCQoHnHUWKRi7u078nZJslta/q4hJpqcph+KHP6RqCL3lsLZNeMT/DPmfL79CoL+nvwbou6USKXbJ5O5lU9uR7FJAF0IEQWD+YynlTertWXalCyGGAXAu6CQAPa1mVL23WyQgBGKGgRYocQiAw8PNlvqvUiks0ep0ASoX+uOREfzLtm1EYKLe94NM2bJQs22cn0yic2AAQcPAlVu3YqbRQM220a1oRdlaChgGoqZJIQD1tFkgd/OCXVB4iUz497PFIr49OQlLSiwJh/Hny5Z5rDGAwgemEG6zFkEx7csHB/Fj1dxEKsATUAQv2vfvy2bxcD7v8U4INU7YMBz3P4vj9dBDCIaBIF93BnelwByTSOD8/n58Z3oar1UqTo9zwWDnk5Zto2zbnnVWFLWoBIUhQoaB705OIs1KmCQmNQAEvgDmGw1cMzWFv1q2DGVuuAPX+6BusrvDgyhaWTpME6d2d+Mnc3Pg5DMhKAlS957YGrg791K7hwCo375tO/kUflXxrkwGT5VKnrH4HsQMAx9V9zKvQhdXbduGvlAI/7hyJeV9KOaP+7JZPKxofp0cCmnjrru8xCwAud1/8APgH/+R3NC27TZ2CQZdq/X97/fOVW+04t+kdfE93s57fNsffJCUidNOW/hdlhUrgK99jZSAu+8GvvhFyrxnJYPH57/vvNNtPsLnrdfJatePNwwKAXzhC7Tx33ILHXf++aRA+NMRsllvEt/O8IS/G6VaJY/LT39KiYcnn0y5DIuVLb4ZyWSAq6+m/Aj/v7i/P78f3BexH5x7aZrtQzzt5LHHKIv+9NPpGd4Oc/FulV2Z5S4A/ADAS1LKf9M++jWAjwP4Z/X7Fu39PxVC/BSUDJffHfFzlmO7unB4IoHbMxncxR0sDMMBSt0S9ktTkrvb4mN8T1bDtvGT2VmsVwQtFyaTOLW3F3+xdCmqloV/n5jAxwYHnYYgFcvCykgE/6o2XACAELg1lcK92ayzqbfzevJcp+p1fH9mhog6pMR0o4FvTUzgf4yOAkIgoAD3g/39KLVaeFqFHABSbgKGgZBhQHLTGXVOdoezmEIrddP+40whcFFfHw7zBRJ10hG2XEeCQazp7MSvmK5JkwcLBXQHg/jM8DCklPjKtm2Y4Ox8LUkMoP7ujxcKeKJYxOdGRog6FNST/3GVVn2OCknoDGgS5DVgDneuKLClxCP5PO7OZp2Mcr4XDJhQxxpC4NPDw9hPmS1T9Tr+bWKCaF0BT+dBGwTueg5BO9eTx9oW3oRBPS9CVzJ4XjzeZQMDWBmNkgdHKYNsfdcsC08XCvjx7Kyj2PA6nbUFJCxI6HSqPA19g33qKeB736NNc6+9gD/+Y/dYpkEFqAXntm3ecYLBhUDqt6wAbx9501x8M9bFNInedOlSAl2dipWFlYt2Vp9fsdA9BuvWUZMcPm79euDMMynxj6XVAv7lXyj+LyVZ9X/5l78/Hdd0ufVW6s5n2+SWvu8+Cnv4yW3erFSrLuWs/94vpvT57yPg3iNdYVzsXjLIt7P8pSQFsKNj93HR70h2pYV+LICPAnheCPGMeu+vQUD+cyHEJwFsBcCRidtBJWubQGVrn9iFc9uhGIISrEy4m6/ugmbxu9In63VcMzWFpuqpzW5VJ5FJStw0P4+ni0VYKrv95lQK3cEgjursRMw08cWxMfQqFfDVchlXT04iGQzib5ct82zcQWU9W/rmra8B3rainKQlBPVV31Kr4a82b4aQEod1dOCPhoYQEAKXDQ6iYtt4uVyGBHDpwAAOSSTQFQjg3xWDGIMG07ryHE7p6UGm2cSDuRzF0RVb2tm9vThBy4gHKD77H5OTRGGqxoJhwDAMREzTdZ9Ltx69KSVuS6fRHQzimK4u/MnICK7atg3pVgtCUJjicyMjeL1Ww39MTDilgH+1eTMOicfRHQzi8XweLTXf36TTuCOTwXm9vSjbNmb1dr900517yMDIAMy9CRYLJQQNAyG13qhpoqW8AjyW3uzHgJeC1/OM6UoTn5dLJIXASDiMz2iIIP27lfrbUIrZ1ePjyKrrxcdnLAt//frrEFKiCVJonFAHW+dCQByehagEgWd6IKWbYBaPk7s7maT48rXXuhzer7xC5CU8lWOPpSQj2/a23xSCwPyjHyWXdrnsegJMc2F3L1YibJuywE84YcEtaCvr1wP/9V9et6t/XB0A9PnpooOMbdN6m033OO4zrwNEq+UeB5Br/uqrgb/7u93Ppf1GhdehP247yqXYGeFn6lvfIjc30N5Fzu+zaPq1Mxe/4uX/12DR76WuRDLA68/EzTeTJ+Koo976Wt9u2ZVZ7g8Bi3a0PLXN8RLA53bVfN6M3J/LYW0+D6HKtvguO250LLSkloRCuHRgAD+YnkYLC13St6fTWJfPEwCqcZtSepLmGMyn63V8e2oKJctCxbLw9YkJTzbz6b29SLdaeKpQcMCPz2UIgY8ODGCvWAwj4TBmtBavfJwthMMH/lSphA2bN+O47m6c39eHjw0NoaJqQ4ZCITSlxH9NT1OGOIOYGm+m0cBfvPYa9o5G8dmREZzX14eyZeHJUgnv7ezEyT096AsGEWij0tZ8SgYfcUxnJ2YbDdyby3ksfqGuV0tZj9fOzDjgxHX2yWAQW2s11JTFDylh2zbWl8swQPWUbN41lZfl1kwG5/f347ZUyqn9Ftq8GHiPVvO6J5fzgCzgPuxsdX9nagoCBOYfHxhwnh8phMNFz6Dp8KRLygdwlEj1w8AaMQxcnEziF6kUalKiLxDAZ0dGvKETIRxFSvcknNfXh8M7OjAQCuHr4+PIWxb0hjR1RZsqtB999xNSQgYlTj9LYs25tPF+85vkNv/4x6mDGbAQqG2bEplY1q0jy3RmhhLYWPhUS5eS1To9TRzkHR3EwBYIAE88QTXH+uYuBMU1t1cyx7JxI8Viq9X27T/94/IlCAapy9tNN3m72vGmPzHhbSTDn69dSy71k06iY6+6itzI+vc5yev3TfwA+nbJ9ddT9z5uBczn8FvZwOIudf/7/vf83pV2nh79HPrfjYabKf9ukz2d4rYjddtGVVF/App2op6Ac5PJBaVoQhAbGBN78NPALsuqbTsNaCClQ53qfy4zzSa+OTmJjOrbLgG8pnqHf1a1iuoIBIjbGhpIgxK/9o1G8et0GitiMQc42NXrAITtcnw3pERDKQ4AgSK3Z7WkxNe2baO2qVIu8ARISQl9L5TLjsW/byyGL46NYSAYdIhBdClbFr4+Po6q9p+hs489Uyrh8ULBbdKj3QP93CUtmKsDpNNyFXDc6c125oP6btm20W2a5F5XNed8nl5VvdAZCCBimjgzmcQhiQQeyOXQlBKn9vTg1nQar6gkNL6+FcuCbdsoWxZ+MD3tEL7o98vf8x7ab2ed0iXXqds2bstk0JBUFWCo502XU3t6cFA8jt9mMtigvCyndnfjlJ4eBA0DS8JhR3nwn1uq+8DPSjsTqLsTGOulTc00Cbxvvplqdvv7F26Sfj2uVqMmMWyt8tBC0Ourr3Yt71iMyF64Drinhxp9PPwwNQZpt6FvTyzLzbTn8/o3+mOOoQQ/TvrjY1euBC6/nLwPOghLSfPWN3kes1ajlq933unmFegu4O2B4X330XUCiDDm7LPf2Fp3h7xVcP/Zz6j+u9FYGPbwW9GLndPvnNKfxcVA2v/3Yufi5/JnP6O8kN1FYbuY7AH07chRnZ2YqNcpnqzdWQnaiB8rFHBgPI6xcBjfmJxES0pc3NeHH2uJT+zyNkGtPJ8sFj0tY6X6zG+7SgAFRV3KQGVLiZfKZfzjli0QAA6Kx2Folh1bnU8UCnimWMRco4HvTE4iEQjgU8PD+EB/P26an0dLSo+FCLgguFhYiFnTGBxZGIQNQYl0eUXN+UKlguFwGPsulhIqiRaUgZOBOBkM4tKBAUzW69S3XM3JcVXDG/pgGlH+nBPMDHVdnfVJiVXRKBKmiec0E4vvJXeSM6CoZzVloGxZuC+Xw58qReqlSgXjtRrO6+93EgzvzeU8/POspPH6Cq2WZ258zaDN0bOrGAbV2bPSpca2QVns/P125Yg9SomKmyaOUw1mVkYiiCl2krBh4IqhIaRbLfx8bs7x0ujz4vsBvu7q+qzp7HQ62333u2SdmybVYl9zDfCnf0rL4Npgrun1PaKOxe5rdgcpyQ3Nr3t6vO1XEwkCxZdf9o67bh0lxS1GyapfYp3GleeqSzJJ7vtcjizyI46ghjCjo7SJmya5XI8/nrwFvI7FwKxa9RLP8LHsWbj00oXsdI8/TmNz0iB3itvZsMI7ISefTF6Jl1+m1+99L3DQQTv+3ssvUyIdX4PVq4GLLqLPikVSjvTOe/68X36fRb+e/kQ5zbnnGcuvcPJxbKH7trkFVnylQkrHu032APp2JBkMYkBZPzrtqICKOzYauGF2FhEhsK1Wgw3gxrk5TCv2NAi3ZM0AcZwXWy0PaEoAazo6sI8P+ISyvvRNX4AS6mYUsUjZslyXvmbxM9+1LSXmm02km038aHYWZ/b00JjsTlYWow5Ez5XLWJrP4xhVc/Pj2Vlsq1aJd1ytu52LmRUC3TrNatSp7YSBg69HVyCADyqQXJvLuePyudT1P6qrC/srzwh/X1cKAGAsHMbJPT3OOKPhMM7v70fEMFC3bbyi2vryveQ9gBUdfX2QEnONBjZWKrg1naZafZt6xp/R24s702lsqdUcljnHXa9eA4AwDMRNE2sSCTxYKDjXm8+xMhJB0DCwkc1HzXujK2wBn8lb0Ohr/TIaiSzaO/+gjg5kGw3cpJ5N/RnwXHMl7H4fDoUcj8vcHFmlvCHOzdGGuWQJJQ3de693Y9Q3Sr8VxU1B/OBeLlO8+1Ofoo30ueeA228noNPHLha9XOWLydgYAdHatV7Q4PMB5NbfuJES2j74QfIOdHbSXK6/3k2ke+45Ol5vRKKvicfVG+OwQiEl9b1vR+uqr0df3913U5z93dL+dWiI5n/44bSe/fbbuVK9apWy1gFaX7lMysrpp3tBWb9m+vVpZ6H7k9n894CfMf/zx/eNlbx291IX/bu33ELK3/ZaHr/TsgfQdyAHx+MIKuB5plTCeL0O7mcupMSM8r3ZIIBktzSEIksxDPQHgzi+uxuTKhtbf0oOisdxRjK5wC2dCARwYnc37slknGx5Q8VFeWydEx3q/GxxctmZqay68VrNExcW0GhJNYs3p3i1AeDXqRSeKRZRUAqC51yLgTrPT81nMfF8X/3dsG08nMuhrACU5wo9TABgeSTihANO6+nBLQpkQ0LglJ4eRAwDAcPA6b296A4EIEB95llpOquvD/tWq3i6WKR7onIZfpfJoKySGZ08AXUvC6q2e0IpU5ASD+ZyiBsGphsNuhcaeHNtvT7/kBA4KJHAQyrnQb+XyVAIEcPAK4qYhtu8ClWiBlA/em50xPNqSklzehMSDwRwbl8fUo0GHsznYYG63HEPdwFgY6WCV6tVSBVGWRhicjfORgP4+c8pnn7ggeQy9h+jTX3BOPyjb7zhMLD33rTJbtpEmdXp9MJkpf32c2P4AFl5P/4xWdlSEhD+0R8RcBxwAAG6HxR4nuk0ZaEzO1pnJ63tJz+hDnlSkiIzPe393mKWnQ5Q/C8RCpFicdJJi9O6+sMV6TT9vJtk5cr2BDaLyeSk69UAXEB/6CG6v1NTC93dumLUzqPCr3UFrZ0C4Fcu2ykG+ueLHcdjzszQ3N9NsgfQdyDLo1EsV/6w6UaDWmQCtOFLlxLTcQ1rYN+hyq+GQiEc1tGBKQZ0Pg7AXtGoh6+aJWQYODOZRNGy8Hg+D4sTpRSo6xYrALCLmJ9Aob0fEgLv6exE2L9zCK3EzG+VgmL2ZQ4M8rG+8Z3f2lhsUb5areKH09NY09npAQJLStySSqGuriGfuyYlXqxUPIqL1M6zIhrF8kjE4V8HgBN7elCVFMMPGwZO6+lBwDAw22jg9nTaUQT0jn77xmLYNxbD1lrNzQuQEttqNRycSODlahW1Vsu5l5ASdSkxru6fAHkEcs2m854hvfXeB3d04NVKBWUNjGu2jbtVCWQ7etjV8Ther1bd50S71hKAaRg4NB7H8+UyWupccdPEe9qQu+yMhJTSs61Ww0Oqe54lqb3uWDiM9/X2YkU0ilG1a62Ox7FUu/b+DdSyKKNde2Qc0YGPv9duQ+X3GPzCYYofC0Egq3OxA27Lz732clnIpKSY/tNPk2vUMCipjWP+xeLiJUf6Rj45SQlsK1aQ6/3ZZ73tRfXjdfe/fz3+cXmMY45ZWKOuf9fvQXgjMj1NRDR83Zcv3zFT3TshkQiFUbZu9a4rk6FQiv58GIZb5QC0B2Td3e4X/Rj9ueJUGv2e+Z/PdvdOH5fHufNOUkb42EQCuPji3VenvgfQ34Dsp0Ag1Ww67ndd+B3+LG6a+KDWE3t5NIrRcJjqpoXAWDjsbJD3ZrOYVUGZgBA4q7cXiUAA5/f1oWLbeL5Y9JQ6sQyFw1gWiWBTpYK0KifTQQcgN/IpPT1OAprenYwBCNCUBN+uxO/r3dv471gggINjMRiGgVSjgddrNWf9U40GphsNZJtNPF8qwQCxfSWDQTxTLDqNYxb8L9o2pKoZ5zksCYdxWm8vDtbq2KWU+G0mQ2V0ynplyTSbeLxQAEAu+S21GhKmiTWqw8lTxSJ1bdP+c4Vh4ISuLvQGg8g2m3iuXKa2twDiBjHMvVqtehMY1Xc9tKpS4pjOTgyGw0g1Gni+VEJDJUNurFYd1zzPjeWgRAJly8IryipOaw1s+D4e39WFlysVtCRlvB/X3Y1T32pnElbQhEDJsvBEoYB8LIb39fY6ys9ioj1KzutslmLafheo/tt3+gVj+l2tMzPAk08uPG5oiLLiddKTe+6hpDmNLBGNhsvFrisM7Sxqf7LanXdSAh6XoPnX004pMQwKPYyOkpLDRDJCULb+6tWUvb+YjI3RuvQa/XbXqp1kMuRdYPYxIehavBsAPZmkEMOzz3qvlZ6v6vfm7EhJ0r+zs8qP/7ll2Vl3O3/20kve84fDlBfx/vfvnjr1PYD+BuSYri6ULAv3ZrPIK/BcEYkgaprYWKlQCZQGumXLwrpcDseqxKQD43FsLJcx0WjAlhIHxuPYNxbD44UC7s5kkFI10EHDQMOy0BEMQkgiWFkRjRKdqxBOHFtIiRXRKC4bHMQ92SzuzmSIMlS3bIXA3rEYwoYBQwjsHY3iZabKEm5egAew1W8ppZMJz+56qf4Dg0Jg31gM/cEgLh0chCEE7stmHUDX3dWbqlW8WqsBto2CZeFc1XrVoWWFq0zoNdq6x2FlJOIBcwBYl8/jnnQaBVXfHTYMtGwbUcMg5Ua6/PWpRgPPlEoOoBdbLdRt21k/A3HQMHBRfz9+k0rh+XIZUkpETBNHdnZiVSRCsXd1LXqCQSyLRinurdbD13B9sYjBcBjn9vbi5UoFDSmd66I/I3wNZut1/CaVggDQFwxiixa64br3FZEIni2VnJ7+hhDoUvX67WRDuYyttZpHATm6owN9WlZ81bLwcKHgNkHyjfFateqsb99YzNOzf599CKiYDAUg0Pvtb8k61m7fAgvJb82q0y/4zX/X625imW5VrVpFSWUsjz1GgK67QvXz83fZWm8HFvx7dJRKzm69lSx9P9Do1rZuWfL7e+9N1tqdd7o88Ny85KKL3O557WTvvQn0t21bOMcdSSbjksLwWlIpogPdmaS1XS29vbQ2vh/z897mMfozw1Y6S7tnxu/pWewY/XnSn8F2yoN+P/1j8n30J9gBlOy5di3lEuwOBWoPoL9BOa23FyXLwmYFXB/o78eScBjfm5oiAFAAKKVE1bbxeq2GY0E0lRsrFUwpMBcAxut1rM1mcV8uh7QWp27YNu7P5ZwnrcM0cVg8jtdV4h279nUmtFN7elBstfBgLoeyNgcDwIV9fQgbBp7O5zESDmNjpUJMXb6dtScQQF8wiAFmd4Ni9FKf29rTHlGZ8zFfn0eHlU6tUf+OFAJPFAo4TLnf9dIt5291rE4GA1Cv9bXZLPaOxbBE+SnvUYoVt0Kt2TZ19VNr0sMhUp8LgJN6evBSuYyMlkAoJfGOv1qp4O5s1ikvDAiB/mAQk40GVkajTsJjfzCIhm0jpXIOGGhtKfFIoQBTCOSaTTRVfbyuQBlCqyiQEtvqdYyrWLh+X211L00h8L7ublw7O4uaeq/UauG+XA6n9PQsAPXXq1XckU5js4rJSzW/XLOJkXAYR3V2Im6aqNk2tqr8CmhzLLRaWJvN4oVyGS8qNHs9HseF6nkHiKyiWCTLt9EgyzOZJGuWN2a/pa23jB0YIOBlytN2myxbbsuWUYb3DTe4a5SSgGDbNrJmAdpMM5n2SoNew9wOAKSk+ff00N8rVtDYrLD4E/d0a01/7QeYM8/Em5L+flJY9DH7+nbuuww4PI9ikUII7wZAX7YM+JM/ob+npyk8MjNDr3WQ1F/7AVW3PVhJ0terg7T+m58B/3h+y7vd/fQrmu2UAynpeb7nHmp2tLMeg7dL9gD6m5AL+/sXvHf50BD+Y2ICm1QCEQDEhMDhyqp8vlzGf8/POxnZEsBzpRKeUz23+T1OrtMBN99qUba22tjZO8VjsRyUSODZUgmlet0BKAarh3I5/HJ+Hg2V9CVAYM3KQSIQwHFdXR7iEqk+s+Ba0yw8Z134eIAsSk7g4rVBrcsQLoMaf8cP5oDrahag5KyXKxUc2dGBQxIJ7KXal+rgB20MqHE5kz5mmk7FAgBsrdWcxirc4tWWEr9TgTzHigZVJ/xsZviRqH8AACAASURBVAYD4TA+PjyMfWIxzDUa+OXcHH7O7G/aWvjvhpS4L5fDQDCIOU1hM1jh4oZF6npJvl58bfV7SAfRa9uluOX7oO8bM8raf035nG1t3PtyORhSomRZOL23Fz3BID4+NIRvT05iotGAYduwDQPTzSZ+OjvrhEQkgP/L3nuH21KV9+OfNbN7Ob3efmnSy6VJFxEuiKFE0ViQ2JP4NcYYExPzjSX5RdOsMQmJGokKkSKCiICCiKAUkXoFBG4595x7yj5tn73PbrNn1u+Pd70z78yZcwtcLvp97vs8+9l7z55Zs8rs9Xn7++TiItZnMj6gWxY5wFWrpGbPZMgxbceOQKqSbhtyY9SaVK9PP00e5bzRRjdcxyHJ8oQTSEXd1RUkZtGarr31VirM0t8fABkQZByzLKrI5nnkic/9YA99pmKRyp6efTY5aH3zm4GT1nL9Y42DBBRuOyop7imdeOILy0aWThPwm6q8UIoYnhfKWMTR5CQxcS/Gw3tujuz8jz8ez8jtjmStNc09M4pRJm65NmRbTLyW8l0+r0zRvvExZvZkRb99TfsBfS9R3rbxqu5ubG000DRPSqndxrWlEj5RKFBcsgFHH7wQAEEI1MXT49vMVRDCZpk2XB3YjcebTXx7cpI8nhmElYLreXi8UsEts7MUb2zasM3TyLbfFakU1mYy+FW16m/iGYsKtbS8oIqZB3poekQlOCbfjq3ii8PAjG2k0aDseGZ8HuDHUyulULRtVIzpQJmx8j/rgUoFDyws4PKhIXQlk775QgI5S/2WOaYArEmn8buGEZsypUW3sD1c61BOetagMHMFpeAphQnHwTWTk7isvx8PVCp4rFoNnPfE9ZJZUaC6APdXKqh7VMGtrTXKnHFPzF2w6EF7WlMN9J5EAgnDdLR5XMrUC4jQfeUynpA6YjPX3DcPFMFwbKGAVZkMhtJpvGN4GF8eG8OsicnSYu51ZC23NxqU0GciA920sXFjoH7/+MeXpnOVUo+UnEqlQJXNoBndPMtlilk+9lhS0557LnmbS8e6Rx+lOO4TTySVp0xIojWB+Yc+RIzHv/87ecDbNr3L8Le1a6neOUA2aGZM5OYdlbii0mIUTF4OWr2ayrx+7WsBKBmr34umSoXSsf7856Rdectbdl9rIGlxkUwzUpsDLAXKaFldqfWJzn0U6IHwuZLJWk6DIxm8uKx0cc9qlFGQDN1+Cf23lH61uIirxsfJ8cwAmgJVDGNnN18CM0+S/L/70p2QFhXgA6mUdFkNLhPD3Dg97dvYmXmAUmhqjWtLpXCiENMHltChNZ6u1fwYaK7SdX5vL6ZaLcw6TihzWt628YFVq5CPeM1LCZ1uRH3P2jaSpl8L7Ta+Mz0dPt889QXbxop0Gud0d+Obk5Oou244DtuM2wNVBPuDFSvw8S1bMMEZHsw5GiLEy8xR03Ux4zhIKoUbSiXyI+DzjSTsCaaC65xroylhrcFos4nPj46G5lHuCpIpgKKwsptnZvB369ejN5XC9mYTv67V8L9TU745wPd4F+KAPy8muuGK4WGsTqdRTCSgQZn2iraNDttGy/NQdl0/Y1zWtpFRCnUzJzoyTr8+uli71ZkMNvb04JrJSXq+zBzy2vA4y+02rhofx9ZmE+53V0A/3YFEAnj3u8mTWkqzvPnyYyJBVilyXJMbY9QxSp7LlE4TeHPJU6af/Yy8jfkazyOP42yW1Ls9PfT69KeDa+6+myqn8TWPP05mgMsui9+oC4Ug1p3vbVl0D84cxy92jnq56JBDgM98Zu+3+9BDxGAx3Xgj+TDsSaBFq0Uq6bvuCq8zS7hy7uOkY0mS0Yp7l+dEn82oB728RxyTFmXW4p4ROYaXg/YD+l6gmuviv8bHUWevbbOaGsBks4mv7NiBw+S/O7LJqpiniqVxHyAVqdvTloWWsckqUC70Ghf9MOczSOiYtrjIBhedkUDP53AI1h0mBv69w8M4IeLBY8l/gKFQO6Z9aI1zOjvxOiMd/83mzRhrtcD2fXYiK1oW/mz1aqzMZGAphZrr4rpSicZq+sQ2aqUUbp2ZCaU89Z3MzDk0ZQHDs6XRwLcmJtCZSOAxzvzH5wrNhwcTAy7ZcrkOCNTj/HvUuU9HzuHselvrdXx62za0dVA+1T9ftiXmkkF/3nEwmEzig6tWYdJxcGOphD9ZtQoNz8O95TLuL5fxvpUroQGckM9jotnE/Zw6l/uvVDCPkTFtrtdx0/Q0hcfx2phzlJlfrSnhjxb9U4ok3auuouIs2Ww4g5bkU5ikVzNPM2+iUkqSUtTsLIHtqafS55tuCtqWbfF9CgXg/e8nm/jOQtTE3wsAMQqlEo1JgoNtA+96FzEAnLOdpfqNG0k9z+kAEgkqQHPhhfH33ds0MxPM1QuRlneXHGdp8p6HH6Y55kxvcTQ9HfTL84jxuuWWnYNenLZDSvLyPEm7kooFz7wkkZE8J+5eO2s/+ozta8mcaT+g7yUKVQUTT6MGhUxtbzQCkDVx1hrwJeqQVMa/AT7Ye1pjOJnE6V1duG5qyt9wH6lWUSyVCKCNTVZKyhIwpJQtNQC2OAeAL6F55jhnEovSorFBQ7StEIAQg1wbxPTkbDvUNx98lULZdfHvO3bgb02WijO6ulBxXdw+O0uFWPimKqh4Bk1e6YqPm++OWA9/vCCvb8vMt5yr6PxLD31pS+d2mBHi9ZPgLdtSlgUNCkPkczWCpDEugrS/flsINDE8Xs/z8N/j4/6cFWwbH1i5EhOOg8+OjPiV1z5uqqWxBG4rBdv0AVr7nvYJy6KqfWJND87lcMXQEP7LpDD2+xJZS+4DmYV0SELp7ydQ/+d/DnK08++s4o7bqD2PQLDVCqs3+a80PU1FYP7yL0nqTSbpfOn4JSUwpSiJTF8fSamtFvDhD1P/ZF6cOEbg/vvJU54rxXFfUinyXP7AB4B/+ieSyPv6qAZ6sUgq5BtuoLaOO44ArtUKqpEpRX3f26FMMzOkdZifp7F95CPknf9S0MMPU255CWo7k0YXF8lO/m//RuVze3vJ2/+aa5YyeHLtuF0pjUf4a/9YVCqPA2zZlmTgoveTzGU0Fl0mBYqzj0cZQ88jc1IkMOclp/2AvpfIMps/A4vMh+2xypYlRmGj5CeBr9eRJ4NrfPP1aSHNs9R07/w8zuruxoGZDDbX6wS+tu2XZ+XN2M+BHqMJkKAVTXMbx25uazTwTyMjcDwqZLI+k8GxhYIvBTKoWUrhTuPJ/xYRkw9NVcUktSJ9OK+nB6d2duLm6Wn83CQ+Yf8CC5Ro5Q+Hh/HPo6OYdRzkTOrYayYn4SAc4w0zfmZk+Lsy53nMkCjlq8uhVEiSlnPEQM8e72ym8BkoM7akZeF9K1ZgIJVC1YR/ydA1nl8FQHme74Ng87xrMgX4/hKG+blyfByX9fdTfXXTFhef4bz0J3d04KL+ft9k8eWxMWxvNvGOoSEcms+jYNtwPM8vybuhWMTvDQ3hmslJf1y8I8oysYFWJ/xcKEUSukyfadvA+vWUTezOO5c8Rv4m/MY3kk21VIqPSd6xg0D9wx8mp7XZWVKZxzEAAAH+5z9PzIBSBHrd3WTj58g7lrzlxh7NVmzb1NYVV9A4lAI+9angelZcnXkmATlA6nbLImn+4YeD/r3tbcDJJy+dgxdCXNjmc5+jOQMI1L/wBeBjH9t7NnNJJ5xA97r55mC+oqlSmVot4O/+jhiOZpM+9/SQOUM6sEkwZoqCozweZSSiUnb0M7cjts3Qb9xOVCKPnh/VKEUZ0+ix6Wngi1+kssH7kvYD+l6gnG3jD1aswBdHR7HouqGEKAwKAHwVM2+IPqgjkJjZVsne59wGS9OwLF9q56Idrta4d2EBrudhY08PTurogKM1vjQ2hhlpXzZ94vtzbPO6bBY/npsL1U5nWk4rprRG0/Mo8YrWeL7RQMaycHZ3N348Nxcad9uAe5vHYPrPKm7fjhy5R8qy0GNZyLBGQ7TJqvGvTU5i3nGglELd83C9KT7D5ySMDZoZJmaiQuVKI+w6A1fesvCBVatQtG08ZezefH8FAu3jCgVc2NuLa6em8HStFkpuo0GMSTeH9on2ZXEW/zcVxKef09WFU7u68J87dmCkXg8xAFpTjv6vT0z4YYxcLlX5zSk8uriIFZkMzjeRC3+wciWanofhdBo118XfbNkCDeAv1qxBdzIJSymc3tmJg7NZsO/GXXNzuG9hIZSsSIsu66DrAMgh62/+JryOjz4aTvcJhKUnpUiSicuMxpuk6wKbN1NBmPe/n0A5ToKSElqjERxvNsMlXAHK1FYqUY70KLDw50svBY45hhLY8G8mM26IEonw8RtuoEQ48u+3Nz2fv/QlknaND6tP09PAZz8L/PVf71452T2hRIK0DPJ+J5xAOdjjqNEI6sO3WrQGEhDli48xSAPhELaoR3p0neLSvsr15DYkRfsS9zkq+Uevj3v++N4vMCPzi6L9gL6X6BX5PD60ejXY6WhbvY5rSqVAdYtAXSsdqXiDvKivD4fl84GaE8CU4+Cq8XE0QYVi3jU0hK3NZuBBrQJJmpOkdCWTWJnJoO15oVAwmPNDpEn9emFvL+quiweMbdmvagbgwt5eHBnJ3z3VauHLRjWrAV9yfrpeDzzHxX25LQvAu4aH8cWxMcwZ73QgUMHHaQLk9axC9wEF8L3h+TdOVWtF/tX+3PPcATjVVNPbanZ6Sym8eWAAazgsSykcZHwfSkZ3qmUfNTn8rTSFVUL/d3Pfpuvin0ZGUEwm8UZTW5QT9sBco03f5PiLiQRWptO+pKyAwNnR9L9m1pjv52erM+0ckc/jlcL3oT+Vgqs1PrNtG6YdB/MmlO7zo6P48zVrkLdtJC0rVNAlb8wkcu54/rTZIBMJkj655DunYGV6+ukwmErQlUsfBWR5PkASqbThRs8DKDTrqKMoS5u8h9x8f/UrKviiVKASX456epaOh+n558lrHiANBMdWA+FqXNwHLrl5xBHL32936MorKVyPcw/xPXiM1eqLa39ndPLJQcw/QGr0yPYQojhHRwnc0TWX6xRdPz4WBX55vrxXHAhHz4+T2Pm6OE942Sd5PM5883KEru0H9L1IBwnHN09rJMwmaAnJi1WW/NlTCq8ytaqzloUvjI5ixsT9uObVk0jg94eGsCabRWcyiY6VK/FIpYIHKpUApKXUCZJMLx8cxNd27PCLuDCosnRuKYWxRgOfGx3FYrvt29mPLxZxgnFbPTCbRca20fQ8fGF0FPPtNjytMeM4vtTGkv1wMonVmQzuN7HZkoFgIFqfzeK9w8NouC4qrov/nZpC3fPQkUjgbYODseVAffMFayYQgEoiIrUrca5v+hDzzwzTcYUCXtfXh4V228/6BwCH5XLILFMtQ7allMIhuRxeY8Syi/r6cHpnJ26ansbhuRzWZbO4ZmoK860WFtptVFwX35yYgKc12c1N5j4p+fJzobXGj+fm0J9KQYHs7OB1E+OPriX3i9vqSibRFZNUesF1Sath7rej2cS/jY3hg6tWhXLeA8CZXV04MJvFT+bn8VStFmg4ANivLsFqJXHpSVmcfPLy9mFWy8Ztyvz9O9+hEDK5SfOmLTdw/v300ynO+rHHwuc/9BB5eH/1q0HOdr7WtimH+De+QeppbjtuuXkzX25MY2MUFjY3R+c89RQ5xr3tbeExy7bq9XBI3wulcjls55UV7woFCiXb29I5U6FAWex2h9hcwWtTqwUe8rK6Wdw8S/CX2QVlvDdfz20st14yP4DMU8Akt5zosxr9PWp3jzIi8llNvAzouh/Q9wI1jJd7ud1GwbbxnhUraONnda85b10mg4Jt41e1mi8JK1Bpz4Jt47/Gx/FcvQ5HVu5SCg3Pw8/LZRyZz6MzkcCxxSLGTIlWXwqPspoAji4U8FZT9/qW6emQjVqBpK4WKAsbGPy0xtZGg+KRlYI2TmkawFizSdnTAL/eOLdlgap3DaVS/rgknSRKnvK7q40Tm+cha9s4ehlW/9SODow3m6G0q6d0dOCgbNa/t884AUF9cxWuF86fLKXQl0yi17yunpzESKOBNw4MLAvmPGYlXp2JhF9YZ00mgzVGUl+RSqFo27h+airEiMyYErRcyEeB7N3SlADz+4LrompqqLMJIFRKN0bb4ucpUAoHZbM4dZncohaCuH/WfJQiouqjlQrumJ31GYQ5x/H76JsuOttQaRcHH7zzYhRHHUXSLKeDtSyKKeeY7zvuoIxhUmrj86QkZtsE4v/4j/R9eJjA+9lng420XAauu45Uz3Lzz2aBiy+mzwzC0Y07NEfmnt//Pjm/RSuKOQ7Zh7mddptMCzt20HiZ4XjuuaCt007bs8pkyxHfk+eJH9l0mhLsbNjw4u/xYimZpPm+7jpyjsvnafx3370UhIGlgAgEQK51kKY3et2uAFhey9tqHPPAxyQjFqdGZ5+LOEc++b4zZvClpP2A/iLJ1RpXT07i6WoVLQNQ/z46CsewbvxsKaXQmUjgvJ4eNLTGZuMgdXShgEOMl85ovQ7X8/wKXxwD7VoWnlxcxLVTU3jT4CCeqFbxcKWydGOP6JeUUjipsxMtz0NCKcw4Du6en/eZCd8eLJ5apRTmHIcAHbThe8ZZy7/GAL/vGGaOT7Za6EskcHg+j2dqNRyUyfhV1o4sFNAT2fVtpfy66zuj1ZkMLujtRaNUwvZmE0fn8zivtxc9qRTO6+nBd0slVFw3AHTzj/PnHgDHczMDZCmF22Zm8Gy9jpFGA5V2G9dOTeHtQ0N+JjRJazMZnNzRgQdNkeqV6TROiQHMI8x4254HWylYkbVkZzk5l/46qCC73lG5HDY3Gphtt0MV8XyTjWmr08z3Q6YQDTMb/UZbsoQMg+nXPzdt1lwX10xN4fKhITxbr+N7MzOYaDaXJOjhe7AWYDlAlNTXR6FdUlV78MHA+vX0+b77wjZ1puiGD5DqfutW+h6NCWfiuuy8oabTpIo/7TQCWCAMAnFjYGCZnFyqwp6bI40C/234Va+TBqC/H7jggsCrm8c1MkLOfDvL4b47JKVazyPwfNWr6H6nnfbyAEmUlKIQw3ab5iWbJe/7e+5ZqnmJW2cJ7MDSuZYSfBTg41Tp/FtUjS77EHd/eT95D2Y0JGO13Fj2Je0H9BdJnqYyoQ5oQ3a19mtaA2GJTgE4IJvFxu5uPJvJAFrj2GIRQwZApLqUN28GpqRSGEil8Hy9jltnZlBqtXwwD22wMX1MWRbO7enxS4xy0hQFYKzVwtOLiwFQmH6EiqOosEc2dLj4Cr9X2238ul7H4bkczunujq2fvSva3mjgjrk5rEqnsVFUEXtFLoeNPT3Y2mhgQ7Hox6Cf0dWFuudhptXCzxYW4EQlWQimx7yvTadxeD6Pu+fm8LSo4rG1XscNU1N469BQiPmYbLXwvelpzLfbsBTlcD+vp8fXNMQR7yOhObNEuKLW6EmlcLyJa9ncaGBLowELZOYoJhJ4cnERDeNk6WcY5LUwu3bb8zBrjLVy7TfX6/ja+DiON6lymSylcFpnJ26fnaW8CWYtHc/Dc0YDMuM4GBdlYdkr30KQMAcAlAagd2/XWr1652lC5QabTJL0/thj4VKs0c1ybi74HCV5LJEgBzhWgUY3fXl+9HicwqZeD4A6ei/u44MPErjLc0ZHSdX/YumVrwTWraMQu4MPJobp3HP3fYjUrsi2KSKBqVwGjj+ezCJMck3lsShjF6WoA2V0/aLgKsF3Oala9ieuX7I/fP+45DR87fw8pba94IL4dl4K2g/oL5KiUgt7HfMxVj/3JJM40vzjjiwUcGShgPvLZTywsIAHFhZwhokziSaU4XYLto1XdXXhvnIZIyYtFYNw1Kt+OSomEqE89CONBkZKpUCli8A+6/8rDJAfXijg+XqdCoOA0sIelMlgE4dimbHOtduoex7eMjS0x3NZarVw8/Q0NlWr2JJKoWDbOE1I8McVizgukpLKUgrndHfjxlLJnzNPKUptK+ZCaiFcUDU0P8OckXgB4OlaDZV2OwToc46DXwpx0AN8T/pHKxU/w94JxaJfjez2mRnU2TFRMFBs+y4mEnhVVxfWZ7N4aGHBdxBkRubOuTnf94HHydf6awSg7nl4rtEIhb9Ba0w5DqYcBzOOg5xl4WDj36EUhQNWXBf3lMtk3kHwHO9oNvELo4WAaUsm9PGZEwAaasnjtrBAmxj7Kg4OAuecE13p5Yk33tNPJ+c1x4kHVXm+GXqIoqDA1NsLHHYY2byj4C2vk5LZfffRODjdab0ehLdFN3++5+bN4QQ0y/XzhdCpp9J7Ok3pbmU06G8ydXYSYyUBXdKupGVgKWDHXSevjV7Dv4ntLZYh4+ukBB7tC7DUMU/2pVoFNm3aD+i/laRhQFhRda7D8nnUPA9b6nUUbRund3WFwOmJahV3zs1hh7FfL7TbWJVKYdZx0BTAxLZgHbmXDIGLq83+k7k5X6I8rbMT9y8s+IVZAPKaVwCe5VKqSwYU/AsspXB2dzd6k0mqbgZiME4oFvGUKUYjK6ZNtlrYtLjoq593hxbabXx/ZgZPVqtQloVZx8Fds7Mo2DYyluX384h8HuuzWfy8XMa048DTGi3Pw73z86QlAQJzAMhLuzeZxKipjteXTKLpurjXOO75/0CLKra5mrKh9S4u4jU9PWh5Hh5cWPDXVmmNmVYLjy0uoieZxI/m5rDVePZPt1pYk8lAa42pVivIA4Bwut6MZeGUzk4cns/jxlIJT9VqvoYlZX7nNWDw9kFV9NnX6IidR0NoJEBJjbbU6z6gA+Qw+bq+Pjy4sICWOTdlWTg8l8OE4+DpWi0Uc8/35D7ArPX6bAYdBybQ0UEe3T/6EamU77svnLmsWKRCLMvRoYfSOb/+NXmUd3bS65BDqDBL3KYbJ4FJ4ik5+GCKyWYrSn8/cMklpAWYn6e85LK9wUHyHmfp3/NIU5BIUF84V3zcveX3tWuJaZiZ2btgLmlfZaLbW7SwQJoLCc4SMCXQ8u/yfWe0nKQuKZpcJq4fUfBeLg0sXxf9Lj9HmYF9QfsBfS/QAZkMOmwbW42kdGShgHcOD2NHs4kbSyWsMXmyJf2iUsEOkwIVSuGRSgWndHQgZVmot9u+ZM+gXnNdPCxidvwCIiyVmU3W0xq/rFRwy/Q0ysapquw4+HmlQg5t5rqVmQxe090dSlcKANAa/akUukzOcIDApDuRwNsiUvfzBog0AC3znTcauH1mBjnLwnpRP1uSpzXuK5fRMEzGTLuN+8tl6o/rwrIsTLRauG16Ghnbxq+NFDzWbOKQbBY/np/HJM8f9yGyG6RsG4dks7AsCyMmxqeYSMACMGVs03LczJT9dH4eSduGBeCYYhFjrVaQIU6R+nm21cLtojwplMKmxUVsqtWgPQ/vHh7G840GasY0okT/CraNS/v6cG+5jE1G5c/g/Hy97h/jNWXQ1ma9+Te/39GdQzBjoTS1giwAB+ZyqBoGrTuZxBsGBvDE4iI8kyyI59Uy8yLnCgBe09ONDW9Iw3Upacz3vx/OjAYQoN17784B/aKLaOP85jeB17yGQs4AcqratCmcdS4qPceBKb9bFtntu7pIupZTd8ghZIvneuPcXnc32eZnZ4P2PY+kyjggl/3J56nvq1aRJPrMM0HSF4Cc+F6I/XzzZlLfA8ArXhHMz28DtdvE4LET4f33h9ctzms8bi35XfpGyN+kpCyBfDkQj8a183nR9qM2cnl9XFt8fbSdfUX7Af1FUsKy8J6VK1Ftt/HvO3YAAC4fHETagNmfSk8gQX5lMvH5Hi7FqUVGMyMZzTsOvlMq4fB8HuzMxCDjx1ibJ+wHMzOYa7cBrdHWlByEVeL+/c3LzwBGNwYAHFso4JKYErFR4uQuLBXKNKHP1et4sFJZFtAbrosbSiWKIzfEfbRVEAe/1WRnYEB7tFrFI8ZLKVQ+VQfZzNgMkbUsFBIJ/GRuzge2zUbE4ix1UgLWyhTF0RqNdhu3zc7i9K4uXNrXh89u3+5rIKCpfjnPJrfBc+iZOXDNmizJ/ifmIZTTXxwbNP4SNc4ep3VISpf+Fn6RFwTMGZfe9bTGWLOJWccJmRGyto0/igmwZo0GRzlw+/44xY7J42g0qIRpNNac33dHSrEsqhAmKZUiiXn79qXtSpIbLW/6WpPW4Pbbl2b5YhV+VJWvNan5ZX/Z6iHruEtSiqT34WGqXf57vxf81t8fpGHVmjQD7Ai4uzQ+TnnPn36a2jjxRGonIh/8RpLWVDTn2mvDMfNKhQF4uXhyrYP5l85s0iFNStHR9YnGwEfbigtfk0wF30u2we/cB8n0SVBXiu5XrVIUxu6G+r1Y2g/oe4kKiQQ+sgx4V9ptjDQa0CBV93A6LfJhBwDuIdg0GbAB+N7Qk46D0vx84OwVAaZiIoG8bfvteEr5kjN7T8O8Nz0P482mfw+/4tce6ohk6VI/85r5jYFstNEIxXoDQK3dhmtU3H6/VGADtkB2+qRlodpu+5qIUJlSniueNwHInmVhut0OmBk+jwFX/DuZMeF59Mxuwe2nLAsdto15E6sv076GNAMsRfNxUBQE543ncba0xuZ6HTnbRiGZxEK7TUVTlELetlFMJHBcNounFxdRNfO2s7W0zPgsBBodv09K4f5yGeuyWbw6LsVZhF6Ry+ENAwO4wfhWsPmi2m6jJsQdmeSGN7WogxBvmsIVYI9oYIDiuj/zmSB+WDq0aU1ScT5P6lxZOlVutABCVd9k6lHuq2QKmOI2bj4ur+3ro3Sr5TLw5JPB9WvXkqT+QsF3bo7A/Fe/Co7dfz85xL361S+szX1J9TplzJNWPZ5rWYteSrtRlXvcS0rBQNhBbnfXUh5fTl0fZ1aJtiUBP64/O3ZQqtwPf3jP5u6F0n5Af4mp6Xn44ewsfjAzAyiFI3M5vH14mLJyKao1nrVt1F03lHvdBfyqZ6xm9e3pOsiTUSG/cgAAIABJREFUzpS1KOPbiR0duHVmhhgE80Qy4PIzaQGYaLVwO4OdYSlZ+trdBEe2AaCyUZtLtXTatpE1O+gNpRKeNGpkJ8KG+yVjgZAnfVIpHF0sojuRwO1G/8kgKbULDM4K8AFPm8/UVJAMxs+3LyVO2Z7pAzMGvBYHZLN46+Ag/nVsjObGzJdrfl+uvaJtY9qy/DzvPP8lx8FVExP41AEHYLrVwg2lku/Ffnguh/WZDL41OYntJmzM8zwUEglUhZSuzfwXDQNX59BCzwuBOTRls8vsgf7vsFwO7xoeRsG20Z9KYd5x8M3JSTxerfpj4KxyTHJTlhtiIkH28Wo1KGiyJ8SbppSuMpkgO9mpp5LE+vWvk33WdcMMhJSe5OYr45njNn8+PwomcrzSaxogsL3hhuAcpcgh6pJL9mzMTPfdR6r+OFD5bSCeGznvUsPCUrYEVakil2sfBVupdYn7PQrAy724rWhcv/xd9luOST4Ty8Wm8/O/r+g3IGLx/216pFLBD2Zn/VjtTbUabpmZwZsGBrChWETetnFZfz8lSWFVppD+lNYEuELCZJBioE4phfO6u/EqI4HlbdsHMSBQZUvgYUmYAZTb8o/tBh2QzeLdK1agw7aRt20UbBv5RAI5y8IpxSIu6u3FQrsdSkPL9+W+RJPQKCNhr8tk8M6hIWQsCym+hsdiXpYYI7cVaofPQxA9IFGHf/PbFb/lLQsFCYKinzLnPffXQpB4JmtZqLkuLh8cRNG2l5wLTRqZyWYTtQj6PLiwgM9s20a17bldy8LbBwcxnMkExWK0RmcigX848EBcYeYpupasFdhoitwsR1PNJkrG47/tebhzfh6f2bYN/zM+jslmE98plfBEteqPM2NZ+L2BgVA4XD6/dPNTilTRl15Knu/XXENx3SJScJfEmyoDsG0DBx0E/MM/kOR+8cV03hVXUNIWqc7l66O8TDQGXYKMvF4CftzY5DWtFkmiOztvTymVCpz5uN8vts19TfK5AILELba9NIELEL8uUuqOG7s0ncj14nOXSzyzq3WOm+9ov4DwMxZta/Vq4D3v2fU87S3aL6HvZWKbcDGRgNYaNZMqNfQcGlC+fHAQ7sAAUpaFE4pFfGLLFkwYJyqWDgH46UKlepk374RSOKOrCxtNAQ4A+MCqVfjE5s0YN5u0Bnkxtz0vVDULkba43GZc+lU5Nh4fABxVKOAfDzxwybkJw3R8e2oKz3J+d2ZrlULKtn2tgMMOdUYC9QFQKWw08fM/mZ+HI8bvl6s13yHeQ97fiIQSMtNk/nkJwPdu57bSto2PrF2Llek00pwIxozJia6lkfqlLf/Cri7cPjuLyVaL7OhiLWH6Mt5s4uNbtwZMlArytUNqCMwYcraND61ahU9v2+aHs2Usy0/VWzE+CaEUt7wWAtGq/Dya/ky0Wvjc6ChySuGja9di0+IibpqaggbwTL2OT27d6mtueL1e29ODk4V3V6EAfPCDwN/+LYG13NQmJ4FPfjLYkJ9+muKoL76Yko2wWb9aDc6RjmPcTlTNGgfETFLyUiq4BxcKSSYDyS6RIEBw3aXOd/w5quaPbtpak3R+++1LQefFSNXnnEMOenfeSX1LJOj1m5A4BiAmptEgjUlcqtl8nsrLfuYzwXPBBVp4zqV3uVxfuQaWtdSHIbpFRZ3Q5LlyDeKAW0rpO2srrp24z2zeWa6vLyXtB/S9RC3PQ6nVwj+OjCBt2/jYmjXY2mziWt5k2dYtnoqkZYHdlNKWFdp4mTQCgOLvLKV5WuO4YhEX9/WFQDhtWQTM5t+Stiy8ZXAQ35qaQj1SzMSX3g0wvL6/H7mYwN9Ku42/27YNc46DvG3jL9as8RPiLJcu1TVgLVXZAJU9fe/wMA7O5zHZauGz27ejIYqqKASpSW2lcEl/PxZdFw+Z4jFs744zDfg53OmLP1/+mIVWQmmN83t78fjiIjazx74KnM2UWVeAMvq9cWAA3zZgB61DOQMUgrW9aXYWTa1Da+2vpXnXANomu52UqiWz5Z+rKPFLZyKBj0e8qjj3+pldXdhQLC65J0AaGwBwPA//sn17EF1hqOV5qAH42ObN/ryyJsU173IMKcOsSWJJUmbjGh6mDGbXXRdsnI5DHucPPUTZ4177WgKrz36W8qMnEsQcHHKI6Vtr6aa/HMnHkDftTAb44z+m71/8Iqk//+IvKBTtoYeoTjdAtup77gmDNVPUeSo6xVqHy67urY3ctoHXvY6YnQcfBM46iyqb/SYkkPE84MYbySxw9tkUQseMkqShIWLogKCYjeuG1zLKLEnAjTJXcdJ6nEe5XKMosC8H9Hxsd9qKPhPydzmGneVQeCloP6DvBaq5Lv5u61aU221UXReW6+Kzo6N4XW8v2lrDsoI0pHZU/2NIKYX3r1yJfx4ZoeIs5qlgYPFLbprztdawLQsp2/alSEkfXLXKT34CAEOpFNKWhf8cG0ObbhhSHR+Zz+ONAwO+5C3J1RqfHR3FlKmQ5miNL46N4a/WrEFhJxUIrjcx1uD7Ich41pFMojORoEQ14t+hAKzKZPCO4WH/WNqycNnAAM7v7cX/Tk1RZjsETIJ06PM8z2cGFEDZ2RQ5lNlmHXxveK2RMRoJSym/b47W+KeREQDEYPQmk/jrtWtxRlcXKp6HW6enoZTC0YUCEkrhl9UqPLPOnqbIAgUEeQJAWgC/X3LutfYzv0nmjU0uAADPw39OTCBtWfjwqlWhamhMtqLUwnHE4P2vY2MYaTTgiuP8fAHkrKdAz2go3axkcpaJx5GSDj/elhWvivc8ktRuu40A9pFHyJO93abjV14J/Nmf0cZ/5ZUE6kahs1O64grgy1+mcDStiTl45zspzh2gWui2TXb8M88Ejj02qBueSoWzf0nGRH6meQtLdZOTlMf93HOBH/5w533cU+L87OeeS30V6QReVrrtNgLzVgv48Y+Bn/6UfAXOPHPpuWztKRaX13wAS0Fda1ovlkGikrB8yW01GlPOEj5rBfh6SRLIl2uL7yXbks+E7BdADO273/3C5veF0n5A30tUdV1UzdPiaY3tzSb+Z2KCnJoQSHy+KjWG+pPJUJY2/5kTGypgJD3jOa0BXDs1hYcrFV8V/ebBQWyI8cToTyZ9JytbBU5wWpETWhyY++Nrt31g0MZD/l9GR/HRNWtiGQqAGB0uCKMjL/4HDCST+PO1azHSbOJbExPoSaXwnhUrluR9LyYSKCYSfkgbYMLmEJZ6OaMag6MvORuNBs8nq6ZvnZ31S8/KeV4Utu2a6+Ijzz+PtZkM3rdiBY40u2pnIgFLUfGcJ2s1eFrj/J4ePFyphMwdLM37ayvWUe5UWikcXyjgnO5uXFsqYbMxVWilsNhuYxHAl8bG8JHVq9EX0XE+uLCA60slHJ7L4feHh3HT9DR+JsIgoRQW2m0fzH2TCzNF5hgzP1H08p0XlcItMzPotG2cYHTji4vAv/wLvUclW7mxRmOCm02S8hoN2rT5eKVCMcv/8z/kOS4Zhaj0LKmnB3jXu6hW+MQEtXnttWTHHBqiDZYpmaSscQCBMCfD4SWJk+Lkdwk8rkuhSckkOenddx8dP/HEveONXij8ZkjlkhqNIEyR3znx4nLE8xVVScfNdfSZWU4yZylYPh/RqAoZ6SApKqFzv2Rb0tmTr+F7RJ+NqHbBsoJnbF/RfkDfC6SAkFMUrywniGGHOC3Oi6OEZeGKoSF8dccOVF3XByUABArmu2W+A5RxzgOBDgPut6emcN3UFBJGtc1FOlam03h9Xx++Uyr5pVaV1jggk8Hv7ML9OFTq0wB4xYRx7WpeZMiZDarS5uehVwo528aP5ubgKYVKu43/2LEDB0SkdCbb9AVa+5K1rRQu7Omh/OQsdZpzeE3WZTIYSqWouArdmOLNjV+BTKTCbct/Z9Vx8Kzn4dMjI1CgjHVvMjk33zAwgFeZjCrrslk8Ua2SzwOIgfMdHMVaKjGnWmu8IpfDud3dGEynMZBK4S2Dg/j6xARG63Xfqc/TGpV2e4mpYdPiIq4vlVB2HDxareLjW7ag6rpYNKINj8kD/eE1Ar8CBmnu40AyiQ0dHbhrbi7ERErfg5rrohnZHXlaJU9o28Axx5BK9p57lj4fWgchTdIxynXJa312NuyNLiWo5WjFirC0PTsL/Md/kBp/uai9DRsoNztXg4tKiVKaXM5RyvOIgbjwQuC44+jcoaGdJ5KZmyMzQL1OG/+f/MnOq9b9ptC995LPAIOy1FbsiljTEpWG45g0KfXy96hEL9uQ0rFsQz5bUck9uo7cN7neUSZOnh8FezkHL4evw35A30tkKxVSqfLq8mbMKt6DslmcyXq+GDq6UMDbh4cx7Ti4eXoaTa0pXE20BcCXqGqGVfRDspSimuWGzbxqYgLvW7EC/akUkpaFs7q7UXVd3Dk/7ydrmXYcfHNiAkopnFgs4uyYnc8yIMebfM5kO0ua/jxZreJ7MzOwAByezyNv25RCFPAd+lizcG53N1YJ99221ph1HFiabLalZhNdyxifmElgoPKM5PhQtYo2zAOtRC10M2cz7TYWjbd9NMGOrYIwQQC+rR9i3ZRFRW2mjdPiwwsLKNo2XtvbixXpdKhC2+v7+/GNyUnMOU7oXvlEApf09aHkOH58PI9pstXChOPgqGIRN01P46nFRapZrpQfvsgMjPTs31yv49uTk1hwHFhKoem6KBmwDeV/N3PBNnFb/MZj7bFtvHlwEFXXDUUSyGx8UpIPrYsK7IVyYywWye67bh2lTuX65f78m6FIic3zCOwkYMg2t2+n2N6LLqLzv/99aptpdjaQtPg7f962DfjWt4J+HnQQcNllJN3H2U4lAEQlOu4bA8DYWKB63hUtLlLN9rExYmDKZeArXyGbfhSUftPo4IMpouDxx5cH0uUoypTFebpLJoHXUT4H0XVgYJbMVpxUHV1P2Ybsn2Qy5NiWYzokQ8OgrhQ9d9/6FvDWt+7e3OwN2g/oe4lYEgxlTONj4rfZdhs3TU8DIJA7q7s7FP6jlMIJHR2YabXwg9lZtNmAhIi0JDdV8bsMedNaY7zVCklTOdvGuT096EomMdJo4OFKBbV2GyPGIFR1XRRsGyd2dOC7pRK2NZtwPQ8VscmnbRsX9PbilZ2dUEphc72O705PY4fJ6lZut5FQyi9p6nGfQbXNz+3p8WPU5RzJOVxS79vQWV1dmHYcTLRaARhpCgGTUnDo3UjiDS9IaQodOAUysLFt3VWBXV7G+8u1XPQ8TLZauGd+Ho9Vq75a+zXd3dhUq6Fu5ott/FDk2PbLSgV11tfxOLVG1XUx1Wrhrrk5PFAu+4l4EkoFvhCmja9NTCBhdpp6u42pVot8BXQ4Sc+SqAgQ06IsC6/q7EQ3m3gMddg2jikW8bNyOdC8RJ5lDSqSc7DMAJhyYb+2BHVrL9RiEkoRkJ9/Pv3c10evlSspm9pjj1E6U7nRsp06KuVEJSKAJNqJieD4xASlcDXdDUlgkiYngf/9XwJ1BoD5eeprHCht2ED9NH/XJRJaNFSpo2P3M4JpTWCudWCPNYkmf+NpcJDi6xsNMjUAVN/+sMN2ft3QENnYf/IT+h6VtHkepcpbHgeWSs0SyIGldvm4Z0x6zUvglvfidqPMg2xTXh+nPSoWdz0ne5v2A/peoJRl4YyuLvxwbg4t86RI6Uu+l9ttv8AJQEk84gxkhUQCp3R04Cdzc3ABdCQSOK2jAxOtFp40kq+/UUdYSP7NNZv6LTMzeOvgoG8j70omcW5PD+4rlymNqtj8Zx0Hd8zO4rFqFc/V66iYOHKoQL19Vmcnzunu9jUS8+22D+YKVGjFBzIx9qMKBZzf07PEeYvHEWWC4mjKcdDgBDI67HwWuo4Bk9sV/2yeM54nOj1IW8saABkCBgTaEe7jtkYD25tNTAnjYcMAc4vnjNdHa7QBPFOrBZoDbstIy883Gni2VsOCqICmzX2PLRbxlGEUNouAZ5bcOXGOFE+imf8UgOM6OjCQSuHs7u5YJ7qRRgP3mWyEsg+szXlFLofze3rQL2z4aVvh9MNSuPM2C02QI9x551F4mqS1a+k1MUEqbr9fKnB8khTdQOM2zfvvJzCP23zlpq8Uqfc5Jzofq1YpPWk6vfT+s7NkF45u2rJvcmPv6gKOOGLJlMZSHCjsrpT7m0Br11KUAgddHHVU2EchjgoFckT86U/D8yedDqNAzu/RtYxKzVGPeSY5tzL0LWo6ibYrr2eSz1fUWS5KStH/YMOGnc/J3qb9gL4XyFYKG3t7seC6+Fm5TDHNOqh/rSAAwxCD2JO1GqYdx08TCsAHk7rnIZNI4JhcDn2pFI4rFnF9qbTkKeJNl1XFvnrbtPmrxUU0PA9xCYskmHI7460WxlutIK4bCFXaauggdCxoSPtSIl8jQfOgbBYX9vZiICZgVQIxv884Du6dn8fpEfPEMwbweE616BePPVRXHmFgCwFV3DHxDsAvTGJF9XlKYdrYzWXVu5FmMzbvgD/POpwYSDINU61WLCNzRKGAC3t7sbVeR0P0TTIvMrRxMJ3GQZEc+jzWV3V14eFqFTdPTwNaI2Pb+J3eXj/00FaKQuGic2aYE0frINufoYRl4YL+HjivJOm5WKQiK7sLUFLVKjfu6AYqj3Pbzz9PDnRMUsJjaSxuk5btzM7G94vrmS+3qUvJ/Jhjgrztu0O76tNvAx122J5LoP39xPQ89VQwXumDCYTnIZplToJonCQexwQwyTh3vibKWMWBury/vFYel8/dy7mG+wF9L1HasnBJXx9aWqPpuqh5Hp4zsc1QwharhHe51niuXsdz9To8k9wEECpgpZCzLORsGy3Pw/enp/HM4mIIsNj72NIaKzMZlNttX8ILVciKocFkEqtTKWw3JVwlKMonU0porDaWIAAzRoj7ScCEUjg2n8eamHArHq8WnwGqQf5YtboE0A/J5TBicsMzUHPsON83JJ0DQSU4MS42AzBDoMx8R/uiDOitTqexudHwAbg7kUDasjBhNBNyTXneJWPF6ViVUn41M9YIRNdSRXaEMzo70Z9MBmlfRVsA2eYPyWZhKUXaAqVwdD6PI2M0P67n4cFy2dcqpG0bbc9DxrLQkUjg1d3dOLGjg6rb8RzwnGiNLY0GRhoNHBBhGDhELZEgSWxnG9oBB1CpVGnb3tkmujNJiNuSanF+502WjxWLFN/OhU6iUl70O3+WL9kv/r2jA3jLW5YfbxwlEsDhhwMPP0ySYyZD339bwDyOHniAtC88T2efHYSsMfX3A6edFs5PHwXwODAGln9WlgPana1jjDwSm0xGag/irol7l8/svqb9gL4XqZBI4J1G7zTWbOKqHTuobKchH+SUUBcLEOd3jSBpSs3z8EPOuY4gqYwPSgCgNXpTKVzQ04PNjQbunZ9H1Ry3LAvrMhk/AYmkg3I5nNTRgZFSiZgIya4KUIJSIYnc00HSlHK7jSeFDdkfJ49HMC5HtlohVa1P4l4hhidmjs/p7kal3ca95TKV/lQK67JZpJRC2/OwtdHwc6ezd31sCVEtqt2ZMXkIO4FxXwq2jUv7+/G9mRl/HQ7J5dCVSODWmRnMGg0LxFoyIycTuHCOeCjyK2BPeF+Cl0yReV+RTmNzvY6xRgMtTtIDYj44YU9nIoH3rlyJcVOu98nFRdw+O4ucbS8B3gcXFlD3gjrtTc/zHfQ6Egm0tQ7Kxcq1ZMYpZpdytcadUwu49UdFtKoWCgXKAnfGGTELCArtqlZJSuNp41CjdpskY8cJfpPSNt+eN/dTTiFw4DKlvJFKqa/dJjVvNkuAIgE9TiMQBwRRikqVe0qZDPCmN1HoXqNB6vrXv/63G9B/9jOyqfPatFqUGCcaO9/ZSdEIO3aEJeTo2satZXQt4rzr93QtGbijv0kNQbS95Y5L5nM/oP8/RN2JBA7K5bCNJTgvqBceqsMNhKQ53jxlfDVgNn9Fnuy+Sl9IlBuKRRxrXhXXxUi9Dq0otekR+Tw2mcIaWcvCBhFLIzOj+Wp10S/ui+ybDz4AtjcauG9hgVSyok/yX6gBPFytImvbuKivDx1R261gUNgjO1R2NEKX9Pej4rrYWq/DsixcPjiI1ZkMSq0WPrFlS7gKm5CSlQSkOHBS5IDGZgMXlM1vRTqNdZkMPrR69ZK+bK7Xca9xIpPMj88MmWNDqZRvxuBqaAzI0DqUV4DXQimF9ek07l9YoHKm4neYc9K2jeFUCnOOg9tnZ/GESbrzXL2OW6en8fqBAQwLD/wfzM4Gle902Gmu3G7j+lKJwixZEyDWkucrGjbXcF3csTCNhpWB0mlUKsB3v0sS8bHHxi4hzjuPXlGq14F/+zcC3agK0/MIlAcHycGKaTlJia9vNoGbbgpv/DLeWFIcYEhQiUrp6fSubcfLUbEIvP/9L+zal4IefZS8720bOP74PQuhe/ZZYtI8L3j9+MfkRxEtQnnAAZTW9qqr4oFUrl8UQPmzVK1Hvy8npe9sLflcywqeDelgJ+SVWEk/CuIyamNf0n5Af4lopNHAD+fmSErGUjAB4Jeg5M3dB01zjnT6cgFYngfPsnznJ1eLFKdaY6TRQFcigSvEbveLchn/PTGBuolr7zSq265EAoOpFCaMd7i/wRtg6bBtNLRG0/PgGm+SaDW2lpCIlZCIIYHdPPVaKdwzP49Dczk/IYkkl+fDXJOxLPSkUphxHLQ8LwRKAHC53NHNPD5Tq8HVOnDi06J2uBJOeiwJM2Nl5jfEVJlzckrh3cPDcLTGrzjhuFLoSSSwKpNBZyJBxViM1KvFOLid1ZkM3jIwgKunptDyPDieh7KJJ2fVu2dC6qQ2xNUady8swHPdwKygFGyzfnPtNnJK4cyuLjy5uIifLyyEJOpN9TpWLSyEatv3JpOYdJzAeVMyCUr5lffYZCF9EsDrGUOu1vBW1oC5FKAV5ueB669fHtDj6Ne/JkB/5SvJNl6vB7+xLXx4GPjzPw9f191N4FOvBxt8VxeptaenAzCQns2cDS7qFc0kk5FEJcX+/qCdgQHgHe/Y/THuK2o0aD4LBQLQXdEzzwDf+AaZQdJp0pCcdtrux1LfdhswOhrMk3xfjhj4eS6ZJFBGK+fJa+V10oMdCI7HrSP3TZ4nHeaWS0QTfRaiEr8EddddPj3sS0n7Af0lIq012cQNAFpCQo+TgjUCtS8QSG5aPOm2MglARFscVzzfbuNbExM4OJfDcYUCVmUy2NZo4H8mJymDnWlrrt3Gv46NYSiVwskdHbjbxKOHJGsAxxQKeN7YSz3BXEBruJ6HbfU6iskkfjQ3h7amuGYfzMUYpeS/3H/bUgoF28aM40CBnKwOzGZxSkcHvjc9jWq7jQt6e9GXSi2b3nTBdXH1xAQaQhfmAT4wcSx3SIUsPvtSvVIoJBI+o1RMJNDyPNxXLuO66WnKBaA1ji4W8X9WrcLr+vow3mrhoYWFkHStzf1TloULurtxSC6HTxqX4F8tLuJKTh4knwsxb35fZd/Me9aycE5PD66fmsJ0u42vjY/jpGLRZ6x4l8oqtSQv//9ZtQp/8fzzmDbPBJsAeP1tw/zwmsm15H5Nt1qou24QeqgUvIQLffYU9OY8UCGzyp5saCMjlOaVbeGmWZ+W22QBUlVXq8CmTZTgJpOhlK/r1gEf/SjFeHN7PJS4mPPo93yeJFaOMtSa1Mcf+ciel4Hdl9RuU2jYtdeSo94VVxCwDwxQfL90AhwaonFedVXgXNhoAFdfvXMNS5SKRZJKWy34JU+7umieOaKBPeIBSv6TzQZMWxQYo8Apf4tuo3xu1OmN1zuuLek4yddyW8s5U0bvKRmS6O/A0ox1+4L2A/pLRQZIlOdBmafD3xx38qQw4GgQ0OVM+U2tNZpaIwEgm0iEwryanoefLSxAa40tzSZun53FG/r7ccfcHBZMNjdW9fLnqVYLtxqbsLR9M/2UvfUBP76Z/yt118VXxsfxp6tX4x1DQ/ji6Kg/hmg5VDkmiHdJK9Np/OHKlfji6Cg8rXFALocjcjn8g4kxUgCeWFzE6V1deE13NxSAnmQy5BewnA2fz5Cqah+oQBXLHE0laAFSsX9w5UqsE7bnX5lMbL6mRVE8+EK7jQ5TLrZg26gbx0Yp9bY8D1dNTqKYSOAI46R2SDaLCwwgM5jyzuCr2yPzF53Lb09O+uBdMjkL5DMBAKd0dOC8np4l8523LEzLOTHPnvTP4LYSJqGRX2QHwI/m5pC3bWzs6UHCaHQKto2qcqGhYBnJaU/qQH/lKwQ0cX8Nf+yKwGp8fOlvl10GvP3twBe+AJx+OjmY1WoEVgsL4Y1WqWDD5gpmjcZS9e8ll5DamGPcuW8TE7RZDw7u/vj2FWlNhVyuu44+j41Rqdkjj6S0uDfeSM5rAI3nsstI/c3SLoNfo0GJb/7kT8jpTyh5YumKK4j5efxxur6jA3jDG+jYF79IYP+hDwXaguOPp/W59tqgAhuTZLyi4B33fMigjDh7erRduQVHf5dAvtyzGKd14Db5ZVkvT7re/YD+UpEAQR9IGXDMkyXt5NFwq5SRFD+6Zg26kkk8Ua3iq+PjOCCTwQdEfIxSlA3uoYUFOmBUv9dNTYWKs/j2bK391Kmekaxh+uZXhTO/Kakx4DFZVESk4rr48tgYLurrQ1opymhnztE8RkEKpFlYTgd3UC6Hfz7oIP/7T+fnl5go7pufx8/LZWitcUl/P87t6fFjw8scysbnmzl3NaWbhflNhuclLAsX9vbiicVFHJPP45hCAZ2JBNIRqVaZlyfaeqpWww2lEt4xPIw3DQzgsoEBfGl0FE/XaiHmwgKl5b1yxw58av16dCWTeHJxEd+ZmvLnVIY0ytmJi1LIWBbeuWIFvjo+jnq7HWThM22wGn+5ebaVwofXrMHfb9uGWcdBk1X9Wji9CcZoQ6GAtFL4KfsJmHNvmp7GYCqFEzo6UEgk8K7hYXw02ETNAAAgAElEQVTuuXHUbTqnr4/AACAQljXQM5lwne9gvGa+VVglyi/PI3D95CeDTZOvufBCer3//UH62VyOQOSTnwzfn9uybYqnXrMGuPvusPTF50Xj02s1qgzX1UWSeqGwZ4zLcqQ1MR4A9euFgsG2bZQ2l6VDKaHefjtVmGNyXZKQZcESXn6tSdvx939PYP7RjwaFbOJIKeC97wU+/3lKxnP55QTWc3NBjoEvf5kq3ZmMyTj9dPr9+98Pt7Wcb4MEd6lqX+432VYckxDnNc/Pxc7akr9FpXN+ZgYHKd3wvqb9gP4SEUvY0mOaJdjQJq4DOy4DTtKy8KerVmFlOu0ng9lgsnPZloWkZcFhm21URWvadMU/UxbfYHWqn0fcfJaqfgXa+B3ZvpC8+bpp44h1cV8friuVfM9tT4zV9yRfTjMhKOSJH2GCtCYbvWfG/d3paRRtG6d2dWHWcfCl0VEqqCKYIimV+7nolfIzwp3T1YUzOjtxYrGIL4+N4cbpafzBihU4OmaH5rVktbQcSSKiKVBiLf05FWP3mTgIqThCPF/REDb+zOYEXm8et0x+45nnIFrfviORwMfWrsV4s4nPjY5S+mD5PJi2D8/n8ZbBQXzX6MFlVAA/wwAV7rlyfBw1z4Xncux+UCP73nuBG24Iun/KKcAb3xguLRlVpcsNWKpUWdUpQV/rQNUbdeRKpYLrZbvc1pYtQbKZ8PzT+wc+AHz605Rlju/ruqRN+MQnaONmBuPF0JNPkjOg1pRN74MfjK8xvjskgVnOo+sGtmKeu1tvpTzyth1OlgLQu+uSRuILX6BqdTujZBL4oz+ia9Jp4Jprwg5os7PUzqc+FeR0T6fDmdui4CzHJEkyHnJNo23ESfhxn/ldMghxDnu76kNPD/l45PPEuO5r2k2Xh/20p3RoPo83DwyE1e0M8DpcSY0lQB+gjQpTVj9TSqEjmUTetnHP/Dz+cvNm/NXmzfir55/HY5VK4PAlAB7ymA4ng+H+MJhw2lWAamv/zbp1OCCbDSR4rUMMBLe9o9XCTTMz4CI0WrTFaWiZSfAzme0OiXnje/pZ1wC0PQ9XT07ikYUFfHZkBNOcrc3zoDwvzEiZeT80n8dJpobjqcUizu/pwVUTE/jk1q3Y1mj4poQtRsrmV2gtTX8OyeXwu729xGAwgybm0J8zTQ5+7x0aQmciAa01jiwU8IaBgUAqF3PqPx8qcOLzveW1RsN18d/j42ixKUVr37HOZ1bM/X9WLuOOmKwp3K+vjo8HaWghGCClsD6TwduHhpC1bX+MO1tLrhsft7xcFrXVIlXuffcBP/hBWHqUzUlwlxvtcrnWlSLVeFwBGNmnKF/EwMaV3qKbt9YkfX/oQ8DHPra0/CeP6cXS1q3Af/4nScutFjkEfuUrL6yttWuB3/990lLEZUMDwuDXbtMc/PEfA//3/1JIWRyIcRjhriifJ3V73Fp63tJ2zjorCG9cTtKOgq4cC3+OYwLiSD5z8lh0a+JnY1dMhexbPg/84R+SX8Le0Nq8ENovob9EZCmF07q6sDabhdYaD1UquGtubinbDIQSowDYJei1tPYLaPjXMTiItniDls5SPmMB+PXEpSR4QkcHXmdCy94xNISa6+I/duxASYRNhVTEWqOYSOCUYhG3zMwEBVw4a5wYywW9vTh6N3WJp3R0YKrVwo/n5ghII/8sDaCuySSwaHYNCUg8F/z5gFwObx0cxJ1zc9BK4eHFRTy5dSuqpsQrS7wLppY9S/OdiQT+cs0anGZqod9spNXNjQb+v5ERKEVe5jOOg+eMm3V0T2lojSsnJnzb/1+tWYOzurqw6Lq4zURCAFgiEkRDChnAtdb4vaEh3FAqoR1ZE45SgNGw1COir6s1/n7rVpQch1L06rD2gJ9N27LQlUzitpkZPFSt+ozRcmtJc60NUxI094tfBCpVBhLHAX70IwJg3hTn55dusvJz1KtZhhVpDZx0EpUrjVLkr+a3F5Wu4v5y3/0ugdNtt5GUyhXlom29WNI68NAHSFrlKnR7SkrRPFQqlLeex/jMM6QKj47z3HPJnp3JkJnkfe8D/vVfyUTB1/b2kjr9hYxLvnd2Au95T1gzk8mQc9zOrpdjWw7oo5qXaDtxAB53XvScnVG0DaUCc8LLRfsB/SWklGX5iT2eq9dD6lApfbLalQG2bWyuH1q9OlQX/IlqFd+emgrsnkr5NlR+ujyl/KphntEPJpQKVOZCIveLjuhA/d9h236c+FA6jf8YG8N8u02Z6xAGTLZHlx0H91UqfnpUtqVDUfW2k0yY2vpMZtna6ZIerVRwfamEpieKqSgV0mJ4AC7s7SUzhAqqqmnTL5i+srZgR7OJL42NYdF1YSuqStYQcyGZgarIQrfYbuNLY2P401WrcE53N2rtNn48P4+21n5GvprrotJu+06EtpGUtVLwXBeWZaFiKqex813GFMk5JJfDQ5UK7l9Y8PvBGgWuTe7nljd6yYRlYWUy6T8zPHZl5l06Jkb3pC+NjmJ7swnHC1eeC12jFLY3Gvi/W7ag7rpouK7/uwfg7O5unNvdHVpLS2tYaQ/69TuAa1fDNrv2oYcS2N59d1hS5IQqcrON5uOWm6pS8ZWweAN/7DECrUsvDfJn1+tUG73ZXOq1zN+Xk8wsi66/+mqybXP5BVkBTvb5xRCPg+9rWeQrcPXVe56BDiDp/MwziSm49VY6xur2I46g/j/1FIWlbdwYVg0feig5uH3968H6pFJUWGdPKJUiTcFXv0oRCPk8fY8rXsPjl98lE8dzLgE7bi3lGkeBdmfgH20rqtmIa0dWgQPIdPDWt+68XO6+oP2Avgd0y/Q0Hl9cJClEKfxuXx8Oy+d369rji0WMNJv4ZaUSVssaEAql/QQw1277KvKRRgPfmJhAzXUptEupcAUwIZHaCOyrUgKXgNydSOB4U/hFbuRahaucXT05iac4fS3C9lmIe3gAKqZ8pzL3OCSXwymdnViTySyJId8Z/bpW88OxuC0Gcgk4CsDDCwvYtLiI0zs6cNf8PBYZeMw/jwHdMtJqyeRKD82ZGQd4DTRFEnjmHK0UJXUBkLdtrMhkQv4HCsAvKxU4WvsMBddXl06H3K+qUev/wYoVKCYSOLJQQH8qhcNzOTywsICnOeWqCswXnlKwI2upDOPG45O2eLlKDy0sYGU67TNVs47j989nkMQ88/G21ig1m9Q2o57WOMl4zheEOShv27h8aAjTjoMf2fO48O0aPbnAuWtoaOlmK4GaN8VzzwV+/nMCUL5l1KYpN9tDDqHvzzxDEmW1CnznOwRib3gDcMcdgYe6LO3Km7kEdLnRy6Qg7CEfBRbu/zJVfveI5Ni4X/39wMknv/A2d+wg7UiUkenqIjX3KadQWF8UgJQiiR0IwvXy+Rc2zmOOIeag1SKAXy4E7pWvJG/8TZuCNZCak6hqO8pEybXkMUSBXb7Ltebj8nmTzFoU8PmauOfg4IN3LdW/1LQf0PeAphwHY40GANoEry+VcLllhUKclqO+VArn9/Sg7nl4plZbki0OyiT1AJC0bZzW0YFvTk7C0ZTcZaTRCGyXEKCkw5nQfDBhCUyoYG3znrFtHJjJ4J6IJHdQNosTjfHne9PTeKRSgeN5vmc3S44M5ADQmUjgyHwe91cqobCnnlQKJ0cTOe+CtjcauHZqCrNGIyBBM1TMxBybdhx4jgMbwGu6u+FpTRXvfHuuChgWcZ38tx9bKGC02cQkZ3FDmLGxtEal3cY3Jyf9hD08Z2xe4Op5SrTvGclaS9OD1sjaNo7J5/FcvY5bZmZwQDaLi/v6MJhK4flGAyqia7UtC2d0dKA3kcBd5TJO6ehAXzKJrEkwZJk+hwrjiOei3G7jttlZ/KxcxgW9vaEStewcaEXWVCnKbDeYSuEJk2GQf1uZTqM34nmmFJlqWp6HjkQCJx9swxYb2+GHE8DKTfYXvyCwldL5YYcRkN10U1jlHLfZKhUuyiKPeR55WEtHNrnsUqLj41FVrWURsD33HMXGy8xyShHIbdxITlAvdhPv76e853fdRW0NDAAXX0zOcS+UenuJ4eE5kswLV71bjmybtCovlpQCTjhh1+cNDQGvfS1pBJ5/PtxfgBgLLjEb1eJE1zHKNEaBPZp9MMoExIWtScYiOj6mvaGp2Ru0H9B3k35eLmObAXMAvhq3sgfZA1ZnMliZSuEZs1tJifPofB7PLi5i0fNwUrGIGcfBM4uLcM15vlQmrgFiAM/87qtvxTGtKIHLEfk8fsI1rwWoL7gufjg3hxPbbYw0myTxIjALAAilZ83ZNjb29qI7kcADlUpIulfRp383aMGUYdWIMQf4DYf/oUprjLZaeMvgIAZTKTS0xj1zc2ibfnK//Xh0MV4N4IBsFgum5KmvemcwN/dztcajlQoanoesbePwXA5PiTRmB2WzODiXw8OVCkqOE2I6fAaCmSmTSvbaqSlsazRQarUw2Wrh0Fwu5KHK0r7WGkcXCjiyUEDGME93zc3h8Hwer+zs9GuXy3liRojbmGg2MQlyXCtHNB/KzAnMuPlYZyKBtZkMnqhWfQ9/pRR+WalgXSaDQ2M0UynLwqkxTNzg4NKY7a1bge3bw8csi6THhQXgzjtJVS5/i27Ac3PBZyaeaq41zkAdBwRRIF63DnjFK4LjGzaQ1HXTTdQnyyJP7jPOoOx0r3nN3pHQc7lA9c3S+dFHv7g2OztJvX7//fRdKVKb7+tynrtL69cD558P3HwzZZwDlkrpci2lZA0sNcVEgZyJTRvyOYpqjOLembjtKGPgOBTjf/nly/sE7AvaD+i7SZvrdV/dDdBG+0IZc97kNTUEaI3jCgWsTqdRdl2c3tGBz23fTqldLYvityNOZlJNKp9e3syPLBR8+7t8pou2jfWZDH46NwdtWaH866VWC9OOg/l2O3C6U+G64bJ4StaycFZXF6YdB4fm8/i18aQZSqX8JCpMP56bw5gpLcpq3pRl4aLeXl99O5BKUd75Ws3/p4QYEzF+zsYWTCppHi7s7cWi5+EX5XIAVOYUyeCszmSwJp3G+mwWj5qUrry2tlI4PJfDplrNz8TX0BqPV6s40EjUjelpbGs0sDKdxnk9PTiyUEBXIoHRZhOPVSqY10FYlxZ9rLkubp6extZ6HUoplNttPF6tIqkUjisU8Fy9jgmufidIa405x8FNpRI21WqouC7SSuH0zk7MOQ4xGGIt5bPJn7c2m/DMunJ/fI2LPN/Mw4HZLNaZjIO8e401m/jB7Cyyto21mQxcrfH9mRmc1dW1bBa/ODrqKJJyzfQCIMnUtkla/clPaJOUatE4Wr2a7NsTE+G2zPAAEAhv2EAS3iOPLFWj8ia9ahVw0UXh9leuJAcz9l1Mp6ngyN7etLu6SCrfmzQ8TMwHj/WAA3a/XvvLQUccQZqZW24JMgbGhTNG36X0HPddXhunQo/+LiX+6PlM0iGTJft8/uWX1PcD+m7SQbkcnq3XMc11wpXCgZlMyGltd2h9NotNtRrVvuYNXyk8Uq2iP5XCxp4eJE37gPFQN2pzBiiK8w2exJCUrjUOzGZxaX8/VsUEQs47Dq4rlUL2+hBzoiknvCe+85PNEiODRt3zcNfsLDb29uJ3entxl1EDH5TNYkMkbuPRahXP1mpw2SNdKSQU2fwv6e9H2rLQn0rh1M5ObFpcDKm/JfkAJLUB3E8AxUQCF/f2gh3nyu02tppCNXyuBnBwNovV6TQer1Qwbxg1zzBQR+TzuKS/H6npafyyUgklXJlrtzHTbuO8nh78olLBEaJM6Rkm88aWeh1zjhPyUue5rIMyz4Wkaa0x3myiO5HwzRvRbHFtrfHAwgLKpt2HFhaQt22c0tGBpFC3h5gcZlK4/5G1ZF+Q6B7EbSTMGgEITERK4blaDTuaTWyu1VByHDywsIDZVgu9qRTO6+nZLcfHk05aqtb1PPIon5kJwqnMLUMkN/N16wiI77iDrpMbMA85maSsb5UKFSDhKRLTAYBMAE8/TY5hkl772l0O5zeSVqwA3vzml7sXe0Ynnki+ED/8IaXsZYk68lj7FCdJy3WNMoN8LMoARK+Tv0Wl/mhYJTN6GzfGJ0zal7Qf0HeTTu7oQNV1cefsLObbbaxIpXBxXx+21et4olqFrRTO6OwM1wiPoQ3FIhZdF49VqwCA0WYTC+02Hq1UoEGVq17d3U3gqQKnMx+0lAplmOs1xVbmBYCc3tmJ7Y0GnlxchNYaScvCGV1dcLXGbbOz+GWl4jtsadEWxD2jpMU5TI7WmDSlQw/IZpeU6mR6olr17cxQQfERx/Pw03IZKSNdKwCLrotX5HI+gzLrOH7tbr5egjpL/JL6Uim8e8UKAMAD5TK21Ou+dK8B9CWTaHoe7pibw1ijEYxNUfz1GwcG0J9K4cLeXjxerVISH8NUzToO7piZwaX9/X6pXEm/rFRQNdXM5D0h1hIAEoJBU6B4/jHOgcnXxUjpXPiF5+pOLq2rwv4Q8vnx19a08f+z96ZhklRl2vB9IiL3yqysysrau5tuAdkXZRPZF1kEEQRBHREVBWcYndFv5vWamfed75vtRccZZ1xQHDdkUUBRQNlEkF32fWmgm+6qrj23yj0zMuJ8P55zTpzIyu6uamiahnquq66qzMrYTkSe+zzbfeuvGbzFgwrVc65y76/K1II2Q8rxund+nsaPMTxQKCAWCOCIRAKhbWREuf9+AnTJ6hYIUB5ZTqa1micAwhjlr0dHKV+sH5Jz/2JgzRo/wYz8f/u6Y2yMpFjbAX3ZOtu6dfTDGI2ZFCN8+GEvHWJZFCVYCtAdeyx1LcioiL4AAxaG4HXu9fbPStPFVPTPyf+1e9zyfX2h0N9PqRbAz+8PULpkW4mA3khbBvQl2OpwGAmhcrUyFMJ0o4Gbslnkmk0EDAMt18WJgt96S3ZkMokjk0m8Wq3i2tlZzIuWJ84Y7p2fx/5dXRgNhfBSpaJY18DYQo8PVE3eFwjgD4UCKkLB65VaDc+Uy8jbNklsGgZczrFrNIp7RH4Z0EL2uvE2SloIARjGMCiUuqSgR9Qw8LH+/oXV7232aq1GPc/QAEV8W+qOg1uyWcV41hcM4s8GBpTXe2Mmg9uzWe+6te1dEGgNBgIdF1LFVguv1GqqUFB++3oDAWxqNLBJ1ES4nCvvfNdwGC9VKnipWkXBtkkFTswasiVwqtnEb7PZjnrjd+bzyNg2wCj3rqRHQaAbADASDqPhOJiTixxonrW4PgnEumCPnIPk9ejhcot5Ajr6AlCPWHJACe0o4h1t8SBrBlzGMCVy+5DPgrbvdCCAbsvyzk08u/K52BaT3rlY5yqv56KLPBrUF14Abr7Z22Z4mCbi3/2OqrrlfvSJdpddqJ0oHqd9Dw1R7l6fwKUXCBBP/NTUtkuivlNscpLIgV54gV4feCBFQWZmqH9f6tMHgwR+xx/vUfJuzV55xev51xdn0iQAyyI5HaS31Icun41OrHQ66+DmqGcPPBA4/fTFXcOOsmVAX4Ldlc9jo5ALvb9YxH2FgmoRa7gubs5mcUgigdQil2p35PMYk/lJeBN1xDBwweAg/vdrr6EhQ9TiqdR7rWOmiZ5AAPt2deHZchnrhJd+b6EAR4jCuK6LuuPgpmwWnw8EaEIXT7bLGCJiHwzEz16RKlzMy/G7nMMySEf9+UqFPgMCxsVM4Gem01hXq+HlWs2TFtW9R8bUtzTTbOLa2VlFXTvXaPhBiXN0BwJqkcIAnD80hAFtzJuuixcrFayv13FPoeALeTPG8KLQDAc8kHNdF47r4reiLFjvudcr5eV5b6jXcXc+vwDQlRetASVjDGkxxqlAAJ8bGsLvczncmc+r9rj2PL/cPikq2gEhUSrvifgtawwc0bIHeMVw8pq59nzJynap8pYKBlWY3+aUp5fj5HTw+hOWhdP7+rB3LEYqcPLHJWnfpdprr1FolXPKmbdPrABVP69dS++dfDLlm1etokr2n/3Mvz+pgMU55eMPOoioXScnSaDknHOIi719Ypf27LPE7b4tE3c+77XJAbQo2NFEI9vLHn6Y6GqlPf44RUIefpjAXIJovU6FbgcfTBGVxdjtt1NRo7yXgP+5MAxKyUhQ1yvgO9Uo6167vuDTFwJ6SF3vaJD/6+ra8T3mi7FlQF+kZW0b5VbLJ3giPRTe9rNYi5smLMNA03WVJ8hc18dfLo/HAAr7wiNuebeQGL1yehqviNCoBBQHNMlKk8Dpui5szpVMZjoYxHn9/ei2LDxULOJBoTXJQQDf4MQH7ghP+lODgzi8uxsGY2i5Ll6Tlf+cI2ZZGN5MbC1hWTA4h62FfBW4cE7ypuLap5pNfGdiomPRIQe1qJ24hdlhvNEg5TaIHm3tPunpBOn1cxCASw/Tx48vzokJz1WFytu8X2lxoYTncE9vPQLgrL4+9AYC4JzjwWIRt+Tzqgpfp43VnyGXc5zR14c9YjHUHQfJQEDl8wvCu3e0e2lq28lxlYsyeY2O5r6GDANfGR1FKhjEa7UaXq3VcN3cnHrGlGqceG6iloVTentxsJjZui0LU80mUcOKBUB7mmBLNj0NXHUVAa4+ectdOA5N3A89RLzgMqy6Zg0ph+mgrx9WgrXjEDkL5zQhX3yxN/kz5vfmpJcejdLPUq1apQjD3Xd7wHDcccC55y59XzuL6WkN3fttB+J2D3sx+9XFZVotqtqX68VCgX7L+yjPoT1AJM9Lmh4m18FcB3T5Ohj0A/hRR1EHxlvdlgF9kXZ3Pk99wjIXKjwc6c0siPMswj45OIia61JO23URNU2ETLNjyxkD3ayQYaAhBEoYY7g1m/UVWUlwdPVzYh4rnc79bQDYWK/j38fGsHcshguGhvAhTej5a2NjeFW0zhmMJEOvnJnBrpEIBkMhrK/V8PXxcRVZWB0O489HRtDToVDw88PD+JdmE+uqVQU2poggyPPRr1fmnNVL8Z4LCqVPNBpgnKM/GIRlGJgWFfQAMNto0H0S96idbEbfFwBAnENHMJIzBNcY/cRnG66LUqvl49z/85ERfMNx8GKlorzoOue4bGLCdy06/7o8bsAwEBKLNohnICCOHzZN/OPq1QCAqUYD/z42hqxG5qP7xjKaIK8T0J4L+J8tF8Cmeh1f27gRNkhcR7WvyTHjHCHTxAd6e7F/VxcmxPfg3HQa35qYQKbZ9KIMW4nYlMuePvkvfkEebfsmOtA+9hhw7bW+RxkbNgC//CXlzuUkLydu2frlul74HqC8/A9/SIps8njtxCGmSexpxx+/xUvoaI8+SmD+TjLdq5XTTTzuT2GY5kJec9sGZmfpXqVSm9+vvOeWRaIve+xB3v/Xv04LqHLZu//t7G76by0A6FsE6MfTzbKIFOeii5Y+JjvalgF9kXZmOo2sbePxYlGFjLkAVnCOkGEgrE3Gi7XPDA6i4jh4tVbDpwcHsX9XFwKGgXXVqpd7FZ/tDwRwaDyOm0RYuClIXzjgixFxwAdODCTHCsYQtizy+jmHLYDMBVVeXz87i08MDHj5aOmhwVsMMFDl+GAoRHlxmSMWYeirZ2ZwiSbvqpvPM2+Lc+nj5nJOzGjifb0nG5zjjlwOd+Ry4ADOHxzELuEwvjE+rnqs1TEAuOL8dNIUfV/y2Fy7Vm/gvLw04EU/5L6er1RwUyaDT4h72HRdxE0TXxwZwb+NjWG62URdpjDgr1mQx3O189ovGsV5g4MIGYb6nNUBIIeEfvw3x8dRcxz1LKqx1Lbhbc+Fel97PuZFfYNMtRjaM8QYVbq/v7sbR3V344dTU0oi1pERJC2yMS9IgZIdFnWNBjG5PfSQuEeuN9lq6yb1u1wGrr7aC7NKj1x+NhAgwJCCH4EA8JGPAIcfTm1P3/iGn3ymUKDjy320e3imue2FTbqn904wy/KkZeW1BwIkX3vppR4PQDgMXHKJF27nnMR5rrmGIi0XXUT70aMiwSBtJ6ln5SINoB79f/s3Ksb79rf9ZDOdBGT050mvlpfnHQrRcfT7tmIFRYB2RlsG9EWayRgCjKRNVYhWeLmmYeAvRkawOhz2UWIuxizDwEXDw3A4KazJ3uTvTEygKj1M0MQ81WjgFvHUMsbwnBTNoDdUWJ1rTy0Xi41LRkexJhLBpUIs5olyGVeK5l0Jco+VSoiZJs7p74cpJnIVfRDHqbkuvjc5iU8PDuJns7NwBADIzzlbiFTIkLI8Z7f9PS2CAO19CTbQtgMAx3VxhbgGW2wroxFM20b3Nn1npwMfb1MR4xRal33ovmiHdq0256i1WrhyZgbPVio4M5XC8b29+JuVK1FptfC18XGfB8u0hZE8N3keT1Yq6M3l8JH+fkrDcC/lEmkr+tstGsWFQ0O4fHJS8cYztIG5/ltzeeRnTMYw02zif6amfHK76pkT2x8Qj+PMdBpXzsz4Ig8t7Z5wADXHwb9u3Ii+QAD/3+rVC9rXbr+diE70cKo8tS3lPiXoS8lN+f5ppxHhy6OP0nunnkq52maTpEh1MHddApWzz6a8u9ToZoyAiDFqo/vgBxeex7baNgTtdho7+WTgmGP870UF3e+Xv+xfeOmUFC+8QGDuuiQW8w//QMVmF17ofeb88+l5+Pa3KRrzyU8S4Q9A4AtQseOll3rPw4MPUjGepOeVoX95Dpblqb3JBWIkQlz5V11FHr98PxhcfAHfW8120tPeMXZOfz9OS6V8IWz5xKSFWMa2WLRtsmagli4le8qYCtFKznC4LlptYEoba3KsEqQAdIljyAVHpI1cW07S9xcKiJsmPtjXh88ODeFbmzapti8Jkk3XxYpQCB9Pp/E/U1O+lrctVbwrwAQ8kIY/RKwzuun5ZcjXWggZYpz0/ctFgQRxXXxE78WW/+eAotw1DQOfGRzEbsJdmKzX8d2JCdjaNvI40lN/rFjE06USGq6LFqgqP25ZOLS7GzHTxF8MD+M/xsZIEU4Cr/wR1yVrB0RQUfEAACAASURBVGTUxHFd/GhqCi+JfnwOCuW36wYcEI/jn9eswdNCtAeAV40vxguieFCOuVxUhEwTnxscxEAwCNtx1PXo4y/H0uLEb990HDW2KpqhL1BAsrYN7Z7o1mp5wK0/Jrpnq68fdUBs53HnnCbds88mIHddyrN+//vkvWlEfr7j7b8/5bV//nN6veuuBCCSGGSJtBILzk8LOu1wkpHtaYHA5sdqS/IWEmzlIq3ZXChBK+tML76YAFiG5XM58s4rFWof+/u/94517LG0MJB2001UqCe/8qefTgWS//mfVLz4+c9TYWU6TQB++eX0fI6MkCLczmrLgL4Ei1uWL1+6PU2FgMXkL6MCip7UMBZQlaLttckYLhgYwMpwGOn2b5/mjelgZYPkWQEKm35+eBj/OT6OWdv2VWLLCml9Zl4dDuO8LZT1fnZoCHUx2ddcF5dNTKCos+/Bi3rIvRrarHhYPI65Vgsv67O1NvvrlenSKzU0YJIEPQzAYd3dOFYQwVwxPY2xRgMfTadxSCKh2g4bAmhduTBgnuKYXJjYoOiAPP8G57hmdhYx08Q+XV3oEzFc5SVr16r60LUQ+0OlEh6vVFB1HNiyNQ/+ELk0gzH0B4PE0CYXHdp4vTeRwIk9Pbh2dhbrpLQrp46FPxsYwH7xOPX4a/dfTzvIRceTlQpeFMprMvYtF2Q6772qK9mKtX9E3sKPfYz6yi+7jPLsErh7eymHKtvWGCMCmK9+lURGzjyT1MGee47C9DrRnvT+DYP2+fd/74VeGaOCvDvuIFB/PXbwwVQdr1+j7FleNs80H2LBAqjd9PGrVoFvfpNy6IyRp/4f/wH83d/Rwi4Y9HcUnH02UfNKGxighcIXv0iLhJUrvQXXgQfSc8EY7UdMCzulLQP6W9EE+DDX9Xm0Op+7BNV9YzEcm0zi1lwOr9RqCrBMxvCRdBqHdXcj0MFVUGHoDq/vn59HXyCAI5NJDIZCCIjFAwAEQMV8SctCRuMtB2MICra3zdmgqIBvui6+MT6OipAW1c2AVzCmPHjxOx4IoCQiFxJ4fOOjbaMXuUnhkt0iEZzS2wsGYEATGblgaAgVx8GacBiWYaDYauHyyUn1WVObhRTHfNssJM+HgwhfZOohbpr4yxUr0OIcM80mfjE7ixbnVHwGP3gCtIioi7y4uvYtzXoA9onFcFIqhTtzOXUvGIAXKxWsiUTwiYEB/HhqCpsaDVqcABgKBlFstVQEQmkEwHvG5OuG66KpgThEZEjuiwOqsFClJNrsvvsoLNoWTKJ7JYqoBgfJa9K9cYB6km+8kXLjcihsm36kiEupRD8SKKRnxrnHt865Vygnj9Fq+YvnttUiEQoDL9uWTd4//Wdr1mwC//3f1Hqoc/J30qeXlkx2BuZOMrCm+fa5d8uAvp3thUoFv5qbU2DMOcdBiQRO2kzbVcN18YOpKdICh+apiklWAooJ4tr+cF8fRsNh3Dc/r/7viv/vEg53BHMA2CMaxbHJJO4tFNR5MTEbvjsaxR5alcrHBwZII1zsf0UohEvHxogRjXmKX0vRqcjrYigAAYHw+EzAn8tlDHvHYnh/IoGq6+LQRMLTjucc183OoiQY61SfPrwohvy7yzQXcMwDwCpBkWu7Lv57fBwZ28aMbaMwO0sV/PCArl24hWsLCNkudkwyiXeJuCFjTAmZxGo1Gict5K/ujoyYyP2K65bRAH3eyzSb9IyI3vOhUAjnptOoOg7+VCyqz9Y4x535PLpMExFRn2GK8/3J1BQMRi2CcmGhi7MY2jn50iTi7yPFNV4/N4e8bVP0AySlem46rXTfAeCJJ0jStFrdPPGHPrFLsNVpP9eto5y4ninSTkd9Xu5bRv3bCWQ6Acq6dcQfftppCx6NZXsDLJejkHa9Tq1gJ57odQSsXAmccsqWt2eM9gH4GeHeCGGct5stA/p2trLjYFLIRslJM9OeNNLM5RzTjQYVh4lQrAJ1uQ+5H9vGK7UaRsNhX37XAPUb/zqTQY9lKc9Yt7hl4QO9vai6Lp4ol32FZIOBgM/T3lsAUs1x8OPpadyeyynlOVOAmMEYJptN3JzJ4HSt9a2jcWpZ8ympiZlYesPyWmWldV8goK5DJ3PhIqf/69lZlF3XGx8xRqZ2zA31Or45Pu7v9xZowjnHh1IpTDWbyApPVjKlSY9XX2BBO4Y8DxltWB2JLCiOzNk2rp+dVZGWdo8Y8Ie75QKnk9mcY6JeV6H+vG3jdsvCaCgEpm3HOEfRtlGwba/nHQB3XczK6iBx73zPD/Py+3Ls2gvldgmHlTxuxXHU+0HDwMGJhLqml16iMLkeQu8E6gCpVQWDlP+WE7cshHMcqlLX16e6h9cevmWM+sDTafLkbr+dPD35GT0KkE57RVfb0zgHrriCWrYCAQrzd2rb0u3BB4mshTFKL7wejfQdYeUyXfOGDXQPMxmKZsj+/P7+rXvHlkWLrRtu8KIp8h4uxsN/J9kyoL8JpoMy2v7e3GfVvCVzoyIMr1ux1cI9hQJiwvvy9SKLSb+2BVYHqdE+FAz6wKS9+EqawznWV6soOo7XKy9mZ8Y5YqaJEQG6DxeLpGIGArujk0lvv/rMLo+r5cKlR62AhDGsrdVw+eSk8oiP7+nBbtEoGGM4JJHAbzMZKhAU+5LApgr1OOmaS7lbCXr66JQcR42zLMhTwCvTINo90gv75Oc5gLtyOTxaLKq0RzoYRMQwsCIUwmv1OuWaxb76g0GkLAtrazW1z3dHo3hXJIKHi0WsCIcxGAxiSCywyq0Wfj03p0L6DJS3H6vX0dfV5QNmyHNkHQrxtM/p26je8zbkXFCnIf4+tLsbDdfFz2dmUG61ELMsHJJIICQ+n88TgEkQb+8X1slJOslmaqewYPsVK7xCqMMOI9rR2Vn636GHUiV2dzeByB/+QOF1YCEQ9PSQTOr2tt/8hsRhRDkDfvELAvX2Pm1pTz9N5z09TedcLlMl+b77bv9zfaOs1aI6BXlPZQ/6SSdR7cLatcThD5Bk7Pvet3AfjAFHHEELs9tuo9dHH+3nIVg2smVAX4L9aX6e8tTi9bHJJFZ0UDTTzefpwO+RdTKLMRyaSODuQgFNjbxGhT+1GCIDhV/H6nV/LlaESffv6kLPVor4RsPhjqps7dYStKhNGZoXCwjZ7w3G0GNZSmVtY72OZzVi7orjIGIY2EV612xhFbv+vrw+mUuebTYxJ8LDLufYLxbDbtEoHM7x22wWDXleYh8+YJPg1yFWK//PtffVokC+1jxUMIZUIIBu0ySWvDZwBICxRgOucDMZgI8NDCBhWTg8mcR98/M+0Oy1LJyUSmFAJAQ5qHr93dEouiwLu0YiapEEUP3B2mpVsd7JhUvOtvGkiLR4l0KRh2dKJRTEYkX9T46BjJLoCyrttW/xAm8hJF87nOPXc3N4vFRCw3EQMgxEDQMfSadhGUbH8DjnwOrVFG7lnIArl1vowYfDRCby3HMLNaz7+yl0K4VUDjyQqp9l//NxxxGYAwSCBx1E/e/6AqKnh6reHYdal+TwvPe9wF574Q23l17y2q5cF3j+eeBXvwLOO4+utd0mJ2mRIsdF8szvTICu3zP999wc3Vf90UsmOwM6QJGaE08kLgPDoM/dcQcVQw4N0eJt2ZYBfdH2bLmMu/J5TDabyhPbMxrdIqDPNpt4rFhUk6ScBDfW6/jV7CwYgAPjcazWQsiWYeCkVAoPFIto6rMP50rxqz3U+1q9TgQm0KqhGcORyWRHgo+lGucct+dyeHB+Hg1RsKUAU5v4AWBdrYanymVsFCpmEjjW1+uYajaxSyQCk5H2d7bZxHOVCmzA5wlzDTzVokELS3NtLF3O8WSpRGMlQVruQ4bTtfPj2r7azWAM+3V14YH5eTRdFwHDwN7RKBqcY62QAJPRE98+uEcHy5gnosNdF89WKjjdcZCwLMRNE3vGYni+XEbMsrBXNIqV4TB2j0axewe+0aM7VPWETRP7xGJ4olz26GwZcRfMt1pIBQKIGgY2NRrqfDYKRbQFUW49ed1pTLR0hL7tLpEIhsQio+W6eLpcRlN0BNQ5x5PlMj7c1wcLHkjqQD00RJPz/vvT/zZu9PeMSwsEyDPTOcP10273zo44YuE+AAL0M84gEM3nKRKw115UUb/77uQ5r1vn7VuGhVev7ry/12tyTGybFjNnndUZ0PXP76yhZf3+S6tUqK5CmgT1zWSYlIVCRBxULtNC6OGHaZtkknrdN3f/30m2DOiLtGfKZUw0Gr6J7ZlKBXNCH50xBgvAMT09CIqZZqrZxDOC/EUWV3EQ1ea4oC6NW5YCdM457isUkLFttLTeZFXopX2zZQ98XyCAktDolqaHgqXVHQf3FApoifPoCQRwmHRhtmIPFAr4Qz7vC9/LHLT03lxASW6+UKmo46tCKgEKD83PI2fbiJsmIpEIGpwr5jHIzwKIWRZWhULI2DbmJAUU5z71OfUe4FP64iAJUK59doF+PF+oKvd0qYRuy4IFCmMboGrwsutirXYPZ5tNZOR9F+cs9ycZ9fQCMjkWvYEAzk6nAQADwSA+uhXljseKRbp2AcxHdHejy7JwTn8/Ko6Dl6pV1WMujzUcDGIwFMKmRsPrJZfnoC0M1TiAokK7RaOoO46nHc8Yuk1T1Wfo/PbH9vT4FqHquhnzfVa79WDM87J3280Dc/1znYQ49CHUH+npafLQolEC5a1ZdzdwwQXk9d57L+VwGSPd7Vde8QPmq6/Sz/YAdO02wDSpD35zQTSd2awTMO4MFgjQ/X7mGbrO3Xaj65+f956HpS5W5ueBRx7xyIHm5+k+dnUBBxywfa5jZ7FlQG+z+VYLG+t17NdeDS1BlQkOcNfFI/PzpGomACMoctnH9fRQcVcbqEjyFq718m6s15FpNjHRbCJn27hpbg5F+aSK47qa9xk2DMRNE1XOMRAIYCgYxHSziazUQxcLAc6ISU72GRebTdxaKMButcAYQ28wqIqqpI2GQr6J+vlKBTnbxs3ZLEpiwSCvQQ+TS0dpRpPcVMMG/wR/ezaLTYILPGAYOKK7G6/UarDFfgGiqT0oHsdR3d349dzcgn1yUCRg72gUMdPEymCQpE41cJOpgKBhYDgcRrnVojY7sT3nHH2hEOKmialGAxXHwX1axT8YQ81x8LtcTlGh6pKlSvVMFPfJyIGUdZVjMxIMqgUeQLStf7kZalzdXqxU8LtsFlNCbc5kDLuLYrtkIIAz+/tx6YYNAEiIRxIMKeIdffzluInzVXl/xjAaCqEnEMCFQ0OqrU7WY+wWieDDYgGyOWOgSISuAqc/VTooiY9gepq8ZDkMo6P0WtIL6ODFeWeFLIAKrW69lUB9EUMKgCRXzzvPfz767+1psmVqfJy889WrSdp1c2Iw6TR9RmqL9/RsvYjurWaxGLGx1esUhfjc56jw8aGHqFhOTo+JBPWKL8XkPXMcyq+vX78M6MuALqzhuni+UsHGWg1PlMtouS56AgEFcD2WhYhpoioqeuUPGFNAUnNd3JTJ4LB4HPFAQHF1qzyklp+UE+1jpRJ2EzrlY/W68hpd7TOyKIlzEiM5rqcHG+p1fLS/H9dMT+NVIXgiKUDBiZb0tnweXCOslhM+YwzZZhNXTk97oW7GcFh3t7reDbUabpidxbgQPeGMqX1xEMDo4BZiDDHDUCpgOv96t5AAzdg2kbUIURbXcXBXoYCU7GkX26SF93pbNouny2XViqfngO8vFLBnNIqDEwl8ZngY356YwLzQIW9xTgscEQH5SDqNuWYTd+iJWsZwXDKJI5JJXLpxI14V3PlOG/owAC1oufj2NID4vAQ0iGt2QOB9/tCQIpcBKFLyQqVCxXpibNolWAHg9lwOUyIKII+1tlrFCuExv1QuwwEU9a7kYC+2WkhZFmKmibL2rAKefCrEvQxbFi4cHsZMo4EXRUrhzHQa716C3NgL1aqi3ZUFiOlg0JeS0D0xxgi8X3nFA+GPf5yKo6pVv2ylzvWut6FJoQ2Atnv44cUDerslElSUVizSa8bo9faQyjz/fPIqv/1tAqALLtgyickhh9DPbbeRd6sTpexM1tMDfOUr/ve6ujzNecaoBuLYYxe3P85pHOUCzzCIw+Css97Y894ZbRnQhW1qNHDZpk1ogcDqexMT2LOrC19esQIAcGoqhbLj4MlyWYVUXU760a4ENwFWcgINGgaipomKJhoCaIVFcjsBwA68UDoXLViAtxgIGAa6RRXxIYkE8raNeUHbqec5pXyqfNrlMfS8N2fUk+wACphKrRZytg2bc/xCgDkHFMjJPCwDkBDRCOm5ro5EsEckQvKbAjjkguTIZBIfTKXw3YkJZGwbOm13hDGcPziIn83MoCW26RYxSAlGalzkNYnXk40G1larWBMO428FTZfDOe4pFHDN9DRaACabTVw/O4uvrlq12RSDq91P/Z7oCy/ZyuVq46xAHlAopIf/P9Hf7ytoA4ANgk4WoMjGLtEoPjs4uKC1UEqdynvZcl3cPT+PA0TR4Q2ZDBVNivsi7+X6Wg09gQCOSibxYLGoogfzrZaS6JX3ssU5npifx025HGwhNDQYCuEvRka2WuwJcawrpqeRl+kezjESDOKEZBKviVLuQtBALBZWyljBIElRtk/eUu4U8EtxSmBvtTye7nYrFCjsusgMks8OO4xyun/4g+f9n3DC9msPsyzgr/+acvaFAnnflkWh983Z27Hga999t724zzRpkVCv0xiGQv4FWC7nMcoBtHDYXCfB282WAV2amPiYWP5zRnKhOdtGrygs+2h/vy/vyTnHV9etw4wM5RoGUcOKJ2mvWAyfHBzE1dPTYIz0w+tyEgZN6CHTREBUQytueAnEglxG9jSvDIfxBRG3qzgObspk8JSojpbEJjJ/3C62IYljdPEOvZgNAF6uVnHf/DymGg2sF6xz6lpBpCFgDHHTxP9auXKBTOqduZzaxkcaI+ycdBrj9TqmhViJHEMA+Nq73rXgloQMQ8nF6iF+OU43zs3hpmwWFwwM4P3JJBGlNBr4+fQ0WhrLnu26yNq2YoZrN6lhLs+XMQZHEvtoAC5zy8oz16Mo2m/1d1ty0OXEFidTMZxRwdr1c3MqDF+wbZQE7SvE8SEWcxcNDWEoFMKkkKHyjS/zV/af3tfn4wP4+tgYXhZeuFwoNTnH9XNzPm96ttnEFdPT+MqKFcho8lUMQNKyfP31P5qc9MCcMYQZw1HJJL4zMQGHU+3Hyp4wjjtutdIJP/hg4tVut055VOmdSy9MC6747NFHKZx71FH0v/7+pammHX/8tkmmbqutX0/MZ6JeFskk8H//77bpsEubnPQ40k3T837fjjY0RMIsa9cSTfDKlcT7Lu33vwfuust7Xk44gZ65RaxRd3pbBnRhHPADHicikhszGXy6w7ej2GoprwYg8IlaFr44OoqENum9Nx7HfqJf+vZcDrfmchSiFNt+IJnE+xIJ3JfPw2Ck7sVAniu3LCQtC3+/ahWibSW9d+XzeFAIS3PAk3IFVIW3LGFjAJgQ6ZDhf+lt6hXbB3R14fRUCj+YnPQ8YbEYMBnDJSMjSrjEai8x1ozB6+9mnKNm28g1m7h8chLTgmQnYBhoctIK/8HUFL48OopukR+WdmwyiYrj4M58Hrbr+qQ65W/HdXHV7Czl/6NRX/uZBLxNzSZ+OTeHC4TMacQwfOplXxgexr+PjeGVWo30xxlDQ45lu0uoAakCbddVERI9rTEvrluO74ZaDVdPTytOfmiLBYCIe66dncWTpZJKvTiguomAQLS8SCuETBN1EfmReWwwBkssgmqOQ9zr4v45WvRBP6bsN+eupwnPATxYLOJ60Ykhx+HkVMqfUxfPk7zXDc7x89lZBeacc4yPM0RfAL72NfrM5h4bXcdcDms47N1u/X3t8Or3vfdSP7PrAh/9KEUA3oo9yq0WCchIoh1Z0pDLbTugj40Rr7mkQu3pAf72b5eek97ZbLfdiMsd8HdJiClGPTN33kmLvKOPfnPPb0fYMqALY4zBMk20bBswDBiMYY9oFB/v8K1ouC6+PzGB9bUaGpzDYgx/OTqK1ZEIebFtFhDvndTbi5Lj4P75eXDhlQdE6PqS0VFcKjS0A6aJL42MqP7wTvuUNK+KYETMdIYIpQKe92bSB1QYXgK5yTyWMNMwEBTXLYFKDy0zEIhvCcgNRsxsKqQuzumuQgF3FYukIAcSXDmvvx83ZDIot1qYt238y4YN6A8G8U9r1ii2OMYYTuntxXHJJK6fm8OjxSLAGGx4CzDAX+Fe1xKvnFMB4opAAOf09eF32SzuyudxWCKB8/r71bVYhoEvjo7i6+PjOCOVwkAohH/dsAFVUWBocE9RzDIMleJQBXTiWhXIizH+8cyM0n6XtQ1qUSLOmYFETaqOg1/NzeGpUsnn8TPG8NH+fuzX1QWHc3xjbAyfGx7GpwYH8b2JCRWJCJomDBAp0Hn9/fhNJoP7CgXaDyeZVyaeG59SqTwPEeExGUNQLgJcTb5X/F83S4iR64WGrnwWVQSDg/Oty1F+6UvAP/0TkY5YFoHbJZd4cpcy9K7zsst8unwt86rXXw/09S2spN8R1miQ5xwOewsM6UnL65qfp7z6P//z0vXY63XaVqztAdDi4NvfpvF8Ky5q3ih76imilJULI8BPF6yP8TvFlgFd2JpwGOcPDOAnQl97dSSCTw8NLdB0BoAfTk5iXb3ueeiMUVHYVsiFLcPAGX19OKGnR0383cIjjVkWvrxihQLovq18s0/o7UXetnHf/LwCFcYYTkul8ESppPrAAwZptZuGgW+Pj6PJPTnWVZEI9o3FsL5exycHBjyvtcO3QHp5W7IjuruRbTYp9C73IxYYruyTBwHYtbOzSmaTgwhTOsluykWEkq4Fjf+rtdoCgJlpNvHdiQk0HUdFJQzGMN5s4v/dsAGu68LmHA8VCugyDJyRTquFUMyy8KXRUSQti7oM2seBMQyHwzg0kcCvBTc/tHEH1+oRhDVdl2oh9Ly8DNkzBlMsyNZWq/i7detgc0/GVj5XjBNX/fWiNqHuurh0bIzkc8X+RkIhfHFkBIwxhAwDQcbQdBzYAjn0cVL719MpErjFvi4aHsajghNeRnqMDvf+z4eH8e/j45iTLpG2P+86tw4qjkN5UIDanC66iKq7UylqHxNDobxZOcRtNA3q7y0wK7+p5rrAlVcCzz5LBVvSQ9Spb+W1dNKDX6xpXZ1qofNWGYPtZevWAT/6EV27DuhMK77UozuvZ3x3JlsGdGGSQnQ0FKJ8oGH4Que61V1XVR4zUPHS96em8P+sWKEINzZnESGU0cnac9JbspAApIrr4gkRazu2pwfH9/Tg4EQCDU14ejQUAgPwd6tWYUOjgd9ms/jz4WFETBMJ00TVdf35ZfEN0fPHLqBC45tjlguKtIPK5zKvEl7PqXPAA2/uSbh2TKQKi5mmWjCZ4tvbHo1gnKMmBEvUdQhAU547SPL07vl5xCwLJ2oiObJWwqf0pkUsDLFoM/UZQ14ntLoBPbqhkQHJ82GgCviD4nHcks3CAVDXQ+G6t88YGhKAxVjWZVsjY0hZFr4wMoK09tzdks3i0VLJo+Vti7ao3DtjVDMiwDdlWfjC6CiSgYBSTgMowtOJmOYn09MkzMK82gZdZpYxhpERhgsu2OxtBQB897uUVy4WafL9+c+Bv/kb7//6ULdzuevAqH/mreCVXXcdeZHNJvGQd3URC93FFwPf+hZ51ZxTMd/FFy/dO2+3rXyF3nYmuJPUNbcDuf5M3Hgj1Sq85z077nzfDFsGdM0sw8DKDi1E7ebLQYJC16f19qL/9X4jl2iyJeso0fsyHAohYpr40dSU8tCDhoG/HBnBUCiElYLhazQUQslxcPnkpFqU7NfVhXNFwZ8+JzBAfUPyzSYum5zEX4+OdpRJ/dP8PO4STbMMNMEfmkhgzraxXlQ9M+4Vkknwow38BXpbso/19+N/pqawqV5H0DBwbjqNuGnivzZtUpXf7TSvqjBQnIOUKe1oIgeswE/8Pd1s4uZs1j9r6OOkLwTE++0a4QaAVCCAzw4NYUxQx+rbSOIclbsGfCF7tVASn593HPzX+LiSnGWc6hKaerRD25ZxTT1NAjCos+Bzw8MYFguDQxIJTDQaeEDWaWifBYDLRMpJX6QZ+swKoC8QwGdGB5HeSjGSlD6Vm+pypqtWETvYr36lbo3PC9NND7FeeSVVPq9Zs+Vjby+76SbqtZYeZK0GXH01MdDttRe1rF11FV1bOr3lKvctWTgMfPrTwA9+QG1/coHwmc+8vcPt+v3v9Lvt649q9e0ftQCWAX2L9nCxiDtyOTAAB2uSp6bwOCXJDOccv8/ncbfIWSYsC5eMjCD8Juj79QWDC8Lz846DnMYw9p2JCcRNE38xMoK4ZWFNJILHikXkpCQqgEeLRcRNE6emUlBCL8zrOQfIS5ttNPCDyUn8ZVvx3zPlMm7KZFAVHrIBUmk7ra8PVcfBVdPT2CS52EW4+QM9Pbgrn0edU7/4JwYG8Gqthvvn5/HZLZTpjobD+MTAAEqtFjhIDzxoGDiupwfXzc6iqfK3nkKafj0SmDZHARsXbGzXzs6iLvYle8/LgphnwbbavmVPtuu6sDTwla5j0DCwSziMTfW6eo4kKEpvWHno0oMWSOZbBAEA58i3Wj7FNL3QUV+UuGLfgLegAmPoMk18vL9fybzKMTg1lcJesZjarx6ZKbRa4NxjMYR+fgB6LQvnDw4qadpOdt11VKk8N+fPL7daFE695BICp2OOIdBbt86rXm6fxPWcOmPk/f7kJ8AXvkBkMm+GPfIIFaidfTYtUHR1N4Da4ySo7L8/1QPsvffiZUBvuIGu5bDD/O/vtx9w4YXe4iEU8jju387GmKfGJ8dwzRpqUXvmGe8z8tl4Oy9wpC0D+mbs2XIZt2Qyinb03nwezwnXYbrZ9eLmnAAAIABJREFUVN7Isd3d6LUs3JLLoSzAbM628dPpaVw4NLTFIrLtYdfNziLTbBLxCwg8ZxsNZA0D3xofRyoYVGCpy4FWHQf3FwroMk0cm0wi02xiU6OxwGvkAMbqdfx0agqfHx5Wi5ay4yCvLRDASGt7fa2G93V3IyaKtlzmqaCtrdXwwb4+hA0DFdfFHbkcqo6DecfBN8bGsOsWmMo6cZ/vE4vhV4zBkC1fhqEY1Dq10T1SLOKlahWnpFJKIrbhuvihaMeSNLkGPJBtlxqV5vNUAVVgpnPKx0wTp/b2IhUIKMCVY7VAWEa8lmI88l6abcAJzn3XqJ9XO4mRvA7OuUoFhE0TZ/T14eAOTCpysXhPoYDHSiVq+gXVbxht+9fTDl2miTPTaYQNA9fMzHQsLL31VuDxxwn4dDCWveoHHkgsYwB5tQcd5PWii8MBINrXAw8EHniAGNh0UM9klu6V3X475bwB6v/eZ5+tb7N2LUnElkpUpLZxoyc2oy843v9+KDpZwyAgXoz98Y80VnNzBNb3309yohK0GfNU594pNjhILHsTE9ThMDhInQ2Dg/S8NBpEXiQjOocc8uYo6u1oWwb0zVi+1UJW5AcZ5yg4DgqS6UKzPWIxPFMu0+QvJmPHdSmcugTjnMhc5gTDWcQw8LGBAeprX4KN1+vUty3O0xS/Xc6xsV5HSQvFqtC0+My842Cm2cRRySROTaXwm0wGmWZTUZ/KCTtgGFgTiVArlXcBXgsVPC9tQEQPpBcqf8A5pppNfDYex0AohPXVKm7NZlXL0/paDeGtLIY457hmZgYZ20bIMHBmXx9OS6XQEjP6XLOJR0ol75jaYgIAsraNjG3jxkwGEcPAqnAYv5iZwfPlMmQHtiEL2bgmq6rtD4BSPdNzyLJ2QI5HxDBwUk8PTujtVRGP9lQAOMeukQiqrqsoXxn3SHpM7dlTvP5iH7KTQc/9MxCA7xWNYk0kAg7gj7kcsfmJfVqMaHZVuLzNHisW8cd8niiEhf0um6XcuV5YJ44ZMQwckkjgiXIZxVYLecdBYcrEmuk+H0HK5CQU2Uw7k5xpUq65WASuvdYr+urqIhCU3hdAwBmJAKeeSnrqUp2sU0het3XrCLzlsO25JxXkPfCA1wJ100207w4UCT6bn6caAN0TX7WKQv5jY97n9thj28hvpqa8/ZdKtFD5zW+Aj32MjvNOtGiUigyzWcqNp9PEcSDt5JNJ415+TQ44gNr5OtncHHVGyAzcypUk5rMz2jKgb8YUAOivNdO9tp5AAKZhwHAc5f0s+XiM4aVyGVOiDcgA5b/P6+9fUuj+kEQCGdH/LCd41XrF2vLU2ownPcaXq1VcMTUFAIrcRH0GXkg532rB4RxzzSZuy+WQE+cNeLnU0VBI0ZoqMJehcAEk+oJFH2vl6W/BbhatbDXHATMMBBjDxwYGEBXj9VSphEf05GzbcaRN1OvI2jZWhEJYW62iJY8tXUdOsrW65zsQDGKvaBSv1mpK2QwQix7DABcMfnIbizEckUx6YA5gl3AY745E8HK1CgBYEQ7jpN5e3C/IfVREA/CEVrR7qZfxctP0k8xwry9+12hUpYsemZ9X9LyLqVuYaDQwJ++t8MInNCU9CelBw8BB8ThSwSB2j0TwrU2b4IgoxQuFGrCePvfEE+QBS+1zeWvkpRgGgTlAefNnnvHavLq7qbdaB+tslvqMP/QhmrDn5rz2tve+F9BqHn2WzZJ8qTy+DPvnct5nNm2i11sD9E4LiJERWiRIFTdg2+lpJamObuPjxDL3TgV0aakULebabY89Fpd2KBaBX/7Si8oAO3eufRnQN2Mc8HlZ+mSpeLuFN3liTw/uLxQ2T0ayWNNmthYoL31WX9+SAP3IZBJlx8Gd2SxKGo97+xkNBIPYNRLBOqmIIWy60cCkYDNTuVrpgYrXNucq574mEsEjmkRsJ81xgMLh080m8oKf3GAMxyaTCnzbw8hMo9DdnD1dLqMm5Fw5J9nOuGXhzL4+FB0Hf5KtV5rH7DPxek04rCIJuinQBHxh6l7LwrHJJFKBAF4VjHr6WKl9a1607br4fS6HM/r6VBpmRTiMk1Mp1FwX42JR8EKlgphposeykNMU9NqL8CCvS/zWlfgk+BqcL+DAPzAex2g4jOcrFewaiSBpWR1bM33Hgfb8aBEL+Z7BGPbv6sLZ/f2IWxbx4sO7lzxuY65vHr/8ZTdee41C0lvynm0buOUWqhDXW9PKZfJ+221igjzhffYh0Jeh+w9+sDMn+8wM8Nhj3pByTv3v2uV1Gu6tWvs2Bx74xobC9XPb1ilm2fxWLgMvvOBFiQyDFkoPPggcfviOPrul2zKgb8b6AwEMh0KYajQAxtAbCBB3uwhXuoxhl1AI3aaJ+woF1AWRiR72XIrpDF6qyAgLgXgxdkoqhT/NzxOgi31xxhAA8O5oFAwEJofG43i1WvWpiOmel877rZ8XZwy1Vgv3FArICn14vTUKYh+b6nWsr9WwJhLBCb29KDsOxut1OJw03O+bn0d3uYyQaWLvaBRrwmECSLFoyto2ni2Xsa9QvntOALYqtNJC3i6AhuPgT/PzODKRwJ2FAp4uldT9UB4387dwAZTzfa5SwVOlErW3yfvB/MVorusq4pWy4+DxUom8c+8mQtLG+lTHOEcDwH2FAhKWhRN6etQCYPdoFPvGYhgTkrrjjQbeHYtht0gEZbEgsx0Hr4pUinz24LoUCdKvSQv9K0GbtmfojHQatuvi+rk5nNzbq1r1OtnGeh0bhGCQvrDV4zYMVDR5ciqloi26dj0YA7pszPaUMPsbL96sf0U4B3bZhULL9ToVl6n9MyzYpn3tJP9eLH3r9DR5ZHrvsm5yv4sFzlSKvO/xce+1kBZ4Q6ydKGXZ3lhrf45sW5WL7HS2DOibsT1iMZziuqpy/T1dXUqRTd7/D/X1YTgYxPcmJlARlcYS3GT1+1JMgrrsW1asW9tgq8JhhAwDYwIgV4RCSAYCPoa0nkAA6UAAcxqRCoOn7qVrYEtT0QnG0HBdbNJyvUbbZ9fVargtm8U+sRg4KHqQCgTQdBz8w2uv4W7R4hYxTbipFFaEw3i1XofsuZ5qNvGnYhH7dnVhfa2GW7JZdFsW9ojFsHc0qgRMmOC9Z4zB5hx3FQq4v1DwqsahgZDmYUrQfapcxhOlkmoxU6FyDdQlGQ+HaF/LZAB4iwK1+BFkQ5JcRYnUcI6K6+IP+TyO6e5W7IEzzSbGtBA2ALxUqeDD6TQ+IOLFDuf4h3XrUHUcjzJW27dUWtMXVY74X69loeQ4uCefBxjD/rEYkoHAgkK1iuPg0WJRpZLAifBmbbWq0ko9mmjOZKOBsrjG0XAYkbZ6Cjneim5YFGHo3i/nlPuMxahtK5Oh9iJgoSeqh7Tl+/J1by8xwy3W4nEqnpqe9oNkTw+Bp2Rd6+9fXM57zRriCr/tNtrfPvsARxyx+PPZmg0N0THm5jxVuP7+1y848uKLXmRi//23rPz2drVQiBZj69d7kZ2+vs5h/J3BlgF9C/aeeBzvafvWnNwmSNxyXQwEg8gKFTE5eSt5zSXYgCC1kdal50WXaJ8ZHkbDdfGdTZtQ5xxfGB5e4I3tFYvhyGQSN8zOEhAIWVNFKSsWGLqkq5xJLcbwrkgExyaT+P7kpEpFqDCvqHZ6plLBs5UKOOc4utHA6X19eLVWQ8NxFGVr1XFw/ews5YdBixhZgZ+3bTxWLOKB+Xml4PVkqYSjkknyWJknSsNA1fp35/PqtaqF0ADPJ7gCbSGlL8SYv9pcLnYcbV9ye924/BFjCXl8eM+Ffo831uvYILz8hTx5ZM9XKioC5DAv7y3vi2RxixgGVdBr2+4ejeLlahV31uvgjCHT04NTUykfuVHTdXF3LoebBMmNJb1xuUgxDOwTi+E8sQh4tVrFz6anUWy1YAHYNxZTCnnyvJRQkD5Orkf6wjkByGmnUQX7P/6jRywjhkz95tx7X74n34/HgeOOoyrmxdqaNdT//etfe/3JPT10Hs0m8PTT9LnTT6fCqsXYPvssriJ+W+yYY+jn+9+nNAQAnHTS1nP7W7KNG+n6Bekg5uYIxF6PQMz2NM7pvtg2UQMfcMDSUyKdLJUCPvlJapNstej51PS3djpbBvTXaZZh4OKREXx9bAwb63W0OBUC9Qg9dN0Kto1pkUMGaDIeDoWQsCxIPvc30kKGga8sIvbnAKq4SgK3y6g/OWgYyNq24iDnIAAZCYdx8cgIJhoNdJkmSsLLV+F5TS3MEUB6V6GANZEIbshkUBRa3cx1lfcL10WLMVjco1GdaDZx9cwMqpI6VpzD3YUCnae4BslPr1IGzGvjkoAnvUWI69VpUKV4iTwGE+Mic/DSzE770q5b/k/2fMtx4IB6LnTbIxrFntEoHhJ1CPJ69OjO7bkcijKlA+F9MwbDMBAXYx8wDOzb1bWgf/+yTZvUgoG7Lm7N5RAzTRzf04OAEOx5YH4eN2azqk1PeveqJkKkEKYbDRRaLVw/N4epZlPJsd6UySBhmjhCKN4FGEPCspC1bZjKQ+dg3K9tfsQRBKKAJ5EqbskCT17+6CxwsRhw5JGL19HWTQLw/fcTqH38497/PvShpe9ve9vsLOV75djJ9MS2KojdeiuBulws3XorFZFtr0XJ67Xnnwd++EOqoYhGqe/+japPGBkB/s//eWP2taPtHdBqv/0tYpr4qxUrMBAMIm5ZGAmF8Fejo762rqrj4OZsFt8cH8d/bdqE/xofx3+Oj+PWbBZjtVpHHvPtaU3XxcZ6HTXXRVgUewHw8rGui9P6+nDpu96FtAQhAVph08QlIyPI2jbChoEP9PQot0mpjsEr2tKTVJxzdEk2CECFvRU4tyVK94hGcaisbNJAU3rWrgBk/diuOH8Zupdes29bcayIaS7kChBAp+oYRBKTy59OA6r9T9enjwnJ0YRlYTgUwpe056LuOLg9m8WDgo9fdiFETNNXqPbXo6NImia4XijIOeKGgZOFLkA6EMCJySSyto1Sq4WNtRrG6nVFsiO3a4n8+QbRVvlUsYifT08T77z8nARy7onSVB0Hv5idxX+OjXktmeKeNTnH1TMzqrVtl0gEFw4NoduyEBXiQ+23NhJZCEY6cMu8sS6+on8OIPDpJMW6FDviCD+Yv1Xt5ps9XnvOqWhQVulvq7WnLnjHB3vHW7MJ/PjHtKDhnH7/9KfEvrdsflv20N8g67Ys/IvGM6m3JwHA/YUC7isUVOhVfpPuzudxV6GACwYH8b5taVLdBnM5x72FAq6bmcGaSASHJxK4q1BQuXDW9s2WV6I8NxDxzjUzM5B12DJXLSdvB+R9y23l/yqui08NDOD7k5PItVqoi1a/9kUAawNOxYcuPysWEL5zFUDEtePq2+uhb3BibDuttxePlkp4VcwOqoZBm+XkNrLXW/+/WrBoLqWMEMRNE18cHVWte4D/uXikVMLvczkN8GjsDxE1ArbrKvCPmiYyon1Mgm3RtnHd7CwYY5hsNPAvGzdir1gM6UAA9xYK1PamXbusB5CpjEyziR/NzBBvgRbKl/UFkpzIAfWjqzSFdv0qFdP2vO8Ri+E/dt0VDxeLuHp6GrbJgUgLbo2mnCOOoFC5NCmfKm+nHLL2SVv+3zQ9QZd3ir2Ra34tk7RTmL740Ospls1vyx76G2gmY+qn3XwqV4z5gM/lXAHbm2GPFou4Xkh7rqvVcI8oINPD0ZYgUQGISCYgXgcMAw3XxVUzM2hyktjkuneneXWmaSqRD1lk94uZGfzbxo04b2AA/3vVKuXB6aDGtX04rkv98AJcLMNAsEMbH9O2ld94CyKXK/ndtf9bpokTentxXG+vF0kRs4aeO+fa7OFq5yWv02QkXRpkDEHDUNcZZAyfGRrCbtFox+dC3nN1zWLcwRgeKJXwjxs34q58HuVWCw3XJS/dsrxFjgBoPUrBOceLlYqSTXXUR/35ftt18ZPpabwiFOt8Y6+F9uU+ZT+8PK6vYFKMR6BDm6HJGA5LJKhVb7QG4/gZWBYxwQUCdLm1GoVRv/QlCn0GAtRq9jd/A3z5y55gie61A8SH/tGPLngM3tbWVnf4ukzeA9OknLS8H29la0+/LNtCW/bQ3yRTOUkAcF04jPkY2G7IZNAbDGJ/0aK1vezlahU/nZ4mz1qcj2S5Azxv65jubhyeSKDlurhkZAQ1UWB33sAAwoaB/xofJ41tCcaM2uJUMRmAQ+NxVF0XTwl+T3ksF0CYMa8HXQ2Sf+nNAbxQrXqCLpzj6O5unNDbi8unppRX7W3uhf1TloVT+vrwi9lZfwQAADjH++JxnJJKwWQMFw0P4z9EKFnqgrdHABTlK/MqyU3GcEg8jg/19anZ5gdTU3itVsMnBgZwwBbu5ROlEm4SlfIcgoZX5Kpb4nxvymRwczYLzjkuGBrykR25kuNdstdpY6Qb18BXeeCgEPpPp6bQkMAMKPIcfXGnquq1KASDv4AvwBg+PzzcUWnQYAzH9vSg7Dj4vdnCgQeSlGg4TDnz//kf4OWXqTDpr/6KQDsQoCI1SeWqrSXAGIFQMPj61cl2JtPHoRPRzFLtvPOAM8+k8d9/f2JZ2xyT2lvBTNNPD/wmyGTslLYM6G+SHZVMImPbuGd+XvGDQ3CNuwBO6e3FXm9CiSkH5Tw5PJBiAPaMxRA1TTwumNUeKJUwHA7j0VIJr9VqcEGtTVfNzOCCgQGvsE/MLKOhEC4UBVkSELotCy3OUXUcvFyteiCgAbdeIKiAhHmV5I4OJIzhwXIZQ5EIPjc8jG+OjysSHPkZ6VHCMHB4IoGa6+LGuTkF0HRQ7stTJywLfz4ygm+MjyNj29TDLcLf3UKkRIa29Qr3fbq68JH+fp9IzYVDQ2iIzoctdTns19WFE3t78TvJM6ovIMRvW3jfjuviiulpVITOuysWYDo/gCnAWIG2Nmaq0E7/n+sSmOspDGE6X708t/Y+fn1xAMbQvYUZNmAYODmVwsHvd5Fg1Ap21VVUsV2pELBfeSVN1JZFwN7TQ+1En/oU5U8lR9KqVfTedl73vuXsrLOAE08kxbbddycA7kCRv2iLxejns5+lToG3cvoiGAS+8hVi7fvJT+j+p1JeWmbZPFsG9DfJIqaJuGV5uWEJQOL/vYGAnxt9O9macBhn9/Xhl3Nzvgl6nVD+kp5YzXHwq7k51KXMqPD0sraNH01N+eU5Qd7q5nTSA1q+VRZSfXfTJjDGfBKmC0Bd/kMD7JrjoOm6SAeDKoStFM60Y+ZtG9+ZnMRfiOjC7/N5f3Fe2zkOhEL4/NAQvjsxgYLg8OcASo6DW7JZ1ZKnFgacI2QYPjAH0FFWtpMFDQMn9Pai6rqqzU6F+TWvWVpZ6AjI8VF1BloUQUUPtLHjbb99YyAWLXpdgK/mQTu+igC07Ytr92BLFjVNRJMe6Fcq1KYmw6dS+pNz0kf/yleofeiQQ2jyFpeFWIxC8zuz1evA179Ov7/85cX10Hd308/55xMAd3XRIudf/5X43b/4xW1TlVtK//6OtKEh4g74wheo7fCtnh7YUbYM6G+SPVws4oFCwTdpHpxI4MB4HAzAu96k5WbAMHB0Tw8qGsitiUTQEwjgyVKJQu9iopatYqboFZIQbnOOM/r6cFsup3jPO9UNSFORAC2PLAHKpzomPiMjGBJAVC9z2zEMkSdXIKMtHDiA9bUavjY2hobrqtY16cXqYiQ/mZrChnod56bTCDCm8v4yvFwRhC6G3I771c221WKmib5AwBfOpwvlKlSuqt9lbt40cU46jabr4oZMhsRsDGOBVy3HWy6GVA2HYaiiN1+KQ7smJtJBajzbxlWek6zKN9s9+q3Y735HYXbT9Bc7yTx5qeQVgFnW4nvBdwZrtYDLLiOeeNcFLr+cwHixJDF6V+L3vkctdzJ1cckl3uJnR9iVVxJ3PeeUQtlWjffNGWOvr/f+nWDLgP4mWbHVQlEAg5wY08EgDlwi3dNz5TJeqlZx9utgP4iaJk7o6VH85f3BIF6oVHxeFgN8fdkcXtWzZRjYLRrFHfk8ea0A5ppNXD87i3M6nNcHentRaLUw2Wgo8JVjAAEeMkR/ejrtA8pnKxU8Xy77PMZ7CwU8US6T4pf2PtfAlnGOFqBaqRTAyb859U+vrVYx1WyiYtu4bm4OhyUSuHd+HhXB+CaRxmhbaADAK7Ua/pDP4/htTD4+Vy7jfpGC0VXvlB66AGITXuTEEv3mVceBmcl4ZDxaPUN7FbrM0UthF8C7l3p6Q46xIa7Xd07wLxAYiIDGFWmBq2dmcNHw8II++3a75x7q/RYcQdCGmM7LfP261ZxTz7KUG73ggh0LdNI4B664gtrPJHf42BiFkT//+aX1lF91FfDSS95+JiZoPxdfvGPSEb/5DaVQqlWKHFxzDYXzd/Zoys5my4D+JpruhQJ+L3Extq5Ww02ZDIqtFmZsG3tGozhuG8EkGQjgSI3r8cVKRYXE5VlJj1xPEzDO0XRd/DaToTA0CATqrosJnddcs3dHo4gL1juZnwZjCBgGjunuRpcIW3dZFo7s7gZjDC9UKrg7n1etWjqgZ4TsqSJ3YcwXepZhZENU0OgFbmpfjGGi0cCGWo1eGwamGg0EGUNLyM/6RE90kBQRi1KrhZlFSDOVWi1cPTODuqi27wsG8fH+fuRbLcxpCw5dDU8fc0dsF2BUrJiwLFRFq5kCX9DCLB0I4AXJn6qF1yXpDte8czVe8rX4uy8YxIpQCM+Uy/6oh/ZsqO3EfjfUavjh5CR6AgH82cAAMraNX2cy6jPvikRwSiqF2VkKtbc/+rLgyTSBo456fTSk111HXO31Oh3nmmsI1F8vVeq22i23ELXoyScTgNu2t2hxXY+tbSk2Pk770QIs2LTJI+d5s21y0htvwyDiG1mzWixS7l9OD/39O0fv/85oy4D+JpleRKS3/CzWJhsN3Dg3p0DzxXLZz5/9es+vPewrTZv0TcZwSCIBBuqh1qunDcYw22zi7nwex/b04PFSCc+Wy7QLCE+Z+XvELcZwdE8PBkIhZJpN5TGbjCFn24r8RBVlifNR4K6BbPu1SNCW3rgELH3EdW9bXucGwX2v7pPmwepFhBzASCiEfWOxLY5rXdDaPl0qwWEMcF3ETBNxw0CPoGltvxb9PBmA98bj6AkGYTGGD/T0IGQY6LEsHBSP4zFxH3qCQbw7GiXSF3ne7fvTAVyMa/sYJiwL+8RipP6mfda32BD7U2yAoMXfy5UKgpaFluui7rrEAy9C/DPNJhEKIdnxEZPrJtMEDjts2xnQAOCVVzzwcF3yiHeEJOaDDwJr11IYOp+ncxJfCWWhEKl6LbVi//DDqUhM8s4Db528Muceo1+tRlrjUgYXoAVMdzep4S3VymWS1ZW1sOk0VesvG9kyoL9JtjIcxqpwWLVGjYRCWL2EvHnWtvFqve7zkCYbDTxVKuGAN8D1WBMO49BEAhvrdaKn5Ry7RqPoEypzHATAH+rrQ91x8ES5THSpYnvOOeYdB/cUCphrNvGaUA+D9v/dIhFsqtdRFnzx+8ZiiAnq0t9kMnhckJcYbWAkdqByyWvCYfQHAni2VEJJ6I5DA3EZYlabytdyn5zj1VoNVRmvFO8BVLkeMgw8XS57evBtM6VcyAyHQkoJ7uFiERvrdRyaSHhqcKB6g+cqFUWryhlDzXXxbKWCjw8MYHUkgg16+51W4CbP67Dubuzfdo8TloUP9fXBEtdrMoaZZnMBEOvWcb4Xefn9YjFYjGocMnIx1ZaCab9+FcFQH6LoxpMiRWLK62EMuVYLfywUELJMAHHfpQIEAHvuSdXtnSRPl2J6Tn5Hgtwrr5BMq7T16ymvvGGD570GAqQSZy1xJj76aOCuuzyxFsui9rPXsxB6o0wf82aToiXyq2QYVBD53HPbBui1GvDEE17KJh6n52WxSntvd1sG9DfJdo9G8YHeXjwserLfE49j7614d7rpNKYATaZN10VlK7G6llD4kuHemNDybg/379vVhX27unBfoYDnRfj96GQSe2zmHHePRvGcmLj1XPNcs4l7mk2f1yc93cMSCYyHQpizbViGgbPSaXRZFsbrdTxZKsGBFxqWoX9fVTeAoWAQp6ZS2CsWwy+FdG1NI0NRx2zbTu+j5oxho5gRXG07A8BJvb1YEQ7DFPzxddfFOhGW171mXdnsuXIZd+ZymGw2kbNtjIZCeF93N1KBgHdezJOolXdstQhD31coYM62MSN5/jU04tpx2q0vGMT5okrqgUIBD83Pq0iEGHhP073DfrhwpUKGgU8MDlLbYrGIH05PK44Eee5M30aOozZ2+rX56hq0ezndaICZZXAeV5P+6KgXXj/9dHr9ek3PyYvD7xDrdA5HH00hZyE0iEiEQO+3vyVwOvroxe9/r73IQ+WcgFz29+8I22038rxlxGDNGgJaed1ymnJd4mLflkJH2wbuu89LKzBGHvsjj2xfQH/pJa/YTz5br7dtcHvZdgN0xtiPAZwGYJZzvo94rxfAtQB2AbABwEc553lGs/9/AzgVQBXABZzzJ7bXue0oOyAe32ZvOmFZGAqFMNloqF7kVZEI3r8Vuti6EOQot1pgjARXIozh8M0kKY9MJn259U7WbVk4t78fJcfBa9UqcYy7rgrLyhCsno/nnOQ4V0UiSAeDOCSRQEz0LqsQbhsAQduX/EzKstAr3Jmz0mnMt1p4vFSCLUVetG9dyDSxKhxG2XEwY9teyJlzH+EKxL5dAE9XKhgNh/FZ0QP0Wq2Gr2/c6NM352IfWdvGH/N5PFoqYVOzCcY5nqlU8HS5jBWhEFKBACzDwC7hMJ6vVJRSmzyvuWYTc80m9ohGYVcqmGo0fPrr8vyeK5eRtW0A5Ikf3t2NoJZuydk21larfhpaeZ762ALeAkECtvZbmbiXch/y/qUDAdQ4R7nVwmAwqDQANjYaaIoFo46dDmumzBnuAAAgAElEQVSrlgfgcvFbTPCHHbZt4ipbsl12Ic9340YK865aRa93lMmUgiRF+cQn6H3bprD8PfcAd95JxWyxmCdYszU799ztd85LtRNPpO6EDRvoWs8+mwCvXKbxf/FFr4Cvuxv48IeXfoxqlcaq0aD9OI5XezE7C7zwgvfZlStpUfFG2BNPAA884H1FOKd2v3cUoAP4KYDvAPiZ9t5XAfyBc34pY+yr4vX/AnAKgN3Ez6EAvid+L5uwXSIRnJ1O41dzcwDIQ0xvZZbinOOZcpmYx8TEXWq18LtsFod1dy+5KE+3dDCIs9Np/PvYmAfmwuPtMk2YjKHYavkm+IeKRTwgNLfLjoOTensRFJShEjAdzlUYWQKSvkB4ulLBrtEoBkMhGIzh/MFBvFSpICeoc10B1CHGsGcshi+MjOCP+TyuFcLPej5ZgSZjcMTfd+ZyiBkGTtQoYV3GwB3HRwPLGMP6Wg3rqlXlwUpRGAnIT5dKqLsu9onFUHZdNEQSsT8YxOmpFH6by+H+fF55+3HLQrfGVcABzDQa+GOhANk2GDYMNDlHt2kibBjYPx7HTLOJdZLGVb//4pp6RBFdzXGQCgSQt23Y4n+GYWCwjQRHRgmkxrw8n+FQCDkh/rJrJILT+/rwSrWKn01NQVQ7+AsYOUfYNNEfDKLiOJhvtYCuFpyeBlg29IaE1zvZn/0ZgeVll1F491Of2j7H2Zr19ZHHXKl4r/WAV7FIBXwyuzE/T576YgH9rWZnnbXwva4uGv/LL/euc1uAkHMK3UvvXG9rTCSAO+4g0AXoGHvsAZxzjr/Nb3PWaHiSudKf6O312u7kgkz+vJVtuwE65/xextgubW+fAeAY8fcVAP4IAvQzAPyMk0vyJ8ZYkjE2xDmf2l7ntzPanrEY/mEJYfqG6+K6mRlURJ5Z+nRvRCFsy3UVg5z0WOG6iBgGDkkk0GWaFAIGMN9qoS6+gY6gvb05k8H+XV1YKWKEUm1NssOp4jMJKJtZfMjFgO88AIQtC8eJSMP/3965RzdW3ff++ztHT8uSbcny24NnAilvGKAFArmhEMIAEwjlcSE0IZACzeUR7qW9adI/8li9t2QlbSCQtkAutyGhEDppCLAgXFLe0CQ8JuWZMMBkGL8fsmRZlizpnH3/2HsfbXlmPLYZWx7591lLS9KRdLS1Zeu39+/x/TXYtherN5O8qnap6nZRCPxsYgJHR6PoCgahtdLLRLBhuJOFkZynx6HeWwiBHYUCnkqnMVksImzbuKyzE8fHYp7h/FUmg6cmJytJewDWBYO4uK0NrSpD6v1CAbcPDCBXLMrxqmSze4aHvVasfxuJ4JBIBJsSCfxoeBguVTLz9fUhDQ3IOA4Gi0Vc0taGR1MpjKskxQbLwlVdXQjbNgqOg/dnZ+Xn0DtuI2SwdXram6cn02mEbBvPZTLIqFa4ejFjq8Vd3O9HXyiEL3R349l0Gg+MjSHXnYM4Kg3xi3acdBJw7LGL/MNbIH6/1IffE7mc3E0SSZf/UkRZ9sbmzXLX+sYb0hBcdJF0TQPy/rZtcpdpRs2KRelZOOCAfT+eWhGPA1/+8gc7h06u0wJERNKYH3igXADddVdFEtd15W79+eelp2BvDAzIMsdSSSZl6gXBjTfKx82WvyvcFHPRrHQMvd0w0sMA9FqtG8BO43n96tguBp2IrgJwFQCsW0Cv77VOLBBAOp/3DJBFhNg+EEIeLhbx81QKjtYUV4ZqQziMi1Qt+lmq+PeWnTvxutqmaJevADBQKKBXG03I3bmljLcnj2sGrtTr5i6StUGHYfgnSyXcNTSEG3p7cUQkgveiUTwxOVnVccyM++oe8NDHzfdTRl1nuWtPgF5weLt9NZ6wZeHB8XGvz3zOcfCDoSFEfb6qvIkqfXQivJ3P4z+mpvBJJd/18MQExsrlSjczqgjEaEnYgdlZrA+HvYUNaW+JEVZ4Jp3Gpzs6cK0KUOtEvvcLBbiui6lyGU0+H96emcFD4+Oe50a7y/WiRSdH6nl5aHx8l7AFQWrRJ3w+XNvdjV61YPtoczNyjoMn0mkgaAFNtYv3FosyoezBB+X9gw8GLr9cGp59zSWX7P749LQUYplbjz82Jsd13XX7fiz7OzpsoW/HYnLR9uKLlWOLzZ1wXVlyp3Su9vgac4fe0FDbEM581KzbmtqNL9qBIYS4QwhxnBDiuGQyuQwjqx9Cto3/0dNTpSveFQzixt7eRbvbJ0oljKnsdwDoCYVwdVeX7JYGucv0QUrcmmTLZZSFqIqvEoCS4+Du4WGZgEeEiG0jovTIdTlZVXKXet+gZVXFjydKJWj9dq/0TkhJ09FiEbcPDuLxVApPTU7CkzoVqqsYgIDaUerSOBJClgPOyfD2FhdCwK/GIIxjemwNto1Pt7VVwiHKWOddF/eoRDtAxsJDau70vBzb2IjNhgLKn3d1ocf47szPCUjPx49GRlB03Yr3wZgDfQnZNnxzvu/f5/P41vvv4+s7duBbO3bITHuqlPmZsW/vc+iLeUwvSPS41BjT5TJ+ODJS9Z5nxOO4af163LS5FTfdVLvM5NdfrxhzIlla9sADKzsGIaT7fe6/YWcncM01KzuWyUkZgx4bkxchKrdV/6CaYziKqi6A3KmbWvRE0uAuRJ9+aAjYsqVSUqdfbxIKye+qoUFeX3DB8nmWPigrvUMf0a50IuoEMKqODwDoNZ7Xo44xH5Bmvx/X9fTgW++/D79l4b/39iK0yB36VLmMb73/PjLlMq7r6cGhapf5B5EIrujsxJ2Dg7CIcGA4jM91dHivm1Xu4W26MYsypFrRbMZ1cfvgIL62fj2+uWED3szlcMfgYFXMH0BF+hTAHzc346MqEdAVArf29yOtAmvegkEla0EIDBQKGJqdhdmNTAvN+ABsTiTw2swM3lVj9FkWPtPRgZhtY7pcrmSHG7v5DaEQDmxowGOplDyfZcEGcHY8jlNaWhAkwgPj41Whgoht4/rubq/D3MbGRgy1tOBBJb6i8wdKQrZeBQzhIfVrZjZagXJpX9/djf/IZPDg2FglCY4qNfeCCKe3tFQlTxYcB7cNDGBKzduU4+AfBgZkGRyMtqtzfkW1N8JsBawNuWO8t02EVr8fX5yTsm62ya015k5Mu2lXklhMKqndfHOlhE2vwVagpYPHzIyUkH3vPXnf75eyrffdJ8MSiQTwpS+tDqW9YLCSWEdUqd0/7jiZf/DTn1bq3088cWFlcd3dwNVXA3/3d9WiPKYuwHnnyfAJUP2+q5GVNugPArgMwE3q+mfG8WuJ6D7IZLgMx8/3HT2hEL6+fj0AaeAXQ8l1cfPOnRgtFuEIgdsHB/EXPT3oVTX0G6NR3PShD0FAKpmZu+efjI7i1VzOa2wiYJR7qV+uso552zYOj0RwUVsb7hkeruxKAdjKMJJ6D9O7YCysK+c3DY0ONajneK1GiXBqSwtOi8dxUnOzN46pUgnfGxrCZKmEmM+Hi5NJWEQyQx1S195vWdgUjyNTLuOFdBonxmLY3NqKmM8n+4ILAduyYFkWyspSCMi6bF3b/nI2i0dTKc8IWupYs8+HP1Ehi5IhP2vqr+vPYBHh3UIB942Oei1wtbyrJ4SjvpO5HpmyTt5TczVaLOLnExM4v60NW8bGKnX7enElKvKw+ru0qCIX61OfW7+LRYTQSlqmRaAjOOYub6WZmpL661rUD9g3bVEXy513Atu3V3rNl8tSnlZrOo2OykXHX/+1jC/XytUciUit+m9/Wy40mpulBr6u3z/llOpdczi88BarwaB8rlYq7OuTIRiNz7d4nYBasZxla/dCJsC1ElE/gK9CGvL7iejzAHYAuEg9/RHIkrV3IMvWLt/lhMwHYrGGXHPbwAB2FAreD3emXMatg4P48rp1aFGNRfZ07llRSXQzYyt692i6yd/N5/Hd/n5pmJQhMBPj9vYL7DUUMd6DgErjEsMgQQicGIvhrEQCfmWgAWlAv7NzpxdamCiV8MD4OC5MJnHf6Cg6g0F8obsbIcvC05kMXp6eBhHh5elpdIRCOEMFYYkIX+zpQbZcxi1qJ5x3XXxtx46qGFfJiMsTEcpqh665fXBQlinqmL0KB1yQTOLnqRS+0N2NbLmMovISeNLCloWjGxtxTiIBQeSV+c2ZsKqwBhFhsFTCS9ksvtbX58XhfzE5ieenpqqTAc3FFhEu7+jA+lCo4v2AXPjM17Cnlugp1s1hamFIgUr5lR7TQmK++xqtoGc6Y8xxEcmksb/8S9kY5YYbVn6MmgMOAL7ylcp3ZjgD4fMtvZ97Xx/w1a9W7gcC+2973uXMct9DOgh2iZypePoKR44YzZOTk3gslao6tkm5j7PlcnUNtpB1yAvBq32ekzSm8RPhsvZ2zDgO/rG/HxnVttTGnEWA+rU5PhrFx4waeYsIf9bZie/19yNdKkEQodGycFYigZ+OjXmlaPo8Hw6Hca5KOGv1+z33t0lOlb9pwz9ULOKFqSnc2NuLmM+HrmAQz6bTeHRiwmsrO+u6eHRiAg2W5dXwtwUCiCqLoeP/OeXC1zF5S7nQzaYnJlnHQRmoyvDPu67U83ccfLe/X3ogdDhBu8aF7PfetZesM51TYDbh2V4o4Cfj47iupwdPTk7iFSUepMen3fjaQ3BhMonjYzH4VulufHccdhhwzjmyRKyvT2affxDt+KUQi0l9+TvvrLjcgZU36npnbr6//lM0b+dyMmO/1iykDG2xWNbyVDnUgv3EkbB6+PHoKN7M5SAAXJRM4vD9dSlnMOO6skbY4JFUChHb9rqMATKLOmxZuLi9HU0L8EFtSiQwWS5jm1Ja078cZnnaY5OTOL+1FdOOA1vpfmtXsbmDFABa/P5d+o8f1NCAq7q6UFTZ3T4ihC0LthK70edaHwrhorY2L+t6Lv/Q34/+2VlkXReWGqN2Ow8Ui3h8chLX9/Tg5WwWD4+Po+C6lR2oMup518VdQ0N4T2mYX9rWBhuqWxoqyWwQompedXIZobphj62eoz0aOnSRVSJBUyoRTif16YS1Dzc04BPzbFdCto0rOzsxUSrhvpERFIxfdAEZUwfk4qHgOJV2suozfLSpCYdHIhCQjXf2J2MOSBfraadJY97UJK9rwcaNUhzm/vvlrjgelzX0K4ltV7exBSpxaPNrJVrZ2D6zNNigL4JHJibw0tSU7OUN4P7RUYRte8V6mS+FGcfBHYODcIXAVV1dXmczk6rMZcV0uYxpx8GnWlvxw+FhTJRK0sBYFg4OhxeUJd8VDOJTyaSUNBUCj6VSGFZdFbSRGldyp0QEUgZYG0ozA1zHmLuDQRw/RyXk0DmLqrzj4LL2duwsFvFEKoXOQAB/kkxi5+ws7h0ZAYhwYizm7abvGRnBb3M55NVuVxtbHRZwXRfjSqkt5ziYcpzKYkPHmQE8m05j2nHkLl+dN+84lcWJMrpQ7nEvKx/YZf71MQtS2AZCIGRZOKW5GU9nMtLNLiq18LYaC0Eq+XXuJcX3sMZGOEIg77p4YHwcJVV37pULQjaF2VEoeE12CFLt8Ix43KuV319pbJTa57XEsoCTTpILDCGWLon6QTj7bNnk5aGHZEa7ThDUazjTecQGffXDBn2BPJNO44VMBjPqxxwAxkol6UZdpRRdFz8aHsbvcjkIIvxoZASXd3YiOOc/0zQ4QKX2miCblZyXTCJdLkMIWbLVtIh4/IfCYW/BE7QsbBkdlTKmRlw77vPhY01NeCqdriptA6ozvdOlkieBOh9h28ZHmpuRKpXQYtuYchz8v4kJpMplrz963nEQsW0cE41icHbWM5AEePFqrwyLCBnHwf2jo+gMBOSY5vhGCfDGZkFm4I/MznpytMI8JyoxfpuqJWj15354fByjxaJMPFNeA4sIh0UieD6bRUnt0s0yNn3O9/J5PJ1OV4UnHhgbw47ZWTiuCx8RLmprQ0cwiKOjUTw8Po6ysbjQfx3tgQDOjMcx67p4J5/HH4TDOHOVGPOXXgJ++cvK13DSScAxx9R2TEshEJCd02rFIYfIa59PGvTHH5fudTNlhUjGp08/XT5369ZqKVQAOPlk6XFgagsb9AUS9/nQYFlIG/FG0yW82nCFwJaxMbyay8FR/rTXczlsGR3Fxe3te01YMh89YS968QvluFgMj4yPI4Xq8qqwbePM1lYEbRuuEBgtlaQQjS7X2sO49kbc78cnEgk8MzmJJ5XEqrkYeyyVQoOqffcS8ICqDG+96511HLxXKKArEKh4FMwgo8b4u7CIcFw0itdyOS/eDuO9zN25qYMPAL8vFLzFowWpi05EaPD58JFYDM9kMl6f9Lnny5TL2KnTlBXbCwVsm5mRr7EsbBkbw6Xt7YjaNj4ej2OiVMJbMzM4PhpFh7G77wuH8Yl4HH0zMzgsEkHPamjnBSkG8tvfVu5PT8sd7sEH125M+zPHHy935o4jjbo+pp2P8XhFknZkpHruiSoyqUxtYYO+QA5vbMSL2azXHMWrT16luELgzelpFLX/TAgUXRevT0/DaWurMujrQyGsC4XQPzvruXDXNzR4sqz7kmOjUaTKZcwKgaMiETT7/QhZFsK2jfOUUNAvMxm8oQWwFQKyBe36JYxJGNfmgmCgUMCjauduJpYR4NV7a0EZIkKmXMYr2ay309YLAIHqmm2dDW5ZFj7W3IwG28YLmYxcWOkxkCwrE8a5AOCdfB73jYxgTHesU6/Ru+YgETYnEnAhs/KnHAdvmnOlXmPWez+XTkupV4UrZKOcnOOgxe/HuckkJkoltGUyODuR2KVW/NBIxNMeWA1s2wa8/Xb1sYEBeWGDvnQsC9i0SZasua5soLKnfzczcW6VFjOsSdigL4IPhcN4N5/HZLmMdcEgmn0+tKxSDUCLZHOSyakplIzdIQwjoTk4EsGM6+JlI431xFgMG5YhN+DM1lZkXRdZx8H5ySTiu5m/ZCCAIxsbMV4qYbBQQKvfj+5QCMdGo1XtXJ9Np73Ytp8Ip7W07KJUB8Azsjp73Uu0I8K2QsEzgNqT4WV9a3e4mrOMakwC47iYu0sXlXr7g8JhJPx+fKq1FdlyGa9OT1eJtpglfXpMOwsFvG/0ItfflAtZO/7k5CTifj8uSCYRsCwMzM6i4DjYrsRzLHPcirzrejXs+r3ntptN+P3YrCoAVjvbt1e6ehnTzuwDAgGZ9T8fRo4kgNWvb76WYIO+CP5LczOmHQdvz8zgE/H4qtq1zMUiwvnJJH6TzSKtDJpPZXvvLqHtmGgUxyyxtetix3XxXtot6bj71mwWz6TTOLyxEafNydrems3ikYkJb+fpV1Ksm4wuaYBMuntnZsZLLLNMY2cYeNe4rYVh+oJB2JaFguNgQCX2mTF+04ZUKdspt/aZiYQXb/6v7e0oCYFMuVzlCdG7e7NFrBbHMY05VFObpzMZhCwLPsvCx1ta0B0M4rxkEveOjmJINVZp9vnQbbjNT4/H8Vouh4zqDCeEwLpQaNUKv+yNZFJ2LhsfrxjyZFJe5iOXq8TdtVFKJoEjjlj+MdcT5vwzqws26IvkrETCazqy2iEidIdCaFSJe1GfD5/p6FjWMqPRYhE5x8F6tbvfns9jVGeyQ3aMm1t6VnRdvJnL4ajGxip378ZoFBv3sMh4LJVCSnUME0Kg4Dh4PJXCSU1NiBufb1s+j5eyWW8X7qoddZPfj5htY6xYxIyhyKZbkB4QDOLPu7vR7Pfjt7kcvrNzp1d6p2u3o0qLXZePCeOSVKEETdzvx/W9vfhdLoe/37nTc+l7zVRQ6ftuNprRSnf6vYUQmFF17yfHYmjw+XBgQwPObW3FwypN+YhIBH88ZwHUEQgg77peUtylHR2rIrltKWzcKI3zs88C6bR0FZ9xBnDkkfO/bmREduwql+VrbFu66NmgL45jjpHz/9xzlWNLFXVh9i1s0OuYoGXhht7evT9xHrbn87JMT1TkPg+JRHZJqpsul7G9UMDL2SwmSiVPNe2xyUm8rbLshRC4uqsLx6myMyEE3szlMFws4tGJCTiqvv3Ahoa9jsuFlH3VBthrobob32tVO1Zl2NcHgzixqQkPjo9jenbWe62AdN9/rrPTU8ALWBZiPh8mdXY+EUJEOCEWQ4AIv85mMVUqIa8WBkJ5Rw7YTQBSnyulQgVehrraNrrqtifIo267qqxPUKU7nLn4ObKxEUfOo4nw6aU0oV7FnHyyvDz0kEzcOvHEhb1OC6nsL+0wV4p335UCN0LIzPe9yUx89KPywqwu2KAz8/LDkRFsz+c9Qxm0bXyjrw/tc+qct05P4wdDQ56U6dtK/kq/TvcKHykWUXJd+C0LmXIZt/X3Y0b9qt7W34/OUAj/e8OGvY7LFcLroa53/85ujLknyGKMBULgpWwWL2Wz3u7XbMEa9furkuc2hMP4XGcn7hgclAI4roveSAQbGxvR7PNhc2sr7h4exmu5nGeAJ8plbJ+Z8XbffaEQiAi9wSA+3tyMH4+Py3FpTXVAJgdaFnKOg1k1VkEkW9RCGnXtotdtaDV5x8FQsYhG20bbKt15T5XLsrRPCCQCgV08NUvhk59c+HOFqPQf59h7hcFB4PbbpW67zwdccYVcIHGy2/4HG3RmXnRplucS3k1aa1llW2sHs4B0JXsZ36gkfT2aSqEvHMZhSmnMrKH2zr8AtIqduSOP7EaUWyvHzSqj6MWp535O9d4+IlzU2or2OUbxsEgEF7e14a6hIYAIb8/M4Js7duDYWAyfbm/HZ4wd8HixiG/398uadyFlWK/t7sahjY3Yls/j/rGxyvsbn/+whgZc2t6Ofx0bw6+npry2s+b8aRU77TFxhMB4qYQXMxk8NDGBDzc04Mqurn1iLPclBcfBQ+PjeDadhiDCyU1NuDCZXHTnvw+K/vOwLGm8FuAMqmsmJ4Hvf1+2SiWS4Yi77wY2bKjWSl/rZLMyzKBJJFZnT/TV9V/PrDq0q7yqTGuO0X01l8MjesdpZHn7iODqHaZyH1/S3o7DVDJhlYiMqusWC9wWXNPdjZt27ED/7CwAGV64pqcHiTn/ZX8YjSJdKmHLqOrUa5y/6rOo67IQ+JexMXSHw1WJZfo5ZkcxAeCVqSlELAsXtrV5yXj/NDiIETUuQKrL/dPgIL6xfn1lS+i6gNmhjAj/OT2NuM+HP21vR85x8Fo262XFVyXIESGoXvvrTAb/d3gYUN6Kt2dmcO/ICC7r6FhxY7knyq6LR1MpPJfJyDl3XbyQySDh969oPopKpQAgp3/dOtnCdC1z113Azp3ViYJMNbOzsp3s1q3yvhDAZZcBJ5xQ23HtDjbozPzoWK3KEt9dQt3RjY04N5HAv+qdpzL+JzQ1Ydpx8Nr0tIz9CoHgHCNjGyVlgNxRL4SA6u2uXflEtNtdKRHBb4zfbAOqH9cZ7oBsA3pFR8euxrz6pN4O2RUCz2UyaPb5cJYq+9I7a69LHIxwgE6Ao4pynPZSlFV5md+ycHlnJ25TYjaOksTV79modvwDxSLuHhlBWcu2qvP8JptF1OdbNXHz5zMZ/CKV8soCdXjEXcEAdn+/bITiOBWtcqLVuctaacz1OYcgduWBB4BXXqnul26s11cVbNCZedHNTQSAgG3jio6OKiUx/ZwGn6+qP/bR0SjOa22FA2DGqG9OGr+gcb8fV3d14db+fsy6Llr9flzT3b3gsUUX6Fb2dv20a8MXQLZYFTrTfZ7zms1V9LkslYluCgCbCxTztVWCLaanQBl2ywgZRGwbV6qmM9/t78emRAIfVpUDRIT2QABvGcJBc4VuSqso26uMyiJHVyXo2yuJVg0WYuG9suudyy8Hbr21Utev4Z16hVJJLgTNf/8tW2Rm/94qK1YaNujMvFzR2Ym80WNx/R7EZo6NRtEVCHg/2M1+v9cIZr7ObIdFIvjSunUQaic9N3a9LzghFsNIsYgnlPyr52onwqnNzfijWMwzwATMr5A3R/DFQcWIav6sqws39/d7ojchAJd3dKDJ50PEsnBBWxu2jI3tcuq54QwtuvPfurvRHghU1dfvLBRw59AQHFVbr8vciAiHhMP45CoSiTkhFsPw7CyeTqcri5oaWwzeiUriceDKK4FbbpFlfX4/8NnPAm1ttR7Z6sEQ2/Q4/fTVqUrIBp2Zl669dO3ShG0bG5aQYURES3rdYmiwbZwZj+PwhoYqZTYBoDcY9MrT9sbhkQiu7+7Gq7kcnkunK+IyZLRCBdATCuHqri4UXNfbfeu8AZ+Sg+0OBvHG9DSeSqcBYzy7E/3ZnX66IwSmddMWVFqobgiFcGF7+24V+GpFxLZxViKBguvixWwWFhGOaWzER/ZRj4CFoKMuek3E8eIKXV3SqOuytUMP5bkxOeMM2ZHurbfkvLiuXPCsxmISNujMmqDZ71+w4d4TjT4fjoxG0RkM4sBwGK9ksygJgeNjMfTNMboHzbNICds2jmhsRHcwiL45Ho899WufS2cwiLPicTyaSnmlbN3BIM5PJhe8CFtJWlQCnF7YrAuFVnzRYduVOnQtLMNIuLnKnmlvl7r2OglOiNW5OwfYoDPMokkGAkgGAmgPBOAIsSAhnN0R9/uX3MkuaFk4raUFEcvCQLGI9woFnNPauuSxrASdweBe+7QvF4kEcP75MhYKSIO+nwg+MquAvj55We2wQWeYJbKnfIKVIurz4fREAiPFIt7N53HUPEpxa51QCDjttFqPgmGWFzboDLOf0668BQzDrG32z3ZLDMMwDMNUwQadYRiGYeoANugMwzAMUwewQWcYhmGYOoANOsMwDMPUAWzQGYZhGKYOYIPOMAzDMHUAG3SGYRiGqQPYoDMMwzBMHcAGnWEYhmHqADboDMMwDFMHsEFnGIZhmDqADTrDMAzD1AFs0BmGYRimDmCDzjAMwzB1ABt0hmEYhqkD2KAzDMMwTB3ABp1hGIZh6gA26AzDMAxTB7BBZxiGYZg6gA06wzAMw9QBJISo9RiWDBGNAep+zIcAAATkSURBVNixiJe0AhhfpuEw88NzXzt47msHz33tqNe5P0AIkdzdA/u1QV8sRPSSEOK4Wo9jLcJzXzt47msHz33tWItzzy53hmEYhqkD2KAzDMMwTB2w1gz6HbUewBqG57528NzXDp772rHm5n5NxdAZhmEYpl5Zazt0hmEYhqlL1oRBJ6JNRPQ7InqHiP6q1uOpR4joLiIaJaLXjWNxInqciLap6xZ1nIjou+r7eJWIjqndyPdviKiXiJ4kojeJ6A0i+qI6znO/zBBRiIh+TUT/qeb+6+r4eiL6lZrjHxNRQB0PqvvvqMf7ajn+eoCIbCLaSkQPq/treu7r3qATkQ3gewDOBHAogEuI6NDajqou+WcAm+Yc+ysA/y6EOAjAv6v7gPwuDlKXqwD84wqNsR4pA7hRCHEogBMAXKP+vnnul59ZAKcKIY4CcDSATUR0AoBvAviOEOJAAJMAPq+e/3kAk+r4d9TzmA/GFwG8Zdxf03Nf9wYdwB8BeEcI8Z4QogjgPgDn1nhMdYcQ4hkAqTmHzwXwA3X7BwA+ZRy/W0h+CaCZiDpXZqT1hRBiSAjxirqdhfxx6wbP/bKj5nBa3fWriwBwKoAt6vjcudffyRYApxERrdBw6w4i6gFwNoDvq/uENT73a8GgdwPYadzvV8eY5addCDGkbg8DaFe3+TtZBpQbcSOAX4HnfkVQLt/fABgF8DiAdwGkhRBl9RRzfr25V49nACRWdsR1xc0A/icAV91PYI3P/Vow6MwqQMhyCi6pWCaIqBHATwDcIISYMh/juV8+hBCOEOJoAD2Q3sCDazykNQERbQYwKoR4udZjWU2sBYM+AKDXuN+jjjHLz4h256rrUXWcv5N9CBH5IY35PUKIf1OHee5XECFEGsCTAE6EDGP41EPm/Hpzrx5vAjCxwkOtF04CcA4R/R4yjHoqgFuwxud+LRj0FwEcpLIfAwAuBvBgjce0VngQwGXq9mUAfmYc/6zKuD4BQMZwDzOLQMUB/w+At4QQf288xHO/zBBRkoia1e0wgNMhcxieBHCBetrcudffyQUAnhAsBLIkhBBfFkL0CCH6IH/TnxBCXIo1PvdrQliGiM6CjLfYAO4SQvyvGg+p7iCiewGcAtnhaATAVwE8AOB+AOsgu+JdJIRIKSN0G2RW/AyAy4UQL9Vi3Ps7RHQygGcBvIZKLPErkHF0nvtlhIiOhEy0siE3R/cLIb5BRBsgd41xAFsB/KkQYpaIQgB+CJnnkAJwsRDivdqMvn4golMA/IUQYvNan/s1YdAZhmEYpt5ZCy53hmEYhql72KAzDMMwTB3ABp1hGIZh6gA26AzDMAxTB7BBZxiGYZg6gA06wzAMw9QBbNAZhmEYpg5gg84wzIIhoj9UfdRDRBRRfcAPr/W4GIZhYRmGYRYJEf0NgBCAMIB+IcTf1nhIDMOADTrDMItE9UR4EUABwEeEEE6Nh8QwDNjlzjDM4kkAaAQQhdypMwyzCuAdOsMwi4KIHoRsgLEeQKcQ4toaD4lhGAC+vT+FYRhGQkSfBVASQvwLEdkAXiCiU4UQT9R6bAyz1uEdOsMwDMPUARxDZxiGYZg6gA06wzAMw9QBbNAZhmEYpg5gg84wDMMwdQAbdIZhGIapA9igMwzDMEwdwAadYRiGYeoANugMwzAMUwf8f+OcjBkHo6iaAAAAAElFTkSuQmCC\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "plot(data3, n_samples3, km3, 'Clustered points in dataset n. 2')" + ] }, { "cell_type": "markdown", @@ -706,12 +13067,68 @@ }, { "cell_type": "code", - "execution_count": null, - "metadata": { - "collapsed": true - }, + "execution_count": 11, + "metadata": {}, "outputs": [], - "source": [] + "source": [ + "from numpy.linalg import norm\n", + "from pandas import DataFrame\n", + "from functools import partial\n", + "from multiprocessing import Pool\n", + "\n", + "def sse(X, Y):\n", + " return norm(X - Y)**2\n", + "\n", + "def wss(clusters, means):\n", + " wss = 0\n", + " for mean, cluster in zip(means, clusters):\n", + " wss += sum(map(lambda example: sse(example, mean), cluster))\n", + " return wss\n", + "\n", + "def compute_clusters(labels, data):\n", + " df = DataFrame(zip(labels, data)) # tuple \n", + " return map(lambda nd: nd[1]. # list of lists\n", + " drop(0, axis=1). # cluster is the index, drop it from tuple\n", + " values.flatten(), # flatten list of single values\n", + " df.groupby(0)) # group by cluster \n", + "\n", + "def wss_from_data(n_clusters, data, seed):\n", + " km = KMeans(n_clusters=n_clusters, random_state=seed).fit(data)\n", + " means = km.cluster_centers_\n", + "\n", + " clusters = list(compute_clusters(km.labels_, data))\n", + " #return min(map(lambda clu: wss(clu, means), clusters))\n", + " #print(seed, clusters)\n", + " return wss(clusters, means)\n", + "\n", + "\n", + "def eval_kmeans(n_clusters, data, n_iter = 10):\n", + " for seed in range(n_iter):\n", + " yield wss_from_data(n_clusters, data, seed)" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "data1 evaluation: 62525.0\n", + "data2 evaluation: 2586285.25177\n", + "data3 evaluation: 490062743.13532\n" + ] + } + ], + "source": [ + "print(\"data1 evaluation:\", round(min(eval_kmeans(3, data1, 10)), 5))\n", + "print(\"data2 evaluation:\", round(min(eval_kmeans(4, data2, 10)), 5))\n", + "print(\"data3 evaluation:\", round(min(eval_kmeans(5, data3, 2)), 5))" + ] }, { "cell_type": "markdown", @@ -722,10 +13139,302 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 13, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[540169.52, 225889.10891089108, 62525.00000000001, 45950.0, 30325.0, 15625.0, 12565.0, 9660.0, 6953.0]\n" + ] + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "evaluations = list(min(eval_kmeans(k, data1, 3)) for k in range(1, 10))\n", + "print(evaluations)\n", + "plt.plot([k for k in range(1,10)], evaluations)\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[11537928.01912388, 6012536.807920576, 3327557.113231734, 2586444.7018824834, 2055550.8429586363, 1660581.039399704, 1290968.9205037882, 1108531.0806234784, 943472.771559587]\n" + ] + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "evaluations = list(min(eval_kmeans(k, data2, 3)) for k in range(1, 10))\n", + "print(evaluations)\n", + "plt.plot([k for k in range(1,10)], evaluations)\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 15, "metadata": { - "collapsed": true + "scrolled": true }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[2418979043.1137347, 1425203644.7839656, 916173975.2419217, 632521252.3875252, 490062743.1353237, 384262182.9810209, 314207574.83201766, 275304694.96816456, 240376762.14969328]\n" + ] + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "evaluations = list(min(eval_kmeans(k, data3, 1)) for k in range(1, 10))\n", + "print(evaluations)\n", + "plt.plot([k for k in range(1,10)], evaluations)\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# DBSCAN" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "from sklearn.cluster import DBSCAN\n", + "cls = DBSCAN(eps = 3)\n", + "cls.fit(data1)\n", + "\n", + "plot(data1, n_samples1, cls, 'DBScan 1')" + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "DBScan2 = DBSCAN(eps = 1.4, min_samples = 5)\n", + "DBScan2.fit(data2)\n", + "\n", + "fig = plt.figure(figsize=(8,8))\n", + "\n", + "ax = fig.add_subplot(111)\n", + "fig.subplots_adjust(top=1)\n", + "ax.set_title('DBSCAN dataset 2')\n", + "\n", + "ax.set_xlabel('x')\n", + "ax.set_ylabel('y')\n", + "\n", + "for clu in range(-1, max(DBScan2.labels_) + 1):\n", + " data_list_x = [data2[i,0] for i in range(n_samples2) if DBScan2.labels_[i]==clu]\n", + " data_list_y = [data2[i,1] for i in range(n_samples2) if DBScan2.labels_[i]==clu]\n", + " plt.scatter(data_list_x, data_list_y, s=2, edgecolors='none', c=color[clu])" + ] + }, + { + "cell_type": "code", + "execution_count": 18, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "DBScan3 = DBSCAN(eps = 4, min_samples = 20)\n", + "DBScan3.fit(data3)\n", + "\n", + "fig = plt.figure(figsize=(8,8))\n", + "\n", + "ax = fig.add_subplot(111)\n", + "fig.subplots_adjust(top=1)\n", + "ax.set_title('DBSCAN 3')\n", + "\n", + "ax.set_xlabel('x')\n", + "ax.set_ylabel('y')\n", + "\n", + "for clu in range(-1, max(DBScan3.labels_) + 1):\n", + " data_list_x = [data3[i,0] for i in range(n_samples3) if DBScan3.labels_[i]==clu]\n", + " data_list_y = [data3[i,1] for i in range(n_samples3) if DBScan3.labels_[i]==clu]\n", + " plt.scatter(data_list_x, data_list_y, s=2, edgecolors='none', c=color[clu % len(color)])" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Reachability Distance" + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "metadata": {}, + "outputs": [], + "source": [ + "from sklearn.neighbors import LocalOutlierFactor\n", + "def reachability_distance(x, n):\n", + " lof = LocalOutlierFactor(n_neighbors = n)\n", + " lof.fit(x)\n", + " kneighbors = lof.kneighbors()[0]\n", + " distances = list(max(kneighbor) for kneighbor in kneighbors)\n", + " distances = sorted(distances)\n", + " return distances" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "[]" + ] + }, + "execution_count": 20, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "plt.plot(reachability_distance(data2, 5))" + ] + }, + { + "cell_type": "code", + "execution_count": 21, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "[]" + ] + }, + "execution_count": 21, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "plt.plot(reachability_distance(data3, 20))" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, "outputs": [], "source": [] } diff --git a/anno3/apprendimento_automatico/esercizi/meo/3-clusters.csv b/anno3/apprendimento_automatico/esercizi/meo/3-clusters.csv new file mode 120000 index 0000000..d775e10 --- /dev/null +++ b/anno3/apprendimento_automatico/esercizi/meo/3-clusters.csv @@ -0,0 +1 @@ +Datasets/3-clusters.csv \ No newline at end of file diff --git a/anno3/apprendimento_automatico/esercizi/meo/CURE-complete.csv b/anno3/apprendimento_automatico/esercizi/meo/CURE-complete.csv new file mode 120000 index 0000000..68f4b45 --- /dev/null +++ b/anno3/apprendimento_automatico/esercizi/meo/CURE-complete.csv @@ -0,0 +1 @@ +Datasets/CURE-complete.csv \ No newline at end of file diff --git a/anno3/apprendimento_automatico/esercizi/meo/clustering-copy_for_students_aa_19_20.ipynb b/anno3/apprendimento_automatico/esercizi/meo/clustering-copy_for_students_aa_19_20.ipynb index fea11f6..704609e 100644 --- a/anno3/apprendimento_automatico/esercizi/meo/clustering-copy_for_students_aa_19_20.ipynb +++ b/anno3/apprendimento_automatico/esercizi/meo/clustering-copy_for_students_aa_19_20.ipynb @@ -21,7 +21,7 @@ }, { "cell_type": "code", - "execution_count": 22, + "execution_count": 1, "metadata": {}, "outputs": [ { @@ -35,11 +35,11 @@ "n_samples=\n", "86558\n", "dataset n. 1: n samples, n features\n", - "(150, 2)\n", + "150 2\n", "dataset n. 2: n samples, n features\n", - "(6118, 2)\n", + "6118 2\n", "dataset n. 3: n samples, n features\n", - "(86558, 2)\n" + "86558 2\n" ] } ], @@ -49,6 +49,8 @@ "from os.path import join\n", "\n", "import numpy as np\n", + "color=['b','g','r','c','m','y','k','w']\n", + "\n", " \n", "# this function reads the data file, loads the configuration attributes specifiefd in the heading\n", "# (numer of examples and features), the list of feature names\n", @@ -109,12 +111,14 @@ "outputs": [ { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfgAAAI1CAYAAAA6tuiIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3XuUXWWd5vHnkQjIJSGRmEkIQxgbcSW0F6ak8dZNA7ao\njGHNmtWD02i0WcPoOIq92gtoj7fxQne7HJ3VSxwG1EyD0CzEJm23DnR0dBwFLFAREuhEQQhJSDkI\nAS/I5Td/7F1yrNTlnKrznr3fd38/a9Wqc/bep877VqV+v9rvc/aJI0IAAKAsT2l6AAAAYPho8AAA\nFIgGDwBAgWjwAAAUiAYPAECBaPAAABSIBg9gaFz5rO2f2r6xj+PX2A7bi0Yxvvmy/VLbdzQ9DmAQ\nNHh0lu27bP/C9kO2H7D9LdtvtN3X78WomlPq57G9v+2r6u9H2D5pAV/uJZJeJml1RJwwnBFWbJ9k\ne8cwv2a/IuL/RMSx/Rzb5DiBXjR4dN2/iohDJR0l6QJJ75J0SbNDasQ3JZ0lafcCv85Rku6KiJ8t\nfEgAFoIGD0iKiAcjYpOkfytpg+3jJMn2q2x/1/Ze2/fYfn/Pw75Rf37A9sO2X2j7mba/avv/2f6J\n7ctsHzb5ANvvsn1vvWpwh+1T6u1PsX2e7R/Wj73S9rKZnmfq+G2/v37M/6y/9m22x/qc+68i4hMR\n8U1Jj891vO1VtjfZvt/2dtv/vt5+tqSLJb2wHucHpnnsfrY/Vn9vfiTpVVP2v8H21noOP7L9H+rt\nB0v6sqRV9dd+uB7HCba/Xa/A7LL9V7b3n2Hckysh59jeWR//9p79B9j+RL1vZ337gHrfb5yV16sd\nb7d9i+0Hbf+N7QPnGOd4/e/oPtsfn+v7DCxYRPDBRyc/JN0l6dRptt8t6U317ZMk/baqP4afI+k+\nSWfU+9ZICkmLeh77W6qWqA+QtFxVc/5Eve9YSfdIWtXz+GfWt8+VdL2k1fVj/7uky2d6nmnG/H5J\nv5T0Skn7SfqopOvn8T3ZIemkOY75hqRPSTpQ0vMkTUg6ud73eknfnOWxb5R0u6QjJS2T9LXeualq\n+M+UZEm/J+nnko7v+VnsmPL1/qWkEyUtqr9PWyW9bYbnnvw+Xi7p4PrnOjH5b0DSB+ufwTPqn923\nJP2X6Z67/rdzo6RV9Ty2SnrjLOP8tqTX1rcPkXRi0//++Sj/gzN4YF87VRVtRcT/jogfRMQTEXGL\nqubwezM9MCK2R8R1EfFIRExI+njP8Y+rat5rbT81Iu6KiB/W+94o6T0RsSMiHlHVsP/NgLn7NyPi\nHyLicUl/Lem5Azy2L7aPlPRiSe+KiF9GxPdUnbW/rs8v8Yeq/uC5JyLuV/WHyK9FxN9HxA+j8nVJ\n10p66UxfLCJuiojrI+KxiLhL1R9GM/58ah+IiJ9FxA8kfVbSa+rtfyTpgxGxp/7ZfUDSa2f5Ov8t\nInbW8/g7VX/szORRSb9l+/CIeDgirp9jjMCC0eCBfR0h6X5Jsv07tr9me8L2g6oa8eEzPdD2CttX\n1MvweyVdOnl8RGyX9DZVzXtPfdyq+qFHSfpivdT8gKozwsclrRhg3L35+c8lHZjghXmrJN0fEQ/1\nbPuxqu9Zv4+/Z8pjf832K2xfXy//P6BqRWK27/ezbH/J9u76+/2R2Y6vTX3+yZ/Bqinj6d03nanf\n70NmOfZsSc+SdLvt79g+fY4xAgtGgwd62H6Bqmb1zXrT5yVtknRkRCyR9GlVy8dStdw71Ufq7b8d\nEYtVvXBt8nhFxOcj4iWqGnpI+vN61z2SXhERh/V8HBgR987wPE3ZKWmZ7UN7tv1zSff2+fhdqpbn\nex8rqcrAJX1B0sckrYiIwyT9g2b/fl+oasn/mPr7/e6e42cy9fl31rd3qvq5TLdvEPuMMyK2RcRr\nVC3//7mkq+q8HkiGBg9Isr24Pqu6QtKl9fKtJB2q6oz1l7ZPkPTveh42IekJSf+iZ9uhkh6W9KDt\nIyS9o+c5jrV9ct3IfinpF/XjpeoPhw/bPqo+drnt9bM8z1DVLzA7sL67f/2CsX0aZUTcoyqb/mh9\nzHNUnZ1e2udTXSnprbZX214q6byeffurijAmJD1m+xWS/qBn/32Snm57Sc+2QyXtlfSw7WdLelMf\nY/jPtg+yvU7SGyT9Tb39ckl/Vn/vD5f03gHm1Wufcdo+y/byiHhC0gP15iemfTQwJDR4dN3f2X5I\n1Rn0e1Rl5m/o2f8fJX2wPua9qhqUJCkifi7pw5L+b720fqKq3PZ4SQ9K+ntJV/d8rQNUXYr3E1XL\nu8+QdH6975OqVgqurZ/rekm/M8vzDMT2l22/e5ZD7lD1B8cRkv5XffuoGY59jaoXrO2U9EVJ74uI\nf+xzKP+j/vrfl3Szer4/9bL/W1V9j3+q6o+pTT37b1fVhH9Ufx9WSXp7fdxD9deebNaz+bqk7ZI2\nS/pYRFxbb/+QpHFJt0j6QT2+D/U5r1+bYZynSbrN9sOqftZnRsQvBv3awCAc0abVPwBIw/YaSXdK\nempEPNbsaID0OIMHAKBANHgAAArEEj0AAAXiDB4AgALR4AEAKFCr/w/muRx++OGxZs2apocBAMDI\n3HTTTT+JiOVzHZd1g1+zZo3Gx8ebHgYAACNj+8dzH8USPQAARaLBAwBQIBo8AAAFosEDAFAgGjwA\nAAWiwQMAUCAaPAAABaLBAwBQIBo8AAAFosEDAFAgGjwAAAWiwQMAUCAaPAAABaLBAwBQIBo8AAAF\nosEDAFAgGjwAAAWiwQMAUCAaPAAABaLBAwBQIBo8AGQqInTbzgcVEU0PBS1EgweATG3ZtVdvuvRm\nbdm1t+mhoIVo8ACQqbUrF+vCs47X2pWLmx4KWmhR0wMAAMyPba1btaTpYaClOIMHgMyQvaMfNHgA\nyAzZO/pBgweAzJC9ox80eADIQO+y/GT2brvpYaHFaPAAkAGW5TEoGjwAZIBleQyKy+QAIANcEodB\ncQYPAC3GJXGYLxo8ALQY2TvmiwYPAC1G9o75IoMHgBYje8d8cQYPAC1D7o5hoMEDQMuQu2MYaPAA\n0DLk7hgGMngAaBlydwwDZ/AA0BJk7xgmGjwAtATZO4aJBg8ALUH2jmEigweAliB7xzBxBg8ADSJ3\nRyo0eABoELk7UqHBA0CDyN2RChk8ADSI3B2pcAYPAA0ge0dqNHgAaADZO1KjwQNAA8jekRoZPAA0\ngOwdqXEGDwAjQu6OUUrW4G1/xvYe27f2bPtL27fbvsX2F20f1rPvfNvbbd9h++WpxgUATSF3xyil\nPIP/nKTTpmy7TtJxEfEcSf8k6XxJsr1W0pmS1tWP+ZTt/RKODQBGjtwdo5SswUfENyTdP2XbtRHx\nWH33ekmr69vrJV0REY9ExJ2Stks6IdXYAGCUJpfmJWndqiWy3fCI0AVNZvB/LOnL9e0jJN3Ts29H\nvW0fts+xPW57fGJiIvEQAWDhWJpHExpp8LbfI+kxSZcN+tiIuCgixiJibPny5cMfHAAMGUvzaMLI\nL5Oz/XpJp0s6JZ58Kem9ko7sOWx1vQ0AssclcWjCSM/gbZ8m6Z2SXh0RP+/ZtUnSmbYPsH20pGMk\n3TjKsQHAMHFJHJqW8jK5yyV9W9KxtnfYPlvSX0k6VNJ1tr9n+9OSFBG3SbpS0hZJX5H05oh4PNXY\nACA1cnc0zTn/dTk2Nhbj4+NNDwMA9hER2rJrr9auXMyr5jFUtm+KiLG5juOtagEgAXJ3NI23qgWA\nISJ7R1vQ4AFgiMje0RY0eAAYIq55R1uQwQPAEJG9oy04gweABSJ3RxvR4AFggcjd0UY0eABYIHJ3\ntBEZPAAsELk72ogzeACYB3J3tB0NHgDmgdwdbUeDB4B5IHdH25HBA8A8kLuj7TiDB4ABkL0jFzR4\nABgA2TtyQYMHgAGQvSMXNHgAmEPvsvxk9m676WEBs6LBA8AcWJZHjmjwADAHluWRIy6TA4A5cEkc\ncsQZPADMgEvikDMaPADMgOwdOaPBA8AMyN6RMzJ4AJgB2Ttyxhk8APQgd0cpaPAA0IPcHaWgwQNA\nD3J3lIIMHgB6kLujFJzBA4DI3lEeGjwAiOwd5aHBA4DI3lEeMngAENk7ysMZPIDOIndHyWjwADqL\n3B0lo8ED6Cxyd5SMDB5AZ5G7o2ScwQPoHLJ3dAENHkDnkL2jC2jwADqH7B1dQAYPoHPI3tEFnMED\n6ARyd3QNDR5AJ5C7o2to8AA6gdwdXUODB1C0yaV5SVq3aolsNzwiYDRo8ACKxtI8uooGD6BoLM2j\nq7hMDkDRuCQOXcUZPIDicEkcQIMHUCByd4AGD6BA5O4AGTyAApG7A5zBAygI2TvwJBo8gGKQvQNP\nosEDKAbZO/AkMngAxSB7B57EGTyArJG7A9OjwQPIGrk7MD0aPICskbsD0yODB5A1cndgepzBA8gO\nuTswNxo8gOyQuwNzo8EDyA65OzA3MngA2SF3B+bGGTyAbJC9A/2jwQPIBtk70D8aPIBskL0D/SOD\nB5ANsnegf5zBA2g1cndgfpI1eNufsb3H9q0925bZvs72tvrz0p5959vebvsO2y9PNS4AeSF3B+Yn\n5Rn85ySdNmXbeZI2R8QxkjbX92V7raQzJa2rH/Mp2/slHBuATJC7A/OTrMFHxDck3T9l83pJG+vb\nGyWd0bP9ioh4JCLulLRd0gmpxgag/SaX5iVp3aolst3wiIC8jDqDXxERu+rbuyWtqG8fIemenuN2\n1Nv2Yfsc2+O2xycmJtKNFECjWJoHFqaxF9lF9YqZgV81ExEXRcRYRIwtX748wcgAtAFL88DCjLrB\n32d7pSTVn/fU2++VdGTPcavrbQA6avKSOJbmgfkZdYPfJGlDfXuDpGt6tp9p+wDbR0s6RtKNIx4b\ngIZxSRwwPCkvk7tc0rclHWt7h+2zJV0g6WW2t0k6tb6viLhN0pWStkj6iqQ3R8TjqcYGoJ3I3YHh\ncc5/KY+NjcX4+HjTwwAwJBGhLbv2au3KxSzNAzOwfVNEjM11HG9VC6A1eCtaYHh4q1oAjSN7B4aP\nBg+gcWTvwPDR4AE0jmvegeEjgwfQOLJ3YPg4gwfQCHJ3IC0aPIBGkLsDadHgATSC3B1IiwweQCPI\n3YG0OIMHMFJk78Bo0OABjBTZOzAaNHgAI0X2DowGGTyAkSJ7B0aDM3gAyZG7A6NHgweQHLk7MHo0\neADJkbsDo0eDB5DM5NK8JK1btUS2Gx4R0B00eADJsDQPNIcGDyAZluaB5nCZHIBkuCQOaA5n8ACG\nikvigHagwQMYKnJ3oB1o8ACGitwdaAcyeABDRe4OtANn8ACGguwdaBcaPIChIHsH2oUGD2AoyN6B\ndiGDBzAUZO9Au3AGD2DeyN2B9qLBA5g3cnegvWjwAOaN3B1oLzJ4APNG7g60F2fwAAZG9g60Hw0e\nwMDI3oH2o8EDGBjZO9B+ZPAABkb2DrQfZ/AA+kLuDuSFBg+gL+TuQF5o8AD6Qu4O5IUMHkBfyN2B\nvHAGD2BG5O5AvmjwAGZE7g7kiwYPYEbk7kC+aPAA9jG5NC9J61Ytke2GRwRgUDR4APtgaR7IHw0e\nwD5Ymgfyx2VyAPbBJXFA/jiDByCJS+KA0tDgAUgidwdKQ4MHIIncHSgNGTwASeTuQGk4gwc6juwd\nKBMNHug4snegTDR4oOPI3oEykcEDHUf2DpSJM3igg8jdgfLR4IEOIncHykeDBzqI3B0oHxk80EHk\n7kD5OIMHOoTsHegOGjzQIWTvQHfQ4IEOIXsHuoMMHugQsnegOziDBwpH7g50Ew0eKBy5O9BNNHig\ncOTuQDeRwQOFI3cHuokzeKBQZO9AtzXS4G3/ie3bbN9q+3LbB9peZvs629vqz0ubGBtQCrJ3oNtG\n3uBtHyHprZLGIuI4SftJOlPSeZI2R8QxkjbX9wHME9k70G1NLdEvkvQ024skHSRpp6T1kjbW+zdK\nOqOhsQHZ6l2Wn8zebTc9LAANGHmDj4h7JX1M0t2Sdkl6MCKulbQiInbVh+2WtGK6x9s+x/a47fGJ\niYmRjBnIBcvyACY1sUS/VNXZ+tGSVkk62PZZvcdE9aqgaV8ZFBEXRcRYRIwtX748+XiBnLAsD2BS\nE0v0p0q6MyImIuJRSVdLepGk+2yvlKT6854GxgZkjWV5AJOaaPB3SzrR9kGuqtApkrZK2iRpQ33M\nBknXNDA2AACKMPI3uomIG2xfJelmSY9J+q6kiyQdIulK22dL+rGkPxz12AAAKEUj72QXEe+T9L4p\nmx9RdTYPAAAWiHeyAwCgQDR4AAAKRIMHAKBANHgAAApEgwcAoEA0eAAACkSDBwCgQDR4AAAKRIMH\nAKBANHgAAApEgwcAoEA0eAAACkSDBwCgQDR4AAAKRIMHAKBANHgAAApEgwcAoEA0eAAACkSDBwCg\nQDR4AAAKRIMHAKBANHgAAApEgwcAoEA0eAAACkSDBwCgQDR4AAAKRIMHAKBANHgAAApEgwcAoEA0\neAAACkSDBwCgQDR4AAAKRIMHAKBAczZ422+xvXQUgwEAAMPRzxn8CknfsX2l7dNsO/WgAADAwszZ\n4CPizyQdI+kSSa+XtM32R2w/M/HYAADAPPWVwUdESNpdfzwmaamkq2z/RcKxAQCAeVo01wG2z5X0\nOkk/kXSxpHdExKO2nyJpm6R3ph0iAAAY1JwNXtIySf86In7cuzEinrB9epphAQCAhZizwUfE+2bZ\nt3W4wwEAAMPAdfAAABSIBg8AQIFo8AAAFIgGDwBAgWjwAAAUiAYPAECBaPAAABSIBg8AQIFo8AAA\nFIgGDwBAgWjwAAAUiAYPAECBaPAAABSIBg8AQIFo8AAAFIgGDwBAgWjwAAAUiAYPAECBaPAAABSI\nBg8AQIFo8AAAFIgGDwBAgWjwAAAUiAYPAECBaPAAABSIBg8AQIFo8AAAFIgGDwBAgRpp8LYPs32V\n7dttb7X9QtvLbF9ne1v9eWkTYwMAoARNncF/UtJXIuLZkp4raauk8yRtjohjJG2u7wMAgHkYeYO3\nvUTS70q6RJIi4lcR8YCk9ZI21odtlHTGqMcGAMCwRIRu2/mgIqKR52/iDP5oSROSPmv7u7Yvtn2w\npBURsas+ZrekFdM92PY5tsdtj09MTIxoyAAADGbLrr1606U3a8uuvY08fxMNfpGk4yVdGBHPl/Qz\nTVmOj+rPnWn/5ImIiyJiLCLGli9fnnywAADMx9qVi3XhWcdr7crFjTx/Ew1+h6QdEXFDff8qVQ3/\nPtsrJan+vKeBsQEAMBS2tW7VEtlu5PlH3uAjYreke2wfW286RdIWSZskbai3bZB0zajHBgDAQjWd\nvU9q6lX0b5F0me1bJD1P0kckXSDpZba3STq1vg8AQFaazt4nLWriSSPie5LGptl1yqjHAgDAMDWd\nvU9qpMEDAFCqyey9abxVLTqlLdkYgLK0sbbQ4NEpbcnGAJSljbXFbfprY1BjY2MxPj7e9DCQkYjQ\nll17tXbl4sYuXQFQnlHWFts3RcR0r2P7DWTw6JS2ZGMAytLG2sISPTqhjfkYgPy1ubbQ4NEJbczH\nAOSvzbWFDB6dQPYOIIUmagsZPNCjjfkYgPy1ubawRI9itTkbA5CvXGoLDR7FanM2BiBfudQWMngU\ni9wdQApN1xYyeHRem7MxAPnKpbawRI/i5JKPAchLbrWFBo/i5JKPAchLbrWFDB7FaTofA1CmttSW\nfjN4zuBRhN6ls8l8jOYOYKFyri00eBQht6UzAHnIubawRI8itGXpDEBZ2lhbuEwOnZLLZSsA8pJz\nbWGJHlnL7bIVAHkoobbQ4JG1nPMxAO1VQm0hg0fW2piPAchfm2sLGTw6Ied8DEB7lVBbWKJHdkrI\nxgC0T2m1hQaP7JSQjQFon9JqCxk8stPmbAxAvnKpLWTwKFYJ2RiA9imttrBEj2yUlo8BaIdSawsN\nHtkoLR8D0A6l1hYyeGQjl3wMQF5yqy1k8ChOafkYgHYotbawRI9WKzUbA9CsLtQWGjxardRsDECz\nulBbyODRarllYwDykHNtIYNHEUrNxgA0qwu1hSV6tFIX8jEAo9el2kKDRyt1IR8DMHpdqi1k8Gil\nnPMxAO1VQm0hg0fWupCPARi9LtUWlujRGl3KxgCMTldrCw0erdGlbAzA6HS1tpDBozVKyMYAtE9p\ntaXfDJ4zeDSqd+lsMhsr4RcQQLOoLTR4NKyrS2cA0qK2sESPhpW2dAagHUquLVwmhyx06ZIVAKND\nbWGJHg3p6mUrANKitjyJBo9GkI8BSIHa8iQyeDSi5HwMQHO6UFvI4NFq5GMAUqC2PIkleowM2RiA\nFKgt06PBY2TIxgCkQG2ZHhk8RqYL2RiA0etabSGDR+uQjQFIgdoyPZbokRz5GIAUqC2zo8EjOfIx\nAClQW2ZHBo/kupaPARiNrtYWMni0BvkYgBSoLbNjiR5JkI0BSIHa0j8aPJIgGwOQArWlf2TwSKKr\n2RiAtKgtZPBoGNkYgBSoLf1jiR5DRT4GIAVqy+Bo8Bgq8jEAKVBbBkcGj6EiHwOQArXlSWTwaAT5\nGIAUqC2DY4keC0Y2BiAFasvCNNbgbe9n+7u2v1TfX2b7Otvb6s9LmxobBkM2BiAFasvCNHkGf66k\nrT33z5O0OSKOkbS5vo8MrF25WBeedbzWrlzc9FAAFITasjCNNHjbqyW9StLFPZvXS9pY394o6YxR\njwuDmVw+k6R1q5Z0/oUvAIaD2jIcTZ3Bf0LSOyU90bNtRUTsqm/vlrRiugfaPsf2uO3xiYmJxMPE\nbFg+A5ACtWU4Rt7gbZ8uaU9E3DTTMVG9omLaV1VExEURMRYRY8uXL081TPSB5TMAKVBbhqOJy+Re\nLOnVtl8p6UBJi21fKuk+2ysjYpftlZL2NDA2DIDLVgCkQG0ZjpGfwUfE+RGxOiLWSDpT0lcj4ixJ\nmyRtqA/bIOmaUY8Nc+OyFQApUFuGr03XwV8g6WW2t0k6tb6PliEbA5ACtWX4eKtaDIS3iwSQArWl\nf7xVLZIgGwOQArVl+Nq0RI8WIx8DkAK1JR0aPPpCPgYgBWpLOmTw6Av5GIAUqC2DI4PHUJGPAUiB\n2pIOS/SYEdkYgBSoLaNBg8eMyMYApEBtGQ0yeMyIbAxACtSWhSGDx4KRjQFIgdoyGizRYx/kYwBS\noLaMFg0e+yAfA5ACtWW0yOCxD/IxAClQW4aDDB7zRj4GIAVqy2ixRA9JZGMA0qC2NIcGD0lkYwDS\noLY0hwweksjGAKRBbRm+fjN4zuA7rHfpbDIb4xcQwEJRW9qBBt9hLJ0BSIHa0g4s0XcYS2cAUqC2\npMVlcpgTl6wASIHa0g4s0XcQl60ASIHa0i40+A4iHwOQArWlXcjgO4h8DEAK1JbRIIPHjMjHAKRA\nbWkXlug7gmwMQArUlvaiwXcE2RiAFKgt7UUG3xFkYwBSoLaMHhk8fgPZGIAUqC3txRJ94cjHAKRA\nbWk/GnzhyMcApEBtaT8y+MKRjwFIgdrSHDJ4SCIfA5AGtaX9WKIvENkYgBSoLXmhwReIbAxACtSW\nvJDBF4hsDEAK1JZ2IIPvMLIxAClQW/LCEn1ByMcApEBtyRMNviDkYwBSoLbkiQy+IORjAFKgtrQL\nGXwHkY8BSIHakieW6DNHNgYgBWpL/mjwmSMbA5ACtSV/ZPCZIxsDkAK1pb36zeA5g8/U5PKZJK1b\ntYRfQABDQW0pBw0+UyyfAUiB2lIOlugzxfIZgBSoLe3HZXKF47IVAClQW8rBEn1GuGwFQArUljLR\n4DNCNgYgBWpLmcjgM0I2BiAFakteyOALRDYGIAVqS5lYos8A+RiAFKgtZaPBZ4B8DEAK1JaykcFn\ngHwMQArUljyRwReEfAxACtSWsrFE31JkYwBSoLZ0Bw2+pcjGAKRAbekOMviWIhsDkAK1JX9k8Jkj\nGwOQArWlO1iibxnyMQApUFu6hwbfMuRjAFKgtnQPGXzLkI8BSIHaUg4y+EyRjwFIgdrSPSzRtwDZ\nGIAUqC3dRoNvAbIxAClQW7qNDL4FyMYApEBtKRMZfEbIxgCkQG3pNpboG0Q+BiAFagukBhq87SNt\nf832Ftu32T633r7M9nW2t9Wfl456bKNGPgYgBWoLpAYyeNsrJa2MiJttHyrpJklnSHq9pPsj4gLb\n50laGhHvmu1r5Z7Bk48BSIHaUrZ+M/iRn8FHxK6IuLm+/ZCkrZKOkLRe0sb6sI2qmn5xepfOJvMx\nfgEBLBS1BVM1msHbXiPp+ZJukLQiInbVu3ZLWjHDY86xPW57fGJiYiTjHCaWzgCkQG3BVI1dJmf7\nEElfl/ThiLja9gMRcVjP/p9GxKw5fI5L9CydAUiB2tIdrV2ilyTbT5X0BUmXRcTV9eb76nx+Mqff\n08TYUmPpDEAK1BZM1cSr6C3pEklbI+LjPbs2SdpQ394g6ZpRjy0VLlkBkAK1BbNp4gz+xZJeK+lk\n29+rP14p6QJJL7O9TdKp9f0ikI0BSIHagtnwVrUjQDYGIAVqSzfxVrUtwttFAkiB2oLZ8Fa1CZGP\nAUiB2oJ+0OATIh8DkAK1Bf0gg0+IfAxACtSWbiODbwHyMQApUFvQD5boh4xsDEAK1BYMigY/ZGRj\nAFKgtmBQZPBDRjYGIAVqCyaRwTeEbAxACtQWDIol+iEhHwOQArUF80WDHxLyMQApUFswX2TwQ0I+\nBiAFagumIoMfMfIxAClQWzBfLNEvANkYgBSoLRgGGvwCkI0BSIHagmEgg18AsjEAKVBbMJt+M3jO\n4OdhcvlMktatWsIvIIChoLZgmGjw88DyGYAUqC0YJpbo54HlMwApUFvQDy6TS4jLVgCkQG3BMLFE\n3ycuWwGQArUFqdDg+0Q2BiAFagtSIYPvE9kYgBSoLRgUGfyQkY0BSIHaglRYop8D+RiAFKgtSI0G\nPwfyMQApUFuQGhn8HMjHAKRAbcF8kcEPCfkYgBSoLUiNJfppkI0BSIHaglGiwU+DbAxACtQWjBIZ\n/DTIxgCkQG3BMJDBLwDZGIAUqC0YJZboe5CPAUiB2oIm0OB7kI8BSIHagiaQwfcgHwOQArUFw0QG\nPw/kYwDbqUpcAAAGJ0lEQVRSoLagCSzRAwBQIBo8AAAFosEDAFAgGjwAAAWiwQMAUCAaPAAABaLB\nAwBQIBo8AAAFosEDAFAgGjwAAAWiwQMAUCAaPAAABaLBAwBQIBo8AAAFosEDAFAgGjwAAAWiwQMA\nUCAaPAAABaLBAwBQIBo8AAAFckQ0PYZ5sz0h6cdNj2MeDpf0k6YHMSSlzKWUeUjMpa1KmUsp85Dy\nnctREbF8roOybvC5sj0eEWNNj2MYSplLKfOQmEtblTKXUuYhlTWX6bBEDwBAgWjwAAAUiAbfjIua\nHsAQlTKXUuYhMZe2KmUupcxDKmsu+yCDBwCgQJzBAwBQIBp8QraPtP0121ts32b73Hr7MtvX2d5W\nf17a9Fj7ZXs/29+1/aX6fpZzsX2Y7ats3257q+0X5jgX239S/9u61fbltg/MZR62P2N7j+1be7bN\nOHbb59vebvsO2y9vZtTTm2Euf1n/+7rF9hdtH9azL6u59Oz7U9th+/CebdnNxfZb6p/Nbbb/omd7\na+cyHzT4tB6T9KcRsVbSiZLebHutpPMkbY6IYyRtru/n4lxJW3vu5zqXT0r6SkQ8W9JzVc0pq7nY\nPkLSWyWNRcRxkvaTdKbymcfnJJ02Zdu0Y69/b86UtK5+zKds7ze6oc7pc9p3LtdJOi4iniPpnySd\nL2U7F9k+UtIfSLq7Z1t2c7H9+5LWS3puRKyT9LF6e9vnMjAafEIRsSsibq5vP6SqiRyh6h/Xxvqw\njZLOaGaEg7G9WtKrJF3cszm7udheIul3JV0iSRHxq4h4QBnORdIiSU+zvUjSQZJ2KpN5RMQ3JN0/\nZfNMY18v6YqIeCQi7pS0XdIJIxloH6abS0RcGxGP1Xevl7S6vp3dXGr/VdI7JfW+cCvHubxJ0gUR\n8Uh9zJ56e6vnMh80+BGxvUbS8yXdIGlFROyqd+2WtKKhYQ3qE6p+wZ/o2ZbjXI6WNCHps3XccLHt\ng5XZXCLiXlVnH3dL2iXpwYi4VpnNY4qZxn6EpHt6jttRb8vFH0v6cn07u7nYXi/p3oj4/pRd2c1F\n0rMkvdT2Dba/bvsF9fYc5zIrGvwI2D5E0hckvS0i9vbui+oyhtZfymD7dEl7IuKmmY7JZS6qznqP\nl3RhRDxf0s80ZRk7h7nU+fR6VX+wrJJ0sO2zeo/JYR4zyXnsvWy/R1Vcd1nTY5kP2wdJerek9zY9\nliFZJGmZqtj0HZKutO1mh5QGDT4x209V1dwvi4ir68332V5Z718pac9Mj2+RF0t6te27JF0h6WTb\nlyrPueyQtCMibqjvX6Wq4ec2l1Ml3RkRExHxqKSrJb1I+c2j10xjv1fSkT3Hra63tZrt10s6XdIf\nxZPXJOc2l2eq+iPy+/Xv/2pJN9v+Z8pvLlL1+391VG5UtSJ5uPKcy6xo8AnVfxVeImlrRHy8Z9cm\nSRvq2xskXTPqsQ0qIs6PiNURsUbVC1G+GhFnKc+57JZ0j+1j602nSNqi/OZyt6QTbR9U/1s7RdXr\nPHKbR6+Zxr5J0pm2D7B9tKRjJN3YwPj6Zvs0VZHWqyPi5z27sppLRPwgIp4REWvq3/8dko6vf4+y\nmkvtbyX9viTZfpak/VX9hzM5zmV2EcFHog9JL1G1xHiLpO/VH6+U9HRVrxDeJukfJS1reqwDzusk\nSV+qb2c5F0nPkzRe/2z+VtLSHOci6QOSbpd0q6S/lnRALvOQdLmq1w48qqppnD3b2CW9R9IPJd0h\n6RVNj7+PuWxXlelO/u5/Ote5TNl/l6TDc52LqoZ+af07c7Okk3OYy3w+eCc7AAAKxBI9AAAFosED\nAFAgGjwAAAWiwQMAUCAaPAAABaLBAwBQIBo8AAAFosED6JvtF9T/v/mBtg+u/z/t45oeF4B98UY3\nAAZi+0OSDpT0NFXv6f/RhocEYBo0eAADsb2/pO9I+qWkF0XE4w0PCcA0WKIHMKinSzpE0qGqzuQB\ntBBn8AAGYnuTqv8y+GhJKyPiPzU8JADTWNT0AADkw/brJD0aEZ+3vZ+kb9k+OSK+2vTYAPwmzuAB\nACgQGTwAAAWiwQMAUCAaPAAABaLBAwBQIBo8AAAFosEDAFAgGjwAAAWiwQMAUKD/D/mpeUrYHKI4\nAAAAAElFTkSuQmCC\n", + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfQAAAI1CAYAAAAgiggGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3debRlZ1nn8e/PhBAZMpEynUrSJDK5EgRNl4iN2sigTE1YvVjd2CBBszoN2oIjg9hOrQjKYnDZYiMIUTBAR5CIYoNAQ9MSsBIxkARMGDNUSNGQBESQwNN/nH31pKpu3enss/d+z/ez1l11zj773PPuu+99nvu+v7NvpaqQJEnT9g1DD0CSJO2cDV2SpAbY0CVJaoANXZKkBtjQJUlqgA1dkqQG2NAlLUxmXpXk80k+sIn9T09SSY5cxvi2K8n3JPno0OOQDseGrpWV5JNJ/iHJF5LcnOSvkjw1yaZ+LpbVjPp+nSRHJbmo+3pUkgfv4NN9N/Bw4NSqesBiRjiT5MFJrlvk59ysqvo/VXWfzew75Di12mzoWnX/tqruCtwdeD7wLOCVww5pEO8FngTcuMPPc3fgk1X19zsfkqStsKFLQFXdUlUXA/8BODfJfQGSPDrJ3yS5Ncm1SX5p7mnv6f69OckXk3xXknskeWeS/5fks0lem+S4tSckeVaS67tVgY8meWi3/RuSPDvJx7rnviHJCeu9zoHjT/JL3XP+oPvcVyTZs8lj/8eqeklVvRf42kb7J9md5OIkn0tyTZL/1G0/D3gF8F3dOH/5EM89IskLu6/Nx4FHH/D4Dye5qjuGjyf5z932OwNvBXZ3n/uL3TgekOR93QrLviS/neSodca9ttJxfpIbuv1/Zu7xOyZ5SffYDd3tO3aP3W7W3a1m/EySy5PckuT1SY7eYJx7u++jzyR50UZfZ2nLqsoPP1byA/gk8LBDbP808LTu9oOBb2X2y+/9gM8Aj+seOx0o4Mi5596T2ZLzHYFdzJrxS7rH7gNcC+yee/49utvPAC4BTu2e+z+AC9d7nUOM+ZeALwOPAo4Afh24ZBtfk+uAB2+wz3uA3wGOBr4N2A88pHvsKcB7D/PcpwIfAU4DTgDeNX9szBr8PYAA/wb4EnD23Lm47oDP96+ABwJHdl+nq4CfWOe1176OFwJ37s7r/rXvAeBXunPwTd25+yvgvx3qtbvvnQ8Au7vjuAp46mHG+T7gh7rbdwEeOPT3vx/tfThDlw52A7MiTVX976r6UFV9vaouZ9YM/s16T6yqa6rq7VX1laraD7xobv+vMWvWZya5Q1V9sqo+1j32VOC5VXVdVX2FWYN+/BZz8/dW1Z9X1deAPwTuv4XnbkqS04AHAc+qqi9X1QeZzcqfvMlP8e+Z/YJzbVV9jtkvHv+kqv6sqj5WM+8G3gZ8z3qfrKourapLquq2qvoks1+E1j0/nV+uqr+vqg8BrwJ+sNv+ROBXquqm7tz9MvBDh/k8v1VVN3TH8afMfrlZz1eBeyY5saq+WFWXbDBGacts6NLBTgE+B5DkO5O8K8n+JLcwa7wnrvfEJCcleV23rH4r8Jq1/avqGuAnmDXrm7r9dndPvTvwpm7p+GZmM76vASdtYdzz+feXgKN7eCPdbuBzVfWFuW2fYvY12+zzrz3guf8kySOTXNIt59/MbMXhcF/veyd5S5Ibu6/38w63f+fA1187B7sPGM/8Y4dy4Nf7LofZ9zzg3sBHkvx1ksdsMEZpy2zo0pwk38GsOb232/RHwMXAaVV1LPC7zJaDYbZ8e6Dnddu/taqOYfZGs7X9qao/qqrvZtbAC3hB99C1wCOr6ri5j6Or6vp1XmcoNwAnJLnr3LZ/CVy/yefvY7bcPv9cYJZhA38MvBA4qaqOA/6cw3+9X8ZsCf9e3df75+b2X8+Br39Dd/sGZuflUI9txUHjrKqrq+oHmS3nvwC4qMvbpYWxoUtAkmO6WdPrgNd0y7EAd2U2I/1ykgcA/3HuafuBrwPfPLftrsAXgVuSnAL87Nxr3CfJQ7rG9WXgH7rnw+wXhV9Lcvdu311JzjnM6yxU94awo7u7R3Vv8DqoMVbVtcyy5V/v9rkfs9nnazb5Um8Anp7k1CTHA8+ee+woZpHEfuC2JI8Evn/u8c8Ad0ty7Ny2uwK3Al9M8i3A0zYxhv+a5E5JzgJ+GHh9t/1C4Oe7r/2JwC9s4bjmHTTOJE9Ksquqvg7c3G3++iGfLW2TDV2r7k+TfIHZDPm5zDLvH557/EeBX+n2+QVmDQmAqvoS8GvA/+2Wyh/ILHc9G7gF+DPgjXOf647MLo37LLPl2m8CntM99lJmKwFv617rEuA7D/M6W5LkrUl+7jC7fJTZLxinAP+ru333dfb9QWZvMLsBeBPwi1X1l5scyu91n/9vgcuY+/p0y/hPZ/Y1/jyzX54unnv8I8ya7se7r8Nu4Ge6/b7Qfe615nw47wauAd4BvLCq3tZt/1VgL3A58KFufL+6yeP6J+uM8xHAFUm+yOxcP6Gq/mGrn1s6nFSNaTVPkvqR5HTgE8Adquq2YUcjLZ4zdEmSGmBDlySpAS65S5LUAGfokiQ1wIYuSVIDRv1/EG/kxBNPrNNPP33oYUiStDSXXnrpZ6tq14HbJ93QTz/9dPbu3Tv0MCRJWpoknzrUdpfcJUlqgA1dkqQG2NAlSWqADV2SpAbY0CVJaoANXZKkBtjQJUlqgA1dkqQG2NAlSWqADV2SpAbY0CVJaoANXZKkBtjQJUlqgA1dkqQG2NAlSWqADV2SpAbY0CVJaoANXZKkBtjQJUlqgA1dkqQG2NAlaaKqiituuIWqGnooGgEbuiRN1JX7buVpr7mMK/fdOvRQNAI2dEmaqDNPPoaXPelszjz5mKGHohE4cugBSJK2Jwln7T526GFoJJyhS9LEmJ3rUGzokjQxZuc6FBu6JE2M2bkOxYYuSRMwv8y+lp0nGXpYGhEbuiRNgMvs2ogNXZImwGV2bcTL1iRpArxETRtxhi5JI+YlatosG7okjZjZuTbLhi5JI2Z2rs0yQ5ekETM712Y5Q5ekkTE313bY0CVpZMzNtR02dEkaGXNzbYcZuiSNjLm5tsMZuiSNhNm5dsKGLkkjYXaunbChS9JImJ1rJ8zQJWkkzM61E87QJWlA5uZaFBu6JA3I3FyLYkOXpAGZm2tRzNAlaUDm5loUZ+iSNACzcy2aDV2SBmB2rkWzoUvSAMzOtWhm6JI0ALNzLZozdElaEnNz9am3hp7k95PclOTDc9t+M8lHklye5E1Jjpt77DlJrkny0SQ/0Ne4JGko5ubqU58z9FcDjzhg29uB+1bV/YC/A54DkORM4AnAWd1zfifJET2OTZKWztxcfeqtoVfVe4DPHbDtbVV1W3f3EuDU7vY5wOuq6itV9QngGuABfY1NkpZlfpl9LTdPMvSw1KAhM/QfAd7a3T4FuHbuseu6bQdJcn6SvUn27t+/v+chStLOuMyuZRmkoSd5LnAb8NqtPreqXl5Ve6pqz65duxY/OElaIJfZtSxLv2wtyVOAxwAPrX9+q+f1wGlzu53abZOkSfPyNC3LUmfoSR4BPBN4bFV9ae6hi4EnJLljkjOAewEfWObYJGmRvERNy9bnZWsXAu8D7pPkuiTnAb8N3BV4e5IPJvldgKq6AngDcCXwF8CPVdXX+hqbJPXN7FzLlin/9rhnz57au3fv0MOQpINUFVfuu5UzTz7Gd7VroZJcWlV7Dtzun36VpB6YnWvZ/NOvkrQg5uYakg1dkhbE3FxDsqFL0oJ4zbmGZIYuSQtibq4hOUOXpB0yO9cY2NAlaYfMzjUGNnRJ2iGzc42BGbok7ZDZucbAGbokbYO5ucbGhi5J22BurrGxoUvSNpiba2zM0CVpG8zNNTbO0CVpC8zONVY2dEnaArNzjZUNXZK2wOxcY2VDl6QNzC+zr2XnSYYelnQ7NnRJ2oDL7JoCG7okbcBldk2Bl61J0ga8RE1T4AxdktbhJWqaEhu6JK3D7FxTYkOXpHWYnWtKzNAlaR1m55oSZ+iSNMfcXFNlQ5ekOebmmiobuiTNMTfXVJmhS9Icc3NNlTN0ScLsXNNnQ5ckzM41fTZ0ScLsXNNnhi5JmJ1r+pyhS1pZ5uZqiQ1d0soyN1dLbOiSVpa5uVpihi5pZZmbqyXO0CWtHLNztciGLmnlmJ2rRTZ0SSvH7FwtMkOXtHLMztUiZ+iSVoK5uVpnQ5e0EszN1TobuqSVYG6u1tnQJTVtbakd4Kzdx5Jk4BFJ/bChS2qaS+1aFTZ0SU1zqV2rwsvWJDXNS9S0KpyhS2qOl6hpFdnQJTXH3FyryIYuqTnm5lpFZuiSmmNurlXkDF1SE8zNteps6JKaYG6uVWdDl9QEc3OtOjN0SU0wN9eqc4YuadLMzqUZG7qkSTM7l2Zs6JImzexcmjFDlzRpZufSjDN0SZNjbi4dzIYuaXLMzaWD2dAlTY65uXQwM3RJk2NuLh3MGbqkyTA7l9ZnQ5c0GWbn0vps6JImw+xcWp8ZuqTJMDuX1ucMXdKomZtLm9NbQ0/y+0luSvLhuW0nJHl7kqu7f4/vtifJbyW5JsnlSc7ua1ySpsXcXNqcPmforwYeccC2ZwPvqKp7Ae/o7gM8ErhX93E+8LIexyVpQszNpc3praFX1XuAzx2w+Rzggu72BcDj5rb/Qc1cAhyX5OS+xiZp/NaW2gHO2n0sSQYekTRuy87QT6qqfd3tG4GTutunANfO7Xddt+0gSc5PsjfJ3v379/c3UkmDcqld2prB3hRXs3e4bPldLlX18qraU1V7du3a1cPIJI2BS+3S1iy7oX9mbSm9+/embvv1wGlz+53abZO0otYuUXOpXdqcZTf0i4Fzu9vnAm+e2/7k7t3uDwRumVual7QivERN2r4+L1u7EHgfcJ8k1yU5D3g+8PAkVwMP6+4D/DnwceAa4PeAH+1rXJLGy9xc2r5M+TfhPXv21N69e4cehqQFqSqu3HcrZ558jEvt0jqSXFpVew7c7p9+lTQa/mlXafv806+SBmd2Lu2cDV3S4MzOpZ2zoUsanNecSztnhi5pcGbn0s45Q5c0CHNzabFs6JIGYW4uLZYNXdIgzM2lxTJDlzQIc3NpsZyhS1oqs3OpHzZ0SUtldi71w4YuaanMzqV+mKFLWiqzc6kfztAl9c7cXOqfDV1S78zNpf7Z0CX1ztxc6p8NXVJv1pbaAc7afSxJBh6R1C4buqTeuNQuLY8NXVJvXGqXlsfL1iT1xkvUpOVxhi5pobxETRqGDV3SQpmbS8OwoUtaKHNzaRhm6JIWytxcGoYzdEk7Zm4uDc+GLmnHzM2l4dnQJe2Yubk0PDN0STtmbi4Nzxm6pG0zO5fGw4YuadvMzqXxsKFL2jazc2k8zNAlbZvZuTQeztAlbYm5uTRONnRJW2JuLo2TDV3SlpibS+Nkhi5pS8zNpXFyhi5pU8zOpXGzoUvaFLNzadxs6JI2xexcGjczdEmbYnYujZszdEnrMjeXpsOGLmld5ubSdNjQJa3L3FyaDhu6pIOsLbUDnLX7WJIMPCJJG7GhSzqIS+3S9NjQJR3EpXZperxsTdJBvERNmh5n6JIAL1GTps6GLgkwN5emzoYuCTA3l6bODF0SYG4uTZ0zdGnFmZ1LbbChSyvO7Fxqgw1dWnFm51IbzNClFWd2LrXBGbq0gszNpfbY0KUVZG4utceGLq0gc3OpPWbo0goyN5fa4wxdWiFm51K7bOjSCjE7l9plQ5dWiNm51C4zdGmFmJ1L7XKGLjXO3FxaDTZ0qXHm5tJqsKFLjTM3l1aDGbrUOHNzaTU4Q5caZXYurZZBGnqSn0xyRZIPJ7kwydFJzkjy/iTXJHl9kqOGGJvUCrNzabUsvaEnOQV4OrCnqu4LHAE8AXgB8OKquifweeC8ZY9NaonZubRahlpyPxL4xiRHAncC9gEPAS7qHr8AeNxAY5Mma36ZfS07TzL0sCQtwdIbelVdD7wQ+DSzRn4LcClwc1Xd1u12HXDKoZ6f5Pwke5Ps3b9//zKGLE2Gy+zS6hpiyf144BzgDGA3cGfgEZt9flW9vKr2VNWeXbt29TRKaZpcZpdW1xCXrT0M+ERV7QdI8kbgQcBxSY7sZumnAtcPMDZp0rxETVpdQ2TonwYemOROmYV7DwWuBN4FPL7b51zgzQOMTZKkSRoiQ38/sze/XQZ8qBvDy4FnAT+V5BrgbsArlz02SZKmapC/FFdVvwj84gGbPw48YIDhSJI0ef6lOEmSGmBDlySpATZ0SZIaYEOXJKkBNnRJkhpgQ5ckqQE2dEmSGmBDlySpATZ0SZIaYEOXJKkBNnRJkhpgQ5ckqQE2dEmSGmBDlySpATZ0SZIaYEOXJKkBNnRJkhpgQ5ckqQE2dEmSGmBDlySpATZ0SZIaYEOXJKkBNnRJkhpgQ5ckqQE2dEmSGmBDlySpATZ0SZIaYEOXJKkBNnRJkhpgQ5ckqQE2dEmSGmBDlySpATZ0SZIasGFDT/LjSY5fxmAkSdL2bGaGfhLw10nekOQRSdL3oCRJ0tZs2NCr6ueBewGvBJ4CXJ3keUnu0fPYJEnSJm0qQ6+qAm7sPm4DjgcuSvIbPY5NkiRt0pEb7ZDkGcCTgc8CrwB+tqq+muQbgKuBZ/Y7REmStJENGzpwAvDvqupT8xur6utJHtPPsCRJ0lZs2NCr6hcP89hVix2OJEnaDq9DlySpATZ0SZIaYEOXJKkBNnRJkhpgQ5ckqQE2dEmSGmBDlySpATZ0SZIaYEOXJKkBNnRJkhpgQ5ckqQE2dEmSGmBDlySpATZ0SZIaYEOXJKkBNnRJkhpgQ5ckqQE2dEmSGmBDlySpATZ0SZIaYEOXJKkBNnRJkhpgQ5ckqQE2dEmSGmBDlySpATZ0SZIaYEOXJKkBNnRJkhowSENPclySi5J8JMlVSb4ryQlJ3p7k6u7f44cYmyRJUzTUDP2lwF9U1bcA9weuAp4NvKOq7gW8o7svSZI2YekNPcmxwPcCrwSoqn+sqpuBc4ALut0uAB637LFJkrQoVcUVN9xCVS3l9YaYoZ8B7AdeleRvkrwiyZ2Bk6pqX7fPjcBJh3pykvOT7E2yd//+/UsasiRJW3Plvlt52msu48p9ty7l9YZo6EcCZwMvq6pvB/6eA5bXa/brzCF/pamql1fVnqras2vXrt4HK0nSdpx58jG87Elnc+bJxyzl9YZo6NcB11XV+7v7FzFr8J9JcjJA9+9NA4xNkqSFSMJZu48lyVJeb+kNvapuBK5Ncp9u00OBK4GLgXO7becCb1722CRJ2qllZ+drhnqX+48Dr01yOfBtwPOA5wMPT3I18LDuviRJk7Ls7HzNkUt9tU5VfRDYc4iHHrrssUiStEjLzs7XDNLQJUlq1Vp2vmz+6VetlKGyLUltG0NtsaFrpQyVbUlq2xhqS6Y8U9mzZ0/t3bt36GFoQqqKK/fdypknH7O0S0kktW+ZtSXJpVV10PvQzNC1UobKtiS1bQy1xSV3rYQx5FuS2jOm2mJD10oYQ74lqT1jqi1m6FoJZueS+jBEbTFD10obQ74lqT1jqi0uuatZY8q2JLVjrLXFhq5mjSnbktSOsdYWM3Q1y9xcUh+Gri1m6Fo5Y8q2JLVjrLXFJXc1Z6z5lqRpG3ttsaGrOWPNtyRN29hrixm6mjN0viWpTWOpLetl6M7Q1YT5pbC1fMtmLmmnplRbbOhqwtiXwiRN05Rqi0vuasJYlsIktWWMtcXL1tS0sV5GImnaplRbXHLXpI39MhJJ0zTF2mJD16RNKd+SNB1TrC1m6Jq0MeZbkqZvzLXFDF1NmlK+JWk6plhbXHLX5Ewx25I0flOvLTZ0Tc4Usy1J4zf12mKGrskZc7YlabqmUlvM0NWMKWZbksZv6rXFJXdNwtSzLUnj1FJtsaFrEqaebUkap5Zqixm6JmEq2ZakaZlibTFD16RNPduSNE4t1RaX3DVqLeVbksajxdpiQ9eotZRvSRqPFmuLGbpGbYr5lqTxm3JtMUPXJLWUb0kajxZri0vuGp0Wsy1Jw2u9ttjQNTotZluShtd6bTFD1+hMOduSNF6t1BYzdE1Gi9mWpOG1XltcctdotJ5vSRrGqtQWG7pGo/V8S9IwVqW2mKFrNFrJtySNS2u1Zb0M3Rm6BjW/FLaWb7XwAydpWKtYW2zoGtSqLIVJWq5VrC0uuWtQrS2FSRqHlmuLl61plFq/jETSMFaxtrjkrkGsymUkkpZrlWuLDV2DWMV8S1L/Vrm2mKFrEC3nW5KGswq1xQxdo7KK+Zak/q1ybXHJXUuzytmWpP5YW2Zs6FqaVc62JPXH2jJjhq6lWYVsS9LyrVptMUPX4FY525LUH2vLjEvu6p35lqQ+WFtuz4au3plvSeqDteX2zNDVu1XLtyQtx6rWFjN0DcZ8S1IfrC2355K7emG2JakP1pb12dDVC7MtSX2wtqzPDF29WNVsS1K/rC1m6Foysy1JfbC2rM8ldy2U+ZakPlhbNmZD10KZb0nqg7VlY2boWijzLUl9sLb8MzN0LYX5lqQ+WFs25pK7dsxsS1IfrC1bM1hDT3JEkr9J8pbu/hlJ3p/kmiSvT3LUUGPT1phtSeqDtWVrhpyhPwO4au7+C4AXV9U9gc8D5w0yKm3ZmScfw8uedDZnnnzM0EOR1BBry9YM0tCTnAo8GnhFdz/AQ4CLul0uAB43xNi0eWvLYQBn7T525d+oImkxrC3bM9QM/SXAM4Gvd/fvBtxcVbd1968DTjnUE5Ocn2Rvkr379+/vf6Ral8thkvpgbdmepTf0JI8BbqqqS7fz/Kp6eVXtqao9u3btWvDotBUuh0nqg7Vle4a4bO1BwGOTPAo4GjgGeClwXJIju1n6qcD1A4xNW+BlJJL6YG3ZnqXP0KvqOVV1alWdDjwBeGdVPRF4F/D4brdzgTcve2zamJeRSOqDtWXnxnQd+rOAn0pyDbNM/ZUDj0eHYLYlqQ/Wlp3zT79qS/zzi5L6YG3ZPP/0qxbCbEtSH6wtOzemJXeNmPmWpD5YWxbHhq5NMd+S1Adry+KYoWtTzLck9cHasnVm6NoR8y1JfbC2LI5L7lqX2ZakPlhb+mFD17rMtiT1wdrSDzN0rctsS1IfrC07Y4auLTPbktQHa0s/XHLX7ZhtSeqDtaV/NnTdjtmWpD5YW/pnhq7bMduS1Adry+KYoWtTzLYk9cHa0j+X3AWYb0nqh7VleWzoAsy3JPXD2rI8ZugCzLck9cPasnjrZejO0FfY/FLYWr7lD5yknbK2DMOGvsJcCpPUB2vLMFxyX2EuhUnqg7WlX162poN4GYmkPlhbhuGS+wryMhJJfbC2DMuGvoLMtyT1wdoyLDP0FWS+JakP1pblMEPXPzHfktQHa8uwXHJfEWZbkvpgbRkPG/qKMNuS1Adry3iYoa8Isy1JfbC2LJ8Z+ooz25LUB2vLeLjk3jjzLUl9sLaMjw29ceZbkvpgbRkfM/TGmW9J6oO1ZThm6CvKfEtSH6wt4+OSe4PMtiT1wdoybjb0BpltSeqDtWXczNAbZLYlqQ/WlnEwQ18hZluS+mBtGTeX3BtiviWpD9aWabChN8R8S1IfrC3TYIbeEPMtSX2wtoyLGfoKMN+S1AdryzS45D5xZluS+mBtmR4b+sSZbUnqg7VleszQJ85sS1IfrC3jtV6G7gx9otaWwwDO2n2sP3CSFsLaMl029IlyOUxSH6wt0+WS+0S5HCapD9aW8fOytcZ4GYmkPlhbpssl9wnxMhJJfbC2tMGGPiFmW5L6YG1pgxn6hJhtSeqDtWVazNAbYLYlqQ/Wlja45D4B5luS+mBtaYsNfQLMtyT1wdrSFjP0CTDfktQHa8s0maFPmPmWpD5YW9rikvtImW1J6oO1pV029JEy25LUB2tLu8zQR8psS1IfrC3TZ4Y+MWZbkvpgbWmXS+4jYrYlqQ/WltVgQx8Rsy1JfbC2rAYz9BEx25LUB2tLW8zQJ8BsS1IfrC2rwSX3ETDfktQHa8tqsaGPgPmWpD5YW1aLGfoImG9J6oO1pU1m6CNmviWpD9aW1eKS+0DMtiT1wdqyupbe0JOcluRdSa5MckWSZ3TbT0jy9iRXd/8ev+yxLZPZlqQ+WFtW19Iz9CQnAydX1WVJ7gpcCjwOeArwuap6fpJnA8dX1bMO97mmnKGbbUnqg7Wlfetl6EufoVfVvqq6rLv9BeAq4BTgHOCCbrcLmDX55qwthwGctftYf+AkLYS1RYNm6ElOB74deD9wUlXt6x66EThpneecn2Rvkr379+9fyjgXyeUwSX2wtmiwy9aS3AV4N/BrVfXGJDdX1XFzj3++qg6bo09xyd3lMEl9sLasjtEsuXeDuQPwx8Brq+qN3ebPdPn6Ws5+0xBj69vaZST+wElaJGuLhniXe4BXAldV1YvmHroYOLe7fS7w5mWPrS9eRiKpD9YWzRtihv4g4IeAhyT5YPfxKOD5wMOTXA08rLvfBLMtSX2wtmief/p1Ccy2JPXB2rKa/NOvA/LPL0rqg7VF8/zTrz0y35LUB2uLDsWG3iPzLUl9sLboUMzQe2S+JakP1pbVZoY+APMtSX2wtuhQXHJfMLMtSX2wtmgjNvQFM9uS1AdrizZihr5gZluS+mBt0Roz9CUx25LUB2uLNuKS+4KYb0nqg7VFm2VDXxDzLUl9sLZos8zQF8R8S1IfrC06kBl6z8y3JPXB2qLNcsl9B8y2JPXB2qLtsKHvgNmWpD5YW7QdZug7YLYlqQ/WFh3Oehm6M/RtWFsOAzhr97H+wElaCGuLdsKGvg0uh0nqg7VFO+GS+za4HCapD9YWbYaXrS2Ql5FI6oO1RTvhkvsmeRmJpD5YW7QoNvRNMtuS1AdrixbFDH2TzLYk9cHaoq0yQ98hsy1JfbC2aFFcct+A+ZakPlhbtGg29A2Yb0nqg7VFi2aGvgHzLUl9sLZou8zQt8l8S1IfrC1aNJfcD8FsS1IfrC3qkw39EMy2JPXB2qI+maEfgtmWpD5YW7QIZuhbYLYlqQ/WFvXJJfeO2ZakPlhbtCw29I7ZlqQ+WFu0LGboHbMtSX2wtmjRzNA3YLYlqQ/WFqz1kZ4AAAaKSURBVC2LS+6SJDXAhi5JUgNs6JIkNcCGLklSA2zokiQ1wIYuSVIDbOiSJDXAhi5JUgNs6JIkNcCGLklSA2zokiQ1wIYuSVIDbOiSJDXAhi5JUgNs6JIkNcCGLklSA2zokiQ1wIYuSVIDbOiSJDXAhi5JUgNSVUOPYduS7Ac+NfQ4tuFE4LNDD2JBWjmWVo4DPJYxauU4wGMZg7tX1a4DN066oU9Vkr1VtWfocSxCK8fSynGAxzJGrRwHeCxj5pK7JEkNsKFLktQAG/owXj70ABaolWNp5TjAYxmjVo4DPJbRMkOXJKkBztAlSWqADb1HSU5L8q4kVya5Iskzuu0nJHl7kqu7f48feqybleSIJH+T5C3d/TOSvD/JNUlen+Sooce4GUmOS3JRko8kuSrJd03xvCT5ye5768NJLkxy9FTOSZLfT3JTkg/PbTvkOcjMb3XHdHmSs4cb+cHWOZbf7L6/Lk/ypiTHzT32nO5YPprkB4YZ9aEd6ljmHvvpJJXkxO7+5M5Lt/3Hu3NzRZLfmNs+2vOyGTb0ft0G/HRVnQk8EPixJGcCzwbeUVX3At7R3Z+KZwBXzd1/AfDiqron8HngvEFGtXUvBf6iqr4FuD+zY5rUeUlyCvB0YE9V3Rc4AngC0zknrwYeccC29c7BI4F7dR/nAy9b0hg369UcfCxvB+5bVfcD/g54DkBXA54AnNU953eSHLG8oW7o1Rx8LCQ5Dfh+4NNzmyd3XpJ8H3AOcP+qOgt4Ybd97OdlQzb0HlXVvqq6rLv9BWZN4xRm30wXdLtdADxumBFuTZJTgUcDr+juB3gIcFG3yySOJcmxwPcCrwSoqn+sqpuZ5nk5EvjGJEcCdwL2MZFzUlXvAT53wOb1zsE5wB/UzCXAcUlOXs5IN3aoY6mqt1XVbd3dS4BTu9vnAK+rqq9U1SeAa4AHLG2wG1jnvAC8GHgmMP/Gq8mdF+BpwPOr6ivdPjd120d9XjbDhr4kSU4Hvh14P3BSVe3rHroROGmgYW3VS5j9QH+9u3834Oa5onUds19Yxu4MYD/wqi4+eEWSOzOx81JV1zObXXyaWSO/BbiUaZ6TNeudg1OAa+f2m9px/Qjw1u725I4lyTnA9VX1twc8NLljAe4NfE8XS707yXd026d4LLdjQ1+CJHcB/hj4iaq6df6xml1mMPpLDZI8Bripqi4deiwLcCRwNvCyqvp24O85YHl9Cuely5fPYfYLym7gzhxiqXSqpnAONiPJc5nFb68deizbkeROwM8BvzD0WBbkSOAEZjHozwJv6FYbJ8+G3rMkd2DWzF9bVW/sNn9mbVmq+/em9Z4/Ig8CHpvkk8DrmC3rvpTZEtuR3T6nAtcPM7wtuQ64rqre392/iFmDn9p5eRjwiaraX1VfBd7I7DxN8ZysWe8cXA+cNrffJI4ryVOAxwBPrH++Rnhqx3IPZr80/m33838qcFmSf8H0jgVmP/9v7GKCDzBbcTyRaR7L7djQe9T91vdK4KqqetHcQxcD53a3zwXevOyxbVVVPaeqTq2q05m9ceSdVfVE4F3A47vdpnIsNwLXJrlPt+mhwJVM77x8Gnhgkjt132trxzG5czJnvXNwMfDk7l3VDwRumVuaH6Ukj2AWUT22qr4099DFwBOS3DHJGczeUPaBIca4GVX1oar6pqo6vfv5vw44u/s5mtx5Af4E+D6AJPcGjmL2H7RM6rwcUlX50dMH8N3MlgwvBz7YfTyKWfb8DuBq4C+BE4Ye6xaP68HAW7rb38zsm/4a4H8Cdxx6fJs8hm8D9nbn5k+A46d4XoBfBj4CfBj4Q+COUzknwIXMsv+vMmsS5613DoAA/x34GPAhZu/sH/wYNjiWa5hlsms/+787t/9zu2P5KPDIoce/0bEc8PgngRMnfF6OAl7T/cxcBjxkCudlMx/+pThJkhrgkrskSQ2woUuS1AAbuiRJDbChS5LUABu6JEkNsKFLktQAG7okSQ2woUvatCTf0f2/10cnuXP3/0nfd+hxScI/LCNpa5L8KnA08I3M/ib+rw88JEnY0CVtUZKjgL8Gvgz866r62sBDkoRL7pK27m7AXYC7MpupSxoBZ+iStiTJxcz+C90zgJOr6r8MPCRJzP6jd0nalCRPBr5aVX+U5Ajgr5I8pKreOfTYpFXnDF2SpAaYoUuS1AAbuiRJDbChS5LUABu6JEkNsKFLktQAG7okSQ2woUuS1AAbuiRJDfj/sZw/giO30nUAAAAASUVORK5CYII=\n", "text/plain": [ - "" + "
" ] }, - "metadata": {}, + "metadata": { + "needs_background": "light" + }, "output_type": "display_data" } ], @@ -150,12 +154,14 @@ "outputs": [ { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAABJYAAAKzCAYAAABMJofMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nH+ab4wDoeeRVAAAAAElFTkSuQmCC\n", + "image/png": "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\n", "text/plain": [ - "" + "
" ] }, - "metadata": {}, + "metadata": { + "needs_background": "light" + }, "output_type": "display_data" } ], @@ -191,12 +197,14 @@ "outputs": [ { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfgAAAI1CAYAAAA6tuiIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n", + "image/png": "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\n", "text/plain": [ - "" + "
" ] }, - "metadata": {}, + "metadata": { + "needs_background": "light" + }, "output_type": "display_data" } ], @@ -228,7 +236,7 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 5, "metadata": {}, "outputs": [ { @@ -602,38 +610,44 @@ "outputs": [ { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfgAAAI1CAYAAAA6tuiIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n", + "image/png": "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\n", "text/plain": [ - "" + "
" ] }, - "metadata": {}, + "metadata": { + "needs_background": "light" + }, "output_type": "display_data" } ], "source": [ "import matplotlib.pyplot as plt\n", "\n", - "fig = plt.figure(figsize=(8,8))\n", + "def plot(data, n_samples, kmeans, title):\n", "\n", - "ax = fig.add_subplot(111)\n", - "fig.subplots_adjust(top=1)\n", - "ax.set_title('Clustered points in dataset n. 1')\n", + " fig = plt.figure(figsize=(8,8))\n", "\n", - "ax.set_xlabel('x')\n", - "ax.set_ylabel('y')\n", + " ax = fig.add_subplot(111)\n", + " fig.subplots_adjust(top=1)\n", + " ax.set_title(title)\n", "\n", - "# set the list of colors to be selected when plotting the different clusters\n", - "color=['b','g','r','c','m','y','k','w']\n", + " ax.set_xlabel('x')\n", + " ax.set_ylabel('y')\n", + "\n", + " # set the list of colors to be selected when plotting the different clusters\n", + " color=['b','g','r','c','m','y','k','w']\n", + "\n", + " #plot the dataset\n", + " for clu in range(k):\n", + " # collect the sequence of cooordinates of the points in each given cluster (determined by clu)\n", + " data_list_x = [data[i,0] for i in range(n_samples) if kmeans.labels_[i]==clu]\n", + " data_list_y = [data[i,1] for i in range(n_samples) if kmeans.labels_[i]==clu]\n", + " plt.scatter(data_list_x, data_list_y, s=2, edgecolors='none', c=color[clu], alpha=0.5)\n", + "\n", + " plt.show()\n", " \n", - "#plot the dataset\n", - "for clu in range(k):\n", - " # collect the sequence of cooordinates of the points in each given cluster (determined by clu)\n", - " data_list_x = [data1[i,0] for i in range(n_samples1) if kmeans1.labels_[i]==clu]\n", - " data_list_y = [data1[i,1] for i in range(n_samples1) if kmeans1.labels_[i]==clu]\n", - " plt.scatter(data_list_x, data_list_y, s=2, edgecolors='none', c=color[clu], alpha=0.5)\n", - "\n", - "plt.show()" + "plot(data1, n_samples1, kmeans1, 'Clustered points in dataset n. 1')" ] }, { @@ -647,10 +661,6163 @@ }, { "cell_type": "code", - "execution_count": 16, + "execution_count": 7, "metadata": {}, - "outputs": [], - "source": [] + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Example n. 0 = (5.0, 482.0): cluster 0\n", + "Example n. 1 = (5.0, 481.0): cluster 0\n", + "Example n. 2 = (5.0, 480.0): cluster 0\n", + "Example n. 3 = (5.0, 479.0): cluster 0\n", + "Example n. 4 = (5.0, 478.0): cluster 0\n", + "Example n. 5 = (5.0, 477.0): cluster 0\n", + "Example n. 6 = (5.0, 476.0): cluster 0\n", + "Example n. 7 = (5.0, 475.0): cluster 0\n", + "Example n. 8 = (5.0, 474.0): cluster 0\n", + "Example n. 9 = (5.0, 473.0): cluster 0\n", + "Example n. 10 = (5.0, 472.0): cluster 0\n", + "Example n. 11 = (5.0, 471.0): cluster 0\n", + "Example n. 12 = (5.0, 470.0): cluster 0\n", + "Example n. 13 = (5.0, 469.0): cluster 0\n", + "Example n. 14 = (5.0, 468.0): cluster 0\n", + "Example n. 15 = (5.0, 467.0): cluster 0\n", + "Example n. 16 = (5.0, 466.0): cluster 0\n", + "Example n. 17 = (5.0, 465.0): cluster 0\n", + "Example n. 18 = (5.0, 464.0): cluster 0\n", + "Example n. 19 = (5.0, 463.0): cluster 0\n", + "Example n. 20 = (5.0, 462.0): cluster 0\n", + "Example n. 21 = (5.0, 461.0): cluster 0\n", + "Example n. 22 = (5.0, 460.0): cluster 0\n", + "Example n. 23 = (6.0, 482.0): cluster 0\n", + "Example n. 24 = (6.0, 481.0): cluster 0\n", + "Example n. 25 = (6.0, 480.0): cluster 0\n", + "Example n. 26 = (6.0, 479.0): cluster 0\n", + "Example n. 27 = (6.0, 478.0): cluster 0\n", + "Example n. 28 = (6.0, 477.0): cluster 0\n", + "Example n. 29 = (6.0, 476.0): cluster 0\n", + "Example n. 30 = (6.0, 475.0): cluster 0\n", + "Example n. 31 = (6.0, 474.0): cluster 0\n", + "Example n. 32 = (6.0, 473.0): cluster 0\n", + "Example n. 33 = (6.0, 472.0): cluster 0\n", + "Example n. 34 = (6.0, 471.0): cluster 0\n", + "Example n. 35 = (6.0, 470.0): cluster 0\n", + "Example n. 36 = (6.0, 469.0): cluster 0\n", + "Example n. 37 = (6.0, 468.0): cluster 0\n", + "Example n. 38 = (6.0, 467.0): cluster 0\n", + "Example n. 39 = (6.0, 466.0): cluster 0\n", + "Example n. 40 = (6.0, 465.0): cluster 0\n", + "Example n. 41 = (6.0, 464.0): cluster 0\n", + "Example n. 42 = (6.0, 463.0): cluster 0\n", + "Example n. 43 = (6.0, 462.0): cluster 0\n", + "Example n. 44 = (6.0, 461.0): cluster 0\n", + "Example n. 45 = (6.0, 460.0): cluster 0\n", + "Example n. 46 = (7.0, 482.0): cluster 0\n", + "Example n. 47 = (7.0, 481.0): cluster 0\n", + "Example n. 48 = (7.0, 480.0): cluster 0\n", + "Example n. 49 = (7.0, 479.0): cluster 0\n", + "Example n. 50 = (7.0, 478.0): cluster 0\n", + "Example n. 51 = (7.0, 477.0): cluster 0\n", + "Example n. 52 = (7.0, 476.0): cluster 0\n", + "Example n. 53 = (7.0, 475.0): cluster 0\n", + "Example n. 54 = (7.0, 474.0): cluster 0\n", + "Example n. 55 = (7.0, 473.0): cluster 0\n", + "Example n. 56 = (7.0, 472.0): cluster 0\n", + "Example n. 57 = (7.0, 471.0): cluster 0\n", + "Example n. 58 = (7.0, 470.0): cluster 0\n", + "Example n. 59 = (7.0, 469.0): cluster 0\n", + "Example n. 60 = (7.0, 468.0): cluster 0\n", + "Example n. 61 = (7.0, 467.0): cluster 0\n", + "Example n. 62 = (7.0, 466.0): cluster 0\n", + "Example n. 63 = (7.0, 465.0): cluster 0\n", + "Example n. 64 = (7.0, 464.0): cluster 0\n", + "Example n. 65 = (7.0, 463.0): cluster 0\n", + "Example n. 66 = (7.0, 462.0): cluster 0\n", + "Example n. 67 = (7.0, 461.0): cluster 0\n", + "Example n. 68 = (7.0, 460.0): cluster 0\n", + "Example n. 69 = (8.0, 490.0): cluster 0\n", + "Example n. 70 = (8.0, 489.0): cluster 0\n", + "Example n. 71 = (8.0, 488.0): cluster 0\n", + "Example n. 72 = (8.0, 487.0): cluster 0\n", + "Example n. 73 = (8.0, 486.0): cluster 0\n", + "Example n. 74 = (8.0, 485.0): cluster 0\n", + "Example n. 75 = (8.0, 484.0): cluster 0\n", + "Example n. 76 = (8.0, 483.0): cluster 0\n", + "Example n. 77 = (8.0, 482.0): cluster 0\n", + "Example n. 78 = (8.0, 481.0): cluster 0\n", + "Example n. 79 = (8.0, 480.0): cluster 0\n", + "Example n. 80 = (8.0, 479.0): cluster 0\n", + "Example n. 81 = (8.0, 478.0): cluster 0\n", + "Example n. 82 = (8.0, 477.0): cluster 0\n", + "Example n. 83 = (8.0, 476.0): cluster 0\n", + "Example n. 84 = (8.0, 475.0): cluster 0\n", + "Example n. 85 = (8.0, 474.0): cluster 0\n", + "Example n. 86 = (8.0, 473.0): cluster 0\n", + "Example n. 87 = (8.0, 472.0): cluster 0\n", + "Example n. 88 = (8.0, 471.0): cluster 0\n", + "Example n. 89 = (8.0, 470.0): cluster 0\n", + "Example n. 90 = (8.0, 469.0): cluster 0\n", + "Example n. 91 = (8.0, 468.0): cluster 0\n", + "Example n. 92 = (8.0, 467.0): cluster 0\n", + "Example n. 93 = (8.0, 466.0): cluster 0\n", + "Example n. 94 = (8.0, 465.0): cluster 0\n", + "Example n. 95 = (8.0, 464.0): cluster 0\n", + "Example n. 96 = (8.0, 463.0): cluster 0\n", + "Example n. 97 = (8.0, 462.0): cluster 0\n", + "Example n. 98 = (8.0, 461.0): cluster 0\n", + "Example n. 99 = (8.0, 460.0): cluster 0\n", + "Example n. 100 = (8.0, 459.0): cluster 0\n", + "Example n. 101 = (8.0, 458.0): cluster 0\n", + "Example n. 102 = (8.0, 457.0): cluster 0\n", + "Example n. 103 = (8.0, 456.0): cluster 0\n", + "Example n. 104 = (9.0, 490.0): cluster 0\n", + "Example n. 105 = (9.0, 489.0): cluster 0\n", + "Example n. 106 = (9.0, 488.0): cluster 0\n", + "Example n. 107 = (9.0, 487.0): cluster 0\n", + "Example n. 108 = (9.0, 486.0): cluster 0\n", + "Example n. 109 = (9.0, 485.0): cluster 0\n", + "Example n. 110 = (9.0, 484.0): cluster 0\n", + "Example n. 111 = (9.0, 483.0): cluster 0\n", + "Example n. 112 = (9.0, 482.0): cluster 0\n", + "Example n. 113 = (9.0, 481.0): cluster 0\n", + "Example n. 114 = (9.0, 480.0): cluster 0\n", + "Example n. 115 = (9.0, 479.0): cluster 0\n", + "Example n. 116 = (9.0, 478.0): cluster 0\n", + "Example n. 117 = (9.0, 477.0): cluster 0\n", + "Example n. 118 = (9.0, 476.0): cluster 0\n", + "Example n. 119 = (9.0, 475.0): cluster 0\n", + "Example n. 120 = (9.0, 474.0): cluster 0\n", + "Example n. 121 = (9.0, 473.0): cluster 0\n", + "Example n. 122 = (9.0, 472.0): cluster 0\n", + "Example n. 123 = (9.0, 471.0): cluster 0\n", + "Example n. 124 = (9.0, 470.0): cluster 0\n", + "Example n. 125 = (9.0, 469.0): cluster 0\n", + "Example n. 126 = (9.0, 468.0): cluster 0\n", + "Example n. 127 = (9.0, 467.0): cluster 0\n", + "Example n. 128 = (9.0, 466.0): cluster 0\n", + "Example n. 129 = (9.0, 465.0): cluster 0\n", + "Example n. 130 = (9.0, 464.0): cluster 0\n", + "Example n. 131 = (9.0, 463.0): cluster 0\n", + "Example n. 132 = (9.0, 462.0): cluster 0\n", + "Example n. 133 = (9.0, 461.0): cluster 0\n", + "Example n. 134 = (9.0, 460.0): cluster 0\n", + "Example n. 135 = (9.0, 459.0): cluster 0\n", + "Example n. 136 = (9.0, 458.0): cluster 0\n", + "Example n. 137 = (9.0, 457.0): cluster 0\n", + "Example n. 138 = (9.0, 456.0): cluster 0\n", + "Example n. 139 = (10.0, 490.0): cluster 0\n", + "Example n. 140 = (10.0, 489.0): cluster 0\n", + "Example n. 141 = (10.0, 488.0): cluster 0\n", + "Example n. 142 = (10.0, 487.0): cluster 0\n", + "Example n. 143 = (10.0, 486.0): cluster 0\n", + "Example n. 144 = (10.0, 485.0): cluster 0\n", + "Example n. 145 = (10.0, 484.0): cluster 0\n", + "Example n. 146 = (10.0, 483.0): cluster 0\n", + "Example n. 147 = (10.0, 482.0): cluster 0\n", + "Example n. 148 = (10.0, 481.0): cluster 0\n", + "Example n. 149 = (10.0, 480.0): cluster 0\n", + "Example n. 150 = (10.0, 479.0): cluster 0\n", + "Example n. 151 = (10.0, 478.0): cluster 0\n", + "Example n. 152 = (10.0, 477.0): cluster 0\n", + "Example n. 153 = (10.0, 476.0): cluster 0\n", + "Example n. 154 = (10.0, 475.0): cluster 0\n", + "Example n. 155 = (10.0, 474.0): cluster 0\n", + "Example n. 156 = (10.0, 473.0): cluster 0\n", + "Example n. 157 = (10.0, 472.0): cluster 0\n", + "Example n. 158 = (10.0, 471.0): cluster 0\n", + "Example n. 159 = (10.0, 470.0): cluster 0\n", + "Example n. 160 = (10.0, 469.0): cluster 0\n", + "Example n. 161 = (10.0, 468.0): cluster 0\n", + "Example n. 162 = (10.0, 467.0): cluster 0\n", + "Example n. 163 = (10.0, 466.0): cluster 0\n", + "Example n. 164 = (10.0, 465.0): cluster 0\n", + "Example n. 165 = (10.0, 464.0): cluster 0\n", + "Example n. 166 = (10.0, 463.0): cluster 0\n", + "Example n. 167 = (10.0, 462.0): cluster 0\n", + "Example n. 168 = (10.0, 461.0): cluster 0\n", + "Example n. 169 = (10.0, 460.0): cluster 0\n", + "Example n. 170 = (10.0, 459.0): cluster 0\n", + "Example n. 171 = (10.0, 458.0): cluster 0\n", + "Example n. 172 = (10.0, 457.0): cluster 0\n", + "Example n. 173 = (10.0, 456.0): cluster 0\n", + "Example n. 174 = (11.0, 490.0): cluster 0\n", + "Example n. 175 = (11.0, 489.0): cluster 0\n", + "Example n. 176 = (11.0, 488.0): cluster 0\n", + "Example n. 177 = (11.0, 487.0): cluster 0\n", + "Example n. 178 = (11.0, 486.0): cluster 0\n", + "Example n. 179 = (11.0, 485.0): cluster 0\n", + "Example n. 180 = (11.0, 484.0): cluster 0\n", + "Example n. 181 = (11.0, 483.0): cluster 0\n", + "Example n. 182 = (11.0, 482.0): cluster 0\n", + "Example n. 183 = (11.0, 481.0): cluster 0\n", + "Example n. 184 = (11.0, 480.0): cluster 0\n", + "Example n. 185 = (11.0, 479.0): cluster 0\n", + "Example n. 186 = (11.0, 478.0): cluster 0\n", + "Example n. 187 = (11.0, 477.0): cluster 0\n", + "Example n. 188 = (11.0, 476.0): cluster 0\n", + "Example n. 189 = (11.0, 475.0): cluster 0\n", + "Example n. 190 = (11.0, 474.0): cluster 0\n", + "Example n. 191 = (11.0, 473.0): cluster 0\n", + "Example n. 192 = (11.0, 472.0): cluster 0\n", + "Example n. 193 = (11.0, 471.0): cluster 0\n", + "Example n. 194 = (11.0, 470.0): cluster 0\n", + "Example n. 195 = (11.0, 469.0): cluster 0\n", + "Example n. 196 = (11.0, 468.0): cluster 0\n", + "Example n. 197 = (11.0, 467.0): cluster 0\n", + "Example n. 198 = (11.0, 466.0): cluster 0\n", + "Example n. 199 = (11.0, 465.0): cluster 0\n", + "Example n. 200 = (11.0, 464.0): cluster 0\n", + "Example n. 201 = (11.0, 463.0): cluster 0\n", + "Example n. 202 = (11.0, 462.0): cluster 0\n", + "Example n. 203 = (11.0, 461.0): cluster 0\n", + "Example n. 204 = (11.0, 460.0): cluster 0\n", + "Example n. 205 = (11.0, 459.0): cluster 0\n", + "Example n. 206 = (11.0, 458.0): cluster 0\n", + "Example n. 207 = (11.0, 457.0): cluster 0\n", + "Example n. 208 = (11.0, 456.0): cluster 0\n", + "Example n. 209 = (12.0, 494.0): cluster 0\n", + "Example n. 210 = (12.0, 493.0): cluster 0\n", + "Example n. 211 = (12.0, 492.0): cluster 0\n", + "Example n. 212 = (12.0, 491.0): cluster 0\n", + "Example n. 213 = (12.0, 490.0): cluster 0\n", + "Example n. 214 = (12.0, 489.0): cluster 0\n", + "Example n. 215 = (12.0, 488.0): cluster 0\n", + "Example n. 216 = (12.0, 487.0): cluster 0\n", + "Example n. 217 = (12.0, 486.0): cluster 0\n", + "Example n. 218 = (12.0, 485.0): cluster 0\n", + "Example n. 219 = (12.0, 484.0): cluster 0\n", + "Example n. 220 = (12.0, 483.0): cluster 0\n", + "Example n. 221 = (12.0, 482.0): cluster 0\n", + "Example n. 222 = (12.0, 481.0): cluster 0\n", + "Example n. 223 = (12.0, 480.0): cluster 0\n", + "Example n. 224 = (12.0, 479.0): cluster 0\n", + "Example n. 225 = (12.0, 478.0): cluster 0\n", + "Example n. 226 = (12.0, 477.0): cluster 0\n", + "Example n. 227 = (12.0, 476.0): cluster 0\n", + "Example n. 228 = (12.0, 475.0): cluster 0\n", + "Example n. 229 = (12.0, 474.0): cluster 0\n", + "Example n. 230 = (12.0, 473.0): cluster 0\n", + "Example n. 231 = (12.0, 472.0): cluster 0\n", + "Example n. 232 = (12.0, 471.0): cluster 0\n", + "Example n. 233 = (12.0, 470.0): cluster 0\n", + "Example n. 234 = (12.0, 469.0): cluster 0\n", + "Example n. 235 = (12.0, 468.0): cluster 0\n", + "Example n. 236 = (12.0, 467.0): cluster 0\n", + "Example n. 237 = (12.0, 466.0): cluster 0\n", + "Example n. 238 = (12.0, 465.0): cluster 0\n", + "Example n. 239 = (12.0, 464.0): cluster 0\n", + "Example n. 240 = (12.0, 463.0): cluster 0\n", + "Example n. 241 = (12.0, 462.0): cluster 0\n", + "Example n. 242 = (12.0, 461.0): cluster 0\n", + "Example n. 243 = (12.0, 460.0): cluster 0\n", + "Example n. 244 = (12.0, 459.0): cluster 0\n", + "Example n. 245 = (12.0, 458.0): cluster 0\n", + "Example n. 246 = (12.0, 457.0): cluster 0\n", + "Example n. 247 = (12.0, 456.0): cluster 0\n", + "Example n. 248 = (13.0, 494.0): cluster 0\n", + "Example n. 249 = (13.0, 493.0): cluster 0\n", + "Example n. 250 = (13.0, 492.0): cluster 0\n", + "Example n. 251 = (13.0, 491.0): cluster 0\n", + "Example n. 252 = (13.0, 490.0): cluster 0\n", + "Example n. 253 = (13.0, 489.0): cluster 0\n", + "Example n. 254 = (13.0, 488.0): cluster 0\n", + "Example n. 255 = (13.0, 487.0): cluster 0\n", + "Example n. 256 = (13.0, 486.0): cluster 0\n", + "Example n. 257 = (13.0, 485.0): cluster 0\n", + "Example n. 258 = (13.0, 484.0): cluster 0\n", + "Example n. 259 = (13.0, 483.0): cluster 0\n", + "Example n. 260 = (13.0, 482.0): cluster 0\n", + "Example n. 261 = (13.0, 481.0): cluster 0\n", + "Example n. 262 = (13.0, 480.0): cluster 0\n", + "Example n. 263 = (13.0, 479.0): cluster 0\n", + "Example n. 264 = (13.0, 478.0): cluster 0\n", + "Example n. 265 = (13.0, 477.0): cluster 0\n", + "Example n. 266 = (13.0, 476.0): cluster 0\n", + "Example n. 267 = (13.0, 475.0): cluster 0\n", + "Example n. 268 = (13.0, 474.0): cluster 0\n", + "Example n. 269 = (13.0, 473.0): cluster 0\n", + "Example n. 270 = (13.0, 472.0): cluster 0\n", + "Example n. 271 = (13.0, 471.0): cluster 0\n", + "Example n. 272 = (13.0, 470.0): cluster 0\n", + "Example n. 273 = (13.0, 469.0): cluster 0\n", + "Example n. 274 = (13.0, 468.0): cluster 0\n", + "Example n. 275 = (13.0, 467.0): cluster 0\n", + "Example n. 276 = (13.0, 466.0): cluster 0\n", + "Example n. 277 = (13.0, 465.0): cluster 0\n", + "Example n. 278 = (13.0, 464.0): cluster 0\n", + "Example n. 279 = (13.0, 463.0): cluster 0\n", + "Example n. 280 = (13.0, 462.0): cluster 0\n", + "Example n. 281 = (13.0, 461.0): cluster 0\n", + "Example n. 282 = (13.0, 460.0): cluster 0\n", + "Example n. 283 = (13.0, 459.0): cluster 0\n", + "Example n. 284 = (13.0, 458.0): cluster 0\n", + "Example n. 285 = (13.0, 457.0): cluster 0\n", + "Example n. 286 = (13.0, 456.0): cluster 0\n", + "Example n. 287 = (14.0, 494.0): cluster 0\n", + "Example n. 288 = (14.0, 493.0): cluster 0\n", + "Example n. 289 = (14.0, 492.0): cluster 0\n", + "Example n. 290 = (14.0, 491.0): cluster 0\n", + "Example n. 291 = (14.0, 490.0): cluster 0\n", + "Example n. 292 = (14.0, 489.0): cluster 0\n", + "Example n. 293 = (14.0, 488.0): cluster 0\n", + "Example n. 294 = (14.0, 487.0): cluster 0\n", + "Example n. 295 = (14.0, 486.0): cluster 0\n", + "Example n. 296 = (14.0, 485.0): cluster 0\n", + "Example n. 297 = (14.0, 484.0): cluster 0\n", + "Example n. 298 = (14.0, 483.0): cluster 0\n", + "Example n. 299 = (14.0, 482.0): cluster 0\n", + "Example n. 300 = (14.0, 481.0): cluster 0\n", + "Example n. 301 = (14.0, 480.0): cluster 0\n", + "Example n. 302 = (14.0, 479.0): cluster 0\n", + "Example n. 303 = (14.0, 478.0): cluster 0\n", + "Example n. 304 = (14.0, 477.0): cluster 0\n", + "Example n. 305 = (14.0, 476.0): cluster 0\n", + "Example n. 306 = (14.0, 475.0): cluster 0\n", + "Example n. 307 = (14.0, 474.0): cluster 0\n", + "Example n. 308 = (14.0, 473.0): cluster 0\n", + "Example n. 309 = (14.0, 472.0): cluster 0\n", + "Example n. 310 = (14.0, 471.0): cluster 0\n", + "Example n. 311 = (14.0, 470.0): cluster 0\n", + "Example n. 312 = (14.0, 469.0): cluster 0\n", + "Example n. 313 = (14.0, 468.0): cluster 0\n", + "Example n. 314 = (14.0, 467.0): cluster 0\n", + "Example n. 315 = (14.0, 466.0): cluster 0\n", + "Example n. 316 = (14.0, 465.0): cluster 0\n", + "Example n. 317 = (14.0, 464.0): cluster 0\n", + "Example n. 318 = (14.0, 463.0): cluster 0\n", + "Example n. 319 = (14.0, 462.0): cluster 0\n", + "Example n. 320 = (14.0, 461.0): cluster 0\n", + "Example n. 321 = (14.0, 460.0): cluster 0\n", + "Example n. 322 = (14.0, 459.0): cluster 0\n", + "Example n. 323 = (14.0, 458.0): cluster 0\n", + "Example n. 324 = (14.0, 457.0): cluster 0\n", + "Example n. 325 = (14.0, 456.0): cluster 0\n", + "Example n. 326 = (15.0, 494.0): cluster 0\n", + "Example n. 327 = (15.0, 493.0): cluster 0\n", + "Example n. 328 = (15.0, 492.0): cluster 0\n", + "Example n. 329 = (15.0, 491.0): cluster 0\n", + "Example n. 330 = (15.0, 490.0): cluster 0\n", + "Example n. 331 = (15.0, 489.0): cluster 0\n", + "Example n. 332 = (15.0, 488.0): cluster 0\n", + "Example n. 333 = (15.0, 487.0): cluster 0\n", + "Example n. 334 = (15.0, 486.0): cluster 0\n", + "Example n. 335 = (15.0, 485.0): cluster 0\n", + "Example n. 336 = (15.0, 484.0): cluster 0\n", + "Example n. 337 = (15.0, 483.0): cluster 0\n", + "Example n. 338 = (15.0, 482.0): cluster 0\n", + "Example n. 339 = (15.0, 481.0): cluster 0\n", + "Example n. 340 = (15.0, 480.0): cluster 0\n", + "Example n. 341 = (15.0, 479.0): cluster 0\n", + "Example n. 342 = (15.0, 478.0): cluster 0\n", + "Example n. 343 = (15.0, 477.0): cluster 0\n", + "Example n. 344 = (15.0, 476.0): cluster 0\n", + "Example n. 345 = (15.0, 475.0): cluster 0\n", + "Example n. 346 = (15.0, 474.0): cluster 0\n", + "Example n. 347 = (15.0, 473.0): cluster 0\n", + "Example n. 348 = (15.0, 472.0): cluster 0\n", + "Example n. 349 = (15.0, 471.0): cluster 0\n", + "Example n. 350 = (15.0, 470.0): cluster 0\n", + "Example n. 351 = (15.0, 469.0): cluster 0\n", + "Example n. 352 = (15.0, 468.0): cluster 0\n", + "Example n. 353 = (15.0, 467.0): cluster 0\n", + "Example n. 354 = (15.0, 466.0): cluster 0\n", + "Example n. 355 = (15.0, 465.0): cluster 0\n", + "Example n. 356 = (15.0, 464.0): cluster 0\n", + "Example n. 357 = (15.0, 463.0): cluster 0\n", + "Example n. 358 = (15.0, 462.0): cluster 0\n", + "Example n. 359 = (15.0, 461.0): cluster 0\n", + "Example n. 360 = (15.0, 460.0): cluster 0\n", + "Example n. 361 = (15.0, 459.0): cluster 0\n", + "Example n. 362 = (15.0, 458.0): cluster 0\n", + "Example n. 363 = (15.0, 457.0): cluster 0\n", + "Example n. 364 = (15.0, 456.0): cluster 0\n", + "Example n. 365 = (16.0, 494.0): cluster 0\n", + "Example n. 366 = (16.0, 493.0): cluster 0\n", + "Example n. 367 = (16.0, 492.0): cluster 0\n", + "Example n. 368 = (16.0, 491.0): cluster 0\n", + "Example n. 369 = (16.0, 490.0): cluster 0\n", + "Example n. 370 = (16.0, 489.0): cluster 0\n", + "Example n. 371 = (16.0, 488.0): cluster 0\n", + "Example n. 372 = (16.0, 487.0): cluster 0\n", + "Example n. 373 = (16.0, 486.0): cluster 0\n", + "Example n. 374 = (16.0, 485.0): cluster 0\n", + "Example n. 375 = (16.0, 484.0): cluster 0\n", + "Example n. 376 = (16.0, 483.0): cluster 0\n", + "Example n. 377 = (16.0, 482.0): cluster 0\n", + "Example n. 378 = (16.0, 481.0): cluster 0\n", + "Example n. 379 = (16.0, 480.0): cluster 0\n", + "Example n. 380 = (16.0, 479.0): cluster 0\n", + "Example n. 381 = (16.0, 478.0): cluster 0\n", + "Example n. 382 = (16.0, 477.0): cluster 0\n", + "Example n. 383 = (16.0, 476.0): cluster 0\n", + "Example n. 384 = (16.0, 475.0): cluster 0\n", + "Example n. 385 = (16.0, 474.0): cluster 0\n", + "Example n. 386 = (16.0, 473.0): cluster 0\n", + "Example n. 387 = (16.0, 472.0): cluster 0\n", + "Example n. 388 = (16.0, 471.0): cluster 0\n", + "Example n. 389 = (16.0, 470.0): cluster 0\n", + "Example n. 390 = (16.0, 469.0): cluster 0\n", + "Example n. 391 = (16.0, 468.0): cluster 0\n", + "Example n. 392 = (16.0, 467.0): cluster 0\n", + "Example n. 393 = (16.0, 466.0): cluster 0\n", + "Example n. 394 = (16.0, 465.0): cluster 0\n", + "Example n. 395 = (16.0, 464.0): cluster 0\n", + "Example n. 396 = (16.0, 463.0): cluster 0\n", + "Example n. 397 = (16.0, 462.0): cluster 0\n", + "Example n. 398 = (16.0, 461.0): cluster 0\n", + "Example n. 399 = (16.0, 460.0): cluster 0\n", + "Example n. 400 = (16.0, 459.0): cluster 0\n", + "Example n. 401 = (16.0, 458.0): cluster 0\n", + "Example n. 402 = (16.0, 457.0): cluster 0\n", + "Example n. 403 = (16.0, 456.0): cluster 0\n", + "Example n. 404 = (17.0, 494.0): cluster 0\n", + "Example n. 405 = (17.0, 493.0): cluster 0\n", + "Example n. 406 = (17.0, 492.0): cluster 0\n", + "Example n. 407 = (17.0, 491.0): cluster 0\n", + "Example n. 408 = (17.0, 490.0): cluster 0\n", + "Example n. 409 = (17.0, 489.0): cluster 0\n", + "Example n. 410 = (17.0, 488.0): cluster 0\n", + "Example n. 411 = (17.0, 487.0): cluster 0\n", + "Example n. 412 = (17.0, 486.0): cluster 0\n", + "Example n. 413 = (17.0, 485.0): cluster 0\n", + "Example n. 414 = (17.0, 484.0): cluster 0\n", + "Example n. 415 = (17.0, 483.0): cluster 0\n", + "Example n. 416 = (17.0, 482.0): cluster 0\n", + "Example n. 417 = (17.0, 481.0): cluster 0\n", + "Example n. 418 = (17.0, 480.0): cluster 0\n", + "Example n. 419 = (17.0, 479.0): cluster 0\n", + "Example n. 420 = (17.0, 478.0): cluster 0\n", + "Example n. 421 = (17.0, 477.0): cluster 0\n", + "Example n. 422 = (17.0, 476.0): cluster 0\n", + "Example n. 423 = (17.0, 475.0): cluster 0\n", + "Example n. 424 = (17.0, 474.0): cluster 0\n", + "Example n. 425 = (17.0, 473.0): cluster 0\n", + "Example n. 426 = (17.0, 472.0): cluster 0\n", + "Example n. 427 = (17.0, 471.0): cluster 0\n", + "Example n. 428 = (17.0, 470.0): cluster 0\n", + "Example n. 429 = (17.0, 469.0): cluster 0\n", + "Example n. 430 = (17.0, 468.0): cluster 0\n", + "Example n. 431 = (17.0, 467.0): cluster 0\n", + "Example n. 432 = (17.0, 466.0): cluster 0\n", + "Example n. 433 = (17.0, 465.0): cluster 0\n", + "Example n. 434 = (17.0, 464.0): cluster 0\n", + "Example n. 435 = (17.0, 463.0): cluster 0\n", + "Example n. 436 = (17.0, 462.0): cluster 0\n", + "Example n. 437 = (17.0, 461.0): cluster 0\n", + "Example n. 438 = (17.0, 460.0): cluster 0\n", + "Example n. 439 = (17.0, 459.0): cluster 0\n", + "Example n. 440 = (17.0, 458.0): cluster 0\n", + "Example n. 441 = (17.0, 457.0): cluster 0\n", + "Example n. 442 = (17.0, 456.0): cluster 0\n", + "Example n. 443 = (18.0, 494.0): cluster 0\n", + "Example n. 444 = (18.0, 493.0): cluster 0\n", + "Example n. 445 = (18.0, 492.0): cluster 0\n", + "Example n. 446 = (18.0, 491.0): cluster 0\n", + "Example n. 447 = (18.0, 490.0): cluster 0\n", + "Example n. 448 = (18.0, 489.0): cluster 0\n", + "Example n. 449 = (18.0, 488.0): cluster 0\n", + "Example n. 450 = (18.0, 487.0): cluster 0\n", + "Example n. 451 = (18.0, 486.0): cluster 0\n", + "Example n. 452 = (18.0, 485.0): cluster 0\n", + "Example n. 453 = (18.0, 484.0): cluster 0\n", + "Example n. 454 = (18.0, 483.0): cluster 0\n", + "Example n. 455 = (18.0, 482.0): cluster 0\n", + "Example n. 456 = (18.0, 481.0): cluster 0\n", + "Example n. 457 = (18.0, 480.0): cluster 0\n", + "Example n. 458 = (18.0, 479.0): cluster 0\n", + "Example n. 459 = (18.0, 478.0): cluster 0\n", + "Example n. 460 = (18.0, 477.0): cluster 0\n", + "Example n. 461 = (18.0, 476.0): cluster 0\n", + "Example n. 462 = (18.0, 475.0): cluster 0\n", + "Example n. 463 = (18.0, 474.0): cluster 0\n", + "Example n. 464 = (18.0, 473.0): cluster 0\n", + "Example n. 465 = (18.0, 472.0): cluster 0\n", + "Example n. 466 = (18.0, 471.0): cluster 0\n", + "Example n. 467 = (18.0, 470.0): cluster 0\n", + "Example n. 468 = (18.0, 469.0): cluster 0\n", + "Example n. 469 = (18.0, 468.0): cluster 0\n", + "Example n. 470 = (18.0, 467.0): cluster 0\n", + "Example n. 471 = (18.0, 466.0): cluster 0\n", + "Example n. 472 = (18.0, 465.0): cluster 0\n", + "Example n. 473 = (18.0, 464.0): cluster 0\n", + "Example n. 474 = (18.0, 463.0): cluster 0\n", + "Example n. 475 = (18.0, 462.0): cluster 0\n", + "Example n. 476 = (18.0, 461.0): cluster 0\n", + "Example n. 477 = (18.0, 460.0): cluster 0\n", + "Example n. 478 = (18.0, 459.0): cluster 0\n", + "Example n. 479 = (18.0, 458.0): cluster 0\n", + "Example n. 480 = (18.0, 457.0): cluster 0\n", + "Example n. 481 = (18.0, 456.0): cluster 0\n", + "Example n. 482 = (19.0, 494.0): cluster 0\n", + "Example n. 483 = (19.0, 493.0): cluster 0\n", + "Example n. 484 = (19.0, 492.0): cluster 0\n", + "Example n. 485 = (19.0, 491.0): cluster 0\n", + "Example n. 486 = (19.0, 490.0): cluster 0\n", + "Example n. 487 = (19.0, 489.0): cluster 0\n", + "Example n. 488 = (19.0, 488.0): cluster 0\n", + "Example n. 489 = (19.0, 487.0): cluster 0\n", + "Example n. 490 = (19.0, 486.0): cluster 0\n", + "Example n. 491 = (19.0, 485.0): cluster 0\n", + "Example n. 492 = (19.0, 484.0): cluster 0\n", + "Example n. 493 = (19.0, 483.0): cluster 0\n", + "Example n. 494 = (19.0, 482.0): cluster 0\n", + "Example n. 495 = (19.0, 481.0): cluster 0\n", + "Example n. 496 = (19.0, 480.0): cluster 0\n", + "Example n. 497 = (19.0, 479.0): cluster 0\n", + "Example n. 498 = (19.0, 478.0): cluster 0\n", + "Example n. 499 = (19.0, 477.0): cluster 0\n", + "Example n. 500 = (19.0, 476.0): cluster 0\n", + "Example n. 501 = (19.0, 475.0): cluster 0\n", + "Example n. 502 = (19.0, 474.0): cluster 0\n", + "Example n. 503 = (19.0, 473.0): cluster 0\n", + "Example n. 504 = (19.0, 472.0): cluster 0\n", + "Example n. 505 = (19.0, 471.0): cluster 0\n", + "Example n. 506 = (19.0, 470.0): cluster 0\n", + "Example n. 507 = (19.0, 469.0): cluster 0\n", + "Example n. 508 = (19.0, 468.0): cluster 0\n", + "Example n. 509 = (19.0, 467.0): cluster 0\n", + "Example n. 510 = (19.0, 466.0): cluster 0\n", + "Example n. 511 = (19.0, 465.0): cluster 0\n", + "Example n. 512 = (19.0, 464.0): cluster 0\n", + "Example n. 513 = (19.0, 463.0): cluster 0\n", + "Example n. 514 = (19.0, 462.0): cluster 0\n", + "Example n. 515 = (19.0, 461.0): cluster 0\n", + "Example n. 516 = (19.0, 460.0): cluster 0\n", + "Example n. 517 = (19.0, 459.0): cluster 0\n", + "Example n. 518 = (19.0, 458.0): cluster 0\n", + "Example n. 519 = (19.0, 457.0): cluster 0\n", + "Example n. 520 = (19.0, 456.0): cluster 0\n", + "Example n. 521 = (20.0, 494.0): cluster 0\n", + "Example n. 522 = (20.0, 493.0): cluster 0\n", + "Example n. 523 = (20.0, 492.0): cluster 0\n", + "Example n. 524 = (20.0, 491.0): cluster 0\n", + "Example n. 525 = (20.0, 490.0): cluster 0\n", + "Example n. 526 = (20.0, 489.0): cluster 0\n", + "Example n. 527 = (20.0, 488.0): cluster 0\n", + "Example n. 528 = (20.0, 487.0): cluster 0\n", + "Example n. 529 = (20.0, 486.0): cluster 0\n", + "Example n. 530 = (20.0, 485.0): cluster 0\n", + "Example n. 531 = (20.0, 484.0): cluster 0\n", + "Example n. 532 = (20.0, 483.0): cluster 0\n", + "Example n. 533 = (20.0, 482.0): cluster 0\n", + "Example n. 534 = (20.0, 481.0): cluster 0\n", + "Example n. 535 = (20.0, 480.0): cluster 0\n", + "Example n. 536 = (20.0, 479.0): cluster 0\n", + "Example n. 537 = (20.0, 478.0): cluster 0\n", + "Example n. 538 = (20.0, 477.0): cluster 0\n", + "Example n. 539 = (20.0, 476.0): cluster 0\n", + "Example n. 540 = (20.0, 475.0): cluster 0\n", + "Example n. 541 = (20.0, 474.0): cluster 0\n", + "Example n. 542 = (20.0, 473.0): cluster 0\n", + "Example n. 543 = (20.0, 472.0): cluster 0\n", + "Example n. 544 = (20.0, 471.0): cluster 0\n", + "Example n. 545 = (21.0, 496.0): cluster 0\n", + "Example n. 546 = (21.0, 494.0): cluster 0\n", + "Example n. 547 = (21.0, 493.0): cluster 0\n", + "Example n. 548 = (21.0, 492.0): cluster 0\n", + "Example n. 549 = (21.0, 491.0): cluster 0\n", + "Example n. 550 = (21.0, 490.0): cluster 0\n", + "Example n. 551 = (21.0, 489.0): cluster 0\n", + "Example n. 552 = (21.0, 488.0): cluster 0\n", + "Example n. 553 = (21.0, 487.0): cluster 0\n", + "Example n. 554 = (21.0, 486.0): cluster 0\n", + "Example n. 555 = (21.0, 485.0): cluster 0\n", + "Example n. 556 = (21.0, 484.0): cluster 0\n", + "Example n. 557 = (21.0, 483.0): cluster 0\n", + "Example n. 558 = (21.0, 482.0): cluster 0\n", + "Example n. 559 = (21.0, 481.0): cluster 0\n", + "Example n. 560 = (21.0, 480.0): cluster 0\n", + "Example n. 561 = (21.0, 479.0): cluster 0\n", + "Example n. 562 = (21.0, 478.0): cluster 0\n", + "Example n. 563 = (21.0, 477.0): cluster 0\n", + "Example n. 564 = (21.0, 476.0): cluster 0\n", + "Example n. 565 = (21.0, 475.0): cluster 0\n", + "Example n. 566 = (21.0, 474.0): cluster 0\n", + "Example n. 567 = (21.0, 473.0): cluster 0\n", + "Example n. 568 = (21.0, 472.0): cluster 0\n", + "Example n. 569 = (21.0, 471.0): cluster 0\n", + "Example n. 570 = (22.0, 494.0): cluster 0\n", + "Example n. 571 = (22.0, 493.0): cluster 0\n", + "Example n. 572 = (22.0, 492.0): cluster 0\n", + "Example n. 573 = (22.0, 491.0): cluster 0\n", + "Example n. 574 = (22.0, 490.0): cluster 0\n", + "Example n. 575 = (22.0, 489.0): cluster 0\n", + "Example n. 576 = (22.0, 488.0): cluster 0\n", + "Example n. 577 = (22.0, 487.0): cluster 0\n", + "Example n. 578 = (22.0, 486.0): cluster 0\n", + "Example n. 579 = (22.0, 485.0): cluster 0\n", + "Example n. 580 = (22.0, 484.0): cluster 0\n", + "Example n. 581 = (22.0, 483.0): cluster 0\n", + "Example n. 582 = (22.0, 482.0): cluster 0\n", + "Example n. 583 = (22.0, 481.0): cluster 0\n", + "Example n. 584 = (22.0, 480.0): cluster 0\n", + "Example n. 585 = (22.0, 479.0): cluster 0\n", + "Example n. 586 = (22.0, 478.0): cluster 0\n", + "Example n. 587 = (22.0, 477.0): cluster 0\n", + "Example n. 588 = (22.0, 476.0): cluster 0\n", + "Example n. 589 = (22.0, 475.0): cluster 0\n", + "Example n. 590 = (22.0, 474.0): cluster 0\n", + "Example n. 591 = (22.0, 473.0): cluster 0\n", + "Example n. 592 = (22.0, 472.0): cluster 0\n", + "Example n. 593 = (22.0, 471.0): cluster 0\n", + "Example n. 594 = (23.0, 497.0): cluster 0\n", + "Example n. 595 = (23.0, 496.0): cluster 0\n", + "Example n. 596 = (23.0, 495.0): cluster 0\n", + "Example n. 597 = (23.0, 494.0): cluster 0\n", + "Example n. 598 = (23.0, 493.0): cluster 0\n", + "Example n. 599 = (23.0, 492.0): cluster 0\n", + "Example n. 600 = (23.0, 491.0): cluster 0\n", + "Example n. 601 = (23.0, 490.0): cluster 0\n", + "Example n. 602 = (23.0, 489.0): cluster 0\n", + "Example n. 603 = (23.0, 488.0): cluster 0\n", + "Example n. 604 = (23.0, 487.0): cluster 0\n", + "Example n. 605 = (23.0, 486.0): cluster 0\n", + "Example n. 606 = (23.0, 485.0): cluster 0\n", + "Example n. 607 = (23.0, 484.0): cluster 0\n", + "Example n. 608 = (23.0, 483.0): cluster 0\n", + "Example n. 609 = (23.0, 482.0): cluster 0\n", + "Example n. 610 = (23.0, 481.0): cluster 0\n", + "Example n. 611 = (23.0, 480.0): cluster 0\n", + "Example n. 612 = (23.0, 479.0): cluster 0\n", + "Example n. 613 = (23.0, 478.0): cluster 0\n", + "Example n. 614 = (23.0, 477.0): cluster 0\n", + "Example n. 615 = (23.0, 476.0): cluster 0\n", + "Example n. 616 = (23.0, 475.0): cluster 0\n", + "Example n. 617 = (23.0, 474.0): cluster 0\n", + "Example n. 618 = (23.0, 473.0): cluster 0\n", + "Example n. 619 = (23.0, 472.0): cluster 0\n", + "Example n. 620 = (23.0, 471.0): cluster 0\n", + "Example n. 621 = (24.0, 498.0): cluster 0\n", + "Example n. 622 = (24.0, 497.0): cluster 0\n", + "Example n. 623 = (24.0, 496.0): cluster 0\n", + "Example n. 624 = (24.0, 495.0): cluster 0\n", + "Example n. 625 = (24.0, 494.0): cluster 0\n", + "Example n. 626 = (24.0, 493.0): cluster 0\n", + "Example n. 627 = (24.0, 492.0): cluster 0\n", + "Example n. 628 = (24.0, 491.0): cluster 0\n", + "Example n. 629 = (24.0, 490.0): cluster 0\n", + "Example n. 630 = (24.0, 489.0): cluster 0\n", + "Example n. 631 = (24.0, 488.0): cluster 0\n", + "Example n. 632 = (24.0, 487.0): cluster 0\n", + "Example n. 633 = (24.0, 486.0): cluster 0\n", + "Example n. 634 = (24.0, 485.0): cluster 0\n", + "Example n. 635 = (24.0, 484.0): cluster 0\n", + "Example n. 636 = (24.0, 483.0): cluster 0\n", + "Example n. 637 = (24.0, 482.0): cluster 0\n", + "Example n. 638 = (24.0, 481.0): cluster 0\n", + "Example n. 639 = (24.0, 480.0): cluster 0\n", + "Example n. 640 = (24.0, 479.0): cluster 0\n", + "Example n. 641 = (25.0, 498.0): cluster 0\n", + "Example n. 642 = (25.0, 497.0): cluster 0\n", + "Example n. 643 = (25.0, 496.0): cluster 0\n", + "Example n. 644 = (25.0, 495.0): cluster 0\n", + "Example n. 645 = (25.0, 494.0): cluster 0\n", + "Example n. 646 = (25.0, 493.0): cluster 0\n", + "Example n. 647 = (25.0, 492.0): cluster 0\n", + "Example n. 648 = (25.0, 491.0): cluster 0\n", + "Example n. 649 = (25.0, 490.0): cluster 0\n", + "Example n. 650 = (25.0, 489.0): cluster 0\n", + "Example n. 651 = (25.0, 488.0): cluster 0\n", + "Example n. 652 = (25.0, 487.0): cluster 0\n", + "Example n. 653 = (25.0, 486.0): cluster 0\n", + "Example n. 654 = (25.0, 485.0): cluster 0\n", + "Example n. 655 = (25.0, 484.0): cluster 0\n", + "Example n. 656 = (25.0, 483.0): cluster 0\n", + "Example n. 657 = (25.0, 482.0): cluster 0\n", + "Example n. 658 = (25.0, 481.0): cluster 0\n", + "Example n. 659 = (25.0, 480.0): cluster 0\n", + "Example n. 660 = (25.0, 479.0): cluster 0\n", + "Example n. 661 = (26.0, 498.0): cluster 0\n", + "Example n. 662 = (26.0, 497.0): cluster 0\n", + "Example n. 663 = (26.0, 496.0): cluster 0\n", + "Example n. 664 = (26.0, 495.0): cluster 0\n", + "Example n. 665 = (26.0, 494.0): cluster 0\n", + "Example n. 666 = (26.0, 493.0): cluster 0\n", + "Example n. 667 = (26.0, 492.0): cluster 0\n", + "Example n. 668 = (26.0, 491.0): cluster 0\n", + "Example n. 669 = (26.0, 490.0): cluster 0\n", + "Example n. 670 = (26.0, 489.0): cluster 0\n", + "Example n. 671 = (26.0, 488.0): cluster 0\n", + "Example n. 672 = (26.0, 487.0): cluster 0\n", + "Example n. 673 = (26.0, 486.0): cluster 0\n", + "Example n. 674 = (26.0, 485.0): cluster 0\n", + "Example n. 675 = (26.0, 484.0): cluster 0\n", + "Example n. 676 = (26.0, 483.0): cluster 0\n", + "Example n. 677 = (26.0, 482.0): cluster 0\n", + "Example n. 678 = (26.0, 481.0): cluster 0\n", + "Example n. 679 = (26.0, 480.0): cluster 0\n", + "Example n. 680 = (26.0, 479.0): cluster 0\n", + "Example n. 681 = (27.0, 498.0): cluster 0\n", + "Example n. 682 = (27.0, 497.0): cluster 0\n", + "Example n. 683 = (27.0, 496.0): cluster 0\n", + "Example n. 684 = (27.0, 495.0): cluster 0\n", + "Example n. 685 = (27.0, 494.0): cluster 0\n", + "Example n. 686 = (27.0, 493.0): cluster 0\n", + "Example n. 687 = (27.0, 492.0): cluster 0\n", + "Example n. 688 = (27.0, 491.0): cluster 0\n", + "Example n. 689 = (27.0, 490.0): cluster 0\n", + "Example n. 690 = (27.0, 489.0): cluster 0\n", + "Example n. 691 = (27.0, 488.0): cluster 0\n", + "Example n. 692 = (27.0, 487.0): cluster 0\n", + "Example n. 693 = (27.0, 486.0): cluster 0\n", + "Example n. 694 = (27.0, 485.0): cluster 0\n", + "Example n. 695 = (27.0, 484.0): cluster 0\n", + "Example n. 696 = (27.0, 483.0): cluster 0\n", + "Example n. 697 = (27.0, 482.0): cluster 0\n", + "Example n. 698 = (27.0, 481.0): cluster 0\n", + "Example n. 699 = (27.0, 480.0): cluster 0\n", + "Example n. 700 = (27.0, 479.0): cluster 0\n", + "Example n. 701 = (28.0, 502.0): cluster 0\n", + "Example n. 702 = (28.0, 501.0): cluster 0\n", + "Example n. 703 = (28.0, 500.0): cluster 0\n", + "Example n. 704 = (28.0, 499.0): cluster 0\n", + "Example n. 705 = (28.0, 498.0): cluster 0\n", + "Example n. 706 = (28.0, 497.0): cluster 0\n", + "Example n. 707 = (28.0, 496.0): cluster 0\n", + "Example n. 708 = (28.0, 495.0): cluster 0\n", + "Example n. 709 = (28.0, 494.0): cluster 0\n", + "Example n. 710 = (28.0, 493.0): cluster 0\n", + "Example n. 711 = (28.0, 492.0): cluster 0\n", + "Example n. 712 = (28.0, 491.0): cluster 0\n", + "Example n. 713 = (28.0, 490.0): cluster 0\n", + "Example n. 714 = (28.0, 489.0): cluster 0\n", + "Example n. 715 = (28.0, 488.0): cluster 0\n", + "Example n. 716 = (28.0, 487.0): cluster 0\n", + "Example n. 717 = (28.0, 486.0): cluster 0\n", + "Example n. 718 = (28.0, 485.0): cluster 0\n", + "Example n. 719 = (28.0, 484.0): cluster 0\n", + "Example n. 720 = (28.0, 483.0): cluster 0\n", + "Example n. 721 = (28.0, 475.0): cluster 0\n", + "Example n. 722 = (28.0, 474.0): cluster 0\n", + "Example n. 723 = (28.0, 473.0): cluster 0\n", + "Example n. 724 = (28.0, 472.0): cluster 0\n", + "Example n. 725 = (28.0, 471.0): cluster 0\n", + "Example n. 726 = (28.0, 470.0): cluster 0\n", + "Example n. 727 = (28.0, 469.0): cluster 0\n", + "Example n. 728 = (28.0, 468.0): cluster 0\n", + "Example n. 729 = (28.0, 467.0): cluster 0\n", + "Example n. 730 = (28.0, 466.0): cluster 0\n", + "Example n. 731 = (28.0, 465.0): cluster 0\n", + "Example n. 732 = (28.0, 464.0): cluster 0\n", + "Example n. 733 = (28.0, 463.0): cluster 0\n", + "Example n. 734 = (28.0, 462.0): cluster 0\n", + "Example n. 735 = (28.0, 461.0): cluster 0\n", + "Example n. 736 = (28.0, 460.0): cluster 0\n", + "Example n. 737 = (28.0, 459.0): cluster 0\n", + "Example n. 738 = (28.0, 458.0): cluster 0\n", + "Example n. 739 = (28.0, 457.0): cluster 0\n", + "Example n. 740 = (28.0, 456.0): cluster 0\n", + "Example n. 741 = (28.0, 455.0): cluster 0\n", + "Example n. 742 = (28.0, 454.0): cluster 0\n", + "Example n. 743 = (28.0, 453.0): cluster 0\n", + "Example n. 744 = (28.0, 452.0): cluster 0\n", + "Example n. 745 = (29.0, 502.0): cluster 0\n", + "Example n. 746 = (29.0, 501.0): cluster 0\n", + "Example n. 747 = (29.0, 500.0): cluster 0\n", + "Example n. 748 = (29.0, 499.0): cluster 0\n", + "Example n. 749 = (29.0, 498.0): cluster 0\n", + "Example n. 750 = (29.0, 497.0): cluster 0\n", + "Example n. 751 = (29.0, 496.0): cluster 0\n", + "Example n. 752 = (29.0, 495.0): cluster 0\n", + "Example n. 753 = (29.0, 494.0): cluster 0\n", + "Example n. 754 = (29.0, 493.0): cluster 0\n", + "Example n. 755 = (29.0, 492.0): cluster 0\n", + "Example n. 756 = (29.0, 491.0): cluster 0\n", + "Example n. 757 = (29.0, 490.0): cluster 0\n", + "Example n. 758 = (29.0, 489.0): cluster 0\n", + "Example n. 759 = (29.0, 488.0): cluster 0\n", + "Example n. 760 = (29.0, 487.0): cluster 0\n", + "Example n. 761 = (29.0, 486.0): cluster 0\n", + "Example n. 762 = (29.0, 485.0): cluster 0\n", + "Example n. 763 = (29.0, 484.0): cluster 0\n", + "Example n. 764 = (29.0, 483.0): cluster 0\n", + "Example n. 765 = (29.0, 475.0): cluster 0\n", + "Example n. 766 = (29.0, 474.0): cluster 0\n", + "Example n. 767 = (29.0, 473.0): cluster 0\n", + "Example n. 768 = (29.0, 472.0): cluster 0\n", + "Example n. 769 = (29.0, 471.0): cluster 0\n", + "Example n. 770 = (29.0, 470.0): cluster 0\n", + "Example n. 771 = (29.0, 469.0): cluster 0\n", + "Example n. 772 = (29.0, 468.0): cluster 0\n", + "Example n. 773 = (29.0, 467.0): cluster 0\n", + "Example n. 774 = (29.0, 466.0): cluster 0\n", + "Example n. 775 = (29.0, 465.0): cluster 0\n", + "Example n. 776 = (29.0, 464.0): cluster 0\n", + "Example n. 777 = (29.0, 463.0): cluster 0\n", + "Example n. 778 = (29.0, 462.0): cluster 0\n", + "Example n. 779 = (29.0, 461.0): cluster 0\n", + "Example n. 780 = (29.0, 460.0): cluster 0\n", + "Example n. 781 = (29.0, 459.0): cluster 0\n", + "Example n. 782 = (29.0, 458.0): cluster 0\n", + "Example n. 783 = (29.0, 457.0): cluster 0\n", + "Example n. 784 = (29.0, 456.0): cluster 0\n", + "Example n. 785 = (29.0, 455.0): cluster 0\n", + "Example n. 786 = (29.0, 454.0): cluster 0\n", + "Example n. 787 = (29.0, 453.0): cluster 0\n", + "Example n. 788 = (29.0, 452.0): cluster 0\n", + "Example n. 789 = (30.0, 502.0): cluster 0\n", + "Example n. 790 = (30.0, 501.0): cluster 0\n", + "Example n. 791 = (30.0, 500.0): cluster 0\n", + "Example n. 792 = (30.0, 499.0): cluster 0\n", + "Example n. 793 = (30.0, 498.0): cluster 0\n", + "Example n. 794 = (30.0, 497.0): cluster 0\n", + "Example n. 795 = (30.0, 496.0): cluster 0\n", + "Example n. 796 = (30.0, 495.0): cluster 0\n", + "Example n. 797 = (30.0, 494.0): cluster 0\n", + "Example n. 798 = (30.0, 493.0): cluster 0\n", + "Example n. 799 = (30.0, 492.0): cluster 0\n", + "Example n. 800 = (30.0, 491.0): cluster 0\n", + "Example n. 801 = (30.0, 490.0): cluster 0\n", + "Example n. 802 = (30.0, 489.0): cluster 0\n", + "Example n. 803 = (30.0, 488.0): cluster 0\n", + "Example n. 804 = (30.0, 487.0): cluster 0\n", + "Example n. 805 = (30.0, 486.0): cluster 0\n", + "Example n. 806 = (30.0, 485.0): cluster 0\n", + "Example n. 807 = (30.0, 484.0): cluster 0\n", + "Example n. 808 = (30.0, 483.0): cluster 0\n", + "Example n. 809 = (30.0, 475.0): cluster 0\n", + "Example n. 810 = (30.0, 474.0): cluster 0\n", + "Example n. 811 = (30.0, 473.0): cluster 0\n", + "Example n. 812 = (30.0, 472.0): cluster 0\n", + "Example n. 813 = (30.0, 471.0): cluster 0\n", + "Example n. 814 = (30.0, 470.0): cluster 0\n", + "Example n. 815 = (30.0, 469.0): cluster 0\n", + "Example n. 816 = (30.0, 468.0): cluster 0\n", + "Example n. 817 = (30.0, 467.0): cluster 0\n", + "Example n. 818 = (30.0, 466.0): cluster 0\n", + "Example n. 819 = (30.0, 465.0): cluster 0\n", + "Example n. 820 = (30.0, 464.0): cluster 0\n", + "Example n. 821 = (30.0, 463.0): cluster 0\n", + "Example n. 822 = (30.0, 462.0): cluster 0\n", + "Example n. 823 = (30.0, 461.0): cluster 0\n", + "Example n. 824 = (30.0, 460.0): cluster 0\n", + "Example n. 825 = (30.0, 459.0): cluster 0\n", + "Example n. 826 = (30.0, 458.0): cluster 0\n", + "Example n. 827 = (30.0, 457.0): cluster 0\n", + "Example n. 828 = (30.0, 456.0): cluster 0\n", + "Example n. 829 = (30.0, 455.0): cluster 0\n", + "Example n. 830 = (30.0, 454.0): cluster 0\n", + "Example n. 831 = (30.0, 453.0): cluster 0\n", + "Example n. 832 = (30.0, 452.0): cluster 0\n", + "Example n. 833 = (31.0, 502.0): cluster 0\n", + "Example n. 834 = (31.0, 501.0): cluster 0\n", + "Example n. 835 = (31.0, 500.0): cluster 0\n", + "Example n. 836 = (31.0, 499.0): cluster 0\n", + "Example n. 837 = (31.0, 498.0): cluster 0\n", + "Example n. 838 = (31.0, 497.0): cluster 0\n", + "Example n. 839 = (31.0, 496.0): cluster 0\n", + "Example n. 840 = (31.0, 495.0): cluster 0\n", + "Example n. 841 = (31.0, 494.0): cluster 0\n", + "Example n. 842 = (31.0, 493.0): cluster 0\n", + "Example n. 843 = (31.0, 492.0): cluster 0\n", + "Example n. 844 = (31.0, 491.0): cluster 0\n", + "Example n. 845 = (31.0, 490.0): cluster 0\n", + "Example n. 846 = (31.0, 489.0): cluster 0\n", + "Example n. 847 = (31.0, 488.0): cluster 0\n", + "Example n. 848 = (31.0, 487.0): cluster 0\n", + "Example n. 849 = (31.0, 486.0): cluster 0\n", + "Example n. 850 = (31.0, 485.0): cluster 0\n", + "Example n. 851 = (31.0, 484.0): cluster 0\n", + "Example n. 852 = (31.0, 483.0): cluster 0\n", + "Example n. 853 = (31.0, 475.0): cluster 0\n", + "Example n. 854 = (31.0, 474.0): cluster 0\n", + "Example n. 855 = (31.0, 473.0): cluster 0\n", + "Example n. 856 = (31.0, 472.0): cluster 0\n", + "Example n. 857 = (31.0, 471.0): cluster 0\n", + "Example n. 858 = (31.0, 470.0): cluster 0\n", + "Example n. 859 = (31.0, 469.0): cluster 0\n", + "Example n. 860 = (31.0, 468.0): cluster 0\n", + "Example n. 861 = (31.0, 467.0): cluster 0\n", + "Example n. 862 = (31.0, 466.0): cluster 0\n", + "Example n. 863 = (31.0, 465.0): cluster 0\n", + "Example n. 864 = (31.0, 464.0): cluster 0\n", + "Example n. 865 = (31.0, 463.0): cluster 0\n", + "Example n. 866 = (31.0, 462.0): cluster 0\n", + "Example n. 867 = (31.0, 461.0): cluster 0\n", + "Example n. 868 = (31.0, 460.0): cluster 0\n", + "Example n. 869 = (31.0, 459.0): cluster 0\n", + "Example n. 870 = (31.0, 458.0): cluster 0\n", + "Example n. 871 = (31.0, 457.0): cluster 0\n", + "Example n. 872 = (31.0, 456.0): cluster 0\n", + "Example n. 873 = (31.0, 455.0): cluster 0\n", + "Example n. 874 = (31.0, 454.0): cluster 0\n", + "Example n. 875 = (31.0, 453.0): cluster 0\n", + "Example n. 876 = (31.0, 452.0): cluster 0\n", + "Example n. 877 = (32.0, 502.0): cluster 0\n", + "Example n. 878 = (32.0, 501.0): cluster 0\n", + "Example n. 879 = (32.0, 500.0): cluster 0\n", + "Example n. 880 = (32.0, 499.0): cluster 0\n", + "Example n. 881 = (32.0, 498.0): cluster 0\n", + "Example n. 882 = (32.0, 497.0): cluster 0\n", + "Example n. 883 = (32.0, 495.0): cluster 0\n", + "Example n. 884 = (32.0, 494.0): cluster 0\n", + "Example n. 885 = (32.0, 493.0): cluster 0\n", + "Example n. 886 = (32.0, 492.0): cluster 0\n", + "Example n. 887 = (32.0, 491.0): cluster 0\n", + "Example n. 888 = (32.0, 490.0): cluster 0\n", + "Example n. 889 = (32.0, 489.0): cluster 0\n", + "Example n. 890 = (32.0, 488.0): cluster 0\n", + "Example n. 891 = (32.0, 487.0): cluster 0\n", + "Example n. 892 = (32.0, 479.0): cluster 0\n", + "Example n. 893 = (32.0, 478.0): cluster 0\n", + "Example n. 894 = (32.0, 477.0): cluster 0\n", + "Example n. 895 = (32.0, 476.0): cluster 0\n", + "Example n. 896 = (32.0, 475.0): cluster 0\n", + "Example n. 897 = (32.0, 474.0): cluster 0\n", + "Example n. 898 = (32.0, 473.0): cluster 0\n", + "Example n. 899 = (32.0, 472.0): cluster 0\n", + "Example n. 900 = (32.0, 471.0): cluster 0\n", + "Example n. 901 = (32.0, 470.0): cluster 0\n", + "Example n. 902 = (32.0, 469.0): cluster 0\n", + "Example n. 903 = (32.0, 468.0): cluster 0\n", + "Example n. 904 = (32.0, 467.0): cluster 0\n", + "Example n. 905 = (32.0, 466.0): cluster 0\n", + "Example n. 906 = (32.0, 465.0): cluster 0\n", + "Example n. 907 = (32.0, 464.0): cluster 0\n", + "Example n. 908 = (32.0, 463.0): cluster 0\n", + "Example n. 909 = (32.0, 462.0): cluster 0\n", + "Example n. 910 = (32.0, 461.0): cluster 0\n", + "Example n. 911 = (32.0, 460.0): cluster 0\n", + "Example n. 912 = (32.0, 459.0): cluster 0\n", + "Example n. 913 = (32.0, 458.0): cluster 0\n", + "Example n. 914 = (32.0, 457.0): cluster 0\n", + "Example n. 915 = (32.0, 456.0): cluster 0\n", + "Example n. 916 = (32.0, 455.0): cluster 0\n", + "Example n. 917 = (32.0, 454.0): cluster 0\n", + "Example n. 918 = (32.0, 453.0): cluster 0\n", + "Example n. 919 = (32.0, 452.0): cluster 0\n", + "Example n. 920 = (32.0, 451.0): cluster 0\n", + "Example n. 921 = (32.0, 450.0): cluster 0\n", + "Example n. 922 = (32.0, 449.0): cluster 0\n", + "Example n. 923 = (32.0, 448.0): cluster 0\n", + "Example n. 924 = (32.0, 421.0): cluster 3\n", + "Example n. 925 = (32.0, 420.0): cluster 3\n", + "Example n. 926 = (32.0, 419.0): cluster 3\n", + "Example n. 927 = (32.0, 418.0): cluster 3\n", + "Example n. 928 = (32.0, 417.0): cluster 3\n", + "Example n. 929 = (32.0, 416.0): cluster 3\n", + "Example n. 930 = (32.0, 415.0): cluster 3\n", + "Example n. 931 = (32.0, 414.0): cluster 3\n", + "Example n. 932 = (32.0, 413.0): cluster 3\n", + "Example n. 933 = (32.0, 412.0): cluster 3\n", + "Example n. 934 = (32.0, 411.0): cluster 3\n", + "Example n. 935 = (32.0, 410.0): cluster 3\n", + "Example n. 936 = (33.0, 502.0): cluster 0\n", + "Example n. 937 = (33.0, 501.0): cluster 0\n", + "Example n. 938 = (33.0, 500.0): cluster 0\n", + "Example n. 939 = (33.0, 499.0): cluster 0\n", + "Example n. 940 = (33.0, 498.0): cluster 0\n", + "Example n. 941 = (33.0, 494.0): cluster 0\n", + "Example n. 942 = (33.0, 493.0): cluster 0\n", + "Example n. 943 = (33.0, 492.0): cluster 0\n", + "Example n. 944 = (33.0, 491.0): cluster 0\n", + "Example n. 945 = (33.0, 490.0): cluster 0\n", + "Example n. 946 = (33.0, 489.0): cluster 0\n", + "Example n. 947 = (33.0, 488.0): cluster 0\n", + "Example n. 948 = (33.0, 487.0): cluster 0\n", + "Example n. 949 = (33.0, 479.0): cluster 0\n", + "Example n. 950 = (33.0, 478.0): cluster 0\n", + "Example n. 951 = (33.0, 477.0): cluster 0\n", + "Example n. 952 = (33.0, 476.0): cluster 0\n", + "Example n. 953 = (33.0, 475.0): cluster 0\n", + "Example n. 954 = (33.0, 474.0): cluster 0\n", + "Example n. 955 = (33.0, 473.0): cluster 0\n", + "Example n. 956 = (33.0, 472.0): cluster 0\n", + "Example n. 957 = (33.0, 471.0): cluster 0\n", + "Example n. 958 = (33.0, 470.0): cluster 0\n", + "Example n. 959 = (33.0, 469.0): cluster 0\n", + "Example n. 960 = (33.0, 468.0): cluster 0\n", + "Example n. 961 = (33.0, 467.0): cluster 0\n", + "Example n. 962 = (33.0, 466.0): cluster 0\n", + "Example n. 963 = (33.0, 465.0): cluster 0\n", + "Example n. 964 = (33.0, 464.0): cluster 0\n", + "Example n. 965 = (33.0, 463.0): cluster 0\n", + "Example n. 966 = (33.0, 462.0): cluster 0\n", + "Example n. 967 = (33.0, 461.0): cluster 0\n", + "Example n. 968 = (33.0, 460.0): cluster 0\n", + "Example n. 969 = (33.0, 459.0): cluster 0\n", + "Example n. 970 = (33.0, 458.0): cluster 0\n", + "Example n. 971 = (33.0, 457.0): cluster 0\n", + "Example n. 972 = (33.0, 456.0): cluster 0\n", + "Example n. 973 = (33.0, 455.0): cluster 0\n", + "Example n. 974 = (33.0, 454.0): cluster 0\n", + "Example n. 975 = (33.0, 453.0): cluster 0\n", + "Example n. 976 = (33.0, 452.0): cluster 0\n", + "Example n. 977 = (33.0, 451.0): cluster 0\n", + "Example n. 978 = (33.0, 450.0): cluster 0\n", + "Example n. 979 = (33.0, 449.0): cluster 0\n", + "Example n. 980 = (33.0, 448.0): cluster 0\n", + "Example n. 981 = (33.0, 421.0): cluster 3\n", + "Example n. 982 = (33.0, 420.0): cluster 3\n", + "Example n. 983 = (33.0, 419.0): cluster 3\n", + "Example n. 984 = (33.0, 418.0): cluster 3\n", + "Example n. 985 = (33.0, 417.0): cluster 3\n", + "Example n. 986 = (33.0, 416.0): cluster 3\n", + "Example n. 987 = (33.0, 415.0): cluster 3\n", + "Example n. 988 = (33.0, 414.0): cluster 3\n", + "Example n. 989 = (33.0, 413.0): cluster 3\n", + "Example n. 990 = (33.0, 412.0): cluster 3\n", + "Example n. 991 = (33.0, 411.0): cluster 3\n", + "Example n. 992 = (33.0, 410.0): cluster 3\n", + "Example n. 993 = (34.0, 502.0): cluster 0\n", + "Example n. 994 = (34.0, 501.0): cluster 0\n", + "Example n. 995 = (34.0, 500.0): cluster 0\n", + "Example n. 996 = (34.0, 499.0): cluster 0\n", + "Example n. 997 = (34.0, 498.0): cluster 0\n", + "Example n. 998 = (34.0, 494.0): cluster 0\n", + "Example n. 999 = (34.0, 493.0): cluster 0\n", + "Example n. 1000 = (34.0, 492.0): cluster 0\n", + "Example n. 1001 = (34.0, 491.0): cluster 0\n", + "Example n. 1002 = (34.0, 490.0): cluster 0\n", + "Example n. 1003 = (34.0, 489.0): cluster 0\n", + "Example n. 1004 = (34.0, 488.0): cluster 0\n", + "Example n. 1005 = (34.0, 487.0): cluster 0\n", + "Example n. 1006 = (34.0, 479.0): cluster 0\n", + "Example n. 1007 = (34.0, 478.0): cluster 0\n", + "Example n. 1008 = (34.0, 477.0): cluster 0\n", + "Example n. 1009 = (34.0, 476.0): cluster 0\n", + "Example n. 1010 = (34.0, 475.0): cluster 0\n", + "Example n. 1011 = (34.0, 474.0): cluster 0\n", + "Example n. 1012 = (34.0, 473.0): cluster 0\n", + "Example n. 1013 = (34.0, 472.0): cluster 0\n", + "Example n. 1014 = (34.0, 471.0): cluster 0\n", + "Example n. 1015 = (34.0, 470.0): cluster 0\n", + "Example n. 1016 = (34.0, 469.0): cluster 0\n", + "Example n. 1017 = (34.0, 468.0): cluster 0\n", + "Example n. 1018 = (34.0, 467.0): cluster 0\n", + "Example n. 1019 = (34.0, 466.0): cluster 0\n", + "Example n. 1020 = (34.0, 465.0): cluster 0\n", + "Example n. 1021 = (34.0, 464.0): cluster 0\n", + "Example n. 1022 = (34.0, 463.0): cluster 0\n", + "Example n. 1023 = (34.0, 462.0): cluster 0\n", + "Example n. 1024 = (34.0, 461.0): cluster 0\n", + "Example n. 1025 = (34.0, 460.0): cluster 0\n", + "Example n. 1026 = (34.0, 459.0): cluster 0\n", + "Example n. 1027 = (34.0, 458.0): cluster 0\n", + "Example n. 1028 = (34.0, 457.0): cluster 0\n", + "Example n. 1029 = (34.0, 456.0): cluster 0\n", + "Example n. 1030 = (34.0, 455.0): cluster 0\n", + "Example n. 1031 = (34.0, 454.0): cluster 0\n", + "Example n. 1032 = (34.0, 453.0): cluster 0\n", + "Example n. 1033 = (34.0, 452.0): cluster 0\n", + "Example n. 1034 = (34.0, 451.0): cluster 0\n", + "Example n. 1035 = (34.0, 450.0): cluster 0\n", + "Example n. 1036 = (34.0, 449.0): cluster 0\n", + "Example n. 1037 = (34.0, 448.0): cluster 0\n", + "Example n. 1038 = (34.0, 421.0): cluster 3\n", + "Example n. 1039 = (34.0, 420.0): cluster 3\n", + "Example n. 1040 = (34.0, 419.0): cluster 3\n", + "Example n. 1041 = (34.0, 418.0): cluster 3\n", + "Example n. 1042 = (34.0, 417.0): cluster 3\n", + "Example n. 1043 = (34.0, 416.0): cluster 3\n", + "Example n. 1044 = (34.0, 415.0): cluster 3\n", + "Example n. 1045 = (34.0, 414.0): cluster 3\n", + "Example n. 1046 = (34.0, 413.0): cluster 3\n", + "Example n. 1047 = (34.0, 412.0): cluster 3\n", + "Example n. 1048 = (34.0, 411.0): cluster 3\n", + "Example n. 1049 = (34.0, 410.0): cluster 3\n", + "Example n. 1050 = (35.0, 502.0): cluster 0\n", + "Example n. 1051 = (35.0, 501.0): cluster 0\n", + "Example n. 1052 = (35.0, 500.0): cluster 0\n", + "Example n. 1053 = (35.0, 499.0): cluster 0\n", + "Example n. 1054 = (35.0, 498.0): cluster 0\n", + "Example n. 1055 = (35.0, 497.0): cluster 0\n", + "Example n. 1056 = (35.0, 496.0): cluster 0\n", + "Example n. 1057 = (35.0, 495.0): cluster 0\n", + "Example n. 1058 = (35.0, 494.0): cluster 0\n", + "Example n. 1059 = (35.0, 493.0): cluster 0\n", + "Example n. 1060 = (35.0, 492.0): cluster 0\n", + "Example n. 1061 = (35.0, 491.0): cluster 0\n", + "Example n. 1062 = (35.0, 490.0): cluster 0\n", + "Example n. 1063 = (35.0, 489.0): cluster 0\n", + "Example n. 1064 = (35.0, 488.0): cluster 0\n", + "Example n. 1065 = (35.0, 487.0): cluster 0\n", + "Example n. 1066 = (35.0, 479.0): cluster 0\n", + "Example n. 1067 = (35.0, 478.0): cluster 0\n", + "Example n. 1068 = (35.0, 477.0): cluster 0\n", + "Example n. 1069 = (35.0, 476.0): cluster 0\n", + "Example n. 1070 = (35.0, 475.0): cluster 0\n", + "Example n. 1071 = (35.0, 474.0): cluster 0\n", + "Example n. 1072 = (35.0, 473.0): cluster 0\n", + "Example n. 1073 = (35.0, 472.0): cluster 0\n", + "Example n. 1074 = (35.0, 471.0): cluster 0\n", + "Example n. 1075 = (35.0, 470.0): cluster 0\n", + "Example n. 1076 = (35.0, 469.0): cluster 0\n", + "Example n. 1077 = (35.0, 468.0): cluster 0\n", + "Example n. 1078 = (35.0, 467.0): cluster 0\n", + "Example n. 1079 = (35.0, 466.0): cluster 0\n", + "Example n. 1080 = (35.0, 465.0): cluster 0\n", + "Example n. 1081 = (35.0, 464.0): cluster 0\n", + "Example n. 1082 = (35.0, 463.0): cluster 0\n", + "Example n. 1083 = (35.0, 462.0): cluster 0\n", + "Example n. 1084 = (35.0, 461.0): cluster 0\n", + "Example n. 1085 = (35.0, 460.0): cluster 0\n", + "Example n. 1086 = (35.0, 459.0): cluster 0\n", + "Example n. 1087 = (35.0, 458.0): cluster 0\n", + "Example n. 1088 = (35.0, 457.0): cluster 0\n", + "Example n. 1089 = (35.0, 456.0): cluster 0\n", + "Example n. 1090 = (35.0, 455.0): cluster 0\n", + "Example n. 1091 = (35.0, 454.0): cluster 0\n", + "Example n. 1092 = (35.0, 453.0): cluster 0\n", + "Example n. 1093 = (35.0, 452.0): cluster 0\n", + "Example n. 1094 = (35.0, 451.0): cluster 0\n", + "Example n. 1095 = (35.0, 450.0): cluster 0\n", + "Example n. 1096 = (35.0, 449.0): cluster 1\n", + "Example n. 1097 = (35.0, 448.0): cluster 1\n", + "Example n. 1098 = (35.0, 425.0): cluster 3\n", + "Example n. 1099 = (35.0, 424.0): cluster 3\n", + "Example n. 1100 = (35.0, 423.0): cluster 3\n", + "Example n. 1101 = (35.0, 422.0): cluster 3\n", + "Example n. 1102 = (35.0, 421.0): cluster 3\n", + "Example n. 1103 = (35.0, 420.0): cluster 3\n", + "Example n. 1104 = (35.0, 419.0): cluster 3\n", + "Example n. 1105 = (35.0, 418.0): cluster 3\n", + "Example n. 1106 = (35.0, 417.0): cluster 3\n", + "Example n. 1107 = (35.0, 416.0): cluster 3\n", + "Example n. 1108 = (35.0, 415.0): cluster 3\n", + "Example n. 1109 = (35.0, 414.0): cluster 3\n", + "Example n. 1110 = (35.0, 413.0): cluster 3\n", + "Example n. 1111 = (35.0, 412.0): cluster 3\n", + "Example n. 1112 = (35.0, 411.0): cluster 3\n", + "Example n. 1113 = (35.0, 410.0): cluster 3\n", + "Example n. 1114 = (35.0, 409.0): cluster 3\n", + "Example n. 1115 = (35.0, 408.0): cluster 3\n", + "Example n. 1116 = (35.0, 407.0): cluster 3\n", + "Example n. 1117 = (35.0, 406.0): cluster 3\n", + "Example n. 1118 = (35.0, 405.0): cluster 3\n", + "Example n. 1119 = (35.0, 404.0): cluster 3\n", + "Example n. 1120 = (35.0, 403.0): cluster 3\n", + "Example n. 1121 = (35.0, 402.0): cluster 3\n", + "Example n. 1122 = (36.0, 502.0): cluster 0\n", + "Example n. 1123 = (36.0, 501.0): cluster 0\n", + "Example n. 1124 = (36.0, 500.0): cluster 0\n", + "Example n. 1125 = (36.0, 499.0): cluster 0\n", + "Example n. 1126 = (36.0, 498.0): cluster 0\n", + "Example n. 1127 = (36.0, 497.0): cluster 0\n", + "Example n. 1128 = (36.0, 496.0): cluster 0\n", + "Example n. 1129 = (36.0, 495.0): cluster 0\n", + "Example n. 1130 = (36.0, 494.0): cluster 0\n", + "Example n. 1131 = (36.0, 493.0): cluster 0\n", + "Example n. 1132 = (36.0, 492.0): cluster 0\n", + "Example n. 1133 = (36.0, 491.0): cluster 0\n", + "Example n. 1134 = (36.0, 490.0): cluster 0\n", + "Example n. 1135 = (36.0, 489.0): cluster 0\n", + "Example n. 1136 = (36.0, 488.0): cluster 0\n", + "Example n. 1137 = (36.0, 487.0): cluster 0\n", + "Example n. 1138 = (36.0, 479.0): cluster 0\n", + "Example n. 1139 = (36.0, 478.0): cluster 0\n", + "Example n. 1140 = (36.0, 477.0): cluster 0\n", + "Example n. 1141 = (36.0, 476.0): cluster 0\n", + "Example n. 1142 = (36.0, 475.0): cluster 0\n", + "Example n. 1143 = (36.0, 474.0): cluster 0\n", + "Example n. 1144 = (36.0, 473.0): cluster 0\n", + "Example n. 1145 = (36.0, 472.0): cluster 0\n", + "Example n. 1146 = (36.0, 471.0): cluster 0\n", + "Example n. 1147 = (36.0, 470.0): cluster 0\n", + "Example n. 1148 = (36.0, 469.0): cluster 0\n", + "Example n. 1149 = (36.0, 468.0): cluster 0\n", + "Example n. 1150 = (36.0, 467.0): cluster 0\n", + "Example n. 1151 = (36.0, 466.0): cluster 0\n", + "Example n. 1152 = (36.0, 465.0): cluster 0\n", + "Example n. 1153 = (36.0, 464.0): cluster 0\n", + "Example n. 1154 = (36.0, 463.0): cluster 0\n", + "Example n. 1155 = (36.0, 462.0): cluster 0\n", + "Example n. 1156 = (36.0, 461.0): cluster 0\n", + "Example n. 1157 = (36.0, 460.0): cluster 0\n", + "Example n. 1158 = (36.0, 459.0): cluster 0\n", + "Example n. 1159 = (36.0, 458.0): cluster 0\n", + "Example n. 1160 = (36.0, 457.0): cluster 0\n", + "Example n. 1161 = (36.0, 456.0): cluster 0\n", + "Example n. 1162 = (36.0, 455.0): cluster 0\n", + "Example n. 1163 = (36.0, 454.0): cluster 0\n", + "Example n. 1164 = (36.0, 453.0): cluster 0\n", + "Example n. 1165 = (36.0, 452.0): cluster 1\n", + "Example n. 1166 = (36.0, 451.0): cluster 1\n", + "Example n. 1167 = (36.0, 450.0): cluster 1\n", + "Example n. 1168 = (36.0, 449.0): cluster 1\n", + "Example n. 1169 = (36.0, 448.0): cluster 1\n", + "Example n. 1170 = (36.0, 425.0): cluster 3\n", + "Example n. 1171 = (36.0, 424.0): cluster 3\n", + "Example n. 1172 = (36.0, 423.0): cluster 3\n", + "Example n. 1173 = (36.0, 422.0): cluster 3\n", + "Example n. 1174 = (36.0, 421.0): cluster 3\n", + "Example n. 1175 = (36.0, 420.0): cluster 3\n", + "Example n. 1176 = (36.0, 419.0): cluster 3\n", + "Example n. 1177 = (36.0, 418.0): cluster 3\n", + "Example n. 1178 = (36.0, 417.0): cluster 3\n", + "Example n. 1179 = (36.0, 416.0): cluster 3\n", + "Example n. 1180 = (36.0, 415.0): cluster 3\n", + "Example n. 1181 = (36.0, 414.0): cluster 3\n", + "Example n. 1182 = (36.0, 413.0): cluster 3\n", + "Example n. 1183 = (36.0, 412.0): cluster 3\n", + "Example n. 1184 = (36.0, 411.0): cluster 3\n", + "Example n. 1185 = (36.0, 410.0): cluster 3\n", + "Example n. 1186 = (36.0, 409.0): cluster 3\n", + "Example n. 1187 = (36.0, 408.0): cluster 3\n", + "Example n. 1188 = (36.0, 407.0): cluster 3\n", + "Example n. 1189 = (36.0, 406.0): cluster 3\n", + "Example n. 1190 = (36.0, 405.0): cluster 3\n", + "Example n. 1191 = (36.0, 404.0): cluster 3\n", + "Example n. 1192 = (36.0, 403.0): cluster 3\n", + "Example n. 1193 = (36.0, 402.0): cluster 3\n", + "Example n. 1194 = (37.0, 502.0): cluster 0\n", + "Example n. 1195 = (37.0, 501.0): cluster 0\n", + "Example n. 1196 = (37.0, 500.0): cluster 0\n", + "Example n. 1197 = (37.0, 499.0): cluster 0\n", + "Example n. 1198 = (37.0, 498.0): cluster 0\n", + "Example n. 1199 = (37.0, 497.0): cluster 0\n", + "Example n. 1200 = (37.0, 496.0): cluster 0\n", + "Example n. 1201 = (37.0, 495.0): cluster 0\n", + "Example n. 1202 = (37.0, 494.0): cluster 0\n", + "Example n. 1203 = (37.0, 493.0): cluster 0\n", + "Example n. 1204 = (37.0, 492.0): cluster 0\n", + "Example n. 1205 = (37.0, 491.0): cluster 0\n", + "Example n. 1206 = (37.0, 490.0): cluster 0\n", + "Example n. 1207 = (37.0, 489.0): cluster 0\n", + "Example n. 1208 = (37.0, 488.0): cluster 0\n", + "Example n. 1209 = (37.0, 487.0): cluster 0\n", + "Example n. 1210 = (37.0, 479.0): cluster 0\n", + "Example n. 1211 = (37.0, 478.0): cluster 0\n", + "Example n. 1212 = (37.0, 477.0): cluster 0\n", + "Example n. 1213 = (37.0, 476.0): cluster 0\n", + "Example n. 1214 = (37.0, 475.0): cluster 0\n", + "Example n. 1215 = (37.0, 474.0): cluster 0\n", + "Example n. 1216 = (37.0, 473.0): cluster 0\n", + "Example n. 1217 = (37.0, 472.0): cluster 0\n", + "Example n. 1218 = (37.0, 471.0): cluster 0\n", + "Example n. 1219 = (37.0, 470.0): cluster 0\n", + "Example n. 1220 = (37.0, 469.0): cluster 0\n", + "Example n. 1221 = (37.0, 468.0): cluster 0\n", + "Example n. 1222 = (37.0, 467.0): cluster 0\n", + "Example n. 1223 = (37.0, 466.0): cluster 0\n", + "Example n. 1224 = (37.0, 465.0): cluster 0\n", + "Example n. 1225 = (37.0, 464.0): cluster 0\n", + "Example n. 1226 = (37.0, 463.0): cluster 0\n", + "Example n. 1227 = (37.0, 462.0): cluster 0\n", + "Example n. 1228 = (37.0, 461.0): cluster 0\n", + "Example n. 1229 = (37.0, 460.0): cluster 0\n", + "Example n. 1230 = (37.0, 459.0): cluster 0\n", + "Example n. 1231 = (37.0, 458.0): cluster 0\n", + "Example n. 1232 = (37.0, 457.0): cluster 0\n", + "Example n. 1233 = (37.0, 456.0): cluster 0\n", + "Example n. 1234 = (37.0, 455.0): cluster 1\n", + "Example n. 1235 = (37.0, 454.0): cluster 1\n", + "Example n. 1236 = (37.0, 453.0): cluster 1\n", + "Example n. 1237 = (37.0, 452.0): cluster 1\n", + "Example n. 1238 = (37.0, 451.0): cluster 1\n", + "Example n. 1239 = (37.0, 450.0): cluster 1\n", + "Example n. 1240 = (37.0, 449.0): cluster 1\n", + "Example n. 1241 = (37.0, 448.0): cluster 1\n", + "Example n. 1242 = (37.0, 425.0): cluster 3\n", + "Example n. 1243 = (37.0, 424.0): cluster 3\n", + "Example n. 1244 = (37.0, 423.0): cluster 3\n", + "Example n. 1245 = (37.0, 422.0): cluster 3\n", + "Example n. 1246 = (37.0, 421.0): cluster 3\n", + "Example n. 1247 = (37.0, 420.0): cluster 3\n", + "Example n. 1248 = (37.0, 419.0): cluster 3\n", + "Example n. 1249 = (37.0, 418.0): cluster 3\n", + "Example n. 1250 = (37.0, 417.0): cluster 3\n", + "Example n. 1251 = (37.0, 416.0): cluster 3\n", + "Example n. 1252 = (37.0, 415.0): cluster 3\n", + "Example n. 1253 = (37.0, 414.0): cluster 3\n", + "Example n. 1254 = (37.0, 413.0): cluster 3\n", + "Example n. 1255 = (37.0, 412.0): cluster 3\n", + "Example n. 1256 = (37.0, 411.0): cluster 3\n", + "Example n. 1257 = (37.0, 410.0): cluster 3\n", + "Example n. 1258 = (37.0, 409.0): cluster 3\n", + "Example n. 1259 = (37.0, 408.0): cluster 3\n", + "Example n. 1260 = (37.0, 407.0): cluster 3\n", + "Example n. 1261 = (37.0, 406.0): cluster 3\n", + "Example n. 1262 = (37.0, 405.0): cluster 3\n", + "Example n. 1263 = (37.0, 404.0): cluster 3\n", + "Example n. 1264 = (37.0, 403.0): cluster 3\n", + "Example n. 1265 = (37.0, 402.0): cluster 3\n", + "Example n. 1266 = (38.0, 502.0): cluster 0\n", + "Example n. 1267 = (38.0, 501.0): cluster 0\n", + "Example n. 1268 = (38.0, 500.0): cluster 0\n", + "Example n. 1269 = (38.0, 499.0): cluster 0\n", + "Example n. 1270 = (38.0, 498.0): cluster 0\n", + "Example n. 1271 = (38.0, 497.0): cluster 0\n", + "Example n. 1272 = (38.0, 496.0): cluster 0\n", + "Example n. 1273 = (38.0, 495.0): cluster 0\n", + "Example n. 1274 = (38.0, 494.0): cluster 0\n", + "Example n. 1275 = (38.0, 493.0): cluster 0\n", + "Example n. 1276 = (38.0, 492.0): cluster 0\n", + "Example n. 1277 = (38.0, 491.0): cluster 0\n", + "Example n. 1278 = (38.0, 490.0): cluster 0\n", + "Example n. 1279 = (38.0, 489.0): cluster 0\n", + "Example n. 1280 = (38.0, 488.0): cluster 0\n", + "Example n. 1281 = (38.0, 487.0): cluster 0\n", + "Example n. 1282 = (38.0, 479.0): cluster 0\n", + "Example n. 1283 = (38.0, 478.0): cluster 0\n", + "Example n. 1284 = (38.0, 477.0): cluster 0\n", + "Example n. 1285 = (38.0, 476.0): cluster 0\n", + "Example n. 1286 = (38.0, 475.0): cluster 0\n", + "Example n. 1287 = (38.0, 474.0): cluster 0\n", + "Example n. 1288 = (38.0, 473.0): cluster 0\n", + "Example n. 1289 = (38.0, 472.0): cluster 0\n", + "Example n. 1290 = (38.0, 471.0): cluster 0\n", + "Example n. 1291 = (38.0, 470.0): cluster 0\n", + "Example n. 1292 = (38.0, 469.0): cluster 0\n", + "Example n. 1293 = (38.0, 468.0): cluster 0\n", + "Example n. 1294 = (38.0, 467.0): cluster 0\n", + "Example n. 1295 = (38.0, 466.0): cluster 0\n", + "Example n. 1296 = (38.0, 465.0): cluster 0\n", + "Example n. 1297 = (38.0, 464.0): cluster 0\n", + "Example n. 1298 = (38.0, 463.0): cluster 0\n", + "Example n. 1299 = (38.0, 462.0): cluster 0\n", + "Example n. 1300 = (38.0, 461.0): cluster 0\n", + "Example n. 1301 = (38.0, 460.0): cluster 0\n", + "Example n. 1302 = (38.0, 459.0): cluster 0\n", + "Example n. 1303 = (38.0, 458.0): cluster 0\n", + "Example n. 1304 = (38.0, 457.0): cluster 1\n", + "Example n. 1305 = (38.0, 456.0): cluster 1\n", + "Example n. 1306 = (38.0, 455.0): cluster 1\n", + "Example n. 1307 = (38.0, 454.0): cluster 1\n", + "Example n. 1308 = (38.0, 453.0): cluster 1\n", + "Example n. 1309 = (38.0, 452.0): cluster 1\n", + "Example n. 1310 = (38.0, 451.0): cluster 1\n", + "Example n. 1311 = (38.0, 450.0): cluster 1\n", + "Example n. 1312 = (38.0, 449.0): cluster 1\n", + "Example n. 1313 = (38.0, 448.0): cluster 1\n", + "Example n. 1314 = (38.0, 425.0): cluster 3\n", + "Example n. 1315 = (38.0, 424.0): cluster 3\n", + "Example n. 1316 = (38.0, 423.0): cluster 3\n", + "Example n. 1317 = (38.0, 422.0): cluster 3\n", + "Example n. 1318 = (38.0, 421.0): cluster 3\n", + "Example n. 1319 = (38.0, 420.0): cluster 3\n", + "Example n. 1320 = (38.0, 419.0): cluster 3\n", + "Example n. 1321 = (38.0, 418.0): cluster 3\n", + "Example n. 1322 = (38.0, 417.0): cluster 3\n", + "Example n. 1323 = (38.0, 416.0): cluster 3\n", + "Example n. 1324 = (38.0, 415.0): cluster 3\n", + "Example n. 1325 = (38.0, 414.0): cluster 3\n", + "Example n. 1326 = (38.0, 413.0): cluster 3\n", + "Example n. 1327 = (38.0, 412.0): cluster 3\n", + "Example n. 1328 = (38.0, 411.0): cluster 3\n", + "Example n. 1329 = (38.0, 410.0): cluster 3\n", + "Example n. 1330 = (38.0, 409.0): cluster 3\n", + "Example n. 1331 = (38.0, 408.0): cluster 3\n", + "Example n. 1332 = (38.0, 407.0): cluster 3\n", + "Example n. 1333 = (38.0, 406.0): cluster 3\n", + "Example n. 1334 = (38.0, 405.0): cluster 3\n", + "Example n. 1335 = (38.0, 404.0): cluster 3\n", + "Example n. 1336 = (38.0, 403.0): cluster 3\n", + "Example n. 1337 = (38.0, 402.0): cluster 3\n", + "Example n. 1338 = (39.0, 506.0): cluster 0\n", + "Example n. 1339 = (39.0, 505.0): cluster 0\n", + "Example n. 1340 = (39.0, 504.0): cluster 0\n", + "Example n. 1341 = (39.0, 503.0): cluster 0\n", + "Example n. 1342 = (39.0, 502.0): cluster 0\n", + "Example n. 1343 = (39.0, 501.0): cluster 0\n", + "Example n. 1344 = (39.0, 500.0): cluster 0\n", + "Example n. 1345 = (39.0, 499.0): cluster 0\n", + "Example n. 1346 = (39.0, 498.0): cluster 0\n", + "Example n. 1347 = (39.0, 497.0): cluster 0\n", + "Example n. 1348 = (39.0, 496.0): cluster 0\n", + "Example n. 1349 = (39.0, 495.0): cluster 0\n", + "Example n. 1350 = (39.0, 494.0): cluster 0\n", + "Example n. 1351 = (39.0, 493.0): cluster 0\n", + "Example n. 1352 = (39.0, 492.0): cluster 0\n", + "Example n. 1353 = (39.0, 491.0): cluster 0\n", + "Example n. 1354 = (39.0, 490.0): cluster 0\n", + "Example n. 1355 = (39.0, 489.0): cluster 0\n", + "Example n. 1356 = (39.0, 488.0): cluster 0\n", + "Example n. 1357 = (39.0, 487.0): cluster 0\n", + "Example n. 1358 = (39.0, 479.0): cluster 0\n", + "Example n. 1359 = (39.0, 478.0): cluster 0\n", + "Example n. 1360 = (39.0, 477.0): cluster 0\n", + "Example n. 1361 = (39.0, 476.0): cluster 0\n", + "Example n. 1362 = (39.0, 475.0): cluster 0\n", + "Example n. 1363 = (39.0, 474.0): cluster 0\n", + "Example n. 1364 = (39.0, 473.0): cluster 0\n", + "Example n. 1365 = (39.0, 472.0): cluster 0\n", + "Example n. 1366 = (39.0, 471.0): cluster 0\n", + "Example n. 1367 = (39.0, 470.0): cluster 0\n", + "Example n. 1368 = (39.0, 469.0): cluster 0\n", + "Example n. 1369 = (39.0, 468.0): cluster 0\n", + "Example n. 1370 = (39.0, 467.0): cluster 0\n", + "Example n. 1371 = (39.0, 466.0): cluster 0\n", + "Example n. 1372 = (39.0, 465.0): cluster 0\n", + "Example n. 1373 = (39.0, 464.0): cluster 0\n", + "Example n. 1374 = (39.0, 463.0): cluster 0\n", + "Example n. 1375 = (39.0, 462.0): cluster 0\n", + "Example n. 1376 = (39.0, 461.0): cluster 0\n", + "Example n. 1377 = (39.0, 460.0): cluster 1\n", + "Example n. 1378 = (39.0, 459.0): cluster 1\n", + "Example n. 1379 = (39.0, 458.0): cluster 1\n", + "Example n. 1380 = (39.0, 457.0): cluster 1\n", + "Example n. 1381 = (39.0, 456.0): cluster 1\n", + "Example n. 1382 = (39.0, 455.0): cluster 1\n", + "Example n. 1383 = (39.0, 454.0): cluster 1\n", + "Example n. 1384 = (39.0, 453.0): cluster 1\n", + "Example n. 1385 = (39.0, 452.0): cluster 1\n", + "Example n. 1386 = (39.0, 451.0): cluster 1\n", + "Example n. 1387 = (39.0, 450.0): cluster 1\n", + "Example n. 1388 = (39.0, 449.0): cluster 1\n", + "Example n. 1389 = (39.0, 448.0): cluster 1\n", + "Example n. 1390 = (39.0, 429.0): cluster 3\n", + "Example n. 1391 = (39.0, 428.0): cluster 3\n", + "Example n. 1392 = (39.0, 427.0): cluster 3\n", + "Example n. 1393 = (39.0, 426.0): cluster 3\n", + "Example n. 1394 = (39.0, 425.0): cluster 3\n", + "Example n. 1395 = (39.0, 424.0): cluster 3\n", + "Example n. 1396 = (39.0, 423.0): cluster 3\n", + "Example n. 1397 = (39.0, 422.0): cluster 3\n", + "Example n. 1398 = (39.0, 421.0): cluster 3\n", + "Example n. 1399 = (39.0, 420.0): cluster 3\n", + "Example n. 1400 = (39.0, 419.0): cluster 3\n", + "Example n. 1401 = (39.0, 418.0): cluster 3\n", + "Example n. 1402 = (39.0, 417.0): cluster 3\n", + "Example n. 1403 = (39.0, 416.0): cluster 3\n", + "Example n. 1404 = (39.0, 415.0): cluster 3\n", + "Example n. 1405 = (39.0, 414.0): cluster 3\n", + "Example n. 1406 = (39.0, 413.0): cluster 3\n", + "Example n. 1407 = (39.0, 412.0): cluster 3\n", + "Example n. 1408 = (39.0, 411.0): cluster 3\n", + "Example n. 1409 = (39.0, 410.0): cluster 3\n", + "Example n. 1410 = (39.0, 409.0): cluster 3\n", + "Example n. 1411 = (39.0, 408.0): cluster 3\n", + "Example n. 1412 = (39.0, 407.0): cluster 3\n", + "Example n. 1413 = (39.0, 406.0): cluster 3\n", + "Example n. 1414 = (39.0, 405.0): cluster 3\n", + "Example n. 1415 = (39.0, 404.0): cluster 3\n", + "Example n. 1416 = (39.0, 403.0): cluster 3\n", + "Example n. 1417 = (39.0, 402.0): cluster 3\n", + "Example n. 1418 = (39.0, 401.0): cluster 3\n", + "Example n. 1419 = (39.0, 400.0): cluster 3\n", + "Example n. 1420 = (39.0, 399.0): cluster 3\n", + "Example n. 1421 = (39.0, 398.0): cluster 3\n", + "Example n. 1422 = (40.0, 506.0): cluster 0\n", + "Example n. 1423 = (40.0, 505.0): cluster 0\n", + "Example n. 1424 = (40.0, 504.0): cluster 0\n", + "Example n. 1425 = (40.0, 503.0): cluster 0\n", + "Example n. 1426 = (40.0, 502.0): cluster 0\n", + "Example n. 1427 = (40.0, 501.0): cluster 0\n", + "Example n. 1428 = (40.0, 500.0): cluster 0\n", + "Example n. 1429 = (40.0, 499.0): cluster 0\n", + "Example n. 1430 = (40.0, 498.0): cluster 0\n", + "Example n. 1431 = (40.0, 497.0): cluster 0\n", + "Example n. 1432 = (40.0, 496.0): cluster 0\n", + "Example n. 1433 = (40.0, 495.0): cluster 0\n", + "Example n. 1434 = (40.0, 494.0): cluster 0\n", + "Example n. 1435 = (40.0, 493.0): cluster 0\n", + "Example n. 1436 = (40.0, 492.0): cluster 0\n", + "Example n. 1437 = (40.0, 491.0): cluster 0\n", + "Example n. 1438 = (40.0, 490.0): cluster 0\n", + "Example n. 1439 = (40.0, 489.0): cluster 0\n", + "Example n. 1440 = (40.0, 488.0): cluster 0\n", + "Example n. 1441 = (40.0, 487.0): cluster 0\n", + "Example n. 1442 = (40.0, 475.0): cluster 0\n", + "Example n. 1443 = (40.0, 474.0): cluster 0\n", + "Example n. 1444 = (40.0, 473.0): cluster 0\n", + "Example n. 1445 = (40.0, 472.0): cluster 0\n", + "Example n. 1446 = (40.0, 471.0): cluster 0\n", + "Example n. 1447 = (40.0, 470.0): cluster 0\n", + "Example n. 1448 = (40.0, 469.0): cluster 0\n", + "Example n. 1449 = (40.0, 468.0): cluster 0\n", + "Example n. 1450 = (40.0, 467.0): cluster 0\n", + "Example n. 1451 = (40.0, 466.0): cluster 0\n", + "Example n. 1452 = (40.0, 465.0): cluster 0\n", + "Example n. 1453 = (40.0, 464.0): cluster 0\n", + "Example n. 1454 = (40.0, 459.0): cluster 1\n", + "Example n. 1455 = (40.0, 458.0): cluster 1\n", + "Example n. 1456 = (40.0, 457.0): cluster 1\n", + "Example n. 1457 = (40.0, 456.0): cluster 1\n", + "Example n. 1458 = (40.0, 455.0): cluster 1\n", + "Example n. 1459 = (40.0, 454.0): cluster 1\n", + "Example n. 1460 = (40.0, 453.0): cluster 1\n", + "Example n. 1461 = (40.0, 452.0): cluster 1\n", + "Example n. 1462 = (40.0, 451.0): cluster 1\n", + "Example n. 1463 = (40.0, 450.0): cluster 1\n", + "Example n. 1464 = (40.0, 449.0): cluster 1\n", + "Example n. 1465 = (40.0, 448.0): cluster 1\n", + "Example n. 1466 = (40.0, 429.0): cluster 3\n", + "Example n. 1467 = (40.0, 428.0): cluster 3\n", + "Example n. 1468 = (40.0, 427.0): cluster 3\n", + "Example n. 1469 = (40.0, 426.0): cluster 3\n", + "Example n. 1470 = (40.0, 425.0): cluster 3\n", + "Example n. 1471 = (40.0, 424.0): cluster 3\n", + "Example n. 1472 = (40.0, 423.0): cluster 3\n", + "Example n. 1473 = (40.0, 422.0): cluster 3\n", + "Example n. 1474 = (40.0, 421.0): cluster 3\n", + "Example n. 1475 = (40.0, 420.0): cluster 3\n", + "Example n. 1476 = (40.0, 419.0): cluster 3\n", + "Example n. 1477 = (40.0, 418.0): cluster 3\n", + "Example n. 1478 = (40.0, 417.0): cluster 3\n", + "Example n. 1479 = (40.0, 416.0): cluster 3\n", + "Example n. 1480 = (40.0, 415.0): cluster 3\n", + "Example n. 1481 = (40.0, 414.0): cluster 3\n", + "Example n. 1482 = (40.0, 413.0): cluster 3\n", + "Example n. 1483 = (40.0, 412.0): cluster 3\n", + "Example n. 1484 = (40.0, 411.0): cluster 3\n", + "Example n. 1485 = (40.0, 410.0): cluster 3\n", + "Example n. 1486 = (40.0, 409.0): cluster 3\n", + "Example n. 1487 = (40.0, 408.0): cluster 3\n", + "Example n. 1488 = (40.0, 407.0): cluster 3\n", + "Example n. 1489 = (40.0, 406.0): cluster 3\n", + "Example n. 1490 = (40.0, 405.0): cluster 3\n", + "Example n. 1491 = (40.0, 404.0): cluster 3\n", + "Example n. 1492 = (40.0, 403.0): cluster 3\n", + "Example n. 1493 = (40.0, 402.0): cluster 3\n", + "Example n. 1494 = (40.0, 401.0): cluster 3\n", + "Example n. 1495 = (40.0, 400.0): cluster 3\n", + "Example n. 1496 = (40.0, 399.0): cluster 3\n", + "Example n. 1497 = (40.0, 398.0): cluster 3\n", + "Example n. 1498 = (41.0, 506.0): cluster 0\n", + "Example n. 1499 = (41.0, 505.0): cluster 0\n", + "Example n. 1500 = (41.0, 504.0): cluster 0\n", + "Example n. 1501 = (41.0, 503.0): cluster 0\n", + "Example n. 1502 = (41.0, 502.0): cluster 0\n", + "Example n. 1503 = (41.0, 501.0): cluster 0\n", + "Example n. 1504 = (41.0, 500.0): cluster 0\n", + "Example n. 1505 = (41.0, 499.0): cluster 0\n", + "Example n. 1506 = (41.0, 498.0): cluster 0\n", + "Example n. 1507 = (41.0, 497.0): cluster 0\n", + "Example n. 1508 = (41.0, 496.0): cluster 0\n", + "Example n. 1509 = (41.0, 495.0): cluster 0\n", + "Example n. 1510 = (41.0, 494.0): cluster 0\n", + "Example n. 1511 = (41.0, 493.0): cluster 0\n", + "Example n. 1512 = (41.0, 492.0): cluster 0\n", + "Example n. 1513 = (41.0, 491.0): cluster 0\n", + "Example n. 1514 = (41.0, 490.0): cluster 0\n", + "Example n. 1515 = (41.0, 489.0): cluster 0\n", + "Example n. 1516 = (41.0, 488.0): cluster 0\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Example n. 1517 = (41.0, 487.0): cluster 0\n", + "Example n. 1518 = (41.0, 475.0): cluster 0\n", + "Example n. 1519 = (41.0, 474.0): cluster 0\n", + "Example n. 1520 = (41.0, 473.0): cluster 0\n", + "Example n. 1521 = (41.0, 472.0): cluster 0\n", + "Example n. 1522 = (41.0, 471.0): cluster 0\n", + "Example n. 1523 = (41.0, 470.0): cluster 0\n", + "Example n. 1524 = (41.0, 469.0): cluster 0\n", + "Example n. 1525 = (41.0, 468.0): cluster 0\n", + "Example n. 1526 = (41.0, 467.0): cluster 0\n", + "Example n. 1527 = (41.0, 466.0): cluster 0\n", + "Example n. 1528 = (41.0, 465.0): cluster 1\n", + "Example n. 1529 = (41.0, 464.0): cluster 1\n", + "Example n. 1530 = (41.0, 459.0): cluster 1\n", + "Example n. 1531 = (41.0, 458.0): cluster 1\n", + "Example n. 1532 = (41.0, 457.0): cluster 1\n", + "Example n. 1533 = (41.0, 456.0): cluster 1\n", + "Example n. 1534 = (41.0, 455.0): cluster 1\n", + "Example n. 1535 = (41.0, 454.0): cluster 1\n", + "Example n. 1536 = (41.0, 453.0): cluster 1\n", + "Example n. 1537 = (41.0, 452.0): cluster 1\n", + "Example n. 1538 = (41.0, 451.0): cluster 1\n", + "Example n. 1539 = (41.0, 450.0): cluster 1\n", + "Example n. 1540 = (41.0, 449.0): cluster 1\n", + "Example n. 1541 = (41.0, 448.0): cluster 1\n", + "Example n. 1542 = (41.0, 429.0): cluster 3\n", + "Example n. 1543 = (41.0, 428.0): cluster 3\n", + "Example n. 1544 = (41.0, 427.0): cluster 3\n", + "Example n. 1545 = (41.0, 426.0): cluster 3\n", + "Example n. 1546 = (41.0, 425.0): cluster 3\n", + "Example n. 1547 = (41.0, 424.0): cluster 3\n", + "Example n. 1548 = (41.0, 423.0): cluster 3\n", + "Example n. 1549 = (41.0, 422.0): cluster 3\n", + "Example n. 1550 = (41.0, 421.0): cluster 3\n", + "Example n. 1551 = (41.0, 420.0): cluster 3\n", + "Example n. 1552 = (41.0, 419.0): cluster 3\n", + "Example n. 1553 = (41.0, 418.0): cluster 3\n", + "Example n. 1554 = (41.0, 417.0): cluster 3\n", + "Example n. 1555 = (41.0, 416.0): cluster 3\n", + "Example n. 1556 = (41.0, 415.0): cluster 3\n", + "Example n. 1557 = (41.0, 414.0): cluster 3\n", + "Example n. 1558 = (41.0, 413.0): cluster 3\n", + "Example n. 1559 = (41.0, 412.0): cluster 3\n", + "Example n. 1560 = (41.0, 411.0): cluster 3\n", + "Example n. 1561 = (41.0, 410.0): cluster 3\n", + "Example n. 1562 = (41.0, 409.0): cluster 3\n", + "Example n. 1563 = (41.0, 408.0): cluster 3\n", + "Example n. 1564 = (41.0, 407.0): cluster 3\n", + "Example n. 1565 = (41.0, 406.0): cluster 3\n", + "Example n. 1566 = (41.0, 405.0): cluster 3\n", + "Example n. 1567 = (41.0, 404.0): cluster 3\n", + "Example n. 1568 = (41.0, 403.0): cluster 3\n", + "Example n. 1569 = (41.0, 402.0): cluster 3\n", + "Example n. 1570 = (41.0, 401.0): cluster 3\n", + "Example n. 1571 = (41.0, 400.0): cluster 3\n", + "Example n. 1572 = (41.0, 399.0): cluster 3\n", + "Example n. 1573 = (41.0, 398.0): cluster 3\n", + "Example n. 1574 = (42.0, 506.0): cluster 0\n", + "Example n. 1575 = (42.0, 505.0): cluster 0\n", + "Example n. 1576 = (42.0, 504.0): cluster 0\n", + "Example n. 1577 = (42.0, 503.0): cluster 0\n", + "Example n. 1578 = (42.0, 502.0): cluster 0\n", + "Example n. 1579 = (42.0, 501.0): cluster 0\n", + "Example n. 1580 = (42.0, 500.0): cluster 0\n", + "Example n. 1581 = (42.0, 499.0): cluster 0\n", + "Example n. 1582 = (42.0, 498.0): cluster 0\n", + "Example n. 1583 = (42.0, 497.0): cluster 0\n", + "Example n. 1584 = (42.0, 496.0): cluster 0\n", + "Example n. 1585 = (42.0, 495.0): cluster 0\n", + "Example n. 1586 = (42.0, 494.0): cluster 0\n", + "Example n. 1587 = (42.0, 493.0): cluster 0\n", + "Example n. 1588 = (42.0, 492.0): cluster 0\n", + "Example n. 1589 = (42.0, 491.0): cluster 0\n", + "Example n. 1590 = (42.0, 490.0): cluster 0\n", + "Example n. 1591 = (42.0, 489.0): cluster 0\n", + "Example n. 1592 = (42.0, 488.0): cluster 0\n", + "Example n. 1593 = (42.0, 487.0): cluster 0\n", + "Example n. 1594 = (42.0, 475.0): cluster 0\n", + "Example n. 1595 = (42.0, 474.0): cluster 0\n", + "Example n. 1596 = (42.0, 473.0): cluster 0\n", + "Example n. 1597 = (42.0, 472.0): cluster 0\n", + "Example n. 1598 = (42.0, 471.0): cluster 0\n", + "Example n. 1599 = (42.0, 470.0): cluster 0\n", + "Example n. 1600 = (42.0, 469.0): cluster 0\n", + "Example n. 1601 = (42.0, 468.0): cluster 0\n", + "Example n. 1602 = (42.0, 467.0): cluster 1\n", + "Example n. 1603 = (42.0, 466.0): cluster 1\n", + "Example n. 1604 = (42.0, 465.0): cluster 1\n", + "Example n. 1605 = (42.0, 464.0): cluster 1\n", + "Example n. 1606 = (42.0, 459.0): cluster 1\n", + "Example n. 1607 = (42.0, 458.0): cluster 1\n", + "Example n. 1608 = (42.0, 457.0): cluster 1\n", + "Example n. 1609 = (42.0, 456.0): cluster 1\n", + "Example n. 1610 = (42.0, 455.0): cluster 1\n", + "Example n. 1611 = (42.0, 454.0): cluster 1\n", + "Example n. 1612 = (42.0, 453.0): cluster 1\n", + "Example n. 1613 = (42.0, 452.0): cluster 1\n", + "Example n. 1614 = (42.0, 451.0): cluster 1\n", + "Example n. 1615 = (42.0, 450.0): cluster 1\n", + "Example n. 1616 = (42.0, 449.0): cluster 1\n", + "Example n. 1617 = (42.0, 448.0): cluster 1\n", + "Example n. 1618 = (42.0, 429.0): cluster 3\n", + "Example n. 1619 = (42.0, 428.0): cluster 3\n", + "Example n. 1620 = (42.0, 427.0): cluster 3\n", + "Example n. 1621 = (42.0, 426.0): cluster 3\n", + "Example n. 1622 = (42.0, 425.0): cluster 3\n", + "Example n. 1623 = (42.0, 424.0): cluster 3\n", + "Example n. 1624 = (42.0, 423.0): cluster 3\n", + "Example n. 1625 = (42.0, 422.0): cluster 3\n", + "Example n. 1626 = (42.0, 421.0): cluster 3\n", + "Example n. 1627 = (42.0, 420.0): cluster 3\n", + "Example n. 1628 = (42.0, 419.0): cluster 3\n", + "Example n. 1629 = (42.0, 418.0): cluster 3\n", + "Example n. 1630 = (42.0, 417.0): cluster 3\n", + "Example n. 1631 = (42.0, 416.0): cluster 3\n", + "Example n. 1632 = (42.0, 415.0): cluster 3\n", + "Example n. 1633 = (42.0, 414.0): cluster 3\n", + "Example n. 1634 = (42.0, 413.0): cluster 3\n", + "Example n. 1635 = (42.0, 412.0): cluster 3\n", + "Example n. 1636 = (42.0, 411.0): cluster 3\n", + "Example n. 1637 = (42.0, 410.0): cluster 3\n", + "Example n. 1638 = (42.0, 409.0): cluster 3\n", + "Example n. 1639 = (42.0, 408.0): cluster 3\n", + "Example n. 1640 = (42.0, 406.0): cluster 3\n", + "Example n. 1641 = (42.0, 405.0): cluster 3\n", + "Example n. 1642 = (42.0, 404.0): cluster 3\n", + "Example n. 1643 = (42.0, 403.0): cluster 3\n", + "Example n. 1644 = (42.0, 402.0): cluster 3\n", + "Example n. 1645 = (42.0, 401.0): cluster 3\n", + "Example n. 1646 = (42.0, 400.0): cluster 3\n", + "Example n. 1647 = (42.0, 399.0): cluster 3\n", + "Example n. 1648 = (42.0, 398.0): cluster 3\n", + "Example n. 1649 = (43.0, 506.0): cluster 0\n", + "Example n. 1650 = (43.0, 505.0): cluster 0\n", + "Example n. 1651 = (43.0, 504.0): cluster 0\n", + "Example n. 1652 = (43.0, 503.0): cluster 0\n", + "Example n. 1653 = (43.0, 502.0): cluster 0\n", + "Example n. 1654 = (43.0, 501.0): cluster 0\n", + "Example n. 1655 = (43.0, 500.0): cluster 0\n", + "Example n. 1656 = (43.0, 499.0): cluster 0\n", + "Example n. 1657 = (43.0, 498.0): cluster 0\n", + "Example n. 1658 = (43.0, 497.0): cluster 0\n", + "Example n. 1659 = (43.0, 496.0): cluster 0\n", + "Example n. 1660 = (43.0, 495.0): cluster 0\n", + "Example n. 1661 = (43.0, 494.0): cluster 0\n", + "Example n. 1662 = (43.0, 493.0): cluster 0\n", + "Example n. 1663 = (43.0, 492.0): cluster 0\n", + "Example n. 1664 = (43.0, 491.0): cluster 0\n", + "Example n. 1665 = (43.0, 490.0): cluster 0\n", + "Example n. 1666 = (43.0, 489.0): cluster 0\n", + "Example n. 1667 = (43.0, 488.0): cluster 0\n", + "Example n. 1668 = (43.0, 487.0): cluster 0\n", + "Example n. 1669 = (43.0, 475.0): cluster 0\n", + "Example n. 1670 = (43.0, 474.0): cluster 0\n", + "Example n. 1671 = (43.0, 473.0): cluster 0\n", + "Example n. 1672 = (43.0, 472.0): cluster 0\n", + "Example n. 1673 = (43.0, 471.0): cluster 0\n", + "Example n. 1674 = (43.0, 470.0): cluster 1\n", + "Example n. 1675 = (43.0, 469.0): cluster 1\n", + "Example n. 1676 = (43.0, 468.0): cluster 1\n", + "Example n. 1677 = (43.0, 467.0): cluster 1\n", + "Example n. 1678 = (43.0, 466.0): cluster 1\n", + "Example n. 1679 = (43.0, 465.0): cluster 1\n", + "Example n. 1680 = (43.0, 464.0): cluster 1\n", + "Example n. 1681 = (43.0, 463.0): cluster 1\n", + "Example n. 1682 = (43.0, 462.0): cluster 1\n", + "Example n. 1683 = (43.0, 461.0): cluster 1\n", + "Example n. 1684 = (43.0, 460.0): cluster 1\n", + "Example n. 1685 = (43.0, 459.0): cluster 1\n", + "Example n. 1686 = (43.0, 458.0): cluster 1\n", + "Example n. 1687 = (43.0, 457.0): cluster 1\n", + "Example n. 1688 = (43.0, 456.0): cluster 1\n", + "Example n. 1689 = (43.0, 455.0): cluster 1\n", + "Example n. 1690 = (43.0, 454.0): cluster 1\n", + "Example n. 1691 = (43.0, 453.0): cluster 1\n", + "Example n. 1692 = (43.0, 452.0): cluster 1\n", + "Example n. 1693 = (43.0, 451.0): cluster 1\n", + "Example n. 1694 = (43.0, 450.0): cluster 1\n", + "Example n. 1695 = (43.0, 449.0): cluster 1\n", + "Example n. 1696 = (43.0, 448.0): cluster 1\n", + "Example n. 1697 = (43.0, 432.0): cluster 3\n", + "Example n. 1698 = (43.0, 431.0): cluster 3\n", + "Example n. 1699 = (43.0, 430.0): cluster 3\n", + "Example n. 1700 = (43.0, 429.0): cluster 3\n", + "Example n. 1701 = (43.0, 428.0): cluster 3\n", + "Example n. 1702 = (43.0, 427.0): cluster 3\n", + "Example n. 1703 = (43.0, 426.0): cluster 3\n", + "Example n. 1704 = (43.0, 425.0): cluster 3\n", + "Example n. 1705 = (43.0, 424.0): cluster 3\n", + "Example n. 1706 = (43.0, 423.0): cluster 3\n", + "Example n. 1707 = (43.0, 422.0): cluster 3\n", + "Example n. 1708 = (43.0, 421.0): cluster 3\n", + "Example n. 1709 = (43.0, 420.0): cluster 3\n", + "Example n. 1710 = (43.0, 419.0): cluster 3\n", + "Example n. 1711 = (43.0, 418.0): cluster 3\n", + "Example n. 1712 = (43.0, 417.0): cluster 3\n", + "Example n. 1713 = (43.0, 416.0): cluster 3\n", + "Example n. 1714 = (43.0, 415.0): cluster 3\n", + "Example n. 1715 = (43.0, 414.0): cluster 3\n", + "Example n. 1716 = (43.0, 413.0): cluster 3\n", + "Example n. 1717 = (43.0, 412.0): cluster 3\n", + "Example n. 1718 = (43.0, 411.0): cluster 3\n", + "Example n. 1719 = (43.0, 410.0): cluster 3\n", + "Example n. 1720 = (43.0, 409.0): cluster 3\n", + "Example n. 1721 = (43.0, 408.0): cluster 3\n", + "Example n. 1722 = (43.0, 407.0): cluster 3\n", + "Example n. 1723 = (43.0, 406.0): cluster 3\n", + "Example n. 1724 = (43.0, 405.0): cluster 3\n", + "Example n. 1725 = (43.0, 404.0): cluster 3\n", + "Example n. 1726 = (43.0, 403.0): cluster 3\n", + "Example n. 1727 = (43.0, 402.0): cluster 3\n", + "Example n. 1728 = (43.0, 401.0): cluster 3\n", + "Example n. 1729 = (43.0, 400.0): cluster 3\n", + "Example n. 1730 = (43.0, 399.0): cluster 3\n", + "Example n. 1731 = (43.0, 398.0): cluster 3\n", + "Example n. 1732 = (44.0, 502.0): cluster 0\n", + "Example n. 1733 = (44.0, 501.0): cluster 0\n", + "Example n. 1734 = (44.0, 500.0): cluster 0\n", + "Example n. 1735 = (44.0, 499.0): cluster 0\n", + "Example n. 1736 = (44.0, 498.0): cluster 0\n", + "Example n. 1737 = (44.0, 497.0): cluster 0\n", + "Example n. 1738 = (44.0, 496.0): cluster 0\n", + "Example n. 1739 = (44.0, 495.0): cluster 0\n", + "Example n. 1740 = (44.0, 494.0): cluster 0\n", + "Example n. 1741 = (44.0, 493.0): cluster 0\n", + "Example n. 1742 = (44.0, 492.0): cluster 0\n", + "Example n. 1743 = (44.0, 491.0): cluster 0\n", + "Example n. 1744 = (44.0, 490.0): cluster 0\n", + "Example n. 1745 = (44.0, 489.0): cluster 0\n", + "Example n. 1746 = (44.0, 488.0): cluster 0\n", + "Example n. 1747 = (44.0, 487.0): cluster 0\n", + "Example n. 1748 = (44.0, 475.0): cluster 0\n", + "Example n. 1749 = (44.0, 474.0): cluster 0\n", + "Example n. 1750 = (44.0, 473.0): cluster 0\n", + "Example n. 1751 = (44.0, 472.0): cluster 1\n", + "Example n. 1752 = (44.0, 471.0): cluster 1\n", + "Example n. 1753 = (44.0, 470.0): cluster 1\n", + "Example n. 1754 = (44.0, 469.0): cluster 1\n", + "Example n. 1755 = (44.0, 468.0): cluster 1\n", + "Example n. 1756 = (44.0, 463.0): cluster 1\n", + "Example n. 1757 = (44.0, 462.0): cluster 1\n", + "Example n. 1758 = (44.0, 461.0): cluster 1\n", + "Example n. 1759 = (44.0, 460.0): cluster 1\n", + "Example n. 1760 = (44.0, 459.0): cluster 1\n", + "Example n. 1761 = (44.0, 458.0): cluster 1\n", + "Example n. 1762 = (44.0, 457.0): cluster 1\n", + "Example n. 1763 = (44.0, 456.0): cluster 1\n", + "Example n. 1764 = (44.0, 455.0): cluster 1\n", + "Example n. 1765 = (44.0, 454.0): cluster 1\n", + "Example n. 1766 = (44.0, 453.0): cluster 1\n", + "Example n. 1767 = (44.0, 452.0): cluster 1\n", + "Example n. 1768 = (44.0, 451.0): cluster 1\n", + "Example n. 1769 = (44.0, 450.0): cluster 1\n", + "Example n. 1770 = (44.0, 449.0): cluster 1\n", + "Example n. 1771 = (44.0, 448.0): cluster 1\n", + "Example n. 1772 = (44.0, 432.0): cluster 3\n", + "Example n. 1773 = (44.0, 431.0): cluster 3\n", + "Example n. 1774 = (44.0, 430.0): cluster 3\n", + "Example n. 1775 = (44.0, 429.0): cluster 3\n", + "Example n. 1776 = (44.0, 428.0): cluster 3\n", + "Example n. 1777 = (44.0, 427.0): cluster 3\n", + "Example n. 1778 = (44.0, 426.0): cluster 3\n", + "Example n. 1779 = (44.0, 425.0): cluster 3\n", + "Example n. 1780 = (44.0, 424.0): cluster 3\n", + "Example n. 1781 = (44.0, 423.0): cluster 3\n", + "Example n. 1782 = (44.0, 422.0): cluster 3\n", + "Example n. 1783 = (44.0, 421.0): cluster 3\n", + "Example n. 1784 = (44.0, 420.0): cluster 3\n", + "Example n. 1785 = (44.0, 419.0): cluster 3\n", + "Example n. 1786 = (44.0, 418.0): cluster 3\n", + "Example n. 1787 = (44.0, 417.0): cluster 3\n", + "Example n. 1788 = (44.0, 416.0): cluster 3\n", + "Example n. 1789 = (44.0, 415.0): cluster 3\n", + "Example n. 1790 = (44.0, 414.0): cluster 3\n", + "Example n. 1791 = (44.0, 409.0): cluster 3\n", + "Example n. 1792 = (44.0, 408.0): cluster 3\n", + "Example n. 1793 = (44.0, 407.0): cluster 3\n", + "Example n. 1794 = (44.0, 406.0): cluster 3\n", + "Example n. 1795 = (44.0, 405.0): cluster 3\n", + "Example n. 1796 = (44.0, 404.0): cluster 3\n", + "Example n. 1797 = (44.0, 403.0): cluster 3\n", + "Example n. 1798 = (44.0, 402.0): cluster 3\n", + "Example n. 1799 = (44.0, 401.0): cluster 3\n", + "Example n. 1800 = (44.0, 400.0): cluster 3\n", + "Example n. 1801 = (44.0, 399.0): cluster 3\n", + "Example n. 1802 = (44.0, 398.0): cluster 3\n", + "Example n. 1803 = (45.0, 502.0): cluster 0\n", + "Example n. 1804 = (45.0, 501.0): cluster 0\n", + "Example n. 1805 = (45.0, 500.0): cluster 0\n", + "Example n. 1806 = (45.0, 499.0): cluster 0\n", + "Example n. 1807 = (45.0, 498.0): cluster 0\n", + "Example n. 1808 = (45.0, 497.0): cluster 0\n", + "Example n. 1809 = (45.0, 496.0): cluster 0\n", + "Example n. 1810 = (45.0, 495.0): cluster 0\n", + "Example n. 1811 = (45.0, 494.0): cluster 0\n", + "Example n. 1812 = (45.0, 493.0): cluster 0\n", + "Example n. 1813 = (45.0, 492.0): cluster 0\n", + "Example n. 1814 = (45.0, 491.0): cluster 0\n", + "Example n. 1815 = (45.0, 490.0): cluster 0\n", + "Example n. 1816 = (45.0, 489.0): cluster 0\n", + "Example n. 1817 = (45.0, 488.0): cluster 0\n", + "Example n. 1818 = (45.0, 487.0): cluster 0\n", + "Example n. 1819 = (45.0, 475.0): cluster 1\n", + "Example n. 1820 = (45.0, 474.0): cluster 1\n", + "Example n. 1821 = (45.0, 473.0): cluster 1\n", + "Example n. 1822 = (45.0, 472.0): cluster 1\n", + "Example n. 1823 = (45.0, 471.0): cluster 1\n", + "Example n. 1824 = (45.0, 470.0): cluster 1\n", + "Example n. 1825 = (45.0, 469.0): cluster 1\n", + "Example n. 1826 = (45.0, 468.0): cluster 1\n", + "Example n. 1827 = (45.0, 463.0): cluster 1\n", + "Example n. 1828 = (45.0, 462.0): cluster 1\n", + "Example n. 1829 = (45.0, 461.0): cluster 1\n", + "Example n. 1830 = (45.0, 460.0): cluster 1\n", + "Example n. 1831 = (45.0, 459.0): cluster 1\n", + "Example n. 1832 = (45.0, 458.0): cluster 1\n", + "Example n. 1833 = (45.0, 457.0): cluster 1\n", + "Example n. 1834 = (45.0, 456.0): cluster 1\n", + "Example n. 1835 = (45.0, 455.0): cluster 1\n", + "Example n. 1836 = (45.0, 454.0): cluster 1\n", + "Example n. 1837 = (45.0, 453.0): cluster 1\n", + "Example n. 1838 = (45.0, 452.0): cluster 1\n", + "Example n. 1839 = (45.0, 451.0): cluster 1\n", + "Example n. 1840 = (45.0, 450.0): cluster 1\n", + "Example n. 1841 = (45.0, 449.0): cluster 1\n", + "Example n. 1842 = (45.0, 448.0): cluster 1\n", + "Example n. 1843 = (45.0, 432.0): cluster 3\n", + "Example n. 1844 = (45.0, 431.0): cluster 3\n", + "Example n. 1845 = (45.0, 430.0): cluster 3\n", + "Example n. 1846 = (45.0, 429.0): cluster 3\n", + "Example n. 1847 = (45.0, 428.0): cluster 3\n", + "Example n. 1848 = (45.0, 427.0): cluster 3\n", + "Example n. 1849 = (45.0, 426.0): cluster 3\n", + "Example n. 1850 = (45.0, 425.0): cluster 3\n", + "Example n. 1851 = (45.0, 424.0): cluster 3\n", + "Example n. 1852 = (45.0, 423.0): cluster 3\n", + "Example n. 1853 = (45.0, 422.0): cluster 3\n", + "Example n. 1854 = (45.0, 421.0): cluster 3\n", + "Example n. 1855 = (45.0, 420.0): cluster 3\n", + "Example n. 1856 = (45.0, 419.0): cluster 3\n", + "Example n. 1857 = (45.0, 418.0): cluster 3\n", + "Example n. 1858 = (45.0, 417.0): cluster 3\n", + "Example n. 1859 = (45.0, 416.0): cluster 3\n", + "Example n. 1860 = (45.0, 415.0): cluster 3\n", + "Example n. 1861 = (45.0, 414.0): cluster 3\n", + "Example n. 1862 = (45.0, 409.0): cluster 3\n", + "Example n. 1863 = (45.0, 408.0): cluster 3\n", + "Example n. 1864 = (45.0, 407.0): cluster 3\n", + "Example n. 1865 = (45.0, 406.0): cluster 3\n", + "Example n. 1866 = (45.0, 405.0): cluster 3\n", + "Example n. 1867 = (45.0, 404.0): cluster 3\n", + "Example n. 1868 = (45.0, 403.0): cluster 3\n", + "Example n. 1869 = (45.0, 402.0): cluster 3\n", + "Example n. 1870 = (45.0, 401.0): cluster 3\n", + "Example n. 1871 = (45.0, 400.0): cluster 3\n", + "Example n. 1872 = (45.0, 399.0): cluster 3\n", + "Example n. 1873 = (45.0, 398.0): cluster 3\n", + "Example n. 1874 = (46.0, 502.0): cluster 0\n", + "Example n. 1875 = (46.0, 501.0): cluster 0\n", + "Example n. 1876 = (46.0, 500.0): cluster 0\n", + "Example n. 1877 = (46.0, 499.0): cluster 0\n", + "Example n. 1878 = (46.0, 498.0): cluster 0\n", + "Example n. 1879 = (46.0, 497.0): cluster 0\n", + "Example n. 1880 = (46.0, 496.0): cluster 0\n", + "Example n. 1881 = (46.0, 495.0): cluster 0\n", + "Example n. 1882 = (46.0, 494.0): cluster 0\n", + "Example n. 1883 = (46.0, 493.0): cluster 0\n", + "Example n. 1884 = (46.0, 492.0): cluster 0\n", + "Example n. 1885 = (46.0, 491.0): cluster 0\n", + "Example n. 1886 = (46.0, 490.0): cluster 0\n", + "Example n. 1887 = (46.0, 489.0): cluster 0\n", + "Example n. 1888 = (46.0, 488.0): cluster 0\n", + "Example n. 1889 = (46.0, 487.0): cluster 0\n", + "Example n. 1890 = (46.0, 475.0): cluster 1\n", + "Example n. 1891 = (46.0, 474.0): cluster 1\n", + "Example n. 1892 = (46.0, 473.0): cluster 1\n", + "Example n. 1893 = (46.0, 472.0): cluster 1\n", + "Example n. 1894 = (46.0, 471.0): cluster 1\n", + "Example n. 1895 = (46.0, 470.0): cluster 1\n", + "Example n. 1896 = (46.0, 469.0): cluster 1\n", + "Example n. 1897 = (46.0, 468.0): cluster 1\n", + "Example n. 1898 = (46.0, 463.0): cluster 1\n", + "Example n. 1899 = (46.0, 462.0): cluster 1\n", + "Example n. 1900 = (46.0, 461.0): cluster 1\n", + "Example n. 1901 = (46.0, 460.0): cluster 1\n", + "Example n. 1902 = (46.0, 459.0): cluster 1\n", + "Example n. 1903 = (46.0, 458.0): cluster 1\n", + "Example n. 1904 = (46.0, 457.0): cluster 1\n", + "Example n. 1905 = (46.0, 456.0): cluster 1\n", + "Example n. 1906 = (46.0, 455.0): cluster 1\n", + "Example n. 1907 = (46.0, 454.0): cluster 1\n", + "Example n. 1908 = (46.0, 453.0): cluster 1\n", + "Example n. 1909 = (46.0, 452.0): cluster 1\n", + "Example n. 1910 = (46.0, 451.0): cluster 1\n", + "Example n. 1911 = (46.0, 450.0): cluster 1\n", + "Example n. 1912 = (46.0, 449.0): cluster 1\n", + "Example n. 1913 = (46.0, 448.0): cluster 1\n", + "Example n. 1914 = (46.0, 432.0): cluster 3\n", + "Example n. 1915 = (46.0, 431.0): cluster 3\n", + "Example n. 1916 = (46.0, 430.0): cluster 3\n", + "Example n. 1917 = (46.0, 429.0): cluster 3\n", + "Example n. 1918 = (46.0, 428.0): cluster 3\n", + "Example n. 1919 = (46.0, 427.0): cluster 3\n", + "Example n. 1920 = (46.0, 426.0): cluster 3\n", + "Example n. 1921 = (46.0, 425.0): cluster 3\n", + "Example n. 1922 = (46.0, 424.0): cluster 3\n", + "Example n. 1923 = (46.0, 423.0): cluster 3\n", + "Example n. 1924 = (46.0, 422.0): cluster 3\n", + "Example n. 1925 = (46.0, 421.0): cluster 3\n", + "Example n. 1926 = (46.0, 420.0): cluster 3\n", + "Example n. 1927 = (46.0, 419.0): cluster 3\n", + "Example n. 1928 = (46.0, 418.0): cluster 3\n", + "Example n. 1929 = (46.0, 417.0): cluster 3\n", + "Example n. 1930 = (46.0, 416.0): cluster 3\n", + "Example n. 1931 = (46.0, 415.0): cluster 3\n", + "Example n. 1932 = (46.0, 414.0): cluster 3\n", + "Example n. 1933 = (46.0, 409.0): cluster 3\n", + "Example n. 1934 = (46.0, 408.0): cluster 3\n", + "Example n. 1935 = (46.0, 407.0): cluster 3\n", + "Example n. 1936 = (46.0, 406.0): cluster 3\n", + "Example n. 1937 = (46.0, 405.0): cluster 3\n", + "Example n. 1938 = (46.0, 404.0): cluster 3\n", + "Example n. 1939 = (46.0, 403.0): cluster 3\n", + "Example n. 1940 = (46.0, 402.0): cluster 3\n", + "Example n. 1941 = (46.0, 401.0): cluster 3\n", + "Example n. 1942 = (46.0, 400.0): cluster 3\n", + "Example n. 1943 = (46.0, 399.0): cluster 3\n", + "Example n. 1944 = (46.0, 398.0): cluster 3\n", + "Example n. 1945 = (47.0, 502.0): cluster 0\n", + "Example n. 1946 = (47.0, 501.0): cluster 0\n", + "Example n. 1947 = (47.0, 500.0): cluster 0\n", + "Example n. 1948 = (47.0, 499.0): cluster 0\n", + "Example n. 1949 = (47.0, 498.0): cluster 0\n", + "Example n. 1950 = (47.0, 497.0): cluster 0\n", + "Example n. 1951 = (47.0, 496.0): cluster 0\n", + "Example n. 1952 = (47.0, 495.0): cluster 0\n", + "Example n. 1953 = (47.0, 494.0): cluster 0\n", + "Example n. 1954 = (47.0, 493.0): cluster 0\n", + "Example n. 1955 = (47.0, 492.0): cluster 0\n", + "Example n. 1956 = (47.0, 491.0): cluster 0\n", + "Example n. 1957 = (47.0, 490.0): cluster 0\n", + "Example n. 1958 = (47.0, 489.0): cluster 0\n", + "Example n. 1959 = (47.0, 488.0): cluster 0\n", + "Example n. 1960 = (47.0, 487.0): cluster 0\n", + "Example n. 1961 = (47.0, 486.0): cluster 0\n", + "Example n. 1962 = (47.0, 485.0): cluster 0\n", + "Example n. 1963 = (47.0, 484.0): cluster 0\n", + "Example n. 1964 = (47.0, 483.0): cluster 0\n", + "Example n. 1965 = (47.0, 475.0): cluster 1\n", + "Example n. 1966 = (47.0, 474.0): cluster 1\n", + "Example n. 1967 = (47.0, 473.0): cluster 1\n", + "Example n. 1968 = (47.0, 472.0): cluster 1\n", + "Example n. 1969 = (47.0, 471.0): cluster 1\n", + "Example n. 1970 = (47.0, 470.0): cluster 1\n", + "Example n. 1971 = (47.0, 469.0): cluster 1\n", + "Example n. 1972 = (47.0, 468.0): cluster 1\n", + "Example n. 1973 = (47.0, 463.0): cluster 1\n", + "Example n. 1974 = (47.0, 462.0): cluster 1\n", + "Example n. 1975 = (47.0, 461.0): cluster 1\n", + "Example n. 1976 = (47.0, 460.0): cluster 1\n", + "Example n. 1977 = (47.0, 459.0): cluster 1\n", + "Example n. 1978 = (47.0, 458.0): cluster 1\n", + "Example n. 1979 = (47.0, 457.0): cluster 1\n", + "Example n. 1980 = (47.0, 456.0): cluster 1\n", + "Example n. 1981 = (47.0, 455.0): cluster 1\n", + "Example n. 1982 = (47.0, 454.0): cluster 1\n", + "Example n. 1983 = (47.0, 453.0): cluster 1\n", + "Example n. 1984 = (47.0, 452.0): cluster 1\n", + "Example n. 1985 = (47.0, 451.0): cluster 1\n", + "Example n. 1986 = (47.0, 450.0): cluster 1\n", + "Example n. 1987 = (47.0, 449.0): cluster 1\n", + "Example n. 1988 = (47.0, 448.0): cluster 1\n", + "Example n. 1989 = (47.0, 436.0): cluster 3\n", + "Example n. 1990 = (47.0, 435.0): cluster 3\n", + "Example n. 1991 = (47.0, 434.0): cluster 3\n", + "Example n. 1992 = (47.0, 433.0): cluster 3\n", + "Example n. 1993 = (47.0, 432.0): cluster 3\n", + "Example n. 1994 = (47.0, 431.0): cluster 3\n", + "Example n. 1995 = (47.0, 430.0): cluster 3\n", + "Example n. 1996 = (47.0, 429.0): cluster 3\n", + "Example n. 1997 = (47.0, 428.0): cluster 3\n", + "Example n. 1998 = (47.0, 427.0): cluster 3\n", + "Example n. 1999 = (47.0, 426.0): cluster 3\n", + "Example n. 2000 = (47.0, 425.0): cluster 3\n", + "Example n. 2001 = (47.0, 424.0): cluster 3\n", + "Example n. 2002 = (47.0, 423.0): cluster 3\n", + "Example n. 2003 = (47.0, 422.0): cluster 3\n", + "Example n. 2004 = (47.0, 421.0): cluster 3\n", + "Example n. 2005 = (47.0, 420.0): cluster 3\n", + "Example n. 2006 = (47.0, 419.0): cluster 3\n", + "Example n. 2007 = (47.0, 418.0): cluster 3\n", + "Example n. 2008 = (47.0, 417.0): cluster 3\n", + "Example n. 2009 = (47.0, 416.0): cluster 3\n", + "Example n. 2010 = (47.0, 415.0): cluster 3\n", + "Example n. 2011 = (47.0, 414.0): cluster 3\n", + "Example n. 2012 = (47.0, 409.0): cluster 3\n", + "Example n. 2013 = (47.0, 408.0): cluster 3\n", + "Example n. 2014 = (47.0, 407.0): cluster 3\n", + "Example n. 2015 = (47.0, 406.0): cluster 3\n", + "Example n. 2016 = (47.0, 405.0): cluster 3\n", + "Example n. 2017 = (47.0, 404.0): cluster 3\n", + "Example n. 2018 = (47.0, 403.0): cluster 3\n", + "Example n. 2019 = (47.0, 402.0): cluster 3\n", + "Example n. 2020 = (47.0, 401.0): cluster 3\n", + "Example n. 2021 = (47.0, 400.0): cluster 3\n", + "Example n. 2022 = (47.0, 399.0): cluster 3\n", + "Example n. 2023 = (47.0, 398.0): cluster 3\n", + "Example n. 2024 = (48.0, 502.0): cluster 0\n", + "Example n. 2025 = (48.0, 501.0): cluster 0\n", + "Example n. 2026 = (48.0, 500.0): cluster 0\n", + "Example n. 2027 = (48.0, 499.0): cluster 0\n", + "Example n. 2028 = (48.0, 498.0): cluster 0\n", + "Example n. 2029 = (48.0, 497.0): cluster 0\n", + "Example n. 2030 = (48.0, 496.0): cluster 0\n", + "Example n. 2031 = (48.0, 495.0): cluster 0\n", + "Example n. 2032 = (48.0, 494.0): cluster 0\n", + "Example n. 2033 = (48.0, 493.0): cluster 0\n", + "Example n. 2034 = (48.0, 492.0): cluster 0\n", + "Example n. 2035 = (48.0, 491.0): cluster 0\n", + "Example n. 2036 = (48.0, 490.0): cluster 0\n", + "Example n. 2037 = (48.0, 489.0): cluster 0\n", + "Example n. 2038 = (48.0, 488.0): cluster 0\n", + "Example n. 2039 = (48.0, 487.0): cluster 0\n", + "Example n. 2040 = (48.0, 486.0): cluster 0\n", + "Example n. 2041 = (48.0, 485.0): cluster 0\n", + "Example n. 2042 = (48.0, 484.0): cluster 0\n", + "Example n. 2043 = (48.0, 483.0): cluster 0\n", + "Example n. 2044 = (48.0, 463.0): cluster 1\n", + "Example n. 2045 = (48.0, 462.0): cluster 1\n", + "Example n. 2046 = (48.0, 461.0): cluster 1\n", + "Example n. 2047 = (48.0, 460.0): cluster 1\n", + "Example n. 2048 = (48.0, 459.0): cluster 1\n", + "Example n. 2049 = (48.0, 458.0): cluster 1\n", + "Example n. 2050 = (48.0, 457.0): cluster 1\n", + "Example n. 2051 = (48.0, 456.0): cluster 1\n", + "Example n. 2052 = (48.0, 455.0): cluster 1\n", + "Example n. 2053 = (48.0, 454.0): cluster 1\n", + "Example n. 2054 = (48.0, 453.0): cluster 1\n", + "Example n. 2055 = (48.0, 452.0): cluster 1\n", + "Example n. 2056 = (48.0, 451.0): cluster 1\n", + "Example n. 2057 = (48.0, 450.0): cluster 1\n", + "Example n. 2058 = (48.0, 449.0): cluster 1\n", + "Example n. 2059 = (48.0, 448.0): cluster 1\n", + "Example n. 2060 = (48.0, 436.0): cluster 3\n", + "Example n. 2061 = (48.0, 435.0): cluster 3\n", + "Example n. 2062 = (48.0, 434.0): cluster 3\n", + "Example n. 2063 = (48.0, 433.0): cluster 3\n", + "Example n. 2064 = (48.0, 432.0): cluster 3\n", + "Example n. 2065 = (48.0, 431.0): cluster 3\n", + "Example n. 2066 = (48.0, 430.0): cluster 3\n", + "Example n. 2067 = (48.0, 429.0): cluster 3\n", + "Example n. 2068 = (48.0, 428.0): cluster 3\n", + "Example n. 2069 = (48.0, 427.0): cluster 3\n", + "Example n. 2070 = (48.0, 426.0): cluster 3\n", + "Example n. 2071 = (48.0, 425.0): cluster 3\n", + "Example n. 2072 = (48.0, 424.0): cluster 3\n", + "Example n. 2073 = (48.0, 423.0): cluster 3\n", + "Example n. 2074 = (48.0, 422.0): cluster 3\n", + "Example n. 2075 = (48.0, 421.0): cluster 3\n", + "Example n. 2076 = (48.0, 420.0): cluster 3\n", + "Example n. 2077 = (48.0, 419.0): cluster 3\n", + "Example n. 2078 = (48.0, 418.0): cluster 3\n", + "Example n. 2079 = (48.0, 417.0): cluster 3\n", + "Example n. 2080 = (48.0, 409.0): cluster 3\n", + "Example n. 2081 = (48.0, 408.0): cluster 3\n", + "Example n. 2082 = (48.0, 407.0): cluster 3\n", + "Example n. 2083 = (48.0, 406.0): cluster 3\n", + "Example n. 2084 = (48.0, 405.0): cluster 3\n", + "Example n. 2085 = (48.0, 404.0): cluster 3\n", + "Example n. 2086 = (48.0, 403.0): cluster 3\n", + "Example n. 2087 = (48.0, 402.0): cluster 3\n", + "Example n. 2088 = (48.0, 401.0): cluster 3\n", + "Example n. 2089 = (48.0, 400.0): cluster 3\n", + "Example n. 2090 = (48.0, 399.0): cluster 3\n", + "Example n. 2091 = (48.0, 398.0): cluster 3\n", + "Example n. 2092 = (49.0, 502.0): cluster 0\n", + "Example n. 2093 = (49.0, 501.0): cluster 0\n", + "Example n. 2094 = (49.0, 500.0): cluster 0\n", + "Example n. 2095 = (49.0, 499.0): cluster 0\n", + "Example n. 2096 = (49.0, 498.0): cluster 0\n", + "Example n. 2097 = (49.0, 497.0): cluster 0\n", + "Example n. 2098 = (49.0, 496.0): cluster 0\n", + "Example n. 2099 = (49.0, 495.0): cluster 0\n", + "Example n. 2100 = (49.0, 494.0): cluster 0\n", + "Example n. 2101 = (49.0, 493.0): cluster 0\n", + "Example n. 2102 = (49.0, 492.0): cluster 0\n", + "Example n. 2103 = (49.0, 491.0): cluster 0\n", + "Example n. 2104 = (49.0, 490.0): cluster 0\n", + "Example n. 2105 = (49.0, 489.0): cluster 0\n", + "Example n. 2106 = (49.0, 488.0): cluster 0\n", + "Example n. 2107 = (49.0, 487.0): cluster 0\n", + "Example n. 2108 = (49.0, 486.0): cluster 0\n", + "Example n. 2109 = (49.0, 485.0): cluster 1\n", + "Example n. 2110 = (49.0, 484.0): cluster 1\n", + "Example n. 2111 = (49.0, 483.0): cluster 1\n", + "Example n. 2112 = (49.0, 463.0): cluster 1\n", + "Example n. 2113 = (49.0, 462.0): cluster 1\n", + "Example n. 2114 = (49.0, 461.0): cluster 1\n", + "Example n. 2115 = (49.0, 460.0): cluster 1\n", + "Example n. 2116 = (49.0, 459.0): cluster 1\n", + "Example n. 2117 = (49.0, 458.0): cluster 1\n", + "Example n. 2118 = (49.0, 457.0): cluster 1\n", + "Example n. 2119 = (49.0, 456.0): cluster 1\n", + "Example n. 2120 = (49.0, 455.0): cluster 1\n", + "Example n. 2121 = (49.0, 454.0): cluster 1\n", + "Example n. 2122 = (49.0, 453.0): cluster 1\n", + "Example n. 2123 = (49.0, 452.0): cluster 1\n", + "Example n. 2124 = (49.0, 451.0): cluster 1\n", + "Example n. 2125 = (49.0, 450.0): cluster 1\n", + "Example n. 2126 = (49.0, 449.0): cluster 1\n", + "Example n. 2127 = (49.0, 448.0): cluster 1\n", + "Example n. 2128 = (49.0, 436.0): cluster 3\n", + "Example n. 2129 = (49.0, 435.0): cluster 3\n", + "Example n. 2130 = (49.0, 434.0): cluster 3\n", + "Example n. 2131 = (49.0, 433.0): cluster 3\n", + "Example n. 2132 = (49.0, 432.0): cluster 3\n", + "Example n. 2133 = (49.0, 431.0): cluster 3\n", + "Example n. 2134 = (49.0, 430.0): cluster 3\n", + "Example n. 2135 = (49.0, 429.0): cluster 3\n", + "Example n. 2136 = (49.0, 428.0): cluster 3\n", + "Example n. 2137 = (49.0, 427.0): cluster 3\n", + "Example n. 2138 = (49.0, 426.0): cluster 3\n", + "Example n. 2139 = (49.0, 425.0): cluster 3\n", + "Example n. 2140 = (49.0, 424.0): cluster 3\n", + "Example n. 2141 = (49.0, 423.0): cluster 3\n", + "Example n. 2142 = (49.0, 422.0): cluster 3\n", + "Example n. 2143 = (49.0, 421.0): cluster 3\n", + "Example n. 2144 = (49.0, 420.0): cluster 3\n", + "Example n. 2145 = (49.0, 419.0): cluster 3\n", + "Example n. 2146 = (49.0, 418.0): cluster 3\n", + "Example n. 2147 = (49.0, 417.0): cluster 3\n", + "Example n. 2148 = (49.0, 409.0): cluster 3\n", + "Example n. 2149 = (49.0, 408.0): cluster 3\n", + "Example n. 2150 = (49.0, 407.0): cluster 3\n", + "Example n. 2151 = (49.0, 406.0): cluster 3\n", + "Example n. 2152 = (49.0, 405.0): cluster 3\n", + "Example n. 2153 = (49.0, 404.0): cluster 3\n", + "Example n. 2154 = (49.0, 403.0): cluster 3\n", + "Example n. 2155 = (49.0, 402.0): cluster 3\n", + "Example n. 2156 = (49.0, 401.0): cluster 3\n", + "Example n. 2157 = (49.0, 400.0): cluster 3\n", + "Example n. 2158 = (49.0, 399.0): cluster 3\n", + "Example n. 2159 = (49.0, 398.0): cluster 3\n", + "Example n. 2160 = (50.0, 502.0): cluster 0\n", + "Example n. 2161 = (50.0, 501.0): cluster 0\n", + "Example n. 2162 = (50.0, 500.0): cluster 0\n", + "Example n. 2163 = (50.0, 499.0): cluster 0\n", + "Example n. 2164 = (50.0, 498.0): cluster 0\n", + "Example n. 2165 = (50.0, 497.0): cluster 0\n", + "Example n. 2166 = (50.0, 496.0): cluster 0\n", + "Example n. 2167 = (50.0, 495.0): cluster 0\n", + "Example n. 2168 = (50.0, 494.0): cluster 0\n", + "Example n. 2169 = (50.0, 493.0): cluster 0\n", + "Example n. 2170 = (50.0, 492.0): cluster 0\n", + "Example n. 2171 = (50.0, 491.0): cluster 0\n", + "Example n. 2172 = (50.0, 490.0): cluster 0\n", + "Example n. 2173 = (50.0, 489.0): cluster 0\n", + "Example n. 2174 = (50.0, 488.0): cluster 1\n", + "Example n. 2175 = (50.0, 487.0): cluster 1\n", + "Example n. 2176 = (50.0, 486.0): cluster 1\n", + "Example n. 2177 = (50.0, 485.0): cluster 1\n", + "Example n. 2178 = (50.0, 484.0): cluster 1\n", + "Example n. 2179 = (50.0, 483.0): cluster 1\n", + "Example n. 2180 = (50.0, 463.0): cluster 1\n", + "Example n. 2181 = (50.0, 462.0): cluster 1\n", + "Example n. 2182 = (50.0, 461.0): cluster 1\n", + "Example n. 2183 = (50.0, 460.0): cluster 1\n", + "Example n. 2184 = (50.0, 459.0): cluster 1\n", + "Example n. 2185 = (50.0, 458.0): cluster 1\n", + "Example n. 2186 = (50.0, 457.0): cluster 1\n", + "Example n. 2187 = (50.0, 456.0): cluster 1\n", + "Example n. 2188 = (50.0, 455.0): cluster 1\n", + "Example n. 2189 = (50.0, 454.0): cluster 1\n", + "Example n. 2190 = (50.0, 453.0): cluster 1\n", + "Example n. 2191 = (50.0, 452.0): cluster 1\n", + "Example n. 2192 = (50.0, 451.0): cluster 1\n", + "Example n. 2193 = (50.0, 450.0): cluster 1\n", + "Example n. 2194 = (50.0, 449.0): cluster 1\n", + "Example n. 2195 = (50.0, 448.0): cluster 1\n", + "Example n. 2196 = (50.0, 436.0): cluster 3\n", + "Example n. 2197 = (50.0, 435.0): cluster 3\n", + "Example n. 2198 = (50.0, 434.0): cluster 3\n", + "Example n. 2199 = (50.0, 433.0): cluster 3\n", + "Example n. 2200 = (50.0, 432.0): cluster 3\n", + "Example n. 2201 = (50.0, 431.0): cluster 3\n", + "Example n. 2202 = (50.0, 430.0): cluster 3\n", + "Example n. 2203 = (50.0, 429.0): cluster 3\n", + "Example n. 2204 = (50.0, 428.0): cluster 3\n", + "Example n. 2205 = (50.0, 427.0): cluster 3\n", + "Example n. 2206 = (50.0, 426.0): cluster 3\n", + "Example n. 2207 = (50.0, 425.0): cluster 3\n", + "Example n. 2208 = (50.0, 424.0): cluster 3\n", + "Example n. 2209 = (50.0, 423.0): cluster 3\n", + "Example n. 2210 = (50.0, 422.0): cluster 3\n", + "Example n. 2211 = (50.0, 421.0): cluster 3\n", + "Example n. 2212 = (50.0, 420.0): cluster 3\n", + "Example n. 2213 = (50.0, 419.0): cluster 3\n", + "Example n. 2214 = (50.0, 418.0): cluster 3\n", + "Example n. 2215 = (50.0, 417.0): cluster 3\n", + "Example n. 2216 = (50.0, 409.0): cluster 3\n", + "Example n. 2217 = (50.0, 408.0): cluster 3\n", + "Example n. 2218 = (50.0, 407.0): cluster 3\n", + "Example n. 2219 = (50.0, 406.0): cluster 3\n", + "Example n. 2220 = (50.0, 405.0): cluster 3\n", + "Example n. 2221 = (50.0, 404.0): cluster 3\n", + "Example n. 2222 = (50.0, 403.0): cluster 3\n", + "Example n. 2223 = (50.0, 402.0): cluster 3\n", + "Example n. 2224 = (50.0, 401.0): cluster 3\n", + "Example n. 2225 = (50.0, 400.0): cluster 3\n", + "Example n. 2226 = (50.0, 399.0): cluster 3\n", + "Example n. 2227 = (50.0, 398.0): cluster 3\n", + "Example n. 2228 = (51.0, 502.0): cluster 0\n", + "Example n. 2229 = (51.0, 501.0): cluster 0\n", + "Example n. 2230 = (51.0, 500.0): cluster 0\n", + "Example n. 2231 = (51.0, 499.0): cluster 0\n", + "Example n. 2232 = (51.0, 498.0): cluster 0\n", + "Example n. 2233 = (51.0, 497.0): cluster 0\n", + "Example n. 2234 = (51.0, 496.0): cluster 0\n", + "Example n. 2235 = (51.0, 495.0): cluster 0\n", + "Example n. 2236 = (51.0, 494.0): cluster 0\n", + "Example n. 2237 = (51.0, 493.0): cluster 0\n", + "Example n. 2238 = (51.0, 492.0): cluster 0\n", + "Example n. 2239 = (51.0, 491.0): cluster 0\n", + "Example n. 2240 = (51.0, 490.0): cluster 1\n", + "Example n. 2241 = (51.0, 489.0): cluster 1\n", + "Example n. 2242 = (51.0, 488.0): cluster 1\n", + "Example n. 2243 = (51.0, 487.0): cluster 1\n", + "Example n. 2244 = (51.0, 486.0): cluster 1\n", + "Example n. 2245 = (51.0, 485.0): cluster 1\n", + "Example n. 2246 = (51.0, 484.0): cluster 1\n", + "Example n. 2247 = (51.0, 483.0): cluster 1\n", + "Example n. 2248 = (51.0, 482.0): cluster 1\n", + "Example n. 2249 = (51.0, 481.0): cluster 1\n", + "Example n. 2250 = (51.0, 479.0): cluster 1\n", + "Example n. 2251 = (51.0, 459.0): cluster 1\n", + "Example n. 2252 = (51.0, 458.0): cluster 1\n", + "Example n. 2253 = (51.0, 457.0): cluster 1\n", + "Example n. 2254 = (51.0, 456.0): cluster 1\n", + "Example n. 2255 = (51.0, 455.0): cluster 1\n", + "Example n. 2256 = (51.0, 454.0): cluster 1\n", + "Example n. 2257 = (51.0, 453.0): cluster 1\n", + "Example n. 2258 = (51.0, 452.0): cluster 1\n", + "Example n. 2259 = (51.0, 440.0): cluster 3\n", + "Example n. 2260 = (51.0, 439.0): cluster 3\n", + "Example n. 2261 = (51.0, 438.0): cluster 3\n", + "Example n. 2262 = (51.0, 437.0): cluster 3\n", + "Example n. 2263 = (51.0, 436.0): cluster 3\n", + "Example n. 2264 = (51.0, 435.0): cluster 3\n", + "Example n. 2265 = (51.0, 434.0): cluster 3\n", + "Example n. 2266 = (51.0, 433.0): cluster 3\n", + "Example n. 2267 = (51.0, 432.0): cluster 3\n", + "Example n. 2268 = (51.0, 431.0): cluster 3\n", + "Example n. 2269 = (51.0, 430.0): cluster 3\n", + "Example n. 2270 = (51.0, 429.0): cluster 3\n", + "Example n. 2271 = (51.0, 428.0): cluster 3\n", + "Example n. 2272 = (51.0, 427.0): cluster 3\n", + "Example n. 2273 = (51.0, 426.0): cluster 3\n", + "Example n. 2274 = (51.0, 425.0): cluster 3\n", + "Example n. 2275 = (51.0, 424.0): cluster 3\n", + "Example n. 2276 = (51.0, 423.0): cluster 3\n", + "Example n. 2277 = (51.0, 422.0): cluster 3\n", + "Example n. 2278 = (51.0, 421.0): cluster 3\n", + "Example n. 2279 = (51.0, 409.0): cluster 3\n", + "Example n. 2280 = (51.0, 408.0): cluster 3\n", + "Example n. 2281 = (51.0, 407.0): cluster 3\n", + "Example n. 2282 = (51.0, 406.0): cluster 3\n", + "Example n. 2283 = (51.0, 405.0): cluster 3\n", + "Example n. 2284 = (51.0, 404.0): cluster 3\n", + "Example n. 2285 = (51.0, 403.0): cluster 3\n", + "Example n. 2286 = (51.0, 402.0): cluster 3\n", + "Example n. 2287 = (51.0, 401.0): cluster 3\n", + "Example n. 2288 = (51.0, 400.0): cluster 3\n", + "Example n. 2289 = (51.0, 399.0): cluster 3\n", + "Example n. 2290 = (51.0, 398.0): cluster 3\n", + "Example n. 2291 = (52.0, 502.0): cluster 0\n", + "Example n. 2292 = (52.0, 501.0): cluster 0\n", + "Example n. 2293 = (52.0, 500.0): cluster 0\n", + "Example n. 2294 = (52.0, 499.0): cluster 0\n", + "Example n. 2295 = (52.0, 498.0): cluster 0\n", + "Example n. 2296 = (52.0, 497.0): cluster 0\n", + "Example n. 2297 = (52.0, 496.0): cluster 0\n", + "Example n. 2298 = (52.0, 495.0): cluster 0\n", + "Example n. 2299 = (52.0, 494.0): cluster 0\n", + "Example n. 2300 = (52.0, 493.0): cluster 1\n", + "Example n. 2301 = (52.0, 492.0): cluster 1\n", + "Example n. 2302 = (52.0, 491.0): cluster 1\n", + "Example n. 2303 = (52.0, 490.0): cluster 1\n", + "Example n. 2304 = (52.0, 489.0): cluster 1\n", + "Example n. 2305 = (52.0, 488.0): cluster 1\n", + "Example n. 2306 = (52.0, 487.0): cluster 1\n", + "Example n. 2307 = (52.0, 486.0): cluster 1\n", + "Example n. 2308 = (52.0, 485.0): cluster 1\n", + "Example n. 2309 = (52.0, 484.0): cluster 1\n", + "Example n. 2310 = (52.0, 483.0): cluster 1\n", + "Example n. 2311 = (52.0, 482.0): cluster 1\n", + "Example n. 2312 = (52.0, 481.0): cluster 1\n", + "Example n. 2313 = (52.0, 480.0): cluster 1\n", + "Example n. 2314 = (52.0, 459.0): cluster 1\n", + "Example n. 2315 = (52.0, 458.0): cluster 1\n", + "Example n. 2316 = (52.0, 457.0): cluster 1\n", + "Example n. 2317 = (52.0, 456.0): cluster 1\n", + "Example n. 2318 = (52.0, 455.0): cluster 1\n", + "Example n. 2319 = (52.0, 454.0): cluster 1\n", + "Example n. 2320 = (52.0, 453.0): cluster 1\n", + "Example n. 2321 = (52.0, 452.0): cluster 1\n", + "Example n. 2322 = (52.0, 440.0): cluster 3\n", + "Example n. 2323 = (52.0, 439.0): cluster 3\n", + "Example n. 2324 = (52.0, 438.0): cluster 3\n", + "Example n. 2325 = (52.0, 437.0): cluster 3\n", + "Example n. 2326 = (52.0, 436.0): cluster 3\n", + "Example n. 2327 = (52.0, 435.0): cluster 3\n", + "Example n. 2328 = (52.0, 434.0): cluster 3\n", + "Example n. 2329 = (52.0, 433.0): cluster 3\n", + "Example n. 2330 = (52.0, 432.0): cluster 3\n", + "Example n. 2331 = (52.0, 431.0): cluster 3\n", + "Example n. 2332 = (52.0, 430.0): cluster 3\n", + "Example n. 2333 = (52.0, 429.0): cluster 3\n", + "Example n. 2334 = (52.0, 428.0): cluster 3\n", + "Example n. 2335 = (52.0, 427.0): cluster 3\n", + "Example n. 2336 = (52.0, 426.0): cluster 3\n", + "Example n. 2337 = (52.0, 425.0): cluster 3\n", + "Example n. 2338 = (52.0, 424.0): cluster 3\n", + "Example n. 2339 = (52.0, 423.0): cluster 3\n", + "Example n. 2340 = (52.0, 422.0): cluster 3\n", + "Example n. 2341 = (52.0, 421.0): cluster 3\n", + "Example n. 2342 = (52.0, 409.0): cluster 3\n", + "Example n. 2343 = (52.0, 408.0): cluster 3\n", + "Example n. 2344 = (52.0, 407.0): cluster 3\n", + "Example n. 2345 = (52.0, 406.0): cluster 3\n", + "Example n. 2346 = (52.0, 405.0): cluster 3\n", + "Example n. 2347 = (52.0, 404.0): cluster 3\n", + "Example n. 2348 = (52.0, 403.0): cluster 3\n", + "Example n. 2349 = (52.0, 402.0): cluster 3\n", + "Example n. 2350 = (52.0, 401.0): cluster 3\n", + "Example n. 2351 = (52.0, 400.0): cluster 3\n", + "Example n. 2352 = (52.0, 399.0): cluster 3\n", + "Example n. 2353 = (52.0, 398.0): cluster 3\n", + "Example n. 2354 = (53.0, 502.0): cluster 0\n", + "Example n. 2355 = (53.0, 501.0): cluster 0\n", + "Example n. 2356 = (53.0, 500.0): cluster 0\n", + "Example n. 2357 = (53.0, 499.0): cluster 0\n", + "Example n. 2358 = (53.0, 498.0): cluster 0\n", + "Example n. 2359 = (53.0, 497.0): cluster 0\n", + "Example n. 2360 = (53.0, 496.0): cluster 0\n", + "Example n. 2361 = (53.0, 495.0): cluster 1\n", + "Example n. 2362 = (53.0, 494.0): cluster 1\n", + "Example n. 2363 = (53.0, 493.0): cluster 1\n", + "Example n. 2364 = (53.0, 492.0): cluster 1\n", + "Example n. 2365 = (53.0, 491.0): cluster 1\n", + "Example n. 2366 = (53.0, 490.0): cluster 1\n", + "Example n. 2367 = (53.0, 489.0): cluster 1\n", + "Example n. 2368 = (53.0, 488.0): cluster 1\n", + "Example n. 2369 = (53.0, 487.0): cluster 1\n", + "Example n. 2370 = (53.0, 486.0): cluster 1\n", + "Example n. 2371 = (53.0, 485.0): cluster 1\n", + "Example n. 2372 = (53.0, 484.0): cluster 1\n", + "Example n. 2373 = (53.0, 483.0): cluster 1\n", + "Example n. 2374 = (53.0, 481.0): cluster 1\n", + "Example n. 2375 = (53.0, 459.0): cluster 1\n", + "Example n. 2376 = (53.0, 458.0): cluster 1\n", + "Example n. 2377 = (53.0, 457.0): cluster 1\n", + "Example n. 2378 = (53.0, 456.0): cluster 1\n", + "Example n. 2379 = (53.0, 455.0): cluster 1\n", + "Example n. 2380 = (53.0, 454.0): cluster 1\n", + "Example n. 2381 = (53.0, 453.0): cluster 1\n", + "Example n. 2382 = (53.0, 452.0): cluster 1\n", + "Example n. 2383 = (53.0, 440.0): cluster 3\n", + "Example n. 2384 = (53.0, 439.0): cluster 3\n", + "Example n. 2385 = (53.0, 438.0): cluster 3\n", + "Example n. 2386 = (53.0, 437.0): cluster 3\n", + "Example n. 2387 = (53.0, 436.0): cluster 3\n", + "Example n. 2388 = (53.0, 435.0): cluster 3\n", + "Example n. 2389 = (53.0, 434.0): cluster 3\n", + "Example n. 2390 = (53.0, 433.0): cluster 3\n", + "Example n. 2391 = (53.0, 432.0): cluster 3\n", + "Example n. 2392 = (53.0, 431.0): cluster 3\n", + "Example n. 2393 = (53.0, 430.0): cluster 3\n", + "Example n. 2394 = (53.0, 429.0): cluster 3\n", + "Example n. 2395 = (53.0, 428.0): cluster 3\n", + "Example n. 2396 = (53.0, 427.0): cluster 3\n", + "Example n. 2397 = (53.0, 426.0): cluster 3\n", + "Example n. 2398 = (53.0, 425.0): cluster 3\n", + "Example n. 2399 = (53.0, 424.0): cluster 3\n", + "Example n. 2400 = (53.0, 423.0): cluster 3\n", + "Example n. 2401 = (53.0, 422.0): cluster 3\n", + "Example n. 2402 = (53.0, 421.0): cluster 3\n", + "Example n. 2403 = (53.0, 409.0): cluster 3\n", + "Example n. 2404 = (53.0, 408.0): cluster 3\n", + "Example n. 2405 = (53.0, 407.0): cluster 3\n", + "Example n. 2406 = (53.0, 406.0): cluster 3\n", + "Example n. 2407 = (53.0, 405.0): cluster 3\n", + "Example n. 2408 = (53.0, 404.0): cluster 3\n", + "Example n. 2409 = (53.0, 403.0): cluster 3\n", + "Example n. 2410 = (53.0, 402.0): cluster 3\n", + "Example n. 2411 = (53.0, 401.0): cluster 3\n", + "Example n. 2412 = (53.0, 400.0): cluster 3\n", + "Example n. 2413 = (53.0, 399.0): cluster 3\n", + "Example n. 2414 = (53.0, 398.0): cluster 3\n", + "Example n. 2415 = (54.0, 502.0): cluster 0\n", + "Example n. 2416 = (54.0, 501.0): cluster 0\n", + "Example n. 2417 = (54.0, 500.0): cluster 0\n", + "Example n. 2418 = (54.0, 499.0): cluster 0\n", + "Example n. 2419 = (54.0, 498.0): cluster 1\n", + "Example n. 2420 = (54.0, 497.0): cluster 1\n", + "Example n. 2421 = (54.0, 496.0): cluster 1\n", + "Example n. 2422 = (54.0, 495.0): cluster 1\n", + "Example n. 2423 = (54.0, 494.0): cluster 1\n", + "Example n. 2424 = (54.0, 493.0): cluster 1\n", + "Example n. 2425 = (54.0, 492.0): cluster 1\n", + "Example n. 2426 = (54.0, 491.0): cluster 1\n", + "Example n. 2427 = (54.0, 490.0): cluster 1\n", + "Example n. 2428 = (54.0, 489.0): cluster 1\n", + "Example n. 2429 = (54.0, 488.0): cluster 1\n", + "Example n. 2430 = (54.0, 487.0): cluster 1\n", + "Example n. 2431 = (54.0, 486.0): cluster 1\n", + "Example n. 2432 = (54.0, 485.0): cluster 1\n", + "Example n. 2433 = (54.0, 484.0): cluster 1\n", + "Example n. 2434 = (54.0, 483.0): cluster 1\n", + "Example n. 2435 = (54.0, 482.0): cluster 1\n", + "Example n. 2436 = (54.0, 481.0): cluster 1\n", + "Example n. 2437 = (54.0, 459.0): cluster 1\n", + "Example n. 2438 = (54.0, 458.0): cluster 1\n", + "Example n. 2439 = (54.0, 457.0): cluster 1\n", + "Example n. 2440 = (54.0, 456.0): cluster 1\n", + "Example n. 2441 = (54.0, 455.0): cluster 1\n", + "Example n. 2442 = (54.0, 454.0): cluster 1\n", + "Example n. 2443 = (54.0, 453.0): cluster 1\n", + "Example n. 2444 = (54.0, 452.0): cluster 1\n", + "Example n. 2445 = (54.0, 440.0): cluster 3\n", + "Example n. 2446 = (54.0, 439.0): cluster 3\n", + "Example n. 2447 = (54.0, 438.0): cluster 3\n", + "Example n. 2448 = (54.0, 437.0): cluster 3\n", + "Example n. 2449 = (54.0, 436.0): cluster 3\n", + "Example n. 2450 = (54.0, 435.0): cluster 3\n", + "Example n. 2451 = (54.0, 434.0): cluster 3\n", + "Example n. 2452 = (54.0, 433.0): cluster 3\n", + "Example n. 2453 = (54.0, 432.0): cluster 3\n", + "Example n. 2454 = (54.0, 431.0): cluster 3\n", + "Example n. 2455 = (54.0, 430.0): cluster 3\n", + "Example n. 2456 = (54.0, 429.0): cluster 3\n", + "Example n. 2457 = (54.0, 428.0): cluster 3\n", + "Example n. 2458 = (54.0, 427.0): cluster 3\n", + "Example n. 2459 = (54.0, 426.0): cluster 3\n", + "Example n. 2460 = (54.0, 425.0): cluster 3\n", + "Example n. 2461 = (54.0, 424.0): cluster 3\n", + "Example n. 2462 = (54.0, 423.0): cluster 3\n", + "Example n. 2463 = (54.0, 422.0): cluster 3\n", + "Example n. 2464 = (54.0, 421.0): cluster 3\n", + "Example n. 2465 = (54.0, 409.0): cluster 3\n", + "Example n. 2466 = (54.0, 408.0): cluster 3\n", + "Example n. 2467 = (54.0, 407.0): cluster 3\n", + "Example n. 2468 = (54.0, 406.0): cluster 3\n", + "Example n. 2469 = (54.0, 405.0): cluster 3\n", + "Example n. 2470 = (54.0, 404.0): cluster 3\n", + "Example n. 2471 = (54.0, 403.0): cluster 3\n", + "Example n. 2472 = (54.0, 402.0): cluster 3\n", + "Example n. 2473 = (54.0, 401.0): cluster 3\n", + "Example n. 2474 = (54.0, 400.0): cluster 3\n", + "Example n. 2475 = (54.0, 399.0): cluster 3\n", + "Example n. 2476 = (54.0, 398.0): cluster 3\n", + "Example n. 2477 = (55.0, 498.0): cluster 1\n", + "Example n. 2478 = (55.0, 497.0): cluster 1\n", + "Example n. 2479 = (55.0, 496.0): cluster 1\n", + "Example n. 2480 = (55.0, 495.0): cluster 1\n", + "Example n. 2481 = (55.0, 494.0): cluster 1\n", + "Example n. 2482 = (55.0, 493.0): cluster 1\n", + "Example n. 2483 = (55.0, 492.0): cluster 1\n", + "Example n. 2484 = (55.0, 491.0): cluster 1\n", + "Example n. 2485 = (55.0, 490.0): cluster 1\n", + "Example n. 2486 = (55.0, 489.0): cluster 1\n", + "Example n. 2487 = (55.0, 488.0): cluster 1\n", + "Example n. 2488 = (55.0, 487.0): cluster 1\n", + "Example n. 2489 = (55.0, 486.0): cluster 1\n", + "Example n. 2490 = (55.0, 485.0): cluster 1\n", + "Example n. 2491 = (55.0, 484.0): cluster 1\n", + "Example n. 2492 = (55.0, 483.0): cluster 1\n", + "Example n. 2493 = (55.0, 482.0): cluster 1\n", + "Example n. 2494 = (55.0, 481.0): cluster 1\n", + "Example n. 2495 = (55.0, 480.0): cluster 1\n", + "Example n. 2496 = (55.0, 479.0): cluster 1\n", + "Example n. 2497 = (55.0, 478.0): cluster 1\n", + "Example n. 2498 = (55.0, 477.0): cluster 1\n", + "Example n. 2499 = (55.0, 476.0): cluster 1\n", + "Example n. 2500 = (55.0, 475.0): cluster 1\n", + "Example n. 2501 = (55.0, 440.0): cluster 3\n", + "Example n. 2502 = (55.0, 439.0): cluster 3\n", + "Example n. 2503 = (55.0, 438.0): cluster 3\n", + "Example n. 2504 = (55.0, 437.0): cluster 3\n", + "Example n. 2505 = (55.0, 436.0): cluster 3\n", + "Example n. 2506 = (55.0, 435.0): cluster 3\n", + "Example n. 2507 = (55.0, 434.0): cluster 3\n", + "Example n. 2508 = (55.0, 433.0): cluster 3\n", + "Example n. 2509 = (55.0, 432.0): cluster 3\n", + "Example n. 2510 = (55.0, 431.0): cluster 3\n", + "Example n. 2511 = (55.0, 430.0): cluster 3\n", + "Example n. 2512 = (55.0, 429.0): cluster 3\n", + "Example n. 2513 = (55.0, 428.0): cluster 3\n", + "Example n. 2514 = (55.0, 427.0): cluster 3\n", + "Example n. 2515 = (55.0, 426.0): cluster 3\n", + "Example n. 2516 = (55.0, 425.0): cluster 3\n", + "Example n. 2517 = (55.0, 405.0): cluster 3\n", + "Example n. 2518 = (55.0, 404.0): cluster 3\n", + "Example n. 2519 = (55.0, 403.0): cluster 3\n", + "Example n. 2520 = (55.0, 402.0): cluster 3\n", + "Example n. 2521 = (55.0, 401.0): cluster 3\n", + "Example n. 2522 = (55.0, 400.0): cluster 3\n", + "Example n. 2523 = (55.0, 399.0): cluster 3\n", + "Example n. 2524 = (55.0, 398.0): cluster 3\n", + "Example n. 2525 = (55.0, 397.0): cluster 3\n", + "Example n. 2526 = (55.0, 396.0): cluster 3\n", + "Example n. 2527 = (55.0, 395.0): cluster 3\n", + "Example n. 2528 = (55.0, 394.0): cluster 3\n", + "Example n. 2529 = (56.0, 498.0): cluster 1\n", + "Example n. 2530 = (56.0, 497.0): cluster 1\n", + "Example n. 2531 = (56.0, 496.0): cluster 1\n", + "Example n. 2532 = (56.0, 495.0): cluster 1\n", + "Example n. 2533 = (56.0, 494.0): cluster 1\n", + "Example n. 2534 = (56.0, 493.0): cluster 1\n", + "Example n. 2535 = (56.0, 492.0): cluster 1\n", + "Example n. 2536 = (56.0, 491.0): cluster 1\n", + "Example n. 2537 = (56.0, 490.0): cluster 1\n", + "Example n. 2538 = (56.0, 489.0): cluster 1\n", + "Example n. 2539 = (56.0, 488.0): cluster 1\n", + "Example n. 2540 = (56.0, 487.0): cluster 1\n", + "Example n. 2541 = (56.0, 486.0): cluster 1\n", + "Example n. 2542 = (56.0, 485.0): cluster 1\n", + "Example n. 2543 = (56.0, 484.0): cluster 1\n", + "Example n. 2544 = (56.0, 483.0): cluster 1\n", + "Example n. 2545 = (56.0, 482.0): cluster 1\n", + "Example n. 2546 = (56.0, 481.0): cluster 1\n", + "Example n. 2547 = (56.0, 480.0): cluster 1\n", + "Example n. 2548 = (56.0, 479.0): cluster 1\n", + "Example n. 2549 = (56.0, 478.0): cluster 1\n", + "Example n. 2550 = (56.0, 477.0): cluster 1\n", + "Example n. 2551 = (56.0, 476.0): cluster 1\n", + "Example n. 2552 = (56.0, 475.0): cluster 1\n", + "Example n. 2553 = (56.0, 440.0): cluster 3\n", + "Example n. 2554 = (56.0, 439.0): cluster 3\n", + "Example n. 2555 = (56.0, 438.0): cluster 3\n", + "Example n. 2556 = (56.0, 437.0): cluster 3\n", + "Example n. 2557 = (56.0, 436.0): cluster 3\n", + "Example n. 2558 = (56.0, 435.0): cluster 3\n", + "Example n. 2559 = (56.0, 434.0): cluster 3\n", + "Example n. 2560 = (56.0, 433.0): cluster 3\n", + "Example n. 2561 = (56.0, 432.0): cluster 3\n", + "Example n. 2562 = (56.0, 431.0): cluster 3\n", + "Example n. 2563 = (56.0, 430.0): cluster 3\n", + "Example n. 2564 = (56.0, 429.0): cluster 3\n", + "Example n. 2565 = (56.0, 428.0): cluster 3\n", + "Example n. 2566 = (56.0, 427.0): cluster 3\n", + "Example n. 2567 = (56.0, 426.0): cluster 3\n", + "Example n. 2568 = (56.0, 425.0): cluster 3\n", + "Example n. 2569 = (56.0, 405.0): cluster 3\n", + "Example n. 2570 = (56.0, 404.0): cluster 3\n", + "Example n. 2571 = (56.0, 403.0): cluster 3\n", + "Example n. 2572 = (56.0, 402.0): cluster 3\n", + "Example n. 2573 = (56.0, 401.0): cluster 3\n", + "Example n. 2574 = (56.0, 400.0): cluster 3\n", + "Example n. 2575 = (56.0, 399.0): cluster 3\n", + "Example n. 2576 = (56.0, 398.0): cluster 3\n", + "Example n. 2577 = (56.0, 397.0): cluster 3\n", + "Example n. 2578 = (56.0, 396.0): cluster 3\n", + "Example n. 2579 = (56.0, 395.0): cluster 3\n", + "Example n. 2580 = (56.0, 394.0): cluster 3\n", + "Example n. 2581 = (57.0, 498.0): cluster 1\n", + "Example n. 2582 = (57.0, 497.0): cluster 1\n", + "Example n. 2583 = (57.0, 496.0): cluster 1\n", + "Example n. 2584 = (57.0, 495.0): cluster 1\n", + "Example n. 2585 = (57.0, 494.0): cluster 1\n", + "Example n. 2586 = (57.0, 493.0): cluster 1\n", + "Example n. 2587 = (57.0, 492.0): cluster 1\n", + "Example n. 2588 = (57.0, 491.0): cluster 1\n", + "Example n. 2589 = (57.0, 490.0): cluster 1\n", + "Example n. 2590 = (57.0, 489.0): cluster 1\n", + "Example n. 2591 = (57.0, 488.0): cluster 1\n", + "Example n. 2592 = (57.0, 487.0): cluster 1\n", + "Example n. 2593 = (57.0, 486.0): cluster 1\n", + "Example n. 2594 = (57.0, 485.0): cluster 1\n", + "Example n. 2595 = (57.0, 484.0): cluster 1\n", + "Example n. 2596 = (57.0, 483.0): cluster 1\n", + "Example n. 2597 = (57.0, 482.0): cluster 1\n", + "Example n. 2598 = (57.0, 481.0): cluster 1\n", + "Example n. 2599 = (57.0, 480.0): cluster 1\n", + "Example n. 2600 = (57.0, 479.0): cluster 1\n", + "Example n. 2601 = (57.0, 478.0): cluster 1\n", + "Example n. 2602 = (57.0, 477.0): cluster 1\n", + "Example n. 2603 = (57.0, 476.0): cluster 1\n", + "Example n. 2604 = (57.0, 475.0): cluster 1\n", + "Example n. 2605 = (57.0, 440.0): cluster 3\n", + "Example n. 2606 = (57.0, 439.0): cluster 3\n", + "Example n. 2607 = (57.0, 438.0): cluster 3\n", + "Example n. 2608 = (57.0, 437.0): cluster 3\n", + "Example n. 2609 = (57.0, 436.0): cluster 3\n", + "Example n. 2610 = (57.0, 435.0): cluster 3\n", + "Example n. 2611 = (57.0, 434.0): cluster 3\n", + "Example n. 2612 = (57.0, 433.0): cluster 3\n", + "Example n. 2613 = (57.0, 432.0): cluster 3\n", + "Example n. 2614 = (57.0, 431.0): cluster 3\n", + "Example n. 2615 = (57.0, 430.0): cluster 3\n", + "Example n. 2616 = (57.0, 429.0): cluster 3\n", + "Example n. 2617 = (57.0, 428.0): cluster 3\n", + "Example n. 2618 = (57.0, 427.0): cluster 3\n", + "Example n. 2619 = (57.0, 426.0): cluster 3\n", + "Example n. 2620 = (57.0, 425.0): cluster 3\n", + "Example n. 2621 = (57.0, 405.0): cluster 3\n", + "Example n. 2622 = (57.0, 404.0): cluster 3\n", + "Example n. 2623 = (57.0, 403.0): cluster 3\n", + "Example n. 2624 = (57.0, 402.0): cluster 3\n", + "Example n. 2625 = (57.0, 401.0): cluster 3\n", + "Example n. 2626 = (57.0, 400.0): cluster 3\n", + "Example n. 2627 = (57.0, 399.0): cluster 3\n", + "Example n. 2628 = (57.0, 398.0): cluster 3\n", + "Example n. 2629 = (57.0, 397.0): cluster 3\n", + "Example n. 2630 = (57.0, 396.0): cluster 3\n", + "Example n. 2631 = (57.0, 395.0): cluster 3\n", + "Example n. 2632 = (57.0, 394.0): cluster 3\n", + "Example n. 2633 = (58.0, 498.0): cluster 1\n", + "Example n. 2634 = (58.0, 497.0): cluster 1\n", + "Example n. 2635 = (58.0, 496.0): cluster 1\n", + "Example n. 2636 = (58.0, 495.0): cluster 1\n", + "Example n. 2637 = (58.0, 494.0): cluster 1\n", + "Example n. 2638 = (58.0, 493.0): cluster 1\n", + "Example n. 2639 = (58.0, 492.0): cluster 1\n", + "Example n. 2640 = (58.0, 491.0): cluster 1\n", + "Example n. 2641 = (58.0, 490.0): cluster 1\n", + "Example n. 2642 = (58.0, 489.0): cluster 1\n", + "Example n. 2643 = (58.0, 488.0): cluster 1\n", + "Example n. 2644 = (58.0, 487.0): cluster 1\n", + "Example n. 2645 = (58.0, 486.0): cluster 1\n", + "Example n. 2646 = (58.0, 485.0): cluster 1\n", + "Example n. 2647 = (58.0, 484.0): cluster 1\n", + "Example n. 2648 = (58.0, 483.0): cluster 1\n", + "Example n. 2649 = (58.0, 482.0): cluster 1\n", + "Example n. 2650 = (58.0, 481.0): cluster 1\n", + "Example n. 2651 = (58.0, 480.0): cluster 1\n", + "Example n. 2652 = (58.0, 479.0): cluster 1\n", + "Example n. 2653 = (58.0, 478.0): cluster 1\n", + "Example n. 2654 = (58.0, 477.0): cluster 1\n", + "Example n. 2655 = (58.0, 476.0): cluster 1\n", + "Example n. 2656 = (58.0, 475.0): cluster 1\n", + "Example n. 2657 = (58.0, 440.0): cluster 3\n", + "Example n. 2658 = (58.0, 439.0): cluster 3\n", + "Example n. 2659 = (58.0, 438.0): cluster 3\n", + "Example n. 2660 = (58.0, 437.0): cluster 3\n", + "Example n. 2661 = (58.0, 436.0): cluster 3\n", + "Example n. 2662 = (58.0, 435.0): cluster 3\n", + "Example n. 2663 = (58.0, 434.0): cluster 3\n", + "Example n. 2664 = (58.0, 433.0): cluster 3\n", + "Example n. 2665 = (58.0, 432.0): cluster 3\n", + "Example n. 2666 = (58.0, 431.0): cluster 3\n", + "Example n. 2667 = (58.0, 430.0): cluster 3\n", + "Example n. 2668 = (58.0, 429.0): cluster 3\n", + "Example n. 2669 = (58.0, 428.0): cluster 3\n", + "Example n. 2670 = (58.0, 427.0): cluster 3\n", + "Example n. 2671 = (58.0, 426.0): cluster 3\n", + "Example n. 2672 = (58.0, 425.0): cluster 3\n", + "Example n. 2673 = (58.0, 405.0): cluster 3\n", + "Example n. 2674 = (58.0, 404.0): cluster 3\n", + "Example n. 2675 = (58.0, 403.0): cluster 3\n", + "Example n. 2676 = (58.0, 402.0): cluster 3\n", + "Example n. 2677 = (58.0, 401.0): cluster 3\n", + "Example n. 2678 = (58.0, 400.0): cluster 3\n", + "Example n. 2679 = (58.0, 399.0): cluster 3\n", + "Example n. 2680 = (58.0, 398.0): cluster 3\n", + "Example n. 2681 = (58.0, 397.0): cluster 3\n", + "Example n. 2682 = (58.0, 396.0): cluster 3\n", + "Example n. 2683 = (58.0, 395.0): cluster 3\n", + "Example n. 2684 = (58.0, 394.0): cluster 3\n", + "Example n. 2685 = (59.0, 494.0): cluster 1\n", + "Example n. 2686 = (59.0, 493.0): cluster 1\n", + "Example n. 2687 = (59.0, 492.0): cluster 1\n", + "Example n. 2688 = (59.0, 491.0): cluster 1\n", + "Example n. 2689 = (59.0, 490.0): cluster 1\n", + "Example n. 2690 = (59.0, 489.0): cluster 1\n", + "Example n. 2691 = (59.0, 488.0): cluster 1\n", + "Example n. 2692 = (59.0, 487.0): cluster 1\n", + "Example n. 2693 = (59.0, 486.0): cluster 1\n", + "Example n. 2694 = (59.0, 485.0): cluster 1\n", + "Example n. 2695 = (59.0, 484.0): cluster 1\n", + "Example n. 2696 = (59.0, 483.0): cluster 1\n", + "Example n. 2697 = (59.0, 482.0): cluster 1\n", + "Example n. 2698 = (59.0, 481.0): cluster 1\n", + "Example n. 2699 = (59.0, 480.0): cluster 1\n", + "Example n. 2700 = (59.0, 479.0): cluster 1\n", + "Example n. 2701 = (59.0, 478.0): cluster 1\n", + "Example n. 2702 = (59.0, 477.0): cluster 1\n", + "Example n. 2703 = (59.0, 476.0): cluster 1\n", + "Example n. 2704 = (59.0, 475.0): cluster 1\n", + "Example n. 2705 = (59.0, 474.0): cluster 1\n", + "Example n. 2706 = (59.0, 473.0): cluster 1\n", + "Example n. 2707 = (59.0, 472.0): cluster 1\n", + "Example n. 2708 = (59.0, 471.0): cluster 1\n", + "Example n. 2709 = (59.0, 470.0): cluster 1\n", + "Example n. 2710 = (59.0, 469.0): cluster 1\n", + "Example n. 2711 = (59.0, 468.0): cluster 1\n", + "Example n. 2712 = (59.0, 448.0): cluster 1\n", + "Example n. 2713 = (59.0, 447.0): cluster 1\n", + "Example n. 2714 = (59.0, 446.0): cluster 1\n", + "Example n. 2715 = (59.0, 445.0): cluster 1\n", + "Example n. 2716 = (59.0, 444.0): cluster 1\n", + "Example n. 2717 = (59.0, 443.0): cluster 1\n", + "Example n. 2718 = (59.0, 442.0): cluster 1\n", + "Example n. 2719 = (59.0, 441.0): cluster 3\n", + "Example n. 2720 = (59.0, 440.0): cluster 3\n", + "Example n. 2721 = (59.0, 439.0): cluster 3\n", + "Example n. 2722 = (59.0, 438.0): cluster 3\n", + "Example n. 2723 = (59.0, 437.0): cluster 3\n", + "Example n. 2724 = (59.0, 436.0): cluster 3\n", + "Example n. 2725 = (59.0, 435.0): cluster 3\n", + "Example n. 2726 = (59.0, 434.0): cluster 3\n", + "Example n. 2727 = (59.0, 433.0): cluster 3\n", + "Example n. 2728 = (59.0, 432.0): cluster 3\n", + "Example n. 2729 = (59.0, 431.0): cluster 3\n", + "Example n. 2730 = (59.0, 430.0): cluster 3\n", + "Example n. 2731 = (59.0, 429.0): cluster 3\n", + "Example n. 2732 = (59.0, 425.0): cluster 3\n", + "Example n. 2733 = (59.0, 424.0): cluster 3\n", + "Example n. 2734 = (59.0, 423.0): cluster 3\n", + "Example n. 2735 = (59.0, 422.0): cluster 3\n", + "Example n. 2736 = (59.0, 421.0): cluster 3\n", + "Example n. 2737 = (59.0, 420.0): cluster 3\n", + "Example n. 2738 = (59.0, 419.0): cluster 3\n", + "Example n. 2739 = (59.0, 418.0): cluster 3\n", + "Example n. 2740 = (59.0, 417.0): cluster 3\n", + "Example n. 2741 = (59.0, 409.0): cluster 3\n", + "Example n. 2742 = (59.0, 408.0): cluster 3\n", + "Example n. 2743 = (59.0, 407.0): cluster 3\n", + "Example n. 2744 = (59.0, 406.0): cluster 3\n", + "Example n. 2745 = (59.0, 405.0): cluster 3\n", + "Example n. 2746 = (59.0, 404.0): cluster 3\n", + "Example n. 2747 = (59.0, 403.0): cluster 3\n", + "Example n. 2748 = (59.0, 402.0): cluster 3\n", + "Example n. 2749 = (59.0, 401.0): cluster 3\n", + "Example n. 2750 = (59.0, 400.0): cluster 3\n", + "Example n. 2751 = (59.0, 399.0): cluster 3\n", + "Example n. 2752 = (59.0, 398.0): cluster 3\n", + "Example n. 2753 = (59.0, 397.0): cluster 3\n", + "Example n. 2754 = (59.0, 396.0): cluster 3\n", + "Example n. 2755 = (59.0, 395.0): cluster 3\n", + "Example n. 2756 = (59.0, 394.0): cluster 3\n", + "Example n. 2757 = (60.0, 494.0): cluster 1\n", + "Example n. 2758 = (60.0, 493.0): cluster 1\n", + "Example n. 2759 = (60.0, 492.0): cluster 1\n", + "Example n. 2760 = (60.0, 491.0): cluster 1\n", + "Example n. 2761 = (60.0, 490.0): cluster 1\n", + "Example n. 2762 = (60.0, 489.0): cluster 1\n", + "Example n. 2763 = (60.0, 488.0): cluster 1\n", + "Example n. 2764 = (60.0, 487.0): cluster 1\n", + "Example n. 2765 = (60.0, 486.0): cluster 1\n", + "Example n. 2766 = (60.0, 485.0): cluster 1\n", + "Example n. 2767 = (60.0, 484.0): cluster 1\n", + "Example n. 2768 = (60.0, 483.0): cluster 1\n", + "Example n. 2769 = (60.0, 482.0): cluster 1\n", + "Example n. 2770 = (60.0, 481.0): cluster 1\n", + "Example n. 2771 = (60.0, 480.0): cluster 1\n", + "Example n. 2772 = (60.0, 479.0): cluster 1\n", + "Example n. 2773 = (60.0, 478.0): cluster 1\n", + "Example n. 2774 = (60.0, 477.0): cluster 1\n", + "Example n. 2775 = (60.0, 476.0): cluster 1\n", + "Example n. 2776 = (60.0, 475.0): cluster 1\n", + "Example n. 2777 = (60.0, 474.0): cluster 1\n", + "Example n. 2778 = (60.0, 473.0): cluster 1\n", + "Example n. 2779 = (60.0, 472.0): cluster 1\n", + "Example n. 2780 = (60.0, 471.0): cluster 1\n", + "Example n. 2781 = (60.0, 470.0): cluster 1\n", + "Example n. 2782 = (60.0, 469.0): cluster 1\n", + "Example n. 2783 = (60.0, 468.0): cluster 1\n", + "Example n. 2784 = (60.0, 448.0): cluster 1\n", + "Example n. 2785 = (60.0, 447.0): cluster 1\n", + "Example n. 2786 = (60.0, 446.0): cluster 1\n", + "Example n. 2787 = (60.0, 445.0): cluster 1\n", + "Example n. 2788 = (60.0, 444.0): cluster 1\n", + "Example n. 2789 = (60.0, 443.0): cluster 1\n", + "Example n. 2790 = (60.0, 442.0): cluster 1\n", + "Example n. 2791 = (60.0, 441.0): cluster 3\n", + "Example n. 2792 = (60.0, 440.0): cluster 3\n", + "Example n. 2793 = (60.0, 439.0): cluster 3\n", + "Example n. 2794 = (60.0, 438.0): cluster 3\n", + "Example n. 2795 = (60.0, 437.0): cluster 3\n", + "Example n. 2796 = (60.0, 436.0): cluster 3\n", + "Example n. 2797 = (60.0, 435.0): cluster 3\n", + "Example n. 2798 = (60.0, 434.0): cluster 3\n", + "Example n. 2799 = (60.0, 433.0): cluster 3\n", + "Example n. 2800 = (60.0, 432.0): cluster 3\n", + "Example n. 2801 = (60.0, 431.0): cluster 3\n", + "Example n. 2802 = (60.0, 430.0): cluster 3\n", + "Example n. 2803 = (60.0, 429.0): cluster 3\n", + "Example n. 2804 = (60.0, 425.0): cluster 3\n", + "Example n. 2805 = (60.0, 424.0): cluster 3\n", + "Example n. 2806 = (60.0, 423.0): cluster 3\n", + "Example n. 2807 = (60.0, 422.0): cluster 3\n", + "Example n. 2808 = (60.0, 421.0): cluster 3\n", + "Example n. 2809 = (60.0, 420.0): cluster 3\n", + "Example n. 2810 = (60.0, 419.0): cluster 3\n", + "Example n. 2811 = (60.0, 418.0): cluster 3\n", + "Example n. 2812 = (60.0, 417.0): cluster 3\n", + "Example n. 2813 = (60.0, 409.0): cluster 3\n", + "Example n. 2814 = (60.0, 408.0): cluster 3\n", + "Example n. 2815 = (60.0, 407.0): cluster 3\n", + "Example n. 2816 = (60.0, 406.0): cluster 3\n", + "Example n. 2817 = (60.0, 405.0): cluster 3\n", + "Example n. 2818 = (60.0, 404.0): cluster 3\n", + "Example n. 2819 = (60.0, 403.0): cluster 3\n", + "Example n. 2820 = (60.0, 402.0): cluster 3\n", + "Example n. 2821 = (60.0, 401.0): cluster 3\n", + "Example n. 2822 = (60.0, 400.0): cluster 3\n", + "Example n. 2823 = (60.0, 399.0): cluster 3\n", + "Example n. 2824 = (60.0, 398.0): cluster 3\n", + "Example n. 2825 = (60.0, 397.0): cluster 3\n", + "Example n. 2826 = (60.0, 396.0): cluster 3\n", + "Example n. 2827 = (60.0, 395.0): cluster 3\n", + "Example n. 2828 = (60.0, 394.0): cluster 3\n", + "Example n. 2829 = (61.0, 494.0): cluster 1\n", + "Example n. 2830 = (61.0, 493.0): cluster 1\n", + "Example n. 2831 = (61.0, 492.0): cluster 1\n", + "Example n. 2832 = (61.0, 491.0): cluster 1\n", + "Example n. 2833 = (61.0, 490.0): cluster 1\n", + "Example n. 2834 = (61.0, 489.0): cluster 1\n", + "Example n. 2835 = (61.0, 488.0): cluster 1\n", + "Example n. 2836 = (61.0, 487.0): cluster 1\n", + "Example n. 2837 = (61.0, 486.0): cluster 1\n", + "Example n. 2838 = (61.0, 485.0): cluster 1\n", + "Example n. 2839 = (61.0, 484.0): cluster 1\n", + "Example n. 2840 = (61.0, 483.0): cluster 1\n", + "Example n. 2841 = (61.0, 482.0): cluster 1\n", + "Example n. 2842 = (61.0, 481.0): cluster 1\n", + "Example n. 2843 = (61.0, 480.0): cluster 1\n", + "Example n. 2844 = (61.0, 479.0): cluster 1\n", + "Example n. 2845 = (61.0, 478.0): cluster 1\n", + "Example n. 2846 = (61.0, 477.0): cluster 1\n", + "Example n. 2847 = (61.0, 476.0): cluster 1\n", + "Example n. 2848 = (61.0, 475.0): cluster 1\n", + "Example n. 2849 = (61.0, 474.0): cluster 1\n", + "Example n. 2850 = (61.0, 473.0): cluster 1\n", + "Example n. 2851 = (61.0, 472.0): cluster 1\n", + "Example n. 2852 = (61.0, 471.0): cluster 1\n", + "Example n. 2853 = (61.0, 470.0): cluster 1\n", + "Example n. 2854 = (61.0, 469.0): cluster 1\n", + "Example n. 2855 = (61.0, 468.0): cluster 1\n", + "Example n. 2856 = (61.0, 448.0): cluster 1\n", + "Example n. 2857 = (61.0, 447.0): cluster 1\n", + "Example n. 2858 = (61.0, 446.0): cluster 1\n", + "Example n. 2859 = (61.0, 445.0): cluster 1\n", + "Example n. 2860 = (61.0, 444.0): cluster 1\n", + "Example n. 2861 = (61.0, 443.0): cluster 1\n", + "Example n. 2862 = (61.0, 442.0): cluster 1\n", + "Example n. 2863 = (61.0, 441.0): cluster 3\n", + "Example n. 2864 = (61.0, 440.0): cluster 3\n", + "Example n. 2865 = (61.0, 439.0): cluster 3\n", + "Example n. 2866 = (61.0, 438.0): cluster 3\n", + "Example n. 2867 = (61.0, 437.0): cluster 3\n", + "Example n. 2868 = (61.0, 436.0): cluster 3\n", + "Example n. 2869 = (61.0, 435.0): cluster 3\n", + "Example n. 2870 = (61.0, 434.0): cluster 3\n", + "Example n. 2871 = (61.0, 433.0): cluster 3\n", + "Example n. 2872 = (61.0, 432.0): cluster 3\n", + "Example n. 2873 = (61.0, 431.0): cluster 3\n", + "Example n. 2874 = (61.0, 430.0): cluster 3\n", + "Example n. 2875 = (61.0, 429.0): cluster 3\n", + "Example n. 2876 = (61.0, 425.0): cluster 3\n", + "Example n. 2877 = (61.0, 424.0): cluster 3\n", + "Example n. 2878 = (61.0, 423.0): cluster 3\n", + "Example n. 2879 = (61.0, 422.0): cluster 3\n", + "Example n. 2880 = (61.0, 421.0): cluster 3\n", + "Example n. 2881 = (61.0, 420.0): cluster 3\n", + "Example n. 2882 = (61.0, 419.0): cluster 3\n", + "Example n. 2883 = (61.0, 418.0): cluster 3\n", + "Example n. 2884 = (61.0, 417.0): cluster 3\n", + "Example n. 2885 = (61.0, 409.0): cluster 3\n", + "Example n. 2886 = (61.0, 408.0): cluster 3\n", + "Example n. 2887 = (61.0, 407.0): cluster 3\n", + "Example n. 2888 = (61.0, 406.0): cluster 3\n", + "Example n. 2889 = (61.0, 405.0): cluster 3\n", + "Example n. 2890 = (61.0, 404.0): cluster 3\n", + "Example n. 2891 = (61.0, 403.0): cluster 3\n", + "Example n. 2892 = (61.0, 402.0): cluster 3\n", + "Example n. 2893 = (61.0, 401.0): cluster 3\n", + "Example n. 2894 = (61.0, 400.0): cluster 3\n", + "Example n. 2895 = (61.0, 399.0): cluster 3\n", + "Example n. 2896 = (61.0, 398.0): cluster 3\n", + "Example n. 2897 = (61.0, 397.0): cluster 3\n", + "Example n. 2898 = (61.0, 396.0): cluster 3\n", + "Example n. 2899 = (61.0, 395.0): cluster 3\n", + "Example n. 2900 = (61.0, 394.0): cluster 3\n", + "Example n. 2901 = (62.0, 494.0): cluster 1\n", + "Example n. 2902 = (62.0, 493.0): cluster 1\n", + "Example n. 2903 = (62.0, 492.0): cluster 1\n", + "Example n. 2904 = (62.0, 491.0): cluster 1\n", + "Example n. 2905 = (62.0, 490.0): cluster 1\n", + "Example n. 2906 = (62.0, 489.0): cluster 1\n", + "Example n. 2907 = (62.0, 488.0): cluster 1\n", + "Example n. 2908 = (62.0, 487.0): cluster 1\n", + "Example n. 2909 = (62.0, 486.0): cluster 1\n", + "Example n. 2910 = (62.0, 485.0): cluster 1\n", + "Example n. 2911 = (62.0, 484.0): cluster 1\n", + "Example n. 2912 = (62.0, 483.0): cluster 1\n", + "Example n. 2913 = (62.0, 482.0): cluster 1\n", + "Example n. 2914 = (62.0, 481.0): cluster 1\n", + "Example n. 2915 = (62.0, 480.0): cluster 1\n", + "Example n. 2916 = (62.0, 479.0): cluster 1\n", + "Example n. 2917 = (62.0, 478.0): cluster 1\n", + "Example n. 2918 = (62.0, 477.0): cluster 1\n", + "Example n. 2919 = (62.0, 476.0): cluster 1\n", + "Example n. 2920 = (62.0, 475.0): cluster 1\n", + "Example n. 2921 = (62.0, 474.0): cluster 1\n", + "Example n. 2922 = (62.0, 473.0): cluster 1\n", + "Example n. 2923 = (62.0, 472.0): cluster 1\n", + "Example n. 2924 = (62.0, 471.0): cluster 1\n", + "Example n. 2925 = (62.0, 470.0): cluster 1\n", + "Example n. 2926 = (62.0, 469.0): cluster 1\n", + "Example n. 2927 = (62.0, 468.0): cluster 1\n", + "Example n. 2928 = (62.0, 467.0): cluster 1\n", + "Example n. 2929 = (62.0, 466.0): cluster 1\n", + "Example n. 2930 = (62.0, 465.0): cluster 1\n", + "Example n. 2931 = (62.0, 464.0): cluster 1\n", + "Example n. 2932 = (62.0, 463.0): cluster 1\n", + "Example n. 2933 = (62.0, 462.0): cluster 1\n", + "Example n. 2934 = (62.0, 461.0): cluster 1\n", + "Example n. 2935 = (62.0, 460.0): cluster 1\n", + "Example n. 2936 = (62.0, 459.0): cluster 1\n", + "Example n. 2937 = (62.0, 458.0): cluster 1\n", + "Example n. 2938 = (62.0, 457.0): cluster 1\n", + "Example n. 2939 = (62.0, 456.0): cluster 1\n", + "Example n. 2940 = (62.0, 455.0): cluster 1\n", + "Example n. 2941 = (62.0, 454.0): cluster 1\n", + "Example n. 2942 = (62.0, 453.0): cluster 1\n", + "Example n. 2943 = (62.0, 452.0): cluster 1\n", + "Example n. 2944 = (62.0, 451.0): cluster 1\n", + "Example n. 2945 = (62.0, 450.0): cluster 1\n", + "Example n. 2946 = (62.0, 449.0): cluster 1\n", + "Example n. 2947 = (62.0, 448.0): cluster 1\n", + "Example n. 2948 = (62.0, 447.0): cluster 1\n", + "Example n. 2949 = (62.0, 446.0): cluster 1\n", + "Example n. 2950 = (62.0, 445.0): cluster 1\n", + "Example n. 2951 = (62.0, 444.0): cluster 1\n", + "Example n. 2952 = (62.0, 443.0): cluster 1\n", + "Example n. 2953 = (62.0, 442.0): cluster 1\n", + "Example n. 2954 = (62.0, 441.0): cluster 3\n", + "Example n. 2955 = (62.0, 440.0): cluster 3\n", + "Example n. 2956 = (62.0, 439.0): cluster 3\n", + "Example n. 2957 = (62.0, 438.0): cluster 3\n", + "Example n. 2958 = (62.0, 437.0): cluster 3\n", + "Example n. 2959 = (62.0, 436.0): cluster 3\n", + "Example n. 2960 = (62.0, 435.0): cluster 3\n", + "Example n. 2961 = (62.0, 434.0): cluster 3\n", + "Example n. 2962 = (62.0, 433.0): cluster 3\n", + "Example n. 2963 = (62.0, 432.0): cluster 3\n", + "Example n. 2964 = (62.0, 431.0): cluster 3\n", + "Example n. 2965 = (62.0, 430.0): cluster 3\n", + "Example n. 2966 = (62.0, 429.0): cluster 3\n", + "Example n. 2967 = (62.0, 425.0): cluster 3\n", + "Example n. 2968 = (62.0, 424.0): cluster 3\n", + "Example n. 2969 = (62.0, 423.0): cluster 3\n", + "Example n. 2970 = (62.0, 422.0): cluster 3\n", + "Example n. 2971 = (62.0, 421.0): cluster 3\n", + "Example n. 2972 = (62.0, 420.0): cluster 3\n", + "Example n. 2973 = (62.0, 419.0): cluster 3\n", + "Example n. 2974 = (62.0, 418.0): cluster 3\n", + "Example n. 2975 = (62.0, 417.0): cluster 3\n", + "Example n. 2976 = (62.0, 416.0): cluster 3\n", + "Example n. 2977 = (62.0, 415.0): cluster 3\n", + "Example n. 2978 = (62.0, 414.0): cluster 3\n", + "Example n. 2979 = (62.0, 409.0): cluster 3\n", + "Example n. 2980 = (62.0, 408.0): cluster 3\n", + "Example n. 2981 = (62.0, 407.0): cluster 3\n", + "Example n. 2982 = (62.0, 406.0): cluster 3\n", + "Example n. 2983 = (62.0, 405.0): cluster 3\n", + "Example n. 2984 = (62.0, 404.0): cluster 3\n", + "Example n. 2985 = (62.0, 403.0): cluster 3\n", + "Example n. 2986 = (62.0, 402.0): cluster 3\n", + "Example n. 2987 = (62.0, 401.0): cluster 3\n", + "Example n. 2988 = (62.0, 400.0): cluster 3\n", + "Example n. 2989 = (62.0, 399.0): cluster 3\n", + "Example n. 2990 = (62.0, 398.0): cluster 3\n", + "Example n. 2991 = (62.0, 397.0): cluster 3\n", + "Example n. 2992 = (62.0, 396.0): cluster 3\n", + "Example n. 2993 = (62.0, 395.0): cluster 3\n", + "Example n. 2994 = (62.0, 394.0): cluster 3\n", + "Example n. 2995 = (63.0, 494.0): cluster 1\n", + "Example n. 2996 = (63.0, 493.0): cluster 1\n", + "Example n. 2997 = (63.0, 492.0): cluster 1\n", + "Example n. 2998 = (63.0, 491.0): cluster 1\n", + "Example n. 2999 = (63.0, 490.0): cluster 1\n", + "Example n. 3000 = (63.0, 489.0): cluster 1\n", + "Example n. 3001 = (63.0, 488.0): cluster 1\n", + "Example n. 3002 = (63.0, 487.0): cluster 1\n", + "Example n. 3003 = (63.0, 486.0): cluster 1\n", + "Example n. 3004 = (63.0, 485.0): cluster 1\n", + "Example n. 3005 = (63.0, 484.0): cluster 1\n", + "Example n. 3006 = (63.0, 483.0): cluster 1\n", + "Example n. 3007 = (63.0, 482.0): cluster 1\n", + "Example n. 3008 = (63.0, 481.0): cluster 1\n", + "Example n. 3009 = (63.0, 480.0): cluster 1\n", + "Example n. 3010 = (63.0, 479.0): cluster 1\n", + "Example n. 3011 = (63.0, 478.0): cluster 1\n", + "Example n. 3012 = (63.0, 477.0): cluster 1\n", + "Example n. 3013 = (63.0, 476.0): cluster 1\n", + "Example n. 3014 = (63.0, 475.0): cluster 1\n", + "Example n. 3015 = (63.0, 474.0): cluster 1\n", + "Example n. 3016 = (63.0, 473.0): cluster 1\n", + "Example n. 3017 = (63.0, 472.0): cluster 1\n", + "Example n. 3018 = (63.0, 471.0): cluster 1\n", + "Example n. 3019 = (63.0, 470.0): cluster 1\n", + "Example n. 3020 = (63.0, 469.0): cluster 1\n", + "Example n. 3021 = (63.0, 468.0): cluster 1\n", + "Example n. 3022 = (63.0, 467.0): cluster 1\n", + "Example n. 3023 = (63.0, 466.0): cluster 1\n", + "Example n. 3024 = (63.0, 465.0): cluster 1\n", + "Example n. 3025 = (63.0, 464.0): cluster 1\n", + "Example n. 3026 = (63.0, 463.0): cluster 1\n", + "Example n. 3027 = (63.0, 462.0): cluster 1\n", + "Example n. 3028 = (63.0, 461.0): cluster 1\n", + "Example n. 3029 = (63.0, 460.0): cluster 1\n", + "Example n. 3030 = (63.0, 459.0): cluster 1\n", + "Example n. 3031 = (63.0, 458.0): cluster 1\n", + "Example n. 3032 = (63.0, 457.0): cluster 1\n", + "Example n. 3033 = (63.0, 456.0): cluster 1\n", + "Example n. 3034 = (63.0, 455.0): cluster 1\n", + "Example n. 3035 = (63.0, 454.0): cluster 1\n", + "Example n. 3036 = (63.0, 453.0): cluster 1\n", + "Example n. 3037 = (63.0, 452.0): cluster 1\n", + "Example n. 3038 = (63.0, 451.0): cluster 1\n", + "Example n. 3039 = (63.0, 450.0): cluster 1\n", + "Example n. 3040 = (63.0, 449.0): cluster 1\n", + "Example n. 3041 = (63.0, 448.0): cluster 1\n", + "Example n. 3042 = (63.0, 447.0): cluster 1\n", + "Example n. 3043 = (63.0, 446.0): cluster 1\n", + "Example n. 3044 = (63.0, 445.0): cluster 1\n", + "Example n. 3045 = (63.0, 444.0): cluster 1\n", + "Example n. 3046 = (63.0, 443.0): cluster 1\n", + "Example n. 3047 = (63.0, 442.0): cluster 1\n", + "Example n. 3048 = (63.0, 441.0): cluster 3\n", + "Example n. 3049 = (63.0, 440.0): cluster 3\n", + "Example n. 3050 = (63.0, 439.0): cluster 3\n", + "Example n. 3051 = (63.0, 438.0): cluster 3\n", + "Example n. 3052 = (63.0, 437.0): cluster 3\n", + "Example n. 3053 = (63.0, 436.0): cluster 3\n", + "Example n. 3054 = (63.0, 435.0): cluster 3\n", + "Example n. 3055 = (63.0, 434.0): cluster 3\n", + "Example n. 3056 = (63.0, 433.0): cluster 3\n", + "Example n. 3057 = (63.0, 425.0): cluster 3\n", + "Example n. 3058 = (63.0, 424.0): cluster 3\n", + "Example n. 3059 = (63.0, 423.0): cluster 3\n", + "Example n. 3060 = (63.0, 422.0): cluster 3\n", + "Example n. 3061 = (63.0, 421.0): cluster 3\n", + "Example n. 3062 = (63.0, 420.0): cluster 3\n", + "Example n. 3063 = (63.0, 419.0): cluster 3\n", + "Example n. 3064 = (63.0, 418.0): cluster 3\n", + "Example n. 3065 = (63.0, 417.0): cluster 3\n", + "Example n. 3066 = (63.0, 416.0): cluster 3\n", + "Example n. 3067 = (63.0, 415.0): cluster 3\n", + "Example n. 3068 = (63.0, 414.0): cluster 3\n", + "Example n. 3069 = (63.0, 409.0): cluster 3\n", + "Example n. 3070 = (63.0, 408.0): cluster 3\n", + "Example n. 3071 = (63.0, 407.0): cluster 3\n", + "Example n. 3072 = (63.0, 406.0): cluster 3\n", + "Example n. 3073 = (63.0, 405.0): cluster 3\n", + "Example n. 3074 = (63.0, 404.0): cluster 3\n", + "Example n. 3075 = (63.0, 403.0): cluster 3\n", + "Example n. 3076 = (63.0, 402.0): cluster 3\n", + "Example n. 3077 = (63.0, 401.0): cluster 3\n", + "Example n. 3078 = (63.0, 400.0): cluster 3\n", + "Example n. 3079 = (63.0, 399.0): cluster 3\n", + "Example n. 3080 = (63.0, 398.0): cluster 3\n", + "Example n. 3081 = (63.0, 397.0): cluster 3\n", + "Example n. 3082 = (63.0, 396.0): cluster 3\n", + "Example n. 3083 = (63.0, 395.0): cluster 3\n", + "Example n. 3084 = (63.0, 394.0): cluster 3\n", + "Example n. 3085 = (64.0, 494.0): cluster 1\n", + "Example n. 3086 = (64.0, 493.0): cluster 1\n", + "Example n. 3087 = (64.0, 492.0): cluster 1\n", + "Example n. 3088 = (64.0, 491.0): cluster 1\n", + "Example n. 3089 = (64.0, 490.0): cluster 1\n", + "Example n. 3090 = (64.0, 489.0): cluster 1\n", + "Example n. 3091 = (64.0, 488.0): cluster 1\n", + "Example n. 3092 = (64.0, 487.0): cluster 1\n", + "Example n. 3093 = (64.0, 486.0): cluster 1\n", + "Example n. 3094 = (64.0, 485.0): cluster 1\n", + "Example n. 3095 = (64.0, 484.0): cluster 1\n", + "Example n. 3096 = (64.0, 483.0): cluster 1\n", + "Example n. 3097 = (64.0, 482.0): cluster 1\n", + "Example n. 3098 = (64.0, 481.0): cluster 1\n", + "Example n. 3099 = (64.0, 480.0): cluster 1\n", + "Example n. 3100 = (64.0, 479.0): cluster 1\n", + "Example n. 3101 = (64.0, 478.0): cluster 1\n", + "Example n. 3102 = (64.0, 477.0): cluster 1\n", + "Example n. 3103 = (64.0, 476.0): cluster 1\n", + "Example n. 3104 = (64.0, 475.0): cluster 1\n", + "Example n. 3105 = (64.0, 474.0): cluster 1\n", + "Example n. 3106 = (64.0, 473.0): cluster 1\n", + "Example n. 3107 = (64.0, 472.0): cluster 1\n", + "Example n. 3108 = (64.0, 471.0): cluster 1\n", + "Example n. 3109 = (64.0, 470.0): cluster 1\n", + "Example n. 3110 = (64.0, 469.0): cluster 1\n", + "Example n. 3111 = (64.0, 468.0): cluster 1\n", + "Example n. 3112 = (64.0, 467.0): cluster 1\n", + "Example n. 3113 = (64.0, 466.0): cluster 1\n", + "Example n. 3114 = (64.0, 465.0): cluster 1\n", + "Example n. 3115 = (64.0, 464.0): cluster 1\n", + "Example n. 3116 = (64.0, 463.0): cluster 1\n", + "Example n. 3117 = (64.0, 462.0): cluster 1\n", + "Example n. 3118 = (64.0, 461.0): cluster 1\n", + "Example n. 3119 = (64.0, 460.0): cluster 1\n", + "Example n. 3120 = (64.0, 459.0): cluster 1\n", + "Example n. 3121 = (64.0, 458.0): cluster 1\n", + "Example n. 3122 = (64.0, 457.0): cluster 1\n", + "Example n. 3123 = (64.0, 456.0): cluster 1\n", + "Example n. 3124 = (64.0, 455.0): cluster 1\n", + "Example n. 3125 = (64.0, 454.0): cluster 1\n", + "Example n. 3126 = (64.0, 453.0): cluster 1\n", + "Example n. 3127 = (64.0, 452.0): cluster 1\n", + "Example n. 3128 = (64.0, 451.0): cluster 1\n", + "Example n. 3129 = (64.0, 450.0): cluster 1\n", + "Example n. 3130 = (64.0, 449.0): cluster 1\n", + "Example n. 3131 = (64.0, 448.0): cluster 1\n", + "Example n. 3132 = (64.0, 447.0): cluster 1\n", + "Example n. 3133 = (64.0, 446.0): cluster 1\n", + "Example n. 3134 = (64.0, 445.0): cluster 1\n", + "Example n. 3135 = (64.0, 444.0): cluster 1\n", + "Example n. 3136 = (64.0, 443.0): cluster 1\n", + "Example n. 3137 = (64.0, 442.0): cluster 1\n", + "Example n. 3138 = (64.0, 441.0): cluster 3\n", + "Example n. 3139 = (64.0, 440.0): cluster 3\n", + "Example n. 3140 = (64.0, 439.0): cluster 3\n", + "Example n. 3141 = (64.0, 438.0): cluster 3\n", + "Example n. 3142 = (64.0, 437.0): cluster 3\n", + "Example n. 3143 = (64.0, 436.0): cluster 3\n", + "Example n. 3144 = (64.0, 435.0): cluster 3\n", + "Example n. 3145 = (64.0, 434.0): cluster 3\n", + "Example n. 3146 = (64.0, 433.0): cluster 3\n", + "Example n. 3147 = (64.0, 425.0): cluster 3\n", + "Example n. 3148 = (64.0, 424.0): cluster 3\n", + "Example n. 3149 = (64.0, 423.0): cluster 3\n", + "Example n. 3150 = (64.0, 422.0): cluster 3\n", + "Example n. 3151 = (64.0, 421.0): cluster 3\n", + "Example n. 3152 = (64.0, 420.0): cluster 3\n", + "Example n. 3153 = (64.0, 419.0): cluster 3\n", + "Example n. 3154 = (64.0, 418.0): cluster 3\n", + "Example n. 3155 = (64.0, 417.0): cluster 3\n", + "Example n. 3156 = (64.0, 416.0): cluster 3\n", + "Example n. 3157 = (64.0, 415.0): cluster 3\n", + "Example n. 3158 = (64.0, 414.0): cluster 3\n", + "Example n. 3159 = (64.0, 409.0): cluster 3\n", + "Example n. 3160 = (64.0, 408.0): cluster 3\n", + "Example n. 3161 = (64.0, 407.0): cluster 3\n", + "Example n. 3162 = (64.0, 406.0): cluster 3\n", + "Example n. 3163 = (64.0, 405.0): cluster 3\n", + "Example n. 3164 = (64.0, 404.0): cluster 3\n", + "Example n. 3165 = (64.0, 403.0): cluster 3\n", + "Example n. 3166 = (64.0, 402.0): cluster 3\n", + "Example n. 3167 = (64.0, 401.0): cluster 3\n", + "Example n. 3168 = (64.0, 400.0): cluster 3\n", + "Example n. 3169 = (64.0, 399.0): cluster 3\n", + "Example n. 3170 = (64.0, 398.0): cluster 3\n", + "Example n. 3171 = (64.0, 397.0): cluster 3\n", + "Example n. 3172 = (64.0, 396.0): cluster 3\n", + "Example n. 3173 = (64.0, 395.0): cluster 3\n", + "Example n. 3174 = (64.0, 394.0): cluster 3\n", + "Example n. 3175 = (65.0, 494.0): cluster 1\n", + "Example n. 3176 = (65.0, 493.0): cluster 1\n", + "Example n. 3177 = (65.0, 492.0): cluster 1\n", + "Example n. 3178 = (65.0, 491.0): cluster 1\n", + "Example n. 3179 = (65.0, 490.0): cluster 1\n", + "Example n. 3180 = (65.0, 489.0): cluster 1\n", + "Example n. 3181 = (65.0, 488.0): cluster 1\n", + "Example n. 3182 = (65.0, 487.0): cluster 1\n", + "Example n. 3183 = (65.0, 486.0): cluster 1\n", + "Example n. 3184 = (65.0, 485.0): cluster 1\n", + "Example n. 3185 = (65.0, 484.0): cluster 1\n", + "Example n. 3186 = (65.0, 483.0): cluster 1\n", + "Example n. 3187 = (65.0, 482.0): cluster 1\n", + "Example n. 3188 = (65.0, 481.0): cluster 1\n", + "Example n. 3189 = (65.0, 480.0): cluster 1\n", + "Example n. 3190 = (65.0, 479.0): cluster 1\n", + "Example n. 3191 = (65.0, 478.0): cluster 1\n", + "Example n. 3192 = (65.0, 477.0): cluster 1\n", + "Example n. 3193 = (65.0, 476.0): cluster 1\n", + "Example n. 3194 = (65.0, 475.0): cluster 1\n", + "Example n. 3195 = (65.0, 474.0): cluster 1\n", + "Example n. 3196 = (65.0, 473.0): cluster 1\n", + "Example n. 3197 = (65.0, 472.0): cluster 1\n", + "Example n. 3198 = (65.0, 471.0): cluster 1\n", + "Example n. 3199 = (65.0, 470.0): cluster 1\n", + "Example n. 3200 = (65.0, 469.0): cluster 1\n", + "Example n. 3201 = (65.0, 468.0): cluster 1\n", + "Example n. 3202 = (65.0, 467.0): cluster 1\n", + "Example n. 3203 = (65.0, 466.0): cluster 1\n", + "Example n. 3204 = (65.0, 465.0): cluster 1\n", + "Example n. 3205 = (65.0, 464.0): cluster 1\n", + "Example n. 3206 = (65.0, 463.0): cluster 1\n", + "Example n. 3207 = (65.0, 462.0): cluster 1\n", + "Example n. 3208 = (65.0, 461.0): cluster 1\n", + "Example n. 3209 = (65.0, 460.0): cluster 1\n", + "Example n. 3210 = (65.0, 459.0): cluster 1\n", + "Example n. 3211 = (65.0, 458.0): cluster 1\n", + "Example n. 3212 = (65.0, 457.0): cluster 1\n", + "Example n. 3213 = (65.0, 456.0): cluster 1\n", + "Example n. 3214 = (65.0, 455.0): cluster 1\n", + "Example n. 3215 = (65.0, 454.0): cluster 1\n", + "Example n. 3216 = (65.0, 453.0): cluster 1\n", + "Example n. 3217 = (65.0, 452.0): cluster 1\n", + "Example n. 3218 = (65.0, 451.0): cluster 1\n", + "Example n. 3219 = (65.0, 450.0): cluster 1\n", + "Example n. 3220 = (65.0, 449.0): cluster 1\n", + "Example n. 3221 = (65.0, 448.0): cluster 1\n", + "Example n. 3222 = (65.0, 447.0): cluster 1\n", + "Example n. 3223 = (65.0, 446.0): cluster 1\n", + "Example n. 3224 = (65.0, 445.0): cluster 1\n", + "Example n. 3225 = (65.0, 444.0): cluster 1\n", + "Example n. 3226 = (65.0, 443.0): cluster 1\n", + "Example n. 3227 = (65.0, 442.0): cluster 1\n", + "Example n. 3228 = (65.0, 441.0): cluster 3\n", + "Example n. 3229 = (65.0, 440.0): cluster 3\n", + "Example n. 3230 = (65.0, 439.0): cluster 3\n", + "Example n. 3231 = (65.0, 438.0): cluster 3\n", + "Example n. 3232 = (65.0, 437.0): cluster 3\n", + "Example n. 3233 = (65.0, 436.0): cluster 3\n", + "Example n. 3234 = (65.0, 435.0): cluster 3\n", + "Example n. 3235 = (65.0, 434.0): cluster 3\n", + "Example n. 3236 = (65.0, 433.0): cluster 3\n", + "Example n. 3237 = (65.0, 425.0): cluster 3\n", + "Example n. 3238 = (65.0, 424.0): cluster 3\n", + "Example n. 3239 = (65.0, 423.0): cluster 3\n", + "Example n. 3240 = (65.0, 422.0): cluster 3\n", + "Example n. 3241 = (65.0, 421.0): cluster 3\n", + "Example n. 3242 = (65.0, 420.0): cluster 3\n", + "Example n. 3243 = (65.0, 419.0): cluster 3\n", + "Example n. 3244 = (65.0, 418.0): cluster 3\n", + "Example n. 3245 = (65.0, 417.0): cluster 3\n", + "Example n. 3246 = (65.0, 416.0): cluster 3\n", + "Example n. 3247 = (65.0, 415.0): cluster 3\n", + "Example n. 3248 = (65.0, 414.0): cluster 3\n", + "Example n. 3249 = (65.0, 409.0): cluster 3\n", + "Example n. 3250 = (65.0, 408.0): cluster 3\n", + "Example n. 3251 = (65.0, 407.0): cluster 3\n", + "Example n. 3252 = (65.0, 406.0): cluster 3\n", + "Example n. 3253 = (65.0, 405.0): cluster 3\n", + "Example n. 3254 = (65.0, 404.0): cluster 3\n", + "Example n. 3255 = (65.0, 403.0): cluster 3\n", + "Example n. 3256 = (65.0, 402.0): cluster 3\n", + "Example n. 3257 = (65.0, 401.0): cluster 3\n", + "Example n. 3258 = (65.0, 400.0): cluster 3\n", + "Example n. 3259 = (65.0, 399.0): cluster 3\n", + "Example n. 3260 = (65.0, 398.0): cluster 3\n", + "Example n. 3261 = (65.0, 397.0): cluster 3\n", + "Example n. 3262 = (65.0, 396.0): cluster 3\n", + "Example n. 3263 = (65.0, 395.0): cluster 3\n", + "Example n. 3264 = (65.0, 394.0): cluster 3\n", + "Example n. 3265 = (66.0, 494.0): cluster 1\n", + "Example n. 3266 = (66.0, 493.0): cluster 1\n", + "Example n. 3267 = (66.0, 492.0): cluster 1\n", + "Example n. 3268 = (66.0, 491.0): cluster 1\n", + "Example n. 3269 = (66.0, 490.0): cluster 1\n", + "Example n. 3270 = (66.0, 489.0): cluster 1\n", + "Example n. 3271 = (66.0, 488.0): cluster 1\n", + "Example n. 3272 = (66.0, 487.0): cluster 1\n", + "Example n. 3273 = (66.0, 486.0): cluster 1\n", + "Example n. 3274 = (66.0, 485.0): cluster 1\n", + "Example n. 3275 = (66.0, 484.0): cluster 1\n", + "Example n. 3276 = (66.0, 483.0): cluster 1\n", + "Example n. 3277 = (66.0, 482.0): cluster 1\n", + "Example n. 3278 = (66.0, 481.0): cluster 1\n", + "Example n. 3279 = (66.0, 480.0): cluster 1\n", + "Example n. 3280 = (66.0, 479.0): cluster 1\n", + "Example n. 3281 = (66.0, 478.0): cluster 1\n", + "Example n. 3282 = (66.0, 477.0): cluster 1\n", + "Example n. 3283 = (66.0, 476.0): cluster 1\n", + "Example n. 3284 = (66.0, 475.0): cluster 1\n", + "Example n. 3285 = (66.0, 474.0): cluster 1\n", + "Example n. 3286 = (66.0, 473.0): cluster 1\n", + "Example n. 3287 = (66.0, 472.0): cluster 1\n", + "Example n. 3288 = (66.0, 471.0): cluster 1\n", + "Example n. 3289 = (66.0, 470.0): cluster 1\n", + "Example n. 3290 = (66.0, 469.0): cluster 1\n", + "Example n. 3291 = (66.0, 468.0): cluster 1\n", + "Example n. 3292 = (66.0, 467.0): cluster 1\n", + "Example n. 3293 = (66.0, 466.0): cluster 1\n", + "Example n. 3294 = (66.0, 465.0): cluster 1\n", + "Example n. 3295 = (66.0, 464.0): cluster 1\n", + "Example n. 3296 = (66.0, 463.0): cluster 1\n", + "Example n. 3297 = (66.0, 462.0): cluster 1\n", + "Example n. 3298 = (66.0, 461.0): cluster 1\n", + "Example n. 3299 = (66.0, 460.0): cluster 1\n", + "Example n. 3300 = (66.0, 459.0): cluster 1\n", + "Example n. 3301 = (66.0, 458.0): cluster 1\n", + "Example n. 3302 = (66.0, 457.0): cluster 1\n", + "Example n. 3303 = (66.0, 456.0): cluster 1\n", + "Example n. 3304 = (66.0, 455.0): cluster 1\n", + "Example n. 3305 = (66.0, 454.0): cluster 1\n", + "Example n. 3306 = (66.0, 453.0): cluster 1\n", + "Example n. 3307 = (66.0, 452.0): cluster 1\n", + "Example n. 3308 = (66.0, 451.0): cluster 1\n", + "Example n. 3309 = (66.0, 450.0): cluster 1\n", + "Example n. 3310 = (66.0, 449.0): cluster 1\n", + "Example n. 3311 = (66.0, 448.0): cluster 1\n", + "Example n. 3312 = (66.0, 447.0): cluster 1\n", + "Example n. 3313 = (66.0, 446.0): cluster 1\n", + "Example n. 3314 = (66.0, 445.0): cluster 1\n", + "Example n. 3315 = (66.0, 444.0): cluster 1\n", + "Example n. 3316 = (66.0, 443.0): cluster 1\n", + "Example n. 3317 = (66.0, 442.0): cluster 1\n", + "Example n. 3318 = (66.0, 441.0): cluster 3\n", + "Example n. 3319 = (66.0, 440.0): cluster 3\n", + "Example n. 3320 = (66.0, 439.0): cluster 3\n", + "Example n. 3321 = (66.0, 438.0): cluster 3\n", + "Example n. 3322 = (66.0, 437.0): cluster 3\n", + "Example n. 3323 = (66.0, 436.0): cluster 3\n", + "Example n. 3324 = (66.0, 435.0): cluster 3\n", + "Example n. 3325 = (66.0, 434.0): cluster 3\n", + "Example n. 3326 = (66.0, 433.0): cluster 3\n", + "Example n. 3327 = (66.0, 429.0): cluster 3\n", + "Example n. 3328 = (66.0, 428.0): cluster 3\n", + "Example n. 3329 = (66.0, 427.0): cluster 3\n", + "Example n. 3330 = (66.0, 426.0): cluster 3\n", + "Example n. 3331 = (66.0, 425.0): cluster 3\n", + "Example n. 3332 = (66.0, 424.0): cluster 3\n", + "Example n. 3333 = (66.0, 423.0): cluster 3\n", + "Example n. 3334 = (66.0, 422.0): cluster 3\n", + "Example n. 3335 = (66.0, 421.0): cluster 3\n", + "Example n. 3336 = (66.0, 420.0): cluster 3\n", + "Example n. 3337 = (66.0, 419.0): cluster 3\n", + "Example n. 3338 = (66.0, 418.0): cluster 3\n", + "Example n. 3339 = (66.0, 417.0): cluster 3\n", + "Example n. 3340 = (66.0, 416.0): cluster 3\n", + "Example n. 3341 = (66.0, 415.0): cluster 3\n", + "Example n. 3342 = (66.0, 414.0): cluster 3\n", + "Example n. 3343 = (66.0, 409.0): cluster 3\n", + "Example n. 3344 = (66.0, 408.0): cluster 3\n", + "Example n. 3345 = (66.0, 407.0): cluster 3\n", + "Example n. 3346 = (66.0, 406.0): cluster 3\n", + "Example n. 3347 = (66.0, 405.0): cluster 3\n", + "Example n. 3348 = (66.0, 404.0): cluster 3\n", + "Example n. 3349 = (66.0, 403.0): cluster 3\n", + "Example n. 3350 = (66.0, 402.0): cluster 3\n", + "Example n. 3351 = (66.0, 401.0): cluster 3\n", + "Example n. 3352 = (66.0, 400.0): cluster 3\n", + "Example n. 3353 = (66.0, 399.0): cluster 3\n", + "Example n. 3354 = (66.0, 398.0): cluster 3\n", + "Example n. 3355 = (66.0, 397.0): cluster 3\n", + "Example n. 3356 = (66.0, 396.0): cluster 3\n", + "Example n. 3357 = (66.0, 395.0): cluster 3\n", + "Example n. 3358 = (66.0, 394.0): cluster 3\n", + "Example n. 3359 = (67.0, 486.0): cluster 1\n", + "Example n. 3360 = (67.0, 485.0): cluster 1\n", + "Example n. 3361 = (67.0, 484.0): cluster 1\n", + "Example n. 3362 = (67.0, 483.0): cluster 1\n", + "Example n. 3363 = (67.0, 482.0): cluster 1\n", + "Example n. 3364 = (67.0, 481.0): cluster 1\n", + "Example n. 3365 = (67.0, 480.0): cluster 1\n", + "Example n. 3366 = (67.0, 479.0): cluster 1\n", + "Example n. 3367 = (67.0, 478.0): cluster 1\n", + "Example n. 3368 = (67.0, 477.0): cluster 1\n", + "Example n. 3369 = (67.0, 476.0): cluster 1\n", + "Example n. 3370 = (67.0, 475.0): cluster 1\n", + "Example n. 3371 = (67.0, 474.0): cluster 1\n", + "Example n. 3372 = (67.0, 473.0): cluster 1\n", + "Example n. 3373 = (67.0, 472.0): cluster 1\n", + "Example n. 3374 = (67.0, 471.0): cluster 1\n", + "Example n. 3375 = (67.0, 470.0): cluster 1\n", + "Example n. 3376 = (67.0, 469.0): cluster 1\n", + "Example n. 3377 = (67.0, 468.0): cluster 1\n", + "Example n. 3378 = (67.0, 467.0): cluster 1\n", + "Example n. 3379 = (67.0, 466.0): cluster 1\n", + "Example n. 3380 = (67.0, 465.0): cluster 1\n", + "Example n. 3381 = (67.0, 464.0): cluster 1\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Example n. 3382 = (67.0, 463.0): cluster 1\n", + "Example n. 3383 = (67.0, 462.0): cluster 1\n", + "Example n. 3384 = (67.0, 461.0): cluster 1\n", + "Example n. 3385 = (67.0, 460.0): cluster 1\n", + "Example n. 3386 = (67.0, 459.0): cluster 1\n", + "Example n. 3387 = (67.0, 458.0): cluster 1\n", + "Example n. 3388 = (67.0, 457.0): cluster 1\n", + "Example n. 3389 = (67.0, 456.0): cluster 1\n", + "Example n. 3390 = (67.0, 455.0): cluster 1\n", + "Example n. 3391 = (67.0, 454.0): cluster 1\n", + "Example n. 3392 = (67.0, 453.0): cluster 1\n", + "Example n. 3393 = (67.0, 452.0): cluster 1\n", + "Example n. 3394 = (67.0, 451.0): cluster 1\n", + "Example n. 3395 = (67.0, 450.0): cluster 1\n", + "Example n. 3396 = (67.0, 449.0): cluster 1\n", + "Example n. 3397 = (67.0, 448.0): cluster 1\n", + "Example n. 3398 = (67.0, 447.0): cluster 1\n", + "Example n. 3399 = (67.0, 446.0): cluster 1\n", + "Example n. 3400 = (67.0, 445.0): cluster 1\n", + "Example n. 3401 = (67.0, 444.0): cluster 1\n", + "Example n. 3402 = (67.0, 443.0): cluster 1\n", + "Example n. 3403 = (67.0, 442.0): cluster 1\n", + "Example n. 3404 = (67.0, 441.0): cluster 3\n", + "Example n. 3405 = (67.0, 440.0): cluster 3\n", + "Example n. 3406 = (67.0, 439.0): cluster 3\n", + "Example n. 3407 = (67.0, 438.0): cluster 3\n", + "Example n. 3408 = (67.0, 437.0): cluster 3\n", + "Example n. 3409 = (67.0, 429.0): cluster 3\n", + "Example n. 3410 = (67.0, 428.0): cluster 3\n", + "Example n. 3411 = (67.0, 427.0): cluster 3\n", + "Example n. 3412 = (67.0, 426.0): cluster 3\n", + "Example n. 3413 = (67.0, 425.0): cluster 3\n", + "Example n. 3414 = (67.0, 424.0): cluster 3\n", + "Example n. 3415 = (67.0, 423.0): cluster 3\n", + "Example n. 3416 = (67.0, 422.0): cluster 3\n", + "Example n. 3417 = (67.0, 421.0): cluster 3\n", + "Example n. 3418 = (67.0, 420.0): cluster 3\n", + "Example n. 3419 = (67.0, 419.0): cluster 3\n", + "Example n. 3420 = (67.0, 418.0): cluster 3\n", + "Example n. 3421 = (67.0, 417.0): cluster 3\n", + "Example n. 3422 = (67.0, 409.0): cluster 3\n", + "Example n. 3423 = (67.0, 408.0): cluster 3\n", + "Example n. 3424 = (67.0, 407.0): cluster 3\n", + "Example n. 3425 = (67.0, 406.0): cluster 3\n", + "Example n. 3426 = (67.0, 405.0): cluster 3\n", + "Example n. 3427 = (67.0, 404.0): cluster 3\n", + "Example n. 3428 = (67.0, 403.0): cluster 3\n", + "Example n. 3429 = (67.0, 402.0): cluster 3\n", + "Example n. 3430 = (67.0, 401.0): cluster 3\n", + "Example n. 3431 = (67.0, 400.0): cluster 3\n", + "Example n. 3432 = (67.0, 399.0): cluster 3\n", + "Example n. 3433 = (67.0, 398.0): cluster 3\n", + "Example n. 3434 = (67.0, 397.0): cluster 3\n", + "Example n. 3435 = (67.0, 396.0): cluster 3\n", + "Example n. 3436 = (67.0, 395.0): cluster 3\n", + "Example n. 3437 = (67.0, 394.0): cluster 3\n", + "Example n. 3438 = (68.0, 486.0): cluster 1\n", + "Example n. 3439 = (68.0, 485.0): cluster 1\n", + "Example n. 3440 = (68.0, 484.0): cluster 1\n", + "Example n. 3441 = (68.0, 483.0): cluster 1\n", + "Example n. 3442 = (68.0, 482.0): cluster 1\n", + "Example n. 3443 = (68.0, 481.0): cluster 1\n", + "Example n. 3444 = (68.0, 480.0): cluster 1\n", + "Example n. 3445 = (68.0, 479.0): cluster 1\n", + "Example n. 3446 = (68.0, 478.0): cluster 1\n", + "Example n. 3447 = (68.0, 477.0): cluster 1\n", + "Example n. 3448 = (68.0, 476.0): cluster 1\n", + "Example n. 3449 = (68.0, 475.0): cluster 1\n", + "Example n. 3450 = (68.0, 474.0): cluster 1\n", + "Example n. 3451 = (68.0, 473.0): cluster 1\n", + "Example n. 3452 = (68.0, 472.0): cluster 1\n", + "Example n. 3453 = (68.0, 471.0): cluster 1\n", + "Example n. 3454 = (68.0, 470.0): cluster 1\n", + "Example n. 3455 = (68.0, 469.0): cluster 1\n", + "Example n. 3456 = (68.0, 468.0): cluster 1\n", + "Example n. 3457 = (68.0, 467.0): cluster 1\n", + "Example n. 3458 = (68.0, 466.0): cluster 1\n", + "Example n. 3459 = (68.0, 465.0): cluster 1\n", + "Example n. 3460 = (68.0, 464.0): cluster 1\n", + "Example n. 3461 = (68.0, 463.0): cluster 1\n", + "Example n. 3462 = (68.0, 462.0): cluster 1\n", + "Example n. 3463 = (68.0, 461.0): cluster 1\n", + "Example n. 3464 = (68.0, 460.0): cluster 1\n", + "Example n. 3465 = (68.0, 459.0): cluster 1\n", + "Example n. 3466 = (68.0, 458.0): cluster 1\n", + "Example n. 3467 = (68.0, 457.0): cluster 1\n", + "Example n. 3468 = (68.0, 456.0): cluster 1\n", + "Example n. 3469 = (68.0, 455.0): cluster 1\n", + "Example n. 3470 = (68.0, 454.0): cluster 1\n", + "Example n. 3471 = (68.0, 453.0): cluster 1\n", + "Example n. 3472 = (68.0, 452.0): cluster 1\n", + "Example n. 3473 = (68.0, 451.0): cluster 1\n", + "Example n. 3474 = (68.0, 450.0): cluster 1\n", + "Example n. 3475 = (68.0, 449.0): cluster 1\n", + "Example n. 3476 = (68.0, 448.0): cluster 1\n", + "Example n. 3477 = (68.0, 447.0): cluster 1\n", + "Example n. 3478 = (68.0, 446.0): cluster 1\n", + "Example n. 3479 = (68.0, 445.0): cluster 1\n", + "Example n. 3480 = (68.0, 444.0): cluster 1\n", + "Example n. 3481 = (68.0, 443.0): cluster 1\n", + "Example n. 3482 = (68.0, 442.0): cluster 1\n", + "Example n. 3483 = (68.0, 441.0): cluster 3\n", + "Example n. 3484 = (68.0, 440.0): cluster 3\n", + "Example n. 3485 = (68.0, 439.0): cluster 3\n", + "Example n. 3486 = (68.0, 438.0): cluster 3\n", + "Example n. 3487 = (68.0, 437.0): cluster 3\n", + "Example n. 3488 = (68.0, 429.0): cluster 3\n", + "Example n. 3489 = (68.0, 428.0): cluster 3\n", + "Example n. 3490 = (68.0, 427.0): cluster 3\n", + "Example n. 3491 = (68.0, 426.0): cluster 3\n", + "Example n. 3492 = (68.0, 425.0): cluster 3\n", + "Example n. 3493 = (68.0, 424.0): cluster 3\n", + "Example n. 3494 = (68.0, 423.0): cluster 3\n", + "Example n. 3495 = (68.0, 422.0): cluster 3\n", + "Example n. 3496 = (68.0, 421.0): cluster 3\n", + "Example n. 3497 = (68.0, 420.0): cluster 3\n", + "Example n. 3498 = (68.0, 419.0): cluster 3\n", + "Example n. 3499 = (68.0, 418.0): cluster 3\n", + "Example n. 3500 = (68.0, 417.0): cluster 3\n", + "Example n. 3501 = (68.0, 409.0): cluster 3\n", + "Example n. 3502 = (68.0, 408.0): cluster 3\n", + "Example n. 3503 = (68.0, 407.0): cluster 3\n", + "Example n. 3504 = (68.0, 406.0): cluster 3\n", + "Example n. 3505 = (68.0, 405.0): cluster 3\n", + "Example n. 3506 = (68.0, 404.0): cluster 3\n", + "Example n. 3507 = (68.0, 403.0): cluster 3\n", + "Example n. 3508 = (68.0, 402.0): cluster 3\n", + "Example n. 3509 = (68.0, 401.0): cluster 3\n", + "Example n. 3510 = (68.0, 400.0): cluster 3\n", + "Example n. 3511 = (68.0, 399.0): cluster 3\n", + "Example n. 3512 = (68.0, 398.0): cluster 3\n", + "Example n. 3513 = (68.0, 397.0): cluster 3\n", + "Example n. 3514 = (68.0, 396.0): cluster 3\n", + "Example n. 3515 = (68.0, 395.0): cluster 3\n", + "Example n. 3516 = (68.0, 394.0): cluster 3\n", + "Example n. 3517 = (69.0, 486.0): cluster 1\n", + "Example n. 3518 = (69.0, 485.0): cluster 1\n", + "Example n. 3519 = (69.0, 484.0): cluster 1\n", + "Example n. 3520 = (69.0, 483.0): cluster 1\n", + "Example n. 3521 = (69.0, 482.0): cluster 1\n", + "Example n. 3522 = (69.0, 481.0): cluster 1\n", + "Example n. 3523 = (69.0, 480.0): cluster 1\n", + "Example n. 3524 = (69.0, 479.0): cluster 1\n", + "Example n. 3525 = (69.0, 478.0): cluster 1\n", + "Example n. 3526 = (69.0, 477.0): cluster 1\n", + "Example n. 3527 = (69.0, 476.0): cluster 1\n", + "Example n. 3528 = (69.0, 475.0): cluster 1\n", + "Example n. 3529 = (69.0, 474.0): cluster 1\n", + "Example n. 3530 = (69.0, 473.0): cluster 1\n", + "Example n. 3531 = (69.0, 472.0): cluster 1\n", + "Example n. 3532 = (69.0, 471.0): cluster 1\n", + "Example n. 3533 = (69.0, 470.0): cluster 1\n", + "Example n. 3534 = (69.0, 469.0): cluster 1\n", + "Example n. 3535 = (69.0, 468.0): cluster 1\n", + "Example n. 3536 = (69.0, 467.0): cluster 1\n", + "Example n. 3537 = (69.0, 466.0): cluster 1\n", + "Example n. 3538 = (69.0, 465.0): cluster 1\n", + "Example n. 3539 = (69.0, 464.0): cluster 1\n", + "Example n. 3540 = (69.0, 463.0): cluster 1\n", + "Example n. 3541 = (69.0, 462.0): cluster 1\n", + "Example n. 3542 = (69.0, 461.0): cluster 1\n", + "Example n. 3543 = (69.0, 460.0): cluster 1\n", + "Example n. 3544 = (69.0, 459.0): cluster 1\n", + "Example n. 3545 = (69.0, 458.0): cluster 1\n", + "Example n. 3546 = (69.0, 457.0): cluster 1\n", + "Example n. 3547 = (69.0, 456.0): cluster 1\n", + "Example n. 3548 = (69.0, 455.0): cluster 1\n", + "Example n. 3549 = (69.0, 454.0): cluster 1\n", + "Example n. 3550 = (69.0, 453.0): cluster 1\n", + "Example n. 3551 = (69.0, 452.0): cluster 1\n", + "Example n. 3552 = (69.0, 451.0): cluster 1\n", + "Example n. 3553 = (69.0, 450.0): cluster 1\n", + "Example n. 3554 = (69.0, 449.0): cluster 1\n", + "Example n. 3555 = (69.0, 448.0): cluster 1\n", + "Example n. 3556 = (69.0, 447.0): cluster 1\n", + "Example n. 3557 = (69.0, 446.0): cluster 1\n", + "Example n. 3558 = (69.0, 445.0): cluster 1\n", + "Example n. 3559 = (69.0, 444.0): cluster 1\n", + "Example n. 3560 = (69.0, 443.0): cluster 1\n", + "Example n. 3561 = (69.0, 442.0): cluster 1\n", + "Example n. 3562 = (69.0, 441.0): cluster 3\n", + "Example n. 3563 = (69.0, 440.0): cluster 3\n", + "Example n. 3564 = (69.0, 439.0): cluster 3\n", + "Example n. 3565 = (69.0, 438.0): cluster 3\n", + "Example n. 3566 = (69.0, 437.0): cluster 3\n", + "Example n. 3567 = (69.0, 429.0): cluster 3\n", + "Example n. 3568 = (69.0, 428.0): cluster 3\n", + "Example n. 3569 = (69.0, 427.0): cluster 3\n", + "Example n. 3570 = (69.0, 426.0): cluster 3\n", + "Example n. 3571 = (69.0, 425.0): cluster 3\n", + "Example n. 3572 = (69.0, 424.0): cluster 3\n", + "Example n. 3573 = (69.0, 423.0): cluster 3\n", + "Example n. 3574 = (69.0, 422.0): cluster 3\n", + "Example n. 3575 = (69.0, 421.0): cluster 3\n", + "Example n. 3576 = (69.0, 420.0): cluster 3\n", + "Example n. 3577 = (69.0, 419.0): cluster 3\n", + "Example n. 3578 = (69.0, 418.0): cluster 3\n", + "Example n. 3579 = (69.0, 417.0): cluster 3\n", + "Example n. 3580 = (69.0, 409.0): cluster 3\n", + "Example n. 3581 = (69.0, 408.0): cluster 3\n", + "Example n. 3582 = (69.0, 407.0): cluster 3\n", + "Example n. 3583 = (69.0, 406.0): cluster 3\n", + "Example n. 3584 = (69.0, 405.0): cluster 3\n", + "Example n. 3585 = (69.0, 404.0): cluster 3\n", + "Example n. 3586 = (69.0, 403.0): cluster 3\n", + "Example n. 3587 = (69.0, 402.0): cluster 3\n", + "Example n. 3588 = (69.0, 401.0): cluster 3\n", + "Example n. 3589 = (69.0, 400.0): cluster 3\n", + "Example n. 3590 = (69.0, 399.0): cluster 3\n", + "Example n. 3591 = (69.0, 398.0): cluster 3\n", + "Example n. 3592 = (69.0, 397.0): cluster 3\n", + "Example n. 3593 = (69.0, 396.0): cluster 3\n", + "Example n. 3594 = (69.0, 395.0): cluster 3\n", + "Example n. 3595 = (69.0, 394.0): cluster 3\n", + "Example n. 3596 = (70.0, 486.0): cluster 1\n", + "Example n. 3597 = (70.0, 485.0): cluster 1\n", + "Example n. 3598 = (70.0, 484.0): cluster 1\n", + "Example n. 3599 = (70.0, 483.0): cluster 1\n", + "Example n. 3600 = (70.0, 482.0): cluster 1\n", + "Example n. 3601 = (70.0, 481.0): cluster 1\n", + "Example n. 3602 = (70.0, 480.0): cluster 1\n", + "Example n. 3603 = (70.0, 479.0): cluster 1\n", + "Example n. 3604 = (70.0, 478.0): cluster 1\n", + "Example n. 3605 = (70.0, 477.0): cluster 1\n", + "Example n. 3606 = (70.0, 476.0): cluster 1\n", + "Example n. 3607 = (70.0, 475.0): cluster 1\n", + "Example n. 3608 = (70.0, 474.0): cluster 1\n", + "Example n. 3609 = (70.0, 473.0): cluster 1\n", + "Example n. 3610 = (70.0, 472.0): cluster 1\n", + "Example n. 3611 = (70.0, 471.0): cluster 1\n", + "Example n. 3612 = (70.0, 470.0): cluster 1\n", + "Example n. 3613 = (70.0, 469.0): cluster 1\n", + "Example n. 3614 = (70.0, 468.0): cluster 1\n", + "Example n. 3615 = (70.0, 467.0): cluster 1\n", + "Example n. 3616 = (70.0, 466.0): cluster 1\n", + "Example n. 3617 = (70.0, 465.0): cluster 1\n", + "Example n. 3618 = (70.0, 464.0): cluster 1\n", + "Example n. 3619 = (70.0, 463.0): cluster 1\n", + "Example n. 3620 = (70.0, 462.0): cluster 1\n", + "Example n. 3621 = (70.0, 461.0): cluster 1\n", + "Example n. 3622 = (70.0, 460.0): cluster 1\n", + "Example n. 3623 = (70.0, 459.0): cluster 1\n", + "Example n. 3624 = (70.0, 458.0): cluster 1\n", + "Example n. 3625 = (70.0, 457.0): cluster 1\n", + "Example n. 3626 = (70.0, 456.0): cluster 1\n", + "Example n. 3627 = (70.0, 455.0): cluster 1\n", + "Example n. 3628 = (70.0, 454.0): cluster 1\n", + "Example n. 3629 = (70.0, 453.0): cluster 1\n", + "Example n. 3630 = (70.0, 452.0): cluster 1\n", + "Example n. 3631 = (70.0, 451.0): cluster 1\n", + "Example n. 3632 = (70.0, 450.0): cluster 1\n", + "Example n. 3633 = (70.0, 449.0): cluster 1\n", + "Example n. 3634 = (70.0, 448.0): cluster 1\n", + "Example n. 3635 = (70.0, 447.0): cluster 1\n", + "Example n. 3636 = (70.0, 446.0): cluster 1\n", + "Example n. 3637 = (70.0, 445.0): cluster 1\n", + "Example n. 3638 = (70.0, 444.0): cluster 1\n", + "Example n. 3639 = (70.0, 443.0): cluster 1\n", + "Example n. 3640 = (70.0, 442.0): cluster 1\n", + "Example n. 3641 = (70.0, 441.0): cluster 3\n", + "Example n. 3642 = (70.0, 440.0): cluster 3\n", + "Example n. 3643 = (70.0, 439.0): cluster 3\n", + "Example n. 3644 = (70.0, 438.0): cluster 3\n", + "Example n. 3645 = (70.0, 437.0): cluster 3\n", + "Example n. 3646 = (70.0, 432.0): cluster 3\n", + "Example n. 3647 = (70.0, 431.0): cluster 3\n", + "Example n. 3648 = (70.0, 430.0): cluster 3\n", + "Example n. 3649 = (70.0, 429.0): cluster 3\n", + "Example n. 3650 = (70.0, 428.0): cluster 3\n", + "Example n. 3651 = (70.0, 427.0): cluster 3\n", + "Example n. 3652 = (70.0, 426.0): cluster 3\n", + "Example n. 3653 = (70.0, 425.0): cluster 3\n", + "Example n. 3654 = (70.0, 424.0): cluster 3\n", + "Example n. 3655 = (70.0, 423.0): cluster 3\n", + "Example n. 3656 = (70.0, 422.0): cluster 3\n", + "Example n. 3657 = (70.0, 421.0): cluster 3\n", + "Example n. 3658 = (70.0, 420.0): cluster 3\n", + "Example n. 3659 = (70.0, 419.0): cluster 3\n", + "Example n. 3660 = (70.0, 418.0): cluster 3\n", + "Example n. 3661 = (70.0, 417.0): cluster 3\n", + "Example n. 3662 = (70.0, 413.0): cluster 3\n", + "Example n. 3663 = (70.0, 412.0): cluster 3\n", + "Example n. 3664 = (70.0, 411.0): cluster 3\n", + "Example n. 3665 = (70.0, 410.0): cluster 3\n", + "Example n. 3666 = (70.0, 409.0): cluster 3\n", + "Example n. 3667 = (70.0, 408.0): cluster 3\n", + "Example n. 3668 = (70.0, 407.0): cluster 3\n", + "Example n. 3669 = (70.0, 406.0): cluster 3\n", + "Example n. 3670 = (70.0, 405.0): cluster 3\n", + "Example n. 3671 = (70.0, 404.0): cluster 3\n", + "Example n. 3672 = (70.0, 403.0): cluster 3\n", + "Example n. 3673 = (70.0, 402.0): cluster 3\n", + "Example n. 3674 = (70.0, 401.0): cluster 3\n", + "Example n. 3675 = (70.0, 400.0): cluster 3\n", + "Example n. 3676 = (70.0, 399.0): cluster 3\n", + "Example n. 3677 = (70.0, 398.0): cluster 3\n", + "Example n. 3678 = (70.0, 397.0): cluster 3\n", + "Example n. 3679 = (70.0, 396.0): cluster 3\n", + "Example n. 3680 = (70.0, 395.0): cluster 3\n", + "Example n. 3681 = (70.0, 394.0): cluster 3\n", + "Example n. 3682 = (71.0, 482.0): cluster 1\n", + "Example n. 3683 = (71.0, 481.0): cluster 1\n", + "Example n. 3684 = (71.0, 480.0): cluster 1\n", + "Example n. 3685 = (71.0, 479.0): cluster 1\n", + "Example n. 3686 = (71.0, 478.0): cluster 1\n", + "Example n. 3687 = (71.0, 477.0): cluster 1\n", + "Example n. 3688 = (71.0, 476.0): cluster 1\n", + "Example n. 3689 = (71.0, 475.0): cluster 1\n", + "Example n. 3690 = (71.0, 474.0): cluster 1\n", + "Example n. 3691 = (71.0, 473.0): cluster 1\n", + "Example n. 3692 = (71.0, 472.0): cluster 1\n", + "Example n. 3693 = (71.0, 471.0): cluster 1\n", + "Example n. 3694 = (71.0, 470.0): cluster 1\n", + "Example n. 3695 = (71.0, 469.0): cluster 1\n", + "Example n. 3696 = (71.0, 468.0): cluster 1\n", + "Example n. 3697 = (71.0, 467.0): cluster 1\n", + "Example n. 3698 = (71.0, 466.0): cluster 1\n", + "Example n. 3699 = (71.0, 465.0): cluster 1\n", + "Example n. 3700 = (71.0, 464.0): cluster 1\n", + "Example n. 3701 = (71.0, 463.0): cluster 1\n", + "Example n. 3702 = (71.0, 462.0): cluster 1\n", + "Example n. 3703 = (71.0, 461.0): cluster 1\n", + "Example n. 3704 = (71.0, 460.0): cluster 1\n", + "Example n. 3705 = (71.0, 459.0): cluster 1\n", + "Example n. 3706 = (71.0, 458.0): cluster 1\n", + "Example n. 3707 = (71.0, 457.0): cluster 1\n", + "Example n. 3708 = (71.0, 456.0): cluster 1\n", + "Example n. 3709 = (71.0, 455.0): cluster 1\n", + "Example n. 3710 = (71.0, 454.0): cluster 1\n", + "Example n. 3711 = (71.0, 453.0): cluster 1\n", + "Example n. 3712 = (71.0, 452.0): cluster 1\n", + "Example n. 3713 = (71.0, 451.0): cluster 1\n", + "Example n. 3714 = (71.0, 450.0): cluster 1\n", + "Example n. 3715 = (71.0, 449.0): cluster 1\n", + "Example n. 3716 = (71.0, 448.0): cluster 1\n", + "Example n. 3717 = (71.0, 447.0): cluster 1\n", + "Example n. 3718 = (71.0, 446.0): cluster 1\n", + "Example n. 3719 = (71.0, 445.0): cluster 1\n", + "Example n. 3720 = (71.0, 444.0): cluster 1\n", + "Example n. 3721 = (71.0, 443.0): cluster 1\n", + "Example n. 3722 = (71.0, 442.0): cluster 1\n", + "Example n. 3723 = (71.0, 441.0): cluster 3\n", + "Example n. 3724 = (71.0, 432.0): cluster 3\n", + "Example n. 3725 = (71.0, 431.0): cluster 3\n", + "Example n. 3726 = (71.0, 430.0): cluster 3\n", + "Example n. 3727 = (71.0, 429.0): cluster 3\n", + "Example n. 3728 = (71.0, 428.0): cluster 3\n", + "Example n. 3729 = (71.0, 427.0): cluster 3\n", + "Example n. 3730 = (71.0, 426.0): cluster 3\n", + "Example n. 3731 = (71.0, 425.0): cluster 3\n", + "Example n. 3732 = (71.0, 424.0): cluster 3\n", + "Example n. 3733 = (71.0, 423.0): cluster 3\n", + "Example n. 3734 = (71.0, 422.0): cluster 3\n", + "Example n. 3735 = (71.0, 421.0): cluster 3\n", + "Example n. 3736 = (71.0, 420.0): cluster 3\n", + "Example n. 3737 = (71.0, 419.0): cluster 3\n", + "Example n. 3738 = (71.0, 418.0): cluster 3\n", + "Example n. 3739 = (71.0, 417.0): cluster 3\n", + "Example n. 3740 = (71.0, 413.0): cluster 3\n", + "Example n. 3741 = (71.0, 412.0): cluster 3\n", + "Example n. 3742 = (71.0, 411.0): cluster 3\n", + "Example n. 3743 = (71.0, 410.0): cluster 3\n", + "Example n. 3744 = (71.0, 409.0): cluster 3\n", + "Example n. 3745 = (71.0, 408.0): cluster 3\n", + "Example n. 3746 = (71.0, 407.0): cluster 3\n", + "Example n. 3747 = (71.0, 406.0): cluster 3\n", + "Example n. 3748 = (71.0, 405.0): cluster 3\n", + "Example n. 3749 = (71.0, 404.0): cluster 3\n", + "Example n. 3750 = (71.0, 403.0): cluster 3\n", + "Example n. 3751 = (71.0, 402.0): cluster 3\n", + "Example n. 3752 = (71.0, 401.0): cluster 3\n", + "Example n. 3753 = (71.0, 400.0): cluster 3\n", + "Example n. 3754 = (71.0, 399.0): cluster 3\n", + "Example n. 3755 = (71.0, 398.0): cluster 3\n", + "Example n. 3756 = (72.0, 482.0): cluster 1\n", + "Example n. 3757 = (72.0, 481.0): cluster 1\n", + "Example n. 3758 = (72.0, 480.0): cluster 1\n", + "Example n. 3759 = (72.0, 479.0): cluster 1\n", + "Example n. 3760 = (72.0, 478.0): cluster 1\n", + "Example n. 3761 = (72.0, 477.0): cluster 1\n", + "Example n. 3762 = (72.0, 476.0): cluster 1\n", + "Example n. 3763 = (72.0, 475.0): cluster 1\n", + "Example n. 3764 = (72.0, 474.0): cluster 1\n", + "Example n. 3765 = (72.0, 473.0): cluster 1\n", + "Example n. 3766 = (72.0, 472.0): cluster 1\n", + "Example n. 3767 = (72.0, 471.0): cluster 1\n", + "Example n. 3768 = (72.0, 470.0): cluster 1\n", + "Example n. 3769 = (72.0, 469.0): cluster 1\n", + "Example n. 3770 = (72.0, 468.0): cluster 1\n", + "Example n. 3771 = (72.0, 467.0): cluster 1\n", + "Example n. 3772 = (72.0, 466.0): cluster 1\n", + "Example n. 3773 = (72.0, 465.0): cluster 1\n", + "Example n. 3774 = (72.0, 464.0): cluster 1\n", + "Example n. 3775 = (72.0, 463.0): cluster 1\n", + "Example n. 3776 = (72.0, 462.0): cluster 1\n", + "Example n. 3777 = (72.0, 461.0): cluster 1\n", + "Example n. 3778 = (72.0, 460.0): cluster 1\n", + "Example n. 3779 = (72.0, 459.0): cluster 1\n", + "Example n. 3780 = (72.0, 458.0): cluster 1\n", + "Example n. 3781 = (72.0, 457.0): cluster 1\n", + "Example n. 3782 = (72.0, 456.0): cluster 1\n", + "Example n. 3783 = (72.0, 455.0): cluster 1\n", + "Example n. 3784 = (72.0, 454.0): cluster 1\n", + "Example n. 3785 = (72.0, 453.0): cluster 1\n", + "Example n. 3786 = (72.0, 452.0): cluster 1\n", + "Example n. 3787 = (72.0, 451.0): cluster 1\n", + "Example n. 3788 = (72.0, 450.0): cluster 1\n", + "Example n. 3789 = (72.0, 449.0): cluster 1\n", + "Example n. 3790 = (72.0, 448.0): cluster 1\n", + "Example n. 3791 = (72.0, 447.0): cluster 1\n", + "Example n. 3792 = (72.0, 446.0): cluster 1\n", + "Example n. 3793 = (72.0, 445.0): cluster 1\n", + "Example n. 3794 = (72.0, 444.0): cluster 1\n", + "Example n. 3795 = (72.0, 443.0): cluster 1\n", + "Example n. 3796 = (72.0, 442.0): cluster 1\n", + "Example n. 3797 = (72.0, 441.0): cluster 3\n", + "Example n. 3798 = (72.0, 432.0): cluster 3\n", + "Example n. 3799 = (72.0, 431.0): cluster 3\n", + "Example n. 3800 = (72.0, 430.0): cluster 3\n", + "Example n. 3801 = (72.0, 429.0): cluster 3\n", + "Example n. 3802 = (72.0, 428.0): cluster 3\n", + "Example n. 3803 = (72.0, 427.0): cluster 3\n", + "Example n. 3804 = (72.0, 426.0): cluster 3\n", + "Example n. 3805 = (72.0, 425.0): cluster 3\n", + "Example n. 3806 = (72.0, 424.0): cluster 3\n", + "Example n. 3807 = (72.0, 423.0): cluster 3\n", + "Example n. 3808 = (72.0, 422.0): cluster 3\n", + "Example n. 3809 = (72.0, 421.0): cluster 3\n", + "Example n. 3810 = (72.0, 420.0): cluster 3\n", + "Example n. 3811 = (72.0, 419.0): cluster 3\n", + "Example n. 3812 = (72.0, 418.0): cluster 3\n", + "Example n. 3813 = (72.0, 417.0): cluster 3\n", + "Example n. 3814 = (72.0, 413.0): cluster 3\n", + "Example n. 3815 = (72.0, 412.0): cluster 3\n", + "Example n. 3816 = (72.0, 411.0): cluster 3\n", + "Example n. 3817 = (72.0, 410.0): cluster 3\n", + "Example n. 3818 = (72.0, 409.0): cluster 3\n", + "Example n. 3819 = (72.0, 408.0): cluster 3\n", + "Example n. 3820 = (72.0, 407.0): cluster 3\n", + "Example n. 3821 = (72.0, 406.0): cluster 3\n", + "Example n. 3822 = (72.0, 405.0): cluster 3\n", + "Example n. 3823 = (72.0, 404.0): cluster 3\n", + "Example n. 3824 = (72.0, 403.0): cluster 3\n", + "Example n. 3825 = (72.0, 402.0): cluster 3\n", + "Example n. 3826 = (72.0, 401.0): cluster 3\n", + "Example n. 3827 = (72.0, 400.0): cluster 3\n", + "Example n. 3828 = (72.0, 399.0): cluster 3\n", + "Example n. 3829 = (72.0, 398.0): cluster 3\n", + "Example n. 3830 = (73.0, 482.0): cluster 1\n", + "Example n. 3831 = (73.0, 481.0): cluster 1\n", + "Example n. 3832 = (73.0, 480.0): cluster 1\n", + "Example n. 3833 = (73.0, 479.0): cluster 1\n", + "Example n. 3834 = (73.0, 478.0): cluster 1\n", + "Example n. 3835 = (73.0, 477.0): cluster 1\n", + "Example n. 3836 = (73.0, 476.0): cluster 1\n", + "Example n. 3837 = (73.0, 475.0): cluster 1\n", + "Example n. 3838 = (73.0, 474.0): cluster 1\n", + "Example n. 3839 = (73.0, 473.0): cluster 1\n", + "Example n. 3840 = (73.0, 472.0): cluster 1\n", + "Example n. 3841 = (73.0, 471.0): cluster 1\n", + "Example n. 3842 = (73.0, 470.0): cluster 1\n", + "Example n. 3843 = (73.0, 469.0): cluster 1\n", + "Example n. 3844 = (73.0, 468.0): cluster 1\n", + "Example n. 3845 = (73.0, 467.0): cluster 1\n", + "Example n. 3846 = (73.0, 466.0): cluster 1\n", + "Example n. 3847 = (73.0, 465.0): cluster 1\n", + "Example n. 3848 = (73.0, 464.0): cluster 1\n", + "Example n. 3849 = (73.0, 463.0): cluster 1\n", + "Example n. 3850 = (73.0, 462.0): cluster 1\n", + "Example n. 3851 = (73.0, 461.0): cluster 1\n", + "Example n. 3852 = (73.0, 460.0): cluster 1\n", + "Example n. 3853 = (73.0, 459.0): cluster 1\n", + "Example n. 3854 = (73.0, 458.0): cluster 1\n", + "Example n. 3855 = (73.0, 457.0): cluster 1\n", + "Example n. 3856 = (73.0, 456.0): cluster 1\n", + "Example n. 3857 = (73.0, 455.0): cluster 1\n", + "Example n. 3858 = (73.0, 454.0): cluster 1\n", + "Example n. 3859 = (73.0, 453.0): cluster 1\n", + "Example n. 3860 = (73.0, 452.0): cluster 1\n", + "Example n. 3861 = (73.0, 451.0): cluster 1\n", + "Example n. 3862 = (73.0, 450.0): cluster 1\n", + "Example n. 3863 = (73.0, 449.0): cluster 1\n", + "Example n. 3864 = (73.0, 448.0): cluster 1\n", + "Example n. 3865 = (73.0, 447.0): cluster 1\n", + "Example n. 3866 = (73.0, 446.0): cluster 1\n", + "Example n. 3867 = (73.0, 445.0): cluster 1\n", + "Example n. 3868 = (73.0, 444.0): cluster 1\n", + "Example n. 3869 = (73.0, 443.0): cluster 1\n", + "Example n. 3870 = (73.0, 442.0): cluster 1\n", + "Example n. 3871 = (73.0, 441.0): cluster 3\n", + "Example n. 3872 = (73.0, 432.0): cluster 3\n", + "Example n. 3873 = (73.0, 431.0): cluster 3\n", + "Example n. 3874 = (73.0, 430.0): cluster 3\n", + "Example n. 3875 = (73.0, 429.0): cluster 3\n", + "Example n. 3876 = (73.0, 428.0): cluster 3\n", + "Example n. 3877 = (73.0, 427.0): cluster 3\n", + "Example n. 3878 = (73.0, 426.0): cluster 3\n", + "Example n. 3879 = (73.0, 425.0): cluster 3\n", + "Example n. 3880 = (73.0, 424.0): cluster 3\n", + "Example n. 3881 = (73.0, 423.0): cluster 3\n", + "Example n. 3882 = (73.0, 422.0): cluster 3\n", + "Example n. 3883 = (73.0, 421.0): cluster 3\n", + "Example n. 3884 = (73.0, 420.0): cluster 3\n", + "Example n. 3885 = (73.0, 419.0): cluster 3\n", + "Example n. 3886 = (73.0, 418.0): cluster 3\n", + "Example n. 3887 = (73.0, 417.0): cluster 3\n", + "Example n. 3888 = (73.0, 413.0): cluster 3\n", + "Example n. 3889 = (73.0, 412.0): cluster 3\n", + "Example n. 3890 = (73.0, 411.0): cluster 3\n", + "Example n. 3891 = (73.0, 410.0): cluster 3\n", + "Example n. 3892 = (73.0, 409.0): cluster 3\n", + "Example n. 3893 = (73.0, 408.0): cluster 3\n", + "Example n. 3894 = (73.0, 407.0): cluster 3\n", + "Example n. 3895 = (73.0, 406.0): cluster 3\n", + "Example n. 3896 = (73.0, 405.0): cluster 3\n", + "Example n. 3897 = (73.0, 404.0): cluster 3\n", + "Example n. 3898 = (73.0, 403.0): cluster 3\n", + "Example n. 3899 = (73.0, 402.0): cluster 3\n", + "Example n. 3900 = (73.0, 401.0): cluster 3\n", + "Example n. 3901 = (73.0, 400.0): cluster 3\n", + "Example n. 3902 = (73.0, 399.0): cluster 3\n", + "Example n. 3903 = (73.0, 398.0): cluster 3\n", + "Example n. 3904 = (74.0, 482.0): cluster 1\n", + "Example n. 3905 = (74.0, 481.0): cluster 1\n", + "Example n. 3906 = (74.0, 480.0): cluster 1\n", + "Example n. 3907 = (74.0, 479.0): cluster 1\n", + "Example n. 3908 = (74.0, 478.0): cluster 1\n", + "Example n. 3909 = (74.0, 477.0): cluster 1\n", + "Example n. 3910 = (74.0, 476.0): cluster 1\n", + "Example n. 3911 = (74.0, 475.0): cluster 1\n", + "Example n. 3912 = (74.0, 474.0): cluster 1\n", + "Example n. 3913 = (74.0, 473.0): cluster 1\n", + "Example n. 3914 = (74.0, 472.0): cluster 1\n", + "Example n. 3915 = (74.0, 471.0): cluster 1\n", + "Example n. 3916 = (74.0, 470.0): cluster 1\n", + "Example n. 3917 = (74.0, 469.0): cluster 1\n", + "Example n. 3918 = (74.0, 468.0): cluster 1\n", + "Example n. 3919 = (74.0, 467.0): cluster 1\n", + "Example n. 3920 = (74.0, 466.0): cluster 1\n", + "Example n. 3921 = (74.0, 465.0): cluster 1\n", + "Example n. 3922 = (74.0, 464.0): cluster 1\n", + "Example n. 3923 = (74.0, 463.0): cluster 1\n", + "Example n. 3924 = (74.0, 462.0): cluster 1\n", + "Example n. 3925 = (74.0, 461.0): cluster 1\n", + "Example n. 3926 = (74.0, 460.0): cluster 1\n", + "Example n. 3927 = (74.0, 459.0): cluster 1\n", + "Example n. 3928 = (74.0, 458.0): cluster 1\n", + "Example n. 3929 = (74.0, 457.0): cluster 1\n", + "Example n. 3930 = (74.0, 456.0): cluster 1\n", + "Example n. 3931 = (74.0, 455.0): cluster 1\n", + "Example n. 3932 = (74.0, 454.0): cluster 1\n", + "Example n. 3933 = (74.0, 453.0): cluster 1\n", + "Example n. 3934 = (74.0, 452.0): cluster 1\n", + "Example n. 3935 = (74.0, 451.0): cluster 1\n", + "Example n. 3936 = (74.0, 450.0): cluster 1\n", + "Example n. 3937 = (74.0, 449.0): cluster 1\n", + "Example n. 3938 = (74.0, 448.0): cluster 1\n", + "Example n. 3939 = (74.0, 447.0): cluster 1\n", + "Example n. 3940 = (74.0, 446.0): cluster 1\n", + "Example n. 3941 = (74.0, 445.0): cluster 1\n", + "Example n. 3942 = (74.0, 444.0): cluster 1\n", + "Example n. 3943 = (74.0, 443.0): cluster 1\n", + "Example n. 3944 = (74.0, 442.0): cluster 1\n", + "Example n. 3945 = (74.0, 441.0): cluster 3\n", + "Example n. 3946 = (74.0, 432.0): cluster 3\n", + "Example n. 3947 = (74.0, 431.0): cluster 3\n", + "Example n. 3948 = (74.0, 430.0): cluster 3\n", + "Example n. 3949 = (74.0, 429.0): cluster 3\n", + "Example n. 3950 = (74.0, 428.0): cluster 3\n", + "Example n. 3951 = (74.0, 427.0): cluster 3\n", + "Example n. 3952 = (74.0, 426.0): cluster 3\n", + "Example n. 3953 = (74.0, 425.0): cluster 3\n", + "Example n. 3954 = (74.0, 424.0): cluster 3\n", + "Example n. 3955 = (74.0, 423.0): cluster 3\n", + "Example n. 3956 = (74.0, 422.0): cluster 3\n", + "Example n. 3957 = (74.0, 421.0): cluster 3\n", + "Example n. 3958 = (74.0, 420.0): cluster 3\n", + "Example n. 3959 = (74.0, 419.0): cluster 3\n", + "Example n. 3960 = (74.0, 418.0): cluster 3\n", + "Example n. 3961 = (74.0, 417.0): cluster 3\n", + "Example n. 3962 = (74.0, 416.0): cluster 3\n", + "Example n. 3963 = (74.0, 415.0): cluster 3\n", + "Example n. 3964 = (74.0, 414.0): cluster 3\n", + "Example n. 3965 = (74.0, 413.0): cluster 3\n", + "Example n. 3966 = (74.0, 412.0): cluster 3\n", + "Example n. 3967 = (74.0, 411.0): cluster 3\n", + "Example n. 3968 = (74.0, 410.0): cluster 3\n", + "Example n. 3969 = (74.0, 409.0): cluster 3\n", + "Example n. 3970 = (74.0, 408.0): cluster 3\n", + "Example n. 3971 = (74.0, 407.0): cluster 3\n", + "Example n. 3972 = (74.0, 406.0): cluster 3\n", + "Example n. 3973 = (74.0, 405.0): cluster 3\n", + "Example n. 3974 = (74.0, 404.0): cluster 3\n", + "Example n. 3975 = (74.0, 403.0): cluster 3\n", + "Example n. 3976 = (74.0, 402.0): cluster 3\n", + "Example n. 3977 = (74.0, 401.0): cluster 3\n", + "Example n. 3978 = (74.0, 400.0): cluster 3\n", + "Example n. 3979 = (74.0, 399.0): cluster 3\n", + "Example n. 3980 = (74.0, 398.0): cluster 3\n", + "Example n. 3981 = (75.0, 471.0): cluster 1\n", + "Example n. 3982 = (75.0, 470.0): cluster 1\n", + "Example n. 3983 = (75.0, 469.0): cluster 1\n", + "Example n. 3984 = (75.0, 468.0): cluster 1\n", + "Example n. 3985 = (75.0, 467.0): cluster 1\n", + "Example n. 3986 = (75.0, 466.0): cluster 1\n", + "Example n. 3987 = (75.0, 465.0): cluster 1\n", + "Example n. 3988 = (75.0, 464.0): cluster 1\n", + "Example n. 3989 = (75.0, 463.0): cluster 1\n", + "Example n. 3990 = (75.0, 462.0): cluster 1\n", + "Example n. 3991 = (75.0, 461.0): cluster 1\n", + "Example n. 3992 = (75.0, 460.0): cluster 1\n", + "Example n. 3993 = (75.0, 459.0): cluster 1\n", + "Example n. 3994 = (75.0, 458.0): cluster 1\n", + "Example n. 3995 = (75.0, 457.0): cluster 1\n", + "Example n. 3996 = (75.0, 456.0): cluster 1\n", + "Example n. 3997 = (75.0, 455.0): cluster 1\n", + "Example n. 3998 = (75.0, 454.0): cluster 1\n", + "Example n. 3999 = (75.0, 453.0): cluster 1\n", + "Example n. 4000 = (75.0, 452.0): cluster 1\n", + "Example n. 4001 = (75.0, 451.0): cluster 1\n", + "Example n. 4002 = (75.0, 450.0): cluster 1\n", + "Example n. 4003 = (75.0, 449.0): cluster 1\n", + "Example n. 4004 = (75.0, 448.0): cluster 1\n", + "Example n. 4005 = (75.0, 447.0): cluster 1\n", + "Example n. 4006 = (75.0, 446.0): cluster 1\n", + "Example n. 4007 = (75.0, 445.0): cluster 1\n", + "Example n. 4008 = (75.0, 444.0): cluster 1\n", + "Example n. 4009 = (75.0, 432.0): cluster 3\n", + "Example n. 4010 = (75.0, 431.0): cluster 3\n", + "Example n. 4011 = (75.0, 430.0): cluster 3\n", + "Example n. 4012 = (75.0, 429.0): cluster 3\n", + "Example n. 4013 = (75.0, 428.0): cluster 3\n", + "Example n. 4014 = (75.0, 427.0): cluster 3\n", + "Example n. 4015 = (75.0, 426.0): cluster 3\n", + "Example n. 4016 = (75.0, 425.0): cluster 3\n", + "Example n. 4017 = (75.0, 424.0): cluster 3\n", + "Example n. 4018 = (75.0, 423.0): cluster 3\n", + "Example n. 4019 = (75.0, 422.0): cluster 3\n", + "Example n. 4020 = (75.0, 421.0): cluster 3\n", + "Example n. 4021 = (75.0, 417.0): cluster 3\n", + "Example n. 4022 = (75.0, 416.0): cluster 3\n", + "Example n. 4023 = (75.0, 415.0): cluster 3\n", + "Example n. 4024 = (75.0, 414.0): cluster 3\n", + "Example n. 4025 = (75.0, 413.0): cluster 3\n", + "Example n. 4026 = (75.0, 412.0): cluster 3\n", + "Example n. 4027 = (75.0, 411.0): cluster 3\n", + "Example n. 4028 = (75.0, 410.0): cluster 3\n", + "Example n. 4029 = (75.0, 409.0): cluster 3\n", + "Example n. 4030 = (75.0, 408.0): cluster 3\n", + "Example n. 4031 = (75.0, 407.0): cluster 3\n", + "Example n. 4032 = (75.0, 406.0): cluster 3\n", + "Example n. 4033 = (75.0, 405.0): cluster 3\n", + "Example n. 4034 = (75.0, 404.0): cluster 3\n", + "Example n. 4035 = (75.0, 403.0): cluster 3\n", + "Example n. 4036 = (75.0, 402.0): cluster 3\n", + "Example n. 4037 = (75.0, 401.0): cluster 3\n", + "Example n. 4038 = (75.0, 400.0): cluster 3\n", + "Example n. 4039 = (75.0, 399.0): cluster 3\n", + "Example n. 4040 = (75.0, 398.0): cluster 3\n", + "Example n. 4041 = (76.0, 471.0): cluster 1\n", + "Example n. 4042 = (76.0, 470.0): cluster 1\n", + "Example n. 4043 = (76.0, 469.0): cluster 1\n", + "Example n. 4044 = (76.0, 468.0): cluster 1\n", + "Example n. 4045 = (76.0, 467.0): cluster 1\n", + "Example n. 4046 = (76.0, 466.0): cluster 1\n", + "Example n. 4047 = (76.0, 465.0): cluster 1\n", + "Example n. 4048 = (76.0, 464.0): cluster 1\n", + "Example n. 4049 = (76.0, 463.0): cluster 1\n", + "Example n. 4050 = (76.0, 462.0): cluster 1\n", + "Example n. 4051 = (76.0, 461.0): cluster 1\n", + "Example n. 4052 = (76.0, 460.0): cluster 1\n", + "Example n. 4053 = (76.0, 459.0): cluster 1\n", + "Example n. 4054 = (76.0, 458.0): cluster 1\n", + "Example n. 4055 = (76.0, 457.0): cluster 1\n", + "Example n. 4056 = (76.0, 456.0): cluster 1\n", + "Example n. 4057 = (76.0, 455.0): cluster 1\n", + "Example n. 4058 = (76.0, 454.0): cluster 1\n", + "Example n. 4059 = (76.0, 453.0): cluster 1\n", + "Example n. 4060 = (76.0, 452.0): cluster 1\n", + "Example n. 4061 = (76.0, 451.0): cluster 1\n", + "Example n. 4062 = (76.0, 450.0): cluster 1\n", + "Example n. 4063 = (76.0, 449.0): cluster 1\n", + "Example n. 4064 = (76.0, 448.0): cluster 1\n", + "Example n. 4065 = (76.0, 447.0): cluster 1\n", + "Example n. 4066 = (76.0, 446.0): cluster 1\n", + "Example n. 4067 = (76.0, 445.0): cluster 1\n", + "Example n. 4068 = (76.0, 444.0): cluster 1\n", + "Example n. 4069 = (76.0, 432.0): cluster 3\n", + "Example n. 4070 = (76.0, 431.0): cluster 3\n", + "Example n. 4071 = (76.0, 430.0): cluster 3\n", + "Example n. 4072 = (76.0, 429.0): cluster 3\n", + "Example n. 4073 = (76.0, 428.0): cluster 3\n", + "Example n. 4074 = (76.0, 427.0): cluster 3\n", + "Example n. 4075 = (76.0, 426.0): cluster 3\n", + "Example n. 4076 = (76.0, 425.0): cluster 3\n", + "Example n. 4077 = (76.0, 424.0): cluster 3\n", + "Example n. 4078 = (76.0, 423.0): cluster 3\n", + "Example n. 4079 = (76.0, 422.0): cluster 3\n", + "Example n. 4080 = (76.0, 421.0): cluster 3\n", + "Example n. 4081 = (76.0, 417.0): cluster 3\n", + "Example n. 4082 = (76.0, 416.0): cluster 3\n", + "Example n. 4083 = (76.0, 415.0): cluster 3\n", + "Example n. 4084 = (76.0, 414.0): cluster 3\n", + "Example n. 4085 = (76.0, 413.0): cluster 3\n", + "Example n. 4086 = (76.0, 412.0): cluster 3\n", + "Example n. 4087 = (76.0, 411.0): cluster 3\n", + "Example n. 4088 = (76.0, 410.0): cluster 3\n", + "Example n. 4089 = (76.0, 409.0): cluster 3\n", + "Example n. 4090 = (76.0, 408.0): cluster 3\n", + "Example n. 4091 = (76.0, 407.0): cluster 3\n", + "Example n. 4092 = (76.0, 406.0): cluster 3\n", + "Example n. 4093 = (76.0, 405.0): cluster 3\n", + "Example n. 4094 = (76.0, 404.0): cluster 3\n", + "Example n. 4095 = (76.0, 403.0): cluster 3\n", + "Example n. 4096 = (76.0, 402.0): cluster 3\n", + "Example n. 4097 = (76.0, 401.0): cluster 3\n", + "Example n. 4098 = (76.0, 400.0): cluster 3\n", + "Example n. 4099 = (76.0, 399.0): cluster 3\n", + "Example n. 4100 = (76.0, 398.0): cluster 3\n", + "Example n. 4101 = (77.0, 471.0): cluster 1\n", + "Example n. 4102 = (77.0, 470.0): cluster 1\n", + "Example n. 4103 = (77.0, 469.0): cluster 1\n", + "Example n. 4104 = (77.0, 468.0): cluster 1\n", + "Example n. 4105 = (77.0, 467.0): cluster 1\n", + "Example n. 4106 = (77.0, 466.0): cluster 1\n", + "Example n. 4107 = (77.0, 465.0): cluster 1\n", + "Example n. 4108 = (77.0, 464.0): cluster 1\n", + "Example n. 4109 = (77.0, 463.0): cluster 1\n", + "Example n. 4110 = (77.0, 462.0): cluster 1\n", + "Example n. 4111 = (77.0, 461.0): cluster 1\n", + "Example n. 4112 = (77.0, 460.0): cluster 1\n", + "Example n. 4113 = (77.0, 459.0): cluster 1\n", + "Example n. 4114 = (77.0, 458.0): cluster 1\n", + "Example n. 4115 = (77.0, 457.0): cluster 1\n", + "Example n. 4116 = (77.0, 456.0): cluster 1\n", + "Example n. 4117 = (77.0, 455.0): cluster 1\n", + "Example n. 4118 = (77.0, 454.0): cluster 1\n", + "Example n. 4119 = (77.0, 453.0): cluster 1\n", + "Example n. 4120 = (77.0, 452.0): cluster 1\n", + "Example n. 4121 = (77.0, 451.0): cluster 1\n", + "Example n. 4122 = (77.0, 450.0): cluster 1\n", + "Example n. 4123 = (77.0, 449.0): cluster 1\n", + "Example n. 4124 = (77.0, 448.0): cluster 1\n", + "Example n. 4125 = (77.0, 447.0): cluster 1\n", + "Example n. 4126 = (77.0, 446.0): cluster 1\n", + "Example n. 4127 = (77.0, 445.0): cluster 1\n", + "Example n. 4128 = (77.0, 444.0): cluster 1\n", + "Example n. 4129 = (77.0, 432.0): cluster 3\n", + "Example n. 4130 = (77.0, 431.0): cluster 3\n", + "Example n. 4131 = (77.0, 430.0): cluster 3\n", + "Example n. 4132 = (77.0, 429.0): cluster 3\n", + "Example n. 4133 = (77.0, 428.0): cluster 3\n", + "Example n. 4134 = (77.0, 427.0): cluster 3\n", + "Example n. 4135 = (77.0, 426.0): cluster 3\n", + "Example n. 4136 = (77.0, 425.0): cluster 3\n", + "Example n. 4137 = (77.0, 424.0): cluster 3\n", + "Example n. 4138 = (77.0, 423.0): cluster 3\n", + "Example n. 4139 = (77.0, 422.0): cluster 3\n", + "Example n. 4140 = (77.0, 421.0): cluster 3\n", + "Example n. 4141 = (77.0, 417.0): cluster 3\n", + "Example n. 4142 = (77.0, 416.0): cluster 3\n", + "Example n. 4143 = (77.0, 415.0): cluster 3\n", + "Example n. 4144 = (77.0, 414.0): cluster 3\n", + "Example n. 4145 = (77.0, 413.0): cluster 3\n", + "Example n. 4146 = (77.0, 412.0): cluster 3\n", + "Example n. 4147 = (77.0, 411.0): cluster 3\n", + "Example n. 4148 = (77.0, 410.0): cluster 3\n", + "Example n. 4149 = (77.0, 409.0): cluster 3\n", + "Example n. 4150 = (77.0, 408.0): cluster 3\n", + "Example n. 4151 = (77.0, 407.0): cluster 3\n", + "Example n. 4152 = (77.0, 406.0): cluster 3\n", + "Example n. 4153 = (77.0, 405.0): cluster 3\n", + "Example n. 4154 = (77.0, 404.0): cluster 3\n", + "Example n. 4155 = (77.0, 403.0): cluster 3\n", + "Example n. 4156 = (77.0, 402.0): cluster 3\n", + "Example n. 4157 = (77.0, 401.0): cluster 3\n", + "Example n. 4158 = (77.0, 400.0): cluster 3\n", + "Example n. 4159 = (77.0, 399.0): cluster 3\n", + "Example n. 4160 = (77.0, 398.0): cluster 3\n", + "Example n. 4161 = (78.0, 436.0): cluster 3\n", + "Example n. 4162 = (78.0, 435.0): cluster 3\n", + "Example n. 4163 = (78.0, 434.0): cluster 3\n", + "Example n. 4164 = (78.0, 433.0): cluster 3\n", + "Example n. 4165 = (78.0, 432.0): cluster 3\n", + "Example n. 4166 = (78.0, 431.0): cluster 3\n", + "Example n. 4167 = (78.0, 430.0): cluster 3\n", + "Example n. 4168 = (78.0, 429.0): cluster 3\n", + "Example n. 4169 = (78.0, 428.0): cluster 3\n", + "Example n. 4170 = (78.0, 427.0): cluster 3\n", + "Example n. 4171 = (78.0, 426.0): cluster 3\n", + "Example n. 4172 = (78.0, 425.0): cluster 3\n", + "Example n. 4173 = (78.0, 421.0): cluster 3\n", + "Example n. 4174 = (78.0, 420.0): cluster 3\n", + "Example n. 4175 = (78.0, 419.0): cluster 3\n", + "Example n. 4176 = (78.0, 418.0): cluster 3\n", + "Example n. 4177 = (78.0, 417.0): cluster 3\n", + "Example n. 4178 = (78.0, 416.0): cluster 3\n", + "Example n. 4179 = (78.0, 415.0): cluster 3\n", + "Example n. 4180 = (78.0, 414.0): cluster 3\n", + "Example n. 4181 = (78.0, 413.0): cluster 3\n", + "Example n. 4182 = (78.0, 412.0): cluster 3\n", + "Example n. 4183 = (78.0, 411.0): cluster 3\n", + "Example n. 4184 = (78.0, 410.0): cluster 3\n", + "Example n. 4185 = (78.0, 409.0): cluster 3\n", + "Example n. 4186 = (78.0, 408.0): cluster 3\n", + "Example n. 4187 = (78.0, 407.0): cluster 3\n", + "Example n. 4188 = (78.0, 406.0): cluster 3\n", + "Example n. 4189 = (78.0, 405.0): cluster 3\n", + "Example n. 4190 = (78.0, 404.0): cluster 3\n", + "Example n. 4191 = (78.0, 403.0): cluster 3\n", + "Example n. 4192 = (78.0, 402.0): cluster 3\n", + "Example n. 4193 = (78.0, 401.0): cluster 3\n", + "Example n. 4194 = (78.0, 400.0): cluster 3\n", + "Example n. 4195 = (78.0, 399.0): cluster 3\n", + "Example n. 4196 = (78.0, 398.0): cluster 3\n", + "Example n. 4197 = (79.0, 436.0): cluster 3\n", + "Example n. 4198 = (79.0, 435.0): cluster 3\n", + "Example n. 4199 = (79.0, 434.0): cluster 3\n", + "Example n. 4200 = (79.0, 433.0): cluster 3\n", + "Example n. 4201 = (79.0, 432.0): cluster 3\n", + "Example n. 4202 = (79.0, 431.0): cluster 3\n", + "Example n. 4203 = (79.0, 430.0): cluster 3\n", + "Example n. 4204 = (79.0, 429.0): cluster 3\n", + "Example n. 4205 = (79.0, 428.0): cluster 3\n", + "Example n. 4206 = (79.0, 427.0): cluster 3\n", + "Example n. 4207 = (79.0, 426.0): cluster 3\n", + "Example n. 4208 = (79.0, 425.0): cluster 3\n", + "Example n. 4209 = (79.0, 421.0): cluster 3\n", + "Example n. 4210 = (79.0, 420.0): cluster 3\n", + "Example n. 4211 = (79.0, 419.0): cluster 3\n", + "Example n. 4212 = (79.0, 418.0): cluster 3\n", + "Example n. 4213 = (79.0, 417.0): cluster 3\n", + "Example n. 4214 = (79.0, 416.0): cluster 3\n", + "Example n. 4215 = (79.0, 415.0): cluster 3\n", + "Example n. 4216 = (79.0, 414.0): cluster 3\n", + "Example n. 4217 = (79.0, 413.0): cluster 3\n", + "Example n. 4218 = (79.0, 412.0): cluster 3\n", + "Example n. 4219 = (79.0, 411.0): cluster 3\n", + "Example n. 4220 = (79.0, 410.0): cluster 3\n", + "Example n. 4221 = (79.0, 409.0): cluster 3\n", + "Example n. 4222 = (79.0, 408.0): cluster 3\n", + "Example n. 4223 = (79.0, 407.0): cluster 3\n", + "Example n. 4224 = (79.0, 406.0): cluster 3\n", + "Example n. 4225 = (79.0, 405.0): cluster 3\n", + "Example n. 4226 = (79.0, 404.0): cluster 3\n", + "Example n. 4227 = (79.0, 403.0): cluster 3\n", + "Example n. 4228 = (79.0, 402.0): cluster 3\n", + "Example n. 4229 = (79.0, 401.0): cluster 3\n", + "Example n. 4230 = (79.0, 400.0): cluster 3\n", + "Example n. 4231 = (79.0, 399.0): cluster 3\n", + "Example n. 4232 = (79.0, 398.0): cluster 3\n", + "Example n. 4233 = (80.0, 436.0): cluster 3\n", + "Example n. 4234 = (80.0, 435.0): cluster 3\n", + "Example n. 4235 = (80.0, 434.0): cluster 3\n", + "Example n. 4236 = (80.0, 433.0): cluster 3\n", + "Example n. 4237 = (80.0, 432.0): cluster 3\n", + "Example n. 4238 = (80.0, 431.0): cluster 3\n", + "Example n. 4239 = (80.0, 430.0): cluster 3\n", + "Example n. 4240 = (80.0, 429.0): cluster 3\n", + "Example n. 4241 = (80.0, 428.0): cluster 3\n", + "Example n. 4242 = (80.0, 427.0): cluster 3\n", + "Example n. 4243 = (80.0, 426.0): cluster 3\n", + "Example n. 4244 = (80.0, 425.0): cluster 3\n", + "Example n. 4245 = (80.0, 421.0): cluster 3\n", + "Example n. 4246 = (80.0, 420.0): cluster 3\n", + "Example n. 4247 = (80.0, 419.0): cluster 3\n", + "Example n. 4248 = (80.0, 418.0): cluster 3\n", + "Example n. 4249 = (80.0, 417.0): cluster 3\n", + "Example n. 4250 = (80.0, 416.0): cluster 3\n", + "Example n. 4251 = (80.0, 415.0): cluster 3\n", + "Example n. 4252 = (80.0, 414.0): cluster 3\n", + "Example n. 4253 = (80.0, 413.0): cluster 3\n", + "Example n. 4254 = (80.0, 412.0): cluster 3\n", + "Example n. 4255 = (80.0, 411.0): cluster 3\n", + "Example n. 4256 = (80.0, 410.0): cluster 3\n", + "Example n. 4257 = (80.0, 409.0): cluster 3\n", + "Example n. 4258 = (80.0, 408.0): cluster 3\n", + "Example n. 4259 = (80.0, 407.0): cluster 3\n", + "Example n. 4260 = (80.0, 406.0): cluster 3\n", + "Example n. 4261 = (80.0, 405.0): cluster 3\n", + "Example n. 4262 = (80.0, 404.0): cluster 3\n", + "Example n. 4263 = (80.0, 403.0): cluster 3\n", + "Example n. 4264 = (80.0, 402.0): cluster 3\n", + "Example n. 4265 = (80.0, 401.0): cluster 3\n", + "Example n. 4266 = (80.0, 400.0): cluster 3\n", + "Example n. 4267 = (80.0, 399.0): cluster 3\n", + "Example n. 4268 = (80.0, 398.0): cluster 3\n", + "Example n. 4269 = (81.0, 436.0): cluster 3\n", + "Example n. 4270 = (81.0, 435.0): cluster 3\n", + "Example n. 4271 = (81.0, 434.0): cluster 3\n", + "Example n. 4272 = (81.0, 433.0): cluster 3\n", + "Example n. 4273 = (81.0, 432.0): cluster 3\n", + "Example n. 4274 = (81.0, 431.0): cluster 3\n", + "Example n. 4275 = (81.0, 430.0): cluster 3\n", + "Example n. 4276 = (81.0, 429.0): cluster 3\n", + "Example n. 4277 = (81.0, 428.0): cluster 3\n", + "Example n. 4278 = (81.0, 427.0): cluster 3\n", + "Example n. 4279 = (81.0, 426.0): cluster 3\n", + "Example n. 4280 = (81.0, 425.0): cluster 3\n", + "Example n. 4281 = (81.0, 421.0): cluster 3\n", + "Example n. 4282 = (81.0, 420.0): cluster 3\n", + "Example n. 4283 = (81.0, 419.0): cluster 3\n", + "Example n. 4284 = (81.0, 418.0): cluster 3\n", + "Example n. 4285 = (81.0, 417.0): cluster 3\n", + "Example n. 4286 = (81.0, 416.0): cluster 3\n", + "Example n. 4287 = (81.0, 415.0): cluster 3\n", + "Example n. 4288 = (81.0, 414.0): cluster 3\n", + "Example n. 4289 = (81.0, 413.0): cluster 3\n", + "Example n. 4290 = (81.0, 412.0): cluster 3\n", + "Example n. 4291 = (81.0, 411.0): cluster 3\n", + "Example n. 4292 = (81.0, 410.0): cluster 3\n", + "Example n. 4293 = (81.0, 409.0): cluster 3\n", + "Example n. 4294 = (81.0, 408.0): cluster 3\n", + "Example n. 4295 = (81.0, 407.0): cluster 3\n", + "Example n. 4296 = (81.0, 406.0): cluster 3\n", + "Example n. 4297 = (81.0, 405.0): cluster 3\n", + "Example n. 4298 = (81.0, 404.0): cluster 3\n", + "Example n. 4299 = (81.0, 403.0): cluster 3\n", + "Example n. 4300 = (81.0, 402.0): cluster 3\n", + "Example n. 4301 = (81.0, 401.0): cluster 3\n", + "Example n. 4302 = (81.0, 400.0): cluster 3\n", + "Example n. 4303 = (81.0, 399.0): cluster 3\n", + "Example n. 4304 = (81.0, 398.0): cluster 3\n", + "Example n. 4305 = (82.0, 440.0): cluster 3\n", + "Example n. 4306 = (82.0, 439.0): cluster 3\n", + "Example n. 4307 = (82.0, 438.0): cluster 3\n", + "Example n. 4308 = (82.0, 437.0): cluster 3\n", + "Example n. 4309 = (82.0, 436.0): cluster 3\n", + "Example n. 4310 = (82.0, 435.0): cluster 3\n", + "Example n. 4311 = (82.0, 434.0): cluster 3\n", + "Example n. 4312 = (82.0, 433.0): cluster 3\n", + "Example n. 4313 = (82.0, 432.0): cluster 3\n", + "Example n. 4314 = (82.0, 431.0): cluster 3\n", + "Example n. 4315 = (82.0, 430.0): cluster 3\n", + "Example n. 4316 = (82.0, 429.0): cluster 3\n", + "Example n. 4317 = (82.0, 425.0): cluster 3\n", + "Example n. 4318 = (82.0, 424.0): cluster 3\n", + "Example n. 4319 = (82.0, 423.0): cluster 3\n", + "Example n. 4320 = (82.0, 422.0): cluster 3\n", + "Example n. 4321 = (82.0, 421.0): cluster 3\n", + "Example n. 4322 = (82.0, 420.0): cluster 3\n", + "Example n. 4323 = (82.0, 419.0): cluster 3\n", + "Example n. 4324 = (82.0, 418.0): cluster 3\n", + "Example n. 4325 = (82.0, 417.0): cluster 3\n", + "Example n. 4326 = (82.0, 416.0): cluster 3\n", + "Example n. 4327 = (82.0, 415.0): cluster 3\n", + "Example n. 4328 = (82.0, 414.0): cluster 3\n", + "Example n. 4329 = (82.0, 413.0): cluster 3\n", + "Example n. 4330 = (82.0, 412.0): cluster 3\n", + "Example n. 4331 = (82.0, 411.0): cluster 3\n", + "Example n. 4332 = (82.0, 410.0): cluster 3\n", + "Example n. 4333 = (82.0, 409.0): cluster 3\n", + "Example n. 4334 = (82.0, 408.0): cluster 3\n", + "Example n. 4335 = (82.0, 407.0): cluster 3\n", + "Example n. 4336 = (82.0, 406.0): cluster 3\n", + "Example n. 4337 = (82.0, 405.0): cluster 3\n", + "Example n. 4338 = (82.0, 404.0): cluster 3\n", + "Example n. 4339 = (82.0, 403.0): cluster 3\n", + "Example n. 4340 = (82.0, 402.0): cluster 3\n", + "Example n. 4341 = (83.0, 440.0): cluster 2\n", + "Example n. 4342 = (83.0, 439.0): cluster 3\n", + "Example n. 4343 = (83.0, 438.0): cluster 3\n", + "Example n. 4344 = (83.0, 437.0): cluster 3\n", + "Example n. 4345 = (83.0, 436.0): cluster 3\n", + "Example n. 4346 = (83.0, 435.0): cluster 3\n", + "Example n. 4347 = (83.0, 434.0): cluster 3\n", + "Example n. 4348 = (83.0, 433.0): cluster 3\n", + "Example n. 4349 = (83.0, 432.0): cluster 3\n", + "Example n. 4350 = (83.0, 431.0): cluster 3\n", + "Example n. 4351 = (83.0, 430.0): cluster 3\n", + "Example n. 4352 = (83.0, 429.0): cluster 3\n", + "Example n. 4353 = (83.0, 425.0): cluster 3\n", + "Example n. 4354 = (83.0, 424.0): cluster 3\n", + "Example n. 4355 = (83.0, 423.0): cluster 3\n", + "Example n. 4356 = (83.0, 422.0): cluster 3\n", + "Example n. 4357 = (83.0, 421.0): cluster 3\n", + "Example n. 4358 = (83.0, 420.0): cluster 3\n", + "Example n. 4359 = (83.0, 419.0): cluster 3\n", + "Example n. 4360 = (83.0, 418.0): cluster 3\n", + "Example n. 4361 = (83.0, 417.0): cluster 3\n", + "Example n. 4362 = (83.0, 416.0): cluster 3\n", + "Example n. 4363 = (83.0, 415.0): cluster 3\n", + "Example n. 4364 = (83.0, 414.0): cluster 3\n", + "Example n. 4365 = (83.0, 413.0): cluster 3\n", + "Example n. 4366 = (83.0, 412.0): cluster 3\n", + "Example n. 4367 = (83.0, 411.0): cluster 3\n", + "Example n. 4368 = (83.0, 410.0): cluster 3\n", + "Example n. 4369 = (83.0, 409.0): cluster 3\n", + "Example n. 4370 = (83.0, 408.0): cluster 3\n", + "Example n. 4371 = (83.0, 407.0): cluster 3\n", + "Example n. 4372 = (83.0, 406.0): cluster 3\n", + "Example n. 4373 = (83.0, 405.0): cluster 3\n", + "Example n. 4374 = (83.0, 404.0): cluster 3\n", + "Example n. 4375 = (83.0, 403.0): cluster 3\n", + "Example n. 4376 = (83.0, 402.0): cluster 3\n", + "Example n. 4377 = (84.0, 440.0): cluster 2\n", + "Example n. 4378 = (84.0, 439.0): cluster 2\n", + "Example n. 4379 = (84.0, 438.0): cluster 3\n", + "Example n. 4380 = (84.0, 437.0): cluster 3\n", + "Example n. 4381 = (84.0, 436.0): cluster 3\n", + "Example n. 4382 = (84.0, 435.0): cluster 3\n", + "Example n. 4383 = (84.0, 434.0): cluster 3\n", + "Example n. 4384 = (84.0, 433.0): cluster 3\n", + "Example n. 4385 = (84.0, 432.0): cluster 3\n", + "Example n. 4386 = (84.0, 431.0): cluster 3\n", + "Example n. 4387 = (84.0, 430.0): cluster 3\n", + "Example n. 4388 = (84.0, 429.0): cluster 3\n", + "Example n. 4389 = (84.0, 425.0): cluster 3\n", + "Example n. 4390 = (84.0, 424.0): cluster 3\n", + "Example n. 4391 = (84.0, 423.0): cluster 3\n", + "Example n. 4392 = (84.0, 422.0): cluster 3\n", + "Example n. 4393 = (84.0, 421.0): cluster 3\n", + "Example n. 4394 = (84.0, 420.0): cluster 3\n", + "Example n. 4395 = (84.0, 419.0): cluster 3\n", + "Example n. 4396 = (84.0, 418.0): cluster 3\n", + "Example n. 4397 = (84.0, 417.0): cluster 3\n", + "Example n. 4398 = (84.0, 416.0): cluster 3\n", + "Example n. 4399 = (84.0, 415.0): cluster 3\n", + "Example n. 4400 = (84.0, 414.0): cluster 3\n", + "Example n. 4401 = (84.0, 413.0): cluster 3\n", + "Example n. 4402 = (84.0, 412.0): cluster 3\n", + "Example n. 4403 = (84.0, 411.0): cluster 3\n", + "Example n. 4404 = (84.0, 410.0): cluster 3\n", + "Example n. 4405 = (84.0, 409.0): cluster 3\n", + "Example n. 4406 = (84.0, 408.0): cluster 3\n", + "Example n. 4407 = (84.0, 407.0): cluster 3\n", + "Example n. 4408 = (84.0, 406.0): cluster 3\n", + "Example n. 4409 = (84.0, 405.0): cluster 3\n", + "Example n. 4410 = (84.0, 404.0): cluster 3\n", + "Example n. 4411 = (84.0, 403.0): cluster 3\n", + "Example n. 4412 = (84.0, 402.0): cluster 3\n", + "Example n. 4413 = (85.0, 440.0): cluster 2\n", + "Example n. 4414 = (85.0, 439.0): cluster 2\n", + "Example n. 4415 = (85.0, 438.0): cluster 2\n", + "Example n. 4416 = (85.0, 437.0): cluster 2\n", + "Example n. 4417 = (85.0, 436.0): cluster 3\n", + "Example n. 4418 = (85.0, 435.0): cluster 3\n", + "Example n. 4419 = (85.0, 434.0): cluster 3\n", + "Example n. 4420 = (85.0, 433.0): cluster 3\n", + "Example n. 4421 = (85.0, 432.0): cluster 3\n", + "Example n. 4422 = (85.0, 431.0): cluster 3\n", + "Example n. 4423 = (85.0, 430.0): cluster 3\n", + "Example n. 4424 = (85.0, 429.0): cluster 3\n", + "Example n. 4425 = (85.0, 425.0): cluster 3\n", + "Example n. 4426 = (85.0, 424.0): cluster 3\n", + "Example n. 4427 = (85.0, 423.0): cluster 3\n", + "Example n. 4428 = (85.0, 422.0): cluster 3\n", + "Example n. 4429 = (85.0, 421.0): cluster 3\n", + "Example n. 4430 = (85.0, 420.0): cluster 3\n", + "Example n. 4431 = (85.0, 419.0): cluster 3\n", + "Example n. 4432 = (85.0, 418.0): cluster 3\n", + "Example n. 4433 = (85.0, 417.0): cluster 3\n", + "Example n. 4434 = (85.0, 416.0): cluster 3\n", + "Example n. 4435 = (85.0, 415.0): cluster 3\n", + "Example n. 4436 = (85.0, 414.0): cluster 3\n", + "Example n. 4437 = (85.0, 413.0): cluster 3\n", + "Example n. 4438 = (85.0, 412.0): cluster 3\n", + "Example n. 4439 = (85.0, 411.0): cluster 3\n", + "Example n. 4440 = (85.0, 410.0): cluster 3\n", + "Example n. 4441 = (85.0, 409.0): cluster 3\n", + "Example n. 4442 = (85.0, 408.0): cluster 3\n", + "Example n. 4443 = (85.0, 407.0): cluster 3\n", + "Example n. 4444 = (85.0, 406.0): cluster 3\n", + "Example n. 4445 = (85.0, 405.0): cluster 3\n", + "Example n. 4446 = (85.0, 404.0): cluster 3\n", + "Example n. 4447 = (85.0, 403.0): cluster 3\n", + "Example n. 4448 = (85.0, 402.0): cluster 3\n", + "Example n. 4449 = (86.0, 444.0): cluster 2\n", + "Example n. 4450 = (86.0, 443.0): cluster 2\n", + "Example n. 4451 = (86.0, 442.0): cluster 2\n", + "Example n. 4452 = (86.0, 441.0): cluster 2\n", + "Example n. 4453 = (86.0, 440.0): cluster 2\n", + "Example n. 4454 = (86.0, 439.0): cluster 2\n", + "Example n. 4455 = (86.0, 438.0): cluster 2\n", + "Example n. 4456 = (86.0, 437.0): cluster 2\n", + "Example n. 4457 = (86.0, 436.0): cluster 2\n", + "Example n. 4458 = (86.0, 435.0): cluster 3\n", + "Example n. 4459 = (86.0, 434.0): cluster 3\n", + "Example n. 4460 = (86.0, 433.0): cluster 3\n", + "Example n. 4461 = (86.0, 425.0): cluster 3\n", + "Example n. 4462 = (86.0, 424.0): cluster 3\n", + "Example n. 4463 = (86.0, 423.0): cluster 3\n", + "Example n. 4464 = (86.0, 422.0): cluster 3\n", + "Example n. 4465 = (86.0, 421.0): cluster 3\n", + "Example n. 4466 = (86.0, 420.0): cluster 3\n", + "Example n. 4467 = (86.0, 419.0): cluster 3\n", + "Example n. 4468 = (86.0, 418.0): cluster 3\n", + "Example n. 4469 = (86.0, 417.0): cluster 3\n", + "Example n. 4470 = (86.0, 416.0): cluster 3\n", + "Example n. 4471 = (86.0, 415.0): cluster 3\n", + "Example n. 4472 = (86.0, 414.0): cluster 3\n", + "Example n. 4473 = (86.0, 413.0): cluster 3\n", + "Example n. 4474 = (86.0, 412.0): cluster 3\n", + "Example n. 4475 = (86.0, 411.0): cluster 3\n", + "Example n. 4476 = (86.0, 410.0): cluster 3\n", + "Example n. 4477 = (86.0, 409.0): cluster 3\n", + "Example n. 4478 = (86.0, 408.0): cluster 3\n", + "Example n. 4479 = (86.0, 407.0): cluster 3\n", + "Example n. 4480 = (86.0, 406.0): cluster 3\n", + "Example n. 4481 = (86.0, 405.0): cluster 3\n", + "Example n. 4482 = (86.0, 404.0): cluster 3\n", + "Example n. 4483 = (86.0, 403.0): cluster 3\n", + "Example n. 4484 = (86.0, 402.0): cluster 3\n", + "Example n. 4485 = (87.0, 444.0): cluster 2\n", + "Example n. 4486 = (87.0, 443.0): cluster 2\n", + "Example n. 4487 = (87.0, 442.0): cluster 2\n", + "Example n. 4488 = (87.0, 441.0): cluster 2\n", + "Example n. 4489 = (87.0, 440.0): cluster 2\n", + "Example n. 4490 = (87.0, 439.0): cluster 2\n", + "Example n. 4491 = (87.0, 438.0): cluster 2\n", + "Example n. 4492 = (87.0, 437.0): cluster 2\n", + "Example n. 4493 = (87.0, 436.0): cluster 2\n", + "Example n. 4494 = (87.0, 435.0): cluster 2\n", + "Example n. 4495 = (87.0, 434.0): cluster 3\n", + "Example n. 4496 = (87.0, 433.0): cluster 3\n", + "Example n. 4497 = (87.0, 425.0): cluster 3\n", + "Example n. 4498 = (87.0, 424.0): cluster 3\n", + "Example n. 4499 = (87.0, 423.0): cluster 3\n", + "Example n. 4500 = (87.0, 422.0): cluster 3\n", + "Example n. 4501 = (87.0, 421.0): cluster 3\n", + "Example n. 4502 = (87.0, 420.0): cluster 3\n", + "Example n. 4503 = (87.0, 419.0): cluster 3\n", + "Example n. 4504 = (87.0, 418.0): cluster 3\n", + "Example n. 4505 = (87.0, 417.0): cluster 3\n", + "Example n. 4506 = (87.0, 416.0): cluster 3\n", + "Example n. 4507 = (87.0, 415.0): cluster 3\n", + "Example n. 4508 = (87.0, 414.0): cluster 3\n", + "Example n. 4509 = (87.0, 413.0): cluster 3\n", + "Example n. 4510 = (87.0, 412.0): cluster 3\n", + "Example n. 4511 = (87.0, 411.0): cluster 3\n", + "Example n. 4512 = (87.0, 410.0): cluster 3\n", + "Example n. 4513 = (87.0, 409.0): cluster 3\n", + "Example n. 4514 = (87.0, 408.0): cluster 3\n", + "Example n. 4515 = (87.0, 407.0): cluster 3\n", + "Example n. 4516 = (87.0, 406.0): cluster 3\n", + "Example n. 4517 = (87.0, 405.0): cluster 3\n", + "Example n. 4518 = (87.0, 404.0): cluster 3\n", + "Example n. 4519 = (87.0, 403.0): cluster 3\n", + "Example n. 4520 = (87.0, 402.0): cluster 3\n", + "Example n. 4521 = (88.0, 444.0): cluster 2\n", + "Example n. 4522 = (88.0, 443.0): cluster 2\n", + "Example n. 4523 = (88.0, 442.0): cluster 2\n", + "Example n. 4524 = (88.0, 441.0): cluster 2\n", + "Example n. 4525 = (88.0, 440.0): cluster 2\n", + "Example n. 4526 = (88.0, 439.0): cluster 2\n", + "Example n. 4527 = (88.0, 438.0): cluster 2\n", + "Example n. 4528 = (88.0, 437.0): cluster 2\n", + "Example n. 4529 = (88.0, 436.0): cluster 2\n", + "Example n. 4530 = (88.0, 435.0): cluster 2\n", + "Example n. 4531 = (88.0, 434.0): cluster 2\n", + "Example n. 4532 = (88.0, 433.0): cluster 2\n", + "Example n. 4533 = (88.0, 425.0): cluster 3\n", + "Example n. 4534 = (88.0, 424.0): cluster 3\n", + "Example n. 4535 = (88.0, 423.0): cluster 3\n", + "Example n. 4536 = (88.0, 422.0): cluster 3\n", + "Example n. 4537 = (88.0, 421.0): cluster 3\n", + "Example n. 4538 = (88.0, 420.0): cluster 3\n", + "Example n. 4539 = (88.0, 419.0): cluster 3\n", + "Example n. 4540 = (88.0, 418.0): cluster 3\n", + "Example n. 4541 = (88.0, 417.0): cluster 3\n", + "Example n. 4542 = (88.0, 416.0): cluster 3\n", + "Example n. 4543 = (88.0, 415.0): cluster 3\n", + "Example n. 4544 = (88.0, 414.0): cluster 3\n", + "Example n. 4545 = (88.0, 413.0): cluster 3\n", + "Example n. 4546 = (88.0, 412.0): cluster 3\n", + "Example n. 4547 = (88.0, 411.0): cluster 3\n", + "Example n. 4548 = (88.0, 410.0): cluster 3\n", + "Example n. 4549 = (88.0, 409.0): cluster 3\n", + "Example n. 4550 = (88.0, 408.0): cluster 3\n", + "Example n. 4551 = (88.0, 407.0): cluster 3\n", + "Example n. 4552 = (88.0, 406.0): cluster 3\n", + "Example n. 4553 = (88.0, 405.0): cluster 3\n", + "Example n. 4554 = (88.0, 404.0): cluster 3\n", + "Example n. 4555 = (88.0, 403.0): cluster 3\n", + "Example n. 4556 = (88.0, 402.0): cluster 3\n", + "Example n. 4557 = (89.0, 444.0): cluster 2\n", + "Example n. 4558 = (89.0, 443.0): cluster 2\n", + "Example n. 4559 = (89.0, 442.0): cluster 2\n", + "Example n. 4560 = (89.0, 441.0): cluster 2\n", + "Example n. 4561 = (89.0, 440.0): cluster 2\n", + "Example n. 4562 = (89.0, 439.0): cluster 2\n", + "Example n. 4563 = (89.0, 438.0): cluster 2\n", + "Example n. 4564 = (89.0, 437.0): cluster 2\n", + "Example n. 4565 = (89.0, 436.0): cluster 2\n", + "Example n. 4566 = (89.0, 435.0): cluster 2\n", + "Example n. 4567 = (89.0, 434.0): cluster 2\n", + "Example n. 4568 = (89.0, 433.0): cluster 2\n", + "Example n. 4569 = (89.0, 425.0): cluster 3\n", + "Example n. 4570 = (89.0, 424.0): cluster 3\n", + "Example n. 4571 = (89.0, 423.0): cluster 3\n", + "Example n. 4572 = (89.0, 422.0): cluster 3\n", + "Example n. 4573 = (89.0, 421.0): cluster 3\n", + "Example n. 4574 = (89.0, 420.0): cluster 3\n", + "Example n. 4575 = (89.0, 419.0): cluster 3\n", + "Example n. 4576 = (89.0, 418.0): cluster 3\n", + "Example n. 4577 = (89.0, 417.0): cluster 3\n", + "Example n. 4578 = (89.0, 416.0): cluster 3\n", + "Example n. 4579 = (89.0, 415.0): cluster 3\n", + "Example n. 4580 = (89.0, 414.0): cluster 3\n", + "Example n. 4581 = (89.0, 413.0): cluster 3\n", + "Example n. 4582 = (89.0, 412.0): cluster 3\n", + "Example n. 4583 = (89.0, 411.0): cluster 3\n", + "Example n. 4584 = (89.0, 410.0): cluster 3\n", + "Example n. 4585 = (89.0, 409.0): cluster 3\n", + "Example n. 4586 = (89.0, 408.0): cluster 3\n", + "Example n. 4587 = (89.0, 407.0): cluster 3\n", + "Example n. 4588 = (89.0, 406.0): cluster 3\n", + "Example n. 4589 = (89.0, 405.0): cluster 3\n", + "Example n. 4590 = (89.0, 404.0): cluster 3\n", + "Example n. 4591 = (89.0, 403.0): cluster 3\n", + "Example n. 4592 = (89.0, 402.0): cluster 3\n", + "Example n. 4593 = (90.0, 471.0): cluster 2\n", + "Example n. 4594 = (90.0, 470.0): cluster 2\n", + "Example n. 4595 = (90.0, 469.0): cluster 2\n", + "Example n. 4596 = (90.0, 468.0): cluster 2\n", + "Example n. 4597 = (90.0, 467.0): cluster 2\n", + "Example n. 4598 = (90.0, 466.0): cluster 2\n", + "Example n. 4599 = (90.0, 465.0): cluster 2\n", + "Example n. 4600 = (90.0, 464.0): cluster 2\n", + "Example n. 4601 = (90.0, 463.0): cluster 2\n", + "Example n. 4602 = (90.0, 462.0): cluster 2\n", + "Example n. 4603 = (90.0, 461.0): cluster 2\n", + "Example n. 4604 = (90.0, 460.0): cluster 2\n", + "Example n. 4605 = (90.0, 444.0): cluster 2\n", + "Example n. 4606 = (90.0, 443.0): cluster 2\n", + "Example n. 4607 = (90.0, 442.0): cluster 2\n", + "Example n. 4608 = (90.0, 441.0): cluster 2\n", + "Example n. 4609 = (90.0, 440.0): cluster 2\n", + "Example n. 4610 = (90.0, 439.0): cluster 2\n", + "Example n. 4611 = (90.0, 438.0): cluster 2\n", + "Example n. 4612 = (90.0, 437.0): cluster 2\n", + "Example n. 4613 = (90.0, 436.0): cluster 2\n", + "Example n. 4614 = (90.0, 435.0): cluster 2\n", + "Example n. 4615 = (90.0, 434.0): cluster 2\n", + "Example n. 4616 = (90.0, 433.0): cluster 2\n", + "Example n. 4617 = (90.0, 425.0): cluster 3\n", + "Example n. 4618 = (90.0, 424.0): cluster 3\n", + "Example n. 4619 = (90.0, 423.0): cluster 3\n", + "Example n. 4620 = (90.0, 422.0): cluster 3\n", + "Example n. 4621 = (90.0, 421.0): cluster 3\n", + "Example n. 4622 = (90.0, 420.0): cluster 3\n", + "Example n. 4623 = (90.0, 419.0): cluster 3\n", + "Example n. 4624 = (90.0, 418.0): cluster 3\n", + "Example n. 4625 = (90.0, 417.0): cluster 3\n", + "Example n. 4626 = (90.0, 416.0): cluster 3\n", + "Example n. 4627 = (90.0, 415.0): cluster 3\n", + "Example n. 4628 = (90.0, 414.0): cluster 3\n", + "Example n. 4629 = (90.0, 413.0): cluster 3\n", + "Example n. 4630 = (90.0, 412.0): cluster 3\n", + "Example n. 4631 = (90.0, 411.0): cluster 3\n", + "Example n. 4632 = (90.0, 410.0): cluster 3\n", + "Example n. 4633 = (90.0, 409.0): cluster 3\n", + "Example n. 4634 = (90.0, 408.0): cluster 3\n", + "Example n. 4635 = (90.0, 407.0): cluster 3\n", + "Example n. 4636 = (90.0, 406.0): cluster 3\n", + "Example n. 4637 = (91.0, 471.0): cluster 2\n", + "Example n. 4638 = (91.0, 470.0): cluster 2\n", + "Example n. 4639 = (91.0, 469.0): cluster 2\n", + "Example n. 4640 = (91.0, 468.0): cluster 2\n", + "Example n. 4641 = (91.0, 467.0): cluster 2\n", + "Example n. 4642 = (91.0, 466.0): cluster 2\n", + "Example n. 4643 = (91.0, 465.0): cluster 2\n", + "Example n. 4644 = (91.0, 464.0): cluster 2\n", + "Example n. 4645 = (91.0, 463.0): cluster 2\n", + "Example n. 4646 = (91.0, 462.0): cluster 2\n", + "Example n. 4647 = (91.0, 460.0): cluster 2\n", + "Example n. 4648 = (91.0, 444.0): cluster 2\n", + "Example n. 4649 = (91.0, 443.0): cluster 2\n", + "Example n. 4650 = (91.0, 442.0): cluster 2\n", + "Example n. 4651 = (91.0, 441.0): cluster 2\n", + "Example n. 4652 = (91.0, 440.0): cluster 2\n", + "Example n. 4653 = (91.0, 439.0): cluster 2\n", + "Example n. 4654 = (91.0, 438.0): cluster 2\n", + "Example n. 4655 = (91.0, 437.0): cluster 2\n", + "Example n. 4656 = (91.0, 436.0): cluster 2\n", + "Example n. 4657 = (91.0, 435.0): cluster 2\n", + "Example n. 4658 = (91.0, 434.0): cluster 2\n", + "Example n. 4659 = (91.0, 433.0): cluster 2\n", + "Example n. 4660 = (91.0, 425.0): cluster 3\n", + "Example n. 4661 = (91.0, 424.0): cluster 3\n", + "Example n. 4662 = (91.0, 423.0): cluster 3\n", + "Example n. 4663 = (91.0, 422.0): cluster 3\n", + "Example n. 4664 = (91.0, 421.0): cluster 3\n", + "Example n. 4665 = (91.0, 420.0): cluster 3\n", + "Example n. 4666 = (91.0, 419.0): cluster 3\n", + "Example n. 4667 = (91.0, 418.0): cluster 3\n", + "Example n. 4668 = (91.0, 417.0): cluster 3\n", + "Example n. 4669 = (91.0, 416.0): cluster 3\n", + "Example n. 4670 = (91.0, 415.0): cluster 3\n", + "Example n. 4671 = (91.0, 414.0): cluster 3\n", + "Example n. 4672 = (91.0, 413.0): cluster 3\n", + "Example n. 4673 = (91.0, 412.0): cluster 3\n", + "Example n. 4674 = (91.0, 411.0): cluster 3\n", + "Example n. 4675 = (91.0, 410.0): cluster 3\n", + "Example n. 4676 = (91.0, 409.0): cluster 3\n", + "Example n. 4677 = (91.0, 408.0): cluster 3\n", + "Example n. 4678 = (91.0, 407.0): cluster 3\n", + "Example n. 4679 = (91.0, 406.0): cluster 3\n", + "Example n. 4680 = (92.0, 471.0): cluster 2\n", + "Example n. 4681 = (92.0, 470.0): cluster 2\n", + "Example n. 4682 = (92.0, 469.0): cluster 2\n", + "Example n. 4683 = (92.0, 468.0): cluster 2\n", + "Example n. 4684 = (92.0, 467.0): cluster 2\n", + "Example n. 4685 = (92.0, 466.0): cluster 2\n", + "Example n. 4686 = (92.0, 465.0): cluster 2\n", + "Example n. 4687 = (92.0, 464.0): cluster 2\n", + "Example n. 4688 = (92.0, 463.0): cluster 2\n", + "Example n. 4689 = (92.0, 462.0): cluster 2\n", + "Example n. 4690 = (92.0, 461.0): cluster 2\n", + "Example n. 4691 = (92.0, 460.0): cluster 2\n", + "Example n. 4692 = (92.0, 444.0): cluster 2\n", + "Example n. 4693 = (92.0, 443.0): cluster 2\n", + "Example n. 4694 = (92.0, 442.0): cluster 2\n", + "Example n. 4695 = (92.0, 441.0): cluster 2\n", + "Example n. 4696 = (92.0, 440.0): cluster 2\n", + "Example n. 4697 = (92.0, 439.0): cluster 2\n", + "Example n. 4698 = (92.0, 438.0): cluster 2\n", + "Example n. 4699 = (92.0, 437.0): cluster 2\n", + "Example n. 4700 = (92.0, 436.0): cluster 2\n", + "Example n. 4701 = (92.0, 435.0): cluster 2\n", + "Example n. 4702 = (92.0, 434.0): cluster 2\n", + "Example n. 4703 = (92.0, 433.0): cluster 2\n", + "Example n. 4704 = (92.0, 425.0): cluster 3\n", + "Example n. 4705 = (92.0, 424.0): cluster 3\n", + "Example n. 4706 = (92.0, 423.0): cluster 3\n", + "Example n. 4707 = (92.0, 422.0): cluster 3\n", + "Example n. 4708 = (92.0, 421.0): cluster 3\n", + "Example n. 4709 = (92.0, 420.0): cluster 3\n", + "Example n. 4710 = (92.0, 419.0): cluster 3\n", + "Example n. 4711 = (92.0, 418.0): cluster 3\n", + "Example n. 4712 = (92.0, 417.0): cluster 3\n", + "Example n. 4713 = (92.0, 416.0): cluster 3\n", + "Example n. 4714 = (92.0, 415.0): cluster 3\n", + "Example n. 4715 = (92.0, 414.0): cluster 3\n", + "Example n. 4716 = (92.0, 413.0): cluster 3\n", + "Example n. 4717 = (92.0, 412.0): cluster 3\n", + "Example n. 4718 = (92.0, 411.0): cluster 3\n", + "Example n. 4719 = (92.0, 410.0): cluster 3\n", + "Example n. 4720 = (92.0, 409.0): cluster 3\n", + "Example n. 4721 = (92.0, 408.0): cluster 3\n", + "Example n. 4722 = (92.0, 407.0): cluster 3\n", + "Example n. 4723 = (92.0, 406.0): cluster 3\n", + "Example n. 4724 = (93.0, 479.0): cluster 2\n", + "Example n. 4725 = (93.0, 478.0): cluster 2\n", + "Example n. 4726 = (93.0, 477.0): cluster 2\n", + "Example n. 4727 = (93.0, 476.0): cluster 2\n", + "Example n. 4728 = (93.0, 475.0): cluster 2\n", + "Example n. 4729 = (93.0, 474.0): cluster 2\n", + "Example n. 4730 = (93.0, 473.0): cluster 2\n", + "Example n. 4731 = (93.0, 472.0): cluster 2\n", + "Example n. 4732 = (93.0, 471.0): cluster 2\n", + "Example n. 4733 = (93.0, 470.0): cluster 2\n", + "Example n. 4734 = (93.0, 469.0): cluster 2\n", + "Example n. 4735 = (93.0, 468.0): cluster 2\n", + "Example n. 4736 = (93.0, 467.0): cluster 2\n", + "Example n. 4737 = (93.0, 466.0): cluster 2\n", + "Example n. 4738 = (93.0, 465.0): cluster 2\n", + "Example n. 4739 = (93.0, 464.0): cluster 2\n", + "Example n. 4740 = (93.0, 463.0): cluster 2\n", + "Example n. 4741 = (93.0, 462.0): cluster 2\n", + "Example n. 4742 = (93.0, 461.0): cluster 2\n", + "Example n. 4743 = (93.0, 460.0): cluster 2\n", + "Example n. 4744 = (93.0, 444.0): cluster 2\n", + "Example n. 4745 = (93.0, 443.0): cluster 2\n", + "Example n. 4746 = (93.0, 442.0): cluster 2\n", + "Example n. 4747 = (93.0, 441.0): cluster 2\n", + "Example n. 4748 = (93.0, 440.0): cluster 2\n", + "Example n. 4749 = (93.0, 439.0): cluster 2\n", + "Example n. 4750 = (93.0, 438.0): cluster 2\n", + "Example n. 4751 = (93.0, 437.0): cluster 2\n", + "Example n. 4752 = (93.0, 436.0): cluster 2\n", + "Example n. 4753 = (93.0, 435.0): cluster 2\n", + "Example n. 4754 = (93.0, 434.0): cluster 2\n", + "Example n. 4755 = (93.0, 433.0): cluster 2\n", + "Example n. 4756 = (93.0, 425.0): cluster 3\n", + "Example n. 4757 = (93.0, 424.0): cluster 3\n", + "Example n. 4758 = (93.0, 423.0): cluster 3\n", + "Example n. 4759 = (93.0, 422.0): cluster 3\n", + "Example n. 4760 = (93.0, 421.0): cluster 3\n", + "Example n. 4761 = (93.0, 420.0): cluster 3\n", + "Example n. 4762 = (93.0, 419.0): cluster 3\n", + "Example n. 4763 = (93.0, 418.0): cluster 3\n", + "Example n. 4764 = (93.0, 417.0): cluster 3\n", + "Example n. 4765 = (93.0, 416.0): cluster 3\n", + "Example n. 4766 = (93.0, 415.0): cluster 3\n", + "Example n. 4767 = (93.0, 414.0): cluster 3\n", + "Example n. 4768 = (93.0, 413.0): cluster 3\n", + "Example n. 4769 = (93.0, 412.0): cluster 3\n", + "Example n. 4770 = (93.0, 411.0): cluster 3\n", + "Example n. 4771 = (93.0, 410.0): cluster 3\n", + "Example n. 4772 = (93.0, 409.0): cluster 3\n", + "Example n. 4773 = (93.0, 408.0): cluster 3\n", + "Example n. 4774 = (93.0, 407.0): cluster 3\n", + "Example n. 4775 = (93.0, 406.0): cluster 3\n", + "Example n. 4776 = (94.0, 479.0): cluster 2\n", + "Example n. 4777 = (94.0, 478.0): cluster 2\n", + "Example n. 4778 = (94.0, 477.0): cluster 2\n", + "Example n. 4779 = (94.0, 476.0): cluster 2\n", + "Example n. 4780 = (94.0, 475.0): cluster 2\n", + "Example n. 4781 = (94.0, 474.0): cluster 2\n", + "Example n. 4782 = (94.0, 473.0): cluster 2\n", + "Example n. 4783 = (94.0, 472.0): cluster 2\n", + "Example n. 4784 = (94.0, 471.0): cluster 2\n", + "Example n. 4785 = (94.0, 470.0): cluster 2\n", + "Example n. 4786 = (94.0, 469.0): cluster 2\n", + "Example n. 4787 = (94.0, 468.0): cluster 2\n", + "Example n. 4788 = (94.0, 467.0): cluster 2\n", + "Example n. 4789 = (94.0, 466.0): cluster 2\n", + "Example n. 4790 = (94.0, 465.0): cluster 2\n", + "Example n. 4791 = (94.0, 464.0): cluster 2\n", + "Example n. 4792 = (94.0, 463.0): cluster 2\n", + "Example n. 4793 = (94.0, 462.0): cluster 2\n", + "Example n. 4794 = (94.0, 460.0): cluster 2\n", + "Example n. 4795 = (94.0, 444.0): cluster 2\n", + "Example n. 4796 = (94.0, 443.0): cluster 2\n", + "Example n. 4797 = (94.0, 442.0): cluster 2\n", + "Example n. 4798 = (94.0, 441.0): cluster 2\n", + "Example n. 4799 = (94.0, 440.0): cluster 2\n", + "Example n. 4800 = (94.0, 439.0): cluster 2\n", + "Example n. 4801 = (94.0, 438.0): cluster 2\n", + "Example n. 4802 = (94.0, 437.0): cluster 2\n", + "Example n. 4803 = (94.0, 436.0): cluster 2\n", + "Example n. 4804 = (94.0, 435.0): cluster 2\n", + "Example n. 4805 = (94.0, 434.0): cluster 2\n", + "Example n. 4806 = (94.0, 433.0): cluster 2\n", + "Example n. 4807 = (94.0, 425.0): cluster 2\n", + "Example n. 4808 = (94.0, 424.0): cluster 3\n", + "Example n. 4809 = (94.0, 423.0): cluster 3\n", + "Example n. 4810 = (94.0, 422.0): cluster 3\n", + "Example n. 4811 = (94.0, 421.0): cluster 3\n", + "Example n. 4812 = (94.0, 420.0): cluster 3\n", + "Example n. 4813 = (94.0, 419.0): cluster 3\n", + "Example n. 4814 = (94.0, 418.0): cluster 3\n", + "Example n. 4815 = (94.0, 417.0): cluster 3\n", + "Example n. 4816 = (94.0, 416.0): cluster 3\n", + "Example n. 4817 = (94.0, 415.0): cluster 3\n", + "Example n. 4818 = (94.0, 414.0): cluster 3\n", + "Example n. 4819 = (95.0, 479.0): cluster 2\n", + "Example n. 4820 = (95.0, 478.0): cluster 2\n", + "Example n. 4821 = (95.0, 477.0): cluster 2\n", + "Example n. 4822 = (95.0, 476.0): cluster 2\n", + "Example n. 4823 = (95.0, 475.0): cluster 2\n", + "Example n. 4824 = (95.0, 474.0): cluster 2\n", + "Example n. 4825 = (95.0, 473.0): cluster 2\n", + "Example n. 4826 = (95.0, 472.0): cluster 2\n", + "Example n. 4827 = (95.0, 471.0): cluster 2\n", + "Example n. 4828 = (95.0, 470.0): cluster 2\n", + "Example n. 4829 = (95.0, 469.0): cluster 2\n", + "Example n. 4830 = (95.0, 468.0): cluster 2\n", + "Example n. 4831 = (95.0, 467.0): cluster 2\n", + "Example n. 4832 = (95.0, 466.0): cluster 2\n", + "Example n. 4833 = (95.0, 465.0): cluster 2\n", + "Example n. 4834 = (95.0, 464.0): cluster 2\n", + "Example n. 4835 = (95.0, 463.0): cluster 2\n", + "Example n. 4836 = (95.0, 462.0): cluster 2\n", + "Example n. 4837 = (95.0, 461.0): cluster 2\n", + "Example n. 4838 = (95.0, 460.0): cluster 2\n", + "Example n. 4839 = (95.0, 444.0): cluster 2\n", + "Example n. 4840 = (95.0, 443.0): cluster 2\n", + "Example n. 4841 = (95.0, 442.0): cluster 2\n", + "Example n. 4842 = (95.0, 441.0): cluster 2\n", + "Example n. 4843 = (95.0, 440.0): cluster 2\n", + "Example n. 4844 = (95.0, 439.0): cluster 2\n", + "Example n. 4845 = (95.0, 438.0): cluster 2\n", + "Example n. 4846 = (95.0, 437.0): cluster 2\n", + "Example n. 4847 = (95.0, 436.0): cluster 2\n", + "Example n. 4848 = (95.0, 435.0): cluster 2\n", + "Example n. 4849 = (95.0, 434.0): cluster 2\n", + "Example n. 4850 = (95.0, 433.0): cluster 2\n", + "Example n. 4851 = (95.0, 425.0): cluster 2\n", + "Example n. 4852 = (95.0, 424.0): cluster 2\n", + "Example n. 4853 = (95.0, 423.0): cluster 3\n", + "Example n. 4854 = (95.0, 422.0): cluster 3\n", + "Example n. 4855 = (95.0, 421.0): cluster 3\n", + "Example n. 4856 = (95.0, 420.0): cluster 3\n", + "Example n. 4857 = (95.0, 419.0): cluster 3\n", + "Example n. 4858 = (95.0, 418.0): cluster 3\n", + "Example n. 4859 = (95.0, 417.0): cluster 3\n", + "Example n. 4860 = (95.0, 416.0): cluster 3\n", + "Example n. 4861 = (95.0, 415.0): cluster 3\n", + "Example n. 4862 = (95.0, 414.0): cluster 3\n", + "Example n. 4863 = (96.0, 479.0): cluster 2\n", + "Example n. 4864 = (96.0, 478.0): cluster 2\n", + "Example n. 4865 = (96.0, 477.0): cluster 2\n", + "Example n. 4866 = (96.0, 476.0): cluster 2\n", + "Example n. 4867 = (96.0, 475.0): cluster 2\n", + "Example n. 4868 = (96.0, 474.0): cluster 2\n", + "Example n. 4869 = (96.0, 473.0): cluster 2\n", + "Example n. 4870 = (96.0, 472.0): cluster 2\n", + "Example n. 4871 = (96.0, 471.0): cluster 2\n", + "Example n. 4872 = (96.0, 470.0): cluster 2\n", + "Example n. 4873 = (96.0, 469.0): cluster 2\n", + "Example n. 4874 = (96.0, 468.0): cluster 2\n", + "Example n. 4875 = (96.0, 467.0): cluster 2\n", + "Example n. 4876 = (96.0, 466.0): cluster 2\n", + "Example n. 4877 = (96.0, 465.0): cluster 2\n", + "Example n. 4878 = (96.0, 464.0): cluster 2\n", + "Example n. 4879 = (96.0, 463.0): cluster 2\n", + "Example n. 4880 = (96.0, 462.0): cluster 2\n", + "Example n. 4881 = (96.0, 461.0): cluster 2\n", + "Example n. 4882 = (96.0, 460.0): cluster 2\n", + "Example n. 4883 = (96.0, 444.0): cluster 2\n", + "Example n. 4884 = (96.0, 443.0): cluster 2\n", + "Example n. 4885 = (96.0, 442.0): cluster 2\n", + "Example n. 4886 = (96.0, 441.0): cluster 2\n", + "Example n. 4887 = (96.0, 440.0): cluster 2\n", + "Example n. 4888 = (96.0, 439.0): cluster 2\n", + "Example n. 4889 = (96.0, 438.0): cluster 2\n", + "Example n. 4890 = (96.0, 437.0): cluster 2\n", + "Example n. 4891 = (96.0, 436.0): cluster 2\n", + "Example n. 4892 = (96.0, 435.0): cluster 2\n", + "Example n. 4893 = (96.0, 434.0): cluster 2\n", + "Example n. 4894 = (96.0, 433.0): cluster 2\n", + "Example n. 4895 = (96.0, 425.0): cluster 2\n", + "Example n. 4896 = (96.0, 424.0): cluster 2\n", + "Example n. 4897 = (96.0, 423.0): cluster 2\n", + "Example n. 4898 = (96.0, 422.0): cluster 3\n", + "Example n. 4899 = (96.0, 421.0): cluster 3\n", + "Example n. 4900 = (96.0, 420.0): cluster 3\n", + "Example n. 4901 = (96.0, 419.0): cluster 3\n", + "Example n. 4902 = (96.0, 418.0): cluster 3\n", + "Example n. 4903 = (96.0, 417.0): cluster 3\n", + "Example n. 4904 = (96.0, 416.0): cluster 3\n", + "Example n. 4905 = (96.0, 415.0): cluster 3\n", + "Example n. 4906 = (96.0, 414.0): cluster 3\n", + "Example n. 4907 = (97.0, 486.0): cluster 2\n", + "Example n. 4908 = (97.0, 485.0): cluster 2\n", + "Example n. 4909 = (97.0, 483.0): cluster 2\n", + "Example n. 4910 = (97.0, 482.0): cluster 2\n", + "Example n. 4911 = (97.0, 481.0): cluster 2\n", + "Example n. 4912 = (97.0, 480.0): cluster 2\n", + "Example n. 4913 = (97.0, 479.0): cluster 2\n", + "Example n. 4914 = (97.0, 478.0): cluster 2\n", + "Example n. 4915 = (97.0, 477.0): cluster 2\n", + "Example n. 4916 = (97.0, 476.0): cluster 2\n", + "Example n. 4917 = (97.0, 475.0): cluster 2\n", + "Example n. 4918 = (97.0, 474.0): cluster 2\n", + "Example n. 4919 = (97.0, 473.0): cluster 2\n", + "Example n. 4920 = (97.0, 472.0): cluster 2\n", + "Example n. 4921 = (97.0, 471.0): cluster 2\n", + "Example n. 4922 = (97.0, 470.0): cluster 2\n", + "Example n. 4923 = (97.0, 469.0): cluster 2\n", + "Example n. 4924 = (97.0, 468.0): cluster 2\n", + "Example n. 4925 = (97.0, 467.0): cluster 2\n", + "Example n. 4926 = (97.0, 466.0): cluster 2\n", + "Example n. 4927 = (97.0, 465.0): cluster 2\n", + "Example n. 4928 = (97.0, 464.0): cluster 2\n", + "Example n. 4929 = (97.0, 462.0): cluster 2\n", + "Example n. 4930 = (97.0, 461.0): cluster 2\n", + "Example n. 4931 = (97.0, 460.0): cluster 2\n", + "Example n. 4932 = (97.0, 444.0): cluster 2\n", + "Example n. 4933 = (97.0, 443.0): cluster 2\n", + "Example n. 4934 = (97.0, 442.0): cluster 2\n", + "Example n. 4935 = (97.0, 441.0): cluster 2\n", + "Example n. 4936 = (97.0, 440.0): cluster 2\n", + "Example n. 4937 = (97.0, 439.0): cluster 2\n", + "Example n. 4938 = (97.0, 438.0): cluster 2\n", + "Example n. 4939 = (97.0, 437.0): cluster 2\n", + "Example n. 4940 = (97.0, 436.0): cluster 2\n", + "Example n. 4941 = (97.0, 435.0): cluster 2\n", + "Example n. 4942 = (97.0, 434.0): cluster 2\n", + "Example n. 4943 = (97.0, 433.0): cluster 2\n", + "Example n. 4944 = (97.0, 425.0): cluster 2\n", + "Example n. 4945 = (97.0, 424.0): cluster 2\n", + "Example n. 4946 = (97.0, 423.0): cluster 2\n", + "Example n. 4947 = (97.0, 422.0): cluster 2\n", + "Example n. 4948 = (97.0, 421.0): cluster 2\n", + "Example n. 4949 = (97.0, 420.0): cluster 3\n", + "Example n. 4950 = (97.0, 419.0): cluster 3\n", + "Example n. 4951 = (97.0, 418.0): cluster 3\n", + "Example n. 4952 = (97.0, 417.0): cluster 3\n", + "Example n. 4953 = (97.0, 416.0): cluster 3\n", + "Example n. 4954 = (97.0, 415.0): cluster 3\n", + "Example n. 4955 = (97.0, 414.0): cluster 3\n", + "Example n. 4956 = (98.0, 486.0): cluster 2\n", + "Example n. 4957 = (98.0, 485.0): cluster 2\n", + "Example n. 4958 = (98.0, 484.0): cluster 2\n", + "Example n. 4959 = (98.0, 483.0): cluster 2\n", + "Example n. 4960 = (98.0, 482.0): cluster 2\n", + "Example n. 4961 = (98.0, 481.0): cluster 2\n", + "Example n. 4962 = (98.0, 480.0): cluster 2\n", + "Example n. 4963 = (98.0, 479.0): cluster 2\n", + "Example n. 4964 = (98.0, 478.0): cluster 2\n", + "Example n. 4965 = (98.0, 477.0): cluster 2\n", + "Example n. 4966 = (98.0, 476.0): cluster 2\n", + "Example n. 4967 = (98.0, 475.0): cluster 2\n", + "Example n. 4968 = (98.0, 474.0): cluster 2\n", + "Example n. 4969 = (98.0, 473.0): cluster 2\n", + "Example n. 4970 = (98.0, 472.0): cluster 2\n", + "Example n. 4971 = (98.0, 471.0): cluster 2\n", + "Example n. 4972 = (98.0, 470.0): cluster 2\n", + "Example n. 4973 = (98.0, 469.0): cluster 2\n", + "Example n. 4974 = (98.0, 468.0): cluster 2\n", + "Example n. 4975 = (98.0, 467.0): cluster 2\n", + "Example n. 4976 = (98.0, 466.0): cluster 2\n", + "Example n. 4977 = (98.0, 465.0): cluster 2\n", + "Example n. 4978 = (98.0, 464.0): cluster 2\n", + "Example n. 4979 = (98.0, 444.0): cluster 2\n", + "Example n. 4980 = (98.0, 443.0): cluster 2\n", + "Example n. 4981 = (98.0, 442.0): cluster 2\n", + "Example n. 4982 = (98.0, 441.0): cluster 2\n", + "Example n. 4983 = (98.0, 440.0): cluster 2\n", + "Example n. 4984 = (98.0, 439.0): cluster 2\n", + "Example n. 4985 = (98.0, 438.0): cluster 2\n", + "Example n. 4986 = (98.0, 437.0): cluster 2\n", + "Example n. 4987 = (98.0, 436.0): cluster 2\n", + "Example n. 4988 = (98.0, 435.0): cluster 2\n", + "Example n. 4989 = (98.0, 434.0): cluster 2\n", + "Example n. 4990 = (98.0, 433.0): cluster 2\n", + "Example n. 4991 = (99.0, 486.0): cluster 2\n", + "Example n. 4992 = (99.0, 485.0): cluster 2\n", + "Example n. 4993 = (99.0, 484.0): cluster 2\n", + "Example n. 4994 = (99.0, 483.0): cluster 2\n", + "Example n. 4995 = (99.0, 482.0): cluster 2\n", + "Example n. 4996 = (99.0, 481.0): cluster 2\n", + "Example n. 4997 = (99.0, 480.0): cluster 2\n", + "Example n. 4998 = (99.0, 479.0): cluster 2\n", + "Example n. 4999 = (99.0, 478.0): cluster 2\n", + "Example n. 5000 = (99.0, 477.0): cluster 2\n", + "Example n. 5001 = (99.0, 476.0): cluster 2\n", + "Example n. 5002 = (99.0, 475.0): cluster 2\n", + "Example n. 5003 = (99.0, 474.0): cluster 2\n", + "Example n. 5004 = (99.0, 473.0): cluster 2\n", + "Example n. 5005 = (99.0, 472.0): cluster 2\n", + "Example n. 5006 = (99.0, 471.0): cluster 2\n", + "Example n. 5007 = (99.0, 470.0): cluster 2\n", + "Example n. 5008 = (99.0, 469.0): cluster 2\n", + "Example n. 5009 = (99.0, 468.0): cluster 2\n", + "Example n. 5010 = (99.0, 467.0): cluster 2\n", + "Example n. 5011 = (99.0, 466.0): cluster 2\n", + "Example n. 5012 = (99.0, 465.0): cluster 2\n", + "Example n. 5013 = (99.0, 464.0): cluster 2\n", + "Example n. 5014 = (99.0, 444.0): cluster 2\n", + "Example n. 5015 = (99.0, 443.0): cluster 2\n", + "Example n. 5016 = (99.0, 442.0): cluster 2\n", + "Example n. 5017 = (99.0, 441.0): cluster 2\n", + "Example n. 5018 = (99.0, 440.0): cluster 2\n", + "Example n. 5019 = (99.0, 439.0): cluster 2\n", + "Example n. 5020 = (99.0, 438.0): cluster 2\n", + "Example n. 5021 = (99.0, 437.0): cluster 2\n", + "Example n. 5022 = (99.0, 436.0): cluster 2\n", + "Example n. 5023 = (99.0, 435.0): cluster 2\n", + "Example n. 5024 = (99.0, 434.0): cluster 2\n", + "Example n. 5025 = (99.0, 433.0): cluster 2\n", + "Example n. 5026 = (100.0, 486.0): cluster 2\n", + "Example n. 5027 = (100.0, 485.0): cluster 2\n", + "Example n. 5028 = (100.0, 484.0): cluster 2\n", + "Example n. 5029 = (100.0, 483.0): cluster 2\n", + "Example n. 5030 = (100.0, 482.0): cluster 2\n", + "Example n. 5031 = (100.0, 481.0): cluster 2\n", + "Example n. 5032 = (100.0, 480.0): cluster 2\n", + "Example n. 5033 = (100.0, 479.0): cluster 2\n", + "Example n. 5034 = (100.0, 478.0): cluster 2\n", + "Example n. 5035 = (100.0, 477.0): cluster 2\n", + "Example n. 5036 = (100.0, 476.0): cluster 2\n", + "Example n. 5037 = (100.0, 475.0): cluster 2\n", + "Example n. 5038 = (100.0, 474.0): cluster 2\n", + "Example n. 5039 = (100.0, 473.0): cluster 2\n", + "Example n. 5040 = (100.0, 472.0): cluster 2\n", + "Example n. 5041 = (100.0, 471.0): cluster 2\n", + "Example n. 5042 = (100.0, 470.0): cluster 2\n", + "Example n. 5043 = (100.0, 469.0): cluster 2\n", + "Example n. 5044 = (100.0, 468.0): cluster 2\n", + "Example n. 5045 = (100.0, 467.0): cluster 2\n", + "Example n. 5046 = (100.0, 466.0): cluster 2\n", + "Example n. 5047 = (100.0, 465.0): cluster 2\n", + "Example n. 5048 = (100.0, 464.0): cluster 2\n", + "Example n. 5049 = (100.0, 444.0): cluster 2\n", + "Example n. 5050 = (100.0, 443.0): cluster 2\n", + "Example n. 5051 = (100.0, 442.0): cluster 2\n", + "Example n. 5052 = (100.0, 441.0): cluster 2\n", + "Example n. 5053 = (100.0, 440.0): cluster 2\n", + "Example n. 5054 = (100.0, 439.0): cluster 2\n", + "Example n. 5055 = (100.0, 438.0): cluster 2\n", + "Example n. 5056 = (100.0, 437.0): cluster 2\n", + "Example n. 5057 = (100.0, 436.0): cluster 2\n", + "Example n. 5058 = (100.0, 435.0): cluster 2\n", + "Example n. 5059 = (100.0, 434.0): cluster 2\n", + "Example n. 5060 = (100.0, 433.0): cluster 2\n", + "Example n. 5061 = (101.0, 490.0): cluster 2\n", + "Example n. 5062 = (101.0, 489.0): cluster 2\n", + "Example n. 5063 = (101.0, 488.0): cluster 2\n", + "Example n. 5064 = (101.0, 487.0): cluster 2\n", + "Example n. 5065 = (101.0, 486.0): cluster 2\n", + "Example n. 5066 = (101.0, 485.0): cluster 2\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Example n. 5067 = (101.0, 484.0): cluster 2\n", + "Example n. 5068 = (101.0, 483.0): cluster 2\n", + "Example n. 5069 = (101.0, 482.0): cluster 2\n", + "Example n. 5070 = (101.0, 481.0): cluster 2\n", + "Example n. 5071 = (101.0, 480.0): cluster 2\n", + "Example n. 5072 = (101.0, 479.0): cluster 2\n", + "Example n. 5073 = (101.0, 478.0): cluster 2\n", + "Example n. 5074 = (101.0, 477.0): cluster 2\n", + "Example n. 5075 = (101.0, 476.0): cluster 2\n", + "Example n. 5076 = (101.0, 475.0): cluster 2\n", + "Example n. 5077 = (101.0, 474.0): cluster 2\n", + "Example n. 5078 = (101.0, 473.0): cluster 2\n", + "Example n. 5079 = (101.0, 472.0): cluster 2\n", + "Example n. 5080 = (101.0, 471.0): cluster 2\n", + "Example n. 5081 = (101.0, 470.0): cluster 2\n", + "Example n. 5082 = (101.0, 469.0): cluster 2\n", + "Example n. 5083 = (101.0, 468.0): cluster 2\n", + "Example n. 5084 = (101.0, 467.0): cluster 2\n", + "Example n. 5085 = (101.0, 466.0): cluster 2\n", + "Example n. 5086 = (101.0, 465.0): cluster 2\n", + "Example n. 5087 = (101.0, 464.0): cluster 2\n", + "Example n. 5088 = (101.0, 444.0): cluster 2\n", + "Example n. 5089 = (101.0, 443.0): cluster 2\n", + "Example n. 5090 = (101.0, 442.0): cluster 2\n", + "Example n. 5091 = (101.0, 441.0): cluster 2\n", + "Example n. 5092 = (101.0, 440.0): cluster 2\n", + "Example n. 5093 = (101.0, 439.0): cluster 2\n", + "Example n. 5094 = (101.0, 438.0): cluster 2\n", + "Example n. 5095 = (101.0, 437.0): cluster 2\n", + "Example n. 5096 = (101.0, 436.0): cluster 2\n", + "Example n. 5097 = (101.0, 435.0): cluster 2\n", + "Example n. 5098 = (101.0, 434.0): cluster 2\n", + "Example n. 5099 = (101.0, 433.0): cluster 2\n", + "Example n. 5100 = (101.0, 432.0): cluster 2\n", + "Example n. 5101 = (101.0, 431.0): cluster 2\n", + "Example n. 5102 = (101.0, 430.0): cluster 2\n", + "Example n. 5103 = (101.0, 429.0): cluster 2\n", + "Example n. 5104 = (102.0, 490.0): cluster 2\n", + "Example n. 5105 = (102.0, 489.0): cluster 2\n", + "Example n. 5106 = (102.0, 488.0): cluster 2\n", + "Example n. 5107 = (102.0, 487.0): cluster 2\n", + "Example n. 5108 = (102.0, 486.0): cluster 2\n", + "Example n. 5109 = (102.0, 485.0): cluster 2\n", + "Example n. 5110 = (102.0, 484.0): cluster 2\n", + "Example n. 5111 = (102.0, 483.0): cluster 2\n", + "Example n. 5112 = (102.0, 482.0): cluster 2\n", + "Example n. 5113 = (102.0, 481.0): cluster 2\n", + "Example n. 5114 = (102.0, 480.0): cluster 2\n", + "Example n. 5115 = (102.0, 479.0): cluster 2\n", + "Example n. 5116 = (102.0, 478.0): cluster 2\n", + "Example n. 5117 = (102.0, 477.0): cluster 2\n", + "Example n. 5118 = (102.0, 476.0): cluster 2\n", + "Example n. 5119 = (102.0, 475.0): cluster 2\n", + "Example n. 5120 = (102.0, 474.0): cluster 2\n", + "Example n. 5121 = (102.0, 473.0): cluster 2\n", + "Example n. 5122 = (102.0, 472.0): cluster 2\n", + "Example n. 5123 = (102.0, 471.0): cluster 2\n", + "Example n. 5124 = (102.0, 470.0): cluster 2\n", + "Example n. 5125 = (102.0, 469.0): cluster 2\n", + "Example n. 5126 = (102.0, 468.0): cluster 2\n", + "Example n. 5127 = (102.0, 444.0): cluster 2\n", + "Example n. 5128 = (102.0, 443.0): cluster 2\n", + "Example n. 5129 = (102.0, 442.0): cluster 2\n", + "Example n. 5130 = (102.0, 441.0): cluster 2\n", + "Example n. 5131 = (102.0, 440.0): cluster 2\n", + "Example n. 5132 = (102.0, 439.0): cluster 2\n", + "Example n. 5133 = (102.0, 438.0): cluster 2\n", + "Example n. 5134 = (102.0, 437.0): cluster 2\n", + "Example n. 5135 = (102.0, 436.0): cluster 2\n", + "Example n. 5136 = (102.0, 435.0): cluster 2\n", + "Example n. 5137 = (102.0, 434.0): cluster 2\n", + "Example n. 5138 = (102.0, 433.0): cluster 2\n", + "Example n. 5139 = (102.0, 432.0): cluster 2\n", + "Example n. 5140 = (102.0, 431.0): cluster 2\n", + "Example n. 5141 = (102.0, 430.0): cluster 2\n", + "Example n. 5142 = (102.0, 429.0): cluster 2\n", + "Example n. 5143 = (103.0, 490.0): cluster 2\n", + "Example n. 5144 = (103.0, 489.0): cluster 2\n", + "Example n. 5145 = (103.0, 488.0): cluster 2\n", + "Example n. 5146 = (103.0, 487.0): cluster 2\n", + "Example n. 5147 = (103.0, 486.0): cluster 2\n", + "Example n. 5148 = (103.0, 485.0): cluster 2\n", + "Example n. 5149 = (103.0, 484.0): cluster 2\n", + "Example n. 5150 = (103.0, 483.0): cluster 2\n", + "Example n. 5151 = (103.0, 482.0): cluster 2\n", + "Example n. 5152 = (103.0, 481.0): cluster 2\n", + "Example n. 5153 = (103.0, 480.0): cluster 2\n", + "Example n. 5154 = (103.0, 479.0): cluster 2\n", + "Example n. 5155 = (103.0, 478.0): cluster 2\n", + "Example n. 5156 = (103.0, 477.0): cluster 2\n", + "Example n. 5157 = (103.0, 476.0): cluster 2\n", + "Example n. 5158 = (103.0, 475.0): cluster 2\n", + "Example n. 5159 = (103.0, 474.0): cluster 2\n", + "Example n. 5160 = (103.0, 473.0): cluster 2\n", + "Example n. 5161 = (103.0, 472.0): cluster 2\n", + "Example n. 5162 = (103.0, 471.0): cluster 2\n", + "Example n. 5163 = (103.0, 470.0): cluster 2\n", + "Example n. 5164 = (103.0, 469.0): cluster 2\n", + "Example n. 5165 = (103.0, 468.0): cluster 2\n", + "Example n. 5166 = (103.0, 444.0): cluster 2\n", + "Example n. 5167 = (103.0, 443.0): cluster 2\n", + "Example n. 5168 = (103.0, 442.0): cluster 2\n", + "Example n. 5169 = (103.0, 441.0): cluster 2\n", + "Example n. 5170 = (103.0, 440.0): cluster 2\n", + "Example n. 5171 = (103.0, 439.0): cluster 2\n", + "Example n. 5172 = (103.0, 438.0): cluster 2\n", + "Example n. 5173 = (103.0, 437.0): cluster 2\n", + "Example n. 5174 = (103.0, 436.0): cluster 2\n", + "Example n. 5175 = (103.0, 435.0): cluster 2\n", + "Example n. 5176 = (103.0, 434.0): cluster 2\n", + "Example n. 5177 = (103.0, 433.0): cluster 2\n", + "Example n. 5178 = (103.0, 432.0): cluster 2\n", + "Example n. 5179 = (103.0, 431.0): cluster 2\n", + "Example n. 5180 = (103.0, 430.0): cluster 2\n", + "Example n. 5181 = (103.0, 429.0): cluster 2\n", + "Example n. 5182 = (104.0, 490.0): cluster 2\n", + "Example n. 5183 = (104.0, 489.0): cluster 2\n", + "Example n. 5184 = (104.0, 488.0): cluster 2\n", + "Example n. 5185 = (104.0, 487.0): cluster 2\n", + "Example n. 5186 = (104.0, 486.0): cluster 2\n", + "Example n. 5187 = (104.0, 485.0): cluster 2\n", + "Example n. 5188 = (104.0, 484.0): cluster 2\n", + "Example n. 5189 = (104.0, 483.0): cluster 2\n", + "Example n. 5190 = (104.0, 482.0): cluster 2\n", + "Example n. 5191 = (104.0, 481.0): cluster 2\n", + "Example n. 5192 = (104.0, 480.0): cluster 2\n", + "Example n. 5193 = (104.0, 479.0): cluster 2\n", + "Example n. 5194 = (104.0, 478.0): cluster 2\n", + "Example n. 5195 = (104.0, 477.0): cluster 2\n", + "Example n. 5196 = (104.0, 476.0): cluster 2\n", + "Example n. 5197 = (104.0, 475.0): cluster 2\n", + "Example n. 5198 = (104.0, 474.0): cluster 2\n", + "Example n. 5199 = (104.0, 473.0): cluster 2\n", + "Example n. 5200 = (104.0, 472.0): cluster 2\n", + "Example n. 5201 = (104.0, 471.0): cluster 2\n", + "Example n. 5202 = (104.0, 470.0): cluster 2\n", + "Example n. 5203 = (104.0, 469.0): cluster 2\n", + "Example n. 5204 = (104.0, 468.0): cluster 2\n", + "Example n. 5205 = (104.0, 444.0): cluster 2\n", + "Example n. 5206 = (104.0, 443.0): cluster 2\n", + "Example n. 5207 = (104.0, 442.0): cluster 2\n", + "Example n. 5208 = (104.0, 441.0): cluster 2\n", + "Example n. 5209 = (104.0, 440.0): cluster 2\n", + "Example n. 5210 = (104.0, 439.0): cluster 2\n", + "Example n. 5211 = (104.0, 438.0): cluster 2\n", + "Example n. 5212 = (104.0, 437.0): cluster 2\n", + "Example n. 5213 = (104.0, 436.0): cluster 2\n", + "Example n. 5214 = (104.0, 435.0): cluster 2\n", + "Example n. 5215 = (104.0, 434.0): cluster 2\n", + "Example n. 5216 = (104.0, 433.0): cluster 2\n", + "Example n. 5217 = (104.0, 432.0): cluster 2\n", + "Example n. 5218 = (104.0, 431.0): cluster 2\n", + "Example n. 5219 = (104.0, 430.0): cluster 2\n", + "Example n. 5220 = (104.0, 429.0): cluster 2\n", + "Example n. 5221 = (105.0, 490.0): cluster 2\n", + "Example n. 5222 = (105.0, 489.0): cluster 2\n", + "Example n. 5223 = (105.0, 488.0): cluster 2\n", + "Example n. 5224 = (105.0, 487.0): cluster 2\n", + "Example n. 5225 = (105.0, 486.0): cluster 2\n", + "Example n. 5226 = (105.0, 485.0): cluster 2\n", + "Example n. 5227 = (105.0, 484.0): cluster 2\n", + "Example n. 5228 = (105.0, 483.0): cluster 2\n", + "Example n. 5229 = (105.0, 482.0): cluster 2\n", + "Example n. 5230 = (105.0, 481.0): cluster 2\n", + "Example n. 5231 = (105.0, 480.0): cluster 2\n", + "Example n. 5232 = (105.0, 479.0): cluster 2\n", + "Example n. 5233 = (105.0, 478.0): cluster 2\n", + "Example n. 5234 = (105.0, 477.0): cluster 2\n", + "Example n. 5235 = (105.0, 476.0): cluster 2\n", + "Example n. 5236 = (105.0, 475.0): cluster 2\n", + "Example n. 5237 = (105.0, 474.0): cluster 2\n", + "Example n. 5238 = (105.0, 473.0): cluster 2\n", + "Example n. 5239 = (105.0, 472.0): cluster 2\n", + "Example n. 5240 = (105.0, 471.0): cluster 2\n", + "Example n. 5241 = (105.0, 470.0): cluster 2\n", + "Example n. 5242 = (105.0, 469.0): cluster 2\n", + "Example n. 5243 = (105.0, 468.0): cluster 2\n", + "Example n. 5244 = (105.0, 444.0): cluster 2\n", + "Example n. 5245 = (105.0, 443.0): cluster 2\n", + "Example n. 5246 = (105.0, 442.0): cluster 2\n", + "Example n. 5247 = (105.0, 441.0): cluster 2\n", + "Example n. 5248 = (105.0, 440.0): cluster 2\n", + "Example n. 5249 = (105.0, 439.0): cluster 2\n", + "Example n. 5250 = (105.0, 438.0): cluster 2\n", + "Example n. 5251 = (105.0, 437.0): cluster 2\n", + "Example n. 5252 = (105.0, 436.0): cluster 2\n", + "Example n. 5253 = (105.0, 435.0): cluster 2\n", + "Example n. 5254 = (105.0, 434.0): cluster 2\n", + "Example n. 5255 = (105.0, 433.0): cluster 2\n", + "Example n. 5256 = (105.0, 432.0): cluster 2\n", + "Example n. 5257 = (105.0, 431.0): cluster 2\n", + "Example n. 5258 = (105.0, 430.0): cluster 2\n", + "Example n. 5259 = (105.0, 429.0): cluster 2\n", + "Example n. 5260 = (106.0, 490.0): cluster 2\n", + "Example n. 5261 = (106.0, 489.0): cluster 2\n", + "Example n. 5262 = (106.0, 488.0): cluster 2\n", + "Example n. 5263 = (106.0, 487.0): cluster 2\n", + "Example n. 5264 = (106.0, 486.0): cluster 2\n", + "Example n. 5265 = (106.0, 485.0): cluster 2\n", + "Example n. 5266 = (106.0, 484.0): cluster 2\n", + "Example n. 5267 = (106.0, 483.0): cluster 2\n", + "Example n. 5268 = (106.0, 482.0): cluster 2\n", + "Example n. 5269 = (106.0, 481.0): cluster 2\n", + "Example n. 5270 = (106.0, 480.0): cluster 2\n", + "Example n. 5271 = (106.0, 479.0): cluster 2\n", + "Example n. 5272 = (106.0, 478.0): cluster 2\n", + "Example n. 5273 = (106.0, 477.0): cluster 2\n", + "Example n. 5274 = (106.0, 476.0): cluster 2\n", + "Example n. 5275 = (106.0, 475.0): cluster 2\n", + "Example n. 5276 = (106.0, 440.0): cluster 2\n", + "Example n. 5277 = (106.0, 439.0): cluster 2\n", + "Example n. 5278 = (106.0, 438.0): cluster 2\n", + "Example n. 5279 = (106.0, 437.0): cluster 2\n", + "Example n. 5280 = (106.0, 436.0): cluster 2\n", + "Example n. 5281 = (106.0, 435.0): cluster 2\n", + "Example n. 5282 = (106.0, 434.0): cluster 2\n", + "Example n. 5283 = (106.0, 433.0): cluster 2\n", + "Example n. 5284 = (106.0, 432.0): cluster 2\n", + "Example n. 5285 = (106.0, 431.0): cluster 2\n", + "Example n. 5286 = (106.0, 430.0): cluster 2\n", + "Example n. 5287 = (106.0, 429.0): cluster 2\n", + "Example n. 5288 = (107.0, 490.0): cluster 2\n", + "Example n. 5289 = (107.0, 489.0): cluster 2\n", + "Example n. 5290 = (107.0, 488.0): cluster 2\n", + "Example n. 5291 = (107.0, 487.0): cluster 2\n", + "Example n. 5292 = (107.0, 486.0): cluster 2\n", + "Example n. 5293 = (107.0, 485.0): cluster 2\n", + "Example n. 5294 = (107.0, 484.0): cluster 2\n", + "Example n. 5295 = (107.0, 483.0): cluster 2\n", + "Example n. 5296 = (107.0, 482.0): cluster 2\n", + "Example n. 5297 = (107.0, 481.0): cluster 2\n", + "Example n. 5298 = (107.0, 480.0): cluster 2\n", + "Example n. 5299 = (107.0, 479.0): cluster 2\n", + "Example n. 5300 = (107.0, 478.0): cluster 2\n", + "Example n. 5301 = (107.0, 477.0): cluster 2\n", + "Example n. 5302 = (107.0, 476.0): cluster 2\n", + "Example n. 5303 = (107.0, 475.0): cluster 2\n", + "Example n. 5304 = (107.0, 440.0): cluster 2\n", + "Example n. 5305 = (107.0, 439.0): cluster 2\n", + "Example n. 5306 = (107.0, 438.0): cluster 2\n", + "Example n. 5307 = (107.0, 437.0): cluster 2\n", + "Example n. 5308 = (107.0, 436.0): cluster 2\n", + "Example n. 5309 = (107.0, 435.0): cluster 2\n", + "Example n. 5310 = (107.0, 434.0): cluster 2\n", + "Example n. 5311 = (107.0, 433.0): cluster 2\n", + "Example n. 5312 = (107.0, 432.0): cluster 2\n", + "Example n. 5313 = (107.0, 431.0): cluster 2\n", + "Example n. 5314 = (107.0, 430.0): cluster 2\n", + "Example n. 5315 = (107.0, 429.0): cluster 2\n", + "Example n. 5316 = (108.0, 490.0): cluster 2\n", + "Example n. 5317 = (108.0, 489.0): cluster 2\n", + "Example n. 5318 = (108.0, 488.0): cluster 2\n", + "Example n. 5319 = (108.0, 487.0): cluster 2\n", + "Example n. 5320 = (108.0, 486.0): cluster 2\n", + "Example n. 5321 = (108.0, 485.0): cluster 2\n", + "Example n. 5322 = (108.0, 484.0): cluster 2\n", + "Example n. 5323 = (108.0, 483.0): cluster 2\n", + "Example n. 5324 = (108.0, 482.0): cluster 2\n", + "Example n. 5325 = (108.0, 481.0): cluster 2\n", + "Example n. 5326 = (108.0, 480.0): cluster 2\n", + "Example n. 5327 = (108.0, 479.0): cluster 2\n", + "Example n. 5328 = (108.0, 478.0): cluster 2\n", + "Example n. 5329 = (108.0, 477.0): cluster 2\n", + "Example n. 5330 = (108.0, 476.0): cluster 2\n", + "Example n. 5331 = (108.0, 475.0): cluster 2\n", + "Example n. 5332 = (108.0, 440.0): cluster 2\n", + "Example n. 5333 = (108.0, 439.0): cluster 2\n", + "Example n. 5334 = (108.0, 438.0): cluster 2\n", + "Example n. 5335 = (108.0, 437.0): cluster 2\n", + "Example n. 5336 = (108.0, 436.0): cluster 2\n", + "Example n. 5337 = (108.0, 435.0): cluster 2\n", + "Example n. 5338 = (108.0, 434.0): cluster 2\n", + "Example n. 5339 = (108.0, 433.0): cluster 2\n", + "Example n. 5340 = (108.0, 432.0): cluster 2\n", + "Example n. 5341 = (108.0, 431.0): cluster 2\n", + "Example n. 5342 = (108.0, 430.0): cluster 2\n", + "Example n. 5343 = (108.0, 429.0): cluster 2\n", + "Example n. 5344 = (109.0, 482.0): cluster 2\n", + "Example n. 5345 = (109.0, 481.0): cluster 2\n", + "Example n. 5346 = (109.0, 480.0): cluster 2\n", + "Example n. 5347 = (109.0, 479.0): cluster 2\n", + "Example n. 5348 = (109.0, 478.0): cluster 2\n", + "Example n. 5349 = (109.0, 477.0): cluster 2\n", + "Example n. 5350 = (109.0, 476.0): cluster 2\n", + "Example n. 5351 = (109.0, 475.0): cluster 2\n", + "Example n. 5352 = (109.0, 474.0): cluster 2\n", + "Example n. 5353 = (109.0, 473.0): cluster 2\n", + "Example n. 5354 = (109.0, 472.0): cluster 2\n", + "Example n. 5355 = (109.0, 471.0): cluster 2\n", + "Example n. 5356 = (109.0, 470.0): cluster 2\n", + "Example n. 5357 = (109.0, 469.0): cluster 2\n", + "Example n. 5358 = (109.0, 468.0): cluster 2\n", + "Example n. 5359 = (109.0, 440.0): cluster 2\n", + "Example n. 5360 = (109.0, 439.0): cluster 2\n", + "Example n. 5361 = (109.0, 438.0): cluster 2\n", + "Example n. 5362 = (109.0, 437.0): cluster 2\n", + "Example n. 5363 = (109.0, 436.0): cluster 2\n", + "Example n. 5364 = (109.0, 435.0): cluster 2\n", + "Example n. 5365 = (109.0, 434.0): cluster 2\n", + "Example n. 5366 = (109.0, 433.0): cluster 2\n", + "Example n. 5367 = (109.0, 432.0): cluster 2\n", + "Example n. 5368 = (109.0, 431.0): cluster 2\n", + "Example n. 5369 = (109.0, 430.0): cluster 2\n", + "Example n. 5370 = (109.0, 429.0): cluster 2\n", + "Example n. 5371 = (109.0, 428.0): cluster 2\n", + "Example n. 5372 = (109.0, 427.0): cluster 2\n", + "Example n. 5373 = (109.0, 426.0): cluster 2\n", + "Example n. 5374 = (109.0, 425.0): cluster 2\n", + "Example n. 5375 = (110.0, 482.0): cluster 2\n", + "Example n. 5376 = (110.0, 481.0): cluster 2\n", + "Example n. 5377 = (110.0, 480.0): cluster 2\n", + "Example n. 5378 = (110.0, 479.0): cluster 2\n", + "Example n. 5379 = (110.0, 478.0): cluster 2\n", + "Example n. 5380 = (110.0, 477.0): cluster 2\n", + "Example n. 5381 = (110.0, 476.0): cluster 2\n", + "Example n. 5382 = (110.0, 475.0): cluster 2\n", + "Example n. 5383 = (110.0, 474.0): cluster 2\n", + "Example n. 5384 = (110.0, 473.0): cluster 2\n", + "Example n. 5385 = (110.0, 472.0): cluster 2\n", + "Example n. 5386 = (110.0, 471.0): cluster 2\n", + "Example n. 5387 = (110.0, 470.0): cluster 2\n", + "Example n. 5388 = (110.0, 469.0): cluster 2\n", + "Example n. 5389 = (110.0, 468.0): cluster 2\n", + "Example n. 5390 = (110.0, 440.0): cluster 2\n", + "Example n. 5391 = (110.0, 439.0): cluster 2\n", + "Example n. 5392 = (110.0, 438.0): cluster 2\n", + "Example n. 5393 = (110.0, 437.0): cluster 2\n", + "Example n. 5394 = (110.0, 436.0): cluster 2\n", + "Example n. 5395 = (110.0, 435.0): cluster 2\n", + "Example n. 5396 = (110.0, 434.0): cluster 2\n", + "Example n. 5397 = (110.0, 433.0): cluster 2\n", + "Example n. 5398 = (110.0, 432.0): cluster 2\n", + "Example n. 5399 = (110.0, 431.0): cluster 2\n", + "Example n. 5400 = (110.0, 430.0): cluster 2\n", + "Example n. 5401 = (110.0, 429.0): cluster 2\n", + "Example n. 5402 = (110.0, 428.0): cluster 2\n", + "Example n. 5403 = (110.0, 427.0): cluster 2\n", + "Example n. 5404 = (110.0, 426.0): cluster 2\n", + "Example n. 5405 = (110.0, 425.0): cluster 2\n", + "Example n. 5406 = (111.0, 482.0): cluster 2\n", + "Example n. 5407 = (111.0, 481.0): cluster 2\n", + "Example n. 5408 = (111.0, 480.0): cluster 2\n", + "Example n. 5409 = (111.0, 479.0): cluster 2\n", + "Example n. 5410 = (111.0, 478.0): cluster 2\n", + "Example n. 5411 = (111.0, 477.0): cluster 2\n", + "Example n. 5412 = (111.0, 476.0): cluster 2\n", + "Example n. 5413 = (111.0, 475.0): cluster 2\n", + "Example n. 5414 = (111.0, 474.0): cluster 2\n", + "Example n. 5415 = (111.0, 473.0): cluster 2\n", + "Example n. 5416 = (111.0, 472.0): cluster 2\n", + "Example n. 5417 = (111.0, 471.0): cluster 2\n", + "Example n. 5418 = (111.0, 470.0): cluster 2\n", + "Example n. 5419 = (111.0, 469.0): cluster 2\n", + "Example n. 5420 = (111.0, 468.0): cluster 2\n", + "Example n. 5421 = (111.0, 440.0): cluster 2\n", + "Example n. 5422 = (111.0, 439.0): cluster 2\n", + "Example n. 5423 = (111.0, 438.0): cluster 2\n", + "Example n. 5424 = (111.0, 437.0): cluster 2\n", + "Example n. 5425 = (111.0, 436.0): cluster 2\n", + "Example n. 5426 = (111.0, 435.0): cluster 2\n", + "Example n. 5427 = (111.0, 434.0): cluster 2\n", + "Example n. 5428 = (111.0, 433.0): cluster 2\n", + "Example n. 5429 = (111.0, 432.0): cluster 2\n", + "Example n. 5430 = (111.0, 431.0): cluster 2\n", + "Example n. 5431 = (111.0, 430.0): cluster 2\n", + "Example n. 5432 = (111.0, 429.0): cluster 2\n", + "Example n. 5433 = (111.0, 428.0): cluster 2\n", + "Example n. 5434 = (111.0, 427.0): cluster 2\n", + "Example n. 5435 = (111.0, 426.0): cluster 2\n", + "Example n. 5436 = (111.0, 425.0): cluster 2\n", + "Example n. 5437 = (112.0, 482.0): cluster 2\n", + "Example n. 5438 = (112.0, 481.0): cluster 2\n", + "Example n. 5439 = (112.0, 480.0): cluster 2\n", + "Example n. 5440 = (112.0, 479.0): cluster 2\n", + "Example n. 5441 = (112.0, 478.0): cluster 2\n", + "Example n. 5442 = (112.0, 477.0): cluster 2\n", + "Example n. 5443 = (112.0, 476.0): cluster 2\n", + "Example n. 5444 = (112.0, 475.0): cluster 2\n", + "Example n. 5445 = (112.0, 474.0): cluster 2\n", + "Example n. 5446 = (112.0, 473.0): cluster 2\n", + "Example n. 5447 = (112.0, 472.0): cluster 2\n", + "Example n. 5448 = (112.0, 471.0): cluster 2\n", + "Example n. 5449 = (112.0, 470.0): cluster 2\n", + "Example n. 5450 = (112.0, 469.0): cluster 2\n", + "Example n. 5451 = (112.0, 468.0): cluster 2\n", + "Example n. 5452 = (112.0, 440.0): cluster 2\n", + "Example n. 5453 = (112.0, 439.0): cluster 2\n", + "Example n. 5454 = (112.0, 438.0): cluster 2\n", + "Example n. 5455 = (112.0, 437.0): cluster 2\n", + "Example n. 5456 = (112.0, 436.0): cluster 2\n", + "Example n. 5457 = (112.0, 435.0): cluster 2\n", + "Example n. 5458 = (112.0, 434.0): cluster 2\n", + "Example n. 5459 = (112.0, 433.0): cluster 2\n", + "Example n. 5460 = (112.0, 432.0): cluster 2\n", + "Example n. 5461 = (112.0, 431.0): cluster 2\n", + "Example n. 5462 = (112.0, 430.0): cluster 2\n", + "Example n. 5463 = (112.0, 429.0): cluster 2\n", + "Example n. 5464 = (112.0, 428.0): cluster 2\n", + "Example n. 5465 = (112.0, 427.0): cluster 2\n", + "Example n. 5466 = (112.0, 426.0): cluster 2\n", + "Example n. 5467 = (112.0, 425.0): cluster 2\n", + "Example n. 5468 = (113.0, 482.0): cluster 2\n", + "Example n. 5469 = (113.0, 481.0): cluster 2\n", + "Example n. 5470 = (113.0, 480.0): cluster 2\n", + "Example n. 5471 = (113.0, 479.0): cluster 2\n", + "Example n. 5472 = (113.0, 478.0): cluster 2\n", + "Example n. 5473 = (113.0, 477.0): cluster 2\n", + "Example n. 5474 = (113.0, 476.0): cluster 2\n", + "Example n. 5475 = (113.0, 475.0): cluster 2\n", + "Example n. 5476 = (113.0, 474.0): cluster 2\n", + "Example n. 5477 = (113.0, 473.0): cluster 2\n", + "Example n. 5478 = (113.0, 472.0): cluster 2\n", + "Example n. 5479 = (113.0, 471.0): cluster 2\n", + "Example n. 5480 = (113.0, 470.0): cluster 2\n", + "Example n. 5481 = (113.0, 469.0): cluster 2\n", + "Example n. 5482 = (113.0, 468.0): cluster 2\n", + "Example n. 5483 = (113.0, 467.0): cluster 2\n", + "Example n. 5484 = (113.0, 466.0): cluster 2\n", + "Example n. 5485 = (113.0, 465.0): cluster 2\n", + "Example n. 5486 = (113.0, 464.0): cluster 2\n", + "Example n. 5487 = (113.0, 436.0): cluster 2\n", + "Example n. 5488 = (113.0, 435.0): cluster 2\n", + "Example n. 5489 = (113.0, 434.0): cluster 2\n", + "Example n. 5490 = (113.0, 433.0): cluster 2\n", + "Example n. 5491 = (113.0, 432.0): cluster 2\n", + "Example n. 5492 = (113.0, 431.0): cluster 2\n", + "Example n. 5493 = (113.0, 430.0): cluster 2\n", + "Example n. 5494 = (113.0, 429.0): cluster 2\n", + "Example n. 5495 = (113.0, 428.0): cluster 2\n", + "Example n. 5496 = (113.0, 427.0): cluster 2\n", + "Example n. 5497 = (113.0, 426.0): cluster 2\n", + "Example n. 5498 = (113.0, 425.0): cluster 2\n", + "Example n. 5499 = (114.0, 482.0): cluster 2\n", + "Example n. 5500 = (114.0, 481.0): cluster 2\n", + "Example n. 5501 = (114.0, 480.0): cluster 2\n", + "Example n. 5502 = (114.0, 479.0): cluster 2\n", + "Example n. 5503 = (114.0, 478.0): cluster 2\n", + "Example n. 5504 = (114.0, 477.0): cluster 2\n", + "Example n. 5505 = (114.0, 476.0): cluster 2\n", + "Example n. 5506 = (114.0, 475.0): cluster 2\n", + "Example n. 5507 = (114.0, 474.0): cluster 2\n", + "Example n. 5508 = (114.0, 473.0): cluster 2\n", + "Example n. 5509 = (114.0, 472.0): cluster 2\n", + "Example n. 5510 = (114.0, 471.0): cluster 2\n", + "Example n. 5511 = (114.0, 470.0): cluster 2\n", + "Example n. 5512 = (114.0, 469.0): cluster 2\n", + "Example n. 5513 = (114.0, 468.0): cluster 2\n", + "Example n. 5514 = (114.0, 467.0): cluster 2\n", + "Example n. 5515 = (114.0, 466.0): cluster 2\n", + "Example n. 5516 = (114.0, 465.0): cluster 2\n", + "Example n. 5517 = (114.0, 464.0): cluster 2\n", + "Example n. 5518 = (114.0, 436.0): cluster 2\n", + "Example n. 5519 = (114.0, 435.0): cluster 2\n", + "Example n. 5520 = (114.0, 434.0): cluster 2\n", + "Example n. 5521 = (114.0, 433.0): cluster 2\n", + "Example n. 5522 = (114.0, 432.0): cluster 2\n", + "Example n. 5523 = (114.0, 431.0): cluster 2\n", + "Example n. 5524 = (114.0, 430.0): cluster 2\n", + "Example n. 5525 = (114.0, 429.0): cluster 2\n", + "Example n. 5526 = (114.0, 428.0): cluster 2\n", + "Example n. 5527 = (114.0, 427.0): cluster 2\n", + "Example n. 5528 = (114.0, 426.0): cluster 2\n", + "Example n. 5529 = (114.0, 425.0): cluster 2\n", + "Example n. 5530 = (115.0, 482.0): cluster 2\n", + "Example n. 5531 = (115.0, 481.0): cluster 2\n", + "Example n. 5532 = (115.0, 480.0): cluster 2\n", + "Example n. 5533 = (115.0, 479.0): cluster 2\n", + "Example n. 5534 = (115.0, 478.0): cluster 2\n", + "Example n. 5535 = (115.0, 477.0): cluster 2\n", + "Example n. 5536 = (115.0, 476.0): cluster 2\n", + "Example n. 5537 = (115.0, 475.0): cluster 2\n", + "Example n. 5538 = (115.0, 474.0): cluster 2\n", + "Example n. 5539 = (115.0, 473.0): cluster 2\n", + "Example n. 5540 = (115.0, 472.0): cluster 2\n", + "Example n. 5541 = (115.0, 471.0): cluster 2\n", + "Example n. 5542 = (115.0, 470.0): cluster 2\n", + "Example n. 5543 = (115.0, 469.0): cluster 2\n", + "Example n. 5544 = (115.0, 468.0): cluster 2\n", + "Example n. 5545 = (115.0, 467.0): cluster 2\n", + "Example n. 5546 = (115.0, 466.0): cluster 2\n", + "Example n. 5547 = (115.0, 465.0): cluster 2\n", + "Example n. 5548 = (115.0, 464.0): cluster 2\n", + "Example n. 5549 = (115.0, 436.0): cluster 2\n", + "Example n. 5550 = (115.0, 435.0): cluster 2\n", + "Example n. 5551 = (115.0, 434.0): cluster 2\n", + "Example n. 5552 = (115.0, 433.0): cluster 2\n", + "Example n. 5553 = (115.0, 432.0): cluster 2\n", + "Example n. 5554 = (115.0, 431.0): cluster 2\n", + "Example n. 5555 = (115.0, 430.0): cluster 2\n", + "Example n. 5556 = (115.0, 429.0): cluster 2\n", + "Example n. 5557 = (115.0, 428.0): cluster 2\n", + "Example n. 5558 = (115.0, 427.0): cluster 2\n", + "Example n. 5559 = (115.0, 426.0): cluster 2\n", + "Example n. 5560 = (115.0, 425.0): cluster 2\n", + "Example n. 5561 = (116.0, 482.0): cluster 2\n", + "Example n. 5562 = (116.0, 481.0): cluster 2\n", + "Example n. 5563 = (116.0, 480.0): cluster 2\n", + "Example n. 5564 = (116.0, 479.0): cluster 2\n", + "Example n. 5565 = (116.0, 478.0): cluster 2\n", + "Example n. 5566 = (116.0, 477.0): cluster 2\n", + "Example n. 5567 = (116.0, 476.0): cluster 2\n", + "Example n. 5568 = (116.0, 475.0): cluster 2\n", + "Example n. 5569 = (116.0, 474.0): cluster 2\n", + "Example n. 5570 = (116.0, 473.0): cluster 2\n", + "Example n. 5571 = (116.0, 472.0): cluster 2\n", + "Example n. 5572 = (116.0, 471.0): cluster 2\n", + "Example n. 5573 = (116.0, 470.0): cluster 2\n", + "Example n. 5574 = (116.0, 469.0): cluster 2\n", + "Example n. 5575 = (116.0, 468.0): cluster 2\n", + "Example n. 5576 = (116.0, 467.0): cluster 2\n", + "Example n. 5577 = (116.0, 466.0): cluster 2\n", + "Example n. 5578 = (116.0, 465.0): cluster 2\n", + "Example n. 5579 = (116.0, 464.0): cluster 2\n", + "Example n. 5580 = (116.0, 436.0): cluster 2\n", + "Example n. 5581 = (116.0, 435.0): cluster 2\n", + "Example n. 5582 = (116.0, 434.0): cluster 2\n", + "Example n. 5583 = (116.0, 433.0): cluster 2\n", + "Example n. 5584 = (116.0, 432.0): cluster 2\n", + "Example n. 5585 = (116.0, 431.0): cluster 2\n", + "Example n. 5586 = (116.0, 430.0): cluster 2\n", + "Example n. 5587 = (116.0, 429.0): cluster 2\n", + "Example n. 5588 = (116.0, 428.0): cluster 2\n", + "Example n. 5589 = (116.0, 427.0): cluster 2\n", + "Example n. 5590 = (116.0, 426.0): cluster 2\n", + "Example n. 5591 = (116.0, 425.0): cluster 2\n", + "Example n. 5592 = (117.0, 475.0): cluster 2\n", + "Example n. 5593 = (117.0, 474.0): cluster 2\n", + "Example n. 5594 = (117.0, 473.0): cluster 2\n", + "Example n. 5595 = (117.0, 472.0): cluster 2\n", + "Example n. 5596 = (117.0, 471.0): cluster 2\n", + "Example n. 5597 = (117.0, 470.0): cluster 2\n", + "Example n. 5598 = (117.0, 469.0): cluster 2\n", + "Example n. 5599 = (117.0, 468.0): cluster 2\n", + "Example n. 5600 = (117.0, 467.0): cluster 2\n", + "Example n. 5601 = (117.0, 466.0): cluster 2\n", + "Example n. 5602 = (117.0, 465.0): cluster 2\n", + "Example n. 5603 = (117.0, 464.0): cluster 2\n", + "Example n. 5604 = (117.0, 432.0): cluster 2\n", + "Example n. 5605 = (117.0, 431.0): cluster 2\n", + "Example n. 5606 = (117.0, 430.0): cluster 2\n", + "Example n. 5607 = (117.0, 429.0): cluster 2\n", + "Example n. 5608 = (117.0, 428.0): cluster 2\n", + "Example n. 5609 = (117.0, 427.0): cluster 2\n", + "Example n. 5610 = (117.0, 426.0): cluster 2\n", + "Example n. 5611 = (117.0, 425.0): cluster 2\n", + "Example n. 5612 = (117.0, 424.0): cluster 2\n", + "Example n. 5613 = (117.0, 423.0): cluster 2\n", + "Example n. 5614 = (117.0, 422.0): cluster 2\n", + "Example n. 5615 = (117.0, 421.0): cluster 2\n", + "Example n. 5616 = (118.0, 475.0): cluster 2\n", + "Example n. 5617 = (118.0, 474.0): cluster 2\n", + "Example n. 5618 = (118.0, 473.0): cluster 2\n", + "Example n. 5619 = (118.0, 472.0): cluster 2\n", + "Example n. 5620 = (118.0, 471.0): cluster 2\n", + "Example n. 5621 = (118.0, 470.0): cluster 2\n", + "Example n. 5622 = (118.0, 469.0): cluster 2\n", + "Example n. 5623 = (118.0, 468.0): cluster 2\n", + "Example n. 5624 = (118.0, 467.0): cluster 2\n", + "Example n. 5625 = (118.0, 466.0): cluster 2\n", + "Example n. 5626 = (118.0, 465.0): cluster 2\n", + "Example n. 5627 = (118.0, 464.0): cluster 2\n", + "Example n. 5628 = (118.0, 432.0): cluster 2\n", + "Example n. 5629 = (118.0, 431.0): cluster 2\n", + "Example n. 5630 = (118.0, 430.0): cluster 2\n", + "Example n. 5631 = (118.0, 429.0): cluster 2\n", + "Example n. 5632 = (118.0, 428.0): cluster 2\n", + "Example n. 5633 = (118.0, 427.0): cluster 2\n", + "Example n. 5634 = (118.0, 426.0): cluster 2\n", + "Example n. 5635 = (118.0, 425.0): cluster 2\n", + "Example n. 5636 = (118.0, 424.0): cluster 2\n", + "Example n. 5637 = (118.0, 423.0): cluster 2\n", + "Example n. 5638 = (118.0, 422.0): cluster 2\n", + "Example n. 5639 = (118.0, 421.0): cluster 2\n", + "Example n. 5640 = (119.0, 475.0): cluster 2\n", + "Example n. 5641 = (119.0, 474.0): cluster 2\n", + "Example n. 5642 = (119.0, 473.0): cluster 2\n", + "Example n. 5643 = (119.0, 472.0): cluster 2\n", + "Example n. 5644 = (119.0, 471.0): cluster 2\n", + "Example n. 5645 = (119.0, 470.0): cluster 2\n", + "Example n. 5646 = (119.0, 469.0): cluster 2\n", + "Example n. 5647 = (119.0, 468.0): cluster 2\n", + "Example n. 5648 = (119.0, 467.0): cluster 2\n", + "Example n. 5649 = (119.0, 466.0): cluster 2\n", + "Example n. 5650 = (119.0, 465.0): cluster 2\n", + "Example n. 5651 = (119.0, 464.0): cluster 2\n", + "Example n. 5652 = (119.0, 432.0): cluster 2\n", + "Example n. 5653 = (119.0, 431.0): cluster 2\n", + "Example n. 5654 = (119.0, 430.0): cluster 2\n", + "Example n. 5655 = (119.0, 429.0): cluster 2\n", + "Example n. 5656 = (119.0, 428.0): cluster 2\n", + "Example n. 5657 = (119.0, 427.0): cluster 2\n", + "Example n. 5658 = (119.0, 426.0): cluster 2\n", + "Example n. 5659 = (119.0, 425.0): cluster 2\n", + "Example n. 5660 = (119.0, 424.0): cluster 2\n", + "Example n. 5661 = (119.0, 423.0): cluster 2\n", + "Example n. 5662 = (119.0, 422.0): cluster 2\n", + "Example n. 5663 = (119.0, 421.0): cluster 2\n", + "Example n. 5664 = (120.0, 475.0): cluster 2\n", + "Example n. 5665 = (120.0, 474.0): cluster 2\n", + "Example n. 5666 = (120.0, 473.0): cluster 2\n", + "Example n. 5667 = (120.0, 472.0): cluster 2\n", + "Example n. 5668 = (120.0, 471.0): cluster 2\n", + "Example n. 5669 = (120.0, 470.0): cluster 2\n", + "Example n. 5670 = (120.0, 469.0): cluster 2\n", + "Example n. 5671 = (120.0, 468.0): cluster 2\n", + "Example n. 5672 = (120.0, 467.0): cluster 2\n", + "Example n. 5673 = (120.0, 466.0): cluster 2\n", + "Example n. 5674 = (120.0, 465.0): cluster 2\n", + "Example n. 5675 = (120.0, 464.0): cluster 2\n", + "Example n. 5676 = (120.0, 455.0): cluster 2\n", + "Example n. 5677 = (120.0, 454.0): cluster 2\n", + "Example n. 5678 = (120.0, 453.0): cluster 2\n", + "Example n. 5679 = (120.0, 452.0): cluster 2\n", + "Example n. 5680 = (120.0, 451.0): cluster 2\n", + "Example n. 5681 = (120.0, 450.0): cluster 2\n", + "Example n. 5682 = (120.0, 449.0): cluster 2\n", + "Example n. 5683 = (120.0, 448.0): cluster 2\n", + "Example n. 5684 = (120.0, 447.0): cluster 2\n", + "Example n. 5685 = (120.0, 446.0): cluster 2\n", + "Example n. 5686 = (120.0, 445.0): cluster 2\n", + "Example n. 5687 = (120.0, 444.0): cluster 2\n", + "Example n. 5688 = (120.0, 440.0): cluster 2\n", + "Example n. 5689 = (120.0, 439.0): cluster 2\n", + "Example n. 5690 = (120.0, 438.0): cluster 2\n", + "Example n. 5691 = (120.0, 437.0): cluster 2\n", + "Example n. 5692 = (120.0, 436.0): cluster 2\n", + "Example n. 5693 = (120.0, 435.0): cluster 2\n", + "Example n. 5694 = (120.0, 434.0): cluster 2\n", + "Example n. 5695 = (120.0, 433.0): cluster 2\n", + "Example n. 5696 = (120.0, 432.0): cluster 2\n", + "Example n. 5697 = (120.0, 431.0): cluster 2\n", + "Example n. 5698 = (120.0, 430.0): cluster 2\n", + "Example n. 5699 = (120.0, 429.0): cluster 2\n", + "Example n. 5700 = (120.0, 428.0): cluster 2\n", + "Example n. 5701 = (120.0, 427.0): cluster 2\n", + "Example n. 5702 = (120.0, 426.0): cluster 2\n", + "Example n. 5703 = (120.0, 425.0): cluster 2\n", + "Example n. 5704 = (120.0, 424.0): cluster 2\n", + "Example n. 5705 = (120.0, 423.0): cluster 2\n", + "Example n. 5706 = (120.0, 422.0): cluster 2\n", + "Example n. 5707 = (120.0, 421.0): cluster 2\n", + "Example n. 5708 = (121.0, 471.0): cluster 2\n", + "Example n. 5709 = (121.0, 470.0): cluster 2\n", + "Example n. 5710 = (121.0, 469.0): cluster 2\n", + "Example n. 5711 = (121.0, 468.0): cluster 2\n", + "Example n. 5712 = (121.0, 467.0): cluster 2\n", + "Example n. 5713 = (121.0, 466.0): cluster 2\n", + "Example n. 5714 = (121.0, 465.0): cluster 2\n", + "Example n. 5715 = (121.0, 464.0): cluster 2\n", + "Example n. 5716 = (121.0, 455.0): cluster 2\n", + "Example n. 5717 = (121.0, 454.0): cluster 2\n", + "Example n. 5718 = (121.0, 453.0): cluster 2\n", + "Example n. 5719 = (121.0, 452.0): cluster 2\n", + "Example n. 5720 = (121.0, 451.0): cluster 2\n", + "Example n. 5721 = (121.0, 450.0): cluster 2\n", + "Example n. 5722 = (121.0, 449.0): cluster 2\n", + "Example n. 5723 = (121.0, 448.0): cluster 2\n", + "Example n. 5724 = (121.0, 447.0): cluster 2\n", + "Example n. 5725 = (121.0, 446.0): cluster 2\n", + "Example n. 5726 = (121.0, 445.0): cluster 2\n", + "Example n. 5727 = (121.0, 444.0): cluster 2\n", + "Example n. 5728 = (121.0, 440.0): cluster 2\n", + "Example n. 5729 = (121.0, 439.0): cluster 2\n", + "Example n. 5730 = (121.0, 438.0): cluster 2\n", + "Example n. 5731 = (121.0, 437.0): cluster 2\n", + "Example n. 5732 = (121.0, 436.0): cluster 2\n", + "Example n. 5733 = (121.0, 435.0): cluster 2\n", + "Example n. 5734 = (121.0, 434.0): cluster 2\n", + "Example n. 5735 = (121.0, 433.0): cluster 2\n", + "Example n. 5736 = (121.0, 432.0): cluster 2\n", + "Example n. 5737 = (121.0, 431.0): cluster 2\n", + "Example n. 5738 = (121.0, 430.0): cluster 2\n", + "Example n. 5739 = (121.0, 429.0): cluster 2\n", + "Example n. 5740 = (121.0, 428.0): cluster 2\n", + "Example n. 5741 = (121.0, 427.0): cluster 2\n", + "Example n. 5742 = (121.0, 426.0): cluster 2\n", + "Example n. 5743 = (121.0, 425.0): cluster 2\n", + "Example n. 5744 = (121.0, 424.0): cluster 2\n", + "Example n. 5745 = (121.0, 423.0): cluster 2\n", + "Example n. 5746 = (121.0, 422.0): cluster 2\n", + "Example n. 5747 = (121.0, 421.0): cluster 2\n", + "Example n. 5748 = (122.0, 471.0): cluster 2\n", + "Example n. 5749 = (122.0, 470.0): cluster 2\n", + "Example n. 5750 = (122.0, 469.0): cluster 2\n", + "Example n. 5751 = (122.0, 468.0): cluster 2\n", + "Example n. 5752 = (122.0, 467.0): cluster 2\n", + "Example n. 5753 = (122.0, 466.0): cluster 2\n", + "Example n. 5754 = (122.0, 465.0): cluster 2\n", + "Example n. 5755 = (122.0, 464.0): cluster 2\n", + "Example n. 5756 = (122.0, 455.0): cluster 2\n", + "Example n. 5757 = (122.0, 454.0): cluster 2\n", + "Example n. 5758 = (122.0, 453.0): cluster 2\n", + "Example n. 5759 = (122.0, 452.0): cluster 2\n", + "Example n. 5760 = (122.0, 451.0): cluster 2\n", + "Example n. 5761 = (122.0, 450.0): cluster 2\n", + "Example n. 5762 = (122.0, 449.0): cluster 2\n", + "Example n. 5763 = (122.0, 448.0): cluster 2\n", + "Example n. 5764 = (122.0, 447.0): cluster 2\n", + "Example n. 5765 = (122.0, 446.0): cluster 2\n", + "Example n. 5766 = (122.0, 445.0): cluster 2\n", + "Example n. 5767 = (122.0, 444.0): cluster 2\n", + "Example n. 5768 = (122.0, 440.0): cluster 2\n", + "Example n. 5769 = (122.0, 439.0): cluster 2\n", + "Example n. 5770 = (122.0, 438.0): cluster 2\n", + "Example n. 5771 = (122.0, 437.0): cluster 2\n", + "Example n. 5772 = (122.0, 436.0): cluster 2\n", + "Example n. 5773 = (122.0, 435.0): cluster 2\n", + "Example n. 5774 = (122.0, 434.0): cluster 2\n", + "Example n. 5775 = (122.0, 433.0): cluster 2\n", + "Example n. 5776 = (122.0, 432.0): cluster 2\n", + "Example n. 5777 = (122.0, 431.0): cluster 2\n", + "Example n. 5778 = (122.0, 430.0): cluster 2\n", + "Example n. 5779 = (122.0, 429.0): cluster 2\n", + "Example n. 5780 = (122.0, 428.0): cluster 2\n", + "Example n. 5781 = (122.0, 427.0): cluster 2\n", + "Example n. 5782 = (122.0, 426.0): cluster 2\n", + "Example n. 5783 = (122.0, 425.0): cluster 2\n", + "Example n. 5784 = (122.0, 424.0): cluster 2\n", + "Example n. 5785 = (122.0, 423.0): cluster 2\n", + "Example n. 5786 = (122.0, 422.0): cluster 2\n", + "Example n. 5787 = (122.0, 421.0): cluster 2\n", + "Example n. 5788 = (123.0, 471.0): cluster 2\n", + "Example n. 5789 = (123.0, 470.0): cluster 2\n", + "Example n. 5790 = (123.0, 469.0): cluster 2\n", + "Example n. 5791 = (123.0, 468.0): cluster 2\n", + "Example n. 5792 = (123.0, 467.0): cluster 2\n", + "Example n. 5793 = (123.0, 466.0): cluster 2\n", + "Example n. 5794 = (123.0, 465.0): cluster 2\n", + "Example n. 5795 = (123.0, 464.0): cluster 2\n", + "Example n. 5796 = (123.0, 455.0): cluster 2\n", + "Example n. 5797 = (123.0, 454.0): cluster 2\n", + "Example n. 5798 = (123.0, 453.0): cluster 2\n", + "Example n. 5799 = (123.0, 452.0): cluster 2\n", + "Example n. 5800 = (123.0, 451.0): cluster 2\n", + "Example n. 5801 = (123.0, 450.0): cluster 2\n", + "Example n. 5802 = (123.0, 449.0): cluster 2\n", + "Example n. 5803 = (123.0, 448.0): cluster 2\n", + "Example n. 5804 = (123.0, 447.0): cluster 2\n", + "Example n. 5805 = (123.0, 446.0): cluster 2\n", + "Example n. 5806 = (123.0, 445.0): cluster 2\n", + "Example n. 5807 = (123.0, 444.0): cluster 2\n", + "Example n. 5808 = (123.0, 440.0): cluster 2\n", + "Example n. 5809 = (123.0, 439.0): cluster 2\n", + "Example n. 5810 = (123.0, 438.0): cluster 2\n", + "Example n. 5811 = (123.0, 437.0): cluster 2\n", + "Example n. 5812 = (123.0, 436.0): cluster 2\n", + "Example n. 5813 = (123.0, 435.0): cluster 2\n", + "Example n. 5814 = (123.0, 434.0): cluster 2\n", + "Example n. 5815 = (123.0, 433.0): cluster 2\n", + "Example n. 5816 = (123.0, 432.0): cluster 2\n", + "Example n. 5817 = (123.0, 431.0): cluster 2\n", + "Example n. 5818 = (123.0, 430.0): cluster 2\n", + "Example n. 5819 = (123.0, 429.0): cluster 2\n", + "Example n. 5820 = (123.0, 428.0): cluster 2\n", + "Example n. 5821 = (123.0, 427.0): cluster 2\n", + "Example n. 5822 = (123.0, 426.0): cluster 2\n", + "Example n. 5823 = (123.0, 425.0): cluster 2\n", + "Example n. 5824 = (123.0, 424.0): cluster 2\n", + "Example n. 5825 = (123.0, 423.0): cluster 2\n", + "Example n. 5826 = (123.0, 422.0): cluster 2\n", + "Example n. 5827 = (123.0, 421.0): cluster 2\n", + "Example n. 5828 = (124.0, 471.0): cluster 2\n", + "Example n. 5829 = (124.0, 470.0): cluster 2\n", + "Example n. 5830 = (124.0, 469.0): cluster 2\n", + "Example n. 5831 = (124.0, 468.0): cluster 2\n", + "Example n. 5832 = (124.0, 467.0): cluster 2\n", + "Example n. 5833 = (124.0, 466.0): cluster 2\n", + "Example n. 5834 = (124.0, 465.0): cluster 2\n", + "Example n. 5835 = (124.0, 464.0): cluster 2\n", + "Example n. 5836 = (124.0, 455.0): cluster 2\n", + "Example n. 5837 = (124.0, 454.0): cluster 2\n", + "Example n. 5838 = (124.0, 453.0): cluster 2\n", + "Example n. 5839 = (124.0, 452.0): cluster 2\n", + "Example n. 5840 = (124.0, 451.0): cluster 2\n", + "Example n. 5841 = (124.0, 450.0): cluster 2\n", + "Example n. 5842 = (124.0, 449.0): cluster 2\n", + "Example n. 5843 = (124.0, 448.0): cluster 2\n", + "Example n. 5844 = (124.0, 447.0): cluster 2\n", + "Example n. 5845 = (124.0, 446.0): cluster 2\n", + "Example n. 5846 = (124.0, 445.0): cluster 2\n", + "Example n. 5847 = (124.0, 444.0): cluster 2\n", + "Example n. 5848 = (124.0, 443.0): cluster 2\n", + "Example n. 5849 = (124.0, 442.0): cluster 2\n", + "Example n. 5850 = (124.0, 441.0): cluster 2\n", + "Example n. 5851 = (124.0, 440.0): cluster 2\n", + "Example n. 5852 = (124.0, 439.0): cluster 2\n", + "Example n. 5853 = (124.0, 438.0): cluster 2\n", + "Example n. 5854 = (124.0, 437.0): cluster 2\n", + "Example n. 5855 = (124.0, 436.0): cluster 2\n", + "Example n. 5856 = (124.0, 435.0): cluster 2\n", + "Example n. 5857 = (124.0, 434.0): cluster 2\n", + "Example n. 5858 = (124.0, 433.0): cluster 2\n", + "Example n. 5859 = (124.0, 432.0): cluster 2\n", + "Example n. 5860 = (124.0, 431.0): cluster 2\n", + "Example n. 5861 = (124.0, 430.0): cluster 2\n", + "Example n. 5862 = (124.0, 429.0): cluster 2\n", + "Example n. 5863 = (124.0, 428.0): cluster 2\n", + "Example n. 5864 = (124.0, 427.0): cluster 2\n", + "Example n. 5865 = (124.0, 426.0): cluster 2\n", + "Example n. 5866 = (124.0, 425.0): cluster 2\n", + "Example n. 5867 = (124.0, 424.0): cluster 2\n", + "Example n. 5868 = (124.0, 423.0): cluster 2\n", + "Example n. 5869 = (124.0, 422.0): cluster 2\n", + "Example n. 5870 = (124.0, 421.0): cluster 2\n", + "Example n. 5871 = (125.0, 455.0): cluster 2\n", + "Example n. 5872 = (125.0, 454.0): cluster 2\n", + "Example n. 5873 = (125.0, 453.0): cluster 2\n", + "Example n. 5874 = (125.0, 452.0): cluster 2\n", + "Example n. 5875 = (125.0, 451.0): cluster 2\n", + "Example n. 5876 = (125.0, 450.0): cluster 2\n", + "Example n. 5877 = (125.0, 449.0): cluster 2\n", + "Example n. 5878 = (125.0, 448.0): cluster 2\n", + "Example n. 5879 = (125.0, 447.0): cluster 2\n", + "Example n. 5880 = (125.0, 446.0): cluster 2\n", + "Example n. 5881 = (125.0, 445.0): cluster 2\n", + "Example n. 5882 = (125.0, 444.0): cluster 2\n", + "Example n. 5883 = (125.0, 443.0): cluster 2\n", + "Example n. 5884 = (125.0, 442.0): cluster 2\n", + "Example n. 5885 = (125.0, 441.0): cluster 2\n", + "Example n. 5886 = (125.0, 440.0): cluster 2\n", + "Example n. 5887 = (125.0, 439.0): cluster 2\n", + "Example n. 5888 = (125.0, 438.0): cluster 2\n", + "Example n. 5889 = (125.0, 437.0): cluster 2\n", + "Example n. 5890 = (125.0, 436.0): cluster 2\n", + "Example n. 5891 = (125.0, 435.0): cluster 2\n", + "Example n. 5892 = (125.0, 434.0): cluster 2\n", + "Example n. 5893 = (125.0, 433.0): cluster 2\n", + "Example n. 5894 = (125.0, 432.0): cluster 2\n", + "Example n. 5895 = (125.0, 431.0): cluster 2\n", + "Example n. 5896 = (125.0, 430.0): cluster 2\n", + "Example n. 5897 = (125.0, 429.0): cluster 2\n", + "Example n. 5898 = (125.0, 428.0): cluster 2\n", + "Example n. 5899 = (125.0, 427.0): cluster 2\n", + "Example n. 5900 = (125.0, 426.0): cluster 2\n", + "Example n. 5901 = (125.0, 425.0): cluster 2\n", + "Example n. 5902 = (125.0, 424.0): cluster 2\n", + "Example n. 5903 = (125.0, 423.0): cluster 2\n", + "Example n. 5904 = (125.0, 422.0): cluster 2\n", + "Example n. 5905 = (125.0, 421.0): cluster 2\n", + "Example n. 5906 = (126.0, 455.0): cluster 2\n", + "Example n. 5907 = (126.0, 454.0): cluster 2\n", + "Example n. 5908 = (126.0, 453.0): cluster 2\n", + "Example n. 5909 = (126.0, 452.0): cluster 2\n", + "Example n. 5910 = (126.0, 451.0): cluster 2\n", + "Example n. 5911 = (126.0, 450.0): cluster 2\n", + "Example n. 5912 = (126.0, 449.0): cluster 2\n", + "Example n. 5913 = (126.0, 448.0): cluster 2\n", + "Example n. 5914 = (126.0, 447.0): cluster 2\n", + "Example n. 5915 = (126.0, 446.0): cluster 2\n", + "Example n. 5916 = (126.0, 445.0): cluster 2\n", + "Example n. 5917 = (126.0, 444.0): cluster 2\n", + "Example n. 5918 = (126.0, 443.0): cluster 2\n", + "Example n. 5919 = (126.0, 442.0): cluster 2\n", + "Example n. 5920 = (126.0, 441.0): cluster 2\n", + "Example n. 5921 = (126.0, 440.0): cluster 2\n", + "Example n. 5922 = (126.0, 439.0): cluster 2\n", + "Example n. 5923 = (126.0, 438.0): cluster 2\n", + "Example n. 5924 = (126.0, 437.0): cluster 2\n", + "Example n. 5925 = (126.0, 436.0): cluster 2\n", + "Example n. 5926 = (126.0, 435.0): cluster 2\n", + "Example n. 5927 = (126.0, 434.0): cluster 2\n", + "Example n. 5928 = (126.0, 433.0): cluster 2\n", + "Example n. 5929 = (126.0, 432.0): cluster 2\n", + "Example n. 5930 = (126.0, 431.0): cluster 2\n", + "Example n. 5931 = (126.0, 430.0): cluster 2\n", + "Example n. 5932 = (126.0, 429.0): cluster 2\n", + "Example n. 5933 = (126.0, 428.0): cluster 2\n", + "Example n. 5934 = (126.0, 427.0): cluster 2\n", + "Example n. 5935 = (126.0, 426.0): cluster 2\n", + "Example n. 5936 = (126.0, 425.0): cluster 2\n", + "Example n. 5937 = (126.0, 424.0): cluster 2\n", + "Example n. 5938 = (126.0, 423.0): cluster 2\n", + "Example n. 5939 = (126.0, 422.0): cluster 2\n", + "Example n. 5940 = (126.0, 421.0): cluster 2\n", + "Example n. 5941 = (127.0, 455.0): cluster 2\n", + "Example n. 5942 = (127.0, 454.0): cluster 2\n", + "Example n. 5943 = (127.0, 453.0): cluster 2\n", + "Example n. 5944 = (127.0, 452.0): cluster 2\n", + "Example n. 5945 = (127.0, 451.0): cluster 2\n", + "Example n. 5946 = (127.0, 450.0): cluster 2\n", + "Example n. 5947 = (127.0, 449.0): cluster 2\n", + "Example n. 5948 = (127.0, 448.0): cluster 2\n", + "Example n. 5949 = (127.0, 447.0): cluster 2\n", + "Example n. 5950 = (127.0, 446.0): cluster 2\n", + "Example n. 5951 = (127.0, 445.0): cluster 2\n", + "Example n. 5952 = (127.0, 444.0): cluster 2\n", + "Example n. 5953 = (127.0, 443.0): cluster 2\n", + "Example n. 5954 = (127.0, 442.0): cluster 2\n", + "Example n. 5955 = (127.0, 441.0): cluster 2\n", + "Example n. 5956 = (127.0, 440.0): cluster 2\n", + "Example n. 5957 = (127.0, 439.0): cluster 2\n", + "Example n. 5958 = (127.0, 438.0): cluster 2\n", + "Example n. 5959 = (127.0, 437.0): cluster 2\n", + "Example n. 5960 = (127.0, 436.0): cluster 2\n", + "Example n. 5961 = (127.0, 435.0): cluster 2\n", + "Example n. 5962 = (127.0, 434.0): cluster 2\n", + "Example n. 5963 = (127.0, 433.0): cluster 2\n", + "Example n. 5964 = (127.0, 432.0): cluster 2\n", + "Example n. 5965 = (127.0, 431.0): cluster 2\n", + "Example n. 5966 = (127.0, 430.0): cluster 2\n", + "Example n. 5967 = (127.0, 429.0): cluster 2\n", + "Example n. 5968 = (127.0, 428.0): cluster 2\n", + "Example n. 5969 = (127.0, 427.0): cluster 2\n", + "Example n. 5970 = (127.0, 426.0): cluster 2\n", + "Example n. 5971 = (127.0, 425.0): cluster 2\n", + "Example n. 5972 = (127.0, 424.0): cluster 2\n", + "Example n. 5973 = (127.0, 423.0): cluster 2\n", + "Example n. 5974 = (127.0, 422.0): cluster 2\n", + "Example n. 5975 = (128.0, 455.0): cluster 2\n", + "Example n. 5976 = (128.0, 454.0): cluster 2\n", + "Example n. 5977 = (128.0, 453.0): cluster 2\n", + "Example n. 5978 = (128.0, 452.0): cluster 2\n", + "Example n. 5979 = (128.0, 451.0): cluster 2\n", + "Example n. 5980 = (128.0, 450.0): cluster 2\n", + "Example n. 5981 = (128.0, 449.0): cluster 2\n", + "Example n. 5982 = (128.0, 448.0): cluster 2\n", + "Example n. 5983 = (128.0, 447.0): cluster 2\n", + "Example n. 5984 = (128.0, 446.0): cluster 2\n", + "Example n. 5985 = (128.0, 445.0): cluster 2\n", + "Example n. 5986 = (128.0, 444.0): cluster 2\n", + "Example n. 5987 = (128.0, 443.0): cluster 2\n", + "Example n. 5988 = (128.0, 442.0): cluster 2\n", + "Example n. 5989 = (128.0, 441.0): cluster 2\n", + "Example n. 5990 = (128.0, 440.0): cluster 2\n", + "Example n. 5991 = (128.0, 439.0): cluster 2\n", + "Example n. 5992 = (128.0, 438.0): cluster 2\n", + "Example n. 5993 = (128.0, 437.0): cluster 2\n", + "Example n. 5994 = (128.0, 436.0): cluster 2\n", + "Example n. 5995 = (128.0, 435.0): cluster 2\n", + "Example n. 5996 = (128.0, 434.0): cluster 2\n", + "Example n. 5997 = (128.0, 433.0): cluster 2\n", + "Example n. 5998 = (128.0, 432.0): cluster 2\n", + "Example n. 5999 = (128.0, 431.0): cluster 2\n", + "Example n. 6000 = (128.0, 430.0): cluster 2\n", + "Example n. 6001 = (128.0, 429.0): cluster 2\n", + "Example n. 6002 = (128.0, 428.0): cluster 2\n", + "Example n. 6003 = (128.0, 427.0): cluster 2\n", + "Example n. 6004 = (128.0, 426.0): cluster 2\n", + "Example n. 6005 = (128.0, 425.0): cluster 2\n", + "Example n. 6006 = (128.0, 424.0): cluster 2\n", + "Example n. 6007 = (128.0, 423.0): cluster 2\n", + "Example n. 6008 = (128.0, 422.0): cluster 2\n", + "Example n. 6009 = (128.0, 421.0): cluster 2\n", + "Example n. 6010 = (129.0, 455.0): cluster 2\n", + "Example n. 6011 = (129.0, 454.0): cluster 2\n", + "Example n. 6012 = (129.0, 453.0): cluster 2\n", + "Example n. 6013 = (129.0, 452.0): cluster 2\n", + "Example n. 6014 = (129.0, 451.0): cluster 2\n", + "Example n. 6015 = (129.0, 450.0): cluster 2\n", + "Example n. 6016 = (129.0, 449.0): cluster 2\n", + "Example n. 6017 = (129.0, 448.0): cluster 2\n", + "Example n. 6018 = (129.0, 447.0): cluster 2\n", + "Example n. 6019 = (129.0, 446.0): cluster 2\n", + "Example n. 6020 = (129.0, 445.0): cluster 2\n", + "Example n. 6021 = (129.0, 444.0): cluster 2\n", + "Example n. 6022 = (129.0, 443.0): cluster 2\n", + "Example n. 6023 = (129.0, 442.0): cluster 2\n", + "Example n. 6024 = (129.0, 441.0): cluster 2\n", + "Example n. 6025 = (129.0, 440.0): cluster 2\n", + "Example n. 6026 = (129.0, 439.0): cluster 2\n", + "Example n. 6027 = (129.0, 438.0): cluster 2\n", + "Example n. 6028 = (129.0, 437.0): cluster 2\n", + "Example n. 6029 = (129.0, 436.0): cluster 2\n", + "Example n. 6030 = (129.0, 435.0): cluster 2\n", + "Example n. 6031 = (129.0, 434.0): cluster 2\n", + "Example n. 6032 = (129.0, 433.0): cluster 2\n", + "Example n. 6033 = (129.0, 432.0): cluster 2\n", + "Example n. 6034 = (129.0, 431.0): cluster 2\n", + "Example n. 6035 = (129.0, 430.0): cluster 2\n", + "Example n. 6036 = (129.0, 429.0): cluster 2\n", + "Example n. 6037 = (130.0, 455.0): cluster 2\n", + "Example n. 6038 = (130.0, 454.0): cluster 2\n", + "Example n. 6039 = (130.0, 453.0): cluster 2\n", + "Example n. 6040 = (130.0, 452.0): cluster 2\n", + "Example n. 6041 = (130.0, 451.0): cluster 2\n", + "Example n. 6042 = (130.0, 450.0): cluster 2\n", + "Example n. 6043 = (130.0, 449.0): cluster 2\n", + "Example n. 6044 = (130.0, 448.0): cluster 2\n", + "Example n. 6045 = (130.0, 447.0): cluster 2\n", + "Example n. 6046 = (130.0, 446.0): cluster 2\n", + "Example n. 6047 = (130.0, 445.0): cluster 2\n", + "Example n. 6048 = (130.0, 444.0): cluster 2\n", + "Example n. 6049 = (130.0, 443.0): cluster 2\n", + "Example n. 6050 = (130.0, 442.0): cluster 2\n", + "Example n. 6051 = (130.0, 441.0): cluster 2\n", + "Example n. 6052 = (130.0, 440.0): cluster 2\n", + "Example n. 6053 = (130.0, 439.0): cluster 2\n", + "Example n. 6054 = (130.0, 438.0): cluster 2\n", + "Example n. 6055 = (130.0, 437.0): cluster 2\n", + "Example n. 6056 = (130.0, 436.0): cluster 2\n", + "Example n. 6057 = (130.0, 435.0): cluster 2\n", + "Example n. 6058 = (130.0, 434.0): cluster 2\n", + "Example n. 6059 = (130.0, 433.0): cluster 2\n", + "Example n. 6060 = (130.0, 432.0): cluster 2\n", + "Example n. 6061 = (130.0, 431.0): cluster 2\n", + "Example n. 6062 = (130.0, 430.0): cluster 2\n", + "Example n. 6063 = (130.0, 429.0): cluster 2\n", + "Example n. 6064 = (131.0, 455.0): cluster 2\n", + "Example n. 6065 = (131.0, 454.0): cluster 2\n", + "Example n. 6066 = (131.0, 453.0): cluster 2\n", + "Example n. 6067 = (131.0, 452.0): cluster 2\n", + "Example n. 6068 = (131.0, 451.0): cluster 2\n", + "Example n. 6069 = (131.0, 450.0): cluster 2\n", + "Example n. 6070 = (131.0, 449.0): cluster 2\n", + "Example n. 6071 = (131.0, 448.0): cluster 2\n", + "Example n. 6072 = (131.0, 447.0): cluster 2\n", + "Example n. 6073 = (131.0, 446.0): cluster 2\n", + "Example n. 6074 = (131.0, 445.0): cluster 2\n", + "Example n. 6075 = (131.0, 444.0): cluster 2\n", + "Example n. 6076 = (131.0, 443.0): cluster 2\n", + "Example n. 6077 = (131.0, 442.0): cluster 2\n", + "Example n. 6078 = (131.0, 441.0): cluster 2\n", + "Example n. 6079 = (131.0, 440.0): cluster 2\n", + "Example n. 6080 = (131.0, 439.0): cluster 2\n", + "Example n. 6081 = (131.0, 438.0): cluster 2\n", + "Example n. 6082 = (131.0, 437.0): cluster 2\n", + "Example n. 6083 = (131.0, 436.0): cluster 2\n", + "Example n. 6084 = (131.0, 435.0): cluster 2\n", + "Example n. 6085 = (131.0, 434.0): cluster 2\n", + "Example n. 6086 = (131.0, 433.0): cluster 2\n", + "Example n. 6087 = (131.0, 432.0): cluster 2\n", + "Example n. 6088 = (131.0, 431.0): cluster 2\n", + "Example n. 6089 = (131.0, 430.0): cluster 2\n", + "Example n. 6090 = (131.0, 429.0): cluster 2\n", + "Example n. 6091 = (132.0, 455.0): cluster 2\n", + "Example n. 6092 = (132.0, 454.0): cluster 2\n", + "Example n. 6093 = (132.0, 453.0): cluster 2\n", + "Example n. 6094 = (132.0, 452.0): cluster 2\n", + "Example n. 6095 = (132.0, 451.0): cluster 2\n", + "Example n. 6096 = (132.0, 450.0): cluster 2\n", + "Example n. 6097 = (132.0, 449.0): cluster 2\n", + "Example n. 6098 = (132.0, 448.0): cluster 2\n", + "Example n. 6099 = (132.0, 447.0): cluster 2\n", + "Example n. 6100 = (132.0, 446.0): cluster 2\n", + "Example n. 6101 = (132.0, 445.0): cluster 2\n", + "Example n. 6102 = (132.0, 444.0): cluster 2\n", + "Example n. 6103 = (132.0, 443.0): cluster 2\n", + "Example n. 6104 = (132.0, 442.0): cluster 2\n", + "Example n. 6105 = (132.0, 441.0): cluster 2\n", + "Example n. 6106 = (132.0, 440.0): cluster 2\n", + "Example n. 6107 = (132.0, 439.0): cluster 2\n", + "Example n. 6108 = (132.0, 438.0): cluster 2\n", + "Example n. 6109 = (132.0, 437.0): cluster 2\n", + "Example n. 6110 = (132.0, 436.0): cluster 2\n", + "Example n. 6111 = (132.0, 435.0): cluster 2\n", + "Example n. 6112 = (132.0, 434.0): cluster 2\n", + "Example n. 6113 = (132.0, 433.0): cluster 2\n", + "Example n. 6114 = (132.0, 432.0): cluster 2\n", + "Example n. 6115 = (132.0, 431.0): cluster 2\n", + "Example n. 6116 = (132.0, 430.0): cluster 2\n", + "Example n. 6117 = (132.0, 429.0): cluster 2\n" + ] + } + ], + "source": [ + "data2,feature_names2,n_samples2,n_features2 = load_data(file_path, file_name2)\n", + "\n", + "np.random.seed(5)\n", + "\n", + "k=4\n", + "km2 = KMeans(n_clusters=k, random_state=0).fit(data2)\n", + "\n", + "for i in range(n_samples2):\n", + " print(f'Example n. {i} = ({(data2[i,0])}, {data2[i,1]}): cluster {km2.labels_[i]}')" + ] }, { "cell_type": "markdown", @@ -661,10 +6828,25 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 8, "metadata": {}, - "outputs": [], - "source": [] + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "plot(data2, n_samples2, km2, 'Clustered points in dataset n. 2')" + ] }, { "cell_type": "markdown", @@ -675,10 +6857,6174 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 9, "metadata": {}, - "outputs": [], - "source": [] + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Example n. 0 = (9.0, 480.0): cluster 0\n", + "Example n. 1 = (9.0, 479.0): cluster 0\n", + "Example n. 2 = (9.0, 478.0): cluster 0\n", + "Example n. 3 = (9.0, 477.0): cluster 0\n", + "Example n. 4 = (9.0, 467.0): cluster 0\n", + "Example n. 5 = (9.0, 466.0): cluster 0\n", + "Example n. 6 = (9.0, 465.0): cluster 0\n", + "Example n. 7 = (9.0, 464.0): cluster 0\n", + "Example n. 8 = (9.0, 392.0): cluster 0\n", + "Example n. 9 = (9.0, 391.0): cluster 0\n", + "Example n. 10 = (9.0, 390.0): cluster 0\n", + "Example n. 11 = (9.0, 350.0): cluster 0\n", + "Example n. 12 = (9.0, 349.0): cluster 0\n", + "Example n. 13 = (9.0, 348.0): cluster 0\n", + "Example n. 14 = (9.0, 347.0): cluster 0\n", + "Example n. 15 = (9.0, 290.0): cluster 0\n", + "Example n. 16 = (9.0, 289.0): cluster 0\n", + "Example n. 17 = (9.0, 288.0): cluster 0\n", + "Example n. 18 = (9.0, 287.0): cluster 0\n", + "Example n. 19 = (9.0, 240.0): cluster 0\n", + "Example n. 20 = (9.0, 239.0): cluster 0\n", + "Example n. 21 = (9.0, 238.0): cluster 0\n", + "Example n. 22 = (9.0, 231.0): cluster 0\n", + "Example n. 23 = (9.0, 230.0): cluster 0\n", + "Example n. 24 = (9.0, 229.0): cluster 0\n", + "Example n. 25 = (9.0, 228.0): cluster 0\n", + "Example n. 26 = (9.0, 129.0): cluster 0\n", + "Example n. 27 = (9.0, 128.0): cluster 0\n", + "Example n. 28 = (9.0, 127.0): cluster 0\n", + "Example n. 29 = (9.0, 126.0): cluster 0\n", + "Example n. 30 = (10.0, 480.0): cluster 0\n", + "Example n. 31 = (10.0, 479.0): cluster 0\n", + "Example n. 32 = (10.0, 478.0): cluster 0\n", + "Example n. 33 = (10.0, 477.0): cluster 0\n", + "Example n. 34 = (10.0, 476.0): cluster 0\n", + "Example n. 35 = (10.0, 468.0): cluster 0\n", + "Example n. 36 = (10.0, 467.0): cluster 0\n", + "Example n. 37 = (10.0, 466.0): cluster 0\n", + "Example n. 38 = (10.0, 465.0): cluster 0\n", + "Example n. 39 = (10.0, 464.0): cluster 0\n", + "Example n. 40 = (10.0, 463.0): cluster 0\n", + "Example n. 41 = (10.0, 393.0): cluster 0\n", + "Example n. 42 = (10.0, 392.0): cluster 0\n", + "Example n. 43 = (10.0, 391.0): cluster 0\n", + "Example n. 44 = (10.0, 390.0): cluster 0\n", + "Example n. 45 = (10.0, 389.0): cluster 0\n", + "Example n. 46 = (10.0, 351.0): cluster 0\n", + "Example n. 47 = (10.0, 350.0): cluster 0\n", + "Example n. 48 = (10.0, 349.0): cluster 0\n", + "Example n. 49 = (10.0, 348.0): cluster 0\n", + "Example n. 50 = (10.0, 347.0): cluster 0\n", + "Example n. 51 = (10.0, 346.0): cluster 0\n", + "Example n. 52 = (10.0, 291.0): cluster 0\n", + "Example n. 53 = (10.0, 290.0): cluster 0\n", + "Example n. 54 = (10.0, 289.0): cluster 0\n", + "Example n. 55 = (10.0, 288.0): cluster 0\n", + "Example n. 56 = (10.0, 287.0): cluster 0\n", + "Example n. 57 = (10.0, 286.0): cluster 0\n", + "Example n. 58 = (10.0, 242.0): cluster 0\n", + "Example n. 59 = (10.0, 241.0): cluster 0\n", + "Example n. 60 = (10.0, 240.0): cluster 0\n", + "Example n. 61 = (10.0, 239.0): cluster 0\n", + "Example n. 62 = (10.0, 238.0): cluster 0\n", + "Example n. 63 = (10.0, 237.0): cluster 0\n", + "Example n. 64 = (10.0, 236.0): cluster 0\n", + "Example n. 65 = (10.0, 232.0): cluster 0\n", + "Example n. 66 = (10.0, 231.0): cluster 0\n", + "Example n. 67 = (10.0, 230.0): cluster 0\n", + "Example n. 68 = (10.0, 229.0): cluster 0\n", + "Example n. 69 = (10.0, 228.0): cluster 0\n", + "Example n. 70 = (10.0, 227.0): cluster 0\n", + "Example n. 71 = (10.0, 226.0): cluster 0\n", + "Example n. 72 = (10.0, 225.0): cluster 0\n", + "Example n. 73 = (10.0, 129.0): cluster 0\n", + "Example n. 74 = (10.0, 128.0): cluster 0\n", + "Example n. 75 = (10.0, 127.0): cluster 0\n", + "Example n. 76 = (10.0, 126.0): cluster 0\n", + "Example n. 77 = (10.0, 125.0): cluster 0\n", + "Example n. 78 = (11.0, 481.0): cluster 0\n", + "Example n. 79 = (11.0, 480.0): cluster 0\n", + "Example n. 80 = (11.0, 479.0): cluster 0\n", + "Example n. 81 = (11.0, 478.0): cluster 0\n", + "Example n. 82 = (11.0, 477.0): cluster 0\n", + "Example n. 83 = (11.0, 476.0): cluster 0\n", + "Example n. 84 = (11.0, 468.0): cluster 0\n", + "Example n. 85 = (11.0, 467.0): cluster 0\n", + "Example n. 86 = (11.0, 466.0): cluster 0\n", + "Example n. 87 = (11.0, 465.0): cluster 0\n", + "Example n. 88 = (11.0, 464.0): cluster 0\n", + "Example n. 89 = (11.0, 463.0): cluster 0\n", + "Example n. 90 = (11.0, 452.0): cluster 0\n", + "Example n. 91 = (11.0, 451.0): cluster 0\n", + "Example n. 92 = (11.0, 408.0): cluster 0\n", + "Example n. 93 = (11.0, 407.0): cluster 0\n", + "Example n. 94 = (11.0, 406.0): cluster 0\n", + "Example n. 95 = (11.0, 393.0): cluster 0\n", + "Example n. 96 = (11.0, 392.0): cluster 0\n", + "Example n. 97 = (11.0, 391.0): cluster 0\n", + "Example n. 98 = (11.0, 390.0): cluster 0\n", + "Example n. 99 = (11.0, 389.0): cluster 0\n", + "Example n. 100 = (11.0, 388.0): cluster 0\n", + "Example n. 101 = (11.0, 351.0): cluster 0\n", + "Example n. 102 = (11.0, 350.0): cluster 0\n", + "Example n. 103 = (11.0, 349.0): cluster 0\n", + "Example n. 104 = (11.0, 348.0): cluster 0\n", + "Example n. 105 = (11.0, 347.0): cluster 0\n", + "Example n. 106 = (11.0, 346.0): cluster 0\n", + "Example n. 107 = (11.0, 291.0): cluster 0\n", + "Example n. 108 = (11.0, 290.0): cluster 0\n", + "Example n. 109 = (11.0, 289.0): cluster 0\n", + "Example n. 110 = (11.0, 288.0): cluster 0\n", + "Example n. 111 = (11.0, 287.0): cluster 0\n", + "Example n. 112 = (11.0, 286.0): cluster 0\n", + "Example n. 113 = (11.0, 242.0): cluster 0\n", + "Example n. 114 = (11.0, 241.0): cluster 0\n", + "Example n. 115 = (11.0, 240.0): cluster 0\n", + "Example n. 116 = (11.0, 239.0): cluster 0\n", + "Example n. 117 = (11.0, 238.0): cluster 0\n", + "Example n. 118 = (11.0, 237.0): cluster 0\n", + "Example n. 119 = (11.0, 236.0): cluster 0\n", + "Example n. 120 = (11.0, 235.0): cluster 0\n", + "Example n. 121 = (11.0, 232.0): cluster 0\n", + "Example n. 122 = (11.0, 231.0): cluster 0\n", + "Example n. 123 = (11.0, 230.0): cluster 0\n", + "Example n. 124 = (11.0, 229.0): cluster 0\n", + "Example n. 125 = (11.0, 228.0): cluster 0\n", + "Example n. 126 = (11.0, 227.0): cluster 0\n", + "Example n. 127 = (11.0, 226.0): cluster 0\n", + "Example n. 128 = (11.0, 225.0): cluster 0\n", + "Example n. 129 = (11.0, 224.0): cluster 0\n", + "Example n. 130 = (11.0, 130.0): cluster 0\n", + "Example n. 131 = (11.0, 129.0): cluster 0\n", + "Example n. 132 = (11.0, 128.0): cluster 0\n", + "Example n. 133 = (11.0, 127.0): cluster 0\n", + "Example n. 134 = (11.0, 126.0): cluster 0\n", + "Example n. 135 = (11.0, 125.0): cluster 0\n", + "Example n. 136 = (12.0, 480.0): cluster 0\n", + "Example n. 137 = (12.0, 479.0): cluster 0\n", + "Example n. 138 = (12.0, 478.0): cluster 0\n", + "Example n. 139 = (12.0, 477.0): cluster 0\n", + "Example n. 140 = (12.0, 476.0): cluster 0\n", + "Example n. 141 = (12.0, 468.0): cluster 0\n", + "Example n. 142 = (12.0, 467.0): cluster 0\n", + "Example n. 143 = (12.0, 466.0): cluster 0\n", + "Example n. 144 = (12.0, 465.0): cluster 0\n", + "Example n. 145 = (12.0, 464.0): cluster 0\n", + "Example n. 146 = (12.0, 463.0): cluster 0\n", + "Example n. 147 = (12.0, 453.0): cluster 0\n", + "Example n. 148 = (12.0, 452.0): cluster 0\n", + "Example n. 149 = (12.0, 451.0): cluster 0\n", + "Example n. 150 = (12.0, 450.0): cluster 0\n", + "Example n. 151 = (12.0, 409.0): cluster 0\n", + "Example n. 152 = (12.0, 408.0): cluster 0\n", + "Example n. 153 = (12.0, 407.0): cluster 0\n", + "Example n. 154 = (12.0, 406.0): cluster 0\n", + "Example n. 155 = (12.0, 405.0): cluster 0\n", + "Example n. 156 = (12.0, 393.0): cluster 0\n", + "Example n. 157 = (12.0, 392.0): cluster 0\n", + "Example n. 158 = (12.0, 391.0): cluster 0\n", + "Example n. 159 = (12.0, 390.0): cluster 0\n", + "Example n. 160 = (12.0, 389.0): cluster 0\n", + "Example n. 161 = (12.0, 388.0): cluster 0\n", + "Example n. 162 = (12.0, 351.0): cluster 0\n", + "Example n. 163 = (12.0, 350.0): cluster 0\n", + "Example n. 164 = (12.0, 349.0): cluster 0\n", + "Example n. 165 = (12.0, 348.0): cluster 0\n", + "Example n. 166 = (12.0, 347.0): cluster 0\n", + "Example n. 167 = (12.0, 346.0): cluster 0\n", + "Example n. 168 = (12.0, 291.0): cluster 0\n", + "Example n. 169 = (12.0, 290.0): cluster 0\n", + "Example n. 170 = (12.0, 289.0): cluster 0\n", + "Example n. 171 = (12.0, 288.0): cluster 0\n", + "Example n. 172 = (12.0, 287.0): cluster 0\n", + "Example n. 173 = (12.0, 256.0): cluster 0\n", + "Example n. 174 = (12.0, 255.0): cluster 0\n", + "Example n. 175 = (12.0, 254.0): cluster 0\n", + "Example n. 176 = (12.0, 252.0): cluster 0\n", + "Example n. 177 = (12.0, 251.0): cluster 0\n", + "Example n. 178 = (12.0, 250.0): cluster 0\n", + "Example n. 179 = (12.0, 249.0): cluster 0\n", + "Example n. 180 = (12.0, 242.0): cluster 0\n", + "Example n. 181 = (12.0, 241.0): cluster 0\n", + "Example n. 182 = (12.0, 240.0): cluster 0\n", + "Example n. 183 = (12.0, 239.0): cluster 0\n", + "Example n. 184 = (12.0, 238.0): cluster 0\n", + "Example n. 185 = (12.0, 237.0): cluster 0\n", + "Example n. 186 = (12.0, 236.0): cluster 0\n", + "Example n. 187 = (12.0, 235.0): cluster 0\n", + "Example n. 188 = (12.0, 232.0): cluster 0\n", + "Example n. 189 = (12.0, 231.0): cluster 0\n", + "Example n. 190 = (12.0, 230.0): cluster 0\n", + "Example n. 191 = (12.0, 229.0): cluster 0\n", + "Example n. 192 = (12.0, 228.0): cluster 0\n", + "Example n. 193 = (12.0, 227.0): cluster 0\n", + "Example n. 194 = (12.0, 226.0): cluster 0\n", + "Example n. 195 = (12.0, 225.0): cluster 0\n", + "Example n. 196 = (12.0, 224.0): cluster 0\n", + "Example n. 197 = (12.0, 223.0): cluster 0\n", + "Example n. 198 = (12.0, 206.0): cluster 0\n", + "Example n. 199 = (12.0, 205.0): cluster 0\n", + "Example n. 200 = (12.0, 204.0): cluster 0\n", + "Example n. 201 = (12.0, 200.0): cluster 0\n", + "Example n. 202 = (12.0, 129.0): cluster 0\n", + "Example n. 203 = (12.0, 128.0): cluster 0\n", + "Example n. 204 = (12.0, 127.0): cluster 0\n", + "Example n. 205 = (12.0, 126.0): cluster 0\n", + "Example n. 206 = (12.0, 125.0): cluster 0\n", + "Example n. 207 = (13.0, 480.0): cluster 0\n", + "Example n. 208 = (13.0, 479.0): cluster 0\n", + "Example n. 209 = (13.0, 478.0): cluster 0\n", + "Example n. 210 = (13.0, 477.0): cluster 0\n", + "Example n. 211 = (13.0, 467.0): cluster 0\n", + "Example n. 212 = (13.0, 466.0): cluster 0\n", + "Example n. 213 = (13.0, 465.0): cluster 0\n", + "Example n. 214 = (13.0, 464.0): cluster 0\n", + "Example n. 215 = (13.0, 454.0): cluster 0\n", + "Example n. 216 = (13.0, 453.0): cluster 0\n", + "Example n. 217 = (13.0, 452.0): cluster 0\n", + "Example n. 218 = (13.0, 451.0): cluster 0\n", + "Example n. 219 = (13.0, 450.0): cluster 0\n", + "Example n. 220 = (13.0, 449.0): cluster 0\n", + "Example n. 221 = (13.0, 409.0): cluster 0\n", + "Example n. 222 = (13.0, 408.0): cluster 0\n", + "Example n. 223 = (13.0, 407.0): cluster 0\n", + "Example n. 224 = (13.0, 406.0): cluster 0\n", + "Example n. 225 = (13.0, 405.0): cluster 0\n", + "Example n. 226 = (13.0, 393.0): cluster 0\n", + "Example n. 227 = (13.0, 392.0): cluster 0\n", + "Example n. 228 = (13.0, 391.0): cluster 0\n", + "Example n. 229 = (13.0, 390.0): cluster 0\n", + "Example n. 230 = (13.0, 389.0): cluster 0\n", + "Example n. 231 = (13.0, 350.0): cluster 0\n", + "Example n. 232 = (13.0, 349.0): cluster 0\n", + "Example n. 233 = (13.0, 348.0): cluster 0\n", + "Example n. 234 = (13.0, 347.0): cluster 0\n", + "Example n. 235 = (13.0, 290.0): cluster 0\n", + "Example n. 236 = (13.0, 289.0): cluster 0\n", + "Example n. 237 = (13.0, 288.0): cluster 0\n", + "Example n. 238 = (13.0, 287.0): cluster 0\n", + "Example n. 239 = (13.0, 266.0): cluster 0\n", + "Example n. 240 = (13.0, 265.0): cluster 0\n", + "Example n. 241 = (13.0, 264.0): cluster 0\n", + "Example n. 242 = (13.0, 263.0): cluster 0\n", + "Example n. 243 = (13.0, 257.0): cluster 0\n", + "Example n. 244 = (13.0, 256.0): cluster 0\n", + "Example n. 245 = (13.0, 255.0): cluster 0\n", + "Example n. 246 = (13.0, 254.0): cluster 0\n", + "Example n. 247 = (13.0, 253.0): cluster 0\n", + "Example n. 248 = (13.0, 252.0): cluster 0\n", + "Example n. 249 = (13.0, 251.0): cluster 0\n", + "Example n. 250 = (13.0, 250.0): cluster 0\n", + "Example n. 251 = (13.0, 249.0): cluster 0\n", + "Example n. 252 = (13.0, 248.0): cluster 0\n", + "Example n. 253 = (13.0, 241.0): cluster 0\n", + "Example n. 254 = (13.0, 240.0): cluster 0\n", + "Example n. 255 = (13.0, 239.0): cluster 0\n", + "Example n. 256 = (13.0, 238.0): cluster 0\n", + "Example n. 257 = (13.0, 237.0): cluster 0\n", + "Example n. 258 = (13.0, 236.0): cluster 0\n", + "Example n. 259 = (13.0, 235.0): cluster 0\n", + "Example n. 260 = (13.0, 231.0): cluster 0\n", + "Example n. 261 = (13.0, 230.0): cluster 0\n", + "Example n. 262 = (13.0, 229.0): cluster 0\n", + "Example n. 263 = (13.0, 228.0): cluster 0\n", + "Example n. 264 = (13.0, 227.0): cluster 0\n", + "Example n. 265 = (13.0, 226.0): cluster 0\n", + "Example n. 266 = (13.0, 225.0): cluster 0\n", + "Example n. 267 = (13.0, 224.0): cluster 0\n", + "Example n. 268 = (13.0, 223.0): cluster 0\n", + "Example n. 269 = (13.0, 222.0): cluster 0\n", + "Example n. 270 = (13.0, 207.0): cluster 0\n", + "Example n. 271 = (13.0, 206.0): cluster 0\n", + "Example n. 272 = (13.0, 205.0): cluster 0\n", + "Example n. 273 = (13.0, 204.0): cluster 0\n", + "Example n. 274 = (13.0, 203.0): cluster 0\n", + "Example n. 275 = (13.0, 202.0): cluster 0\n", + "Example n. 276 = (13.0, 201.0): cluster 0\n", + "Example n. 277 = (13.0, 200.0): cluster 0\n", + "Example n. 278 = (13.0, 199.0): cluster 0\n", + "Example n. 279 = (13.0, 198.0): cluster 0\n", + "Example n. 280 = (13.0, 129.0): cluster 0\n", + "Example n. 281 = (13.0, 128.0): cluster 0\n", + "Example n. 282 = (13.0, 127.0): cluster 0\n", + "Example n. 283 = (13.0, 126.0): cluster 0\n", + "Example n. 284 = (13.0, 121.0): cluster 0\n", + "Example n. 285 = (13.0, 120.0): cluster 0\n", + "Example n. 286 = (13.0, 119.0): cluster 0\n", + "Example n. 287 = (13.0, 118.0): cluster 0\n", + "Example n. 288 = (14.0, 454.0): cluster 0\n", + "Example n. 289 = (14.0, 453.0): cluster 0\n", + "Example n. 290 = (14.0, 452.0): cluster 0\n", + "Example n. 291 = (14.0, 451.0): cluster 0\n", + "Example n. 292 = (14.0, 450.0): cluster 0\n", + "Example n. 293 = (14.0, 449.0): cluster 0\n", + "Example n. 294 = (14.0, 409.0): cluster 0\n", + "Example n. 295 = (14.0, 408.0): cluster 0\n", + "Example n. 296 = (14.0, 407.0): cluster 0\n", + "Example n. 297 = (14.0, 406.0): cluster 0\n", + "Example n. 298 = (14.0, 405.0): cluster 0\n", + "Example n. 299 = (14.0, 392.0): cluster 0\n", + "Example n. 300 = (14.0, 391.0): cluster 0\n", + "Example n. 301 = (14.0, 390.0): cluster 0\n", + "Example n. 302 = (14.0, 267.0): cluster 0\n", + "Example n. 303 = (14.0, 266.0): cluster 0\n", + "Example n. 304 = (14.0, 265.0): cluster 0\n", + "Example n. 305 = (14.0, 264.0): cluster 0\n", + "Example n. 306 = (14.0, 263.0): cluster 0\n", + "Example n. 307 = (14.0, 262.0): cluster 0\n", + "Example n. 308 = (14.0, 257.0): cluster 0\n", + "Example n. 309 = (14.0, 256.0): cluster 0\n", + "Example n. 310 = (14.0, 255.0): cluster 0\n", + "Example n. 311 = (14.0, 254.0): cluster 0\n", + "Example n. 312 = (14.0, 253.0): cluster 0\n", + "Example n. 313 = (14.0, 252.0): cluster 0\n", + "Example n. 314 = (14.0, 251.0): cluster 0\n", + "Example n. 315 = (14.0, 250.0): cluster 0\n", + "Example n. 316 = (14.0, 249.0): cluster 0\n", + "Example n. 317 = (14.0, 248.0): cluster 0\n", + "Example n. 318 = (14.0, 244.0): cluster 0\n", + "Example n. 319 = (14.0, 243.0): cluster 0\n", + "Example n. 320 = (14.0, 242.0): cluster 0\n", + "Example n. 321 = (14.0, 241.0): cluster 0\n", + "Example n. 322 = (14.0, 240.0): cluster 0\n", + "Example n. 323 = (14.0, 239.0): cluster 0\n", + "Example n. 324 = (14.0, 238.0): cluster 0\n", + "Example n. 325 = (14.0, 237.0): cluster 0\n", + "Example n. 326 = (14.0, 236.0): cluster 0\n", + "Example n. 327 = (14.0, 228.0): cluster 0\n", + "Example n. 328 = (14.0, 227.0): cluster 0\n", + "Example n. 329 = (14.0, 226.0): cluster 0\n", + "Example n. 330 = (14.0, 225.0): cluster 0\n", + "Example n. 331 = (14.0, 224.0): cluster 0\n", + "Example n. 332 = (14.0, 223.0): cluster 0\n", + "Example n. 333 = (14.0, 222.0): cluster 0\n", + "Example n. 334 = (14.0, 219.0): cluster 0\n", + "Example n. 335 = (14.0, 218.0): cluster 0\n", + "Example n. 336 = (14.0, 207.0): cluster 0\n", + "Example n. 337 = (14.0, 206.0): cluster 0\n", + "Example n. 338 = (14.0, 205.0): cluster 0\n", + "Example n. 339 = (14.0, 204.0): cluster 0\n", + "Example n. 340 = (14.0, 203.0): cluster 0\n", + "Example n. 341 = (14.0, 202.0): cluster 0\n", + "Example n. 342 = (14.0, 201.0): cluster 0\n", + "Example n. 343 = (14.0, 200.0): cluster 0\n", + "Example n. 344 = (14.0, 199.0): cluster 0\n", + "Example n. 345 = (14.0, 198.0): cluster 0\n", + "Example n. 346 = (14.0, 195.0): cluster 0\n", + "Example n. 347 = (14.0, 194.0): cluster 0\n", + "Example n. 348 = (14.0, 127.0): cluster 0\n", + "Example n. 349 = (14.0, 122.0): cluster 0\n", + "Example n. 350 = (14.0, 121.0): cluster 0\n", + "Example n. 351 = (14.0, 120.0): cluster 0\n", + "Example n. 352 = (14.0, 119.0): cluster 0\n", + "Example n. 353 = (14.0, 118.0): cluster 0\n", + "Example n. 354 = (14.0, 117.0): cluster 0\n", + "Example n. 355 = (15.0, 453.0): cluster 0\n", + "Example n. 356 = (15.0, 452.0): cluster 0\n", + "Example n. 357 = (15.0, 451.0): cluster 0\n", + "Example n. 358 = (15.0, 450.0): cluster 0\n", + "Example n. 359 = (15.0, 449.0): cluster 0\n", + "Example n. 360 = (15.0, 409.0): cluster 0\n", + "Example n. 361 = (15.0, 408.0): cluster 0\n", + "Example n. 362 = (15.0, 407.0): cluster 0\n", + "Example n. 363 = (15.0, 406.0): cluster 0\n", + "Example n. 364 = (15.0, 405.0): cluster 0\n", + "Example n. 365 = (15.0, 310.0): cluster 0\n", + "Example n. 366 = (15.0, 309.0): cluster 0\n", + "Example n. 367 = (15.0, 276.0): cluster 0\n", + "Example n. 368 = (15.0, 275.0): cluster 0\n", + "Example n. 369 = (15.0, 268.0): cluster 0\n", + "Example n. 370 = (15.0, 267.0): cluster 0\n", + "Example n. 371 = (15.0, 266.0): cluster 0\n", + "Example n. 372 = (15.0, 265.0): cluster 0\n", + "Example n. 373 = (15.0, 264.0): cluster 0\n", + "Example n. 374 = (15.0, 263.0): cluster 0\n", + "Example n. 375 = (15.0, 262.0): cluster 0\n", + "Example n. 376 = (15.0, 261.0): cluster 0\n", + "Example n. 377 = (15.0, 258.0): cluster 0\n", + "Example n. 378 = (15.0, 257.0): cluster 0\n", + "Example n. 379 = (15.0, 256.0): cluster 0\n", + "Example n. 380 = (15.0, 255.0): cluster 0\n", + "Example n. 381 = (15.0, 254.0): cluster 0\n", + "Example n. 382 = (15.0, 253.0): cluster 0\n", + "Example n. 383 = (15.0, 252.0): cluster 0\n", + "Example n. 384 = (15.0, 251.0): cluster 0\n", + "Example n. 385 = (15.0, 250.0): cluster 0\n", + "Example n. 386 = (15.0, 249.0): cluster 0\n", + "Example n. 387 = (15.0, 248.0): cluster 0\n", + "Example n. 388 = (15.0, 245.0): cluster 0\n", + "Example n. 389 = (15.0, 244.0): cluster 0\n", + "Example n. 390 = (15.0, 243.0): cluster 0\n", + "Example n. 391 = (15.0, 242.0): cluster 0\n", + "Example n. 392 = (15.0, 241.0): cluster 0\n", + "Example n. 393 = (15.0, 238.0): cluster 0\n", + "Example n. 394 = (15.0, 237.0): cluster 0\n", + "Example n. 395 = (15.0, 227.0): cluster 0\n", + "Example n. 396 = (15.0, 226.0): cluster 0\n", + "Example n. 397 = (15.0, 225.0): cluster 0\n", + "Example n. 398 = (15.0, 224.0): cluster 0\n", + "Example n. 399 = (15.0, 223.0): cluster 0\n", + "Example n. 400 = (15.0, 222.0): cluster 0\n", + "Example n. 401 = (15.0, 221.0): cluster 0\n", + "Example n. 402 = (15.0, 220.0): cluster 0\n", + "Example n. 403 = (15.0, 219.0): cluster 0\n", + "Example n. 404 = (15.0, 218.0): cluster 0\n", + "Example n. 405 = (15.0, 217.0): cluster 0\n", + "Example n. 406 = (15.0, 216.0): cluster 0\n", + "Example n. 407 = (15.0, 207.0): cluster 0\n", + "Example n. 408 = (15.0, 206.0): cluster 0\n", + "Example n. 409 = (15.0, 205.0): cluster 0\n", + "Example n. 410 = (15.0, 204.0): cluster 0\n", + "Example n. 411 = (15.0, 203.0): cluster 0\n", + "Example n. 412 = (15.0, 202.0): cluster 0\n", + "Example n. 413 = (15.0, 201.0): cluster 0\n", + "Example n. 414 = (15.0, 200.0): cluster 0\n", + "Example n. 415 = (15.0, 199.0): cluster 0\n", + "Example n. 416 = (15.0, 198.0): cluster 0\n", + "Example n. 417 = (15.0, 197.0): cluster 0\n", + "Example n. 418 = (15.0, 196.0): cluster 0\n", + "Example n. 419 = (15.0, 195.0): cluster 0\n", + "Example n. 420 = (15.0, 194.0): cluster 0\n", + "Example n. 421 = (15.0, 193.0): cluster 0\n", + "Example n. 422 = (15.0, 192.0): cluster 0\n", + "Example n. 423 = (15.0, 122.0): cluster 0\n", + "Example n. 424 = (15.0, 121.0): cluster 0\n", + "Example n. 425 = (15.0, 120.0): cluster 0\n", + "Example n. 426 = (15.0, 119.0): cluster 0\n", + "Example n. 427 = (15.0, 118.0): cluster 0\n", + "Example n. 428 = (15.0, 117.0): cluster 0\n", + "Example n. 429 = (16.0, 453.0): cluster 0\n", + "Example n. 430 = (16.0, 452.0): cluster 0\n", + "Example n. 431 = (16.0, 451.0): cluster 0\n", + "Example n. 432 = (16.0, 450.0): cluster 0\n", + "Example n. 433 = (16.0, 407.0): cluster 0\n", + "Example n. 434 = (16.0, 311.0): cluster 0\n", + "Example n. 435 = (16.0, 310.0): cluster 0\n", + "Example n. 436 = (16.0, 309.0): cluster 0\n", + "Example n. 437 = (16.0, 308.0): cluster 0\n", + "Example n. 438 = (16.0, 278.0): cluster 0\n", + "Example n. 439 = (16.0, 277.0): cluster 0\n", + "Example n. 440 = (16.0, 276.0): cluster 0\n", + "Example n. 441 = (16.0, 275.0): cluster 0\n", + "Example n. 442 = (16.0, 274.0): cluster 0\n", + "Example n. 443 = (16.0, 268.0): cluster 0\n", + "Example n. 444 = (16.0, 267.0): cluster 0\n", + "Example n. 445 = (16.0, 266.0): cluster 0\n", + "Example n. 446 = (16.0, 265.0): cluster 0\n", + "Example n. 447 = (16.0, 264.0): cluster 0\n", + "Example n. 448 = (16.0, 263.0): cluster 0\n", + "Example n. 449 = (16.0, 262.0): cluster 0\n", + "Example n. 450 = (16.0, 261.0): cluster 0\n", + "Example n. 451 = (16.0, 258.0): cluster 0\n", + "Example n. 452 = (16.0, 257.0): cluster 0\n", + "Example n. 453 = (16.0, 256.0): cluster 0\n", + "Example n. 454 = (16.0, 255.0): cluster 0\n", + "Example n. 455 = (16.0, 254.0): cluster 0\n", + "Example n. 456 = (16.0, 253.0): cluster 0\n", + "Example n. 457 = (16.0, 252.0): cluster 0\n", + "Example n. 458 = (16.0, 251.0): cluster 0\n", + "Example n. 459 = (16.0, 250.0): cluster 0\n", + "Example n. 460 = (16.0, 249.0): cluster 0\n", + "Example n. 461 = (16.0, 246.0): cluster 0\n", + "Example n. 462 = (16.0, 245.0): cluster 0\n", + "Example n. 463 = (16.0, 244.0): cluster 0\n", + "Example n. 464 = (16.0, 243.0): cluster 0\n", + "Example n. 465 = (16.0, 242.0): cluster 0\n", + "Example n. 466 = (16.0, 241.0): cluster 0\n", + "Example n. 467 = (16.0, 228.0): cluster 0\n", + "Example n. 468 = (16.0, 227.0): cluster 0\n", + "Example n. 469 = (16.0, 226.0): cluster 0\n", + "Example n. 470 = (16.0, 225.0): cluster 0\n", + "Example n. 471 = (16.0, 224.0): cluster 0\n", + "Example n. 472 = (16.0, 223.0): cluster 0\n", + "Example n. 473 = (16.0, 222.0): cluster 0\n", + "Example n. 474 = (16.0, 221.0): cluster 0\n", + "Example n. 475 = (16.0, 220.0): cluster 0\n", + "Example n. 476 = (16.0, 219.0): cluster 0\n", + "Example n. 477 = (16.0, 218.0): cluster 0\n", + "Example n. 478 = (16.0, 217.0): cluster 0\n", + "Example n. 479 = (16.0, 216.0): cluster 0\n", + "Example n. 480 = (16.0, 207.0): cluster 0\n", + "Example n. 481 = (16.0, 206.0): cluster 0\n", + "Example n. 482 = (16.0, 205.0): cluster 0\n", + "Example n. 483 = (16.0, 204.0): cluster 0\n", + "Example n. 484 = (16.0, 203.0): cluster 0\n", + "Example n. 485 = (16.0, 202.0): cluster 0\n", + "Example n. 486 = (16.0, 201.0): cluster 0\n", + "Example n. 487 = (16.0, 200.0): cluster 0\n", + "Example n. 488 = (16.0, 199.0): cluster 0\n", + "Example n. 489 = (16.0, 198.0): cluster 0\n", + "Example n. 490 = (16.0, 197.0): cluster 0\n", + "Example n. 491 = (16.0, 196.0): cluster 0\n", + "Example n. 492 = (16.0, 195.0): cluster 0\n", + "Example n. 493 = (16.0, 194.0): cluster 0\n", + "Example n. 494 = (16.0, 193.0): cluster 0\n", + "Example n. 495 = (16.0, 192.0): cluster 0\n", + "Example n. 496 = (16.0, 122.0): cluster 0\n", + "Example n. 497 = (16.0, 121.0): cluster 0\n", + "Example n. 498 = (16.0, 120.0): cluster 0\n", + "Example n. 499 = (16.0, 119.0): cluster 0\n", + "Example n. 500 = (16.0, 118.0): cluster 0\n", + "Example n. 501 = (16.0, 117.0): cluster 0\n", + "Example n. 502 = (17.0, 500.0): cluster 0\n", + "Example n. 503 = (17.0, 499.0): cluster 0\n", + "Example n. 504 = (17.0, 498.0): cluster 0\n", + "Example n. 505 = (17.0, 312.0): cluster 0\n", + "Example n. 506 = (17.0, 311.0): cluster 0\n", + "Example n. 507 = (17.0, 310.0): cluster 0\n", + "Example n. 508 = (17.0, 309.0): cluster 0\n", + "Example n. 509 = (17.0, 308.0): cluster 0\n", + "Example n. 510 = (17.0, 307.0): cluster 0\n", + "Example n. 511 = (17.0, 279.0): cluster 0\n", + "Example n. 512 = (17.0, 278.0): cluster 0\n", + "Example n. 513 = (17.0, 277.0): cluster 0\n", + "Example n. 514 = (17.0, 276.0): cluster 0\n", + "Example n. 515 = (17.0, 275.0): cluster 0\n", + "Example n. 516 = (17.0, 274.0): cluster 0\n", + "Example n. 517 = (17.0, 273.0): cluster 0\n", + "Example n. 518 = (17.0, 270.0): cluster 0\n", + "Example n. 519 = (17.0, 269.0): cluster 0\n", + "Example n. 520 = (17.0, 268.0): cluster 0\n", + "Example n. 521 = (17.0, 267.0): cluster 0\n", + "Example n. 522 = (17.0, 266.0): cluster 0\n", + "Example n. 523 = (17.0, 265.0): cluster 0\n", + "Example n. 524 = (17.0, 264.0): cluster 0\n", + "Example n. 525 = (17.0, 263.0): cluster 0\n", + "Example n. 526 = (17.0, 262.0): cluster 0\n", + "Example n. 527 = (17.0, 261.0): cluster 0\n", + "Example n. 528 = (17.0, 260.0): cluster 0\n", + "Example n. 529 = (17.0, 259.0): cluster 0\n", + "Example n. 530 = (17.0, 258.0): cluster 0\n", + "Example n. 531 = (17.0, 257.0): cluster 0\n", + "Example n. 532 = (17.0, 256.0): cluster 0\n", + "Example n. 533 = (17.0, 255.0): cluster 0\n", + "Example n. 534 = (17.0, 254.0): cluster 0\n", + "Example n. 535 = (17.0, 246.0): cluster 0\n", + "Example n. 536 = (17.0, 245.0): cluster 0\n", + "Example n. 537 = (17.0, 244.0): cluster 0\n", + "Example n. 538 = (17.0, 243.0): cluster 0\n", + "Example n. 539 = (17.0, 242.0): cluster 0\n", + "Example n. 540 = (17.0, 241.0): cluster 0\n", + "Example n. 541 = (17.0, 228.0): cluster 0\n", + "Example n. 542 = (17.0, 227.0): cluster 0\n", + "Example n. 543 = (17.0, 226.0): cluster 0\n", + "Example n. 544 = (17.0, 225.0): cluster 0\n", + "Example n. 545 = (17.0, 224.0): cluster 0\n", + "Example n. 546 = (17.0, 223.0): cluster 0\n", + "Example n. 547 = (17.0, 222.0): cluster 0\n", + "Example n. 548 = (17.0, 221.0): cluster 0\n", + "Example n. 549 = (17.0, 220.0): cluster 0\n", + "Example n. 550 = (17.0, 219.0): cluster 0\n", + "Example n. 551 = (17.0, 218.0): cluster 0\n", + "Example n. 552 = (17.0, 217.0): cluster 0\n", + "Example n. 553 = (17.0, 216.0): cluster 0\n", + "Example n. 554 = (17.0, 207.0): cluster 0\n", + "Example n. 555 = (17.0, 206.0): cluster 0\n", + "Example n. 556 = (17.0, 205.0): cluster 0\n", + "Example n. 557 = (17.0, 204.0): cluster 0\n", + "Example n. 558 = (17.0, 203.0): cluster 0\n", + "Example n. 559 = (17.0, 202.0): cluster 0\n", + "Example n. 560 = (17.0, 201.0): cluster 0\n", + "Example n. 561 = (17.0, 200.0): cluster 0\n", + "Example n. 562 = (17.0, 199.0): cluster 0\n", + "Example n. 563 = (17.0, 198.0): cluster 0\n", + "Example n. 564 = (17.0, 197.0): cluster 0\n", + "Example n. 565 = (17.0, 196.0): cluster 0\n", + "Example n. 566 = (17.0, 195.0): cluster 0\n", + "Example n. 567 = (17.0, 194.0): cluster 0\n", + "Example n. 568 = (17.0, 193.0): cluster 0\n", + "Example n. 569 = (17.0, 192.0): cluster 0\n", + "Example n. 570 = (17.0, 148.0): cluster 0\n", + "Example n. 571 = (17.0, 147.0): cluster 0\n", + "Example n. 572 = (17.0, 142.0): cluster 0\n", + "Example n. 573 = (17.0, 132.0): cluster 0\n", + "Example n. 574 = (17.0, 131.0): cluster 0\n", + "Example n. 575 = (17.0, 130.0): cluster 0\n", + "Example n. 576 = (17.0, 121.0): cluster 0\n", + "Example n. 577 = (17.0, 120.0): cluster 0\n", + "Example n. 578 = (17.0, 119.0): cluster 0\n", + "Example n. 579 = (17.0, 118.0): cluster 0\n", + "Example n. 580 = (17.0, 92.0): cluster 0\n", + "Example n. 581 = (18.0, 501.0): cluster 0\n", + "Example n. 582 = (18.0, 500.0): cluster 0\n", + "Example n. 583 = (18.0, 499.0): cluster 0\n", + "Example n. 584 = (18.0, 498.0): cluster 0\n", + "Example n. 585 = (18.0, 497.0): cluster 0\n", + "Example n. 586 = (18.0, 312.0): cluster 0\n", + "Example n. 587 = (18.0, 311.0): cluster 0\n", + "Example n. 588 = (18.0, 310.0): cluster 0\n", + "Example n. 589 = (18.0, 309.0): cluster 0\n", + "Example n. 590 = (18.0, 308.0): cluster 0\n", + "Example n. 591 = (18.0, 307.0): cluster 0\n", + "Example n. 592 = (18.0, 281.0): cluster 0\n", + "Example n. 593 = (18.0, 280.0): cluster 0\n", + "Example n. 594 = (18.0, 279.0): cluster 0\n", + "Example n. 595 = (18.0, 278.0): cluster 0\n", + "Example n. 596 = (18.0, 277.0): cluster 0\n", + "Example n. 597 = (18.0, 276.0): cluster 0\n", + "Example n. 598 = (18.0, 275.0): cluster 0\n", + "Example n. 599 = (18.0, 274.0): cluster 0\n", + "Example n. 600 = (18.0, 273.0): cluster 0\n", + "Example n. 601 = (18.0, 271.0): cluster 0\n", + "Example n. 602 = (18.0, 270.0): cluster 0\n", + "Example n. 603 = (18.0, 269.0): cluster 0\n", + "Example n. 604 = (18.0, 268.0): cluster 0\n", + "Example n. 605 = (18.0, 267.0): cluster 0\n", + "Example n. 606 = (18.0, 266.0): cluster 0\n", + "Example n. 607 = (18.0, 265.0): cluster 0\n", + "Example n. 608 = (18.0, 264.0): cluster 0\n", + "Example n. 609 = (18.0, 263.0): cluster 0\n", + "Example n. 610 = (18.0, 262.0): cluster 0\n", + "Example n. 611 = (18.0, 261.0): cluster 0\n", + "Example n. 612 = (18.0, 260.0): cluster 0\n", + "Example n. 613 = (18.0, 259.0): cluster 0\n", + "Example n. 614 = (18.0, 258.0): cluster 0\n", + "Example n. 615 = (18.0, 257.0): cluster 0\n", + "Example n. 616 = (18.0, 256.0): cluster 0\n", + "Example n. 617 = (18.0, 255.0): cluster 0\n", + "Example n. 618 = (18.0, 254.0): cluster 0\n", + "Example n. 619 = (18.0, 246.0): cluster 0\n", + "Example n. 620 = (18.0, 245.0): cluster 0\n", + "Example n. 621 = (18.0, 244.0): cluster 0\n", + "Example n. 622 = (18.0, 243.0): cluster 0\n", + "Example n. 623 = (18.0, 242.0): cluster 0\n", + "Example n. 624 = (18.0, 241.0): cluster 0\n", + "Example n. 625 = (18.0, 234.0): cluster 0\n", + "Example n. 626 = (18.0, 228.0): cluster 0\n", + "Example n. 627 = (18.0, 227.0): cluster 0\n", + "Example n. 628 = (18.0, 226.0): cluster 0\n", + "Example n. 629 = (18.0, 225.0): cluster 0\n", + "Example n. 630 = (18.0, 224.0): cluster 0\n", + "Example n. 631 = (18.0, 223.0): cluster 0\n", + "Example n. 632 = (18.0, 222.0): cluster 0\n", + "Example n. 633 = (18.0, 221.0): cluster 0\n", + "Example n. 634 = (18.0, 220.0): cluster 0\n", + "Example n. 635 = (18.0, 219.0): cluster 0\n", + "Example n. 636 = (18.0, 218.0): cluster 0\n", + "Example n. 637 = (18.0, 217.0): cluster 0\n", + "Example n. 638 = (18.0, 216.0): cluster 0\n", + "Example n. 639 = (18.0, 215.0): cluster 0\n", + "Example n. 640 = (18.0, 214.0): cluster 0\n", + "Example n. 641 = (18.0, 208.0): cluster 0\n", + "Example n. 642 = (18.0, 207.0): cluster 0\n", + "Example n. 643 = (18.0, 206.0): cluster 0\n", + "Example n. 644 = (18.0, 205.0): cluster 0\n", + "Example n. 645 = (18.0, 204.0): cluster 0\n", + "Example n. 646 = (18.0, 203.0): cluster 0\n", + "Example n. 647 = (18.0, 202.0): cluster 0\n", + "Example n. 648 = (18.0, 201.0): cluster 0\n", + "Example n. 649 = (18.0, 200.0): cluster 0\n", + "Example n. 650 = (18.0, 199.0): cluster 0\n", + "Example n. 651 = (18.0, 198.0): cluster 0\n", + "Example n. 652 = (18.0, 197.0): cluster 0\n", + "Example n. 653 = (18.0, 196.0): cluster 0\n", + "Example n. 654 = (18.0, 195.0): cluster 0\n", + "Example n. 655 = (18.0, 194.0): cluster 0\n", + "Example n. 656 = (18.0, 193.0): cluster 0\n", + "Example n. 657 = (18.0, 192.0): cluster 0\n", + "Example n. 658 = (18.0, 191.0): cluster 0\n", + "Example n. 659 = (18.0, 190.0): cluster 0\n", + "Example n. 660 = (18.0, 150.0): cluster 0\n", + "Example n. 661 = (18.0, 149.0): cluster 0\n", + "Example n. 662 = (18.0, 148.0): cluster 0\n", + "Example n. 663 = (18.0, 147.0): cluster 0\n", + "Example n. 664 = (18.0, 146.0): cluster 0\n", + "Example n. 665 = (18.0, 144.0): cluster 0\n", + "Example n. 666 = (18.0, 143.0): cluster 0\n", + "Example n. 667 = (18.0, 142.0): cluster 0\n", + "Example n. 668 = (18.0, 141.0): cluster 0\n", + "Example n. 669 = (18.0, 140.0): cluster 0\n", + "Example n. 670 = (18.0, 133.0): cluster 0\n", + "Example n. 671 = (18.0, 132.0): cluster 0\n", + "Example n. 672 = (18.0, 131.0): cluster 0\n", + "Example n. 673 = (18.0, 130.0): cluster 0\n", + "Example n. 674 = (18.0, 129.0): cluster 0\n", + "Example n. 675 = (18.0, 121.0): cluster 0\n", + "Example n. 676 = (18.0, 120.0): cluster 0\n", + "Example n. 677 = (18.0, 119.0): cluster 0\n", + "Example n. 678 = (18.0, 118.0): cluster 0\n", + "Example n. 679 = (18.0, 117.0): cluster 0\n", + "Example n. 680 = (18.0, 93.0): cluster 0\n", + "Example n. 681 = (18.0, 92.0): cluster 0\n", + "Example n. 682 = (18.0, 91.0): cluster 0\n", + "Example n. 683 = (18.0, 90.0): cluster 0\n", + "Example n. 684 = (19.0, 502.0): cluster 0\n", + "Example n. 685 = (19.0, 501.0): cluster 0\n", + "Example n. 686 = (19.0, 500.0): cluster 0\n", + "Example n. 687 = (19.0, 499.0): cluster 0\n", + "Example n. 688 = (19.0, 498.0): cluster 0\n", + "Example n. 689 = (19.0, 497.0): cluster 0\n", + "Example n. 690 = (19.0, 382.0): cluster 0\n", + "Example n. 691 = (19.0, 381.0): cluster 0\n", + "Example n. 692 = (19.0, 380.0): cluster 0\n", + "Example n. 693 = (19.0, 350.0): cluster 0\n", + "Example n. 694 = (19.0, 349.0): cluster 0\n", + "Example n. 695 = (19.0, 348.0): cluster 0\n", + "Example n. 696 = (19.0, 347.0): cluster 0\n", + "Example n. 697 = (19.0, 312.0): cluster 0\n", + "Example n. 698 = (19.0, 311.0): cluster 0\n", + "Example n. 699 = (19.0, 310.0): cluster 0\n", + "Example n. 700 = (19.0, 309.0): cluster 0\n", + "Example n. 701 = (19.0, 308.0): cluster 0\n", + "Example n. 702 = (19.0, 307.0): cluster 0\n", + "Example n. 703 = (19.0, 282.0): cluster 0\n", + "Example n. 704 = (19.0, 281.0): cluster 0\n", + "Example n. 705 = (19.0, 280.0): cluster 0\n", + "Example n. 706 = (19.0, 279.0): cluster 0\n", + "Example n. 707 = (19.0, 278.0): cluster 0\n", + "Example n. 708 = (19.0, 277.0): cluster 0\n", + "Example n. 709 = (19.0, 276.0): cluster 0\n", + "Example n. 710 = (19.0, 275.0): cluster 0\n", + "Example n. 711 = (19.0, 274.0): cluster 0\n", + "Example n. 712 = (19.0, 273.0): cluster 0\n", + "Example n. 713 = (19.0, 272.0): cluster 0\n", + "Example n. 714 = (19.0, 271.0): cluster 0\n", + "Example n. 715 = (19.0, 270.0): cluster 0\n", + "Example n. 716 = (19.0, 269.0): cluster 0\n", + "Example n. 717 = (19.0, 268.0): cluster 0\n", + "Example n. 718 = (19.0, 267.0): cluster 0\n", + "Example n. 719 = (19.0, 266.0): cluster 0\n", + "Example n. 720 = (19.0, 265.0): cluster 0\n", + "Example n. 721 = (19.0, 264.0): cluster 0\n", + "Example n. 722 = (19.0, 263.0): cluster 0\n", + "Example n. 723 = (19.0, 262.0): cluster 0\n", + "Example n. 724 = (19.0, 261.0): cluster 0\n", + "Example n. 725 = (19.0, 260.0): cluster 0\n", + "Example n. 726 = (19.0, 259.0): cluster 0\n", + "Example n. 727 = (19.0, 258.0): cluster 0\n", + "Example n. 728 = (19.0, 257.0): cluster 0\n", + "Example n. 729 = (19.0, 256.0): cluster 0\n", + "Example n. 730 = (19.0, 255.0): cluster 0\n", + "Example n. 731 = (19.0, 254.0): cluster 0\n", + "Example n. 732 = (19.0, 253.0): cluster 0\n", + "Example n. 733 = (19.0, 251.0): cluster 0\n", + "Example n. 734 = (19.0, 250.0): cluster 0\n", + "Example n. 735 = (19.0, 249.0): cluster 0\n", + "Example n. 736 = (19.0, 246.0): cluster 0\n", + "Example n. 737 = (19.0, 245.0): cluster 0\n", + "Example n. 738 = (19.0, 244.0): cluster 0\n", + "Example n. 739 = (19.0, 243.0): cluster 0\n", + "Example n. 740 = (19.0, 242.0): cluster 0\n", + "Example n. 741 = (19.0, 236.0): cluster 0\n", + "Example n. 742 = (19.0, 235.0): cluster 0\n", + "Example n. 743 = (19.0, 234.0): cluster 0\n", + "Example n. 744 = (19.0, 233.0): cluster 0\n", + "Example n. 745 = (19.0, 232.0): cluster 0\n", + "Example n. 746 = (19.0, 228.0): cluster 0\n", + "Example n. 747 = (19.0, 227.0): cluster 0\n", + "Example n. 748 = (19.0, 226.0): cluster 0\n", + "Example n. 749 = (19.0, 225.0): cluster 0\n", + "Example n. 750 = (19.0, 224.0): cluster 0\n", + "Example n. 751 = (19.0, 223.0): cluster 0\n", + "Example n. 752 = (19.0, 222.0): cluster 0\n", + "Example n. 753 = (19.0, 221.0): cluster 0\n", + "Example n. 754 = (19.0, 220.0): cluster 0\n", + "Example n. 755 = (19.0, 219.0): cluster 0\n", + "Example n. 756 = (19.0, 218.0): cluster 0\n", + "Example n. 757 = (19.0, 217.0): cluster 0\n", + "Example n. 758 = (19.0, 216.0): cluster 0\n", + "Example n. 759 = (19.0, 215.0): cluster 0\n", + "Example n. 760 = (19.0, 214.0): cluster 0\n", + "Example n. 761 = (19.0, 208.0): cluster 0\n", + "Example n. 762 = (19.0, 207.0): cluster 0\n", + "Example n. 763 = (19.0, 206.0): cluster 0\n", + "Example n. 764 = (19.0, 205.0): cluster 0\n", + "Example n. 765 = (19.0, 204.0): cluster 0\n", + "Example n. 766 = (19.0, 203.0): cluster 0\n", + "Example n. 767 = (19.0, 202.0): cluster 0\n", + "Example n. 768 = (19.0, 201.0): cluster 0\n", + "Example n. 769 = (19.0, 200.0): cluster 0\n", + "Example n. 770 = (19.0, 199.0): cluster 0\n", + "Example n. 771 = (19.0, 198.0): cluster 0\n", + "Example n. 772 = (19.0, 197.0): cluster 0\n", + "Example n. 773 = (19.0, 196.0): cluster 0\n", + "Example n. 774 = (19.0, 195.0): cluster 0\n", + "Example n. 775 = (19.0, 194.0): cluster 0\n", + "Example n. 776 = (19.0, 193.0): cluster 0\n", + "Example n. 777 = (19.0, 192.0): cluster 0\n", + "Example n. 778 = (19.0, 191.0): cluster 0\n", + "Example n. 779 = (19.0, 190.0): cluster 0\n", + "Example n. 780 = (19.0, 189.0): cluster 0\n", + "Example n. 781 = (19.0, 150.0): cluster 0\n", + "Example n. 782 = (19.0, 149.0): cluster 0\n", + "Example n. 783 = (19.0, 148.0): cluster 0\n", + "Example n. 784 = (19.0, 147.0): cluster 0\n", + "Example n. 785 = (19.0, 146.0): cluster 0\n", + "Example n. 786 = (19.0, 145.0): cluster 0\n", + "Example n. 787 = (19.0, 144.0): cluster 0\n", + "Example n. 788 = (19.0, 143.0): cluster 0\n", + "Example n. 789 = (19.0, 142.0): cluster 0\n", + "Example n. 790 = (19.0, 141.0): cluster 0\n", + "Example n. 791 = (19.0, 140.0): cluster 0\n", + "Example n. 792 = (19.0, 133.0): cluster 0\n", + "Example n. 793 = (19.0, 132.0): cluster 0\n", + "Example n. 794 = (19.0, 131.0): cluster 0\n", + "Example n. 795 = (19.0, 130.0): cluster 0\n", + "Example n. 796 = (19.0, 129.0): cluster 0\n", + "Example n. 797 = (19.0, 121.0): cluster 0\n", + "Example n. 798 = (19.0, 120.0): cluster 0\n", + "Example n. 799 = (19.0, 119.0): cluster 0\n", + "Example n. 800 = (19.0, 118.0): cluster 0\n", + "Example n. 801 = (19.0, 117.0): cluster 0\n", + "Example n. 802 = (19.0, 94.0): cluster 0\n", + "Example n. 803 = (19.0, 93.0): cluster 0\n", + "Example n. 804 = (19.0, 92.0): cluster 0\n", + "Example n. 805 = (19.0, 91.0): cluster 0\n", + "Example n. 806 = (19.0, 90.0): cluster 0\n", + "Example n. 807 = (19.0, 89.0): cluster 0\n", + "Example n. 808 = (20.0, 502.0): cluster 0\n", + "Example n. 809 = (20.0, 501.0): cluster 0\n", + "Example n. 810 = (20.0, 500.0): cluster 0\n", + "Example n. 811 = (20.0, 499.0): cluster 0\n", + "Example n. 812 = (20.0, 498.0): cluster 0\n", + "Example n. 813 = (20.0, 497.0): cluster 0\n", + "Example n. 814 = (20.0, 383.0): cluster 0\n", + "Example n. 815 = (20.0, 382.0): cluster 0\n", + "Example n. 816 = (20.0, 381.0): cluster 0\n", + "Example n. 817 = (20.0, 380.0): cluster 0\n", + "Example n. 818 = (20.0, 379.0): cluster 0\n", + "Example n. 819 = (20.0, 350.0): cluster 0\n", + "Example n. 820 = (20.0, 349.0): cluster 0\n", + "Example n. 821 = (20.0, 348.0): cluster 0\n", + "Example n. 822 = (20.0, 347.0): cluster 0\n", + "Example n. 823 = (20.0, 346.0): cluster 0\n", + "Example n. 824 = (20.0, 311.0): cluster 0\n", + "Example n. 825 = (20.0, 310.0): cluster 0\n", + "Example n. 826 = (20.0, 309.0): cluster 0\n", + "Example n. 827 = (20.0, 308.0): cluster 0\n", + "Example n. 828 = (20.0, 287.0): cluster 0\n", + "Example n. 829 = (20.0, 286.0): cluster 0\n", + "Example n. 830 = (20.0, 285.0): cluster 0\n", + "Example n. 831 = (20.0, 284.0): cluster 0\n", + "Example n. 832 = (20.0, 283.0): cluster 0\n", + "Example n. 833 = (20.0, 282.0): cluster 0\n", + "Example n. 834 = (20.0, 281.0): cluster 0\n", + "Example n. 835 = (20.0, 280.0): cluster 0\n", + "Example n. 836 = (20.0, 279.0): cluster 0\n", + "Example n. 837 = (20.0, 278.0): cluster 0\n", + "Example n. 838 = (20.0, 277.0): cluster 0\n", + "Example n. 839 = (20.0, 276.0): cluster 0\n", + "Example n. 840 = (20.0, 275.0): cluster 0\n", + "Example n. 841 = (20.0, 274.0): cluster 0\n", + "Example n. 842 = (20.0, 272.0): cluster 0\n", + "Example n. 843 = (20.0, 271.0): cluster 0\n", + "Example n. 844 = (20.0, 270.0): cluster 0\n", + "Example n. 845 = (20.0, 269.0): cluster 0\n", + "Example n. 846 = (20.0, 268.0): cluster 0\n", + "Example n. 847 = (20.0, 267.0): cluster 0\n", + "Example n. 848 = (20.0, 266.0): cluster 0\n", + "Example n. 849 = (20.0, 265.0): cluster 0\n", + "Example n. 850 = (20.0, 263.0): cluster 0\n", + "Example n. 851 = (20.0, 262.0): cluster 0\n", + "Example n. 852 = (20.0, 261.0): cluster 0\n", + "Example n. 853 = (20.0, 260.0): cluster 0\n", + "Example n. 854 = (20.0, 259.0): cluster 0\n", + "Example n. 855 = (20.0, 258.0): cluster 0\n", + "Example n. 856 = (20.0, 257.0): cluster 0\n", + "Example n. 857 = (20.0, 256.0): cluster 0\n", + "Example n. 858 = (20.0, 255.0): cluster 0\n", + "Example n. 859 = (20.0, 254.0): cluster 0\n", + "Example n. 860 = (20.0, 253.0): cluster 0\n", + "Example n. 861 = (20.0, 252.0): cluster 0\n", + "Example n. 862 = (20.0, 251.0): cluster 0\n", + "Example n. 863 = (20.0, 250.0): cluster 0\n", + "Example n. 864 = (20.0, 249.0): cluster 0\n", + "Example n. 865 = (20.0, 248.0): cluster 0\n", + "Example n. 866 = (20.0, 246.0): cluster 0\n", + "Example n. 867 = (20.0, 245.0): cluster 0\n", + "Example n. 868 = (20.0, 244.0): cluster 0\n", + "Example n. 869 = (20.0, 243.0): cluster 0\n", + "Example n. 870 = (20.0, 242.0): cluster 0\n", + "Example n. 871 = (20.0, 237.0): cluster 0\n", + "Example n. 872 = (20.0, 236.0): cluster 0\n", + "Example n. 873 = (20.0, 235.0): cluster 0\n", + "Example n. 874 = (20.0, 234.0): cluster 0\n", + "Example n. 875 = (20.0, 233.0): cluster 0\n", + "Example n. 876 = (20.0, 232.0): cluster 0\n", + "Example n. 877 = (20.0, 226.0): cluster 0\n", + "Example n. 878 = (20.0, 225.0): cluster 0\n", + "Example n. 879 = (20.0, 224.0): cluster 0\n", + "Example n. 880 = (20.0, 223.0): cluster 0\n", + "Example n. 881 = (20.0, 222.0): cluster 0\n", + "Example n. 882 = (20.0, 221.0): cluster 0\n", + "Example n. 883 = (20.0, 220.0): cluster 0\n", + "Example n. 884 = (20.0, 219.0): cluster 0\n", + "Example n. 885 = (20.0, 218.0): cluster 0\n", + "Example n. 886 = (20.0, 217.0): cluster 0\n", + "Example n. 887 = (20.0, 216.0): cluster 0\n", + "Example n. 888 = (20.0, 215.0): cluster 0\n", + "Example n. 889 = (20.0, 214.0): cluster 0\n", + "Example n. 890 = (20.0, 208.0): cluster 0\n", + "Example n. 891 = (20.0, 207.0): cluster 0\n", + "Example n. 892 = (20.0, 206.0): cluster 0\n", + "Example n. 893 = (20.0, 205.0): cluster 0\n", + "Example n. 894 = (20.0, 204.0): cluster 0\n", + "Example n. 895 = (20.0, 203.0): cluster 0\n", + "Example n. 896 = (20.0, 202.0): cluster 0\n", + "Example n. 897 = (20.0, 201.0): cluster 0\n", + "Example n. 898 = (20.0, 200.0): cluster 0\n", + "Example n. 899 = (20.0, 199.0): cluster 0\n", + "Example n. 900 = (20.0, 198.0): cluster 0\n", + "Example n. 901 = (20.0, 197.0): cluster 0\n", + "Example n. 902 = (20.0, 196.0): cluster 0\n", + "Example n. 903 = (20.0, 195.0): cluster 0\n", + "Example n. 904 = (20.0, 194.0): cluster 0\n", + "Example n. 905 = (20.0, 193.0): cluster 0\n", + "Example n. 906 = (20.0, 192.0): cluster 0\n", + "Example n. 907 = (20.0, 191.0): cluster 0\n", + "Example n. 908 = (20.0, 190.0): cluster 0\n", + "Example n. 909 = (20.0, 189.0): cluster 0\n", + "Example n. 910 = (20.0, 188.0): cluster 0\n", + "Example n. 911 = (20.0, 150.0): cluster 0\n", + "Example n. 912 = (20.0, 149.0): cluster 0\n", + "Example n. 913 = (20.0, 148.0): cluster 0\n", + "Example n. 914 = (20.0, 147.0): cluster 0\n", + "Example n. 915 = (20.0, 146.0): cluster 0\n", + "Example n. 916 = (20.0, 145.0): cluster 0\n", + "Example n. 917 = (20.0, 144.0): cluster 0\n", + "Example n. 918 = (20.0, 143.0): cluster 0\n", + "Example n. 919 = (20.0, 142.0): cluster 0\n", + "Example n. 920 = (20.0, 141.0): cluster 0\n", + "Example n. 921 = (20.0, 140.0): cluster 0\n", + "Example n. 922 = (20.0, 133.0): cluster 0\n", + "Example n. 923 = (20.0, 132.0): cluster 0\n", + "Example n. 924 = (20.0, 131.0): cluster 3\n", + "Example n. 925 = (20.0, 130.0): cluster 3\n", + "Example n. 926 = (20.0, 129.0): cluster 3\n", + "Example n. 927 = (20.0, 125.0): cluster 3\n", + "Example n. 928 = (20.0, 124.0): cluster 3\n", + "Example n. 929 = (20.0, 123.0): cluster 3\n", + "Example n. 930 = (20.0, 122.0): cluster 3\n", + "Example n. 931 = (20.0, 121.0): cluster 3\n", + "Example n. 932 = (20.0, 120.0): cluster 3\n", + "Example n. 933 = (20.0, 119.0): cluster 3\n", + "Example n. 934 = (20.0, 118.0): cluster 3\n", + "Example n. 935 = (20.0, 117.0): cluster 3\n", + "Example n. 936 = (20.0, 94.0): cluster 0\n", + "Example n. 937 = (20.0, 93.0): cluster 0\n", + "Example n. 938 = (20.0, 92.0): cluster 0\n", + "Example n. 939 = (20.0, 91.0): cluster 0\n", + "Example n. 940 = (20.0, 90.0): cluster 0\n", + "Example n. 941 = (20.0, 89.0): cluster 0\n", + "Example n. 942 = (21.0, 501.0): cluster 0\n", + "Example n. 943 = (21.0, 500.0): cluster 0\n", + "Example n. 944 = (21.0, 499.0): cluster 0\n", + "Example n. 945 = (21.0, 498.0): cluster 0\n", + "Example n. 946 = (21.0, 497.0): cluster 0\n", + "Example n. 947 = (21.0, 383.0): cluster 0\n", + "Example n. 948 = (21.0, 382.0): cluster 0\n", + "Example n. 949 = (21.0, 381.0): cluster 0\n", + "Example n. 950 = (21.0, 380.0): cluster 0\n", + "Example n. 951 = (21.0, 379.0): cluster 0\n", + "Example n. 952 = (21.0, 369.0): cluster 0\n", + "Example n. 953 = (21.0, 351.0): cluster 0\n", + "Example n. 954 = (21.0, 350.0): cluster 0\n", + "Example n. 955 = (21.0, 349.0): cluster 0\n", + "Example n. 956 = (21.0, 348.0): cluster 0\n", + "Example n. 957 = (21.0, 347.0): cluster 0\n", + "Example n. 958 = (21.0, 346.0): cluster 0\n", + "Example n. 959 = (21.0, 307.0): cluster 0\n", + "Example n. 960 = (21.0, 306.0): cluster 0\n", + "Example n. 961 = (21.0, 305.0): cluster 0\n", + "Example n. 962 = (21.0, 287.0): cluster 0\n", + "Example n. 963 = (21.0, 286.0): cluster 0\n", + "Example n. 964 = (21.0, 285.0): cluster 0\n", + "Example n. 965 = (21.0, 284.0): cluster 0\n", + "Example n. 966 = (21.0, 283.0): cluster 0\n", + "Example n. 967 = (21.0, 282.0): cluster 0\n", + "Example n. 968 = (21.0, 281.0): cluster 0\n", + "Example n. 969 = (21.0, 280.0): cluster 0\n", + "Example n. 970 = (21.0, 279.0): cluster 0\n", + "Example n. 971 = (21.0, 278.0): cluster 0\n", + "Example n. 972 = (21.0, 277.0): cluster 0\n", + "Example n. 973 = (21.0, 276.0): cluster 0\n", + "Example n. 974 = (21.0, 275.0): cluster 0\n", + "Example n. 975 = (21.0, 271.0): cluster 0\n", + "Example n. 976 = (21.0, 270.0): cluster 0\n", + "Example n. 977 = (21.0, 269.0): cluster 0\n", + "Example n. 978 = (21.0, 268.0): cluster 0\n", + "Example n. 979 = (21.0, 267.0): cluster 0\n", + "Example n. 980 = (21.0, 262.0): cluster 0\n", + "Example n. 981 = (21.0, 261.0): cluster 3\n", + "Example n. 982 = (21.0, 260.0): cluster 3\n", + "Example n. 983 = (21.0, 259.0): cluster 3\n", + "Example n. 984 = (21.0, 258.0): cluster 3\n", + "Example n. 985 = (21.0, 257.0): cluster 3\n", + "Example n. 986 = (21.0, 256.0): cluster 3\n", + "Example n. 987 = (21.0, 255.0): cluster 3\n", + "Example n. 988 = (21.0, 254.0): cluster 3\n", + "Example n. 989 = (21.0, 253.0): cluster 3\n", + "Example n. 990 = (21.0, 252.0): cluster 3\n", + "Example n. 991 = (21.0, 251.0): cluster 3\n", + "Example n. 992 = (21.0, 250.0): cluster 3\n", + "Example n. 993 = (21.0, 249.0): cluster 0\n", + "Example n. 994 = (21.0, 248.0): cluster 0\n", + "Example n. 995 = (21.0, 247.0): cluster 0\n", + "Example n. 996 = (21.0, 244.0): cluster 0\n", + "Example n. 997 = (21.0, 237.0): cluster 0\n", + "Example n. 998 = (21.0, 236.0): cluster 0\n", + "Example n. 999 = (21.0, 235.0): cluster 0\n", + "Example n. 1000 = (21.0, 234.0): cluster 0\n", + "Example n. 1001 = (21.0, 233.0): cluster 0\n", + "Example n. 1002 = (21.0, 232.0): cluster 0\n", + "Example n. 1003 = (21.0, 231.0): cluster 0\n", + "Example n. 1004 = (21.0, 226.0): cluster 0\n", + "Example n. 1005 = (21.0, 225.0): cluster 0\n", + "Example n. 1006 = (21.0, 224.0): cluster 0\n", + "Example n. 1007 = (21.0, 223.0): cluster 0\n", + "Example n. 1008 = (21.0, 222.0): cluster 0\n", + "Example n. 1009 = (21.0, 221.0): cluster 0\n", + "Example n. 1010 = (21.0, 220.0): cluster 0\n", + "Example n. 1011 = (21.0, 219.0): cluster 0\n", + "Example n. 1012 = (21.0, 218.0): cluster 0\n", + "Example n. 1013 = (21.0, 217.0): cluster 0\n", + "Example n. 1014 = (21.0, 216.0): cluster 0\n", + "Example n. 1015 = (21.0, 215.0): cluster 0\n", + "Example n. 1016 = (21.0, 214.0): cluster 0\n", + "Example n. 1017 = (21.0, 213.0): cluster 0\n", + "Example n. 1018 = (21.0, 212.0): cluster 0\n", + "Example n. 1019 = (21.0, 208.0): cluster 0\n", + "Example n. 1020 = (21.0, 207.0): cluster 0\n", + "Example n. 1021 = (21.0, 206.0): cluster 0\n", + "Example n. 1022 = (21.0, 205.0): cluster 0\n", + "Example n. 1023 = (21.0, 204.0): cluster 0\n", + "Example n. 1024 = (21.0, 203.0): cluster 0\n", + "Example n. 1025 = (21.0, 202.0): cluster 0\n", + "Example n. 1026 = (21.0, 201.0): cluster 0\n", + "Example n. 1027 = (21.0, 200.0): cluster 0\n", + "Example n. 1028 = (21.0, 199.0): cluster 0\n", + "Example n. 1029 = (21.0, 198.0): cluster 0\n", + "Example n. 1030 = (21.0, 197.0): cluster 0\n", + "Example n. 1031 = (21.0, 196.0): cluster 0\n", + "Example n. 1032 = (21.0, 195.0): cluster 0\n", + "Example n. 1033 = (21.0, 194.0): cluster 0\n", + "Example n. 1034 = (21.0, 193.0): cluster 0\n", + "Example n. 1035 = (21.0, 192.0): cluster 0\n", + "Example n. 1036 = (21.0, 191.0): cluster 0\n", + "Example n. 1037 = (21.0, 190.0): cluster 0\n", + "Example n. 1038 = (21.0, 189.0): cluster 3\n", + "Example n. 1039 = (21.0, 188.0): cluster 3\n", + "Example n. 1040 = (21.0, 150.0): cluster 3\n", + "Example n. 1041 = (21.0, 149.0): cluster 3\n", + "Example n. 1042 = (21.0, 148.0): cluster 3\n", + "Example n. 1043 = (21.0, 147.0): cluster 3\n", + "Example n. 1044 = (21.0, 146.0): cluster 3\n", + "Example n. 1045 = (21.0, 144.0): cluster 3\n", + "Example n. 1046 = (21.0, 143.0): cluster 3\n", + "Example n. 1047 = (21.0, 142.0): cluster 3\n", + "Example n. 1048 = (21.0, 141.0): cluster 3\n", + "Example n. 1049 = (21.0, 140.0): cluster 3\n", + "Example n. 1050 = (21.0, 133.0): cluster 0\n", + "Example n. 1051 = (21.0, 132.0): cluster 0\n", + "Example n. 1052 = (21.0, 131.0): cluster 0\n", + "Example n. 1053 = (21.0, 130.0): cluster 0\n", + "Example n. 1054 = (21.0, 129.0): cluster 0\n", + "Example n. 1055 = (21.0, 126.0): cluster 0\n", + "Example n. 1056 = (21.0, 125.0): cluster 0\n", + "Example n. 1057 = (21.0, 124.0): cluster 0\n", + "Example n. 1058 = (21.0, 123.0): cluster 0\n", + "Example n. 1059 = (21.0, 122.0): cluster 0\n", + "Example n. 1060 = (21.0, 121.0): cluster 0\n", + "Example n. 1061 = (21.0, 120.0): cluster 0\n", + "Example n. 1062 = (21.0, 119.0): cluster 0\n", + "Example n. 1063 = (21.0, 118.0): cluster 0\n", + "Example n. 1064 = (21.0, 117.0): cluster 0\n", + "Example n. 1065 = (21.0, 94.0): cluster 0\n", + "Example n. 1066 = (21.0, 93.0): cluster 0\n", + "Example n. 1067 = (21.0, 92.0): cluster 0\n", + "Example n. 1068 = (21.0, 91.0): cluster 0\n", + "Example n. 1069 = (21.0, 90.0): cluster 0\n", + "Example n. 1070 = (22.0, 500.0): cluster 0\n", + "Example n. 1071 = (22.0, 499.0): cluster 0\n", + "Example n. 1072 = (22.0, 498.0): cluster 0\n", + "Example n. 1073 = (22.0, 383.0): cluster 0\n", + "Example n. 1074 = (22.0, 382.0): cluster 0\n", + "Example n. 1075 = (22.0, 381.0): cluster 0\n", + "Example n. 1076 = (22.0, 380.0): cluster 0\n", + "Example n. 1077 = (22.0, 379.0): cluster 0\n", + "Example n. 1078 = (22.0, 371.0): cluster 0\n", + "Example n. 1079 = (22.0, 370.0): cluster 0\n", + "Example n. 1080 = (22.0, 369.0): cluster 0\n", + "Example n. 1081 = (22.0, 368.0): cluster 0\n", + "Example n. 1082 = (22.0, 350.0): cluster 0\n", + "Example n. 1083 = (22.0, 349.0): cluster 0\n", + "Example n. 1084 = (22.0, 348.0): cluster 0\n", + "Example n. 1085 = (22.0, 347.0): cluster 0\n", + "Example n. 1086 = (22.0, 346.0): cluster 0\n", + "Example n. 1087 = (22.0, 308.0): cluster 0\n", + "Example n. 1088 = (22.0, 307.0): cluster 0\n", + "Example n. 1089 = (22.0, 306.0): cluster 0\n", + "Example n. 1090 = (22.0, 305.0): cluster 0\n", + "Example n. 1091 = (22.0, 304.0): cluster 0\n", + "Example n. 1092 = (22.0, 302.0): cluster 0\n", + "Example n. 1093 = (22.0, 301.0): cluster 0\n", + "Example n. 1094 = (22.0, 300.0): cluster 0\n", + "Example n. 1095 = (22.0, 295.0): cluster 0\n", + "Example n. 1096 = (22.0, 294.0): cluster 1\n", + "Example n. 1097 = (22.0, 293.0): cluster 1\n", + "Example n. 1098 = (22.0, 292.0): cluster 3\n", + "Example n. 1099 = (22.0, 288.0): cluster 3\n", + "Example n. 1100 = (22.0, 287.0): cluster 3\n", + "Example n. 1101 = (22.0, 286.0): cluster 3\n", + "Example n. 1102 = (22.0, 285.0): cluster 3\n", + "Example n. 1103 = (22.0, 284.0): cluster 3\n", + "Example n. 1104 = (22.0, 283.0): cluster 3\n", + "Example n. 1105 = (22.0, 282.0): cluster 3\n", + "Example n. 1106 = (22.0, 281.0): cluster 3\n", + "Example n. 1107 = (22.0, 280.0): cluster 3\n", + "Example n. 1108 = (22.0, 279.0): cluster 3\n", + "Example n. 1109 = (22.0, 278.0): cluster 3\n", + "Example n. 1110 = (22.0, 277.0): cluster 3\n", + "Example n. 1111 = (22.0, 276.0): cluster 3\n", + "Example n. 1112 = (22.0, 270.0): cluster 3\n", + "Example n. 1113 = (22.0, 269.0): cluster 3\n", + "Example n. 1114 = (22.0, 268.0): cluster 3\n", + "Example n. 1115 = (22.0, 261.0): cluster 3\n", + "Example n. 1116 = (22.0, 260.0): cluster 3\n", + "Example n. 1117 = (22.0, 259.0): cluster 3\n", + "Example n. 1118 = (22.0, 258.0): cluster 3\n", + "Example n. 1119 = (22.0, 257.0): cluster 3\n", + "Example n. 1120 = (22.0, 256.0): cluster 3\n", + "Example n. 1121 = (22.0, 255.0): cluster 3\n", + "Example n. 1122 = (22.0, 254.0): cluster 0\n", + "Example n. 1123 = (22.0, 253.0): cluster 0\n", + "Example n. 1124 = (22.0, 252.0): cluster 0\n", + "Example n. 1125 = (22.0, 251.0): cluster 0\n", + "Example n. 1126 = (22.0, 250.0): cluster 0\n", + "Example n. 1127 = (22.0, 249.0): cluster 0\n", + "Example n. 1128 = (22.0, 248.0): cluster 0\n", + "Example n. 1129 = (22.0, 247.0): cluster 0\n", + "Example n. 1130 = (22.0, 246.0): cluster 0\n", + "Example n. 1131 = (22.0, 245.0): cluster 0\n", + "Example n. 1132 = (22.0, 244.0): cluster 0\n", + "Example n. 1133 = (22.0, 243.0): cluster 0\n", + "Example n. 1134 = (22.0, 237.0): cluster 0\n", + "Example n. 1135 = (22.0, 236.0): cluster 0\n", + "Example n. 1136 = (22.0, 235.0): cluster 0\n", + "Example n. 1137 = (22.0, 234.0): cluster 0\n", + "Example n. 1138 = (22.0, 233.0): cluster 0\n", + "Example n. 1139 = (22.0, 232.0): cluster 0\n", + "Example n. 1140 = (22.0, 231.0): cluster 0\n", + "Example n. 1141 = (22.0, 230.0): cluster 0\n", + "Example n. 1142 = (22.0, 227.0): cluster 0\n", + "Example n. 1143 = (22.0, 226.0): cluster 0\n", + "Example n. 1144 = (22.0, 225.0): cluster 0\n", + "Example n. 1145 = (22.0, 224.0): cluster 0\n", + "Example n. 1146 = (22.0, 223.0): cluster 0\n", + "Example n. 1147 = (22.0, 222.0): cluster 0\n", + "Example n. 1148 = (22.0, 221.0): cluster 0\n", + "Example n. 1149 = (22.0, 220.0): cluster 0\n", + "Example n. 1150 = (22.0, 219.0): cluster 0\n", + "Example n. 1151 = (22.0, 218.0): cluster 0\n", + "Example n. 1152 = (22.0, 217.0): cluster 0\n", + "Example n. 1153 = (22.0, 216.0): cluster 0\n", + "Example n. 1154 = (22.0, 215.0): cluster 0\n", + "Example n. 1155 = (22.0, 214.0): cluster 0\n", + "Example n. 1156 = (22.0, 213.0): cluster 0\n", + "Example n. 1157 = (22.0, 212.0): cluster 0\n", + "Example n. 1158 = (22.0, 211.0): cluster 0\n", + "Example n. 1159 = (22.0, 210.0): cluster 0\n", + "Example n. 1160 = (22.0, 207.0): cluster 0\n", + "Example n. 1161 = (22.0, 206.0): cluster 0\n", + "Example n. 1162 = (22.0, 205.0): cluster 0\n", + "Example n. 1163 = (22.0, 203.0): cluster 0\n", + "Example n. 1164 = (22.0, 202.0): cluster 0\n", + "Example n. 1165 = (22.0, 201.0): cluster 1\n", + "Example n. 1166 = (22.0, 200.0): cluster 1\n", + "Example n. 1167 = (22.0, 198.0): cluster 1\n", + "Example n. 1168 = (22.0, 197.0): cluster 1\n", + "Example n. 1169 = (22.0, 196.0): cluster 1\n", + "Example n. 1170 = (22.0, 195.0): cluster 3\n", + "Example n. 1171 = (22.0, 193.0): cluster 3\n", + "Example n. 1172 = (22.0, 192.0): cluster 3\n", + "Example n. 1173 = (22.0, 191.0): cluster 3\n", + "Example n. 1174 = (22.0, 190.0): cluster 3\n", + "Example n. 1175 = (22.0, 189.0): cluster 3\n", + "Example n. 1176 = (22.0, 188.0): cluster 3\n", + "Example n. 1177 = (22.0, 149.0): cluster 3\n", + "Example n. 1178 = (22.0, 148.0): cluster 3\n", + "Example n. 1179 = (22.0, 147.0): cluster 3\n", + "Example n. 1180 = (22.0, 144.0): cluster 3\n", + "Example n. 1181 = (22.0, 143.0): cluster 3\n", + "Example n. 1182 = (22.0, 142.0): cluster 3\n", + "Example n. 1183 = (22.0, 141.0): cluster 3\n", + "Example n. 1184 = (22.0, 140.0): cluster 3\n", + "Example n. 1185 = (22.0, 131.0): cluster 3\n", + "Example n. 1186 = (22.0, 130.0): cluster 3\n", + "Example n. 1187 = (22.0, 129.0): cluster 3\n", + "Example n. 1188 = (22.0, 126.0): cluster 3\n", + "Example n. 1189 = (22.0, 125.0): cluster 3\n", + "Example n. 1190 = (22.0, 124.0): cluster 3\n", + "Example n. 1191 = (22.0, 123.0): cluster 3\n", + "Example n. 1192 = (22.0, 122.0): cluster 3\n", + "Example n. 1193 = (22.0, 121.0): cluster 3\n", + "Example n. 1194 = (22.0, 120.0): cluster 0\n", + "Example n. 1195 = (22.0, 119.0): cluster 0\n", + "Example n. 1196 = (22.0, 118.0): cluster 0\n", + "Example n. 1197 = (22.0, 93.0): cluster 0\n", + "Example n. 1198 = (22.0, 92.0): cluster 0\n", + "Example n. 1199 = (22.0, 91.0): cluster 0\n", + "Example n. 1200 = (22.0, 90.0): cluster 0\n", + "Example n. 1201 = (23.0, 456.0): cluster 0\n", + "Example n. 1202 = (23.0, 455.0): cluster 0\n", + "Example n. 1203 = (23.0, 454.0): cluster 0\n", + "Example n. 1204 = (23.0, 453.0): cluster 0\n", + "Example n. 1205 = (23.0, 383.0): cluster 0\n", + "Example n. 1206 = (23.0, 382.0): cluster 0\n", + "Example n. 1207 = (23.0, 381.0): cluster 0\n", + "Example n. 1208 = (23.0, 380.0): cluster 0\n", + "Example n. 1209 = (23.0, 379.0): cluster 0\n", + "Example n. 1210 = (23.0, 372.0): cluster 0\n", + "Example n. 1211 = (23.0, 371.0): cluster 0\n", + "Example n. 1212 = (23.0, 370.0): cluster 0\n", + "Example n. 1213 = (23.0, 369.0): cluster 0\n", + "Example n. 1214 = (23.0, 368.0): cluster 0\n", + "Example n. 1215 = (23.0, 367.0): cluster 0\n", + "Example n. 1216 = (23.0, 350.0): cluster 0\n", + "Example n. 1217 = (23.0, 349.0): cluster 0\n", + "Example n. 1218 = (23.0, 348.0): cluster 0\n", + "Example n. 1219 = (23.0, 347.0): cluster 0\n", + "Example n. 1220 = (23.0, 308.0): cluster 0\n", + "Example n. 1221 = (23.0, 307.0): cluster 0\n", + "Example n. 1222 = (23.0, 306.0): cluster 0\n", + "Example n. 1223 = (23.0, 305.0): cluster 0\n", + "Example n. 1224 = (23.0, 304.0): cluster 0\n", + "Example n. 1225 = (23.0, 303.0): cluster 0\n", + "Example n. 1226 = (23.0, 302.0): cluster 0\n", + "Example n. 1227 = (23.0, 301.0): cluster 0\n", + "Example n. 1228 = (23.0, 300.0): cluster 0\n", + "Example n. 1229 = (23.0, 299.0): cluster 0\n", + "Example n. 1230 = (23.0, 295.0): cluster 0\n", + "Example n. 1231 = (23.0, 294.0): cluster 0\n", + "Example n. 1232 = (23.0, 293.0): cluster 0\n", + "Example n. 1233 = (23.0, 292.0): cluster 0\n", + "Example n. 1234 = (23.0, 291.0): cluster 1\n", + "Example n. 1235 = (23.0, 288.0): cluster 1\n", + "Example n. 1236 = (23.0, 287.0): cluster 1\n", + "Example n. 1237 = (23.0, 286.0): cluster 1\n", + "Example n. 1238 = (23.0, 285.0): cluster 1\n", + "Example n. 1239 = (23.0, 284.0): cluster 1\n", + "Example n. 1240 = (23.0, 283.0): cluster 1\n", + "Example n. 1241 = (23.0, 282.0): cluster 1\n", + "Example n. 1242 = (23.0, 281.0): cluster 3\n", + "Example n. 1243 = (23.0, 280.0): cluster 3\n", + "Example n. 1244 = (23.0, 279.0): cluster 3\n", + "Example n. 1245 = (23.0, 278.0): cluster 3\n", + "Example n. 1246 = (23.0, 277.0): cluster 3\n", + "Example n. 1247 = (23.0, 276.0): cluster 3\n", + "Example n. 1248 = (23.0, 275.0): cluster 3\n", + "Example n. 1249 = (23.0, 267.0): cluster 3\n", + "Example n. 1250 = (23.0, 266.0): cluster 3\n", + "Example n. 1251 = (23.0, 265.0): cluster 3\n", + "Example n. 1252 = (23.0, 261.0): cluster 3\n", + "Example n. 1253 = (23.0, 260.0): cluster 3\n", + "Example n. 1254 = (23.0, 259.0): cluster 3\n", + "Example n. 1255 = (23.0, 258.0): cluster 3\n", + "Example n. 1256 = (23.0, 257.0): cluster 3\n", + "Example n. 1257 = (23.0, 256.0): cluster 3\n", + "Example n. 1258 = (23.0, 255.0): cluster 3\n", + "Example n. 1259 = (23.0, 254.0): cluster 3\n", + "Example n. 1260 = (23.0, 253.0): cluster 3\n", + "Example n. 1261 = (23.0, 252.0): cluster 3\n", + "Example n. 1262 = (23.0, 251.0): cluster 3\n", + "Example n. 1263 = (23.0, 250.0): cluster 3\n", + "Example n. 1264 = (23.0, 249.0): cluster 3\n", + "Example n. 1265 = (23.0, 248.0): cluster 3\n", + "Example n. 1266 = (23.0, 247.0): cluster 0\n", + "Example n. 1267 = (23.0, 246.0): cluster 0\n", + "Example n. 1268 = (23.0, 245.0): cluster 0\n", + "Example n. 1269 = (23.0, 244.0): cluster 0\n", + "Example n. 1270 = (23.0, 243.0): cluster 0\n", + "Example n. 1271 = (23.0, 242.0): cluster 0\n", + "Example n. 1272 = (23.0, 239.0): cluster 0\n", + "Example n. 1273 = (23.0, 238.0): cluster 0\n", + "Example n. 1274 = (23.0, 237.0): cluster 0\n", + "Example n. 1275 = (23.0, 236.0): cluster 0\n", + "Example n. 1276 = (23.0, 235.0): cluster 0\n", + "Example n. 1277 = (23.0, 234.0): cluster 0\n", + "Example n. 1278 = (23.0, 233.0): cluster 0\n", + "Example n. 1279 = (23.0, 232.0): cluster 0\n", + "Example n. 1280 = (23.0, 231.0): cluster 0\n", + "Example n. 1281 = (23.0, 230.0): cluster 0\n", + "Example n. 1282 = (23.0, 229.0): cluster 0\n", + "Example n. 1283 = (23.0, 228.0): cluster 0\n", + "Example n. 1284 = (23.0, 227.0): cluster 0\n", + "Example n. 1285 = (23.0, 226.0): cluster 0\n", + "Example n. 1286 = (23.0, 225.0): cluster 0\n", + "Example n. 1287 = (23.0, 224.0): cluster 0\n", + "Example n. 1288 = (23.0, 223.0): cluster 0\n", + "Example n. 1289 = (23.0, 222.0): cluster 0\n", + "Example n. 1290 = (23.0, 221.0): cluster 0\n", + "Example n. 1291 = (23.0, 220.0): cluster 0\n", + "Example n. 1292 = (23.0, 219.0): cluster 0\n", + "Example n. 1293 = (23.0, 218.0): cluster 0\n", + "Example n. 1294 = (23.0, 217.0): cluster 0\n", + "Example n. 1295 = (23.0, 215.0): cluster 0\n", + "Example n. 1296 = (23.0, 214.0): cluster 0\n", + "Example n. 1297 = (23.0, 213.0): cluster 0\n", + "Example n. 1298 = (23.0, 212.0): cluster 0\n", + "Example n. 1299 = (23.0, 211.0): cluster 0\n", + "Example n. 1300 = (23.0, 210.0): cluster 0\n", + "Example n. 1301 = (23.0, 209.0): cluster 0\n", + "Example n. 1302 = (23.0, 192.0): cluster 0\n", + "Example n. 1303 = (23.0, 191.0): cluster 0\n", + "Example n. 1304 = (23.0, 190.0): cluster 1\n", + "Example n. 1305 = (23.0, 189.0): cluster 1\n", + "Example n. 1306 = (23.0, 145.0): cluster 1\n", + "Example n. 1307 = (23.0, 144.0): cluster 1\n", + "Example n. 1308 = (23.0, 143.0): cluster 1\n", + "Example n. 1309 = (23.0, 142.0): cluster 1\n", + "Example n. 1310 = (23.0, 141.0): cluster 1\n", + "Example n. 1311 = (23.0, 140.0): cluster 1\n", + "Example n. 1312 = (23.0, 131.0): cluster 1\n", + "Example n. 1313 = (23.0, 130.0): cluster 1\n", + "Example n. 1314 = (23.0, 129.0): cluster 3\n", + "Example n. 1315 = (23.0, 128.0): cluster 3\n", + "Example n. 1316 = (23.0, 127.0): cluster 3\n", + "Example n. 1317 = (23.0, 126.0): cluster 3\n", + "Example n. 1318 = (23.0, 125.0): cluster 3\n", + "Example n. 1319 = (23.0, 124.0): cluster 3\n", + "Example n. 1320 = (23.0, 123.0): cluster 3\n", + "Example n. 1321 = (23.0, 122.0): cluster 3\n", + "Example n. 1322 = (24.0, 477.0): cluster 3\n", + "Example n. 1323 = (24.0, 476.0): cluster 3\n", + "Example n. 1324 = (24.0, 475.0): cluster 3\n", + "Example n. 1325 = (24.0, 474.0): cluster 3\n", + "Example n. 1326 = (24.0, 456.0): cluster 3\n", + "Example n. 1327 = (24.0, 455.0): cluster 3\n", + "Example n. 1328 = (24.0, 454.0): cluster 3\n", + "Example n. 1329 = (24.0, 453.0): cluster 3\n", + "Example n. 1330 = (24.0, 452.0): cluster 3\n", + "Example n. 1331 = (24.0, 388.0): cluster 3\n", + "Example n. 1332 = (24.0, 387.0): cluster 3\n", + "Example n. 1333 = (24.0, 386.0): cluster 3\n", + "Example n. 1334 = (24.0, 385.0): cluster 3\n", + "Example n. 1335 = (24.0, 381.0): cluster 3\n", + "Example n. 1336 = (24.0, 372.0): cluster 3\n", + "Example n. 1337 = (24.0, 371.0): cluster 3\n", + "Example n. 1338 = (24.0, 370.0): cluster 0\n", + "Example n. 1339 = (24.0, 369.0): cluster 0\n", + "Example n. 1340 = (24.0, 368.0): cluster 0\n", + "Example n. 1341 = (24.0, 367.0): cluster 0\n", + "Example n. 1342 = (24.0, 308.0): cluster 0\n", + "Example n. 1343 = (24.0, 307.0): cluster 0\n", + "Example n. 1344 = (24.0, 306.0): cluster 0\n", + "Example n. 1345 = (24.0, 305.0): cluster 0\n", + "Example n. 1346 = (24.0, 304.0): cluster 0\n", + "Example n. 1347 = (24.0, 303.0): cluster 0\n", + "Example n. 1348 = (24.0, 302.0): cluster 0\n", + "Example n. 1349 = (24.0, 301.0): cluster 0\n", + "Example n. 1350 = (24.0, 300.0): cluster 0\n", + "Example n. 1351 = (24.0, 299.0): cluster 0\n", + "Example n. 1352 = (24.0, 296.0): cluster 0\n", + "Example n. 1353 = (24.0, 295.0): cluster 0\n", + "Example n. 1354 = (24.0, 294.0): cluster 0\n", + "Example n. 1355 = (24.0, 293.0): cluster 0\n", + "Example n. 1356 = (24.0, 292.0): cluster 0\n", + "Example n. 1357 = (24.0, 291.0): cluster 0\n", + "Example n. 1358 = (24.0, 287.0): cluster 0\n", + "Example n. 1359 = (24.0, 286.0): cluster 0\n", + "Example n. 1360 = (24.0, 285.0): cluster 0\n", + "Example n. 1361 = (24.0, 284.0): cluster 0\n", + "Example n. 1362 = (24.0, 283.0): cluster 0\n", + "Example n. 1363 = (24.0, 282.0): cluster 0\n", + "Example n. 1364 = (24.0, 281.0): cluster 0\n", + "Example n. 1365 = (24.0, 280.0): cluster 0\n", + "Example n. 1366 = (24.0, 279.0): cluster 0\n", + "Example n. 1367 = (24.0, 278.0): cluster 0\n", + "Example n. 1368 = (24.0, 277.0): cluster 0\n", + "Example n. 1369 = (24.0, 276.0): cluster 0\n", + "Example n. 1370 = (24.0, 275.0): cluster 0\n", + "Example n. 1371 = (24.0, 274.0): cluster 0\n", + "Example n. 1372 = (24.0, 268.0): cluster 0\n", + "Example n. 1373 = (24.0, 267.0): cluster 0\n", + "Example n. 1374 = (24.0, 266.0): cluster 0\n", + "Example n. 1375 = (24.0, 265.0): cluster 0\n", + "Example n. 1376 = (24.0, 264.0): cluster 0\n", + "Example n. 1377 = (24.0, 262.0): cluster 1\n", + "Example n. 1378 = (24.0, 261.0): cluster 1\n", + "Example n. 1379 = (24.0, 260.0): cluster 1\n", + "Example n. 1380 = (24.0, 259.0): cluster 1\n", + "Example n. 1381 = (24.0, 258.0): cluster 1\n", + "Example n. 1382 = (24.0, 257.0): cluster 1\n", + "Example n. 1383 = (24.0, 254.0): cluster 1\n", + "Example n. 1384 = (24.0, 253.0): cluster 1\n", + "Example n. 1385 = (24.0, 252.0): cluster 1\n", + "Example n. 1386 = (24.0, 251.0): cluster 1\n", + "Example n. 1387 = (24.0, 250.0): cluster 1\n", + "Example n. 1388 = (24.0, 249.0): cluster 1\n", + "Example n. 1389 = (24.0, 248.0): cluster 1\n", + "Example n. 1390 = (24.0, 247.0): cluster 3\n", + "Example n. 1391 = (24.0, 246.0): cluster 3\n", + "Example n. 1392 = (24.0, 245.0): cluster 3\n", + "Example n. 1393 = (24.0, 244.0): cluster 3\n", + "Example n. 1394 = (24.0, 243.0): cluster 3\n", + "Example n. 1395 = (24.0, 242.0): cluster 3\n", + "Example n. 1396 = (24.0, 241.0): cluster 3\n", + "Example n. 1397 = (24.0, 240.0): cluster 3\n", + "Example n. 1398 = (24.0, 239.0): cluster 3\n", + "Example n. 1399 = (24.0, 238.0): cluster 3\n", + "Example n. 1400 = (24.0, 237.0): cluster 3\n", + "Example n. 1401 = (24.0, 236.0): cluster 3\n", + "Example n. 1402 = (24.0, 235.0): cluster 3\n", + "Example n. 1403 = (24.0, 234.0): cluster 3\n", + "Example n. 1404 = (24.0, 233.0): cluster 3\n", + "Example n. 1405 = (24.0, 232.0): cluster 3\n", + "Example n. 1406 = (24.0, 231.0): cluster 3\n", + "Example n. 1407 = (24.0, 230.0): cluster 3\n", + "Example n. 1408 = (24.0, 229.0): cluster 3\n", + "Example n. 1409 = (24.0, 228.0): cluster 3\n", + "Example n. 1410 = (24.0, 227.0): cluster 3\n", + "Example n. 1411 = (24.0, 226.0): cluster 3\n", + "Example n. 1412 = (24.0, 225.0): cluster 3\n", + "Example n. 1413 = (24.0, 224.0): cluster 3\n", + "Example n. 1414 = (24.0, 223.0): cluster 3\n", + "Example n. 1415 = (24.0, 222.0): cluster 3\n", + "Example n. 1416 = (24.0, 221.0): cluster 3\n", + "Example n. 1417 = (24.0, 220.0): cluster 3\n", + "Example n. 1418 = (24.0, 219.0): cluster 3\n", + "Example n. 1419 = (24.0, 218.0): cluster 3\n", + "Example n. 1420 = (24.0, 217.0): cluster 3\n", + "Example n. 1421 = (24.0, 216.0): cluster 3\n", + "Example n. 1422 = (24.0, 214.0): cluster 0\n", + "Example n. 1423 = (24.0, 213.0): cluster 0\n", + "Example n. 1424 = (24.0, 212.0): cluster 0\n", + "Example n. 1425 = (24.0, 211.0): cluster 0\n", + "Example n. 1426 = (24.0, 210.0): cluster 0\n", + "Example n. 1427 = (24.0, 209.0): cluster 0\n", + "Example n. 1428 = (24.0, 208.0): cluster 0\n", + "Example n. 1429 = (24.0, 205.0): cluster 0\n", + "Example n. 1430 = (24.0, 204.0): cluster 0\n", + "Example n. 1431 = (24.0, 203.0): cluster 0\n", + "Example n. 1432 = (24.0, 192.0): cluster 0\n", + "Example n. 1433 = (24.0, 191.0): cluster 0\n", + "Example n. 1434 = (24.0, 190.0): cluster 0\n", + "Example n. 1435 = (24.0, 189.0): cluster 0\n", + "Example n. 1436 = (24.0, 177.0): cluster 0\n", + "Example n. 1437 = (24.0, 176.0): cluster 0\n", + "Example n. 1438 = (24.0, 175.0): cluster 0\n", + "Example n. 1439 = (24.0, 174.0): cluster 0\n", + "Example n. 1440 = (24.0, 145.0): cluster 0\n", + "Example n. 1441 = (24.0, 144.0): cluster 0\n", + "Example n. 1442 = (24.0, 143.0): cluster 0\n", + "Example n. 1443 = (24.0, 142.0): cluster 0\n", + "Example n. 1444 = (24.0, 141.0): cluster 0\n", + "Example n. 1445 = (24.0, 140.0): cluster 0\n", + "Example n. 1446 = (24.0, 131.0): cluster 0\n", + "Example n. 1447 = (24.0, 130.0): cluster 0\n", + "Example n. 1448 = (24.0, 129.0): cluster 0\n", + "Example n. 1449 = (24.0, 128.0): cluster 0\n", + "Example n. 1450 = (24.0, 127.0): cluster 0\n", + "Example n. 1451 = (24.0, 126.0): cluster 0\n", + "Example n. 1452 = (24.0, 125.0): cluster 0\n", + "Example n. 1453 = (24.0, 124.0): cluster 0\n", + "Example n. 1454 = (24.0, 123.0): cluster 1\n", + "Example n. 1455 = (24.0, 122.0): cluster 1\n", + "Example n. 1456 = (24.0, 120.0): cluster 1\n", + "Example n. 1457 = (24.0, 117.0): cluster 1\n", + "Example n. 1458 = (24.0, 116.0): cluster 1\n", + "Example n. 1459 = (25.0, 478.0): cluster 1\n", + "Example n. 1460 = (25.0, 477.0): cluster 1\n", + "Example n. 1461 = (25.0, 476.0): cluster 1\n", + "Example n. 1462 = (25.0, 475.0): cluster 1\n", + "Example n. 1463 = (25.0, 474.0): cluster 1\n", + "Example n. 1464 = (25.0, 473.0): cluster 1\n", + "Example n. 1465 = (25.0, 472.0): cluster 1\n", + "Example n. 1466 = (25.0, 457.0): cluster 3\n", + "Example n. 1467 = (25.0, 456.0): cluster 3\n", + "Example n. 1468 = (25.0, 455.0): cluster 3\n", + "Example n. 1469 = (25.0, 454.0): cluster 3\n", + "Example n. 1470 = (25.0, 453.0): cluster 3\n", + "Example n. 1471 = (25.0, 452.0): cluster 3\n", + "Example n. 1472 = (25.0, 389.0): cluster 3\n", + "Example n. 1473 = (25.0, 388.0): cluster 3\n", + "Example n. 1474 = (25.0, 387.0): cluster 3\n", + "Example n. 1475 = (25.0, 386.0): cluster 3\n", + "Example n. 1476 = (25.0, 385.0): cluster 3\n", + "Example n. 1477 = (25.0, 384.0): cluster 3\n", + "Example n. 1478 = (25.0, 371.0): cluster 3\n", + "Example n. 1479 = (25.0, 370.0): cluster 3\n", + "Example n. 1480 = (25.0, 369.0): cluster 3\n", + "Example n. 1481 = (25.0, 368.0): cluster 3\n", + "Example n. 1482 = (25.0, 367.0): cluster 3\n", + "Example n. 1483 = (25.0, 308.0): cluster 3\n", + "Example n. 1484 = (25.0, 307.0): cluster 3\n", + "Example n. 1485 = (25.0, 306.0): cluster 3\n", + "Example n. 1486 = (25.0, 305.0): cluster 3\n", + "Example n. 1487 = (25.0, 304.0): cluster 3\n", + "Example n. 1488 = (25.0, 303.0): cluster 3\n", + "Example n. 1489 = (25.0, 302.0): cluster 3\n", + "Example n. 1490 = (25.0, 301.0): cluster 3\n", + "Example n. 1491 = (25.0, 300.0): cluster 3\n", + "Example n. 1492 = (25.0, 299.0): cluster 3\n", + "Example n. 1493 = (25.0, 296.0): cluster 3\n", + "Example n. 1494 = (25.0, 295.0): cluster 3\n", + "Example n. 1495 = (25.0, 294.0): cluster 3\n", + "Example n. 1496 = (25.0, 293.0): cluster 3\n", + "Example n. 1497 = (25.0, 292.0): cluster 3\n", + "Example n. 1498 = (25.0, 291.0): cluster 0\n", + "Example n. 1499 = (25.0, 287.0): cluster 0\n", + "Example n. 1500 = (25.0, 286.0): cluster 0\n", + "Example n. 1501 = (25.0, 285.0): cluster 0\n", + "Example n. 1502 = (25.0, 284.0): cluster 0\n", + "Example n. 1503 = (25.0, 283.0): cluster 0\n", + "Example n. 1504 = (25.0, 282.0): cluster 0\n", + "Example n. 1505 = (25.0, 281.0): cluster 0\n", + "Example n. 1506 = (25.0, 280.0): cluster 0\n", + "Example n. 1507 = (25.0, 279.0): cluster 0\n", + "Example n. 1508 = (25.0, 278.0): cluster 0\n", + "Example n. 1509 = (25.0, 277.0): cluster 0\n", + "Example n. 1510 = (25.0, 276.0): cluster 0\n", + "Example n. 1511 = (25.0, 275.0): cluster 0\n", + "Example n. 1512 = (25.0, 274.0): cluster 0\n", + "Example n. 1513 = (25.0, 273.0): cluster 0\n", + "Example n. 1514 = (25.0, 272.0): cluster 0\n", + "Example n. 1515 = (25.0, 271.0): cluster 0\n", + "Example n. 1516 = (25.0, 268.0): cluster 0\n", + "Example n. 1517 = (25.0, 267.0): cluster 0\n", + "Example n. 1518 = (25.0, 266.0): cluster 0\n", + "Example n. 1519 = (25.0, 265.0): cluster 0\n", + "Example n. 1520 = (25.0, 264.0): cluster 0\n", + "Example n. 1521 = (25.0, 262.0): cluster 0\n", + "Example n. 1522 = (25.0, 261.0): cluster 0\n", + "Example n. 1523 = (25.0, 260.0): cluster 0\n", + "Example n. 1524 = (25.0, 259.0): cluster 0\n", + "Example n. 1525 = (25.0, 258.0): cluster 0\n", + "Example n. 1526 = (25.0, 257.0): cluster 0\n", + "Example n. 1527 = (25.0, 253.0): cluster 0\n", + "Example n. 1528 = (25.0, 252.0): cluster 1\n", + "Example n. 1529 = (25.0, 251.0): cluster 1\n", + "Example n. 1530 = (25.0, 250.0): cluster 1\n", + "Example n. 1531 = (25.0, 249.0): cluster 1\n", + "Example n. 1532 = (25.0, 248.0): cluster 1\n", + "Example n. 1533 = (25.0, 247.0): cluster 1\n", + "Example n. 1534 = (25.0, 246.0): cluster 1\n", + "Example n. 1535 = (25.0, 245.0): cluster 1\n", + "Example n. 1536 = (25.0, 244.0): cluster 1\n", + "Example n. 1537 = (25.0, 243.0): cluster 1\n", + "Example n. 1538 = (25.0, 242.0): cluster 1\n", + "Example n. 1539 = (25.0, 241.0): cluster 1\n", + "Example n. 1540 = (25.0, 240.0): cluster 1\n", + "Example n. 1541 = (25.0, 239.0): cluster 1\n", + "Example n. 1542 = (25.0, 238.0): cluster 3\n", + "Example n. 1543 = (25.0, 237.0): cluster 3\n", + "Example n. 1544 = (25.0, 236.0): cluster 3\n", + "Example n. 1545 = (25.0, 234.0): cluster 3\n", + "Example n. 1546 = (25.0, 233.0): cluster 3\n", + "Example n. 1547 = (25.0, 232.0): cluster 3\n", + "Example n. 1548 = (25.0, 231.0): cluster 3\n", + "Example n. 1549 = (25.0, 230.0): cluster 3\n", + "Example n. 1550 = (25.0, 229.0): cluster 3\n", + "Example n. 1551 = (25.0, 228.0): cluster 3\n", + "Example n. 1552 = (25.0, 227.0): cluster 3\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Example n. 1553 = (25.0, 226.0): cluster 3\n", + "Example n. 1554 = (25.0, 225.0): cluster 3\n", + "Example n. 1555 = (25.0, 224.0): cluster 3\n", + "Example n. 1556 = (25.0, 220.0): cluster 3\n", + "Example n. 1557 = (25.0, 219.0): cluster 3\n", + "Example n. 1558 = (25.0, 218.0): cluster 3\n", + "Example n. 1559 = (25.0, 217.0): cluster 3\n", + "Example n. 1560 = (25.0, 216.0): cluster 3\n", + "Example n. 1561 = (25.0, 214.0): cluster 3\n", + "Example n. 1562 = (25.0, 213.0): cluster 3\n", + "Example n. 1563 = (25.0, 212.0): cluster 3\n", + "Example n. 1564 = (25.0, 211.0): cluster 3\n", + "Example n. 1565 = (25.0, 210.0): cluster 3\n", + "Example n. 1566 = (25.0, 209.0): cluster 3\n", + "Example n. 1567 = (25.0, 208.0): cluster 3\n", + "Example n. 1568 = (25.0, 206.0): cluster 3\n", + "Example n. 1569 = (25.0, 205.0): cluster 3\n", + "Example n. 1570 = (25.0, 204.0): cluster 3\n", + "Example n. 1571 = (25.0, 203.0): cluster 3\n", + "Example n. 1572 = (25.0, 202.0): cluster 3\n", + "Example n. 1573 = (25.0, 193.0): cluster 3\n", + "Example n. 1574 = (25.0, 192.0): cluster 0\n", + "Example n. 1575 = (25.0, 191.0): cluster 0\n", + "Example n. 1576 = (25.0, 190.0): cluster 0\n", + "Example n. 1577 = (25.0, 189.0): cluster 0\n", + "Example n. 1578 = (25.0, 188.0): cluster 0\n", + "Example n. 1579 = (25.0, 184.0): cluster 0\n", + "Example n. 1580 = (25.0, 183.0): cluster 0\n", + "Example n. 1581 = (25.0, 182.0): cluster 0\n", + "Example n. 1582 = (25.0, 178.0): cluster 0\n", + "Example n. 1583 = (25.0, 177.0): cluster 0\n", + "Example n. 1584 = (25.0, 176.0): cluster 0\n", + "Example n. 1585 = (25.0, 175.0): cluster 0\n", + "Example n. 1586 = (25.0, 174.0): cluster 0\n", + "Example n. 1587 = (25.0, 167.0): cluster 0\n", + "Example n. 1588 = (25.0, 166.0): cluster 0\n", + "Example n. 1589 = (25.0, 165.0): cluster 0\n", + "Example n. 1590 = (25.0, 164.0): cluster 0\n", + "Example n. 1591 = (25.0, 163.0): cluster 0\n", + "Example n. 1592 = (25.0, 145.0): cluster 0\n", + "Example n. 1593 = (25.0, 144.0): cluster 0\n", + "Example n. 1594 = (25.0, 143.0): cluster 0\n", + "Example n. 1595 = (25.0, 142.0): cluster 0\n", + "Example n. 1596 = (25.0, 141.0): cluster 0\n", + "Example n. 1597 = (25.0, 131.0): cluster 0\n", + "Example n. 1598 = (25.0, 130.0): cluster 0\n", + "Example n. 1599 = (25.0, 129.0): cluster 0\n", + "Example n. 1600 = (25.0, 128.0): cluster 0\n", + "Example n. 1601 = (25.0, 127.0): cluster 0\n", + "Example n. 1602 = (25.0, 124.0): cluster 1\n", + "Example n. 1603 = (25.0, 122.0): cluster 1\n", + "Example n. 1604 = (25.0, 121.0): cluster 1\n", + "Example n. 1605 = (25.0, 120.0): cluster 1\n", + "Example n. 1606 = (25.0, 119.0): cluster 1\n", + "Example n. 1607 = (25.0, 118.0): cluster 1\n", + "Example n. 1608 = (25.0, 117.0): cluster 1\n", + "Example n. 1609 = (25.0, 116.0): cluster 1\n", + "Example n. 1610 = (25.0, 115.0): cluster 1\n", + "Example n. 1611 = (26.0, 478.0): cluster 1\n", + "Example n. 1612 = (26.0, 477.0): cluster 1\n", + "Example n. 1613 = (26.0, 476.0): cluster 1\n", + "Example n. 1614 = (26.0, 475.0): cluster 1\n", + "Example n. 1615 = (26.0, 474.0): cluster 1\n", + "Example n. 1616 = (26.0, 473.0): cluster 1\n", + "Example n. 1617 = (26.0, 472.0): cluster 1\n", + "Example n. 1618 = (26.0, 471.0): cluster 3\n", + "Example n. 1619 = (26.0, 468.0): cluster 3\n", + "Example n. 1620 = (26.0, 467.0): cluster 3\n", + "Example n. 1621 = (26.0, 466.0): cluster 3\n", + "Example n. 1622 = (26.0, 465.0): cluster 3\n", + "Example n. 1623 = (26.0, 457.0): cluster 3\n", + "Example n. 1624 = (26.0, 456.0): cluster 3\n", + "Example n. 1625 = (26.0, 455.0): cluster 3\n", + "Example n. 1626 = (26.0, 454.0): cluster 3\n", + "Example n. 1627 = (26.0, 453.0): cluster 3\n", + "Example n. 1628 = (26.0, 452.0): cluster 3\n", + "Example n. 1629 = (26.0, 389.0): cluster 3\n", + "Example n. 1630 = (26.0, 388.0): cluster 3\n", + "Example n. 1631 = (26.0, 387.0): cluster 3\n", + "Example n. 1632 = (26.0, 386.0): cluster 3\n", + "Example n. 1633 = (26.0, 385.0): cluster 3\n", + "Example n. 1634 = (26.0, 384.0): cluster 3\n", + "Example n. 1635 = (26.0, 371.0): cluster 3\n", + "Example n. 1636 = (26.0, 370.0): cluster 3\n", + "Example n. 1637 = (26.0, 369.0): cluster 3\n", + "Example n. 1638 = (26.0, 368.0): cluster 3\n", + "Example n. 1639 = (26.0, 367.0): cluster 3\n", + "Example n. 1640 = (26.0, 366.0): cluster 3\n", + "Example n. 1641 = (26.0, 307.0): cluster 3\n", + "Example n. 1642 = (26.0, 306.0): cluster 3\n", + "Example n. 1643 = (26.0, 305.0): cluster 3\n", + "Example n. 1644 = (26.0, 303.0): cluster 3\n", + "Example n. 1645 = (26.0, 302.0): cluster 3\n", + "Example n. 1646 = (26.0, 301.0): cluster 3\n", + "Example n. 1647 = (26.0, 300.0): cluster 3\n", + "Example n. 1648 = (26.0, 299.0): cluster 3\n", + "Example n. 1649 = (26.0, 297.0): cluster 0\n", + "Example n. 1650 = (26.0, 296.0): cluster 0\n", + "Example n. 1651 = (26.0, 295.0): cluster 0\n", + "Example n. 1652 = (26.0, 294.0): cluster 0\n", + "Example n. 1653 = (26.0, 293.0): cluster 0\n", + "Example n. 1654 = (26.0, 292.0): cluster 0\n", + "Example n. 1655 = (26.0, 288.0): cluster 0\n", + "Example n. 1656 = (26.0, 287.0): cluster 0\n", + "Example n. 1657 = (26.0, 286.0): cluster 0\n", + "Example n. 1658 = (26.0, 285.0): cluster 0\n", + "Example n. 1659 = (26.0, 284.0): cluster 0\n", + "Example n. 1660 = (26.0, 283.0): cluster 0\n", + "Example n. 1661 = (26.0, 282.0): cluster 0\n", + "Example n. 1662 = (26.0, 281.0): cluster 0\n", + "Example n. 1663 = (26.0, 280.0): cluster 0\n", + "Example n. 1664 = (26.0, 279.0): cluster 0\n", + "Example n. 1665 = (26.0, 278.0): cluster 0\n", + "Example n. 1666 = (26.0, 277.0): cluster 0\n", + "Example n. 1667 = (26.0, 276.0): cluster 0\n", + "Example n. 1668 = (26.0, 275.0): cluster 0\n", + "Example n. 1669 = (26.0, 274.0): cluster 0\n", + "Example n. 1670 = (26.0, 273.0): cluster 0\n", + "Example n. 1671 = (26.0, 272.0): cluster 0\n", + "Example n. 1672 = (26.0, 271.0): cluster 0\n", + "Example n. 1673 = (26.0, 270.0): cluster 0\n", + "Example n. 1674 = (26.0, 269.0): cluster 1\n", + "Example n. 1675 = (26.0, 268.0): cluster 1\n", + "Example n. 1676 = (26.0, 267.0): cluster 1\n", + "Example n. 1677 = (26.0, 266.0): cluster 1\n", + "Example n. 1678 = (26.0, 265.0): cluster 1\n", + "Example n. 1679 = (26.0, 264.0): cluster 1\n", + "Example n. 1680 = (26.0, 261.0): cluster 1\n", + "Example n. 1681 = (26.0, 260.0): cluster 1\n", + "Example n. 1682 = (26.0, 259.0): cluster 1\n", + "Example n. 1683 = (26.0, 258.0): cluster 1\n", + "Example n. 1684 = (26.0, 254.0): cluster 1\n", + "Example n. 1685 = (26.0, 253.0): cluster 1\n", + "Example n. 1686 = (26.0, 252.0): cluster 1\n", + "Example n. 1687 = (26.0, 251.0): cluster 1\n", + "Example n. 1688 = (26.0, 250.0): cluster 1\n", + "Example n. 1689 = (26.0, 249.0): cluster 1\n", + "Example n. 1690 = (26.0, 248.0): cluster 1\n", + "Example n. 1691 = (26.0, 247.0): cluster 1\n", + "Example n. 1692 = (26.0, 246.0): cluster 1\n", + "Example n. 1693 = (26.0, 245.0): cluster 1\n", + "Example n. 1694 = (26.0, 244.0): cluster 1\n", + "Example n. 1695 = (26.0, 243.0): cluster 1\n", + "Example n. 1696 = (26.0, 242.0): cluster 1\n", + "Example n. 1697 = (26.0, 241.0): cluster 3\n", + "Example n. 1698 = (26.0, 240.0): cluster 3\n", + "Example n. 1699 = (26.0, 239.0): cluster 3\n", + "Example n. 1700 = (26.0, 238.0): cluster 3\n", + "Example n. 1701 = (26.0, 237.0): cluster 3\n", + "Example n. 1702 = (26.0, 236.0): cluster 3\n", + "Example n. 1703 = (26.0, 234.0): cluster 3\n", + "Example n. 1704 = (26.0, 233.0): cluster 3\n", + "Example n. 1705 = (26.0, 232.0): cluster 3\n", + "Example n. 1706 = (26.0, 231.0): cluster 3\n", + "Example n. 1707 = (26.0, 230.0): cluster 3\n", + "Example n. 1708 = (26.0, 229.0): cluster 3\n", + "Example n. 1709 = (26.0, 228.0): cluster 3\n", + "Example n. 1710 = (26.0, 227.0): cluster 3\n", + "Example n. 1711 = (26.0, 226.0): cluster 3\n", + "Example n. 1712 = (26.0, 225.0): cluster 3\n", + "Example n. 1713 = (26.0, 224.0): cluster 3\n", + "Example n. 1714 = (26.0, 220.0): cluster 3\n", + "Example n. 1715 = (26.0, 219.0): cluster 3\n", + "Example n. 1716 = (26.0, 218.0): cluster 3\n", + "Example n. 1717 = (26.0, 217.0): cluster 3\n", + "Example n. 1718 = (26.0, 216.0): cluster 3\n", + "Example n. 1719 = (26.0, 213.0): cluster 3\n", + "Example n. 1720 = (26.0, 212.0): cluster 3\n", + "Example n. 1721 = (26.0, 211.0): cluster 3\n", + "Example n. 1722 = (26.0, 210.0): cluster 3\n", + "Example n. 1723 = (26.0, 209.0): cluster 3\n", + "Example n. 1724 = (26.0, 208.0): cluster 3\n", + "Example n. 1725 = (26.0, 207.0): cluster 3\n", + "Example n. 1726 = (26.0, 206.0): cluster 3\n", + "Example n. 1727 = (26.0, 205.0): cluster 3\n", + "Example n. 1728 = (26.0, 204.0): cluster 3\n", + "Example n. 1729 = (26.0, 203.0): cluster 3\n", + "Example n. 1730 = (26.0, 202.0): cluster 3\n", + "Example n. 1731 = (26.0, 193.0): cluster 3\n", + "Example n. 1732 = (26.0, 192.0): cluster 0\n", + "Example n. 1733 = (26.0, 191.0): cluster 0\n", + "Example n. 1734 = (26.0, 190.0): cluster 0\n", + "Example n. 1735 = (26.0, 189.0): cluster 0\n", + "Example n. 1736 = (26.0, 188.0): cluster 0\n", + "Example n. 1737 = (26.0, 187.0): cluster 0\n", + "Example n. 1738 = (26.0, 185.0): cluster 0\n", + "Example n. 1739 = (26.0, 184.0): cluster 0\n", + "Example n. 1740 = (26.0, 183.0): cluster 0\n", + "Example n. 1741 = (26.0, 182.0): cluster 0\n", + "Example n. 1742 = (26.0, 181.0): cluster 0\n", + "Example n. 1743 = (26.0, 178.0): cluster 0\n", + "Example n. 1744 = (26.0, 177.0): cluster 0\n", + "Example n. 1745 = (26.0, 176.0): cluster 0\n", + "Example n. 1746 = (26.0, 175.0): cluster 0\n", + "Example n. 1747 = (26.0, 174.0): cluster 0\n", + "Example n. 1748 = (26.0, 173.0): cluster 0\n", + "Example n. 1749 = (26.0, 167.0): cluster 0\n", + "Example n. 1750 = (26.0, 166.0): cluster 0\n", + "Example n. 1751 = (26.0, 165.0): cluster 1\n", + "Example n. 1752 = (26.0, 164.0): cluster 1\n", + "Example n. 1753 = (26.0, 163.0): cluster 1\n", + "Example n. 1754 = (26.0, 162.0): cluster 1\n", + "Example n. 1755 = (26.0, 144.0): cluster 1\n", + "Example n. 1756 = (26.0, 143.0): cluster 1\n", + "Example n. 1757 = (26.0, 142.0): cluster 1\n", + "Example n. 1758 = (26.0, 141.0): cluster 1\n", + "Example n. 1759 = (26.0, 131.0): cluster 1\n", + "Example n. 1760 = (26.0, 130.0): cluster 1\n", + "Example n. 1761 = (26.0, 129.0): cluster 1\n", + "Example n. 1762 = (26.0, 128.0): cluster 1\n", + "Example n. 1763 = (26.0, 127.0): cluster 1\n", + "Example n. 1764 = (26.0, 122.0): cluster 1\n", + "Example n. 1765 = (26.0, 121.0): cluster 1\n", + "Example n. 1766 = (26.0, 120.0): cluster 1\n", + "Example n. 1767 = (26.0, 119.0): cluster 1\n", + "Example n. 1768 = (26.0, 118.0): cluster 1\n", + "Example n. 1769 = (26.0, 117.0): cluster 1\n", + "Example n. 1770 = (26.0, 116.0): cluster 1\n", + "Example n. 1771 = (26.0, 115.0): cluster 1\n", + "Example n. 1772 = (26.0, 114.0): cluster 3\n", + "Example n. 1773 = (27.0, 501.0): cluster 3\n", + "Example n. 1774 = (27.0, 500.0): cluster 3\n", + "Example n. 1775 = (27.0, 481.0): cluster 3\n", + "Example n. 1776 = (27.0, 480.0): cluster 3\n", + "Example n. 1777 = (27.0, 479.0): cluster 3\n", + "Example n. 1778 = (27.0, 478.0): cluster 3\n", + "Example n. 1779 = (27.0, 477.0): cluster 3\n", + "Example n. 1780 = (27.0, 476.0): cluster 3\n", + "Example n. 1781 = (27.0, 475.0): cluster 3\n", + "Example n. 1782 = (27.0, 474.0): cluster 3\n", + "Example n. 1783 = (27.0, 473.0): cluster 3\n", + "Example n. 1784 = (27.0, 472.0): cluster 3\n", + "Example n. 1785 = (27.0, 471.0): cluster 3\n", + "Example n. 1786 = (27.0, 470.0): cluster 3\n", + "Example n. 1787 = (27.0, 469.0): cluster 3\n", + "Example n. 1788 = (27.0, 468.0): cluster 3\n", + "Example n. 1789 = (27.0, 467.0): cluster 3\n", + "Example n. 1790 = (27.0, 466.0): cluster 3\n", + "Example n. 1791 = (27.0, 465.0): cluster 3\n", + "Example n. 1792 = (27.0, 464.0): cluster 3\n", + "Example n. 1793 = (27.0, 456.0): cluster 3\n", + "Example n. 1794 = (27.0, 455.0): cluster 3\n", + "Example n. 1795 = (27.0, 454.0): cluster 3\n", + "Example n. 1796 = (27.0, 453.0): cluster 3\n", + "Example n. 1797 = (27.0, 389.0): cluster 3\n", + "Example n. 1798 = (27.0, 388.0): cluster 3\n", + "Example n. 1799 = (27.0, 387.0): cluster 3\n", + "Example n. 1800 = (27.0, 386.0): cluster 3\n", + "Example n. 1801 = (27.0, 385.0): cluster 3\n", + "Example n. 1802 = (27.0, 384.0): cluster 3\n", + "Example n. 1803 = (27.0, 370.0): cluster 0\n", + "Example n. 1804 = (27.0, 369.0): cluster 0\n", + "Example n. 1805 = (27.0, 368.0): cluster 0\n", + "Example n. 1806 = (27.0, 367.0): cluster 0\n", + "Example n. 1807 = (27.0, 366.0): cluster 0\n", + "Example n. 1808 = (27.0, 302.0): cluster 0\n", + "Example n. 1809 = (27.0, 301.0): cluster 0\n", + "Example n. 1810 = (27.0, 297.0): cluster 0\n", + "Example n. 1811 = (27.0, 296.0): cluster 0\n", + "Example n. 1812 = (27.0, 295.0): cluster 0\n", + "Example n. 1813 = (27.0, 294.0): cluster 0\n", + "Example n. 1814 = (27.0, 293.0): cluster 0\n", + "Example n. 1815 = (27.0, 290.0): cluster 0\n", + "Example n. 1816 = (27.0, 289.0): cluster 0\n", + "Example n. 1817 = (27.0, 288.0): cluster 0\n", + "Example n. 1818 = (27.0, 287.0): cluster 0\n", + "Example n. 1819 = (27.0, 286.0): cluster 1\n", + "Example n. 1820 = (27.0, 285.0): cluster 1\n", + "Example n. 1821 = (27.0, 284.0): cluster 1\n", + "Example n. 1822 = (27.0, 283.0): cluster 1\n", + "Example n. 1823 = (27.0, 282.0): cluster 1\n", + "Example n. 1824 = (27.0, 281.0): cluster 1\n", + "Example n. 1825 = (27.0, 280.0): cluster 1\n", + "Example n. 1826 = (27.0, 279.0): cluster 1\n", + "Example n. 1827 = (27.0, 278.0): cluster 1\n", + "Example n. 1828 = (27.0, 277.0): cluster 1\n", + "Example n. 1829 = (27.0, 276.0): cluster 1\n", + "Example n. 1830 = (27.0, 275.0): cluster 1\n", + "Example n. 1831 = (27.0, 274.0): cluster 1\n", + "Example n. 1832 = (27.0, 273.0): cluster 1\n", + "Example n. 1833 = (27.0, 272.0): cluster 1\n", + "Example n. 1834 = (27.0, 271.0): cluster 1\n", + "Example n. 1835 = (27.0, 270.0): cluster 1\n", + "Example n. 1836 = (27.0, 269.0): cluster 1\n", + "Example n. 1837 = (27.0, 268.0): cluster 1\n", + "Example n. 1838 = (27.0, 267.0): cluster 1\n", + "Example n. 1839 = (27.0, 266.0): cluster 1\n", + "Example n. 1840 = (27.0, 265.0): cluster 1\n", + "Example n. 1841 = (27.0, 264.0): cluster 1\n", + "Example n. 1842 = (27.0, 260.0): cluster 1\n", + "Example n. 1843 = (27.0, 259.0): cluster 3\n", + "Example n. 1844 = (27.0, 254.0): cluster 3\n", + "Example n. 1845 = (27.0, 253.0): cluster 3\n", + "Example n. 1846 = (27.0, 252.0): cluster 3\n", + "Example n. 1847 = (27.0, 251.0): cluster 3\n", + "Example n. 1848 = (27.0, 250.0): cluster 3\n", + "Example n. 1849 = (27.0, 249.0): cluster 3\n", + "Example n. 1850 = (27.0, 248.0): cluster 3\n", + "Example n. 1851 = (27.0, 247.0): cluster 3\n", + "Example n. 1852 = (27.0, 246.0): cluster 3\n", + "Example n. 1853 = (27.0, 245.0): cluster 3\n", + "Example n. 1854 = (27.0, 244.0): cluster 3\n", + "Example n. 1855 = (27.0, 243.0): cluster 3\n", + "Example n. 1856 = (27.0, 242.0): cluster 3\n", + "Example n. 1857 = (27.0, 241.0): cluster 3\n", + "Example n. 1858 = (27.0, 240.0): cluster 3\n", + "Example n. 1859 = (27.0, 239.0): cluster 3\n", + "Example n. 1860 = (27.0, 238.0): cluster 3\n", + "Example n. 1861 = (27.0, 237.0): cluster 3\n", + "Example n. 1862 = (27.0, 236.0): cluster 3\n", + "Example n. 1863 = (27.0, 235.0): cluster 3\n", + "Example n. 1864 = (27.0, 234.0): cluster 3\n", + "Example n. 1865 = (27.0, 233.0): cluster 3\n", + "Example n. 1866 = (27.0, 232.0): cluster 3\n", + "Example n. 1867 = (27.0, 231.0): cluster 3\n", + "Example n. 1868 = (27.0, 230.0): cluster 3\n", + "Example n. 1869 = (27.0, 229.0): cluster 3\n", + "Example n. 1870 = (27.0, 226.0): cluster 3\n", + "Example n. 1871 = (27.0, 220.0): cluster 3\n", + "Example n. 1872 = (27.0, 219.0): cluster 3\n", + "Example n. 1873 = (27.0, 218.0): cluster 3\n", + "Example n. 1874 = (27.0, 217.0): cluster 0\n", + "Example n. 1875 = (27.0, 216.0): cluster 0\n", + "Example n. 1876 = (27.0, 215.0): cluster 0\n", + "Example n. 1877 = (27.0, 212.0): cluster 0\n", + "Example n. 1878 = (27.0, 211.0): cluster 0\n", + "Example n. 1879 = (27.0, 210.0): cluster 0\n", + "Example n. 1880 = (27.0, 209.0): cluster 0\n", + "Example n. 1881 = (27.0, 208.0): cluster 0\n", + "Example n. 1882 = (27.0, 206.0): cluster 0\n", + "Example n. 1883 = (27.0, 205.0): cluster 0\n", + "Example n. 1884 = (27.0, 204.0): cluster 0\n", + "Example n. 1885 = (27.0, 203.0): cluster 0\n", + "Example n. 1886 = (27.0, 202.0): cluster 0\n", + "Example n. 1887 = (27.0, 201.0): cluster 0\n", + "Example n. 1888 = (27.0, 192.0): cluster 0\n", + "Example n. 1889 = (27.0, 191.0): cluster 0\n", + "Example n. 1890 = (27.0, 190.0): cluster 1\n", + "Example n. 1891 = (27.0, 189.0): cluster 1\n", + "Example n. 1892 = (27.0, 188.0): cluster 1\n", + "Example n. 1893 = (27.0, 187.0): cluster 1\n", + "Example n. 1894 = (27.0, 186.0): cluster 1\n", + "Example n. 1895 = (27.0, 185.0): cluster 1\n", + "Example n. 1896 = (27.0, 184.0): cluster 1\n", + "Example n. 1897 = (27.0, 183.0): cluster 1\n", + "Example n. 1898 = (27.0, 182.0): cluster 1\n", + "Example n. 1899 = (27.0, 181.0): cluster 1\n", + "Example n. 1900 = (27.0, 178.0): cluster 1\n", + "Example n. 1901 = (27.0, 177.0): cluster 1\n", + "Example n. 1902 = (27.0, 176.0): cluster 1\n", + "Example n. 1903 = (27.0, 175.0): cluster 1\n", + "Example n. 1904 = (27.0, 174.0): cluster 1\n", + "Example n. 1905 = (27.0, 173.0): cluster 1\n", + "Example n. 1906 = (27.0, 168.0): cluster 1\n", + "Example n. 1907 = (27.0, 167.0): cluster 1\n", + "Example n. 1908 = (27.0, 166.0): cluster 1\n", + "Example n. 1909 = (27.0, 165.0): cluster 1\n", + "Example n. 1910 = (27.0, 164.0): cluster 1\n", + "Example n. 1911 = (27.0, 163.0): cluster 1\n", + "Example n. 1912 = (27.0, 162.0): cluster 1\n", + "Example n. 1913 = (27.0, 161.0): cluster 1\n", + "Example n. 1914 = (27.0, 130.0): cluster 3\n", + "Example n. 1915 = (27.0, 129.0): cluster 3\n", + "Example n. 1916 = (27.0, 128.0): cluster 3\n", + "Example n. 1917 = (27.0, 122.0): cluster 3\n", + "Example n. 1918 = (27.0, 121.0): cluster 3\n", + "Example n. 1919 = (27.0, 120.0): cluster 3\n", + "Example n. 1920 = (27.0, 119.0): cluster 3\n", + "Example n. 1921 = (27.0, 118.0): cluster 3\n", + "Example n. 1922 = (27.0, 117.0): cluster 3\n", + "Example n. 1923 = (27.0, 116.0): cluster 3\n", + "Example n. 1924 = (27.0, 115.0): cluster 3\n", + "Example n. 1925 = (27.0, 114.0): cluster 3\n", + "Example n. 1926 = (28.0, 502.0): cluster 3\n", + "Example n. 1927 = (28.0, 501.0): cluster 3\n", + "Example n. 1928 = (28.0, 500.0): cluster 3\n", + "Example n. 1929 = (28.0, 499.0): cluster 3\n", + "Example n. 1930 = (28.0, 498.0): cluster 3\n", + "Example n. 1931 = (28.0, 483.0): cluster 3\n", + "Example n. 1932 = (28.0, 482.0): cluster 3\n", + "Example n. 1933 = (28.0, 481.0): cluster 3\n", + "Example n. 1934 = (28.0, 480.0): cluster 3\n", + "Example n. 1935 = (28.0, 479.0): cluster 3\n", + "Example n. 1936 = (28.0, 478.0): cluster 3\n", + "Example n. 1937 = (28.0, 477.0): cluster 3\n", + "Example n. 1938 = (28.0, 476.0): cluster 3\n", + "Example n. 1939 = (28.0, 475.0): cluster 3\n", + "Example n. 1940 = (28.0, 474.0): cluster 3\n", + "Example n. 1941 = (28.0, 473.0): cluster 3\n", + "Example n. 1942 = (28.0, 472.0): cluster 3\n", + "Example n. 1943 = (28.0, 471.0): cluster 3\n", + "Example n. 1944 = (28.0, 470.0): cluster 3\n", + "Example n. 1945 = (28.0, 469.0): cluster 0\n", + "Example n. 1946 = (28.0, 468.0): cluster 0\n", + "Example n. 1947 = (28.0, 467.0): cluster 0\n", + "Example n. 1948 = (28.0, 466.0): cluster 0\n", + "Example n. 1949 = (28.0, 465.0): cluster 0\n", + "Example n. 1950 = (28.0, 464.0): cluster 0\n", + "Example n. 1951 = (28.0, 455.0): cluster 0\n", + "Example n. 1952 = (28.0, 454.0): cluster 0\n", + "Example n. 1953 = (28.0, 388.0): cluster 0\n", + "Example n. 1954 = (28.0, 387.0): cluster 0\n", + "Example n. 1955 = (28.0, 386.0): cluster 0\n", + "Example n. 1956 = (28.0, 385.0): cluster 0\n", + "Example n. 1957 = (28.0, 370.0): cluster 0\n", + "Example n. 1958 = (28.0, 369.0): cluster 0\n", + "Example n. 1959 = (28.0, 368.0): cluster 0\n", + "Example n. 1960 = (28.0, 367.0): cluster 0\n", + "Example n. 1961 = (28.0, 366.0): cluster 0\n", + "Example n. 1962 = (28.0, 297.0): cluster 0\n", + "Example n. 1963 = (28.0, 296.0): cluster 0\n", + "Example n. 1964 = (28.0, 295.0): cluster 0\n", + "Example n. 1965 = (28.0, 294.0): cluster 1\n", + "Example n. 1966 = (28.0, 293.0): cluster 1\n", + "Example n. 1967 = (28.0, 291.0): cluster 1\n", + "Example n. 1968 = (28.0, 290.0): cluster 1\n", + "Example n. 1969 = (28.0, 289.0): cluster 1\n", + "Example n. 1970 = (28.0, 288.0): cluster 1\n", + "Example n. 1971 = (28.0, 287.0): cluster 1\n", + "Example n. 1972 = (28.0, 286.0): cluster 1\n", + "Example n. 1973 = (28.0, 285.0): cluster 1\n", + "Example n. 1974 = (28.0, 284.0): cluster 1\n", + "Example n. 1975 = (28.0, 283.0): cluster 1\n", + "Example n. 1976 = (28.0, 282.0): cluster 1\n", + "Example n. 1977 = (28.0, 281.0): cluster 1\n", + "Example n. 1978 = (28.0, 280.0): cluster 1\n", + "Example n. 1979 = (28.0, 279.0): cluster 1\n", + "Example n. 1980 = (28.0, 278.0): cluster 1\n", + "Example n. 1981 = (28.0, 277.0): cluster 1\n", + "Example n. 1982 = (28.0, 276.0): cluster 1\n", + "Example n. 1983 = (28.0, 275.0): cluster 1\n", + "Example n. 1984 = (28.0, 274.0): cluster 1\n", + "Example n. 1985 = (28.0, 273.0): cluster 1\n", + "Example n. 1986 = (28.0, 272.0): cluster 1\n", + "Example n. 1987 = (28.0, 271.0): cluster 1\n", + "Example n. 1988 = (28.0, 270.0): cluster 1\n", + "Example n. 1989 = (28.0, 269.0): cluster 3\n", + "Example n. 1990 = (28.0, 268.0): cluster 3\n", + "Example n. 1991 = (28.0, 267.0): cluster 3\n", + "Example n. 1992 = (28.0, 266.0): cluster 3\n", + "Example n. 1993 = (28.0, 265.0): cluster 3\n", + "Example n. 1994 = (28.0, 264.0): cluster 3\n", + "Example n. 1995 = (28.0, 257.0): cluster 3\n", + "Example n. 1996 = (28.0, 256.0): cluster 3\n", + "Example n. 1997 = (28.0, 255.0): cluster 3\n", + "Example n. 1998 = (28.0, 254.0): cluster 3\n", + "Example n. 1999 = (28.0, 253.0): cluster 3\n", + "Example n. 2000 = (28.0, 252.0): cluster 3\n", + "Example n. 2001 = (28.0, 251.0): cluster 3\n", + "Example n. 2002 = (28.0, 250.0): cluster 3\n", + "Example n. 2003 = (28.0, 249.0): cluster 3\n", + "Example n. 2004 = (28.0, 248.0): cluster 3\n", + "Example n. 2005 = (28.0, 247.0): cluster 3\n", + "Example n. 2006 = (28.0, 246.0): cluster 3\n", + "Example n. 2007 = (28.0, 245.0): cluster 3\n", + "Example n. 2008 = (28.0, 244.0): cluster 3\n", + "Example n. 2009 = (28.0, 243.0): cluster 3\n", + "Example n. 2010 = (28.0, 242.0): cluster 3\n", + "Example n. 2011 = (28.0, 241.0): cluster 3\n", + "Example n. 2012 = (28.0, 240.0): cluster 3\n", + "Example n. 2013 = (28.0, 239.0): cluster 3\n", + "Example n. 2014 = (28.0, 238.0): cluster 3\n", + "Example n. 2015 = (28.0, 237.0): cluster 3\n", + "Example n. 2016 = (28.0, 236.0): cluster 3\n", + "Example n. 2017 = (28.0, 235.0): cluster 3\n", + "Example n. 2018 = (28.0, 234.0): cluster 3\n", + "Example n. 2019 = (28.0, 233.0): cluster 3\n", + "Example n. 2020 = (28.0, 232.0): cluster 3\n", + "Example n. 2021 = (28.0, 231.0): cluster 3\n", + "Example n. 2022 = (28.0, 230.0): cluster 3\n", + "Example n. 2023 = (28.0, 229.0): cluster 3\n", + "Example n. 2024 = (28.0, 228.0): cluster 0\n", + "Example n. 2025 = (28.0, 227.0): cluster 0\n", + "Example n. 2026 = (28.0, 226.0): cluster 0\n", + "Example n. 2027 = (28.0, 225.0): cluster 0\n", + "Example n. 2028 = (28.0, 224.0): cluster 0\n", + "Example n. 2029 = (28.0, 220.0): cluster 0\n", + "Example n. 2030 = (28.0, 219.0): cluster 0\n", + "Example n. 2031 = (28.0, 218.0): cluster 0\n", + "Example n. 2032 = (28.0, 217.0): cluster 0\n", + "Example n. 2033 = (28.0, 216.0): cluster 0\n", + "Example n. 2034 = (28.0, 215.0): cluster 0\n", + "Example n. 2035 = (28.0, 214.0): cluster 0\n", + "Example n. 2036 = (28.0, 211.0): cluster 0\n", + "Example n. 2037 = (28.0, 210.0): cluster 0\n", + "Example n. 2038 = (28.0, 209.0): cluster 0\n", + "Example n. 2039 = (28.0, 206.0): cluster 0\n", + "Example n. 2040 = (28.0, 205.0): cluster 0\n", + "Example n. 2041 = (28.0, 204.0): cluster 0\n", + "Example n. 2042 = (28.0, 203.0): cluster 0\n", + "Example n. 2043 = (28.0, 202.0): cluster 0\n", + "Example n. 2044 = (28.0, 201.0): cluster 1\n", + "Example n. 2045 = (28.0, 200.0): cluster 1\n", + "Example n. 2046 = (28.0, 199.0): cluster 1\n", + "Example n. 2047 = (28.0, 196.0): cluster 1\n", + "Example n. 2048 = (28.0, 193.0): cluster 1\n", + "Example n. 2049 = (28.0, 192.0): cluster 1\n", + "Example n. 2050 = (28.0, 191.0): cluster 1\n", + "Example n. 2051 = (28.0, 190.0): cluster 1\n", + "Example n. 2052 = (28.0, 189.0): cluster 1\n", + "Example n. 2053 = (28.0, 188.0): cluster 1\n", + "Example n. 2054 = (28.0, 187.0): cluster 1\n", + "Example n. 2055 = (28.0, 186.0): cluster 1\n", + "Example n. 2056 = (28.0, 185.0): cluster 1\n", + "Example n. 2057 = (28.0, 184.0): cluster 1\n", + "Example n. 2058 = (28.0, 183.0): cluster 1\n", + "Example n. 2059 = (28.0, 182.0): cluster 1\n", + "Example n. 2060 = (28.0, 181.0): cluster 3\n", + "Example n. 2061 = (28.0, 177.0): cluster 3\n", + "Example n. 2062 = (28.0, 176.0): cluster 3\n", + "Example n. 2063 = (28.0, 175.0): cluster 3\n", + "Example n. 2064 = (28.0, 174.0): cluster 3\n", + "Example n. 2065 = (28.0, 168.0): cluster 3\n", + "Example n. 2066 = (28.0, 167.0): cluster 3\n", + "Example n. 2067 = (28.0, 166.0): cluster 3\n", + "Example n. 2068 = (28.0, 165.0): cluster 3\n", + "Example n. 2069 = (28.0, 164.0): cluster 3\n", + "Example n. 2070 = (28.0, 163.0): cluster 3\n", + "Example n. 2071 = (28.0, 162.0): cluster 3\n", + "Example n. 2072 = (28.0, 161.0): cluster 3\n", + "Example n. 2073 = (28.0, 122.0): cluster 3\n", + "Example n. 2074 = (28.0, 121.0): cluster 3\n", + "Example n. 2075 = (28.0, 120.0): cluster 3\n", + "Example n. 2076 = (28.0, 119.0): cluster 3\n", + "Example n. 2077 = (28.0, 118.0): cluster 3\n", + "Example n. 2078 = (28.0, 117.0): cluster 3\n", + "Example n. 2079 = (28.0, 116.0): cluster 3\n", + "Example n. 2080 = (28.0, 115.0): cluster 3\n", + "Example n. 2081 = (28.0, 114.0): cluster 3\n", + "Example n. 2082 = (29.0, 503.0): cluster 3\n", + "Example n. 2083 = (29.0, 502.0): cluster 3\n", + "Example n. 2084 = (29.0, 501.0): cluster 3\n", + "Example n. 2085 = (29.0, 500.0): cluster 3\n", + "Example n. 2086 = (29.0, 499.0): cluster 3\n", + "Example n. 2087 = (29.0, 498.0): cluster 3\n", + "Example n. 2088 = (29.0, 484.0): cluster 3\n", + "Example n. 2089 = (29.0, 483.0): cluster 3\n", + "Example n. 2090 = (29.0, 482.0): cluster 3\n", + "Example n. 2091 = (29.0, 481.0): cluster 3\n", + "Example n. 2092 = (29.0, 480.0): cluster 0\n", + "Example n. 2093 = (29.0, 479.0): cluster 0\n", + "Example n. 2094 = (29.0, 478.0): cluster 0\n", + "Example n. 2095 = (29.0, 477.0): cluster 0\n", + "Example n. 2096 = (29.0, 476.0): cluster 0\n", + "Example n. 2097 = (29.0, 475.0): cluster 0\n", + "Example n. 2098 = (29.0, 474.0): cluster 0\n", + "Example n. 2099 = (29.0, 473.0): cluster 0\n", + "Example n. 2100 = (29.0, 472.0): cluster 0\n", + "Example n. 2101 = (29.0, 471.0): cluster 0\n", + "Example n. 2102 = (29.0, 470.0): cluster 0\n", + "Example n. 2103 = (29.0, 469.0): cluster 0\n", + "Example n. 2104 = (29.0, 468.0): cluster 0\n", + "Example n. 2105 = (29.0, 467.0): cluster 0\n", + "Example n. 2106 = (29.0, 466.0): cluster 0\n", + "Example n. 2107 = (29.0, 465.0): cluster 0\n", + "Example n. 2108 = (29.0, 464.0): cluster 0\n", + "Example n. 2109 = (29.0, 463.0): cluster 1\n", + "Example n. 2110 = (29.0, 370.0): cluster 1\n", + "Example n. 2111 = (29.0, 369.0): cluster 1\n", + "Example n. 2112 = (29.0, 368.0): cluster 1\n", + "Example n. 2113 = (29.0, 367.0): cluster 1\n", + "Example n. 2114 = (29.0, 366.0): cluster 1\n", + "Example n. 2115 = (29.0, 297.0): cluster 1\n", + "Example n. 2116 = (29.0, 296.0): cluster 1\n", + "Example n. 2117 = (29.0, 295.0): cluster 1\n", + "Example n. 2118 = (29.0, 294.0): cluster 1\n", + "Example n. 2119 = (29.0, 293.0): cluster 1\n", + "Example n. 2120 = (29.0, 292.0): cluster 1\n", + "Example n. 2121 = (29.0, 291.0): cluster 1\n", + "Example n. 2122 = (29.0, 290.0): cluster 1\n", + "Example n. 2123 = (29.0, 289.0): cluster 1\n", + "Example n. 2124 = (29.0, 288.0): cluster 1\n", + "Example n. 2125 = (29.0, 287.0): cluster 1\n", + "Example n. 2126 = (29.0, 286.0): cluster 1\n", + "Example n. 2127 = (29.0, 285.0): cluster 1\n", + "Example n. 2128 = (29.0, 284.0): cluster 3\n", + "Example n. 2129 = (29.0, 280.0): cluster 3\n", + "Example n. 2130 = (29.0, 279.0): cluster 3\n", + "Example n. 2131 = (29.0, 278.0): cluster 3\n", + "Example n. 2132 = (29.0, 277.0): cluster 3\n", + "Example n. 2133 = (29.0, 276.0): cluster 3\n", + "Example n. 2134 = (29.0, 274.0): cluster 3\n", + "Example n. 2135 = (29.0, 273.0): cluster 3\n", + "Example n. 2136 = (29.0, 272.0): cluster 3\n", + "Example n. 2137 = (29.0, 271.0): cluster 3\n", + "Example n. 2138 = (29.0, 270.0): cluster 3\n", + "Example n. 2139 = (29.0, 269.0): cluster 3\n", + "Example n. 2140 = (29.0, 268.0): cluster 3\n", + "Example n. 2141 = (29.0, 267.0): cluster 3\n", + "Example n. 2142 = (29.0, 266.0): cluster 3\n", + "Example n. 2143 = (29.0, 265.0): cluster 3\n", + "Example n. 2144 = (29.0, 264.0): cluster 3\n", + "Example n. 2145 = (29.0, 263.0): cluster 3\n", + "Example n. 2146 = (29.0, 262.0): cluster 3\n", + "Example n. 2147 = (29.0, 261.0): cluster 3\n", + "Example n. 2148 = (29.0, 260.0): cluster 3\n", + "Example n. 2149 = (29.0, 259.0): cluster 3\n", + "Example n. 2150 = (29.0, 258.0): cluster 3\n", + "Example n. 2151 = (29.0, 257.0): cluster 3\n", + "Example n. 2152 = (29.0, 256.0): cluster 3\n", + "Example n. 2153 = (29.0, 255.0): cluster 3\n", + "Example n. 2154 = (29.0, 254.0): cluster 3\n", + "Example n. 2155 = (29.0, 253.0): cluster 3\n", + "Example n. 2156 = (29.0, 252.0): cluster 3\n", + "Example n. 2157 = (29.0, 251.0): cluster 3\n", + "Example n. 2158 = (29.0, 250.0): cluster 3\n", + "Example n. 2159 = (29.0, 249.0): cluster 3\n", + "Example n. 2160 = (29.0, 248.0): cluster 0\n", + "Example n. 2161 = (29.0, 247.0): cluster 0\n", + "Example n. 2162 = (29.0, 246.0): cluster 0\n", + "Example n. 2163 = (29.0, 245.0): cluster 0\n", + "Example n. 2164 = (29.0, 244.0): cluster 0\n", + "Example n. 2165 = (29.0, 243.0): cluster 0\n", + "Example n. 2166 = (29.0, 242.0): cluster 0\n", + "Example n. 2167 = (29.0, 241.0): cluster 0\n", + "Example n. 2168 = (29.0, 240.0): cluster 0\n", + "Example n. 2169 = (29.0, 239.0): cluster 0\n", + "Example n. 2170 = (29.0, 238.0): cluster 0\n", + "Example n. 2171 = (29.0, 237.0): cluster 0\n", + "Example n. 2172 = (29.0, 236.0): cluster 0\n", + "Example n. 2173 = (29.0, 235.0): cluster 0\n", + "Example n. 2174 = (29.0, 234.0): cluster 1\n", + "Example n. 2175 = (29.0, 233.0): cluster 1\n", + "Example n. 2176 = (29.0, 232.0): cluster 1\n", + "Example n. 2177 = (29.0, 231.0): cluster 1\n", + "Example n. 2178 = (29.0, 230.0): cluster 1\n", + "Example n. 2179 = (29.0, 229.0): cluster 1\n", + "Example n. 2180 = (29.0, 228.0): cluster 1\n", + "Example n. 2181 = (29.0, 227.0): cluster 1\n", + "Example n. 2182 = (29.0, 226.0): cluster 1\n", + "Example n. 2183 = (29.0, 225.0): cluster 1\n", + "Example n. 2184 = (29.0, 224.0): cluster 1\n", + "Example n. 2185 = (29.0, 218.0): cluster 1\n", + "Example n. 2186 = (29.0, 217.0): cluster 1\n", + "Example n. 2187 = (29.0, 216.0): cluster 1\n", + "Example n. 2188 = (29.0, 215.0): cluster 1\n", + "Example n. 2189 = (29.0, 214.0): cluster 1\n", + "Example n. 2190 = (29.0, 205.0): cluster 1\n", + "Example n. 2191 = (29.0, 204.0): cluster 1\n", + "Example n. 2192 = (29.0, 203.0): cluster 1\n", + "Example n. 2193 = (29.0, 202.0): cluster 1\n", + "Example n. 2194 = (29.0, 201.0): cluster 1\n", + "Example n. 2195 = (29.0, 200.0): cluster 1\n", + "Example n. 2196 = (29.0, 199.0): cluster 3\n", + "Example n. 2197 = (29.0, 198.0): cluster 3\n", + "Example n. 2198 = (29.0, 197.0): cluster 3\n", + "Example n. 2199 = (29.0, 196.0): cluster 3\n", + "Example n. 2200 = (29.0, 195.0): cluster 3\n", + "Example n. 2201 = (29.0, 194.0): cluster 3\n", + "Example n. 2202 = (29.0, 193.0): cluster 3\n", + "Example n. 2203 = (29.0, 192.0): cluster 3\n", + "Example n. 2204 = (29.0, 191.0): cluster 3\n", + "Example n. 2205 = (29.0, 190.0): cluster 3\n", + "Example n. 2206 = (29.0, 189.0): cluster 3\n", + "Example n. 2207 = (29.0, 188.0): cluster 3\n", + "Example n. 2208 = (29.0, 187.0): cluster 3\n", + "Example n. 2209 = (29.0, 186.0): cluster 3\n", + "Example n. 2210 = (29.0, 185.0): cluster 3\n", + "Example n. 2211 = (29.0, 184.0): cluster 3\n", + "Example n. 2212 = (29.0, 183.0): cluster 3\n", + "Example n. 2213 = (29.0, 182.0): cluster 3\n", + "Example n. 2214 = (29.0, 181.0): cluster 3\n", + "Example n. 2215 = (29.0, 177.0): cluster 3\n", + "Example n. 2216 = (29.0, 176.0): cluster 3\n", + "Example n. 2217 = (29.0, 175.0): cluster 3\n", + "Example n. 2218 = (29.0, 167.0): cluster 3\n", + "Example n. 2219 = (29.0, 166.0): cluster 3\n", + "Example n. 2220 = (29.0, 165.0): cluster 3\n", + "Example n. 2221 = (29.0, 164.0): cluster 3\n", + "Example n. 2222 = (29.0, 163.0): cluster 3\n", + "Example n. 2223 = (29.0, 162.0): cluster 3\n", + "Example n. 2224 = (29.0, 161.0): cluster 3\n", + "Example n. 2225 = (29.0, 121.0): cluster 3\n", + "Example n. 2226 = (29.0, 120.0): cluster 3\n", + "Example n. 2227 = (29.0, 119.0): cluster 3\n", + "Example n. 2228 = (29.0, 118.0): cluster 0\n", + "Example n. 2229 = (29.0, 117.0): cluster 0\n", + "Example n. 2230 = (29.0, 116.0): cluster 0\n", + "Example n. 2231 = (29.0, 115.0): cluster 0\n", + "Example n. 2232 = (30.0, 503.0): cluster 0\n", + "Example n. 2233 = (30.0, 502.0): cluster 0\n", + "Example n. 2234 = (30.0, 501.0): cluster 0\n", + "Example n. 2235 = (30.0, 500.0): cluster 0\n", + "Example n. 2236 = (30.0, 499.0): cluster 0\n", + "Example n. 2237 = (30.0, 498.0): cluster 0\n", + "Example n. 2238 = (30.0, 486.0): cluster 0\n", + "Example n. 2239 = (30.0, 485.0): cluster 0\n", + "Example n. 2240 = (30.0, 484.0): cluster 1\n", + "Example n. 2241 = (30.0, 483.0): cluster 1\n", + "Example n. 2242 = (30.0, 482.0): cluster 1\n", + "Example n. 2243 = (30.0, 481.0): cluster 1\n", + "Example n. 2244 = (30.0, 480.0): cluster 1\n", + "Example n. 2245 = (30.0, 479.0): cluster 1\n", + "Example n. 2246 = (30.0, 478.0): cluster 1\n", + "Example n. 2247 = (30.0, 477.0): cluster 1\n", + "Example n. 2248 = (30.0, 476.0): cluster 1\n", + "Example n. 2249 = (30.0, 475.0): cluster 1\n", + "Example n. 2250 = (30.0, 474.0): cluster 1\n", + "Example n. 2251 = (30.0, 473.0): cluster 1\n", + "Example n. 2252 = (30.0, 472.0): cluster 1\n", + "Example n. 2253 = (30.0, 471.0): cluster 1\n", + "Example n. 2254 = (30.0, 470.0): cluster 1\n", + "Example n. 2255 = (30.0, 469.0): cluster 1\n", + "Example n. 2256 = (30.0, 468.0): cluster 1\n", + "Example n. 2257 = (30.0, 467.0): cluster 1\n", + "Example n. 2258 = (30.0, 466.0): cluster 1\n", + "Example n. 2259 = (30.0, 465.0): cluster 3\n", + "Example n. 2260 = (30.0, 464.0): cluster 3\n", + "Example n. 2261 = (30.0, 463.0): cluster 3\n", + "Example n. 2262 = (30.0, 370.0): cluster 3\n", + "Example n. 2263 = (30.0, 369.0): cluster 3\n", + "Example n. 2264 = (30.0, 368.0): cluster 3\n", + "Example n. 2265 = (30.0, 367.0): cluster 3\n", + "Example n. 2266 = (30.0, 302.0): cluster 3\n", + "Example n. 2267 = (30.0, 301.0): cluster 3\n", + "Example n. 2268 = (30.0, 300.0): cluster 3\n", + "Example n. 2269 = (30.0, 299.0): cluster 3\n", + "Example n. 2270 = (30.0, 296.0): cluster 3\n", + "Example n. 2271 = (30.0, 295.0): cluster 3\n", + "Example n. 2272 = (30.0, 294.0): cluster 3\n", + "Example n. 2273 = (30.0, 293.0): cluster 3\n", + "Example n. 2274 = (30.0, 292.0): cluster 3\n", + "Example n. 2275 = (30.0, 291.0): cluster 3\n", + "Example n. 2276 = (30.0, 290.0): cluster 3\n", + "Example n. 2277 = (30.0, 289.0): cluster 3\n", + "Example n. 2278 = (30.0, 288.0): cluster 3\n", + "Example n. 2279 = (30.0, 287.0): cluster 3\n", + "Example n. 2280 = (30.0, 286.0): cluster 3\n", + "Example n. 2281 = (30.0, 281.0): cluster 3\n", + "Example n. 2282 = (30.0, 280.0): cluster 3\n", + "Example n. 2283 = (30.0, 279.0): cluster 3\n", + "Example n. 2284 = (30.0, 278.0): cluster 3\n", + "Example n. 2285 = (30.0, 277.0): cluster 3\n", + "Example n. 2286 = (30.0, 276.0): cluster 3\n", + "Example n. 2287 = (30.0, 273.0): cluster 3\n", + "Example n. 2288 = (30.0, 272.0): cluster 3\n", + "Example n. 2289 = (30.0, 271.0): cluster 3\n", + "Example n. 2290 = (30.0, 270.0): cluster 3\n", + "Example n. 2291 = (30.0, 269.0): cluster 0\n", + "Example n. 2292 = (30.0, 268.0): cluster 0\n", + "Example n. 2293 = (30.0, 267.0): cluster 0\n", + "Example n. 2294 = (30.0, 266.0): cluster 0\n", + "Example n. 2295 = (30.0, 265.0): cluster 0\n", + "Example n. 2296 = (30.0, 264.0): cluster 0\n", + "Example n. 2297 = (30.0, 263.0): cluster 0\n", + "Example n. 2298 = (30.0, 262.0): cluster 0\n", + "Example n. 2299 = (30.0, 261.0): cluster 0\n", + "Example n. 2300 = (30.0, 260.0): cluster 1\n", + "Example n. 2301 = (30.0, 259.0): cluster 1\n", + "Example n. 2302 = (30.0, 258.0): cluster 1\n", + "Example n. 2303 = (30.0, 257.0): cluster 1\n", + "Example n. 2304 = (30.0, 256.0): cluster 1\n", + "Example n. 2305 = (30.0, 255.0): cluster 1\n", + "Example n. 2306 = (30.0, 254.0): cluster 1\n", + "Example n. 2307 = (30.0, 253.0): cluster 1\n", + "Example n. 2308 = (30.0, 252.0): cluster 1\n", + "Example n. 2309 = (30.0, 251.0): cluster 1\n", + "Example n. 2310 = (30.0, 250.0): cluster 1\n", + "Example n. 2311 = (30.0, 249.0): cluster 1\n", + "Example n. 2312 = (30.0, 248.0): cluster 1\n", + "Example n. 2313 = (30.0, 247.0): cluster 1\n", + "Example n. 2314 = (30.0, 246.0): cluster 1\n", + "Example n. 2315 = (30.0, 245.0): cluster 1\n", + "Example n. 2316 = (30.0, 244.0): cluster 1\n", + "Example n. 2317 = (30.0, 243.0): cluster 1\n", + "Example n. 2318 = (30.0, 242.0): cluster 1\n", + "Example n. 2319 = (30.0, 241.0): cluster 1\n", + "Example n. 2320 = (30.0, 240.0): cluster 1\n", + "Example n. 2321 = (30.0, 239.0): cluster 1\n", + "Example n. 2322 = (30.0, 238.0): cluster 3\n", + "Example n. 2323 = (30.0, 237.0): cluster 3\n", + "Example n. 2324 = (30.0, 236.0): cluster 3\n", + "Example n. 2325 = (30.0, 235.0): cluster 3\n", + "Example n. 2326 = (30.0, 234.0): cluster 3\n", + "Example n. 2327 = (30.0, 233.0): cluster 3\n", + "Example n. 2328 = (30.0, 232.0): cluster 3\n", + "Example n. 2329 = (30.0, 231.0): cluster 3\n", + "Example n. 2330 = (30.0, 230.0): cluster 3\n", + "Example n. 2331 = (30.0, 228.0): cluster 3\n", + "Example n. 2332 = (30.0, 227.0): cluster 3\n", + "Example n. 2333 = (30.0, 226.0): cluster 3\n", + "Example n. 2334 = (30.0, 225.0): cluster 3\n", + "Example n. 2335 = (30.0, 224.0): cluster 3\n", + "Example n. 2336 = (30.0, 223.0): cluster 3\n", + "Example n. 2337 = (30.0, 218.0): cluster 3\n", + "Example n. 2338 = (30.0, 217.0): cluster 3\n", + "Example n. 2339 = (30.0, 216.0): cluster 3\n", + "Example n. 2340 = (30.0, 215.0): cluster 3\n", + "Example n. 2341 = (30.0, 214.0): cluster 3\n", + "Example n. 2342 = (30.0, 213.0): cluster 3\n", + "Example n. 2343 = (30.0, 212.0): cluster 3\n", + "Example n. 2344 = (30.0, 211.0): cluster 3\n", + "Example n. 2345 = (30.0, 210.0): cluster 3\n", + "Example n. 2346 = (30.0, 204.0): cluster 3\n", + "Example n. 2347 = (30.0, 203.0): cluster 3\n", + "Example n. 2348 = (30.0, 202.0): cluster 3\n", + "Example n. 2349 = (30.0, 201.0): cluster 3\n", + "Example n. 2350 = (30.0, 200.0): cluster 3\n", + "Example n. 2351 = (30.0, 199.0): cluster 3\n", + "Example n. 2352 = (30.0, 198.0): cluster 3\n", + "Example n. 2353 = (30.0, 197.0): cluster 3\n", + "Example n. 2354 = (30.0, 196.0): cluster 0\n", + "Example n. 2355 = (30.0, 195.0): cluster 0\n", + "Example n. 2356 = (30.0, 194.0): cluster 0\n", + "Example n. 2357 = (30.0, 193.0): cluster 0\n", + "Example n. 2358 = (30.0, 192.0): cluster 0\n", + "Example n. 2359 = (30.0, 191.0): cluster 0\n", + "Example n. 2360 = (30.0, 190.0): cluster 0\n", + "Example n. 2361 = (30.0, 189.0): cluster 1\n", + "Example n. 2362 = (30.0, 188.0): cluster 1\n", + "Example n. 2363 = (30.0, 187.0): cluster 1\n", + "Example n. 2364 = (30.0, 186.0): cluster 1\n", + "Example n. 2365 = (30.0, 185.0): cluster 1\n", + "Example n. 2366 = (30.0, 184.0): cluster 1\n", + "Example n. 2367 = (30.0, 183.0): cluster 1\n", + "Example n. 2368 = (30.0, 182.0): cluster 1\n", + "Example n. 2369 = (30.0, 181.0): cluster 1\n", + "Example n. 2370 = (30.0, 180.0): cluster 1\n", + "Example n. 2371 = (30.0, 179.0): cluster 1\n", + "Example n. 2372 = (30.0, 178.0): cluster 1\n", + "Example n. 2373 = (30.0, 177.0): cluster 1\n", + "Example n. 2374 = (30.0, 176.0): cluster 1\n", + "Example n. 2375 = (30.0, 175.0): cluster 1\n", + "Example n. 2376 = (30.0, 166.0): cluster 1\n", + "Example n. 2377 = (30.0, 165.0): cluster 1\n", + "Example n. 2378 = (30.0, 164.0): cluster 1\n", + "Example n. 2379 = (30.0, 163.0): cluster 1\n", + "Example n. 2380 = (30.0, 162.0): cluster 1\n", + "Example n. 2381 = (30.0, 161.0): cluster 1\n", + "Example n. 2382 = (30.0, 160.0): cluster 1\n", + "Example n. 2383 = (30.0, 158.0): cluster 3\n", + "Example n. 2384 = (30.0, 157.0): cluster 3\n", + "Example n. 2385 = (30.0, 156.0): cluster 3\n", + "Example n. 2386 = (30.0, 155.0): cluster 3\n", + "Example n. 2387 = (30.0, 154.0): cluster 3\n", + "Example n. 2388 = (31.0, 502.0): cluster 3\n", + "Example n. 2389 = (31.0, 501.0): cluster 3\n", + "Example n. 2390 = (31.0, 500.0): cluster 3\n", + "Example n. 2391 = (31.0, 499.0): cluster 3\n", + "Example n. 2392 = (31.0, 498.0): cluster 3\n", + "Example n. 2393 = (31.0, 486.0): cluster 3\n", + "Example n. 2394 = (31.0, 485.0): cluster 3\n", + "Example n. 2395 = (31.0, 484.0): cluster 3\n", + "Example n. 2396 = (31.0, 483.0): cluster 3\n", + "Example n. 2397 = (31.0, 482.0): cluster 3\n", + "Example n. 2398 = (31.0, 481.0): cluster 3\n", + "Example n. 2399 = (31.0, 480.0): cluster 3\n", + "Example n. 2400 = (31.0, 479.0): cluster 3\n", + "Example n. 2401 = (31.0, 478.0): cluster 3\n", + "Example n. 2402 = (31.0, 477.0): cluster 3\n", + "Example n. 2403 = (31.0, 476.0): cluster 3\n", + "Example n. 2404 = (31.0, 475.0): cluster 3\n", + "Example n. 2405 = (31.0, 474.0): cluster 3\n", + "Example n. 2406 = (31.0, 473.0): cluster 3\n", + "Example n. 2407 = (31.0, 472.0): cluster 3\n", + "Example n. 2408 = (31.0, 471.0): cluster 3\n", + "Example n. 2409 = (31.0, 470.0): cluster 3\n", + "Example n. 2410 = (31.0, 469.0): cluster 3\n", + "Example n. 2411 = (31.0, 468.0): cluster 3\n", + "Example n. 2412 = (31.0, 467.0): cluster 3\n", + "Example n. 2413 = (31.0, 466.0): cluster 3\n", + "Example n. 2414 = (31.0, 465.0): cluster 3\n", + "Example n. 2415 = (31.0, 464.0): cluster 0\n", + "Example n. 2416 = (31.0, 463.0): cluster 0\n", + "Example n. 2417 = (31.0, 462.0): cluster 0\n", + "Example n. 2418 = (31.0, 461.0): cluster 0\n", + "Example n. 2419 = (31.0, 396.0): cluster 1\n", + "Example n. 2420 = (31.0, 395.0): cluster 1\n", + "Example n. 2421 = (31.0, 303.0): cluster 1\n", + "Example n. 2422 = (31.0, 302.0): cluster 1\n", + "Example n. 2423 = (31.0, 301.0): cluster 1\n", + "Example n. 2424 = (31.0, 300.0): cluster 1\n", + "Example n. 2425 = (31.0, 299.0): cluster 1\n", + "Example n. 2426 = (31.0, 295.0): cluster 1\n", + "Example n. 2427 = (31.0, 294.0): cluster 1\n", + "Example n. 2428 = (31.0, 293.0): cluster 1\n", + "Example n. 2429 = (31.0, 292.0): cluster 1\n", + "Example n. 2430 = (31.0, 291.0): cluster 1\n", + "Example n. 2431 = (31.0, 290.0): cluster 1\n", + "Example n. 2432 = (31.0, 289.0): cluster 1\n", + "Example n. 2433 = (31.0, 288.0): cluster 1\n", + "Example n. 2434 = (31.0, 287.0): cluster 1\n", + "Example n. 2435 = (31.0, 286.0): cluster 1\n", + "Example n. 2436 = (31.0, 281.0): cluster 1\n", + "Example n. 2437 = (31.0, 280.0): cluster 1\n", + "Example n. 2438 = (31.0, 279.0): cluster 1\n", + "Example n. 2439 = (31.0, 278.0): cluster 1\n", + "Example n. 2440 = (31.0, 277.0): cluster 1\n", + "Example n. 2441 = (31.0, 276.0): cluster 1\n", + "Example n. 2442 = (31.0, 275.0): cluster 1\n", + "Example n. 2443 = (31.0, 271.0): cluster 1\n", + "Example n. 2444 = (31.0, 270.0): cluster 1\n", + "Example n. 2445 = (31.0, 269.0): cluster 3\n", + "Example n. 2446 = (31.0, 268.0): cluster 3\n", + "Example n. 2447 = (31.0, 267.0): cluster 3\n", + "Example n. 2448 = (31.0, 266.0): cluster 3\n", + "Example n. 2449 = (31.0, 265.0): cluster 3\n", + "Example n. 2450 = (31.0, 264.0): cluster 3\n", + "Example n. 2451 = (31.0, 263.0): cluster 3\n", + "Example n. 2452 = (31.0, 262.0): cluster 3\n", + "Example n. 2453 = (31.0, 261.0): cluster 3\n", + "Example n. 2454 = (31.0, 260.0): cluster 3\n", + "Example n. 2455 = (31.0, 259.0): cluster 3\n", + "Example n. 2456 = (31.0, 258.0): cluster 3\n", + "Example n. 2457 = (31.0, 257.0): cluster 3\n", + "Example n. 2458 = (31.0, 256.0): cluster 3\n", + "Example n. 2459 = (31.0, 255.0): cluster 3\n", + "Example n. 2460 = (31.0, 254.0): cluster 3\n", + "Example n. 2461 = (31.0, 253.0): cluster 3\n", + "Example n. 2462 = (31.0, 252.0): cluster 3\n", + "Example n. 2463 = (31.0, 251.0): cluster 3\n", + "Example n. 2464 = (31.0, 250.0): cluster 3\n", + "Example n. 2465 = (31.0, 249.0): cluster 3\n", + "Example n. 2466 = (31.0, 248.0): cluster 3\n", + "Example n. 2467 = (31.0, 247.0): cluster 3\n", + "Example n. 2468 = (31.0, 246.0): cluster 3\n", + "Example n. 2469 = (31.0, 245.0): cluster 3\n", + "Example n. 2470 = (31.0, 244.0): cluster 3\n", + "Example n. 2471 = (31.0, 243.0): cluster 3\n", + "Example n. 2472 = (31.0, 242.0): cluster 3\n", + "Example n. 2473 = (31.0, 241.0): cluster 3\n", + "Example n. 2474 = (31.0, 240.0): cluster 3\n", + "Example n. 2475 = (31.0, 239.0): cluster 3\n", + "Example n. 2476 = (31.0, 238.0): cluster 3\n", + "Example n. 2477 = (31.0, 237.0): cluster 1\n", + "Example n. 2478 = (31.0, 236.0): cluster 1\n", + "Example n. 2479 = (31.0, 235.0): cluster 1\n", + "Example n. 2480 = (31.0, 234.0): cluster 1\n", + "Example n. 2481 = (31.0, 233.0): cluster 1\n", + "Example n. 2482 = (31.0, 232.0): cluster 1\n", + "Example n. 2483 = (31.0, 231.0): cluster 1\n", + "Example n. 2484 = (31.0, 230.0): cluster 1\n", + "Example n. 2485 = (31.0, 228.0): cluster 1\n", + "Example n. 2486 = (31.0, 227.0): cluster 1\n", + "Example n. 2487 = (31.0, 226.0): cluster 1\n", + "Example n. 2488 = (31.0, 225.0): cluster 1\n", + "Example n. 2489 = (31.0, 224.0): cluster 1\n", + "Example n. 2490 = (31.0, 223.0): cluster 1\n", + "Example n. 2491 = (31.0, 218.0): cluster 1\n", + "Example n. 2492 = (31.0, 217.0): cluster 1\n", + "Example n. 2493 = (31.0, 216.0): cluster 1\n", + "Example n. 2494 = (31.0, 215.0): cluster 1\n", + "Example n. 2495 = (31.0, 214.0): cluster 1\n", + "Example n. 2496 = (31.0, 213.0): cluster 1\n", + "Example n. 2497 = (31.0, 212.0): cluster 1\n", + "Example n. 2498 = (31.0, 211.0): cluster 1\n", + "Example n. 2499 = (31.0, 210.0): cluster 1\n", + "Example n. 2500 = (31.0, 209.0): cluster 1\n", + "Example n. 2501 = (31.0, 204.0): cluster 3\n", + "Example n. 2502 = (31.0, 203.0): cluster 3\n", + "Example n. 2503 = (31.0, 202.0): cluster 3\n", + "Example n. 2504 = (31.0, 201.0): cluster 3\n", + "Example n. 2505 = (31.0, 200.0): cluster 3\n", + "Example n. 2506 = (31.0, 199.0): cluster 3\n", + "Example n. 2507 = (31.0, 198.0): cluster 3\n", + "Example n. 2508 = (31.0, 197.0): cluster 3\n", + "Example n. 2509 = (31.0, 196.0): cluster 3\n", + "Example n. 2510 = (31.0, 195.0): cluster 3\n", + "Example n. 2511 = (31.0, 194.0): cluster 3\n", + "Example n. 2512 = (31.0, 193.0): cluster 3\n", + "Example n. 2513 = (31.0, 192.0): cluster 3\n", + "Example n. 2514 = (31.0, 191.0): cluster 3\n", + "Example n. 2515 = (31.0, 188.0): cluster 3\n", + "Example n. 2516 = (31.0, 186.0): cluster 3\n", + "Example n. 2517 = (31.0, 185.0): cluster 3\n", + "Example n. 2518 = (31.0, 184.0): cluster 3\n", + "Example n. 2519 = (31.0, 183.0): cluster 3\n", + "Example n. 2520 = (31.0, 182.0): cluster 3\n", + "Example n. 2521 = (31.0, 181.0): cluster 3\n", + "Example n. 2522 = (31.0, 180.0): cluster 3\n", + "Example n. 2523 = (31.0, 179.0): cluster 3\n", + "Example n. 2524 = (31.0, 178.0): cluster 3\n", + "Example n. 2525 = (31.0, 177.0): cluster 3\n", + "Example n. 2526 = (31.0, 176.0): cluster 3\n", + "Example n. 2527 = (31.0, 175.0): cluster 3\n", + "Example n. 2528 = (31.0, 174.0): cluster 3\n", + "Example n. 2529 = (31.0, 168.0): cluster 1\n", + "Example n. 2530 = (31.0, 167.0): cluster 1\n", + "Example n. 2531 = (31.0, 166.0): cluster 1\n", + "Example n. 2532 = (31.0, 165.0): cluster 1\n", + "Example n. 2533 = (31.0, 164.0): cluster 1\n", + "Example n. 2534 = (31.0, 163.0): cluster 1\n", + "Example n. 2535 = (31.0, 162.0): cluster 1\n", + "Example n. 2536 = (31.0, 161.0): cluster 1\n", + "Example n. 2537 = (31.0, 160.0): cluster 1\n", + "Example n. 2538 = (31.0, 159.0): cluster 1\n", + "Example n. 2539 = (31.0, 158.0): cluster 1\n", + "Example n. 2540 = (31.0, 157.0): cluster 1\n", + "Example n. 2541 = (31.0, 156.0): cluster 1\n", + "Example n. 2542 = (31.0, 155.0): cluster 1\n", + "Example n. 2543 = (31.0, 154.0): cluster 1\n", + "Example n. 2544 = (31.0, 153.0): cluster 1\n", + "Example n. 2545 = (31.0, 132.0): cluster 1\n", + "Example n. 2546 = (31.0, 131.0): cluster 1\n", + "Example n. 2547 = (31.0, 130.0): cluster 1\n", + "Example n. 2548 = (31.0, 129.0): cluster 1\n", + "Example n. 2549 = (32.0, 501.0): cluster 1\n", + "Example n. 2550 = (32.0, 500.0): cluster 1\n", + "Example n. 2551 = (32.0, 499.0): cluster 1\n", + "Example n. 2552 = (32.0, 486.0): cluster 1\n", + "Example n. 2553 = (32.0, 485.0): cluster 3\n", + "Example n. 2554 = (32.0, 484.0): cluster 3\n", + "Example n. 2555 = (32.0, 483.0): cluster 3\n", + "Example n. 2556 = (32.0, 482.0): cluster 3\n", + "Example n. 2557 = (32.0, 481.0): cluster 3\n", + "Example n. 2558 = (32.0, 480.0): cluster 3\n", + "Example n. 2559 = (32.0, 479.0): cluster 3\n", + "Example n. 2560 = (32.0, 478.0): cluster 3\n", + "Example n. 2561 = (32.0, 477.0): cluster 3\n", + "Example n. 2562 = (32.0, 476.0): cluster 3\n", + "Example n. 2563 = (32.0, 475.0): cluster 3\n", + "Example n. 2564 = (32.0, 474.0): cluster 3\n", + "Example n. 2565 = (32.0, 473.0): cluster 3\n", + "Example n. 2566 = (32.0, 472.0): cluster 3\n", + "Example n. 2567 = (32.0, 471.0): cluster 3\n", + "Example n. 2568 = (32.0, 470.0): cluster 3\n", + "Example n. 2569 = (32.0, 469.0): cluster 3\n", + "Example n. 2570 = (32.0, 468.0): cluster 3\n", + "Example n. 2571 = (32.0, 467.0): cluster 3\n", + "Example n. 2572 = (32.0, 466.0): cluster 3\n", + "Example n. 2573 = (32.0, 465.0): cluster 3\n", + "Example n. 2574 = (32.0, 464.0): cluster 3\n", + "Example n. 2575 = (32.0, 463.0): cluster 3\n", + "Example n. 2576 = (32.0, 462.0): cluster 3\n", + "Example n. 2577 = (32.0, 461.0): cluster 3\n", + "Example n. 2578 = (32.0, 460.0): cluster 3\n", + "Example n. 2579 = (32.0, 397.0): cluster 3\n", + "Example n. 2580 = (32.0, 396.0): cluster 3\n", + "Example n. 2581 = (32.0, 395.0): cluster 1\n", + "Example n. 2582 = (32.0, 394.0): cluster 1\n", + "Example n. 2583 = (32.0, 393.0): cluster 1\n", + "Example n. 2584 = (32.0, 303.0): cluster 1\n", + "Example n. 2585 = (32.0, 302.0): cluster 1\n", + "Example n. 2586 = (32.0, 301.0): cluster 1\n", + "Example n. 2587 = (32.0, 300.0): cluster 1\n", + "Example n. 2588 = (32.0, 299.0): cluster 1\n", + "Example n. 2589 = (32.0, 298.0): cluster 1\n", + "Example n. 2590 = (32.0, 295.0): cluster 1\n", + "Example n. 2591 = (32.0, 294.0): cluster 1\n", + "Example n. 2592 = (32.0, 293.0): cluster 1\n", + "Example n. 2593 = (32.0, 292.0): cluster 1\n", + "Example n. 2594 = (32.0, 291.0): cluster 1\n", + "Example n. 2595 = (32.0, 290.0): cluster 1\n", + "Example n. 2596 = (32.0, 289.0): cluster 1\n", + "Example n. 2597 = (32.0, 288.0): cluster 1\n", + "Example n. 2598 = (32.0, 287.0): cluster 1\n", + "Example n. 2599 = (32.0, 286.0): cluster 1\n", + "Example n. 2600 = (32.0, 285.0): cluster 1\n", + "Example n. 2601 = (32.0, 284.0): cluster 1\n", + "Example n. 2602 = (32.0, 281.0): cluster 1\n", + "Example n. 2603 = (32.0, 280.0): cluster 1\n", + "Example n. 2604 = (32.0, 279.0): cluster 1\n", + "Example n. 2605 = (32.0, 278.0): cluster 3\n", + "Example n. 2606 = (32.0, 277.0): cluster 3\n", + "Example n. 2607 = (32.0, 276.0): cluster 3\n", + "Example n. 2608 = (32.0, 275.0): cluster 3\n", + "Example n. 2609 = (32.0, 271.0): cluster 3\n", + "Example n. 2610 = (32.0, 270.0): cluster 3\n", + "Example n. 2611 = (32.0, 269.0): cluster 3\n", + "Example n. 2612 = (32.0, 268.0): cluster 3\n", + "Example n. 2613 = (32.0, 267.0): cluster 3\n", + "Example n. 2614 = (32.0, 266.0): cluster 3\n", + "Example n. 2615 = (32.0, 265.0): cluster 3\n", + "Example n. 2616 = (32.0, 264.0): cluster 3\n", + "Example n. 2617 = (32.0, 263.0): cluster 3\n", + "Example n. 2618 = (32.0, 262.0): cluster 3\n", + "Example n. 2619 = (32.0, 261.0): cluster 3\n", + "Example n. 2620 = (32.0, 260.0): cluster 3\n", + "Example n. 2621 = (32.0, 259.0): cluster 3\n", + "Example n. 2622 = (32.0, 258.0): cluster 3\n", + "Example n. 2623 = (32.0, 257.0): cluster 3\n", + "Example n. 2624 = (32.0, 256.0): cluster 3\n", + "Example n. 2625 = (32.0, 255.0): cluster 3\n", + "Example n. 2626 = (32.0, 254.0): cluster 3\n", + "Example n. 2627 = (32.0, 253.0): cluster 3\n", + "Example n. 2628 = (32.0, 252.0): cluster 3\n", + "Example n. 2629 = (32.0, 251.0): cluster 3\n", + "Example n. 2630 = (32.0, 250.0): cluster 3\n", + "Example n. 2631 = (32.0, 249.0): cluster 3\n", + "Example n. 2632 = (32.0, 248.0): cluster 3\n", + "Example n. 2633 = (32.0, 247.0): cluster 1\n", + "Example n. 2634 = (32.0, 246.0): cluster 1\n", + "Example n. 2635 = (32.0, 245.0): cluster 1\n", + "Example n. 2636 = (32.0, 244.0): cluster 1\n", + "Example n. 2637 = (32.0, 243.0): cluster 1\n", + "Example n. 2638 = (32.0, 242.0): cluster 1\n", + "Example n. 2639 = (32.0, 241.0): cluster 1\n", + "Example n. 2640 = (32.0, 240.0): cluster 1\n", + "Example n. 2641 = (32.0, 239.0): cluster 1\n", + "Example n. 2642 = (32.0, 238.0): cluster 1\n", + "Example n. 2643 = (32.0, 237.0): cluster 1\n", + "Example n. 2644 = (32.0, 236.0): cluster 1\n", + "Example n. 2645 = (32.0, 235.0): cluster 1\n", + "Example n. 2646 = (32.0, 234.0): cluster 1\n", + "Example n. 2647 = (32.0, 233.0): cluster 1\n", + "Example n. 2648 = (32.0, 232.0): cluster 1\n", + "Example n. 2649 = (32.0, 231.0): cluster 1\n", + "Example n. 2650 = (32.0, 230.0): cluster 1\n", + "Example n. 2651 = (32.0, 228.0): cluster 1\n", + "Example n. 2652 = (32.0, 227.0): cluster 1\n", + "Example n. 2653 = (32.0, 226.0): cluster 1\n", + "Example n. 2654 = (32.0, 225.0): cluster 1\n", + "Example n. 2655 = (32.0, 224.0): cluster 1\n", + "Example n. 2656 = (32.0, 223.0): cluster 1\n", + "Example n. 2657 = (32.0, 222.0): cluster 3\n", + "Example n. 2658 = (32.0, 217.0): cluster 3\n", + "Example n. 2659 = (32.0, 216.0): cluster 3\n", + "Example n. 2660 = (32.0, 215.0): cluster 3\n", + "Example n. 2661 = (32.0, 214.0): cluster 3\n", + "Example n. 2662 = (32.0, 213.0): cluster 3\n", + "Example n. 2663 = (32.0, 212.0): cluster 3\n", + "Example n. 2664 = (32.0, 211.0): cluster 3\n", + "Example n. 2665 = (32.0, 210.0): cluster 3\n", + "Example n. 2666 = (32.0, 209.0): cluster 3\n", + "Example n. 2667 = (32.0, 202.0): cluster 3\n", + "Example n. 2668 = (32.0, 201.0): cluster 3\n", + "Example n. 2669 = (32.0, 200.0): cluster 3\n", + "Example n. 2670 = (32.0, 199.0): cluster 3\n", + "Example n. 2671 = (32.0, 198.0): cluster 3\n", + "Example n. 2672 = (32.0, 197.0): cluster 3\n", + "Example n. 2673 = (32.0, 196.0): cluster 3\n", + "Example n. 2674 = (32.0, 195.0): cluster 3\n", + "Example n. 2675 = (32.0, 194.0): cluster 3\n", + "Example n. 2676 = (32.0, 193.0): cluster 3\n", + "Example n. 2677 = (32.0, 192.0): cluster 3\n", + "Example n. 2678 = (32.0, 191.0): cluster 3\n", + "Example n. 2679 = (32.0, 184.0): cluster 3\n", + "Example n. 2680 = (32.0, 183.0): cluster 3\n", + "Example n. 2681 = (32.0, 182.0): cluster 3\n", + "Example n. 2682 = (32.0, 181.0): cluster 3\n", + "Example n. 2683 = (32.0, 180.0): cluster 3\n", + "Example n. 2684 = (32.0, 179.0): cluster 3\n", + "Example n. 2685 = (32.0, 178.0): cluster 1\n", + "Example n. 2686 = (32.0, 177.0): cluster 1\n", + "Example n. 2687 = (32.0, 176.0): cluster 1\n", + "Example n. 2688 = (32.0, 175.0): cluster 1\n", + "Example n. 2689 = (32.0, 174.0): cluster 1\n", + "Example n. 2690 = (32.0, 170.0): cluster 1\n", + "Example n. 2691 = (32.0, 169.0): cluster 1\n", + "Example n. 2692 = (32.0, 168.0): cluster 1\n", + "Example n. 2693 = (32.0, 167.0): cluster 1\n", + "Example n. 2694 = (32.0, 166.0): cluster 1\n", + "Example n. 2695 = (32.0, 165.0): cluster 1\n", + "Example n. 2696 = (32.0, 164.0): cluster 1\n", + "Example n. 2697 = (32.0, 163.0): cluster 1\n", + "Example n. 2698 = (32.0, 162.0): cluster 1\n", + "Example n. 2699 = (32.0, 161.0): cluster 1\n", + "Example n. 2700 = (32.0, 160.0): cluster 1\n", + "Example n. 2701 = (32.0, 159.0): cluster 1\n", + "Example n. 2702 = (32.0, 158.0): cluster 1\n", + "Example n. 2703 = (32.0, 157.0): cluster 1\n", + "Example n. 2704 = (32.0, 156.0): cluster 1\n", + "Example n. 2705 = (32.0, 155.0): cluster 1\n", + "Example n. 2706 = (32.0, 154.0): cluster 1\n", + "Example n. 2707 = (32.0, 153.0): cluster 1\n", + "Example n. 2708 = (32.0, 136.0): cluster 1\n", + "Example n. 2709 = (32.0, 135.0): cluster 1\n", + "Example n. 2710 = (32.0, 134.0): cluster 1\n", + "Example n. 2711 = (32.0, 133.0): cluster 1\n", + "Example n. 2712 = (32.0, 132.0): cluster 1\n", + "Example n. 2713 = (32.0, 131.0): cluster 1\n", + "Example n. 2714 = (32.0, 130.0): cluster 1\n", + "Example n. 2715 = (32.0, 129.0): cluster 1\n", + "Example n. 2716 = (32.0, 128.0): cluster 1\n", + "Example n. 2717 = (33.0, 488.0): cluster 1\n", + "Example n. 2718 = (33.0, 487.0): cluster 1\n", + "Example n. 2719 = (33.0, 486.0): cluster 3\n", + "Example n. 2720 = (33.0, 485.0): cluster 3\n", + "Example n. 2721 = (33.0, 484.0): cluster 3\n", + "Example n. 2722 = (33.0, 483.0): cluster 3\n", + "Example n. 2723 = (33.0, 482.0): cluster 3\n", + "Example n. 2724 = (33.0, 481.0): cluster 3\n", + "Example n. 2725 = (33.0, 480.0): cluster 3\n", + "Example n. 2726 = (33.0, 479.0): cluster 3\n", + "Example n. 2727 = (33.0, 478.0): cluster 3\n", + "Example n. 2728 = (33.0, 477.0): cluster 3\n", + "Example n. 2729 = (33.0, 476.0): cluster 3\n", + "Example n. 2730 = (33.0, 475.0): cluster 3\n", + "Example n. 2731 = (33.0, 474.0): cluster 3\n", + "Example n. 2732 = (33.0, 473.0): cluster 3\n", + "Example n. 2733 = (33.0, 472.0): cluster 3\n", + "Example n. 2734 = (33.0, 471.0): cluster 3\n", + "Example n. 2735 = (33.0, 470.0): cluster 3\n", + "Example n. 2736 = (33.0, 469.0): cluster 3\n", + "Example n. 2737 = (33.0, 468.0): cluster 3\n", + "Example n. 2738 = (33.0, 467.0): cluster 3\n", + "Example n. 2739 = (33.0, 466.0): cluster 3\n", + "Example n. 2740 = (33.0, 465.0): cluster 3\n", + "Example n. 2741 = (33.0, 464.0): cluster 3\n", + "Example n. 2742 = (33.0, 463.0): cluster 3\n", + "Example n. 2743 = (33.0, 462.0): cluster 3\n", + "Example n. 2744 = (33.0, 461.0): cluster 3\n", + "Example n. 2745 = (33.0, 460.0): cluster 3\n", + "Example n. 2746 = (33.0, 459.0): cluster 3\n", + "Example n. 2747 = (33.0, 398.0): cluster 3\n", + "Example n. 2748 = (33.0, 397.0): cluster 3\n", + "Example n. 2749 = (33.0, 396.0): cluster 3\n", + "Example n. 2750 = (33.0, 395.0): cluster 3\n", + "Example n. 2751 = (33.0, 394.0): cluster 3\n", + "Example n. 2752 = (33.0, 393.0): cluster 3\n", + "Example n. 2753 = (33.0, 380.0): cluster 3\n", + "Example n. 2754 = (33.0, 379.0): cluster 3\n", + "Example n. 2755 = (33.0, 378.0): cluster 3\n", + "Example n. 2756 = (33.0, 377.0): cluster 3\n", + "Example n. 2757 = (33.0, 303.0): cluster 1\n", + "Example n. 2758 = (33.0, 302.0): cluster 1\n", + "Example n. 2759 = (33.0, 301.0): cluster 1\n", + "Example n. 2760 = (33.0, 300.0): cluster 1\n", + "Example n. 2761 = (33.0, 299.0): cluster 1\n", + "Example n. 2762 = (33.0, 298.0): cluster 1\n", + "Example n. 2763 = (33.0, 297.0): cluster 1\n", + "Example n. 2764 = (33.0, 295.0): cluster 1\n", + "Example n. 2765 = (33.0, 294.0): cluster 1\n", + "Example n. 2766 = (33.0, 293.0): cluster 1\n", + "Example n. 2767 = (33.0, 292.0): cluster 1\n", + "Example n. 2768 = (33.0, 291.0): cluster 1\n", + "Example n. 2769 = (33.0, 290.0): cluster 1\n", + "Example n. 2770 = (33.0, 289.0): cluster 1\n", + "Example n. 2771 = (33.0, 288.0): cluster 1\n", + "Example n. 2772 = (33.0, 287.0): cluster 1\n", + "Example n. 2773 = (33.0, 286.0): cluster 1\n", + "Example n. 2774 = (33.0, 285.0): cluster 1\n", + "Example n. 2775 = (33.0, 284.0): cluster 1\n", + "Example n. 2776 = (33.0, 283.0): cluster 1\n", + "Example n. 2777 = (33.0, 281.0): cluster 1\n", + "Example n. 2778 = (33.0, 280.0): cluster 1\n", + "Example n. 2779 = (33.0, 279.0): cluster 1\n", + "Example n. 2780 = (33.0, 278.0): cluster 1\n", + "Example n. 2781 = (33.0, 277.0): cluster 1\n", + "Example n. 2782 = (33.0, 276.0): cluster 1\n", + "Example n. 2783 = (33.0, 275.0): cluster 1\n", + "Example n. 2784 = (33.0, 270.0): cluster 1\n", + "Example n. 2785 = (33.0, 269.0): cluster 1\n", + "Example n. 2786 = (33.0, 268.0): cluster 1\n", + "Example n. 2787 = (33.0, 267.0): cluster 1\n", + "Example n. 2788 = (33.0, 266.0): cluster 1\n", + "Example n. 2789 = (33.0, 265.0): cluster 1\n", + "Example n. 2790 = (33.0, 264.0): cluster 1\n", + "Example n. 2791 = (33.0, 263.0): cluster 3\n", + "Example n. 2792 = (33.0, 262.0): cluster 3\n", + "Example n. 2793 = (33.0, 261.0): cluster 3\n", + "Example n. 2794 = (33.0, 260.0): cluster 3\n", + "Example n. 2795 = (33.0, 259.0): cluster 3\n", + "Example n. 2796 = (33.0, 258.0): cluster 3\n", + "Example n. 2797 = (33.0, 257.0): cluster 3\n", + "Example n. 2798 = (33.0, 256.0): cluster 3\n", + "Example n. 2799 = (33.0, 255.0): cluster 3\n", + "Example n. 2800 = (33.0, 254.0): cluster 3\n", + "Example n. 2801 = (33.0, 253.0): cluster 3\n", + "Example n. 2802 = (33.0, 252.0): cluster 3\n", + "Example n. 2803 = (33.0, 251.0): cluster 3\n", + "Example n. 2804 = (33.0, 250.0): cluster 3\n", + "Example n. 2805 = (33.0, 249.0): cluster 3\n", + "Example n. 2806 = (33.0, 248.0): cluster 3\n", + "Example n. 2807 = (33.0, 247.0): cluster 3\n", + "Example n. 2808 = (33.0, 246.0): cluster 3\n", + "Example n. 2809 = (33.0, 245.0): cluster 3\n", + "Example n. 2810 = (33.0, 244.0): cluster 3\n", + "Example n. 2811 = (33.0, 243.0): cluster 3\n", + "Example n. 2812 = (33.0, 242.0): cluster 3\n", + "Example n. 2813 = (33.0, 241.0): cluster 3\n", + "Example n. 2814 = (33.0, 240.0): cluster 3\n", + "Example n. 2815 = (33.0, 239.0): cluster 3\n", + "Example n. 2816 = (33.0, 238.0): cluster 3\n", + "Example n. 2817 = (33.0, 237.0): cluster 3\n", + "Example n. 2818 = (33.0, 236.0): cluster 3\n", + "Example n. 2819 = (33.0, 235.0): cluster 3\n", + "Example n. 2820 = (33.0, 234.0): cluster 3\n", + "Example n. 2821 = (33.0, 233.0): cluster 3\n", + "Example n. 2822 = (33.0, 232.0): cluster 3\n", + "Example n. 2823 = (33.0, 231.0): cluster 3\n", + "Example n. 2824 = (33.0, 230.0): cluster 3\n", + "Example n. 2825 = (33.0, 227.0): cluster 3\n", + "Example n. 2826 = (33.0, 226.0): cluster 3\n", + "Example n. 2827 = (33.0, 225.0): cluster 3\n", + "Example n. 2828 = (33.0, 224.0): cluster 3\n", + "Example n. 2829 = (33.0, 223.0): cluster 1\n", + "Example n. 2830 = (33.0, 222.0): cluster 1\n", + "Example n. 2831 = (33.0, 214.0): cluster 1\n", + "Example n. 2832 = (33.0, 213.0): cluster 1\n", + "Example n. 2833 = (33.0, 212.0): cluster 1\n", + "Example n. 2834 = (33.0, 211.0): cluster 1\n", + "Example n. 2835 = (33.0, 210.0): cluster 1\n", + "Example n. 2836 = (33.0, 209.0): cluster 1\n", + "Example n. 2837 = (33.0, 201.0): cluster 1\n", + "Example n. 2838 = (33.0, 200.0): cluster 1\n", + "Example n. 2839 = (33.0, 199.0): cluster 1\n", + "Example n. 2840 = (33.0, 198.0): cluster 1\n", + "Example n. 2841 = (33.0, 197.0): cluster 1\n", + "Example n. 2842 = (33.0, 196.0): cluster 1\n", + "Example n. 2843 = (33.0, 195.0): cluster 1\n", + "Example n. 2844 = (33.0, 194.0): cluster 1\n", + "Example n. 2845 = (33.0, 193.0): cluster 1\n", + "Example n. 2846 = (33.0, 192.0): cluster 1\n", + "Example n. 2847 = (33.0, 183.0): cluster 1\n", + "Example n. 2848 = (33.0, 182.0): cluster 1\n", + "Example n. 2849 = (33.0, 181.0): cluster 1\n", + "Example n. 2850 = (33.0, 180.0): cluster 1\n", + "Example n. 2851 = (33.0, 179.0): cluster 1\n", + "Example n. 2852 = (33.0, 178.0): cluster 1\n", + "Example n. 2853 = (33.0, 177.0): cluster 1\n", + "Example n. 2854 = (33.0, 176.0): cluster 1\n", + "Example n. 2855 = (33.0, 175.0): cluster 1\n", + "Example n. 2856 = (33.0, 171.0): cluster 1\n", + "Example n. 2857 = (33.0, 170.0): cluster 1\n", + "Example n. 2858 = (33.0, 169.0): cluster 1\n", + "Example n. 2859 = (33.0, 168.0): cluster 1\n", + "Example n. 2860 = (33.0, 167.0): cluster 1\n", + "Example n. 2861 = (33.0, 166.0): cluster 1\n", + "Example n. 2862 = (33.0, 165.0): cluster 1\n", + "Example n. 2863 = (33.0, 164.0): cluster 3\n", + "Example n. 2864 = (33.0, 163.0): cluster 3\n", + "Example n. 2865 = (33.0, 162.0): cluster 3\n", + "Example n. 2866 = (33.0, 161.0): cluster 3\n", + "Example n. 2867 = (33.0, 160.0): cluster 3\n", + "Example n. 2868 = (33.0, 159.0): cluster 3\n", + "Example n. 2869 = (33.0, 158.0): cluster 3\n", + "Example n. 2870 = (33.0, 157.0): cluster 3\n", + "Example n. 2871 = (33.0, 156.0): cluster 3\n", + "Example n. 2872 = (33.0, 155.0): cluster 3\n", + "Example n. 2873 = (33.0, 154.0): cluster 3\n", + "Example n. 2874 = (33.0, 153.0): cluster 3\n", + "Example n. 2875 = (33.0, 136.0): cluster 3\n", + "Example n. 2876 = (33.0, 135.0): cluster 3\n", + "Example n. 2877 = (33.0, 134.0): cluster 3\n", + "Example n. 2878 = (33.0, 133.0): cluster 3\n", + "Example n. 2879 = (33.0, 132.0): cluster 3\n", + "Example n. 2880 = (33.0, 131.0): cluster 3\n", + "Example n. 2881 = (33.0, 130.0): cluster 3\n", + "Example n. 2882 = (33.0, 129.0): cluster 3\n", + "Example n. 2883 = (33.0, 128.0): cluster 3\n", + "Example n. 2884 = (34.0, 488.0): cluster 3\n", + "Example n. 2885 = (34.0, 487.0): cluster 3\n", + "Example n. 2886 = (34.0, 486.0): cluster 3\n", + "Example n. 2887 = (34.0, 485.0): cluster 3\n", + "Example n. 2888 = (34.0, 484.0): cluster 3\n", + "Example n. 2889 = (34.0, 483.0): cluster 3\n", + "Example n. 2890 = (34.0, 482.0): cluster 3\n", + "Example n. 2891 = (34.0, 481.0): cluster 3\n", + "Example n. 2892 = (34.0, 480.0): cluster 3\n", + "Example n. 2893 = (34.0, 479.0): cluster 3\n", + "Example n. 2894 = (34.0, 478.0): cluster 3\n", + "Example n. 2895 = (34.0, 477.0): cluster 3\n", + "Example n. 2896 = (34.0, 476.0): cluster 3\n", + "Example n. 2897 = (34.0, 475.0): cluster 3\n", + "Example n. 2898 = (34.0, 474.0): cluster 3\n", + "Example n. 2899 = (34.0, 473.0): cluster 3\n", + "Example n. 2900 = (34.0, 472.0): cluster 3\n", + "Example n. 2901 = (34.0, 471.0): cluster 1\n", + "Example n. 2902 = (34.0, 470.0): cluster 1\n", + "Example n. 2903 = (34.0, 469.0): cluster 1\n", + "Example n. 2904 = (34.0, 468.0): cluster 1\n", + "Example n. 2905 = (34.0, 467.0): cluster 1\n", + "Example n. 2906 = (34.0, 466.0): cluster 1\n", + "Example n. 2907 = (34.0, 465.0): cluster 1\n", + "Example n. 2908 = (34.0, 464.0): cluster 1\n", + "Example n. 2909 = (34.0, 463.0): cluster 1\n", + "Example n. 2910 = (34.0, 462.0): cluster 1\n", + "Example n. 2911 = (34.0, 461.0): cluster 1\n", + "Example n. 2912 = (34.0, 460.0): cluster 1\n", + "Example n. 2913 = (34.0, 459.0): cluster 1\n", + "Example n. 2914 = (34.0, 398.0): cluster 1\n", + "Example n. 2915 = (34.0, 397.0): cluster 1\n", + "Example n. 2916 = (34.0, 396.0): cluster 1\n", + "Example n. 2917 = (34.0, 395.0): cluster 1\n", + "Example n. 2918 = (34.0, 394.0): cluster 1\n", + "Example n. 2919 = (34.0, 393.0): cluster 1\n", + "Example n. 2920 = (34.0, 382.0): cluster 1\n", + "Example n. 2921 = (34.0, 381.0): cluster 1\n", + "Example n. 2922 = (34.0, 380.0): cluster 1\n", + "Example n. 2923 = (34.0, 379.0): cluster 1\n", + "Example n. 2924 = (34.0, 378.0): cluster 1\n", + "Example n. 2925 = (34.0, 377.0): cluster 1\n", + "Example n. 2926 = (34.0, 376.0): cluster 1\n", + "Example n. 2927 = (34.0, 306.0): cluster 1\n", + "Example n. 2928 = (34.0, 305.0): cluster 1\n", + "Example n. 2929 = (34.0, 304.0): cluster 1\n", + "Example n. 2930 = (34.0, 303.0): cluster 1\n", + "Example n. 2931 = (34.0, 302.0): cluster 1\n", + "Example n. 2932 = (34.0, 301.0): cluster 1\n", + "Example n. 2933 = (34.0, 300.0): cluster 1\n", + "Example n. 2934 = (34.0, 299.0): cluster 1\n", + "Example n. 2935 = (34.0, 298.0): cluster 1\n", + "Example n. 2936 = (34.0, 297.0): cluster 1\n", + "Example n. 2937 = (34.0, 294.0): cluster 1\n", + "Example n. 2938 = (34.0, 293.0): cluster 1\n", + "Example n. 2939 = (34.0, 292.0): cluster 1\n", + "Example n. 2940 = (34.0, 291.0): cluster 1\n", + "Example n. 2941 = (34.0, 290.0): cluster 1\n", + "Example n. 2942 = (34.0, 289.0): cluster 1\n", + "Example n. 2943 = (34.0, 288.0): cluster 1\n", + "Example n. 2944 = (34.0, 287.0): cluster 1\n", + "Example n. 2945 = (34.0, 286.0): cluster 1\n", + "Example n. 2946 = (34.0, 285.0): cluster 1\n", + "Example n. 2947 = (34.0, 284.0): cluster 1\n", + "Example n. 2948 = (34.0, 283.0): cluster 1\n", + "Example n. 2949 = (34.0, 282.0): cluster 1\n", + "Example n. 2950 = (34.0, 281.0): cluster 1\n", + "Example n. 2951 = (34.0, 280.0): cluster 1\n", + "Example n. 2952 = (34.0, 279.0): cluster 1\n", + "Example n. 2953 = (34.0, 278.0): cluster 1\n", + "Example n. 2954 = (34.0, 277.0): cluster 3\n", + "Example n. 2955 = (34.0, 276.0): cluster 3\n", + "Example n. 2956 = (34.0, 269.0): cluster 3\n", + "Example n. 2957 = (34.0, 268.0): cluster 3\n", + "Example n. 2958 = (34.0, 267.0): cluster 3\n", + "Example n. 2959 = (34.0, 266.0): cluster 3\n", + "Example n. 2960 = (34.0, 265.0): cluster 3\n", + "Example n. 2961 = (34.0, 264.0): cluster 3\n", + "Example n. 2962 = (34.0, 263.0): cluster 3\n", + "Example n. 2963 = (34.0, 262.0): cluster 3\n", + "Example n. 2964 = (34.0, 261.0): cluster 3\n", + "Example n. 2965 = (34.0, 260.0): cluster 3\n", + "Example n. 2966 = (34.0, 252.0): cluster 3\n", + "Example n. 2967 = (34.0, 251.0): cluster 3\n", + "Example n. 2968 = (34.0, 248.0): cluster 3\n", + "Example n. 2969 = (34.0, 247.0): cluster 3\n", + "Example n. 2970 = (34.0, 246.0): cluster 3\n", + "Example n. 2971 = (34.0, 245.0): cluster 3\n", + "Example n. 2972 = (34.0, 244.0): cluster 3\n", + "Example n. 2973 = (34.0, 243.0): cluster 3\n", + "Example n. 2974 = (34.0, 242.0): cluster 3\n", + "Example n. 2975 = (34.0, 241.0): cluster 3\n", + "Example n. 2976 = (34.0, 240.0): cluster 3\n", + "Example n. 2977 = (34.0, 239.0): cluster 3\n", + "Example n. 2978 = (34.0, 238.0): cluster 3\n", + "Example n. 2979 = (34.0, 237.0): cluster 3\n", + "Example n. 2980 = (34.0, 236.0): cluster 3\n", + "Example n. 2981 = (34.0, 235.0): cluster 3\n", + "Example n. 2982 = (34.0, 234.0): cluster 3\n", + "Example n. 2983 = (34.0, 233.0): cluster 3\n", + "Example n. 2984 = (34.0, 232.0): cluster 3\n", + "Example n. 2985 = (34.0, 231.0): cluster 3\n", + "Example n. 2986 = (34.0, 227.0): cluster 3\n", + "Example n. 2987 = (34.0, 226.0): cluster 3\n", + "Example n. 2988 = (34.0, 225.0): cluster 3\n", + "Example n. 2989 = (34.0, 224.0): cluster 3\n", + "Example n. 2990 = (34.0, 223.0): cluster 3\n", + "Example n. 2991 = (34.0, 222.0): cluster 3\n", + "Example n. 2992 = (34.0, 221.0): cluster 3\n", + "Example n. 2993 = (34.0, 220.0): cluster 3\n", + "Example n. 2994 = (34.0, 219.0): cluster 3\n", + "Example n. 2995 = (34.0, 218.0): cluster 1\n", + "Example n. 2996 = (34.0, 214.0): cluster 1\n", + "Example n. 2997 = (34.0, 213.0): cluster 1\n", + "Example n. 2998 = (34.0, 212.0): cluster 1\n", + "Example n. 2999 = (34.0, 211.0): cluster 1\n", + "Example n. 3000 = (34.0, 210.0): cluster 1\n", + "Example n. 3001 = (34.0, 187.0): cluster 1\n", + "Example n. 3002 = (34.0, 186.0): cluster 1\n", + "Example n. 3003 = (34.0, 185.0): cluster 1\n", + "Example n. 3004 = (34.0, 184.0): cluster 1\n", + "Example n. 3005 = (34.0, 183.0): cluster 1\n", + "Example n. 3006 = (34.0, 182.0): cluster 1\n", + "Example n. 3007 = (34.0, 181.0): cluster 1\n", + "Example n. 3008 = (34.0, 180.0): cluster 1\n", + "Example n. 3009 = (34.0, 179.0): cluster 1\n", + "Example n. 3010 = (34.0, 178.0): cluster 1\n", + "Example n. 3011 = (34.0, 177.0): cluster 1\n", + "Example n. 3012 = (34.0, 176.0): cluster 1\n", + "Example n. 3013 = (34.0, 171.0): cluster 1\n", + "Example n. 3014 = (34.0, 170.0): cluster 1\n", + "Example n. 3015 = (34.0, 169.0): cluster 1\n", + "Example n. 3016 = (34.0, 168.0): cluster 1\n", + "Example n. 3017 = (34.0, 167.0): cluster 1\n", + "Example n. 3018 = (34.0, 166.0): cluster 1\n", + "Example n. 3019 = (34.0, 165.0): cluster 1\n", + "Example n. 3020 = (34.0, 164.0): cluster 1\n", + "Example n. 3021 = (34.0, 163.0): cluster 1\n", + "Example n. 3022 = (34.0, 162.0): cluster 1\n", + "Example n. 3023 = (34.0, 161.0): cluster 1\n", + "Example n. 3024 = (34.0, 160.0): cluster 1\n", + "Example n. 3025 = (34.0, 159.0): cluster 1\n", + "Example n. 3026 = (34.0, 158.0): cluster 1\n", + "Example n. 3027 = (34.0, 157.0): cluster 1\n", + "Example n. 3028 = (34.0, 156.0): cluster 1\n", + "Example n. 3029 = (34.0, 155.0): cluster 1\n", + "Example n. 3030 = (34.0, 154.0): cluster 1\n", + "Example n. 3031 = (34.0, 153.0): cluster 1\n", + "Example n. 3032 = (34.0, 152.0): cluster 1\n", + "Example n. 3033 = (34.0, 151.0): cluster 1\n", + "Example n. 3034 = (34.0, 136.0): cluster 1\n", + "Example n. 3035 = (34.0, 135.0): cluster 1\n", + "Example n. 3036 = (34.0, 134.0): cluster 1\n", + "Example n. 3037 = (34.0, 133.0): cluster 1\n", + "Example n. 3038 = (34.0, 132.0): cluster 1\n", + "Example n. 3039 = (34.0, 131.0): cluster 1\n", + "Example n. 3040 = (34.0, 130.0): cluster 1\n", + "Example n. 3041 = (34.0, 129.0): cluster 1\n", + "Example n. 3042 = (34.0, 128.0): cluster 1\n", + "Example n. 3043 = (34.0, 113.0): cluster 1\n", + "Example n. 3044 = (34.0, 112.0): cluster 1\n", + "Example n. 3045 = (34.0, 111.0): cluster 1\n", + "Example n. 3046 = (34.0, 108.0): cluster 1\n", + "Example n. 3047 = (34.0, 107.0): cluster 1\n", + "Example n. 3048 = (34.0, 106.0): cluster 3\n", + "Example n. 3049 = (34.0, 105.0): cluster 3\n", + "Example n. 3050 = (35.0, 489.0): cluster 3\n", + "Example n. 3051 = (35.0, 488.0): cluster 3\n", + "Example n. 3052 = (35.0, 487.0): cluster 3\n", + "Example n. 3053 = (35.0, 486.0): cluster 3\n", + "Example n. 3054 = (35.0, 485.0): cluster 3\n", + "Example n. 3055 = (35.0, 484.0): cluster 3\n", + "Example n. 3056 = (35.0, 483.0): cluster 3\n", + "Example n. 3057 = (35.0, 482.0): cluster 3\n", + "Example n. 3058 = (35.0, 481.0): cluster 3\n", + "Example n. 3059 = (35.0, 480.0): cluster 3\n", + "Example n. 3060 = (35.0, 479.0): cluster 3\n", + "Example n. 3061 = (35.0, 478.0): cluster 3\n", + "Example n. 3062 = (35.0, 477.0): cluster 3\n", + "Example n. 3063 = (35.0, 476.0): cluster 3\n", + "Example n. 3064 = (35.0, 475.0): cluster 3\n", + "Example n. 3065 = (35.0, 474.0): cluster 3\n", + "Example n. 3066 = (35.0, 473.0): cluster 3\n", + "Example n. 3067 = (35.0, 472.0): cluster 3\n", + "Example n. 3068 = (35.0, 471.0): cluster 3\n", + "Example n. 3069 = (35.0, 470.0): cluster 3\n", + "Example n. 3070 = (35.0, 469.0): cluster 3\n", + "Example n. 3071 = (35.0, 468.0): cluster 3\n", + "Example n. 3072 = (35.0, 467.0): cluster 3\n", + "Example n. 3073 = (35.0, 466.0): cluster 3\n", + "Example n. 3074 = (35.0, 465.0): cluster 3\n", + "Example n. 3075 = (35.0, 464.0): cluster 3\n", + "Example n. 3076 = (35.0, 463.0): cluster 3\n", + "Example n. 3077 = (35.0, 462.0): cluster 3\n", + "Example n. 3078 = (35.0, 461.0): cluster 3\n", + "Example n. 3079 = (35.0, 460.0): cluster 3\n", + "Example n. 3080 = (35.0, 459.0): cluster 3\n", + "Example n. 3081 = (35.0, 458.0): cluster 3\n", + "Example n. 3082 = (35.0, 397.0): cluster 3\n", + "Example n. 3083 = (35.0, 396.0): cluster 3\n", + "Example n. 3084 = (35.0, 395.0): cluster 3\n", + "Example n. 3085 = (35.0, 394.0): cluster 1\n", + "Example n. 3086 = (35.0, 393.0): cluster 1\n", + "Example n. 3087 = (35.0, 382.0): cluster 1\n", + "Example n. 3088 = (35.0, 381.0): cluster 1\n", + "Example n. 3089 = (35.0, 380.0): cluster 1\n", + "Example n. 3090 = (35.0, 379.0): cluster 1\n", + "Example n. 3091 = (35.0, 378.0): cluster 1\n", + "Example n. 3092 = (35.0, 377.0): cluster 1\n", + "Example n. 3093 = (35.0, 376.0): cluster 1\n", + "Example n. 3094 = (35.0, 375.0): cluster 1\n", + "Example n. 3095 = (35.0, 307.0): cluster 1\n", + "Example n. 3096 = (35.0, 306.0): cluster 1\n", + "Example n. 3097 = (35.0, 305.0): cluster 1\n", + "Example n. 3098 = (35.0, 304.0): cluster 1\n", + "Example n. 3099 = (35.0, 303.0): cluster 1\n", + "Example n. 3100 = (35.0, 302.0): cluster 1\n", + "Example n. 3101 = (35.0, 301.0): cluster 1\n", + "Example n. 3102 = (35.0, 300.0): cluster 1\n", + "Example n. 3103 = (35.0, 299.0): cluster 1\n", + "Example n. 3104 = (35.0, 298.0): cluster 1\n", + "Example n. 3105 = (35.0, 297.0): cluster 1\n", + "Example n. 3106 = (35.0, 292.0): cluster 1\n", + "Example n. 3107 = (35.0, 291.0): cluster 1\n", + "Example n. 3108 = (35.0, 290.0): cluster 1\n", + "Example n. 3109 = (35.0, 289.0): cluster 1\n", + "Example n. 3110 = (35.0, 288.0): cluster 1\n", + "Example n. 3111 = (35.0, 287.0): cluster 1\n", + "Example n. 3112 = (35.0, 286.0): cluster 1\n", + "Example n. 3113 = (35.0, 285.0): cluster 1\n", + "Example n. 3114 = (35.0, 284.0): cluster 1\n", + "Example n. 3115 = (35.0, 283.0): cluster 1\n", + "Example n. 3116 = (35.0, 282.0): cluster 1\n", + "Example n. 3117 = (35.0, 281.0): cluster 1\n", + "Example n. 3118 = (35.0, 280.0): cluster 1\n", + "Example n. 3119 = (35.0, 279.0): cluster 1\n", + "Example n. 3120 = (35.0, 278.0): cluster 1\n", + "Example n. 3121 = (35.0, 277.0): cluster 1\n", + "Example n. 3122 = (35.0, 269.0): cluster 1\n", + "Example n. 3123 = (35.0, 268.0): cluster 1\n", + "Example n. 3124 = (35.0, 267.0): cluster 1\n", + "Example n. 3125 = (35.0, 266.0): cluster 1\n", + "Example n. 3126 = (35.0, 265.0): cluster 1\n", + "Example n. 3127 = (35.0, 264.0): cluster 1\n", + "Example n. 3128 = (35.0, 263.0): cluster 1\n", + "Example n. 3129 = (35.0, 262.0): cluster 1\n", + "Example n. 3130 = (35.0, 261.0): cluster 1\n", + "Example n. 3131 = (35.0, 260.0): cluster 1\n", + "Example n. 3132 = (35.0, 259.0): cluster 1\n", + "Example n. 3133 = (35.0, 258.0): cluster 1\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Example n. 3134 = (35.0, 247.0): cluster 1\n", + "Example n. 3135 = (35.0, 246.0): cluster 1\n", + "Example n. 3136 = (35.0, 245.0): cluster 1\n", + "Example n. 3137 = (35.0, 244.0): cluster 1\n", + "Example n. 3138 = (35.0, 243.0): cluster 3\n", + "Example n. 3139 = (35.0, 242.0): cluster 3\n", + "Example n. 3140 = (35.0, 241.0): cluster 3\n", + "Example n. 3141 = (35.0, 240.0): cluster 3\n", + "Example n. 3142 = (35.0, 239.0): cluster 3\n", + "Example n. 3143 = (35.0, 238.0): cluster 3\n", + "Example n. 3144 = (35.0, 237.0): cluster 3\n", + "Example n. 3145 = (35.0, 236.0): cluster 3\n", + "Example n. 3146 = (35.0, 235.0): cluster 3\n", + "Example n. 3147 = (35.0, 234.0): cluster 3\n", + "Example n. 3148 = (35.0, 233.0): cluster 3\n", + "Example n. 3149 = (35.0, 232.0): cluster 3\n", + "Example n. 3150 = (35.0, 226.0): cluster 3\n", + "Example n. 3151 = (35.0, 225.0): cluster 3\n", + "Example n. 3152 = (35.0, 224.0): cluster 3\n", + "Example n. 3153 = (35.0, 223.0): cluster 3\n", + "Example n. 3154 = (35.0, 222.0): cluster 3\n", + "Example n. 3155 = (35.0, 221.0): cluster 3\n", + "Example n. 3156 = (35.0, 220.0): cluster 3\n", + "Example n. 3157 = (35.0, 219.0): cluster 3\n", + "Example n. 3158 = (35.0, 218.0): cluster 3\n", + "Example n. 3159 = (35.0, 214.0): cluster 3\n", + "Example n. 3160 = (35.0, 213.0): cluster 3\n", + "Example n. 3161 = (35.0, 212.0): cluster 3\n", + "Example n. 3162 = (35.0, 211.0): cluster 3\n", + "Example n. 3163 = (35.0, 210.0): cluster 3\n", + "Example n. 3164 = (35.0, 193.0): cluster 3\n", + "Example n. 3165 = (35.0, 192.0): cluster 3\n", + "Example n. 3166 = (35.0, 191.0): cluster 3\n", + "Example n. 3167 = (35.0, 190.0): cluster 3\n", + "Example n. 3168 = (35.0, 187.0): cluster 3\n", + "Example n. 3169 = (35.0, 186.0): cluster 3\n", + "Example n. 3170 = (35.0, 185.0): cluster 3\n", + "Example n. 3171 = (35.0, 184.0): cluster 3\n", + "Example n. 3172 = (35.0, 183.0): cluster 3\n", + "Example n. 3173 = (35.0, 182.0): cluster 3\n", + "Example n. 3174 = (35.0, 181.0): cluster 3\n", + "Example n. 3175 = (35.0, 180.0): cluster 1\n", + "Example n. 3176 = (35.0, 179.0): cluster 1\n", + "Example n. 3177 = (35.0, 171.0): cluster 1\n", + "Example n. 3178 = (35.0, 170.0): cluster 1\n", + "Example n. 3179 = (35.0, 169.0): cluster 1\n", + "Example n. 3180 = (35.0, 168.0): cluster 1\n", + "Example n. 3181 = (35.0, 167.0): cluster 1\n", + "Example n. 3182 = (35.0, 166.0): cluster 1\n", + "Example n. 3183 = (35.0, 165.0): cluster 1\n", + "Example n. 3184 = (35.0, 164.0): cluster 1\n", + "Example n. 3185 = (35.0, 163.0): cluster 1\n", + "Example n. 3186 = (35.0, 162.0): cluster 1\n", + "Example n. 3187 = (35.0, 161.0): cluster 1\n", + "Example n. 3188 = (35.0, 160.0): cluster 1\n", + "Example n. 3189 = (35.0, 159.0): cluster 1\n", + "Example n. 3190 = (35.0, 158.0): cluster 1\n", + "Example n. 3191 = (35.0, 157.0): cluster 1\n", + "Example n. 3192 = (35.0, 156.0): cluster 1\n", + "Example n. 3193 = (35.0, 155.0): cluster 1\n", + "Example n. 3194 = (35.0, 154.0): cluster 1\n", + "Example n. 3195 = (35.0, 153.0): cluster 1\n", + "Example n. 3196 = (35.0, 152.0): cluster 1\n", + "Example n. 3197 = (35.0, 151.0): cluster 1\n", + "Example n. 3198 = (35.0, 150.0): cluster 1\n", + "Example n. 3199 = (35.0, 136.0): cluster 1\n", + "Example n. 3200 = (35.0, 135.0): cluster 1\n", + "Example n. 3201 = (35.0, 134.0): cluster 1\n", + "Example n. 3202 = (35.0, 133.0): cluster 1\n", + "Example n. 3203 = (35.0, 132.0): cluster 1\n", + "Example n. 3204 = (35.0, 131.0): cluster 1\n", + "Example n. 3205 = (35.0, 130.0): cluster 1\n", + "Example n. 3206 = (35.0, 129.0): cluster 1\n", + "Example n. 3207 = (35.0, 128.0): cluster 1\n", + "Example n. 3208 = (35.0, 127.0): cluster 1\n", + "Example n. 3209 = (35.0, 114.0): cluster 1\n", + "Example n. 3210 = (35.0, 113.0): cluster 1\n", + "Example n. 3211 = (35.0, 112.0): cluster 1\n", + "Example n. 3212 = (35.0, 111.0): cluster 1\n", + "Example n. 3213 = (35.0, 110.0): cluster 1\n", + "Example n. 3214 = (35.0, 109.0): cluster 1\n", + "Example n. 3215 = (35.0, 108.0): cluster 1\n", + "Example n. 3216 = (35.0, 107.0): cluster 1\n", + "Example n. 3217 = (35.0, 106.0): cluster 1\n", + "Example n. 3218 = (35.0, 105.0): cluster 1\n", + "Example n. 3219 = (35.0, 104.0): cluster 1\n", + "Example n. 3220 = (36.0, 489.0): cluster 1\n", + "Example n. 3221 = (36.0, 488.0): cluster 1\n", + "Example n. 3222 = (36.0, 487.0): cluster 1\n", + "Example n. 3223 = (36.0, 486.0): cluster 1\n", + "Example n. 3224 = (36.0, 485.0): cluster 1\n", + "Example n. 3225 = (36.0, 484.0): cluster 1\n", + "Example n. 3226 = (36.0, 483.0): cluster 1\n", + "Example n. 3227 = (36.0, 482.0): cluster 1\n", + "Example n. 3228 = (36.0, 481.0): cluster 3\n", + "Example n. 3229 = (36.0, 480.0): cluster 3\n", + "Example n. 3230 = (36.0, 479.0): cluster 3\n", + "Example n. 3231 = (36.0, 478.0): cluster 3\n", + "Example n. 3232 = (36.0, 477.0): cluster 3\n", + "Example n. 3233 = (36.0, 476.0): cluster 3\n", + "Example n. 3234 = (36.0, 475.0): cluster 3\n", + "Example n. 3235 = (36.0, 474.0): cluster 3\n", + "Example n. 3236 = (36.0, 473.0): cluster 3\n", + "Example n. 3237 = (36.0, 472.0): cluster 3\n", + "Example n. 3238 = (36.0, 471.0): cluster 3\n", + "Example n. 3239 = (36.0, 470.0): cluster 3\n", + "Example n. 3240 = (36.0, 469.0): cluster 3\n", + "Example n. 3241 = (36.0, 468.0): cluster 3\n", + "Example n. 3242 = (36.0, 467.0): cluster 3\n", + "Example n. 3243 = (36.0, 466.0): cluster 3\n", + "Example n. 3244 = (36.0, 465.0): cluster 3\n", + "Example n. 3245 = (36.0, 464.0): cluster 3\n", + "Example n. 3246 = (36.0, 463.0): cluster 3\n", + "Example n. 3247 = (36.0, 462.0): cluster 3\n", + "Example n. 3248 = (36.0, 461.0): cluster 3\n", + "Example n. 3249 = (36.0, 460.0): cluster 3\n", + "Example n. 3250 = (36.0, 459.0): cluster 3\n", + "Example n. 3251 = (36.0, 458.0): cluster 3\n", + "Example n. 3252 = (36.0, 396.0): cluster 3\n", + "Example n. 3253 = (36.0, 395.0): cluster 3\n", + "Example n. 3254 = (36.0, 394.0): cluster 3\n", + "Example n. 3255 = (36.0, 382.0): cluster 3\n", + "Example n. 3256 = (36.0, 381.0): cluster 3\n", + "Example n. 3257 = (36.0, 380.0): cluster 3\n", + "Example n. 3258 = (36.0, 379.0): cluster 3\n", + "Example n. 3259 = (36.0, 378.0): cluster 3\n", + "Example n. 3260 = (36.0, 377.0): cluster 3\n", + "Example n. 3261 = (36.0, 376.0): cluster 3\n", + "Example n. 3262 = (36.0, 375.0): cluster 3\n", + "Example n. 3263 = (36.0, 312.0): cluster 3\n", + "Example n. 3264 = (36.0, 311.0): cluster 3\n", + "Example n. 3265 = (36.0, 310.0): cluster 1\n", + "Example n. 3266 = (36.0, 309.0): cluster 1\n", + "Example n. 3267 = (36.0, 308.0): cluster 1\n", + "Example n. 3268 = (36.0, 307.0): cluster 1\n", + "Example n. 3269 = (36.0, 306.0): cluster 1\n", + "Example n. 3270 = (36.0, 305.0): cluster 1\n", + "Example n. 3271 = (36.0, 304.0): cluster 1\n", + "Example n. 3272 = (36.0, 303.0): cluster 1\n", + "Example n. 3273 = (36.0, 302.0): cluster 1\n", + "Example n. 3274 = (36.0, 301.0): cluster 1\n", + "Example n. 3275 = (36.0, 300.0): cluster 1\n", + "Example n. 3276 = (36.0, 299.0): cluster 1\n", + "Example n. 3277 = (36.0, 298.0): cluster 1\n", + "Example n. 3278 = (36.0, 297.0): cluster 1\n", + "Example n. 3279 = (36.0, 291.0): cluster 1\n", + "Example n. 3280 = (36.0, 290.0): cluster 1\n", + "Example n. 3281 = (36.0, 289.0): cluster 1\n", + "Example n. 3282 = (36.0, 288.0): cluster 1\n", + "Example n. 3283 = (36.0, 287.0): cluster 1\n", + "Example n. 3284 = (36.0, 286.0): cluster 1\n", + "Example n. 3285 = (36.0, 285.0): cluster 1\n", + "Example n. 3286 = (36.0, 284.0): cluster 1\n", + "Example n. 3287 = (36.0, 283.0): cluster 1\n", + "Example n. 3288 = (36.0, 282.0): cluster 1\n", + "Example n. 3289 = (36.0, 281.0): cluster 1\n", + "Example n. 3290 = (36.0, 280.0): cluster 1\n", + "Example n. 3291 = (36.0, 279.0): cluster 1\n", + "Example n. 3292 = (36.0, 278.0): cluster 1\n", + "Example n. 3293 = (36.0, 277.0): cluster 1\n", + "Example n. 3294 = (36.0, 269.0): cluster 1\n", + "Example n. 3295 = (36.0, 268.0): cluster 1\n", + "Example n. 3296 = (36.0, 267.0): cluster 1\n", + "Example n. 3297 = (36.0, 266.0): cluster 1\n", + "Example n. 3298 = (36.0, 265.0): cluster 1\n", + "Example n. 3299 = (36.0, 264.0): cluster 1\n", + "Example n. 3300 = (36.0, 263.0): cluster 1\n", + "Example n. 3301 = (36.0, 262.0): cluster 1\n", + "Example n. 3302 = (36.0, 261.0): cluster 1\n", + "Example n. 3303 = (36.0, 260.0): cluster 1\n", + "Example n. 3304 = (36.0, 259.0): cluster 1\n", + "Example n. 3305 = (36.0, 258.0): cluster 1\n", + "Example n. 3306 = (36.0, 257.0): cluster 1\n", + "Example n. 3307 = (36.0, 256.0): cluster 1\n", + "Example n. 3308 = (36.0, 255.0): cluster 1\n", + "Example n. 3309 = (36.0, 254.0): cluster 1\n", + "Example n. 3310 = (36.0, 253.0): cluster 1\n", + "Example n. 3311 = (36.0, 245.0): cluster 1\n", + "Example n. 3312 = (36.0, 244.0): cluster 1\n", + "Example n. 3313 = (36.0, 243.0): cluster 1\n", + "Example n. 3314 = (36.0, 242.0): cluster 1\n", + "Example n. 3315 = (36.0, 241.0): cluster 1\n", + "Example n. 3316 = (36.0, 240.0): cluster 1\n", + "Example n. 3317 = (36.0, 239.0): cluster 1\n", + "Example n. 3318 = (36.0, 238.0): cluster 3\n", + "Example n. 3319 = (36.0, 237.0): cluster 3\n", + "Example n. 3320 = (36.0, 236.0): cluster 3\n", + "Example n. 3321 = (36.0, 235.0): cluster 3\n", + "Example n. 3322 = (36.0, 234.0): cluster 3\n", + "Example n. 3323 = (36.0, 226.0): cluster 3\n", + "Example n. 3324 = (36.0, 225.0): cluster 3\n", + "Example n. 3325 = (36.0, 224.0): cluster 3\n", + "Example n. 3326 = (36.0, 223.0): cluster 3\n", + "Example n. 3327 = (36.0, 222.0): cluster 3\n", + "Example n. 3328 = (36.0, 221.0): cluster 3\n", + "Example n. 3329 = (36.0, 220.0): cluster 3\n", + "Example n. 3330 = (36.0, 219.0): cluster 3\n", + "Example n. 3331 = (36.0, 218.0): cluster 3\n", + "Example n. 3332 = (36.0, 214.0): cluster 3\n", + "Example n. 3333 = (36.0, 213.0): cluster 3\n", + "Example n. 3334 = (36.0, 212.0): cluster 3\n", + "Example n. 3335 = (36.0, 211.0): cluster 3\n", + "Example n. 3336 = (36.0, 210.0): cluster 3\n", + "Example n. 3337 = (36.0, 195.0): cluster 3\n", + "Example n. 3338 = (36.0, 194.0): cluster 3\n", + "Example n. 3339 = (36.0, 193.0): cluster 3\n", + "Example n. 3340 = (36.0, 192.0): cluster 3\n", + "Example n. 3341 = (36.0, 191.0): cluster 3\n", + "Example n. 3342 = (36.0, 190.0): cluster 3\n", + "Example n. 3343 = (36.0, 189.0): cluster 3\n", + "Example n. 3344 = (36.0, 188.0): cluster 3\n", + "Example n. 3345 = (36.0, 187.0): cluster 3\n", + "Example n. 3346 = (36.0, 186.0): cluster 3\n", + "Example n. 3347 = (36.0, 185.0): cluster 3\n", + "Example n. 3348 = (36.0, 184.0): cluster 3\n", + "Example n. 3349 = (36.0, 183.0): cluster 3\n", + "Example n. 3350 = (36.0, 182.0): cluster 3\n", + "Example n. 3351 = (36.0, 181.0): cluster 3\n", + "Example n. 3352 = (36.0, 180.0): cluster 3\n", + "Example n. 3353 = (36.0, 179.0): cluster 3\n", + "Example n. 3354 = (36.0, 178.0): cluster 3\n", + "Example n. 3355 = (36.0, 170.0): cluster 3\n", + "Example n. 3356 = (36.0, 169.0): cluster 3\n", + "Example n. 3357 = (36.0, 168.0): cluster 3\n", + "Example n. 3358 = (36.0, 167.0): cluster 3\n", + "Example n. 3359 = (36.0, 166.0): cluster 1\n", + "Example n. 3360 = (36.0, 165.0): cluster 1\n", + "Example n. 3361 = (36.0, 163.0): cluster 1\n", + "Example n. 3362 = (36.0, 162.0): cluster 1\n", + "Example n. 3363 = (36.0, 161.0): cluster 1\n", + "Example n. 3364 = (36.0, 160.0): cluster 1\n", + "Example n. 3365 = (36.0, 159.0): cluster 1\n", + "Example n. 3366 = (36.0, 158.0): cluster 1\n", + "Example n. 3367 = (36.0, 157.0): cluster 1\n", + "Example n. 3368 = (36.0, 156.0): cluster 1\n", + "Example n. 3369 = (36.0, 155.0): cluster 1\n", + "Example n. 3370 = (36.0, 154.0): cluster 1\n", + "Example n. 3371 = (36.0, 153.0): cluster 1\n", + "Example n. 3372 = (36.0, 152.0): cluster 1\n", + "Example n. 3373 = (36.0, 151.0): cluster 1\n", + "Example n. 3374 = (36.0, 150.0): cluster 1\n", + "Example n. 3375 = (36.0, 149.0): cluster 1\n", + "Example n. 3376 = (36.0, 148.0): cluster 1\n", + "Example n. 3377 = (36.0, 147.0): cluster 1\n", + "Example n. 3378 = (36.0, 136.0): cluster 1\n", + "Example n. 3379 = (36.0, 135.0): cluster 1\n", + "Example n. 3380 = (36.0, 134.0): cluster 1\n", + "Example n. 3381 = (36.0, 133.0): cluster 1\n", + "Example n. 3382 = (36.0, 132.0): cluster 1\n", + "Example n. 3383 = (36.0, 131.0): cluster 1\n", + "Example n. 3384 = (36.0, 130.0): cluster 1\n", + "Example n. 3385 = (36.0, 129.0): cluster 1\n", + "Example n. 3386 = (36.0, 128.0): cluster 1\n", + "Example n. 3387 = (36.0, 127.0): cluster 1\n", + "Example n. 3388 = (36.0, 126.0): cluster 1\n", + "Example n. 3389 = (36.0, 114.0): cluster 1\n", + "Example n. 3390 = (36.0, 113.0): cluster 1\n", + "Example n. 3391 = (36.0, 112.0): cluster 1\n", + "Example n. 3392 = (36.0, 111.0): cluster 1\n", + "Example n. 3393 = (36.0, 110.0): cluster 1\n", + "Example n. 3394 = (36.0, 109.0): cluster 1\n", + "Example n. 3395 = (36.0, 108.0): cluster 1\n", + "Example n. 3396 = (36.0, 107.0): cluster 1\n", + "Example n. 3397 = (36.0, 106.0): cluster 1\n", + "Example n. 3398 = (36.0, 105.0): cluster 1\n", + "Example n. 3399 = (36.0, 104.0): cluster 1\n", + "Example n. 3400 = (37.0, 490.0): cluster 1\n", + "Example n. 3401 = (37.0, 489.0): cluster 1\n", + "Example n. 3402 = (37.0, 488.0): cluster 1\n", + "Example n. 3403 = (37.0, 487.0): cluster 1\n", + "Example n. 3404 = (37.0, 486.0): cluster 3\n", + "Example n. 3405 = (37.0, 485.0): cluster 3\n", + "Example n. 3406 = (37.0, 484.0): cluster 3\n", + "Example n. 3407 = (37.0, 483.0): cluster 3\n", + "Example n. 3408 = (37.0, 482.0): cluster 3\n", + "Example n. 3409 = (37.0, 481.0): cluster 3\n", + "Example n. 3410 = (37.0, 480.0): cluster 3\n", + "Example n. 3411 = (37.0, 479.0): cluster 3\n", + "Example n. 3412 = (37.0, 478.0): cluster 3\n", + "Example n. 3413 = (37.0, 477.0): cluster 3\n", + "Example n. 3414 = (37.0, 476.0): cluster 3\n", + "Example n. 3415 = (37.0, 475.0): cluster 3\n", + "Example n. 3416 = (37.0, 474.0): cluster 3\n", + "Example n. 3417 = (37.0, 473.0): cluster 3\n", + "Example n. 3418 = (37.0, 472.0): cluster 3\n", + "Example n. 3419 = (37.0, 471.0): cluster 3\n", + "Example n. 3420 = (37.0, 470.0): cluster 3\n", + "Example n. 3421 = (37.0, 469.0): cluster 3\n", + "Example n. 3422 = (37.0, 468.0): cluster 3\n", + "Example n. 3423 = (37.0, 467.0): cluster 3\n", + "Example n. 3424 = (37.0, 466.0): cluster 3\n", + "Example n. 3425 = (37.0, 464.0): cluster 3\n", + "Example n. 3426 = (37.0, 463.0): cluster 3\n", + "Example n. 3427 = (37.0, 462.0): cluster 3\n", + "Example n. 3428 = (37.0, 461.0): cluster 3\n", + "Example n. 3429 = (37.0, 460.0): cluster 3\n", + "Example n. 3430 = (37.0, 459.0): cluster 3\n", + "Example n. 3431 = (37.0, 458.0): cluster 3\n", + "Example n. 3432 = (37.0, 457.0): cluster 3\n", + "Example n. 3433 = (37.0, 382.0): cluster 3\n", + "Example n. 3434 = (37.0, 381.0): cluster 3\n", + "Example n. 3435 = (37.0, 380.0): cluster 3\n", + "Example n. 3436 = (37.0, 379.0): cluster 3\n", + "Example n. 3437 = (37.0, 378.0): cluster 3\n", + "Example n. 3438 = (37.0, 377.0): cluster 1\n", + "Example n. 3439 = (37.0, 376.0): cluster 1\n", + "Example n. 3440 = (37.0, 312.0): cluster 1\n", + "Example n. 3441 = (37.0, 311.0): cluster 1\n", + "Example n. 3442 = (37.0, 310.0): cluster 1\n", + "Example n. 3443 = (37.0, 309.0): cluster 1\n", + "Example n. 3444 = (37.0, 308.0): cluster 1\n", + "Example n. 3445 = (37.0, 307.0): cluster 1\n", + "Example n. 3446 = (37.0, 306.0): cluster 1\n", + "Example n. 3447 = (37.0, 305.0): cluster 1\n", + "Example n. 3448 = (37.0, 304.0): cluster 1\n", + "Example n. 3449 = (37.0, 303.0): cluster 1\n", + "Example n. 3450 = (37.0, 302.0): cluster 1\n", + "Example n. 3451 = (37.0, 301.0): cluster 1\n", + "Example n. 3452 = (37.0, 300.0): cluster 1\n", + "Example n. 3453 = (37.0, 299.0): cluster 1\n", + "Example n. 3454 = (37.0, 298.0): cluster 1\n", + "Example n. 3455 = (37.0, 297.0): cluster 1\n", + "Example n. 3456 = (37.0, 293.0): cluster 1\n", + "Example n. 3457 = (37.0, 292.0): cluster 1\n", + "Example n. 3458 = (37.0, 291.0): cluster 1\n", + "Example n. 3459 = (37.0, 290.0): cluster 1\n", + "Example n. 3460 = (37.0, 289.0): cluster 1\n", + "Example n. 3461 = (37.0, 288.0): cluster 1\n", + "Example n. 3462 = (37.0, 287.0): cluster 1\n", + "Example n. 3463 = (37.0, 286.0): cluster 1\n", + "Example n. 3464 = (37.0, 285.0): cluster 1\n", + "Example n. 3465 = (37.0, 284.0): cluster 1\n", + "Example n. 3466 = (37.0, 281.0): cluster 1\n", + "Example n. 3467 = (37.0, 280.0): cluster 1\n", + "Example n. 3468 = (37.0, 279.0): cluster 1\n", + "Example n. 3469 = (37.0, 278.0): cluster 1\n", + "Example n. 3470 = (37.0, 277.0): cluster 1\n", + "Example n. 3471 = (37.0, 270.0): cluster 1\n", + "Example n. 3472 = (37.0, 269.0): cluster 1\n", + "Example n. 3473 = (37.0, 268.0): cluster 1\n", + "Example n. 3474 = (37.0, 267.0): cluster 1\n", + "Example n. 3475 = (37.0, 266.0): cluster 1\n", + "Example n. 3476 = (37.0, 265.0): cluster 1\n", + "Example n. 3477 = (37.0, 264.0): cluster 1\n", + "Example n. 3478 = (37.0, 263.0): cluster 1\n", + "Example n. 3479 = (37.0, 262.0): cluster 1\n", + "Example n. 3480 = (37.0, 261.0): cluster 1\n", + "Example n. 3481 = (37.0, 260.0): cluster 1\n", + "Example n. 3482 = (37.0, 259.0): cluster 1\n", + "Example n. 3483 = (37.0, 258.0): cluster 3\n", + "Example n. 3484 = (37.0, 257.0): cluster 3\n", + "Example n. 3485 = (37.0, 256.0): cluster 3\n", + "Example n. 3486 = (37.0, 255.0): cluster 3\n", + "Example n. 3487 = (37.0, 254.0): cluster 3\n", + "Example n. 3488 = (37.0, 253.0): cluster 3\n", + "Example n. 3489 = (37.0, 252.0): cluster 3\n", + "Example n. 3490 = (37.0, 245.0): cluster 3\n", + "Example n. 3491 = (37.0, 244.0): cluster 3\n", + "Example n. 3492 = (37.0, 243.0): cluster 3\n", + "Example n. 3493 = (37.0, 242.0): cluster 3\n", + "Example n. 3494 = (37.0, 241.0): cluster 3\n", + "Example n. 3495 = (37.0, 240.0): cluster 3\n", + "Example n. 3496 = (37.0, 239.0): cluster 3\n", + "Example n. 3497 = (37.0, 238.0): cluster 3\n", + "Example n. 3498 = (37.0, 237.0): cluster 3\n", + "Example n. 3499 = (37.0, 236.0): cluster 3\n", + "Example n. 3500 = (37.0, 235.0): cluster 3\n", + "Example n. 3501 = (37.0, 230.0): cluster 3\n", + "Example n. 3502 = (37.0, 229.0): cluster 3\n", + "Example n. 3503 = (37.0, 228.0): cluster 3\n", + "Example n. 3504 = (37.0, 227.0): cluster 3\n", + "Example n. 3505 = (37.0, 226.0): cluster 3\n", + "Example n. 3506 = (37.0, 225.0): cluster 3\n", + "Example n. 3507 = (37.0, 224.0): cluster 3\n", + "Example n. 3508 = (37.0, 223.0): cluster 3\n", + "Example n. 3509 = (37.0, 222.0): cluster 3\n", + "Example n. 3510 = (37.0, 221.0): cluster 3\n", + "Example n. 3511 = (37.0, 220.0): cluster 3\n", + "Example n. 3512 = (37.0, 219.0): cluster 3\n", + "Example n. 3513 = (37.0, 218.0): cluster 3\n", + "Example n. 3514 = (37.0, 214.0): cluster 3\n", + "Example n. 3515 = (37.0, 213.0): cluster 3\n", + "Example n. 3516 = (37.0, 212.0): cluster 3\n", + "Example n. 3517 = (37.0, 211.0): cluster 1\n", + "Example n. 3518 = (37.0, 210.0): cluster 1\n", + "Example n. 3519 = (37.0, 209.0): cluster 1\n", + "Example n. 3520 = (37.0, 208.0): cluster 1\n", + "Example n. 3521 = (37.0, 196.0): cluster 1\n", + "Example n. 3522 = (37.0, 195.0): cluster 1\n", + "Example n. 3523 = (37.0, 194.0): cluster 1\n", + "Example n. 3524 = (37.0, 193.0): cluster 1\n", + "Example n. 3525 = (37.0, 192.0): cluster 1\n", + "Example n. 3526 = (37.0, 191.0): cluster 1\n", + "Example n. 3527 = (37.0, 190.0): cluster 1\n", + "Example n. 3528 = (37.0, 189.0): cluster 1\n", + "Example n. 3529 = (37.0, 188.0): cluster 1\n", + "Example n. 3530 = (37.0, 187.0): cluster 1\n", + "Example n. 3531 = (37.0, 186.0): cluster 1\n", + "Example n. 3532 = (37.0, 185.0): cluster 1\n", + "Example n. 3533 = (37.0, 184.0): cluster 1\n", + "Example n. 3534 = (37.0, 183.0): cluster 1\n", + "Example n. 3535 = (37.0, 182.0): cluster 1\n", + "Example n. 3536 = (37.0, 181.0): cluster 1\n", + "Example n. 3537 = (37.0, 180.0): cluster 1\n", + "Example n. 3538 = (37.0, 179.0): cluster 1\n", + "Example n. 3539 = (37.0, 178.0): cluster 1\n", + "Example n. 3540 = (37.0, 177.0): cluster 1\n", + "Example n. 3541 = (37.0, 163.0): cluster 1\n", + "Example n. 3542 = (37.0, 162.0): cluster 1\n", + "Example n. 3543 = (37.0, 161.0): cluster 1\n", + "Example n. 3544 = (37.0, 160.0): cluster 1\n", + "Example n. 3545 = (37.0, 159.0): cluster 1\n", + "Example n. 3546 = (37.0, 158.0): cluster 1\n", + "Example n. 3547 = (37.0, 157.0): cluster 1\n", + "Example n. 3548 = (37.0, 156.0): cluster 1\n", + "Example n. 3549 = (37.0, 155.0): cluster 1\n", + "Example n. 3550 = (37.0, 154.0): cluster 1\n", + "Example n. 3551 = (37.0, 153.0): cluster 1\n", + "Example n. 3552 = (37.0, 152.0): cluster 1\n", + "Example n. 3553 = (37.0, 151.0): cluster 1\n", + "Example n. 3554 = (37.0, 150.0): cluster 1\n", + "Example n. 3555 = (37.0, 149.0): cluster 1\n", + "Example n. 3556 = (37.0, 148.0): cluster 1\n", + "Example n. 3557 = (37.0, 147.0): cluster 1\n", + "Example n. 3558 = (37.0, 146.0): cluster 1\n", + "Example n. 3559 = (37.0, 132.0): cluster 1\n", + "Example n. 3560 = (37.0, 131.0): cluster 1\n", + "Example n. 3561 = (37.0, 130.0): cluster 1\n", + "Example n. 3562 = (37.0, 129.0): cluster 3\n", + "Example n. 3563 = (37.0, 128.0): cluster 3\n", + "Example n. 3564 = (37.0, 127.0): cluster 3\n", + "Example n. 3565 = (37.0, 126.0): cluster 3\n", + "Example n. 3566 = (37.0, 114.0): cluster 3\n", + "Example n. 3567 = (37.0, 113.0): cluster 3\n", + "Example n. 3568 = (37.0, 112.0): cluster 3\n", + "Example n. 3569 = (37.0, 111.0): cluster 3\n", + "Example n. 3570 = (37.0, 110.0): cluster 3\n", + "Example n. 3571 = (37.0, 109.0): cluster 3\n", + "Example n. 3572 = (37.0, 108.0): cluster 3\n", + "Example n. 3573 = (37.0, 107.0): cluster 3\n", + "Example n. 3574 = (37.0, 106.0): cluster 3\n", + "Example n. 3575 = (37.0, 105.0): cluster 3\n", + "Example n. 3576 = (37.0, 104.0): cluster 3\n", + "Example n. 3577 = (38.0, 491.0): cluster 3\n", + "Example n. 3578 = (38.0, 490.0): cluster 3\n", + "Example n. 3579 = (38.0, 489.0): cluster 3\n", + "Example n. 3580 = (38.0, 488.0): cluster 3\n", + "Example n. 3581 = (38.0, 487.0): cluster 3\n", + "Example n. 3582 = (38.0, 486.0): cluster 3\n", + "Example n. 3583 = (38.0, 485.0): cluster 3\n", + "Example n. 3584 = (38.0, 484.0): cluster 3\n", + "Example n. 3585 = (38.0, 483.0): cluster 3\n", + "Example n. 3586 = (38.0, 482.0): cluster 3\n", + "Example n. 3587 = (38.0, 481.0): cluster 3\n", + "Example n. 3588 = (38.0, 480.0): cluster 3\n", + "Example n. 3589 = (38.0, 479.0): cluster 3\n", + "Example n. 3590 = (38.0, 478.0): cluster 3\n", + "Example n. 3591 = (38.0, 477.0): cluster 3\n", + "Example n. 3592 = (38.0, 476.0): cluster 3\n", + "Example n. 3593 = (38.0, 475.0): cluster 3\n", + "Example n. 3594 = (38.0, 474.0): cluster 3\n", + "Example n. 3595 = (38.0, 473.0): cluster 3\n", + "Example n. 3596 = (38.0, 472.0): cluster 1\n", + "Example n. 3597 = (38.0, 471.0): cluster 1\n", + "Example n. 3598 = (38.0, 470.0): cluster 1\n", + "Example n. 3599 = (38.0, 469.0): cluster 1\n", + "Example n. 3600 = (38.0, 468.0): cluster 1\n", + "Example n. 3601 = (38.0, 467.0): cluster 1\n", + "Example n. 3602 = (38.0, 466.0): cluster 1\n", + "Example n. 3603 = (38.0, 465.0): cluster 1\n", + "Example n. 3604 = (38.0, 464.0): cluster 1\n", + "Example n. 3605 = (38.0, 463.0): cluster 1\n", + "Example n. 3606 = (38.0, 462.0): cluster 1\n", + "Example n. 3607 = (38.0, 461.0): cluster 1\n", + "Example n. 3608 = (38.0, 460.0): cluster 1\n", + "Example n. 3609 = (38.0, 459.0): cluster 1\n", + "Example n. 3610 = (38.0, 458.0): cluster 1\n", + "Example n. 3611 = (38.0, 457.0): cluster 1\n", + "Example n. 3612 = (38.0, 456.0): cluster 1\n", + "Example n. 3613 = (38.0, 381.0): cluster 1\n", + "Example n. 3614 = (38.0, 380.0): cluster 1\n", + "Example n. 3615 = (38.0, 379.0): cluster 1\n", + "Example n. 3616 = (38.0, 378.0): cluster 1\n", + "Example n. 3617 = (38.0, 377.0): cluster 1\n", + "Example n. 3618 = (38.0, 313.0): cluster 1\n", + "Example n. 3619 = (38.0, 312.0): cluster 1\n", + "Example n. 3620 = (38.0, 311.0): cluster 1\n", + "Example n. 3621 = (38.0, 310.0): cluster 1\n", + "Example n. 3622 = (38.0, 309.0): cluster 1\n", + "Example n. 3623 = (38.0, 308.0): cluster 1\n", + "Example n. 3624 = (38.0, 307.0): cluster 1\n", + "Example n. 3625 = (38.0, 306.0): cluster 1\n", + "Example n. 3626 = (38.0, 305.0): cluster 1\n", + "Example n. 3627 = (38.0, 304.0): cluster 1\n", + "Example n. 3628 = (38.0, 303.0): cluster 1\n", + "Example n. 3629 = (38.0, 302.0): cluster 1\n", + "Example n. 3630 = (38.0, 301.0): cluster 1\n", + "Example n. 3631 = (38.0, 300.0): cluster 1\n", + "Example n. 3632 = (38.0, 299.0): cluster 1\n", + "Example n. 3633 = (38.0, 298.0): cluster 1\n", + "Example n. 3634 = (38.0, 297.0): cluster 1\n", + "Example n. 3635 = (38.0, 296.0): cluster 1\n", + "Example n. 3636 = (38.0, 293.0): cluster 1\n", + "Example n. 3637 = (38.0, 292.0): cluster 1\n", + "Example n. 3638 = (38.0, 291.0): cluster 1\n", + "Example n. 3639 = (38.0, 290.0): cluster 1\n", + "Example n. 3640 = (38.0, 289.0): cluster 1\n", + "Example n. 3641 = (38.0, 288.0): cluster 3\n", + "Example n. 3642 = (38.0, 287.0): cluster 3\n", + "Example n. 3643 = (38.0, 286.0): cluster 3\n", + "Example n. 3644 = (38.0, 285.0): cluster 3\n", + "Example n. 3645 = (38.0, 284.0): cluster 3\n", + "Example n. 3646 = (38.0, 281.0): cluster 3\n", + "Example n. 3647 = (38.0, 280.0): cluster 3\n", + "Example n. 3648 = (38.0, 279.0): cluster 3\n", + "Example n. 3649 = (38.0, 278.0): cluster 3\n", + "Example n. 3650 = (38.0, 272.0): cluster 3\n", + "Example n. 3651 = (38.0, 271.0): cluster 3\n", + "Example n. 3652 = (38.0, 270.0): cluster 3\n", + "Example n. 3653 = (38.0, 269.0): cluster 3\n", + "Example n. 3654 = (38.0, 268.0): cluster 3\n", + "Example n. 3655 = (38.0, 267.0): cluster 3\n", + "Example n. 3656 = (38.0, 266.0): cluster 3\n", + "Example n. 3657 = (38.0, 265.0): cluster 3\n", + "Example n. 3658 = (38.0, 264.0): cluster 3\n", + "Example n. 3659 = (38.0, 263.0): cluster 3\n", + "Example n. 3660 = (38.0, 262.0): cluster 3\n", + "Example n. 3661 = (38.0, 261.0): cluster 3\n", + "Example n. 3662 = (38.0, 260.0): cluster 3\n", + "Example n. 3663 = (38.0, 259.0): cluster 3\n", + "Example n. 3664 = (38.0, 258.0): cluster 3\n", + "Example n. 3665 = (38.0, 257.0): cluster 3\n", + "Example n. 3666 = (38.0, 256.0): cluster 3\n", + "Example n. 3667 = (38.0, 255.0): cluster 3\n", + "Example n. 3668 = (38.0, 254.0): cluster 3\n", + "Example n. 3669 = (38.0, 253.0): cluster 3\n", + "Example n. 3670 = (38.0, 252.0): cluster 3\n", + "Example n. 3671 = (38.0, 245.0): cluster 3\n", + "Example n. 3672 = (38.0, 244.0): cluster 3\n", + "Example n. 3673 = (38.0, 243.0): cluster 3\n", + "Example n. 3674 = (38.0, 242.0): cluster 3\n", + "Example n. 3675 = (38.0, 241.0): cluster 3\n", + "Example n. 3676 = (38.0, 240.0): cluster 3\n", + "Example n. 3677 = (38.0, 239.0): cluster 3\n", + "Example n. 3678 = (38.0, 238.0): cluster 3\n", + "Example n. 3679 = (38.0, 237.0): cluster 3\n", + "Example n. 3680 = (38.0, 236.0): cluster 3\n", + "Example n. 3681 = (38.0, 235.0): cluster 3\n", + "Example n. 3682 = (38.0, 234.0): cluster 1\n", + "Example n. 3683 = (38.0, 233.0): cluster 1\n", + "Example n. 3684 = (38.0, 232.0): cluster 1\n", + "Example n. 3685 = (38.0, 231.0): cluster 1\n", + "Example n. 3686 = (38.0, 230.0): cluster 1\n", + "Example n. 3687 = (38.0, 229.0): cluster 1\n", + "Example n. 3688 = (38.0, 228.0): cluster 1\n", + "Example n. 3689 = (38.0, 227.0): cluster 1\n", + "Example n. 3690 = (38.0, 226.0): cluster 1\n", + "Example n. 3691 = (38.0, 224.0): cluster 1\n", + "Example n. 3692 = (38.0, 223.0): cluster 1\n", + "Example n. 3693 = (38.0, 222.0): cluster 1\n", + "Example n. 3694 = (38.0, 221.0): cluster 1\n", + "Example n. 3695 = (38.0, 220.0): cluster 1\n", + "Example n. 3696 = (38.0, 219.0): cluster 1\n", + "Example n. 3697 = (38.0, 218.0): cluster 1\n", + "Example n. 3698 = (38.0, 212.0): cluster 1\n", + "Example n. 3699 = (38.0, 211.0): cluster 1\n", + "Example n. 3700 = (38.0, 210.0): cluster 1\n", + "Example n. 3701 = (38.0, 209.0): cluster 1\n", + "Example n. 3702 = (38.0, 208.0): cluster 1\n", + "Example n. 3703 = (38.0, 207.0): cluster 1\n", + "Example n. 3704 = (38.0, 197.0): cluster 1\n", + "Example n. 3705 = (38.0, 196.0): cluster 1\n", + "Example n. 3706 = (38.0, 195.0): cluster 1\n", + "Example n. 3707 = (38.0, 194.0): cluster 1\n", + "Example n. 3708 = (38.0, 193.0): cluster 1\n", + "Example n. 3709 = (38.0, 192.0): cluster 1\n", + "Example n. 3710 = (38.0, 191.0): cluster 1\n", + "Example n. 3711 = (38.0, 190.0): cluster 1\n", + "Example n. 3712 = (38.0, 189.0): cluster 1\n", + "Example n. 3713 = (38.0, 188.0): cluster 1\n", + "Example n. 3714 = (38.0, 187.0): cluster 1\n", + "Example n. 3715 = (38.0, 186.0): cluster 1\n", + "Example n. 3716 = (38.0, 185.0): cluster 1\n", + "Example n. 3717 = (38.0, 184.0): cluster 1\n", + "Example n. 3718 = (38.0, 183.0): cluster 1\n", + "Example n. 3719 = (38.0, 182.0): cluster 1\n", + "Example n. 3720 = (38.0, 181.0): cluster 1\n", + "Example n. 3721 = (38.0, 180.0): cluster 1\n", + "Example n. 3722 = (38.0, 179.0): cluster 1\n", + "Example n. 3723 = (38.0, 178.0): cluster 3\n", + "Example n. 3724 = (38.0, 177.0): cluster 3\n", + "Example n. 3725 = (38.0, 174.0): cluster 3\n", + "Example n. 3726 = (38.0, 173.0): cluster 3\n", + "Example n. 3727 = (38.0, 172.0): cluster 3\n", + "Example n. 3728 = (38.0, 171.0): cluster 3\n", + "Example n. 3729 = (38.0, 163.0): cluster 3\n", + "Example n. 3730 = (38.0, 162.0): cluster 3\n", + "Example n. 3731 = (38.0, 161.0): cluster 3\n", + "Example n. 3732 = (38.0, 160.0): cluster 3\n", + "Example n. 3733 = (38.0, 159.0): cluster 3\n", + "Example n. 3734 = (38.0, 158.0): cluster 3\n", + "Example n. 3735 = (38.0, 157.0): cluster 3\n", + "Example n. 3736 = (38.0, 156.0): cluster 3\n", + "Example n. 3737 = (38.0, 155.0): cluster 3\n", + "Example n. 3738 = (38.0, 154.0): cluster 3\n", + "Example n. 3739 = (38.0, 153.0): cluster 3\n", + "Example n. 3740 = (38.0, 152.0): cluster 3\n", + "Example n. 3741 = (38.0, 151.0): cluster 3\n", + "Example n. 3742 = (38.0, 150.0): cluster 3\n", + "Example n. 3743 = (38.0, 149.0): cluster 3\n", + "Example n. 3744 = (38.0, 148.0): cluster 3\n", + "Example n. 3745 = (38.0, 147.0): cluster 3\n", + "Example n. 3746 = (38.0, 146.0): cluster 3\n", + "Example n. 3747 = (38.0, 132.0): cluster 3\n", + "Example n. 3748 = (38.0, 131.0): cluster 3\n", + "Example n. 3749 = (38.0, 130.0): cluster 3\n", + "Example n. 3750 = (38.0, 129.0): cluster 3\n", + "Example n. 3751 = (38.0, 128.0): cluster 3\n", + "Example n. 3752 = (38.0, 127.0): cluster 3\n", + "Example n. 3753 = (38.0, 126.0): cluster 3\n", + "Example n. 3754 = (38.0, 114.0): cluster 3\n", + "Example n. 3755 = (38.0, 113.0): cluster 3\n", + "Example n. 3756 = (38.0, 112.0): cluster 1\n", + "Example n. 3757 = (38.0, 111.0): cluster 1\n", + "Example n. 3758 = (38.0, 110.0): cluster 1\n", + "Example n. 3759 = (38.0, 109.0): cluster 1\n", + "Example n. 3760 = (38.0, 108.0): cluster 1\n", + "Example n. 3761 = (38.0, 107.0): cluster 1\n", + "Example n. 3762 = (38.0, 106.0): cluster 1\n", + "Example n. 3763 = (38.0, 105.0): cluster 1\n", + "Example n. 3764 = (39.0, 492.0): cluster 1\n", + "Example n. 3765 = (39.0, 491.0): cluster 1\n", + "Example n. 3766 = (39.0, 490.0): cluster 1\n", + "Example n. 3767 = (39.0, 489.0): cluster 1\n", + "Example n. 3768 = (39.0, 488.0): cluster 1\n", + "Example n. 3769 = (39.0, 487.0): cluster 1\n", + "Example n. 3770 = (39.0, 486.0): cluster 1\n", + "Example n. 3771 = (39.0, 485.0): cluster 1\n", + "Example n. 3772 = (39.0, 484.0): cluster 1\n", + "Example n. 3773 = (39.0, 483.0): cluster 1\n", + "Example n. 3774 = (39.0, 482.0): cluster 1\n", + "Example n. 3775 = (39.0, 481.0): cluster 1\n", + "Example n. 3776 = (39.0, 480.0): cluster 1\n", + "Example n. 3777 = (39.0, 479.0): cluster 1\n", + "Example n. 3778 = (39.0, 478.0): cluster 1\n", + "Example n. 3779 = (39.0, 477.0): cluster 1\n", + "Example n. 3780 = (39.0, 476.0): cluster 1\n", + "Example n. 3781 = (39.0, 475.0): cluster 1\n", + "Example n. 3782 = (39.0, 474.0): cluster 1\n", + "Example n. 3783 = (39.0, 473.0): cluster 1\n", + "Example n. 3784 = (39.0, 472.0): cluster 1\n", + "Example n. 3785 = (39.0, 471.0): cluster 1\n", + "Example n. 3786 = (39.0, 470.0): cluster 1\n", + "Example n. 3787 = (39.0, 469.0): cluster 1\n", + "Example n. 3788 = (39.0, 468.0): cluster 1\n", + "Example n. 3789 = (39.0, 467.0): cluster 1\n", + "Example n. 3790 = (39.0, 466.0): cluster 1\n", + "Example n. 3791 = (39.0, 465.0): cluster 1\n", + "Example n. 3792 = (39.0, 464.0): cluster 1\n", + "Example n. 3793 = (39.0, 463.0): cluster 1\n", + "Example n. 3794 = (39.0, 462.0): cluster 1\n", + "Example n. 3795 = (39.0, 461.0): cluster 1\n", + "Example n. 3796 = (39.0, 460.0): cluster 1\n", + "Example n. 3797 = (39.0, 459.0): cluster 3\n", + "Example n. 3798 = (39.0, 458.0): cluster 3\n", + "Example n. 3799 = (39.0, 457.0): cluster 3\n", + "Example n. 3800 = (39.0, 456.0): cluster 3\n", + "Example n. 3801 = (39.0, 455.0): cluster 3\n", + "Example n. 3802 = (39.0, 318.0): cluster 3\n", + "Example n. 3803 = (39.0, 317.0): cluster 3\n", + "Example n. 3804 = (39.0, 316.0): cluster 3\n", + "Example n. 3805 = (39.0, 315.0): cluster 3\n", + "Example n. 3806 = (39.0, 314.0): cluster 3\n", + "Example n. 3807 = (39.0, 313.0): cluster 3\n", + "Example n. 3808 = (39.0, 312.0): cluster 3\n", + "Example n. 3809 = (39.0, 311.0): cluster 3\n", + "Example n. 3810 = (39.0, 310.0): cluster 3\n", + "Example n. 3811 = (39.0, 309.0): cluster 3\n", + "Example n. 3812 = (39.0, 308.0): cluster 3\n", + "Example n. 3813 = (39.0, 307.0): cluster 3\n", + "Example n. 3814 = (39.0, 306.0): cluster 3\n", + "Example n. 3815 = (39.0, 305.0): cluster 3\n", + "Example n. 3816 = (39.0, 304.0): cluster 3\n", + "Example n. 3817 = (39.0, 303.0): cluster 3\n", + "Example n. 3818 = (39.0, 302.0): cluster 3\n", + "Example n. 3819 = (39.0, 301.0): cluster 3\n", + "Example n. 3820 = (39.0, 300.0): cluster 3\n", + "Example n. 3821 = (39.0, 299.0): cluster 3\n", + "Example n. 3822 = (39.0, 298.0): cluster 3\n", + "Example n. 3823 = (39.0, 297.0): cluster 3\n", + "Example n. 3824 = (39.0, 296.0): cluster 3\n", + "Example n. 3825 = (39.0, 293.0): cluster 3\n", + "Example n. 3826 = (39.0, 292.0): cluster 3\n", + "Example n. 3827 = (39.0, 291.0): cluster 3\n", + "Example n. 3828 = (39.0, 290.0): cluster 3\n", + "Example n. 3829 = (39.0, 289.0): cluster 3\n", + "Example n. 3830 = (39.0, 288.0): cluster 1\n", + "Example n. 3831 = (39.0, 287.0): cluster 1\n", + "Example n. 3832 = (39.0, 286.0): cluster 1\n", + "Example n. 3833 = (39.0, 285.0): cluster 1\n", + "Example n. 3834 = (39.0, 272.0): cluster 1\n", + "Example n. 3835 = (39.0, 271.0): cluster 1\n", + "Example n. 3836 = (39.0, 270.0): cluster 1\n", + "Example n. 3837 = (39.0, 269.0): cluster 1\n", + "Example n. 3838 = (39.0, 268.0): cluster 1\n", + "Example n. 3839 = (39.0, 267.0): cluster 1\n", + "Example n. 3840 = (39.0, 266.0): cluster 1\n", + "Example n. 3841 = (39.0, 265.0): cluster 1\n", + "Example n. 3842 = (39.0, 264.0): cluster 1\n", + "Example n. 3843 = (39.0, 263.0): cluster 1\n", + "Example n. 3844 = (39.0, 262.0): cluster 1\n", + "Example n. 3845 = (39.0, 261.0): cluster 1\n", + "Example n. 3846 = (39.0, 260.0): cluster 1\n", + "Example n. 3847 = (39.0, 259.0): cluster 1\n", + "Example n. 3848 = (39.0, 258.0): cluster 1\n", + "Example n. 3849 = (39.0, 257.0): cluster 1\n", + "Example n. 3850 = (39.0, 256.0): cluster 1\n", + "Example n. 3851 = (39.0, 255.0): cluster 1\n", + "Example n. 3852 = (39.0, 254.0): cluster 1\n", + "Example n. 3853 = (39.0, 253.0): cluster 1\n", + "Example n. 3854 = (39.0, 246.0): cluster 1\n", + "Example n. 3855 = (39.0, 245.0): cluster 1\n", + "Example n. 3856 = (39.0, 244.0): cluster 1\n", + "Example n. 3857 = (39.0, 243.0): cluster 1\n", + "Example n. 3858 = (39.0, 242.0): cluster 1\n", + "Example n. 3859 = (39.0, 241.0): cluster 1\n", + "Example n. 3860 = (39.0, 240.0): cluster 1\n", + "Example n. 3861 = (39.0, 239.0): cluster 1\n", + "Example n. 3862 = (39.0, 238.0): cluster 1\n", + "Example n. 3863 = (39.0, 237.0): cluster 1\n", + "Example n. 3864 = (39.0, 236.0): cluster 1\n", + "Example n. 3865 = (39.0, 235.0): cluster 1\n", + "Example n. 3866 = (39.0, 234.0): cluster 1\n", + "Example n. 3867 = (39.0, 233.0): cluster 1\n", + "Example n. 3868 = (39.0, 232.0): cluster 1\n", + "Example n. 3869 = (39.0, 231.0): cluster 1\n", + "Example n. 3870 = (39.0, 230.0): cluster 1\n", + "Example n. 3871 = (39.0, 229.0): cluster 3\n", + "Example n. 3872 = (39.0, 228.0): cluster 3\n", + "Example n. 3873 = (39.0, 227.0): cluster 3\n", + "Example n. 3874 = (39.0, 226.0): cluster 3\n", + "Example n. 3875 = (39.0, 220.0): cluster 3\n", + "Example n. 3876 = (39.0, 214.0): cluster 3\n", + "Example n. 3877 = (39.0, 213.0): cluster 3\n", + "Example n. 3878 = (39.0, 212.0): cluster 3\n", + "Example n. 3879 = (39.0, 211.0): cluster 3\n", + "Example n. 3880 = (39.0, 210.0): cluster 3\n", + "Example n. 3881 = (39.0, 209.0): cluster 3\n", + "Example n. 3882 = (39.0, 208.0): cluster 3\n", + "Example n. 3883 = (39.0, 207.0): cluster 3\n", + "Example n. 3884 = (39.0, 206.0): cluster 3\n", + "Example n. 3885 = (39.0, 204.0): cluster 3\n", + "Example n. 3886 = (39.0, 203.0): cluster 3\n", + "Example n. 3887 = (39.0, 202.0): cluster 3\n", + "Example n. 3888 = (39.0, 201.0): cluster 3\n", + "Example n. 3889 = (39.0, 200.0): cluster 3\n", + "Example n. 3890 = (39.0, 197.0): cluster 3\n", + "Example n. 3891 = (39.0, 196.0): cluster 3\n", + "Example n. 3892 = (39.0, 195.0): cluster 3\n", + "Example n. 3893 = (39.0, 194.0): cluster 3\n", + "Example n. 3894 = (39.0, 193.0): cluster 3\n", + "Example n. 3895 = (39.0, 192.0): cluster 3\n", + "Example n. 3896 = (39.0, 191.0): cluster 3\n", + "Example n. 3897 = (39.0, 190.0): cluster 3\n", + "Example n. 3898 = (39.0, 187.0): cluster 3\n", + "Example n. 3899 = (39.0, 186.0): cluster 3\n", + "Example n. 3900 = (39.0, 185.0): cluster 3\n", + "Example n. 3901 = (39.0, 184.0): cluster 3\n", + "Example n. 3902 = (39.0, 183.0): cluster 3\n", + "Example n. 3903 = (39.0, 182.0): cluster 3\n", + "Example n. 3904 = (39.0, 181.0): cluster 1\n", + "Example n. 3905 = (39.0, 180.0): cluster 1\n", + "Example n. 3906 = (39.0, 179.0): cluster 1\n", + "Example n. 3907 = (39.0, 178.0): cluster 1\n", + "Example n. 3908 = (39.0, 177.0): cluster 1\n", + "Example n. 3909 = (39.0, 175.0): cluster 1\n", + "Example n. 3910 = (39.0, 174.0): cluster 1\n", + "Example n. 3911 = (39.0, 173.0): cluster 1\n", + "Example n. 3912 = (39.0, 172.0): cluster 1\n", + "Example n. 3913 = (39.0, 171.0): cluster 1\n", + "Example n. 3914 = (39.0, 170.0): cluster 1\n", + "Example n. 3915 = (39.0, 163.0): cluster 1\n", + "Example n. 3916 = (39.0, 162.0): cluster 1\n", + "Example n. 3917 = (39.0, 161.0): cluster 1\n", + "Example n. 3918 = (39.0, 160.0): cluster 1\n", + "Example n. 3919 = (39.0, 159.0): cluster 1\n", + "Example n. 3920 = (39.0, 158.0): cluster 1\n", + "Example n. 3921 = (39.0, 157.0): cluster 1\n", + "Example n. 3922 = (39.0, 156.0): cluster 1\n", + "Example n. 3923 = (39.0, 155.0): cluster 1\n", + "Example n. 3924 = (39.0, 154.0): cluster 1\n", + "Example n. 3925 = (39.0, 153.0): cluster 1\n", + "Example n. 3926 = (39.0, 152.0): cluster 1\n", + "Example n. 3927 = (39.0, 151.0): cluster 1\n", + "Example n. 3928 = (39.0, 150.0): cluster 1\n", + "Example n. 3929 = (39.0, 149.0): cluster 1\n", + "Example n. 3930 = (39.0, 148.0): cluster 1\n", + "Example n. 3931 = (39.0, 147.0): cluster 1\n", + "Example n. 3932 = (39.0, 146.0): cluster 1\n", + "Example n. 3933 = (39.0, 144.0): cluster 1\n", + "Example n. 3934 = (39.0, 143.0): cluster 1\n", + "Example n. 3935 = (39.0, 132.0): cluster 1\n", + "Example n. 3936 = (39.0, 131.0): cluster 1\n", + "Example n. 3937 = (39.0, 130.0): cluster 1\n", + "Example n. 3938 = (39.0, 129.0): cluster 1\n", + "Example n. 3939 = (39.0, 128.0): cluster 1\n", + "Example n. 3940 = (39.0, 127.0): cluster 1\n", + "Example n. 3941 = (39.0, 126.0): cluster 1\n", + "Example n. 3942 = (39.0, 114.0): cluster 1\n", + "Example n. 3943 = (39.0, 113.0): cluster 1\n", + "Example n. 3944 = (39.0, 112.0): cluster 1\n", + "Example n. 3945 = (39.0, 111.0): cluster 3\n", + "Example n. 3946 = (39.0, 110.0): cluster 3\n", + "Example n. 3947 = (39.0, 109.0): cluster 3\n", + "Example n. 3948 = (40.0, 492.0): cluster 3\n", + "Example n. 3949 = (40.0, 491.0): cluster 3\n", + "Example n. 3950 = (40.0, 490.0): cluster 3\n", + "Example n. 3951 = (40.0, 489.0): cluster 3\n", + "Example n. 3952 = (40.0, 488.0): cluster 3\n", + "Example n. 3953 = (40.0, 487.0): cluster 3\n", + "Example n. 3954 = (40.0, 486.0): cluster 3\n", + "Example n. 3955 = (40.0, 485.0): cluster 3\n", + "Example n. 3956 = (40.0, 484.0): cluster 3\n", + "Example n. 3957 = (40.0, 483.0): cluster 3\n", + "Example n. 3958 = (40.0, 482.0): cluster 3\n", + "Example n. 3959 = (40.0, 481.0): cluster 3\n", + "Example n. 3960 = (40.0, 480.0): cluster 3\n", + "Example n. 3961 = (40.0, 479.0): cluster 3\n", + "Example n. 3962 = (40.0, 478.0): cluster 3\n", + "Example n. 3963 = (40.0, 477.0): cluster 3\n", + "Example n. 3964 = (40.0, 476.0): cluster 3\n", + "Example n. 3965 = (40.0, 475.0): cluster 3\n", + "Example n. 3966 = (40.0, 474.0): cluster 3\n", + "Example n. 3967 = (40.0, 473.0): cluster 3\n", + "Example n. 3968 = (40.0, 472.0): cluster 3\n", + "Example n. 3969 = (40.0, 471.0): cluster 3\n", + "Example n. 3970 = (40.0, 470.0): cluster 3\n", + "Example n. 3971 = (40.0, 469.0): cluster 3\n", + "Example n. 3972 = (40.0, 468.0): cluster 3\n", + "Example n. 3973 = (40.0, 467.0): cluster 3\n", + "Example n. 3974 = (40.0, 466.0): cluster 3\n", + "Example n. 3975 = (40.0, 465.0): cluster 3\n", + "Example n. 3976 = (40.0, 464.0): cluster 3\n", + "Example n. 3977 = (40.0, 463.0): cluster 3\n", + "Example n. 3978 = (40.0, 462.0): cluster 3\n", + "Example n. 3979 = (40.0, 461.0): cluster 3\n", + "Example n. 3980 = (40.0, 460.0): cluster 3\n", + "Example n. 3981 = (40.0, 459.0): cluster 1\n", + "Example n. 3982 = (40.0, 458.0): cluster 1\n", + "Example n. 3983 = (40.0, 457.0): cluster 1\n", + "Example n. 3984 = (40.0, 456.0): cluster 1\n", + "Example n. 3985 = (40.0, 455.0): cluster 1\n", + "Example n. 3986 = (40.0, 410.0): cluster 1\n", + "Example n. 3987 = (40.0, 409.0): cluster 1\n", + "Example n. 3988 = (40.0, 408.0): cluster 1\n", + "Example n. 3989 = (40.0, 407.0): cluster 1\n", + "Example n. 3990 = (40.0, 395.0): cluster 1\n", + "Example n. 3991 = (40.0, 394.0): cluster 1\n", + "Example n. 3992 = (40.0, 393.0): cluster 1\n", + "Example n. 3993 = (40.0, 392.0): cluster 1\n", + "Example n. 3994 = (40.0, 345.0): cluster 1\n", + "Example n. 3995 = (40.0, 344.0): cluster 1\n", + "Example n. 3996 = (40.0, 343.0): cluster 1\n", + "Example n. 3997 = (40.0, 342.0): cluster 1\n", + "Example n. 3998 = (40.0, 319.0): cluster 1\n", + "Example n. 3999 = (40.0, 318.0): cluster 1\n", + "Example n. 4000 = (40.0, 317.0): cluster 1\n", + "Example n. 4001 = (40.0, 316.0): cluster 1\n", + "Example n. 4002 = (40.0, 315.0): cluster 1\n", + "Example n. 4003 = (40.0, 314.0): cluster 1\n", + "Example n. 4004 = (40.0, 313.0): cluster 1\n", + "Example n. 4005 = (40.0, 312.0): cluster 1\n", + "Example n. 4006 = (40.0, 311.0): cluster 1\n", + "Example n. 4007 = (40.0, 310.0): cluster 1\n", + "Example n. 4008 = (40.0, 309.0): cluster 1\n", + "Example n. 4009 = (40.0, 308.0): cluster 3\n", + "Example n. 4010 = (40.0, 307.0): cluster 3\n", + "Example n. 4011 = (40.0, 306.0): cluster 3\n", + "Example n. 4012 = (40.0, 305.0): cluster 3\n", + "Example n. 4013 = (40.0, 304.0): cluster 3\n", + "Example n. 4014 = (40.0, 303.0): cluster 3\n", + "Example n. 4015 = (40.0, 302.0): cluster 3\n", + "Example n. 4016 = (40.0, 300.0): cluster 3\n", + "Example n. 4017 = (40.0, 299.0): cluster 3\n", + "Example n. 4018 = (40.0, 298.0): cluster 3\n", + "Example n. 4019 = (40.0, 297.0): cluster 3\n", + "Example n. 4020 = (40.0, 296.0): cluster 3\n", + "Example n. 4021 = (40.0, 293.0): cluster 3\n", + "Example n. 4022 = (40.0, 292.0): cluster 3\n", + "Example n. 4023 = (40.0, 291.0): cluster 3\n", + "Example n. 4024 = (40.0, 290.0): cluster 3\n", + "Example n. 4025 = (40.0, 289.0): cluster 3\n", + "Example n. 4026 = (40.0, 288.0): cluster 3\n", + "Example n. 4027 = (40.0, 287.0): cluster 3\n", + "Example n. 4028 = (40.0, 286.0): cluster 3\n", + "Example n. 4029 = (40.0, 285.0): cluster 3\n", + "Example n. 4030 = (40.0, 284.0): cluster 3\n", + "Example n. 4031 = (40.0, 283.0): cluster 3\n", + "Example n. 4032 = (40.0, 272.0): cluster 3\n", + "Example n. 4033 = (40.0, 271.0): cluster 3\n", + "Example n. 4034 = (40.0, 270.0): cluster 3\n", + "Example n. 4035 = (40.0, 269.0): cluster 3\n", + "Example n. 4036 = (40.0, 268.0): cluster 3\n", + "Example n. 4037 = (40.0, 267.0): cluster 3\n", + "Example n. 4038 = (40.0, 266.0): cluster 3\n", + "Example n. 4039 = (40.0, 265.0): cluster 3\n", + "Example n. 4040 = (40.0, 264.0): cluster 3\n", + "Example n. 4041 = (40.0, 263.0): cluster 1\n", + "Example n. 4042 = (40.0, 260.0): cluster 1\n", + "Example n. 4043 = (40.0, 259.0): cluster 1\n", + "Example n. 4044 = (40.0, 258.0): cluster 1\n", + "Example n. 4045 = (40.0, 257.0): cluster 1\n", + "Example n. 4046 = (40.0, 256.0): cluster 1\n", + "Example n. 4047 = (40.0, 255.0): cluster 1\n", + "Example n. 4048 = (40.0, 254.0): cluster 1\n", + "Example n. 4049 = (40.0, 253.0): cluster 1\n", + "Example n. 4050 = (40.0, 249.0): cluster 1\n", + "Example n. 4051 = (40.0, 248.0): cluster 1\n", + "Example n. 4052 = (40.0, 247.0): cluster 1\n", + "Example n. 4053 = (40.0, 246.0): cluster 1\n", + "Example n. 4054 = (40.0, 245.0): cluster 1\n", + "Example n. 4055 = (40.0, 244.0): cluster 1\n", + "Example n. 4056 = (40.0, 243.0): cluster 1\n", + "Example n. 4057 = (40.0, 242.0): cluster 1\n", + "Example n. 4058 = (40.0, 241.0): cluster 1\n", + "Example n. 4059 = (40.0, 240.0): cluster 1\n", + "Example n. 4060 = (40.0, 239.0): cluster 1\n", + "Example n. 4061 = (40.0, 238.0): cluster 1\n", + "Example n. 4062 = (40.0, 237.0): cluster 1\n", + "Example n. 4063 = (40.0, 236.0): cluster 1\n", + "Example n. 4064 = (40.0, 235.0): cluster 1\n", + "Example n. 4065 = (40.0, 234.0): cluster 1\n", + "Example n. 4066 = (40.0, 233.0): cluster 1\n", + "Example n. 4067 = (40.0, 232.0): cluster 1\n", + "Example n. 4068 = (40.0, 231.0): cluster 1\n", + "Example n. 4069 = (40.0, 230.0): cluster 3\n", + "Example n. 4070 = (40.0, 229.0): cluster 3\n", + "Example n. 4071 = (40.0, 228.0): cluster 3\n", + "Example n. 4072 = (40.0, 227.0): cluster 3\n", + "Example n. 4073 = (40.0, 226.0): cluster 3\n", + "Example n. 4074 = (40.0, 215.0): cluster 3\n", + "Example n. 4075 = (40.0, 214.0): cluster 3\n", + "Example n. 4076 = (40.0, 213.0): cluster 3\n", + "Example n. 4077 = (40.0, 212.0): cluster 3\n", + "Example n. 4078 = (40.0, 211.0): cluster 3\n", + "Example n. 4079 = (40.0, 210.0): cluster 3\n", + "Example n. 4080 = (40.0, 209.0): cluster 3\n", + "Example n. 4081 = (40.0, 208.0): cluster 3\n", + "Example n. 4082 = (40.0, 207.0): cluster 3\n", + "Example n. 4083 = (40.0, 206.0): cluster 3\n", + "Example n. 4084 = (40.0, 205.0): cluster 3\n", + "Example n. 4085 = (40.0, 204.0): cluster 3\n", + "Example n. 4086 = (40.0, 203.0): cluster 3\n", + "Example n. 4087 = (40.0, 202.0): cluster 3\n", + "Example n. 4088 = (40.0, 201.0): cluster 3\n", + "Example n. 4089 = (40.0, 200.0): cluster 3\n", + "Example n. 4090 = (40.0, 199.0): cluster 3\n", + "Example n. 4091 = (40.0, 197.0): cluster 3\n", + "Example n. 4092 = (40.0, 196.0): cluster 3\n", + "Example n. 4093 = (40.0, 195.0): cluster 3\n", + "Example n. 4094 = (40.0, 194.0): cluster 3\n", + "Example n. 4095 = (40.0, 193.0): cluster 3\n", + "Example n. 4096 = (40.0, 192.0): cluster 3\n", + "Example n. 4097 = (40.0, 187.0): cluster 3\n", + "Example n. 4098 = (40.0, 186.0): cluster 3\n", + "Example n. 4099 = (40.0, 185.0): cluster 3\n", + "Example n. 4100 = (40.0, 184.0): cluster 3\n", + "Example n. 4101 = (40.0, 183.0): cluster 1\n", + "Example n. 4102 = (40.0, 182.0): cluster 1\n", + "Example n. 4103 = (40.0, 181.0): cluster 1\n", + "Example n. 4104 = (40.0, 180.0): cluster 1\n", + "Example n. 4105 = (40.0, 179.0): cluster 1\n", + "Example n. 4106 = (40.0, 178.0): cluster 1\n", + "Example n. 4107 = (40.0, 175.0): cluster 1\n", + "Example n. 4108 = (40.0, 174.0): cluster 1\n", + "Example n. 4109 = (40.0, 173.0): cluster 1\n", + "Example n. 4110 = (40.0, 172.0): cluster 1\n", + "Example n. 4111 = (40.0, 171.0): cluster 1\n", + "Example n. 4112 = (40.0, 170.0): cluster 1\n", + "Example n. 4113 = (40.0, 161.0): cluster 1\n", + "Example n. 4114 = (40.0, 160.0): cluster 1\n", + "Example n. 4115 = (40.0, 159.0): cluster 1\n", + "Example n. 4116 = (40.0, 158.0): cluster 1\n", + "Example n. 4117 = (40.0, 157.0): cluster 1\n", + "Example n. 4118 = (40.0, 156.0): cluster 1\n", + "Example n. 4119 = (40.0, 155.0): cluster 1\n", + "Example n. 4120 = (40.0, 154.0): cluster 1\n", + "Example n. 4121 = (40.0, 153.0): cluster 1\n", + "Example n. 4122 = (40.0, 152.0): cluster 1\n", + "Example n. 4123 = (40.0, 151.0): cluster 1\n", + "Example n. 4124 = (40.0, 150.0): cluster 1\n", + "Example n. 4125 = (40.0, 149.0): cluster 1\n", + "Example n. 4126 = (40.0, 148.0): cluster 1\n", + "Example n. 4127 = (40.0, 147.0): cluster 1\n", + "Example n. 4128 = (40.0, 146.0): cluster 1\n", + "Example n. 4129 = (40.0, 145.0): cluster 3\n", + "Example n. 4130 = (40.0, 144.0): cluster 3\n", + "Example n. 4131 = (40.0, 143.0): cluster 3\n", + "Example n. 4132 = (40.0, 142.0): cluster 3\n", + "Example n. 4133 = (40.0, 130.0): cluster 3\n", + "Example n. 4134 = (40.0, 129.0): cluster 3\n", + "Example n. 4135 = (40.0, 128.0): cluster 3\n", + "Example n. 4136 = (40.0, 113.0): cluster 3\n", + "Example n. 4137 = (40.0, 112.0): cluster 3\n", + "Example n. 4138 = (40.0, 111.0): cluster 3\n", + "Example n. 4139 = (40.0, 110.0): cluster 3\n", + "Example n. 4140 = (40.0, 109.0): cluster 3\n", + "Example n. 4141 = (41.0, 491.0): cluster 3\n", + "Example n. 4142 = (41.0, 490.0): cluster 3\n", + "Example n. 4143 = (41.0, 489.0): cluster 3\n", + "Example n. 4144 = (41.0, 488.0): cluster 3\n", + "Example n. 4145 = (41.0, 487.0): cluster 3\n", + "Example n. 4146 = (41.0, 486.0): cluster 3\n", + "Example n. 4147 = (41.0, 485.0): cluster 3\n", + "Example n. 4148 = (41.0, 484.0): cluster 3\n", + "Example n. 4149 = (41.0, 483.0): cluster 3\n", + "Example n. 4150 = (41.0, 482.0): cluster 3\n", + "Example n. 4151 = (41.0, 481.0): cluster 3\n", + "Example n. 4152 = (41.0, 480.0): cluster 3\n", + "Example n. 4153 = (41.0, 479.0): cluster 3\n", + "Example n. 4154 = (41.0, 478.0): cluster 3\n", + "Example n. 4155 = (41.0, 477.0): cluster 3\n", + "Example n. 4156 = (41.0, 476.0): cluster 3\n", + "Example n. 4157 = (41.0, 475.0): cluster 3\n", + "Example n. 4158 = (41.0, 474.0): cluster 3\n", + "Example n. 4159 = (41.0, 473.0): cluster 3\n", + "Example n. 4160 = (41.0, 472.0): cluster 3\n", + "Example n. 4161 = (41.0, 471.0): cluster 3\n", + "Example n. 4162 = (41.0, 470.0): cluster 3\n", + "Example n. 4163 = (41.0, 469.0): cluster 3\n", + "Example n. 4164 = (41.0, 468.0): cluster 3\n", + "Example n. 4165 = (41.0, 467.0): cluster 3\n", + "Example n. 4166 = (41.0, 466.0): cluster 3\n", + "Example n. 4167 = (41.0, 465.0): cluster 3\n", + "Example n. 4168 = (41.0, 464.0): cluster 3\n", + "Example n. 4169 = (41.0, 463.0): cluster 3\n", + "Example n. 4170 = (41.0, 462.0): cluster 3\n", + "Example n. 4171 = (41.0, 461.0): cluster 3\n", + "Example n. 4172 = (41.0, 460.0): cluster 3\n", + "Example n. 4173 = (41.0, 459.0): cluster 3\n", + "Example n. 4174 = (41.0, 458.0): cluster 3\n", + "Example n. 4175 = (41.0, 457.0): cluster 3\n", + "Example n. 4176 = (41.0, 456.0): cluster 3\n", + "Example n. 4177 = (41.0, 455.0): cluster 3\n", + "Example n. 4178 = (41.0, 454.0): cluster 3\n", + "Example n. 4179 = (41.0, 433.0): cluster 3\n", + "Example n. 4180 = (41.0, 432.0): cluster 3\n", + "Example n. 4181 = (41.0, 431.0): cluster 3\n", + "Example n. 4182 = (41.0, 430.0): cluster 3\n", + "Example n. 4183 = (41.0, 411.0): cluster 3\n", + "Example n. 4184 = (41.0, 410.0): cluster 3\n", + "Example n. 4185 = (41.0, 409.0): cluster 3\n", + "Example n. 4186 = (41.0, 408.0): cluster 3\n", + "Example n. 4187 = (41.0, 407.0): cluster 3\n", + "Example n. 4188 = (41.0, 396.0): cluster 3\n", + "Example n. 4189 = (41.0, 395.0): cluster 3\n", + "Example n. 4190 = (41.0, 394.0): cluster 3\n", + "Example n. 4191 = (41.0, 393.0): cluster 3\n", + "Example n. 4192 = (41.0, 392.0): cluster 3\n", + "Example n. 4193 = (41.0, 346.0): cluster 3\n", + "Example n. 4194 = (41.0, 345.0): cluster 3\n", + "Example n. 4195 = (41.0, 344.0): cluster 3\n", + "Example n. 4196 = (41.0, 343.0): cluster 3\n", + "Example n. 4197 = (41.0, 342.0): cluster 3\n", + "Example n. 4198 = (41.0, 341.0): cluster 3\n", + "Example n. 4199 = (41.0, 319.0): cluster 3\n", + "Example n. 4200 = (41.0, 318.0): cluster 3\n", + "Example n. 4201 = (41.0, 317.0): cluster 3\n", + "Example n. 4202 = (41.0, 316.0): cluster 3\n", + "Example n. 4203 = (41.0, 315.0): cluster 3\n", + "Example n. 4204 = (41.0, 314.0): cluster 3\n", + "Example n. 4205 = (41.0, 313.0): cluster 3\n", + "Example n. 4206 = (41.0, 312.0): cluster 3\n", + "Example n. 4207 = (41.0, 311.0): cluster 3\n", + "Example n. 4208 = (41.0, 310.0): cluster 3\n", + "Example n. 4209 = (41.0, 309.0): cluster 3\n", + "Example n. 4210 = (41.0, 308.0): cluster 3\n", + "Example n. 4211 = (41.0, 307.0): cluster 3\n", + "Example n. 4212 = (41.0, 306.0): cluster 3\n", + "Example n. 4213 = (41.0, 305.0): cluster 3\n", + "Example n. 4214 = (41.0, 304.0): cluster 3\n", + "Example n. 4215 = (41.0, 303.0): cluster 3\n", + "Example n. 4216 = (41.0, 302.0): cluster 3\n", + "Example n. 4217 = (41.0, 300.0): cluster 3\n", + "Example n. 4218 = (41.0, 299.0): cluster 3\n", + "Example n. 4219 = (41.0, 298.0): cluster 3\n", + "Example n. 4220 = (41.0, 297.0): cluster 3\n", + "Example n. 4221 = (41.0, 296.0): cluster 3\n", + "Example n. 4222 = (41.0, 292.0): cluster 3\n", + "Example n. 4223 = (41.0, 291.0): cluster 3\n", + "Example n. 4224 = (41.0, 290.0): cluster 3\n", + "Example n. 4225 = (41.0, 289.0): cluster 3\n", + "Example n. 4226 = (41.0, 288.0): cluster 3\n", + "Example n. 4227 = (41.0, 287.0): cluster 3\n", + "Example n. 4228 = (41.0, 286.0): cluster 3\n", + "Example n. 4229 = (41.0, 285.0): cluster 3\n", + "Example n. 4230 = (41.0, 284.0): cluster 3\n", + "Example n. 4231 = (41.0, 283.0): cluster 3\n", + "Example n. 4232 = (41.0, 282.0): cluster 3\n", + "Example n. 4233 = (41.0, 278.0): cluster 3\n", + "Example n. 4234 = (41.0, 277.0): cluster 3\n", + "Example n. 4235 = (41.0, 276.0): cluster 3\n", + "Example n. 4236 = (41.0, 275.0): cluster 3\n", + "Example n. 4237 = (41.0, 272.0): cluster 3\n", + "Example n. 4238 = (41.0, 271.0): cluster 3\n", + "Example n. 4239 = (41.0, 270.0): cluster 3\n", + "Example n. 4240 = (41.0, 269.0): cluster 3\n", + "Example n. 4241 = (41.0, 268.0): cluster 3\n", + "Example n. 4242 = (41.0, 267.0): cluster 3\n", + "Example n. 4243 = (41.0, 266.0): cluster 3\n", + "Example n. 4244 = (41.0, 265.0): cluster 3\n", + "Example n. 4245 = (41.0, 264.0): cluster 3\n", + "Example n. 4246 = (41.0, 263.0): cluster 3\n", + "Example n. 4247 = (41.0, 260.0): cluster 3\n", + "Example n. 4248 = (41.0, 259.0): cluster 3\n", + "Example n. 4249 = (41.0, 258.0): cluster 3\n", + "Example n. 4250 = (41.0, 257.0): cluster 3\n", + "Example n. 4251 = (41.0, 256.0): cluster 3\n", + "Example n. 4252 = (41.0, 255.0): cluster 3\n", + "Example n. 4253 = (41.0, 254.0): cluster 3\n", + "Example n. 4254 = (41.0, 253.0): cluster 3\n", + "Example n. 4255 = (41.0, 250.0): cluster 3\n", + "Example n. 4256 = (41.0, 249.0): cluster 3\n", + "Example n. 4257 = (41.0, 248.0): cluster 3\n", + "Example n. 4258 = (41.0, 247.0): cluster 3\n", + "Example n. 4259 = (41.0, 246.0): cluster 3\n", + "Example n. 4260 = (41.0, 245.0): cluster 3\n", + "Example n. 4261 = (41.0, 244.0): cluster 3\n", + "Example n. 4262 = (41.0, 243.0): cluster 3\n", + "Example n. 4263 = (41.0, 242.0): cluster 3\n", + "Example n. 4264 = (41.0, 241.0): cluster 3\n", + "Example n. 4265 = (41.0, 240.0): cluster 3\n", + "Example n. 4266 = (41.0, 239.0): cluster 3\n", + "Example n. 4267 = (41.0, 238.0): cluster 3\n", + "Example n. 4268 = (41.0, 237.0): cluster 3\n", + "Example n. 4269 = (41.0, 236.0): cluster 3\n", + "Example n. 4270 = (41.0, 235.0): cluster 3\n", + "Example n. 4271 = (41.0, 234.0): cluster 3\n", + "Example n. 4272 = (41.0, 233.0): cluster 3\n", + "Example n. 4273 = (41.0, 232.0): cluster 3\n", + "Example n. 4274 = (41.0, 231.0): cluster 3\n", + "Example n. 4275 = (41.0, 230.0): cluster 3\n", + "Example n. 4276 = (41.0, 229.0): cluster 3\n", + "Example n. 4277 = (41.0, 228.0): cluster 3\n", + "Example n. 4278 = (41.0, 227.0): cluster 3\n", + "Example n. 4279 = (41.0, 226.0): cluster 3\n", + "Example n. 4280 = (41.0, 215.0): cluster 3\n", + "Example n. 4281 = (41.0, 214.0): cluster 3\n", + "Example n. 4282 = (41.0, 213.0): cluster 3\n", + "Example n. 4283 = (41.0, 212.0): cluster 3\n", + "Example n. 4284 = (41.0, 211.0): cluster 3\n", + "Example n. 4285 = (41.0, 210.0): cluster 3\n", + "Example n. 4286 = (41.0, 209.0): cluster 3\n", + "Example n. 4287 = (41.0, 208.0): cluster 3\n", + "Example n. 4288 = (41.0, 207.0): cluster 3\n", + "Example n. 4289 = (41.0, 206.0): cluster 3\n", + "Example n. 4290 = (41.0, 205.0): cluster 3\n", + "Example n. 4291 = (41.0, 204.0): cluster 3\n", + "Example n. 4292 = (41.0, 203.0): cluster 3\n", + "Example n. 4293 = (41.0, 202.0): cluster 3\n", + "Example n. 4294 = (41.0, 201.0): cluster 3\n", + "Example n. 4295 = (41.0, 200.0): cluster 3\n", + "Example n. 4296 = (41.0, 199.0): cluster 3\n", + "Example n. 4297 = (41.0, 196.0): cluster 3\n", + "Example n. 4298 = (41.0, 195.0): cluster 3\n", + "Example n. 4299 = (41.0, 194.0): cluster 3\n", + "Example n. 4300 = (41.0, 193.0): cluster 3\n", + "Example n. 4301 = (41.0, 188.0): cluster 3\n", + "Example n. 4302 = (41.0, 187.0): cluster 3\n", + "Example n. 4303 = (41.0, 186.0): cluster 3\n", + "Example n. 4304 = (41.0, 185.0): cluster 3\n", + "Example n. 4305 = (41.0, 184.0): cluster 3\n", + "Example n. 4306 = (41.0, 183.0): cluster 3\n", + "Example n. 4307 = (41.0, 182.0): cluster 3\n", + "Example n. 4308 = (41.0, 181.0): cluster 3\n", + "Example n. 4309 = (41.0, 180.0): cluster 3\n", + "Example n. 4310 = (41.0, 175.0): cluster 3\n", + "Example n. 4311 = (41.0, 174.0): cluster 3\n", + "Example n. 4312 = (41.0, 173.0): cluster 3\n", + "Example n. 4313 = (41.0, 172.0): cluster 3\n", + "Example n. 4314 = (41.0, 171.0): cluster 3\n", + "Example n. 4315 = (41.0, 161.0): cluster 3\n", + "Example n. 4316 = (41.0, 160.0): cluster 3\n", + "Example n. 4317 = (41.0, 159.0): cluster 3\n", + "Example n. 4318 = (41.0, 158.0): cluster 3\n", + "Example n. 4319 = (41.0, 157.0): cluster 3\n", + "Example n. 4320 = (41.0, 156.0): cluster 3\n", + "Example n. 4321 = (41.0, 155.0): cluster 3\n", + "Example n. 4322 = (41.0, 154.0): cluster 3\n", + "Example n. 4323 = (41.0, 153.0): cluster 3\n", + "Example n. 4324 = (41.0, 152.0): cluster 3\n", + "Example n. 4325 = (41.0, 151.0): cluster 3\n", + "Example n. 4326 = (41.0, 150.0): cluster 3\n", + "Example n. 4327 = (41.0, 149.0): cluster 3\n", + "Example n. 4328 = (41.0, 148.0): cluster 3\n", + "Example n. 4329 = (41.0, 147.0): cluster 3\n", + "Example n. 4330 = (41.0, 146.0): cluster 3\n", + "Example n. 4331 = (41.0, 145.0): cluster 3\n", + "Example n. 4332 = (41.0, 144.0): cluster 3\n", + "Example n. 4333 = (41.0, 143.0): cluster 3\n", + "Example n. 4334 = (41.0, 142.0): cluster 3\n", + "Example n. 4335 = (41.0, 141.0): cluster 3\n", + "Example n. 4336 = (41.0, 113.0): cluster 3\n", + "Example n. 4337 = (41.0, 112.0): cluster 3\n", + "Example n. 4338 = (41.0, 111.0): cluster 3\n", + "Example n. 4339 = (41.0, 110.0): cluster 3\n", + "Example n. 4340 = (41.0, 109.0): cluster 3\n", + "Example n. 4341 = (42.0, 491.0): cluster 2\n", + "Example n. 4342 = (42.0, 490.0): cluster 3\n", + "Example n. 4343 = (42.0, 489.0): cluster 3\n", + "Example n. 4344 = (42.0, 488.0): cluster 3\n", + "Example n. 4345 = (42.0, 487.0): cluster 3\n", + "Example n. 4346 = (42.0, 486.0): cluster 3\n", + "Example n. 4347 = (42.0, 485.0): cluster 3\n", + "Example n. 4348 = (42.0, 484.0): cluster 3\n", + "Example n. 4349 = (42.0, 483.0): cluster 3\n", + "Example n. 4350 = (42.0, 482.0): cluster 3\n", + "Example n. 4351 = (42.0, 481.0): cluster 3\n", + "Example n. 4352 = (42.0, 480.0): cluster 3\n", + "Example n. 4353 = (42.0, 479.0): cluster 3\n", + "Example n. 4354 = (42.0, 478.0): cluster 3\n", + "Example n. 4355 = (42.0, 477.0): cluster 3\n", + "Example n. 4356 = (42.0, 476.0): cluster 3\n", + "Example n. 4357 = (42.0, 475.0): cluster 3\n", + "Example n. 4358 = (42.0, 474.0): cluster 3\n", + "Example n. 4359 = (42.0, 473.0): cluster 3\n", + "Example n. 4360 = (42.0, 472.0): cluster 3\n", + "Example n. 4361 = (42.0, 471.0): cluster 3\n", + "Example n. 4362 = (42.0, 470.0): cluster 3\n", + "Example n. 4363 = (42.0, 469.0): cluster 3\n", + "Example n. 4364 = (42.0, 468.0): cluster 3\n", + "Example n. 4365 = (42.0, 467.0): cluster 3\n", + "Example n. 4366 = (42.0, 466.0): cluster 3\n", + "Example n. 4367 = (42.0, 465.0): cluster 3\n", + "Example n. 4368 = (42.0, 464.0): cluster 3\n", + "Example n. 4369 = (42.0, 463.0): cluster 3\n", + "Example n. 4370 = (42.0, 462.0): cluster 3\n", + "Example n. 4371 = (42.0, 461.0): cluster 3\n", + "Example n. 4372 = (42.0, 460.0): cluster 3\n", + "Example n. 4373 = (42.0, 459.0): cluster 3\n", + "Example n. 4374 = (42.0, 458.0): cluster 3\n", + "Example n. 4375 = (42.0, 457.0): cluster 3\n", + "Example n. 4376 = (42.0, 456.0): cluster 3\n", + "Example n. 4377 = (42.0, 455.0): cluster 2\n", + "Example n. 4378 = (42.0, 454.0): cluster 2\n", + "Example n. 4379 = (42.0, 433.0): cluster 3\n", + "Example n. 4380 = (42.0, 432.0): cluster 3\n", + "Example n. 4381 = (42.0, 431.0): cluster 3\n", + "Example n. 4382 = (42.0, 430.0): cluster 3\n", + "Example n. 4383 = (42.0, 429.0): cluster 3\n", + "Example n. 4384 = (42.0, 411.0): cluster 3\n", + "Example n. 4385 = (42.0, 410.0): cluster 3\n", + "Example n. 4386 = (42.0, 409.0): cluster 3\n", + "Example n. 4387 = (42.0, 408.0): cluster 3\n", + "Example n. 4388 = (42.0, 407.0): cluster 3\n", + "Example n. 4389 = (42.0, 406.0): cluster 3\n", + "Example n. 4390 = (42.0, 396.0): cluster 3\n", + "Example n. 4391 = (42.0, 395.0): cluster 3\n", + "Example n. 4392 = (42.0, 394.0): cluster 3\n", + "Example n. 4393 = (42.0, 393.0): cluster 3\n", + "Example n. 4394 = (42.0, 392.0): cluster 3\n", + "Example n. 4395 = (42.0, 391.0): cluster 3\n", + "Example n. 4396 = (42.0, 346.0): cluster 3\n", + "Example n. 4397 = (42.0, 345.0): cluster 3\n", + "Example n. 4398 = (42.0, 344.0): cluster 3\n", + "Example n. 4399 = (42.0, 343.0): cluster 3\n", + "Example n. 4400 = (42.0, 342.0): cluster 3\n", + "Example n. 4401 = (42.0, 341.0): cluster 3\n", + "Example n. 4402 = (42.0, 319.0): cluster 3\n", + "Example n. 4403 = (42.0, 318.0): cluster 3\n", + "Example n. 4404 = (42.0, 317.0): cluster 3\n", + "Example n. 4405 = (42.0, 316.0): cluster 3\n", + "Example n. 4406 = (42.0, 315.0): cluster 3\n", + "Example n. 4407 = (42.0, 314.0): cluster 3\n", + "Example n. 4408 = (42.0, 313.0): cluster 3\n", + "Example n. 4409 = (42.0, 312.0): cluster 3\n", + "Example n. 4410 = (42.0, 311.0): cluster 3\n", + "Example n. 4411 = (42.0, 310.0): cluster 3\n", + "Example n. 4412 = (42.0, 309.0): cluster 3\n", + "Example n. 4413 = (42.0, 308.0): cluster 2\n", + "Example n. 4414 = (42.0, 307.0): cluster 2\n", + "Example n. 4415 = (42.0, 306.0): cluster 2\n", + "Example n. 4416 = (42.0, 305.0): cluster 2\n", + "Example n. 4417 = (42.0, 304.0): cluster 3\n", + "Example n. 4418 = (42.0, 303.0): cluster 3\n", + "Example n. 4419 = (42.0, 299.0): cluster 3\n", + "Example n. 4420 = (42.0, 298.0): cluster 3\n", + "Example n. 4421 = (42.0, 297.0): cluster 3\n", + "Example n. 4422 = (42.0, 296.0): cluster 3\n", + "Example n. 4423 = (42.0, 291.0): cluster 3\n", + "Example n. 4424 = (42.0, 290.0): cluster 3\n", + "Example n. 4425 = (42.0, 289.0): cluster 3\n", + "Example n. 4426 = (42.0, 288.0): cluster 3\n", + "Example n. 4427 = (42.0, 287.0): cluster 3\n", + "Example n. 4428 = (42.0, 286.0): cluster 3\n", + "Example n. 4429 = (42.0, 285.0): cluster 3\n", + "Example n. 4430 = (42.0, 284.0): cluster 3\n", + "Example n. 4431 = (42.0, 283.0): cluster 3\n", + "Example n. 4432 = (42.0, 282.0): cluster 3\n", + "Example n. 4433 = (42.0, 279.0): cluster 3\n", + "Example n. 4434 = (42.0, 278.0): cluster 3\n", + "Example n. 4435 = (42.0, 277.0): cluster 3\n", + "Example n. 4436 = (42.0, 276.0): cluster 3\n", + "Example n. 4437 = (42.0, 275.0): cluster 3\n", + "Example n. 4438 = (42.0, 274.0): cluster 3\n", + "Example n. 4439 = (42.0, 271.0): cluster 3\n", + "Example n. 4440 = (42.0, 270.0): cluster 3\n", + "Example n. 4441 = (42.0, 269.0): cluster 3\n", + "Example n. 4442 = (42.0, 268.0): cluster 3\n", + "Example n. 4443 = (42.0, 267.0): cluster 3\n", + "Example n. 4444 = (42.0, 266.0): cluster 3\n", + "Example n. 4445 = (42.0, 265.0): cluster 3\n", + "Example n. 4446 = (42.0, 264.0): cluster 3\n", + "Example n. 4447 = (42.0, 263.0): cluster 3\n", + "Example n. 4448 = (42.0, 260.0): cluster 3\n", + "Example n. 4449 = (42.0, 259.0): cluster 2\n", + "Example n. 4450 = (42.0, 258.0): cluster 2\n", + "Example n. 4451 = (42.0, 257.0): cluster 2\n", + "Example n. 4452 = (42.0, 256.0): cluster 2\n", + "Example n. 4453 = (42.0, 255.0): cluster 2\n", + "Example n. 4454 = (42.0, 254.0): cluster 2\n", + "Example n. 4455 = (42.0, 250.0): cluster 2\n", + "Example n. 4456 = (42.0, 249.0): cluster 2\n", + "Example n. 4457 = (42.0, 248.0): cluster 2\n", + "Example n. 4458 = (42.0, 247.0): cluster 3\n", + "Example n. 4459 = (42.0, 246.0): cluster 3\n", + "Example n. 4460 = (42.0, 245.0): cluster 3\n", + "Example n. 4461 = (42.0, 244.0): cluster 3\n", + "Example n. 4462 = (42.0, 243.0): cluster 3\n", + "Example n. 4463 = (42.0, 242.0): cluster 3\n", + "Example n. 4464 = (42.0, 240.0): cluster 3\n", + "Example n. 4465 = (42.0, 238.0): cluster 3\n", + "Example n. 4466 = (42.0, 237.0): cluster 3\n", + "Example n. 4467 = (42.0, 236.0): cluster 3\n", + "Example n. 4468 = (42.0, 235.0): cluster 3\n", + "Example n. 4469 = (42.0, 234.0): cluster 3\n", + "Example n. 4470 = (42.0, 233.0): cluster 3\n", + "Example n. 4471 = (42.0, 232.0): cluster 3\n", + "Example n. 4472 = (42.0, 231.0): cluster 3\n", + "Example n. 4473 = (42.0, 230.0): cluster 3\n", + "Example n. 4474 = (42.0, 229.0): cluster 3\n", + "Example n. 4475 = (42.0, 228.0): cluster 3\n", + "Example n. 4476 = (42.0, 227.0): cluster 3\n", + "Example n. 4477 = (42.0, 226.0): cluster 3\n", + "Example n. 4478 = (42.0, 225.0): cluster 3\n", + "Example n. 4479 = (42.0, 215.0): cluster 3\n", + "Example n. 4480 = (42.0, 214.0): cluster 3\n", + "Example n. 4481 = (42.0, 213.0): cluster 3\n", + "Example n. 4482 = (42.0, 212.0): cluster 3\n", + "Example n. 4483 = (42.0, 211.0): cluster 3\n", + "Example n. 4484 = (42.0, 210.0): cluster 3\n", + "Example n. 4485 = (42.0, 209.0): cluster 2\n", + "Example n. 4486 = (42.0, 208.0): cluster 2\n", + "Example n. 4487 = (42.0, 207.0): cluster 2\n", + "Example n. 4488 = (42.0, 206.0): cluster 2\n", + "Example n. 4489 = (42.0, 205.0): cluster 2\n", + "Example n. 4490 = (42.0, 204.0): cluster 2\n", + "Example n. 4491 = (42.0, 203.0): cluster 2\n", + "Example n. 4492 = (42.0, 202.0): cluster 2\n", + "Example n. 4493 = (42.0, 201.0): cluster 2\n", + "Example n. 4494 = (42.0, 200.0): cluster 2\n", + "Example n. 4495 = (42.0, 199.0): cluster 3\n", + "Example n. 4496 = (42.0, 189.0): cluster 3\n", + "Example n. 4497 = (42.0, 188.0): cluster 3\n", + "Example n. 4498 = (42.0, 187.0): cluster 3\n", + "Example n. 4499 = (42.0, 186.0): cluster 3\n", + "Example n. 4500 = (42.0, 185.0): cluster 3\n", + "Example n. 4501 = (42.0, 184.0): cluster 3\n", + "Example n. 4502 = (42.0, 183.0): cluster 3\n", + "Example n. 4503 = (42.0, 182.0): cluster 3\n", + "Example n. 4504 = (42.0, 181.0): cluster 3\n", + "Example n. 4505 = (42.0, 174.0): cluster 3\n", + "Example n. 4506 = (42.0, 173.0): cluster 3\n", + "Example n. 4507 = (42.0, 172.0): cluster 3\n", + "Example n. 4508 = (42.0, 171.0): cluster 3\n", + "Example n. 4509 = (42.0, 170.0): cluster 3\n", + "Example n. 4510 = (42.0, 161.0): cluster 3\n", + "Example n. 4511 = (42.0, 160.0): cluster 3\n", + "Example n. 4512 = (42.0, 159.0): cluster 3\n", + "Example n. 4513 = (42.0, 158.0): cluster 3\n", + "Example n. 4514 = (42.0, 157.0): cluster 3\n", + "Example n. 4515 = (42.0, 156.0): cluster 3\n", + "Example n. 4516 = (42.0, 155.0): cluster 3\n", + "Example n. 4517 = (42.0, 154.0): cluster 3\n", + "Example n. 4518 = (42.0, 153.0): cluster 3\n", + "Example n. 4519 = (42.0, 152.0): cluster 3\n", + "Example n. 4520 = (42.0, 151.0): cluster 3\n", + "Example n. 4521 = (42.0, 150.0): cluster 2\n", + "Example n. 4522 = (42.0, 149.0): cluster 2\n", + "Example n. 4523 = (42.0, 146.0): cluster 2\n", + "Example n. 4524 = (42.0, 145.0): cluster 2\n", + "Example n. 4525 = (42.0, 144.0): cluster 2\n", + "Example n. 4526 = (42.0, 143.0): cluster 2\n", + "Example n. 4527 = (42.0, 142.0): cluster 2\n", + "Example n. 4528 = (42.0, 141.0): cluster 2\n", + "Example n. 4529 = (42.0, 130.0): cluster 2\n", + "Example n. 4530 = (42.0, 129.0): cluster 2\n", + "Example n. 4531 = (42.0, 128.0): cluster 2\n", + "Example n. 4532 = (42.0, 127.0): cluster 2\n", + "Example n. 4533 = (42.0, 111.0): cluster 3\n", + "Example n. 4534 = (43.0, 494.0): cluster 3\n", + "Example n. 4535 = (43.0, 493.0): cluster 3\n", + "Example n. 4536 = (43.0, 492.0): cluster 3\n", + "Example n. 4537 = (43.0, 491.0): cluster 3\n", + "Example n. 4538 = (43.0, 490.0): cluster 3\n", + "Example n. 4539 = (43.0, 489.0): cluster 3\n", + "Example n. 4540 = (43.0, 488.0): cluster 3\n", + "Example n. 4541 = (43.0, 487.0): cluster 3\n", + "Example n. 4542 = (43.0, 486.0): cluster 3\n", + "Example n. 4543 = (43.0, 485.0): cluster 3\n", + "Example n. 4544 = (43.0, 484.0): cluster 3\n", + "Example n. 4545 = (43.0, 483.0): cluster 3\n", + "Example n. 4546 = (43.0, 482.0): cluster 3\n", + "Example n. 4547 = (43.0, 481.0): cluster 3\n", + "Example n. 4548 = (43.0, 480.0): cluster 3\n", + "Example n. 4549 = (43.0, 479.0): cluster 3\n", + "Example n. 4550 = (43.0, 478.0): cluster 3\n", + "Example n. 4551 = (43.0, 477.0): cluster 3\n", + "Example n. 4552 = (43.0, 476.0): cluster 3\n", + "Example n. 4553 = (43.0, 475.0): cluster 3\n", + "Example n. 4554 = (43.0, 474.0): cluster 3\n", + "Example n. 4555 = (43.0, 473.0): cluster 3\n", + "Example n. 4556 = (43.0, 472.0): cluster 3\n", + "Example n. 4557 = (43.0, 471.0): cluster 2\n", + "Example n. 4558 = (43.0, 470.0): cluster 2\n", + "Example n. 4559 = (43.0, 469.0): cluster 2\n", + "Example n. 4560 = (43.0, 468.0): cluster 2\n", + "Example n. 4561 = (43.0, 467.0): cluster 2\n", + "Example n. 4562 = (43.0, 466.0): cluster 2\n", + "Example n. 4563 = (43.0, 465.0): cluster 2\n", + "Example n. 4564 = (43.0, 464.0): cluster 2\n", + "Example n. 4565 = (43.0, 463.0): cluster 2\n", + "Example n. 4566 = (43.0, 462.0): cluster 2\n", + "Example n. 4567 = (43.0, 461.0): cluster 2\n", + "Example n. 4568 = (43.0, 460.0): cluster 2\n", + "Example n. 4569 = (43.0, 459.0): cluster 3\n", + "Example n. 4570 = (43.0, 458.0): cluster 3\n", + "Example n. 4571 = (43.0, 457.0): cluster 3\n", + "Example n. 4572 = (43.0, 456.0): cluster 3\n", + "Example n. 4573 = (43.0, 455.0): cluster 3\n", + "Example n. 4574 = (43.0, 454.0): cluster 3\n", + "Example n. 4575 = (43.0, 434.0): cluster 3\n", + "Example n. 4576 = (43.0, 433.0): cluster 3\n", + "Example n. 4577 = (43.0, 432.0): cluster 3\n", + "Example n. 4578 = (43.0, 431.0): cluster 3\n", + "Example n. 4579 = (43.0, 430.0): cluster 3\n", + "Example n. 4580 = (43.0, 429.0): cluster 3\n", + "Example n. 4581 = (43.0, 411.0): cluster 3\n", + "Example n. 4582 = (43.0, 410.0): cluster 3\n", + "Example n. 4583 = (43.0, 409.0): cluster 3\n", + "Example n. 4584 = (43.0, 408.0): cluster 3\n", + "Example n. 4585 = (43.0, 407.0): cluster 3\n", + "Example n. 4586 = (43.0, 396.0): cluster 3\n", + "Example n. 4587 = (43.0, 395.0): cluster 3\n", + "Example n. 4588 = (43.0, 394.0): cluster 3\n", + "Example n. 4589 = (43.0, 393.0): cluster 3\n", + "Example n. 4590 = (43.0, 392.0): cluster 3\n", + "Example n. 4591 = (43.0, 346.0): cluster 3\n", + "Example n. 4592 = (43.0, 345.0): cluster 3\n", + "Example n. 4593 = (43.0, 344.0): cluster 2\n", + "Example n. 4594 = (43.0, 343.0): cluster 2\n", + "Example n. 4595 = (43.0, 342.0): cluster 2\n", + "Example n. 4596 = (43.0, 341.0): cluster 2\n", + "Example n. 4597 = (43.0, 340.0): cluster 2\n", + "Example n. 4598 = (43.0, 339.0): cluster 2\n", + "Example n. 4599 = (43.0, 338.0): cluster 2\n", + "Example n. 4600 = (43.0, 319.0): cluster 2\n", + "Example n. 4601 = (43.0, 318.0): cluster 2\n", + "Example n. 4602 = (43.0, 317.0): cluster 2\n", + "Example n. 4603 = (43.0, 316.0): cluster 2\n", + "Example n. 4604 = (43.0, 315.0): cluster 2\n", + "Example n. 4605 = (43.0, 314.0): cluster 2\n", + "Example n. 4606 = (43.0, 313.0): cluster 2\n", + "Example n. 4607 = (43.0, 312.0): cluster 2\n", + "Example n. 4608 = (43.0, 311.0): cluster 2\n", + "Example n. 4609 = (43.0, 310.0): cluster 2\n", + "Example n. 4610 = (43.0, 309.0): cluster 2\n", + "Example n. 4611 = (43.0, 308.0): cluster 2\n", + "Example n. 4612 = (43.0, 307.0): cluster 2\n", + "Example n. 4613 = (43.0, 306.0): cluster 2\n", + "Example n. 4614 = (43.0, 305.0): cluster 2\n", + "Example n. 4615 = (43.0, 304.0): cluster 2\n", + "Example n. 4616 = (43.0, 303.0): cluster 2\n", + "Example n. 4617 = (43.0, 302.0): cluster 3\n", + "Example n. 4618 = (43.0, 301.0): cluster 3\n", + "Example n. 4619 = (43.0, 299.0): cluster 3\n", + "Example n. 4620 = (43.0, 298.0): cluster 3\n", + "Example n. 4621 = (43.0, 297.0): cluster 3\n", + "Example n. 4622 = (43.0, 296.0): cluster 3\n", + "Example n. 4623 = (43.0, 295.0): cluster 3\n", + "Example n. 4624 = (43.0, 291.0): cluster 3\n", + "Example n. 4625 = (43.0, 290.0): cluster 3\n", + "Example n. 4626 = (43.0, 289.0): cluster 3\n", + "Example n. 4627 = (43.0, 288.0): cluster 3\n", + "Example n. 4628 = (43.0, 287.0): cluster 3\n", + "Example n. 4629 = (43.0, 286.0): cluster 3\n", + "Example n. 4630 = (43.0, 285.0): cluster 3\n", + "Example n. 4631 = (43.0, 284.0): cluster 3\n", + "Example n. 4632 = (43.0, 283.0): cluster 3\n", + "Example n. 4633 = (43.0, 282.0): cluster 3\n", + "Example n. 4634 = (43.0, 279.0): cluster 3\n", + "Example n. 4635 = (43.0, 278.0): cluster 3\n", + "Example n. 4636 = (43.0, 277.0): cluster 3\n", + "Example n. 4637 = (43.0, 276.0): cluster 2\n", + "Example n. 4638 = (43.0, 275.0): cluster 2\n", + "Example n. 4639 = (43.0, 274.0): cluster 2\n", + "Example n. 4640 = (43.0, 270.0): cluster 2\n", + "Example n. 4641 = (43.0, 269.0): cluster 2\n", + "Example n. 4642 = (43.0, 268.0): cluster 2\n", + "Example n. 4643 = (43.0, 267.0): cluster 2\n", + "Example n. 4644 = (43.0, 266.0): cluster 2\n", + "Example n. 4645 = (43.0, 265.0): cluster 2\n", + "Example n. 4646 = (43.0, 264.0): cluster 2\n", + "Example n. 4647 = (43.0, 263.0): cluster 2\n", + "Example n. 4648 = (43.0, 262.0): cluster 2\n", + "Example n. 4649 = (43.0, 261.0): cluster 2\n", + "Example n. 4650 = (43.0, 260.0): cluster 2\n", + "Example n. 4651 = (43.0, 259.0): cluster 2\n", + "Example n. 4652 = (43.0, 258.0): cluster 2\n", + "Example n. 4653 = (43.0, 257.0): cluster 2\n", + "Example n. 4654 = (43.0, 256.0): cluster 2\n", + "Example n. 4655 = (43.0, 255.0): cluster 2\n", + "Example n. 4656 = (43.0, 254.0): cluster 2\n", + "Example n. 4657 = (43.0, 253.0): cluster 2\n", + "Example n. 4658 = (43.0, 250.0): cluster 2\n", + "Example n. 4659 = (43.0, 249.0): cluster 2\n", + "Example n. 4660 = (43.0, 248.0): cluster 3\n", + "Example n. 4661 = (43.0, 247.0): cluster 3\n", + "Example n. 4662 = (43.0, 246.0): cluster 3\n", + "Example n. 4663 = (43.0, 245.0): cluster 3\n", + "Example n. 4664 = (43.0, 242.0): cluster 3\n", + "Example n. 4665 = (43.0, 241.0): cluster 3\n", + "Example n. 4666 = (43.0, 240.0): cluster 3\n", + "Example n. 4667 = (43.0, 239.0): cluster 3\n", + "Example n. 4668 = (43.0, 238.0): cluster 3\n", + "Example n. 4669 = (43.0, 237.0): cluster 3\n", + "Example n. 4670 = (43.0, 236.0): cluster 3\n", + "Example n. 4671 = (43.0, 235.0): cluster 3\n", + "Example n. 4672 = (43.0, 234.0): cluster 3\n", + "Example n. 4673 = (43.0, 233.0): cluster 3\n", + "Example n. 4674 = (43.0, 232.0): cluster 3\n", + "Example n. 4675 = (43.0, 229.0): cluster 3\n", + "Example n. 4676 = (43.0, 228.0): cluster 3\n", + "Example n. 4677 = (43.0, 227.0): cluster 3\n", + "Example n. 4678 = (43.0, 226.0): cluster 3\n", + "Example n. 4679 = (43.0, 225.0): cluster 3\n", + "Example n. 4680 = (43.0, 218.0): cluster 2\n", + "Example n. 4681 = (43.0, 215.0): cluster 2\n", + "Example n. 4682 = (43.0, 214.0): cluster 2\n", + "Example n. 4683 = (43.0, 213.0): cluster 2\n", + "Example n. 4684 = (43.0, 212.0): cluster 2\n", + "Example n. 4685 = (43.0, 211.0): cluster 2\n", + "Example n. 4686 = (43.0, 209.0): cluster 2\n", + "Example n. 4687 = (43.0, 208.0): cluster 2\n", + "Example n. 4688 = (43.0, 207.0): cluster 2\n", + "Example n. 4689 = (43.0, 206.0): cluster 2\n", + "Example n. 4690 = (43.0, 205.0): cluster 2\n", + "Example n. 4691 = (43.0, 204.0): cluster 2\n", + "Example n. 4692 = (43.0, 203.0): cluster 2\n", + "Example n. 4693 = (43.0, 202.0): cluster 2\n", + "Example n. 4694 = (43.0, 201.0): cluster 2\n", + "Example n. 4695 = (43.0, 200.0): cluster 2\n", + "Example n. 4696 = (43.0, 199.0): cluster 2\n", + "Example n. 4697 = (43.0, 198.0): cluster 2\n", + "Example n. 4698 = (43.0, 189.0): cluster 2\n", + "Example n. 4699 = (43.0, 188.0): cluster 2\n", + "Example n. 4700 = (43.0, 187.0): cluster 2\n", + "Example n. 4701 = (43.0, 186.0): cluster 2\n", + "Example n. 4702 = (43.0, 185.0): cluster 2\n", + "Example n. 4703 = (43.0, 184.0): cluster 2\n", + "Example n. 4704 = (43.0, 183.0): cluster 3\n", + "Example n. 4705 = (43.0, 182.0): cluster 3\n", + "Example n. 4706 = (43.0, 181.0): cluster 3\n", + "Example n. 4707 = (43.0, 178.0): cluster 3\n", + "Example n. 4708 = (43.0, 177.0): cluster 3\n", + "Example n. 4709 = (43.0, 176.0): cluster 3\n", + "Example n. 4710 = (43.0, 175.0): cluster 3\n", + "Example n. 4711 = (43.0, 173.0): cluster 3\n", + "Example n. 4712 = (43.0, 172.0): cluster 3\n", + "Example n. 4713 = (43.0, 171.0): cluster 3\n", + "Example n. 4714 = (43.0, 170.0): cluster 3\n", + "Example n. 4715 = (43.0, 169.0): cluster 3\n", + "Example n. 4716 = (43.0, 168.0): cluster 3\n", + "Example n. 4717 = (43.0, 161.0): cluster 3\n", + "Example n. 4718 = (43.0, 160.0): cluster 3\n", + "Example n. 4719 = (43.0, 159.0): cluster 3\n", + "Example n. 4720 = (43.0, 158.0): cluster 3\n", + "Example n. 4721 = (43.0, 157.0): cluster 3\n", + "Example n. 4722 = (43.0, 156.0): cluster 3\n", + "Example n. 4723 = (43.0, 155.0): cluster 3\n", + "Example n. 4724 = (43.0, 154.0): cluster 2\n", + "Example n. 4725 = (43.0, 153.0): cluster 2\n", + "Example n. 4726 = (43.0, 152.0): cluster 2\n", + "Example n. 4727 = (43.0, 151.0): cluster 2\n", + "Example n. 4728 = (43.0, 150.0): cluster 2\n", + "Example n. 4729 = (43.0, 149.0): cluster 2\n", + "Example n. 4730 = (43.0, 148.0): cluster 2\n", + "Example n. 4731 = (43.0, 146.0): cluster 2\n", + "Example n. 4732 = (43.0, 145.0): cluster 2\n", + "Example n. 4733 = (43.0, 144.0): cluster 2\n", + "Example n. 4734 = (43.0, 143.0): cluster 2\n", + "Example n. 4735 = (43.0, 142.0): cluster 2\n", + "Example n. 4736 = (43.0, 130.0): cluster 2\n", + "Example n. 4737 = (43.0, 129.0): cluster 2\n", + "Example n. 4738 = (43.0, 128.0): cluster 2\n", + "Example n. 4739 = (43.0, 127.0): cluster 2\n", + "Example n. 4740 = (43.0, 126.0): cluster 2\n", + "Example n. 4741 = (43.0, 120.0): cluster 2\n", + "Example n. 4742 = (43.0, 119.0): cluster 2\n", + "Example n. 4743 = (43.0, 118.0): cluster 2\n", + "Example n. 4744 = (43.0, 117.0): cluster 2\n", + "Example n. 4745 = (43.0, 116.0): cluster 2\n", + "Example n. 4746 = (44.0, 495.0): cluster 2\n", + "Example n. 4747 = (44.0, 494.0): cluster 2\n", + "Example n. 4748 = (44.0, 493.0): cluster 2\n", + "Example n. 4749 = (44.0, 492.0): cluster 2\n", + "Example n. 4750 = (44.0, 491.0): cluster 2\n", + "Example n. 4751 = (44.0, 490.0): cluster 2\n", + "Example n. 4752 = (44.0, 489.0): cluster 2\n", + "Example n. 4753 = (44.0, 488.0): cluster 2\n", + "Example n. 4754 = (44.0, 487.0): cluster 2\n", + "Example n. 4755 = (44.0, 486.0): cluster 2\n", + "Example n. 4756 = (44.0, 485.0): cluster 3\n", + "Example n. 4757 = (44.0, 484.0): cluster 3\n", + "Example n. 4758 = (44.0, 483.0): cluster 3\n", + "Example n. 4759 = (44.0, 482.0): cluster 3\n", + "Example n. 4760 = (44.0, 481.0): cluster 3\n", + "Example n. 4761 = (44.0, 480.0): cluster 3\n", + "Example n. 4762 = (44.0, 479.0): cluster 3\n", + "Example n. 4763 = (44.0, 478.0): cluster 3\n", + "Example n. 4764 = (44.0, 477.0): cluster 3\n", + "Example n. 4765 = (44.0, 476.0): cluster 3\n", + "Example n. 4766 = (44.0, 475.0): cluster 3\n", + "Example n. 4767 = (44.0, 474.0): cluster 3\n", + "Example n. 4768 = (44.0, 473.0): cluster 3\n", + "Example n. 4769 = (44.0, 472.0): cluster 3\n", + "Example n. 4770 = (44.0, 471.0): cluster 3\n", + "Example n. 4771 = (44.0, 470.0): cluster 3\n", + "Example n. 4772 = (44.0, 469.0): cluster 3\n", + "Example n. 4773 = (44.0, 468.0): cluster 3\n", + "Example n. 4774 = (44.0, 467.0): cluster 3\n", + "Example n. 4775 = (44.0, 466.0): cluster 3\n", + "Example n. 4776 = (44.0, 465.0): cluster 2\n", + "Example n. 4777 = (44.0, 464.0): cluster 2\n", + "Example n. 4778 = (44.0, 463.0): cluster 2\n", + "Example n. 4779 = (44.0, 462.0): cluster 2\n", + "Example n. 4780 = (44.0, 461.0): cluster 2\n", + "Example n. 4781 = (44.0, 460.0): cluster 2\n", + "Example n. 4782 = (44.0, 459.0): cluster 2\n", + "Example n. 4783 = (44.0, 458.0): cluster 2\n", + "Example n. 4784 = (44.0, 457.0): cluster 2\n", + "Example n. 4785 = (44.0, 456.0): cluster 2\n", + "Example n. 4786 = (44.0, 455.0): cluster 2\n", + "Example n. 4787 = (44.0, 454.0): cluster 2\n", + "Example n. 4788 = (44.0, 433.0): cluster 2\n", + "Example n. 4789 = (44.0, 432.0): cluster 2\n", + "Example n. 4790 = (44.0, 431.0): cluster 2\n", + "Example n. 4791 = (44.0, 430.0): cluster 2\n", + "Example n. 4792 = (44.0, 429.0): cluster 2\n", + "Example n. 4793 = (44.0, 410.0): cluster 2\n", + "Example n. 4794 = (44.0, 409.0): cluster 2\n", + "Example n. 4795 = (44.0, 408.0): cluster 2\n", + "Example n. 4796 = (44.0, 407.0): cluster 2\n", + "Example n. 4797 = (44.0, 396.0): cluster 2\n", + "Example n. 4798 = (44.0, 395.0): cluster 2\n", + "Example n. 4799 = (44.0, 394.0): cluster 2\n", + "Example n. 4800 = (44.0, 393.0): cluster 2\n", + "Example n. 4801 = (44.0, 392.0): cluster 2\n", + "Example n. 4802 = (44.0, 345.0): cluster 2\n", + "Example n. 4803 = (44.0, 344.0): cluster 2\n", + "Example n. 4804 = (44.0, 343.0): cluster 2\n", + "Example n. 4805 = (44.0, 342.0): cluster 2\n", + "Example n. 4806 = (44.0, 341.0): cluster 2\n", + "Example n. 4807 = (44.0, 340.0): cluster 2\n", + "Example n. 4808 = (44.0, 339.0): cluster 3\n", + "Example n. 4809 = (44.0, 338.0): cluster 3\n", + "Example n. 4810 = (44.0, 337.0): cluster 3\n", + "Example n. 4811 = (44.0, 327.0): cluster 3\n", + "Example n. 4812 = (44.0, 326.0): cluster 3\n", + "Example n. 4813 = (44.0, 325.0): cluster 3\n", + "Example n. 4814 = (44.0, 324.0): cluster 3\n", + "Example n. 4815 = (44.0, 323.0): cluster 3\n", + "Example n. 4816 = (44.0, 322.0): cluster 3\n", + "Example n. 4817 = (44.0, 321.0): cluster 3\n", + "Example n. 4818 = (44.0, 320.0): cluster 3\n", + "Example n. 4819 = (44.0, 319.0): cluster 2\n", + "Example n. 4820 = (44.0, 318.0): cluster 2\n", + "Example n. 4821 = (44.0, 317.0): cluster 2\n", + "Example n. 4822 = (44.0, 316.0): cluster 2\n", + "Example n. 4823 = (44.0, 315.0): cluster 2\n", + "Example n. 4824 = (44.0, 314.0): cluster 2\n", + "Example n. 4825 = (44.0, 313.0): cluster 2\n", + "Example n. 4826 = (44.0, 312.0): cluster 2\n", + "Example n. 4827 = (44.0, 311.0): cluster 2\n", + "Example n. 4828 = (44.0, 310.0): cluster 2\n", + "Example n. 4829 = (44.0, 309.0): cluster 2\n", + "Example n. 4830 = (44.0, 308.0): cluster 2\n", + "Example n. 4831 = (44.0, 307.0): cluster 2\n", + "Example n. 4832 = (44.0, 306.0): cluster 2\n", + "Example n. 4833 = (44.0, 305.0): cluster 2\n", + "Example n. 4834 = (44.0, 304.0): cluster 2\n", + "Example n. 4835 = (44.0, 303.0): cluster 2\n", + "Example n. 4836 = (44.0, 302.0): cluster 2\n", + "Example n. 4837 = (44.0, 301.0): cluster 2\n", + "Example n. 4838 = (44.0, 300.0): cluster 2\n", + "Example n. 4839 = (44.0, 299.0): cluster 2\n", + "Example n. 4840 = (44.0, 298.0): cluster 2\n", + "Example n. 4841 = (44.0, 297.0): cluster 2\n", + "Example n. 4842 = (44.0, 296.0): cluster 2\n", + "Example n. 4843 = (44.0, 295.0): cluster 2\n", + "Example n. 4844 = (44.0, 294.0): cluster 2\n", + "Example n. 4845 = (44.0, 289.0): cluster 2\n", + "Example n. 4846 = (44.0, 288.0): cluster 2\n", + "Example n. 4847 = (44.0, 287.0): cluster 2\n", + "Example n. 4848 = (44.0, 286.0): cluster 2\n", + "Example n. 4849 = (44.0, 285.0): cluster 2\n", + "Example n. 4850 = (44.0, 284.0): cluster 2\n", + "Example n. 4851 = (44.0, 283.0): cluster 2\n", + "Example n. 4852 = (44.0, 282.0): cluster 2\n", + "Example n. 4853 = (44.0, 279.0): cluster 3\n", + "Example n. 4854 = (44.0, 278.0): cluster 3\n", + "Example n. 4855 = (44.0, 277.0): cluster 3\n", + "Example n. 4856 = (44.0, 276.0): cluster 3\n", + "Example n. 4857 = (44.0, 275.0): cluster 3\n", + "Example n. 4858 = (44.0, 274.0): cluster 3\n", + "Example n. 4859 = (44.0, 267.0): cluster 3\n", + "Example n. 4860 = (44.0, 266.0): cluster 3\n", + "Example n. 4861 = (44.0, 265.0): cluster 3\n", + "Example n. 4862 = (44.0, 264.0): cluster 3\n", + "Example n. 4863 = (44.0, 263.0): cluster 2\n", + "Example n. 4864 = (44.0, 262.0): cluster 2\n", + "Example n. 4865 = (44.0, 261.0): cluster 2\n", + "Example n. 4866 = (44.0, 260.0): cluster 2\n", + "Example n. 4867 = (44.0, 259.0): cluster 2\n", + "Example n. 4868 = (44.0, 258.0): cluster 2\n", + "Example n. 4869 = (44.0, 257.0): cluster 2\n", + "Example n. 4870 = (44.0, 256.0): cluster 2\n", + "Example n. 4871 = (44.0, 255.0): cluster 2\n", + "Example n. 4872 = (44.0, 254.0): cluster 2\n", + "Example n. 4873 = (44.0, 253.0): cluster 2\n", + "Example n. 4874 = (44.0, 252.0): cluster 2\n", + "Example n. 4875 = (44.0, 251.0): cluster 2\n", + "Example n. 4876 = (44.0, 250.0): cluster 2\n", + "Example n. 4877 = (44.0, 249.0): cluster 2\n", + "Example n. 4878 = (44.0, 248.0): cluster 2\n", + "Example n. 4879 = (44.0, 247.0): cluster 2\n", + "Example n. 4880 = (44.0, 246.0): cluster 2\n", + "Example n. 4881 = (44.0, 243.0): cluster 2\n", + "Example n. 4882 = (44.0, 242.0): cluster 2\n", + "Example n. 4883 = (44.0, 241.0): cluster 2\n", + "Example n. 4884 = (44.0, 240.0): cluster 2\n", + "Example n. 4885 = (44.0, 239.0): cluster 2\n", + "Example n. 4886 = (44.0, 238.0): cluster 2\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Example n. 4887 = (44.0, 237.0): cluster 2\n", + "Example n. 4888 = (44.0, 236.0): cluster 2\n", + "Example n. 4889 = (44.0, 235.0): cluster 2\n", + "Example n. 4890 = (44.0, 234.0): cluster 2\n", + "Example n. 4891 = (44.0, 229.0): cluster 2\n", + "Example n. 4892 = (44.0, 228.0): cluster 2\n", + "Example n. 4893 = (44.0, 227.0): cluster 2\n", + "Example n. 4894 = (44.0, 226.0): cluster 2\n", + "Example n. 4895 = (44.0, 225.0): cluster 2\n", + "Example n. 4896 = (44.0, 220.0): cluster 2\n", + "Example n. 4897 = (44.0, 219.0): cluster 2\n", + "Example n. 4898 = (44.0, 218.0): cluster 3\n", + "Example n. 4899 = (44.0, 217.0): cluster 3\n", + "Example n. 4900 = (44.0, 216.0): cluster 3\n", + "Example n. 4901 = (44.0, 213.0): cluster 3\n", + "Example n. 4902 = (44.0, 209.0): cluster 3\n", + "Example n. 4903 = (44.0, 208.0): cluster 3\n", + "Example n. 4904 = (44.0, 207.0): cluster 3\n", + "Example n. 4905 = (44.0, 206.0): cluster 3\n", + "Example n. 4906 = (44.0, 205.0): cluster 3\n", + "Example n. 4907 = (44.0, 204.0): cluster 2\n", + "Example n. 4908 = (44.0, 203.0): cluster 2\n", + "Example n. 4909 = (44.0, 202.0): cluster 2\n", + "Example n. 4910 = (44.0, 201.0): cluster 2\n", + "Example n. 4911 = (44.0, 200.0): cluster 2\n", + "Example n. 4912 = (44.0, 199.0): cluster 2\n", + "Example n. 4913 = (44.0, 198.0): cluster 2\n", + "Example n. 4914 = (44.0, 189.0): cluster 2\n", + "Example n. 4915 = (44.0, 188.0): cluster 2\n", + "Example n. 4916 = (44.0, 187.0): cluster 2\n", + "Example n. 4917 = (44.0, 186.0): cluster 2\n", + "Example n. 4918 = (44.0, 185.0): cluster 2\n", + "Example n. 4919 = (44.0, 184.0): cluster 2\n", + "Example n. 4920 = (44.0, 183.0): cluster 2\n", + "Example n. 4921 = (44.0, 182.0): cluster 2\n", + "Example n. 4922 = (44.0, 181.0): cluster 2\n", + "Example n. 4923 = (44.0, 179.0): cluster 2\n", + "Example n. 4924 = (44.0, 178.0): cluster 2\n", + "Example n. 4925 = (44.0, 177.0): cluster 2\n", + "Example n. 4926 = (44.0, 176.0): cluster 2\n", + "Example n. 4927 = (44.0, 175.0): cluster 2\n", + "Example n. 4928 = (44.0, 174.0): cluster 2\n", + "Example n. 4929 = (44.0, 173.0): cluster 2\n", + "Example n. 4930 = (44.0, 172.0): cluster 2\n", + "Example n. 4931 = (44.0, 171.0): cluster 2\n", + "Example n. 4932 = (44.0, 170.0): cluster 2\n", + "Example n. 4933 = (44.0, 169.0): cluster 2\n", + "Example n. 4934 = (44.0, 168.0): cluster 2\n", + "Example n. 4935 = (44.0, 167.0): cluster 2\n", + "Example n. 4936 = (44.0, 166.0): cluster 2\n", + "Example n. 4937 = (44.0, 161.0): cluster 2\n", + "Example n. 4938 = (44.0, 160.0): cluster 2\n", + "Example n. 4939 = (44.0, 159.0): cluster 2\n", + "Example n. 4940 = (44.0, 158.0): cluster 2\n", + "Example n. 4941 = (44.0, 157.0): cluster 2\n", + "Example n. 4942 = (44.0, 156.0): cluster 2\n", + "Example n. 4943 = (44.0, 155.0): cluster 2\n", + "Example n. 4944 = (44.0, 154.0): cluster 2\n", + "Example n. 4945 = (44.0, 153.0): cluster 2\n", + "Example n. 4946 = (44.0, 152.0): cluster 2\n", + "Example n. 4947 = (44.0, 151.0): cluster 2\n", + "Example n. 4948 = (44.0, 150.0): cluster 2\n", + "Example n. 4949 = (44.0, 149.0): cluster 3\n", + "Example n. 4950 = (44.0, 148.0): cluster 3\n", + "Example n. 4951 = (44.0, 145.0): cluster 3\n", + "Example n. 4952 = (44.0, 144.0): cluster 3\n", + "Example n. 4953 = (44.0, 143.0): cluster 3\n", + "Example n. 4954 = (44.0, 142.0): cluster 3\n", + "Example n. 4955 = (44.0, 131.0): cluster 3\n", + "Example n. 4956 = (44.0, 130.0): cluster 2\n", + "Example n. 4957 = (44.0, 129.0): cluster 2\n", + "Example n. 4958 = (44.0, 128.0): cluster 2\n", + "Example n. 4959 = (44.0, 127.0): cluster 2\n", + "Example n. 4960 = (44.0, 126.0): cluster 2\n", + "Example n. 4961 = (44.0, 120.0): cluster 2\n", + "Example n. 4962 = (44.0, 119.0): cluster 2\n", + "Example n. 4963 = (44.0, 118.0): cluster 2\n", + "Example n. 4964 = (44.0, 117.0): cluster 2\n", + "Example n. 4965 = (44.0, 116.0): cluster 2\n", + "Example n. 4966 = (45.0, 496.0): cluster 2\n", + "Example n. 4967 = (45.0, 495.0): cluster 2\n", + "Example n. 4968 = (45.0, 494.0): cluster 2\n", + "Example n. 4969 = (45.0, 493.0): cluster 2\n", + "Example n. 4970 = (45.0, 492.0): cluster 2\n", + "Example n. 4971 = (45.0, 491.0): cluster 2\n", + "Example n. 4972 = (45.0, 490.0): cluster 2\n", + "Example n. 4973 = (45.0, 489.0): cluster 2\n", + "Example n. 4974 = (45.0, 488.0): cluster 2\n", + "Example n. 4975 = (45.0, 487.0): cluster 2\n", + "Example n. 4976 = (45.0, 486.0): cluster 2\n", + "Example n. 4977 = (45.0, 485.0): cluster 2\n", + "Example n. 4978 = (45.0, 484.0): cluster 2\n", + "Example n. 4979 = (45.0, 483.0): cluster 2\n", + "Example n. 4980 = (45.0, 482.0): cluster 2\n", + "Example n. 4981 = (45.0, 481.0): cluster 2\n", + "Example n. 4982 = (45.0, 480.0): cluster 2\n", + "Example n. 4983 = (45.0, 479.0): cluster 2\n", + "Example n. 4984 = (45.0, 478.0): cluster 2\n", + "Example n. 4985 = (45.0, 477.0): cluster 2\n", + "Example n. 4986 = (45.0, 476.0): cluster 2\n", + "Example n. 4987 = (45.0, 475.0): cluster 2\n", + "Example n. 4988 = (45.0, 474.0): cluster 2\n", + "Example n. 4989 = (45.0, 473.0): cluster 2\n", + "Example n. 4990 = (45.0, 472.0): cluster 2\n", + "Example n. 4991 = (45.0, 471.0): cluster 2\n", + "Example n. 4992 = (45.0, 470.0): cluster 2\n", + "Example n. 4993 = (45.0, 469.0): cluster 2\n", + "Example n. 4994 = (45.0, 468.0): cluster 2\n", + "Example n. 4995 = (45.0, 467.0): cluster 2\n", + "Example n. 4996 = (45.0, 466.0): cluster 2\n", + "Example n. 4997 = (45.0, 465.0): cluster 2\n", + "Example n. 4998 = (45.0, 464.0): cluster 2\n", + "Example n. 4999 = (45.0, 463.0): cluster 2\n", + "Example n. 5000 = (45.0, 462.0): cluster 2\n", + "Example n. 5001 = (45.0, 461.0): cluster 2\n", + "Example n. 5002 = (45.0, 460.0): cluster 2\n", + "Example n. 5003 = (45.0, 459.0): cluster 2\n", + "Example n. 5004 = (45.0, 458.0): cluster 2\n", + "Example n. 5005 = (45.0, 457.0): cluster 2\n", + "Example n. 5006 = (45.0, 456.0): cluster 2\n", + "Example n. 5007 = (45.0, 455.0): cluster 2\n", + "Example n. 5008 = (45.0, 454.0): cluster 2\n", + "Example n. 5009 = (45.0, 453.0): cluster 2\n", + "Example n. 5010 = (45.0, 433.0): cluster 2\n", + "Example n. 5011 = (45.0, 432.0): cluster 2\n", + "Example n. 5012 = (45.0, 431.0): cluster 2\n", + "Example n. 5013 = (45.0, 430.0): cluster 2\n", + "Example n. 5014 = (45.0, 394.0): cluster 2\n", + "Example n. 5015 = (45.0, 341.0): cluster 2\n", + "Example n. 5016 = (45.0, 340.0): cluster 2\n", + "Example n. 5017 = (45.0, 339.0): cluster 2\n", + "Example n. 5018 = (45.0, 338.0): cluster 2\n", + "Example n. 5019 = (45.0, 337.0): cluster 2\n", + "Example n. 5020 = (45.0, 336.0): cluster 2\n", + "Example n. 5021 = (45.0, 327.0): cluster 2\n", + "Example n. 5022 = (45.0, 326.0): cluster 2\n", + "Example n. 5023 = (45.0, 325.0): cluster 2\n", + "Example n. 5024 = (45.0, 324.0): cluster 2\n", + "Example n. 5025 = (45.0, 323.0): cluster 2\n", + "Example n. 5026 = (45.0, 322.0): cluster 2\n", + "Example n. 5027 = (45.0, 321.0): cluster 2\n", + "Example n. 5028 = (45.0, 320.0): cluster 2\n", + "Example n. 5029 = (45.0, 319.0): cluster 2\n", + "Example n. 5030 = (45.0, 318.0): cluster 2\n", + "Example n. 5031 = (45.0, 317.0): cluster 2\n", + "Example n. 5032 = (45.0, 316.0): cluster 2\n", + "Example n. 5033 = (45.0, 315.0): cluster 2\n", + "Example n. 5034 = (45.0, 314.0): cluster 2\n", + "Example n. 5035 = (45.0, 313.0): cluster 2\n", + "Example n. 5036 = (45.0, 312.0): cluster 2\n", + "Example n. 5037 = (45.0, 311.0): cluster 2\n", + "Example n. 5038 = (45.0, 310.0): cluster 2\n", + "Example n. 5039 = (45.0, 309.0): cluster 2\n", + "Example n. 5040 = (45.0, 308.0): cluster 2\n", + "Example n. 5041 = (45.0, 307.0): cluster 2\n", + "Example n. 5042 = (45.0, 306.0): cluster 2\n", + "Example n. 5043 = (45.0, 305.0): cluster 2\n", + "Example n. 5044 = (45.0, 304.0): cluster 2\n", + "Example n. 5045 = (45.0, 303.0): cluster 2\n", + "Example n. 5046 = (45.0, 302.0): cluster 2\n", + "Example n. 5047 = (45.0, 301.0): cluster 2\n", + "Example n. 5048 = (45.0, 300.0): cluster 2\n", + "Example n. 5049 = (45.0, 299.0): cluster 2\n", + "Example n. 5050 = (45.0, 298.0): cluster 2\n", + "Example n. 5051 = (45.0, 297.0): cluster 2\n", + "Example n. 5052 = (45.0, 296.0): cluster 2\n", + "Example n. 5053 = (45.0, 295.0): cluster 2\n", + "Example n. 5054 = (45.0, 294.0): cluster 2\n", + "Example n. 5055 = (45.0, 288.0): cluster 2\n", + "Example n. 5056 = (45.0, 287.0): cluster 2\n", + "Example n. 5057 = (45.0, 286.0): cluster 2\n", + "Example n. 5058 = (45.0, 285.0): cluster 2\n", + "Example n. 5059 = (45.0, 284.0): cluster 2\n", + "Example n. 5060 = (45.0, 283.0): cluster 2\n", + "Example n. 5061 = (45.0, 282.0): cluster 2\n", + "Example n. 5062 = (45.0, 281.0): cluster 2\n", + "Example n. 5063 = (45.0, 279.0): cluster 2\n", + "Example n. 5064 = (45.0, 278.0): cluster 2\n", + "Example n. 5065 = (45.0, 277.0): cluster 2\n", + "Example n. 5066 = (45.0, 276.0): cluster 2\n", + "Example n. 5067 = (45.0, 275.0): cluster 2\n", + "Example n. 5068 = (45.0, 266.0): cluster 2\n", + "Example n. 5069 = (45.0, 265.0): cluster 2\n", + "Example n. 5070 = (45.0, 262.0): cluster 2\n", + "Example n. 5071 = (45.0, 261.0): cluster 2\n", + "Example n. 5072 = (45.0, 260.0): cluster 2\n", + "Example n. 5073 = (45.0, 259.0): cluster 2\n", + "Example n. 5074 = (45.0, 258.0): cluster 2\n", + "Example n. 5075 = (45.0, 257.0): cluster 2\n", + "Example n. 5076 = (45.0, 256.0): cluster 2\n", + "Example n. 5077 = (45.0, 255.0): cluster 2\n", + "Example n. 5078 = (45.0, 254.0): cluster 2\n", + "Example n. 5079 = (45.0, 253.0): cluster 2\n", + "Example n. 5080 = (45.0, 252.0): cluster 2\n", + "Example n. 5081 = (45.0, 251.0): cluster 2\n", + "Example n. 5082 = (45.0, 250.0): cluster 2\n", + "Example n. 5083 = (45.0, 249.0): cluster 2\n", + "Example n. 5084 = (45.0, 244.0): cluster 2\n", + "Example n. 5085 = (45.0, 243.0): cluster 2\n", + "Example n. 5086 = (45.0, 242.0): cluster 2\n", + "Example n. 5087 = (45.0, 241.0): cluster 2\n", + "Example n. 5088 = (45.0, 240.0): cluster 2\n", + "Example n. 5089 = (45.0, 239.0): cluster 2\n", + "Example n. 5090 = (45.0, 238.0): cluster 2\n", + "Example n. 5091 = (45.0, 237.0): cluster 2\n", + "Example n. 5092 = (45.0, 236.0): cluster 2\n", + "Example n. 5093 = (45.0, 235.0): cluster 2\n", + "Example n. 5094 = (45.0, 229.0): cluster 2\n", + "Example n. 5095 = (45.0, 228.0): cluster 2\n", + "Example n. 5096 = (45.0, 227.0): cluster 2\n", + "Example n. 5097 = (45.0, 226.0): cluster 2\n", + "Example n. 5098 = (45.0, 225.0): cluster 2\n", + "Example n. 5099 = (45.0, 224.0): cluster 2\n", + "Example n. 5100 = (45.0, 220.0): cluster 2\n", + "Example n. 5101 = (45.0, 219.0): cluster 2\n", + "Example n. 5102 = (45.0, 218.0): cluster 2\n", + "Example n. 5103 = (45.0, 217.0): cluster 2\n", + "Example n. 5104 = (45.0, 216.0): cluster 2\n", + "Example n. 5105 = (45.0, 209.0): cluster 2\n", + "Example n. 5106 = (45.0, 208.0): cluster 2\n", + "Example n. 5107 = (45.0, 207.0): cluster 2\n", + "Example n. 5108 = (45.0, 206.0): cluster 2\n", + "Example n. 5109 = (45.0, 205.0): cluster 2\n", + "Example n. 5110 = (45.0, 202.0): cluster 2\n", + "Example n. 5111 = (45.0, 201.0): cluster 2\n", + "Example n. 5112 = (45.0, 200.0): cluster 2\n", + "Example n. 5113 = (45.0, 199.0): cluster 2\n", + "Example n. 5114 = (45.0, 198.0): cluster 2\n", + "Example n. 5115 = (45.0, 197.0): cluster 2\n", + "Example n. 5116 = (45.0, 189.0): cluster 2\n", + "Example n. 5117 = (45.0, 188.0): cluster 2\n", + "Example n. 5118 = (45.0, 187.0): cluster 2\n", + "Example n. 5119 = (45.0, 186.0): cluster 2\n", + "Example n. 5120 = (45.0, 185.0): cluster 2\n", + "Example n. 5121 = (45.0, 184.0): cluster 2\n", + "Example n. 5122 = (45.0, 183.0): cluster 2\n", + "Example n. 5123 = (45.0, 182.0): cluster 2\n", + "Example n. 5124 = (45.0, 181.0): cluster 2\n", + "Example n. 5125 = (45.0, 179.0): cluster 2\n", + "Example n. 5126 = (45.0, 178.0): cluster 2\n", + "Example n. 5127 = (45.0, 177.0): cluster 2\n", + "Example n. 5128 = (45.0, 176.0): cluster 2\n", + "Example n. 5129 = (45.0, 175.0): cluster 2\n", + "Example n. 5130 = (45.0, 174.0): cluster 2\n", + "Example n. 5131 = (45.0, 173.0): cluster 2\n", + "Example n. 5132 = (45.0, 172.0): cluster 2\n", + "Example n. 5133 = (45.0, 171.0): cluster 2\n", + "Example n. 5134 = (45.0, 170.0): cluster 2\n", + "Example n. 5135 = (45.0, 169.0): cluster 2\n", + "Example n. 5136 = (45.0, 168.0): cluster 2\n", + "Example n. 5137 = (45.0, 167.0): cluster 2\n", + "Example n. 5138 = (45.0, 166.0): cluster 2\n", + "Example n. 5139 = (45.0, 165.0): cluster 2\n", + "Example n. 5140 = (45.0, 161.0): cluster 2\n", + "Example n. 5141 = (45.0, 160.0): cluster 2\n", + "Example n. 5142 = (45.0, 159.0): cluster 2\n", + "Example n. 5143 = (45.0, 158.0): cluster 2\n", + "Example n. 5144 = (45.0, 157.0): cluster 2\n", + "Example n. 5145 = (45.0, 156.0): cluster 2\n", + "Example n. 5146 = (45.0, 155.0): cluster 2\n", + "Example n. 5147 = (45.0, 154.0): cluster 2\n", + "Example n. 5148 = (45.0, 153.0): cluster 2\n", + "Example n. 5149 = (45.0, 152.0): cluster 2\n", + "Example n. 5150 = (45.0, 151.0): cluster 2\n", + "Example n. 5151 = (45.0, 150.0): cluster 2\n", + "Example n. 5152 = (45.0, 149.0): cluster 2\n", + "Example n. 5153 = (45.0, 136.0): cluster 2\n", + "Example n. 5154 = (45.0, 135.0): cluster 2\n", + "Example n. 5155 = (45.0, 134.0): cluster 2\n", + "Example n. 5156 = (45.0, 131.0): cluster 2\n", + "Example n. 5157 = (45.0, 130.0): cluster 2\n", + "Example n. 5158 = (45.0, 129.0): cluster 2\n", + "Example n. 5159 = (45.0, 128.0): cluster 2\n", + "Example n. 5160 = (45.0, 127.0): cluster 2\n", + "Example n. 5161 = (45.0, 126.0): cluster 2\n", + "Example n. 5162 = (45.0, 120.0): cluster 2\n", + "Example n. 5163 = (45.0, 119.0): cluster 2\n", + "Example n. 5164 = (45.0, 118.0): cluster 2\n", + "Example n. 5165 = (45.0, 117.0): cluster 2\n", + "Example n. 5166 = (45.0, 116.0): cluster 2\n", + "Example n. 5167 = (46.0, 496.0): cluster 2\n", + "Example n. 5168 = (46.0, 495.0): cluster 2\n", + "Example n. 5169 = (46.0, 494.0): cluster 2\n", + "Example n. 5170 = (46.0, 493.0): cluster 2\n", + "Example n. 5171 = (46.0, 492.0): cluster 2\n", + "Example n. 5172 = (46.0, 491.0): cluster 2\n", + "Example n. 5173 = (46.0, 490.0): cluster 2\n", + "Example n. 5174 = (46.0, 489.0): cluster 2\n", + "Example n. 5175 = (46.0, 488.0): cluster 2\n", + "Example n. 5176 = (46.0, 487.0): cluster 2\n", + "Example n. 5177 = (46.0, 486.0): cluster 2\n", + "Example n. 5178 = (46.0, 485.0): cluster 2\n", + "Example n. 5179 = (46.0, 484.0): cluster 2\n", + "Example n. 5180 = (46.0, 483.0): cluster 2\n", + "Example n. 5181 = (46.0, 482.0): cluster 2\n", + "Example n. 5182 = (46.0, 481.0): cluster 2\n", + "Example n. 5183 = (46.0, 480.0): cluster 2\n", + "Example n. 5184 = (46.0, 479.0): cluster 2\n", + "Example n. 5185 = (46.0, 478.0): cluster 2\n", + "Example n. 5186 = (46.0, 477.0): cluster 2\n", + "Example n. 5187 = (46.0, 476.0): cluster 2\n", + "Example n. 5188 = (46.0, 475.0): cluster 2\n", + "Example n. 5189 = (46.0, 474.0): cluster 2\n", + "Example n. 5190 = (46.0, 473.0): cluster 2\n", + "Example n. 5191 = (46.0, 472.0): cluster 2\n", + "Example n. 5192 = (46.0, 471.0): cluster 2\n", + "Example n. 5193 = (46.0, 470.0): cluster 2\n", + "Example n. 5194 = (46.0, 469.0): cluster 2\n", + "Example n. 5195 = (46.0, 468.0): cluster 2\n", + "Example n. 5196 = (46.0, 467.0): cluster 2\n", + "Example n. 5197 = (46.0, 466.0): cluster 2\n", + "Example n. 5198 = (46.0, 465.0): cluster 2\n", + "Example n. 5199 = (46.0, 464.0): cluster 2\n", + "Example n. 5200 = (46.0, 463.0): cluster 2\n", + "Example n. 5201 = (46.0, 462.0): cluster 2\n", + "Example n. 5202 = (46.0, 461.0): cluster 2\n", + "Example n. 5203 = (46.0, 460.0): cluster 2\n", + "Example n. 5204 = (46.0, 459.0): cluster 2\n", + "Example n. 5205 = (46.0, 458.0): cluster 2\n", + "Example n. 5206 = (46.0, 457.0): cluster 2\n", + "Example n. 5207 = (46.0, 456.0): cluster 2\n", + "Example n. 5208 = (46.0, 455.0): cluster 2\n", + "Example n. 5209 = (46.0, 454.0): cluster 2\n", + "Example n. 5210 = (46.0, 453.0): cluster 2\n", + "Example n. 5211 = (46.0, 452.0): cluster 2\n", + "Example n. 5212 = (46.0, 341.0): cluster 2\n", + "Example n. 5213 = (46.0, 340.0): cluster 2\n", + "Example n. 5214 = (46.0, 339.0): cluster 2\n", + "Example n. 5215 = (46.0, 338.0): cluster 2\n", + "Example n. 5216 = (46.0, 337.0): cluster 2\n", + "Example n. 5217 = (46.0, 336.0): cluster 2\n", + "Example n. 5218 = (46.0, 327.0): cluster 2\n", + "Example n. 5219 = (46.0, 326.0): cluster 2\n", + "Example n. 5220 = (46.0, 325.0): cluster 2\n", + "Example n. 5221 = (46.0, 324.0): cluster 2\n", + "Example n. 5222 = (46.0, 323.0): cluster 2\n", + "Example n. 5223 = (46.0, 322.0): cluster 2\n", + "Example n. 5224 = (46.0, 321.0): cluster 2\n", + "Example n. 5225 = (46.0, 320.0): cluster 2\n", + "Example n. 5226 = (46.0, 319.0): cluster 2\n", + "Example n. 5227 = (46.0, 318.0): cluster 2\n", + "Example n. 5228 = (46.0, 317.0): cluster 2\n", + "Example n. 5229 = (46.0, 316.0): cluster 2\n", + "Example n. 5230 = (46.0, 315.0): cluster 2\n", + "Example n. 5231 = (46.0, 314.0): cluster 2\n", + "Example n. 5232 = (46.0, 313.0): cluster 2\n", + "Example n. 5233 = (46.0, 312.0): cluster 2\n", + "Example n. 5234 = (46.0, 311.0): cluster 2\n", + "Example n. 5235 = (46.0, 310.0): cluster 2\n", + "Example n. 5236 = (46.0, 309.0): cluster 2\n", + "Example n. 5237 = (46.0, 308.0): cluster 2\n", + "Example n. 5238 = (46.0, 307.0): cluster 2\n", + "Example n. 5239 = (46.0, 306.0): cluster 2\n", + "Example n. 5240 = (46.0, 305.0): cluster 2\n", + "Example n. 5241 = (46.0, 304.0): cluster 2\n", + "Example n. 5242 = (46.0, 303.0): cluster 2\n", + "Example n. 5243 = (46.0, 302.0): cluster 2\n", + "Example n. 5244 = (46.0, 301.0): cluster 2\n", + "Example n. 5245 = (46.0, 300.0): cluster 2\n", + "Example n. 5246 = (46.0, 299.0): cluster 2\n", + "Example n. 5247 = (46.0, 298.0): cluster 2\n", + "Example n. 5248 = (46.0, 297.0): cluster 2\n", + "Example n. 5249 = (46.0, 296.0): cluster 2\n", + "Example n. 5250 = (46.0, 295.0): cluster 2\n", + "Example n. 5251 = (46.0, 294.0): cluster 2\n", + "Example n. 5252 = (46.0, 288.0): cluster 2\n", + "Example n. 5253 = (46.0, 287.0): cluster 2\n", + "Example n. 5254 = (46.0, 286.0): cluster 2\n", + "Example n. 5255 = (46.0, 285.0): cluster 2\n", + "Example n. 5256 = (46.0, 284.0): cluster 2\n", + "Example n. 5257 = (46.0, 283.0): cluster 2\n", + "Example n. 5258 = (46.0, 282.0): cluster 2\n", + "Example n. 5259 = (46.0, 281.0): cluster 2\n", + "Example n. 5260 = (46.0, 280.0): cluster 2\n", + "Example n. 5261 = (46.0, 279.0): cluster 2\n", + "Example n. 5262 = (46.0, 278.0): cluster 2\n", + "Example n. 5263 = (46.0, 277.0): cluster 2\n", + "Example n. 5264 = (46.0, 276.0): cluster 2\n", + "Example n. 5265 = (46.0, 275.0): cluster 2\n", + "Example n. 5266 = (46.0, 274.0): cluster 2\n", + "Example n. 5267 = (46.0, 273.0): cluster 2\n", + "Example n. 5268 = (46.0, 272.0): cluster 2\n", + "Example n. 5269 = (46.0, 271.0): cluster 2\n", + "Example n. 5270 = (46.0, 261.0): cluster 2\n", + "Example n. 5271 = (46.0, 260.0): cluster 2\n", + "Example n. 5272 = (46.0, 259.0): cluster 2\n", + "Example n. 5273 = (46.0, 258.0): cluster 2\n", + "Example n. 5274 = (46.0, 257.0): cluster 2\n", + "Example n. 5275 = (46.0, 256.0): cluster 2\n", + "Example n. 5276 = (46.0, 255.0): cluster 2\n", + "Example n. 5277 = (46.0, 254.0): cluster 2\n", + "Example n. 5278 = (46.0, 253.0): cluster 2\n", + "Example n. 5279 = (46.0, 252.0): cluster 2\n", + "Example n. 5280 = (46.0, 251.0): cluster 2\n", + "Example n. 5281 = (46.0, 250.0): cluster 2\n", + "Example n. 5282 = (46.0, 249.0): cluster 2\n", + "Example n. 5283 = (46.0, 248.0): cluster 2\n", + "Example n. 5284 = (46.0, 245.0): cluster 2\n", + "Example n. 5285 = (46.0, 244.0): cluster 2\n", + "Example n. 5286 = (46.0, 243.0): cluster 2\n", + "Example n. 5287 = (46.0, 242.0): cluster 2\n", + "Example n. 5288 = (46.0, 241.0): cluster 2\n", + "Example n. 5289 = (46.0, 240.0): cluster 2\n", + "Example n. 5290 = (46.0, 239.0): cluster 2\n", + "Example n. 5291 = (46.0, 238.0): cluster 2\n", + "Example n. 5292 = (46.0, 237.0): cluster 2\n", + "Example n. 5293 = (46.0, 236.0): cluster 2\n", + "Example n. 5294 = (46.0, 235.0): cluster 2\n", + "Example n. 5295 = (46.0, 228.0): cluster 2\n", + "Example n. 5296 = (46.0, 227.0): cluster 2\n", + "Example n. 5297 = (46.0, 226.0): cluster 2\n", + "Example n. 5298 = (46.0, 225.0): cluster 2\n", + "Example n. 5299 = (46.0, 224.0): cluster 2\n", + "Example n. 5300 = (46.0, 223.0): cluster 2\n", + "Example n. 5301 = (46.0, 220.0): cluster 2\n", + "Example n. 5302 = (46.0, 219.0): cluster 2\n", + "Example n. 5303 = (46.0, 218.0): cluster 2\n", + "Example n. 5304 = (46.0, 217.0): cluster 2\n", + "Example n. 5305 = (46.0, 216.0): cluster 2\n", + "Example n. 5306 = (46.0, 208.0): cluster 2\n", + "Example n. 5307 = (46.0, 207.0): cluster 2\n", + "Example n. 5308 = (46.0, 206.0): cluster 2\n", + "Example n. 5309 = (46.0, 202.0): cluster 2\n", + "Example n. 5310 = (46.0, 201.0): cluster 2\n", + "Example n. 5311 = (46.0, 200.0): cluster 2\n", + "Example n. 5312 = (46.0, 199.0): cluster 2\n", + "Example n. 5313 = (46.0, 198.0): cluster 2\n", + "Example n. 5314 = (46.0, 188.0): cluster 2\n", + "Example n. 5315 = (46.0, 187.0): cluster 2\n", + "Example n. 5316 = (46.0, 186.0): cluster 2\n", + "Example n. 5317 = (46.0, 185.0): cluster 2\n", + "Example n. 5318 = (46.0, 184.0): cluster 2\n", + "Example n. 5319 = (46.0, 183.0): cluster 2\n", + "Example n. 5320 = (46.0, 182.0): cluster 2\n", + "Example n. 5321 = (46.0, 181.0): cluster 2\n", + "Example n. 5322 = (46.0, 179.0): cluster 2\n", + "Example n. 5323 = (46.0, 178.0): cluster 2\n", + "Example n. 5324 = (46.0, 177.0): cluster 2\n", + "Example n. 5325 = (46.0, 176.0): cluster 2\n", + "Example n. 5326 = (46.0, 175.0): cluster 2\n", + "Example n. 5327 = (46.0, 174.0): cluster 2\n", + "Example n. 5328 = (46.0, 173.0): cluster 2\n", + "Example n. 5329 = (46.0, 172.0): cluster 2\n", + "Example n. 5330 = (46.0, 171.0): cluster 2\n", + "Example n. 5331 = (46.0, 170.0): cluster 2\n", + "Example n. 5332 = (46.0, 169.0): cluster 2\n", + "Example n. 5333 = (46.0, 168.0): cluster 2\n", + "Example n. 5334 = (46.0, 167.0): cluster 2\n", + "Example n. 5335 = (46.0, 166.0): cluster 2\n", + "Example n. 5336 = (46.0, 165.0): cluster 2\n", + "Example n. 5337 = (46.0, 161.0): cluster 2\n", + "Example n. 5338 = (46.0, 160.0): cluster 2\n", + "Example n. 5339 = (46.0, 159.0): cluster 2\n", + "Example n. 5340 = (46.0, 158.0): cluster 2\n", + "Example n. 5341 = (46.0, 157.0): cluster 2\n", + "Example n. 5342 = (46.0, 156.0): cluster 2\n", + "Example n. 5343 = (46.0, 155.0): cluster 2\n", + "Example n. 5344 = (46.0, 151.0): cluster 2\n", + "Example n. 5345 = (46.0, 150.0): cluster 2\n", + "Example n. 5346 = (46.0, 148.0): cluster 2\n", + "Example n. 5347 = (46.0, 147.0): cluster 2\n", + "Example n. 5348 = (46.0, 146.0): cluster 2\n", + "Example n. 5349 = (46.0, 145.0): cluster 2\n", + "Example n. 5350 = (46.0, 140.0): cluster 2\n", + "Example n. 5351 = (46.0, 139.0): cluster 2\n", + "Example n. 5352 = (46.0, 138.0): cluster 2\n", + "Example n. 5353 = (46.0, 137.0): cluster 2\n", + "Example n. 5354 = (46.0, 136.0): cluster 2\n", + "Example n. 5355 = (46.0, 135.0): cluster 2\n", + "Example n. 5356 = (46.0, 134.0): cluster 2\n", + "Example n. 5357 = (46.0, 133.0): cluster 2\n", + "Example n. 5358 = (46.0, 132.0): cluster 2\n", + "Example n. 5359 = (46.0, 131.0): cluster 2\n", + "Example n. 5360 = (46.0, 130.0): cluster 2\n", + "Example n. 5361 = (46.0, 129.0): cluster 2\n", + "Example n. 5362 = (46.0, 128.0): cluster 2\n", + "Example n. 5363 = (46.0, 127.0): cluster 2\n", + "Example n. 5364 = (46.0, 120.0): cluster 2\n", + "Example n. 5365 = (46.0, 119.0): cluster 2\n", + "Example n. 5366 = (46.0, 118.0): cluster 2\n", + "Example n. 5367 = (46.0, 117.0): cluster 2\n", + "Example n. 5368 = (46.0, 116.0): cluster 2\n", + "Example n. 5369 = (47.0, 495.0): cluster 2\n", + "Example n. 5370 = (47.0, 494.0): cluster 2\n", + "Example n. 5371 = (47.0, 493.0): cluster 2\n", + "Example n. 5372 = (47.0, 492.0): cluster 2\n", + "Example n. 5373 = (47.0, 491.0): cluster 2\n", + "Example n. 5374 = (47.0, 490.0): cluster 2\n", + "Example n. 5375 = (47.0, 489.0): cluster 2\n", + "Example n. 5376 = (47.0, 488.0): cluster 2\n", + "Example n. 5377 = (47.0, 487.0): cluster 2\n", + "Example n. 5378 = (47.0, 486.0): cluster 2\n", + "Example n. 5379 = (47.0, 485.0): cluster 2\n", + "Example n. 5380 = (47.0, 484.0): cluster 2\n", + "Example n. 5381 = (47.0, 483.0): cluster 2\n", + "Example n. 5382 = (47.0, 482.0): cluster 2\n", + "Example n. 5383 = (47.0, 481.0): cluster 2\n", + "Example n. 5384 = (47.0, 480.0): cluster 2\n", + "Example n. 5385 = (47.0, 479.0): cluster 2\n", + "Example n. 5386 = (47.0, 478.0): cluster 2\n", + "Example n. 5387 = (47.0, 477.0): cluster 2\n", + "Example n. 5388 = (47.0, 476.0): cluster 2\n", + "Example n. 5389 = (47.0, 475.0): cluster 2\n", + "Example n. 5390 = (47.0, 474.0): cluster 2\n", + "Example n. 5391 = (47.0, 473.0): cluster 2\n", + "Example n. 5392 = (47.0, 472.0): cluster 2\n", + "Example n. 5393 = (47.0, 471.0): cluster 2\n", + "Example n. 5394 = (47.0, 470.0): cluster 2\n", + "Example n. 5395 = (47.0, 469.0): cluster 2\n", + "Example n. 5396 = (47.0, 468.0): cluster 2\n", + "Example n. 5397 = (47.0, 467.0): cluster 2\n", + "Example n. 5398 = (47.0, 466.0): cluster 2\n", + "Example n. 5399 = (47.0, 465.0): cluster 2\n", + "Example n. 5400 = (47.0, 464.0): cluster 2\n", + "Example n. 5401 = (47.0, 463.0): cluster 2\n", + "Example n. 5402 = (47.0, 462.0): cluster 2\n", + "Example n. 5403 = (47.0, 461.0): cluster 2\n", + "Example n. 5404 = (47.0, 460.0): cluster 2\n", + "Example n. 5405 = (47.0, 459.0): cluster 2\n", + "Example n. 5406 = (47.0, 458.0): cluster 2\n", + "Example n. 5407 = (47.0, 457.0): cluster 2\n", + "Example n. 5408 = (47.0, 456.0): cluster 2\n", + "Example n. 5409 = (47.0, 455.0): cluster 2\n", + "Example n. 5410 = (47.0, 454.0): cluster 2\n", + "Example n. 5411 = (47.0, 453.0): cluster 2\n", + "Example n. 5412 = (47.0, 452.0): cluster 2\n", + "Example n. 5413 = (47.0, 341.0): cluster 2\n", + "Example n. 5414 = (47.0, 340.0): cluster 2\n", + "Example n. 5415 = (47.0, 339.0): cluster 2\n", + "Example n. 5416 = (47.0, 338.0): cluster 2\n", + "Example n. 5417 = (47.0, 337.0): cluster 2\n", + "Example n. 5418 = (47.0, 327.0): cluster 2\n", + "Example n. 5419 = (47.0, 326.0): cluster 2\n", + "Example n. 5420 = (47.0, 325.0): cluster 2\n", + "Example n. 5421 = (47.0, 324.0): cluster 2\n", + "Example n. 5422 = (47.0, 323.0): cluster 2\n", + "Example n. 5423 = (47.0, 322.0): cluster 2\n", + "Example n. 5424 = (47.0, 321.0): cluster 2\n", + "Example n. 5425 = (47.0, 320.0): cluster 2\n", + "Example n. 5426 = (47.0, 319.0): cluster 2\n", + "Example n. 5427 = (47.0, 318.0): cluster 2\n", + "Example n. 5428 = (47.0, 317.0): cluster 2\n", + "Example n. 5429 = (47.0, 316.0): cluster 2\n", + "Example n. 5430 = (47.0, 315.0): cluster 2\n", + "Example n. 5431 = (47.0, 314.0): cluster 2\n", + "Example n. 5432 = (47.0, 313.0): cluster 2\n", + "Example n. 5433 = (47.0, 312.0): cluster 2\n", + "Example n. 5434 = (47.0, 311.0): cluster 2\n", + "Example n. 5435 = (47.0, 310.0): cluster 2\n", + "Example n. 5436 = (47.0, 308.0): cluster 2\n", + "Example n. 5437 = (47.0, 307.0): cluster 2\n", + "Example n. 5438 = (47.0, 306.0): cluster 2\n", + "Example n. 5439 = (47.0, 305.0): cluster 2\n", + "Example n. 5440 = (47.0, 304.0): cluster 2\n", + "Example n. 5441 = (47.0, 303.0): cluster 2\n", + "Example n. 5442 = (47.0, 302.0): cluster 2\n", + "Example n. 5443 = (47.0, 301.0): cluster 2\n", + "Example n. 5444 = (47.0, 300.0): cluster 2\n", + "Example n. 5445 = (47.0, 299.0): cluster 2\n", + "Example n. 5446 = (47.0, 298.0): cluster 2\n", + "Example n. 5447 = (47.0, 297.0): cluster 2\n", + "Example n. 5448 = (47.0, 296.0): cluster 2\n", + "Example n. 5449 = (47.0, 295.0): cluster 2\n", + "Example n. 5450 = (47.0, 294.0): cluster 2\n", + "Example n. 5451 = (47.0, 287.0): cluster 2\n", + "Example n. 5452 = (47.0, 286.0): cluster 2\n", + "Example n. 5453 = (47.0, 285.0): cluster 2\n", + "Example n. 5454 = (47.0, 284.0): cluster 2\n", + "Example n. 5455 = (47.0, 283.0): cluster 2\n", + "Example n. 5456 = (47.0, 282.0): cluster 2\n", + "Example n. 5457 = (47.0, 281.0): cluster 2\n", + "Example n. 5458 = (47.0, 280.0): cluster 2\n", + "Example n. 5459 = (47.0, 279.0): cluster 2\n", + "Example n. 5460 = (47.0, 278.0): cluster 2\n", + "Example n. 5461 = (47.0, 277.0): cluster 2\n", + "Example n. 5462 = (47.0, 276.0): cluster 2\n", + "Example n. 5463 = (47.0, 275.0): cluster 2\n", + "Example n. 5464 = (47.0, 274.0): cluster 2\n", + "Example n. 5465 = (47.0, 273.0): cluster 2\n", + "Example n. 5466 = (47.0, 272.0): cluster 2\n", + "Example n. 5467 = (47.0, 271.0): cluster 2\n", + "Example n. 5468 = (47.0, 264.0): cluster 2\n", + "Example n. 5469 = (47.0, 263.0): cluster 2\n", + "Example n. 5470 = (47.0, 262.0): cluster 2\n", + "Example n. 5471 = (47.0, 261.0): cluster 2\n", + "Example n. 5472 = (47.0, 260.0): cluster 2\n", + "Example n. 5473 = (47.0, 259.0): cluster 2\n", + "Example n. 5474 = (47.0, 258.0): cluster 2\n", + "Example n. 5475 = (47.0, 257.0): cluster 2\n", + "Example n. 5476 = (47.0, 256.0): cluster 2\n", + "Example n. 5477 = (47.0, 255.0): cluster 2\n", + "Example n. 5478 = (47.0, 254.0): cluster 2\n", + "Example n. 5479 = (47.0, 253.0): cluster 2\n", + "Example n. 5480 = (47.0, 252.0): cluster 2\n", + "Example n. 5481 = (47.0, 251.0): cluster 2\n", + "Example n. 5482 = (47.0, 250.0): cluster 2\n", + "Example n. 5483 = (47.0, 249.0): cluster 2\n", + "Example n. 5484 = (47.0, 248.0): cluster 2\n", + "Example n. 5485 = (47.0, 247.0): cluster 2\n", + "Example n. 5486 = (47.0, 246.0): cluster 2\n", + "Example n. 5487 = (47.0, 245.0): cluster 2\n", + "Example n. 5488 = (47.0, 244.0): cluster 2\n", + "Example n. 5489 = (47.0, 243.0): cluster 2\n", + "Example n. 5490 = (47.0, 242.0): cluster 2\n", + "Example n. 5491 = (47.0, 241.0): cluster 2\n", + "Example n. 5492 = (47.0, 240.0): cluster 2\n", + "Example n. 5493 = (47.0, 239.0): cluster 2\n", + "Example n. 5494 = (47.0, 238.0): cluster 2\n", + "Example n. 5495 = (47.0, 237.0): cluster 2\n", + "Example n. 5496 = (47.0, 236.0): cluster 2\n", + "Example n. 5497 = (47.0, 227.0): cluster 2\n", + "Example n. 5498 = (47.0, 226.0): cluster 2\n", + "Example n. 5499 = (47.0, 225.0): cluster 2\n", + "Example n. 5500 = (47.0, 224.0): cluster 2\n", + "Example n. 5501 = (47.0, 223.0): cluster 2\n", + "Example n. 5502 = (47.0, 220.0): cluster 2\n", + "Example n. 5503 = (47.0, 219.0): cluster 2\n", + "Example n. 5504 = (47.0, 218.0): cluster 2\n", + "Example n. 5505 = (47.0, 217.0): cluster 2\n", + "Example n. 5506 = (47.0, 216.0): cluster 2\n", + "Example n. 5507 = (47.0, 215.0): cluster 2\n", + "Example n. 5508 = (47.0, 214.0): cluster 2\n", + "Example n. 5509 = (47.0, 213.0): cluster 2\n", + "Example n. 5510 = (47.0, 212.0): cluster 2\n", + "Example n. 5511 = (47.0, 210.0): cluster 2\n", + "Example n. 5512 = (47.0, 209.0): cluster 2\n", + "Example n. 5513 = (47.0, 201.0): cluster 2\n", + "Example n. 5514 = (47.0, 200.0): cluster 2\n", + "Example n. 5515 = (47.0, 199.0): cluster 2\n", + "Example n. 5516 = (47.0, 194.0): cluster 2\n", + "Example n. 5517 = (47.0, 193.0): cluster 2\n", + "Example n. 5518 = (47.0, 192.0): cluster 2\n", + "Example n. 5519 = (47.0, 188.0): cluster 2\n", + "Example n. 5520 = (47.0, 187.0): cluster 2\n", + "Example n. 5521 = (47.0, 186.0): cluster 2\n", + "Example n. 5522 = (47.0, 185.0): cluster 2\n", + "Example n. 5523 = (47.0, 184.0): cluster 2\n", + "Example n. 5524 = (47.0, 183.0): cluster 2\n", + "Example n. 5525 = (47.0, 182.0): cluster 2\n", + "Example n. 5526 = (47.0, 181.0): cluster 2\n", + "Example n. 5527 = (47.0, 179.0): cluster 2\n", + "Example n. 5528 = (47.0, 178.0): cluster 2\n", + "Example n. 5529 = (47.0, 177.0): cluster 2\n", + "Example n. 5530 = (47.0, 176.0): cluster 2\n", + "Example n. 5531 = (47.0, 175.0): cluster 2\n", + "Example n. 5532 = (47.0, 174.0): cluster 2\n", + "Example n. 5533 = (47.0, 173.0): cluster 2\n", + "Example n. 5534 = (47.0, 172.0): cluster 2\n", + "Example n. 5535 = (47.0, 171.0): cluster 2\n", + "Example n. 5536 = (47.0, 170.0): cluster 2\n", + "Example n. 5537 = (47.0, 169.0): cluster 2\n", + "Example n. 5538 = (47.0, 168.0): cluster 2\n", + "Example n. 5539 = (47.0, 167.0): cluster 2\n", + "Example n. 5540 = (47.0, 166.0): cluster 2\n", + "Example n. 5541 = (47.0, 165.0): cluster 2\n", + "Example n. 5542 = (47.0, 164.0): cluster 2\n", + "Example n. 5543 = (47.0, 163.0): cluster 2\n", + "Example n. 5544 = (47.0, 161.0): cluster 2\n", + "Example n. 5545 = (47.0, 160.0): cluster 2\n", + "Example n. 5546 = (47.0, 159.0): cluster 2\n", + "Example n. 5547 = (47.0, 158.0): cluster 2\n", + "Example n. 5548 = (47.0, 157.0): cluster 2\n", + "Example n. 5549 = (47.0, 156.0): cluster 2\n", + "Example n. 5550 = (47.0, 155.0): cluster 2\n", + "Example n. 5551 = (47.0, 149.0): cluster 2\n", + "Example n. 5552 = (47.0, 148.0): cluster 2\n", + "Example n. 5553 = (47.0, 147.0): cluster 2\n", + "Example n. 5554 = (47.0, 146.0): cluster 2\n", + "Example n. 5555 = (47.0, 145.0): cluster 2\n", + "Example n. 5556 = (47.0, 144.0): cluster 2\n", + "Example n. 5557 = (47.0, 143.0): cluster 2\n", + "Example n. 5558 = (47.0, 142.0): cluster 2\n", + "Example n. 5559 = (47.0, 141.0): cluster 2\n", + "Example n. 5560 = (47.0, 140.0): cluster 2\n", + "Example n. 5561 = (47.0, 139.0): cluster 2\n", + "Example n. 5562 = (47.0, 138.0): cluster 2\n", + "Example n. 5563 = (47.0, 137.0): cluster 2\n", + "Example n. 5564 = (47.0, 136.0): cluster 2\n", + "Example n. 5565 = (47.0, 135.0): cluster 2\n", + "Example n. 5566 = (47.0, 134.0): cluster 2\n", + "Example n. 5567 = (47.0, 133.0): cluster 2\n", + "Example n. 5568 = (47.0, 132.0): cluster 2\n", + "Example n. 5569 = (47.0, 131.0): cluster 2\n", + "Example n. 5570 = (47.0, 130.0): cluster 2\n", + "Example n. 5571 = (47.0, 129.0): cluster 2\n", + "Example n. 5572 = (47.0, 128.0): cluster 2\n", + "Example n. 5573 = (47.0, 127.0): cluster 2\n", + "Example n. 5574 = (47.0, 119.0): cluster 2\n", + "Example n. 5575 = (47.0, 118.0): cluster 2\n", + "Example n. 5576 = (47.0, 117.0): cluster 2\n", + "Example n. 5577 = (47.0, 116.0): cluster 2\n", + "Example n. 5578 = (48.0, 494.0): cluster 2\n", + "Example n. 5579 = (48.0, 493.0): cluster 2\n", + "Example n. 5580 = (48.0, 492.0): cluster 2\n", + "Example n. 5581 = (48.0, 491.0): cluster 2\n", + "Example n. 5582 = (48.0, 490.0): cluster 2\n", + "Example n. 5583 = (48.0, 489.0): cluster 2\n", + "Example n. 5584 = (48.0, 488.0): cluster 2\n", + "Example n. 5585 = (48.0, 487.0): cluster 2\n", + "Example n. 5586 = (48.0, 486.0): cluster 2\n", + "Example n. 5587 = (48.0, 485.0): cluster 2\n", + "Example n. 5588 = (48.0, 484.0): cluster 2\n", + "Example n. 5589 = (48.0, 483.0): cluster 2\n", + "Example n. 5590 = (48.0, 482.0): cluster 2\n", + "Example n. 5591 = (48.0, 481.0): cluster 2\n", + "Example n. 5592 = (48.0, 480.0): cluster 2\n", + "Example n. 5593 = (48.0, 479.0): cluster 2\n", + "Example n. 5594 = (48.0, 478.0): cluster 2\n", + "Example n. 5595 = (48.0, 477.0): cluster 2\n", + "Example n. 5596 = (48.0, 476.0): cluster 2\n", + "Example n. 5597 = (48.0, 475.0): cluster 2\n", + "Example n. 5598 = (48.0, 474.0): cluster 2\n", + "Example n. 5599 = (48.0, 473.0): cluster 2\n", + "Example n. 5600 = (48.0, 472.0): cluster 2\n", + "Example n. 5601 = (48.0, 471.0): cluster 2\n", + "Example n. 5602 = (48.0, 470.0): cluster 2\n", + "Example n. 5603 = (48.0, 469.0): cluster 2\n", + "Example n. 5604 = (48.0, 468.0): cluster 2\n", + "Example n. 5605 = (48.0, 467.0): cluster 2\n", + "Example n. 5606 = (48.0, 466.0): cluster 2\n", + "Example n. 5607 = (48.0, 465.0): cluster 2\n", + "Example n. 5608 = (48.0, 464.0): cluster 2\n", + "Example n. 5609 = (48.0, 463.0): cluster 2\n", + "Example n. 5610 = (48.0, 462.0): cluster 2\n", + "Example n. 5611 = (48.0, 461.0): cluster 2\n", + "Example n. 5612 = (48.0, 460.0): cluster 2\n", + "Example n. 5613 = (48.0, 459.0): cluster 2\n", + "Example n. 5614 = (48.0, 458.0): cluster 2\n", + "Example n. 5615 = (48.0, 457.0): cluster 2\n", + "Example n. 5616 = (48.0, 456.0): cluster 2\n", + "Example n. 5617 = (48.0, 455.0): cluster 2\n", + "Example n. 5618 = (48.0, 454.0): cluster 2\n", + "Example n. 5619 = (48.0, 453.0): cluster 2\n", + "Example n. 5620 = (48.0, 452.0): cluster 2\n", + "Example n. 5621 = (48.0, 339.0): cluster 2\n", + "Example n. 5622 = (48.0, 338.0): cluster 2\n", + "Example n. 5623 = (48.0, 327.0): cluster 2\n", + "Example n. 5624 = (48.0, 326.0): cluster 2\n", + "Example n. 5625 = (48.0, 325.0): cluster 2\n", + "Example n. 5626 = (48.0, 324.0): cluster 2\n", + "Example n. 5627 = (48.0, 323.0): cluster 2\n", + "Example n. 5628 = (48.0, 322.0): cluster 2\n", + "Example n. 5629 = (48.0, 321.0): cluster 2\n", + "Example n. 5630 = (48.0, 320.0): cluster 2\n", + "Example n. 5631 = (48.0, 319.0): cluster 2\n", + "Example n. 5632 = (48.0, 318.0): cluster 2\n", + "Example n. 5633 = (48.0, 316.0): cluster 2\n", + "Example n. 5634 = (48.0, 315.0): cluster 2\n", + "Example n. 5635 = (48.0, 314.0): cluster 2\n", + "Example n. 5636 = (48.0, 313.0): cluster 2\n", + "Example n. 5637 = (48.0, 308.0): cluster 2\n", + "Example n. 5638 = (48.0, 307.0): cluster 2\n", + "Example n. 5639 = (48.0, 306.0): cluster 2\n", + "Example n. 5640 = (48.0, 305.0): cluster 2\n", + "Example n. 5641 = (48.0, 304.0): cluster 2\n", + "Example n. 5642 = (48.0, 303.0): cluster 2\n", + "Example n. 5643 = (48.0, 302.0): cluster 2\n", + "Example n. 5644 = (48.0, 301.0): cluster 2\n", + "Example n. 5645 = (48.0, 300.0): cluster 2\n", + "Example n. 5646 = (48.0, 299.0): cluster 2\n", + "Example n. 5647 = (48.0, 298.0): cluster 2\n", + "Example n. 5648 = (48.0, 297.0): cluster 2\n", + "Example n. 5649 = (48.0, 296.0): cluster 2\n", + "Example n. 5650 = (48.0, 295.0): cluster 2\n", + "Example n. 5651 = (48.0, 294.0): cluster 2\n", + "Example n. 5652 = (48.0, 287.0): cluster 2\n", + "Example n. 5653 = (48.0, 286.0): cluster 2\n", + "Example n. 5654 = (48.0, 285.0): cluster 2\n", + "Example n. 5655 = (48.0, 284.0): cluster 2\n", + "Example n. 5656 = (48.0, 283.0): cluster 2\n", + "Example n. 5657 = (48.0, 282.0): cluster 2\n", + "Example n. 5658 = (48.0, 281.0): cluster 2\n", + "Example n. 5659 = (48.0, 280.0): cluster 2\n", + "Example n. 5660 = (48.0, 279.0): cluster 2\n", + "Example n. 5661 = (48.0, 278.0): cluster 2\n", + "Example n. 5662 = (48.0, 277.0): cluster 2\n", + "Example n. 5663 = (48.0, 276.0): cluster 2\n", + "Example n. 5664 = (48.0, 275.0): cluster 2\n", + "Example n. 5665 = (48.0, 274.0): cluster 2\n", + "Example n. 5666 = (48.0, 273.0): cluster 2\n", + "Example n. 5667 = (48.0, 272.0): cluster 2\n", + "Example n. 5668 = (48.0, 271.0): cluster 2\n", + "Example n. 5669 = (48.0, 270.0): cluster 2\n", + "Example n. 5670 = (48.0, 265.0): cluster 2\n", + "Example n. 5671 = (48.0, 264.0): cluster 2\n", + "Example n. 5672 = (48.0, 263.0): cluster 2\n", + "Example n. 5673 = (48.0, 262.0): cluster 2\n", + "Example n. 5674 = (48.0, 261.0): cluster 2\n", + "Example n. 5675 = (48.0, 260.0): cluster 2\n", + "Example n. 5676 = (48.0, 259.0): cluster 2\n", + "Example n. 5677 = (48.0, 258.0): cluster 2\n", + "Example n. 5678 = (48.0, 257.0): cluster 2\n", + "Example n. 5679 = (48.0, 256.0): cluster 2\n", + "Example n. 5680 = (48.0, 255.0): cluster 2\n", + "Example n. 5681 = (48.0, 254.0): cluster 2\n", + "Example n. 5682 = (48.0, 253.0): cluster 2\n", + "Example n. 5683 = (48.0, 252.0): cluster 2\n", + "Example n. 5684 = (48.0, 251.0): cluster 2\n", + "Example n. 5685 = (48.0, 250.0): cluster 2\n", + "Example n. 5686 = (48.0, 249.0): cluster 2\n", + "Example n. 5687 = (48.0, 248.0): cluster 2\n", + "Example n. 5688 = (48.0, 247.0): cluster 2\n", + "Example n. 5689 = (48.0, 246.0): cluster 2\n", + "Example n. 5690 = (48.0, 245.0): cluster 2\n", + "Example n. 5691 = (48.0, 244.0): cluster 2\n", + "Example n. 5692 = (48.0, 243.0): cluster 2\n", + "Example n. 5693 = (48.0, 242.0): cluster 2\n", + "Example n. 5694 = (48.0, 241.0): cluster 2\n", + "Example n. 5695 = (48.0, 240.0): cluster 2\n", + "Example n. 5696 = (48.0, 239.0): cluster 2\n", + "Example n. 5697 = (48.0, 238.0): cluster 2\n", + "Example n. 5698 = (48.0, 237.0): cluster 2\n", + "Example n. 5699 = (48.0, 236.0): cluster 2\n", + "Example n. 5700 = (48.0, 229.0): cluster 2\n", + "Example n. 5701 = (48.0, 228.0): cluster 2\n", + "Example n. 5702 = (48.0, 227.0): cluster 2\n", + "Example n. 5703 = (48.0, 226.0): cluster 2\n", + "Example n. 5704 = (48.0, 225.0): cluster 2\n", + "Example n. 5705 = (48.0, 224.0): cluster 2\n", + "Example n. 5706 = (48.0, 223.0): cluster 2\n", + "Example n. 5707 = (48.0, 219.0): cluster 2\n", + "Example n. 5708 = (48.0, 218.0): cluster 2\n", + "Example n. 5709 = (48.0, 217.0): cluster 2\n", + "Example n. 5710 = (48.0, 216.0): cluster 2\n", + "Example n. 5711 = (48.0, 215.0): cluster 2\n", + "Example n. 5712 = (48.0, 214.0): cluster 2\n", + "Example n. 5713 = (48.0, 213.0): cluster 2\n", + "Example n. 5714 = (48.0, 212.0): cluster 2\n", + "Example n. 5715 = (48.0, 211.0): cluster 2\n", + "Example n. 5716 = (48.0, 210.0): cluster 2\n", + "Example n. 5717 = (48.0, 209.0): cluster 2\n", + "Example n. 5718 = (48.0, 208.0): cluster 2\n", + "Example n. 5719 = (48.0, 195.0): cluster 2\n", + "Example n. 5720 = (48.0, 194.0): cluster 2\n", + "Example n. 5721 = (48.0, 193.0): cluster 2\n", + "Example n. 5722 = (48.0, 192.0): cluster 2\n", + "Example n. 5723 = (48.0, 191.0): cluster 2\n", + "Example n. 5724 = (48.0, 188.0): cluster 2\n", + "Example n. 5725 = (48.0, 187.0): cluster 2\n", + "Example n. 5726 = (48.0, 186.0): cluster 2\n", + "Example n. 5727 = (48.0, 185.0): cluster 2\n", + "Example n. 5728 = (48.0, 184.0): cluster 2\n", + "Example n. 5729 = (48.0, 183.0): cluster 2\n", + "Example n. 5730 = (48.0, 182.0): cluster 2\n", + "Example n. 5731 = (48.0, 181.0): cluster 2\n", + "Example n. 5732 = (48.0, 180.0): cluster 2\n", + "Example n. 5733 = (48.0, 179.0): cluster 2\n", + "Example n. 5734 = (48.0, 178.0): cluster 2\n", + "Example n. 5735 = (48.0, 177.0): cluster 2\n", + "Example n. 5736 = (48.0, 176.0): cluster 2\n", + "Example n. 5737 = (48.0, 175.0): cluster 2\n", + "Example n. 5738 = (48.0, 174.0): cluster 2\n", + "Example n. 5739 = (48.0, 173.0): cluster 2\n", + "Example n. 5740 = (48.0, 172.0): cluster 2\n", + "Example n. 5741 = (48.0, 171.0): cluster 2\n", + "Example n. 5742 = (48.0, 170.0): cluster 2\n", + "Example n. 5743 = (48.0, 169.0): cluster 2\n", + "Example n. 5744 = (48.0, 168.0): cluster 2\n", + "Example n. 5745 = (48.0, 167.0): cluster 2\n", + "Example n. 5746 = (48.0, 166.0): cluster 2\n", + "Example n. 5747 = (48.0, 165.0): cluster 2\n", + "Example n. 5748 = (48.0, 164.0): cluster 2\n", + "Example n. 5749 = (48.0, 163.0): cluster 2\n", + "Example n. 5750 = (48.0, 162.0): cluster 2\n", + "Example n. 5751 = (48.0, 161.0): cluster 2\n", + "Example n. 5752 = (48.0, 160.0): cluster 2\n", + "Example n. 5753 = (48.0, 159.0): cluster 2\n", + "Example n. 5754 = (48.0, 158.0): cluster 2\n", + "Example n. 5755 = (48.0, 157.0): cluster 2\n", + "Example n. 5756 = (48.0, 156.0): cluster 2\n", + "Example n. 5757 = (48.0, 149.0): cluster 2\n", + "Example n. 5758 = (48.0, 148.0): cluster 2\n", + "Example n. 5759 = (48.0, 147.0): cluster 2\n", + "Example n. 5760 = (48.0, 146.0): cluster 2\n", + "Example n. 5761 = (48.0, 145.0): cluster 2\n", + "Example n. 5762 = (48.0, 144.0): cluster 2\n", + "Example n. 5763 = (48.0, 143.0): cluster 2\n", + "Example n. 5764 = (48.0, 142.0): cluster 2\n", + "Example n. 5765 = (48.0, 141.0): cluster 2\n", + "Example n. 5766 = (48.0, 140.0): cluster 2\n", + "Example n. 5767 = (48.0, 139.0): cluster 2\n", + "Example n. 5768 = (48.0, 138.0): cluster 2\n", + "Example n. 5769 = (48.0, 137.0): cluster 2\n", + "Example n. 5770 = (48.0, 136.0): cluster 2\n", + "Example n. 5771 = (48.0, 135.0): cluster 2\n", + "Example n. 5772 = (48.0, 134.0): cluster 2\n", + "Example n. 5773 = (48.0, 133.0): cluster 2\n", + "Example n. 5774 = (48.0, 132.0): cluster 2\n", + "Example n. 5775 = (48.0, 131.0): cluster 2\n", + "Example n. 5776 = (48.0, 130.0): cluster 2\n", + "Example n. 5777 = (48.0, 129.0): cluster 2\n", + "Example n. 5778 = (48.0, 128.0): cluster 2\n", + "Example n. 5779 = (48.0, 127.0): cluster 2\n", + "Example n. 5780 = (49.0, 494.0): cluster 2\n", + "Example n. 5781 = (49.0, 493.0): cluster 2\n", + "Example n. 5782 = (49.0, 492.0): cluster 2\n", + "Example n. 5783 = (49.0, 491.0): cluster 2\n", + "Example n. 5784 = (49.0, 490.0): cluster 2\n", + "Example n. 5785 = (49.0, 489.0): cluster 2\n", + "Example n. 5786 = (49.0, 488.0): cluster 2\n", + "Example n. 5787 = (49.0, 487.0): cluster 2\n", + "Example n. 5788 = (49.0, 486.0): cluster 2\n", + "Example n. 5789 = (49.0, 485.0): cluster 2\n", + "Example n. 5790 = (49.0, 484.0): cluster 2\n", + "Example n. 5791 = (49.0, 483.0): cluster 2\n", + "Example n. 5792 = (49.0, 482.0): cluster 2\n", + "Example n. 5793 = (49.0, 481.0): cluster 2\n", + "Example n. 5794 = (49.0, 480.0): cluster 2\n", + "Example n. 5795 = (49.0, 479.0): cluster 2\n", + "Example n. 5796 = (49.0, 478.0): cluster 2\n", + "Example n. 5797 = (49.0, 477.0): cluster 2\n", + "Example n. 5798 = (49.0, 476.0): cluster 2\n", + "Example n. 5799 = (49.0, 475.0): cluster 2\n", + "Example n. 5800 = (49.0, 474.0): cluster 2\n", + "Example n. 5801 = (49.0, 473.0): cluster 2\n", + "Example n. 5802 = (49.0, 472.0): cluster 2\n", + "Example n. 5803 = (49.0, 471.0): cluster 2\n", + "Example n. 5804 = (49.0, 470.0): cluster 2\n", + "Example n. 5805 = (49.0, 469.0): cluster 2\n", + "Example n. 5806 = (49.0, 468.0): cluster 2\n", + "Example n. 5807 = (49.0, 467.0): cluster 2\n", + "Example n. 5808 = (49.0, 466.0): cluster 2\n", + "Example n. 5809 = (49.0, 465.0): cluster 2\n", + "Example n. 5810 = (49.0, 464.0): cluster 2\n", + "Example n. 5811 = (49.0, 463.0): cluster 2\n", + "Example n. 5812 = (49.0, 462.0): cluster 2\n", + "Example n. 5813 = (49.0, 461.0): cluster 2\n", + "Example n. 5814 = (49.0, 460.0): cluster 2\n", + "Example n. 5815 = (49.0, 459.0): cluster 2\n", + "Example n. 5816 = (49.0, 458.0): cluster 2\n", + "Example n. 5817 = (49.0, 457.0): cluster 2\n", + "Example n. 5818 = (49.0, 456.0): cluster 2\n", + "Example n. 5819 = (49.0, 455.0): cluster 2\n", + "Example n. 5820 = (49.0, 454.0): cluster 2\n", + "Example n. 5821 = (49.0, 453.0): cluster 2\n", + "Example n. 5822 = (49.0, 452.0): cluster 2\n", + "Example n. 5823 = (49.0, 427.0): cluster 2\n", + "Example n. 5824 = (49.0, 426.0): cluster 2\n", + "Example n. 5825 = (49.0, 425.0): cluster 2\n", + "Example n. 5826 = (49.0, 424.0): cluster 2\n", + "Example n. 5827 = (49.0, 326.0): cluster 2\n", + "Example n. 5828 = (49.0, 325.0): cluster 2\n", + "Example n. 5829 = (49.0, 324.0): cluster 2\n", + "Example n. 5830 = (49.0, 323.0): cluster 2\n", + "Example n. 5831 = (49.0, 322.0): cluster 2\n", + "Example n. 5832 = (49.0, 321.0): cluster 2\n", + "Example n. 5833 = (49.0, 320.0): cluster 2\n", + "Example n. 5834 = (49.0, 319.0): cluster 2\n", + "Example n. 5835 = (49.0, 318.0): cluster 2\n", + "Example n. 5836 = (49.0, 308.0): cluster 2\n", + "Example n. 5837 = (49.0, 307.0): cluster 2\n", + "Example n. 5838 = (49.0, 306.0): cluster 2\n", + "Example n. 5839 = (49.0, 305.0): cluster 2\n", + "Example n. 5840 = (49.0, 304.0): cluster 2\n", + "Example n. 5841 = (49.0, 303.0): cluster 2\n", + "Example n. 5842 = (49.0, 302.0): cluster 2\n", + "Example n. 5843 = (49.0, 301.0): cluster 2\n", + "Example n. 5844 = (49.0, 300.0): cluster 2\n", + "Example n. 5845 = (49.0, 299.0): cluster 2\n", + "Example n. 5846 = (49.0, 298.0): cluster 2\n", + "Example n. 5847 = (49.0, 297.0): cluster 2\n", + "Example n. 5848 = (49.0, 296.0): cluster 2\n", + "Example n. 5849 = (49.0, 295.0): cluster 2\n", + "Example n. 5850 = (49.0, 294.0): cluster 2\n", + "Example n. 5851 = (49.0, 290.0): cluster 2\n", + "Example n. 5852 = (49.0, 289.0): cluster 2\n", + "Example n. 5853 = (49.0, 288.0): cluster 2\n", + "Example n. 5854 = (49.0, 287.0): cluster 2\n", + "Example n. 5855 = (49.0, 286.0): cluster 2\n", + "Example n. 5856 = (49.0, 285.0): cluster 2\n", + "Example n. 5857 = (49.0, 284.0): cluster 2\n", + "Example n. 5858 = (49.0, 283.0): cluster 2\n", + "Example n. 5859 = (49.0, 282.0): cluster 2\n", + "Example n. 5860 = (49.0, 281.0): cluster 2\n", + "Example n. 5861 = (49.0, 280.0): cluster 2\n", + "Example n. 5862 = (49.0, 279.0): cluster 2\n", + "Example n. 5863 = (49.0, 278.0): cluster 2\n", + "Example n. 5864 = (49.0, 277.0): cluster 2\n", + "Example n. 5865 = (49.0, 276.0): cluster 2\n", + "Example n. 5866 = (49.0, 275.0): cluster 2\n", + "Example n. 5867 = (49.0, 274.0): cluster 2\n", + "Example n. 5868 = (49.0, 273.0): cluster 2\n", + "Example n. 5869 = (49.0, 272.0): cluster 2\n", + "Example n. 5870 = (49.0, 271.0): cluster 2\n", + "Example n. 5871 = (49.0, 267.0): cluster 2\n", + "Example n. 5872 = (49.0, 266.0): cluster 2\n", + "Example n. 5873 = (49.0, 265.0): cluster 2\n", + "Example n. 5874 = (49.0, 264.0): cluster 2\n", + "Example n. 5875 = (49.0, 263.0): cluster 2\n", + "Example n. 5876 = (49.0, 262.0): cluster 2\n", + "Example n. 5877 = (49.0, 261.0): cluster 2\n", + "Example n. 5878 = (49.0, 260.0): cluster 2\n", + "Example n. 5879 = (49.0, 259.0): cluster 2\n", + "Example n. 5880 = (49.0, 258.0): cluster 2\n", + "Example n. 5881 = (49.0, 257.0): cluster 2\n", + "Example n. 5882 = (49.0, 256.0): cluster 2\n", + "Example n. 5883 = (49.0, 255.0): cluster 2\n", + "Example n. 5884 = (49.0, 254.0): cluster 2\n", + "Example n. 5885 = (49.0, 253.0): cluster 2\n", + "Example n. 5886 = (49.0, 252.0): cluster 2\n", + "Example n. 5887 = (49.0, 251.0): cluster 2\n", + "Example n. 5888 = (49.0, 250.0): cluster 2\n", + "Example n. 5889 = (49.0, 249.0): cluster 2\n", + "Example n. 5890 = (49.0, 248.0): cluster 2\n", + "Example n. 5891 = (49.0, 247.0): cluster 2\n", + "Example n. 5892 = (49.0, 246.0): cluster 2\n", + "Example n. 5893 = (49.0, 245.0): cluster 2\n", + "Example n. 5894 = (49.0, 244.0): cluster 2\n", + "Example n. 5895 = (49.0, 243.0): cluster 2\n", + "Example n. 5896 = (49.0, 242.0): cluster 2\n", + "Example n. 5897 = (49.0, 241.0): cluster 2\n", + "Example n. 5898 = (49.0, 240.0): cluster 2\n", + "Example n. 5899 = (49.0, 239.0): cluster 2\n", + "Example n. 5900 = (49.0, 238.0): cluster 2\n", + "Example n. 5901 = (49.0, 237.0): cluster 2\n", + "Example n. 5902 = (49.0, 236.0): cluster 2\n", + "Example n. 5903 = (49.0, 230.0): cluster 2\n", + "Example n. 5904 = (49.0, 229.0): cluster 2\n", + "Example n. 5905 = (49.0, 228.0): cluster 2\n", + "Example n. 5906 = (49.0, 227.0): cluster 2\n", + "Example n. 5907 = (49.0, 226.0): cluster 2\n", + "Example n. 5908 = (49.0, 225.0): cluster 2\n", + "Example n. 5909 = (49.0, 224.0): cluster 2\n", + "Example n. 5910 = (49.0, 223.0): cluster 2\n", + "Example n. 5911 = (49.0, 215.0): cluster 2\n", + "Example n. 5912 = (49.0, 214.0): cluster 2\n", + "Example n. 5913 = (49.0, 213.0): cluster 2\n", + "Example n. 5914 = (49.0, 212.0): cluster 2\n", + "Example n. 5915 = (49.0, 211.0): cluster 2\n", + "Example n. 5916 = (49.0, 210.0): cluster 2\n", + "Example n. 5917 = (49.0, 209.0): cluster 2\n", + "Example n. 5918 = (49.0, 208.0): cluster 2\n", + "Example n. 5919 = (49.0, 207.0): cluster 2\n", + "Example n. 5920 = (49.0, 196.0): cluster 2\n", + "Example n. 5921 = (49.0, 195.0): cluster 2\n", + "Example n. 5922 = (49.0, 194.0): cluster 2\n", + "Example n. 5923 = (49.0, 193.0): cluster 2\n", + "Example n. 5924 = (49.0, 192.0): cluster 2\n", + "Example n. 5925 = (49.0, 191.0): cluster 2\n", + "Example n. 5926 = (49.0, 190.0): cluster 2\n", + "Example n. 5927 = (49.0, 188.0): cluster 2\n", + "Example n. 5928 = (49.0, 187.0): cluster 2\n", + "Example n. 5929 = (49.0, 186.0): cluster 2\n", + "Example n. 5930 = (49.0, 185.0): cluster 2\n", + "Example n. 5931 = (49.0, 184.0): cluster 2\n", + "Example n. 5932 = (49.0, 183.0): cluster 2\n", + "Example n. 5933 = (49.0, 182.0): cluster 2\n", + "Example n. 5934 = (49.0, 181.0): cluster 2\n", + "Example n. 5935 = (49.0, 180.0): cluster 2\n", + "Example n. 5936 = (49.0, 179.0): cluster 2\n", + "Example n. 5937 = (49.0, 178.0): cluster 2\n", + "Example n. 5938 = (49.0, 177.0): cluster 2\n", + "Example n. 5939 = (49.0, 176.0): cluster 2\n", + "Example n. 5940 = (49.0, 175.0): cluster 2\n", + "Example n. 5941 = (49.0, 174.0): cluster 2\n", + "Example n. 5942 = (49.0, 173.0): cluster 2\n", + "Example n. 5943 = (49.0, 172.0): cluster 2\n", + "Example n. 5944 = (49.0, 171.0): cluster 2\n", + "Example n. 5945 = (49.0, 170.0): cluster 2\n", + "Example n. 5946 = (49.0, 169.0): cluster 2\n", + "Example n. 5947 = (49.0, 168.0): cluster 2\n", + "Example n. 5948 = (49.0, 167.0): cluster 2\n", + "Example n. 5949 = (49.0, 166.0): cluster 2\n", + "Example n. 5950 = (49.0, 165.0): cluster 2\n", + "Example n. 5951 = (49.0, 164.0): cluster 2\n", + "Example n. 5952 = (49.0, 163.0): cluster 2\n", + "Example n. 5953 = (49.0, 162.0): cluster 2\n", + "Example n. 5954 = (49.0, 160.0): cluster 2\n", + "Example n. 5955 = (49.0, 159.0): cluster 2\n", + "Example n. 5956 = (49.0, 158.0): cluster 2\n", + "Example n. 5957 = (49.0, 157.0): cluster 2\n", + "Example n. 5958 = (49.0, 156.0): cluster 2\n", + "Example n. 5959 = (49.0, 149.0): cluster 2\n", + "Example n. 5960 = (49.0, 148.0): cluster 2\n", + "Example n. 5961 = (49.0, 147.0): cluster 2\n", + "Example n. 5962 = (49.0, 146.0): cluster 2\n", + "Example n. 5963 = (49.0, 145.0): cluster 2\n", + "Example n. 5964 = (49.0, 144.0): cluster 2\n", + "Example n. 5965 = (49.0, 143.0): cluster 2\n", + "Example n. 5966 = (49.0, 142.0): cluster 2\n", + "Example n. 5967 = (49.0, 141.0): cluster 2\n", + "Example n. 5968 = (49.0, 140.0): cluster 2\n", + "Example n. 5969 = (49.0, 139.0): cluster 2\n", + "Example n. 5970 = (49.0, 138.0): cluster 2\n", + "Example n. 5971 = (49.0, 137.0): cluster 2\n", + "Example n. 5972 = (49.0, 136.0): cluster 2\n", + "Example n. 5973 = (49.0, 135.0): cluster 2\n", + "Example n. 5974 = (49.0, 134.0): cluster 2\n", + "Example n. 5975 = (49.0, 133.0): cluster 2\n", + "Example n. 5976 = (49.0, 132.0): cluster 2\n", + "Example n. 5977 = (49.0, 131.0): cluster 2\n", + "Example n. 5978 = (49.0, 130.0): cluster 2\n", + "Example n. 5979 = (49.0, 129.0): cluster 2\n", + "Example n. 5980 = (49.0, 128.0): cluster 2\n", + "Example n. 5981 = (50.0, 495.0): cluster 2\n", + "Example n. 5982 = (50.0, 494.0): cluster 2\n", + "Example n. 5983 = (50.0, 493.0): cluster 2\n", + "Example n. 5984 = (50.0, 492.0): cluster 2\n", + "Example n. 5985 = (50.0, 491.0): cluster 2\n", + "Example n. 5986 = (50.0, 490.0): cluster 2\n", + "Example n. 5987 = (50.0, 489.0): cluster 2\n", + "Example n. 5988 = (50.0, 488.0): cluster 2\n", + "Example n. 5989 = (50.0, 487.0): cluster 2\n", + "Example n. 5990 = (50.0, 486.0): cluster 2\n", + "Example n. 5991 = (50.0, 485.0): cluster 2\n", + "Example n. 5992 = (50.0, 484.0): cluster 2\n", + "Example n. 5993 = (50.0, 483.0): cluster 2\n", + "Example n. 5994 = (50.0, 482.0): cluster 2\n", + "Example n. 5995 = (50.0, 481.0): cluster 2\n", + "Example n. 5996 = (50.0, 480.0): cluster 2\n", + "Example n. 5997 = (50.0, 479.0): cluster 2\n", + "Example n. 5998 = (50.0, 478.0): cluster 2\n", + "Example n. 5999 = (50.0, 477.0): cluster 2\n", + "Example n. 6000 = (50.0, 476.0): cluster 2\n", + "Example n. 6001 = (50.0, 475.0): cluster 2\n", + "Example n. 6002 = (50.0, 474.0): cluster 2\n", + "Example n. 6003 = (50.0, 473.0): cluster 2\n", + "Example n. 6004 = (50.0, 472.0): cluster 2\n", + "Example n. 6005 = (50.0, 471.0): cluster 2\n", + "Example n. 6006 = (50.0, 470.0): cluster 2\n", + "Example n. 6007 = (50.0, 469.0): cluster 2\n", + "Example n. 6008 = (50.0, 468.0): cluster 2\n", + "Example n. 6009 = (50.0, 467.0): cluster 2\n", + "Example n. 6010 = (50.0, 466.0): cluster 2\n", + "Example n. 6011 = (50.0, 465.0): cluster 2\n", + "Example n. 6012 = (50.0, 464.0): cluster 2\n", + "Example n. 6013 = (50.0, 463.0): cluster 2\n", + "Example n. 6014 = (50.0, 462.0): cluster 2\n", + "Example n. 6015 = (50.0, 461.0): cluster 2\n", + "Example n. 6016 = (50.0, 460.0): cluster 2\n", + "Example n. 6017 = (50.0, 459.0): cluster 2\n", + "Example n. 6018 = (50.0, 458.0): cluster 2\n", + "Example n. 6019 = (50.0, 457.0): cluster 2\n", + "Example n. 6020 = (50.0, 456.0): cluster 2\n", + "Example n. 6021 = (50.0, 455.0): cluster 2\n", + "Example n. 6022 = (50.0, 454.0): cluster 2\n", + "Example n. 6023 = (50.0, 453.0): cluster 2\n", + "Example n. 6024 = (50.0, 452.0): cluster 2\n", + "Example n. 6025 = (50.0, 428.0): cluster 2\n", + "Example n. 6026 = (50.0, 427.0): cluster 2\n", + "Example n. 6027 = (50.0, 426.0): cluster 2\n", + "Example n. 6028 = (50.0, 425.0): cluster 2\n", + "Example n. 6029 = (50.0, 424.0): cluster 2\n", + "Example n. 6030 = (50.0, 423.0): cluster 2\n", + "Example n. 6031 = (50.0, 329.0): cluster 2\n", + "Example n. 6032 = (50.0, 328.0): cluster 2\n", + "Example n. 6033 = (50.0, 327.0): cluster 2\n", + "Example n. 6034 = (50.0, 326.0): cluster 2\n", + "Example n. 6035 = (50.0, 325.0): cluster 2\n", + "Example n. 6036 = (50.0, 324.0): cluster 2\n", + "Example n. 6037 = (50.0, 323.0): cluster 2\n", + "Example n. 6038 = (50.0, 322.0): cluster 2\n", + "Example n. 6039 = (50.0, 321.0): cluster 2\n", + "Example n. 6040 = (50.0, 320.0): cluster 2\n", + "Example n. 6041 = (50.0, 319.0): cluster 2\n", + "Example n. 6042 = (50.0, 318.0): cluster 2\n", + "Example n. 6043 = (50.0, 307.0): cluster 2\n", + "Example n. 6044 = (50.0, 306.0): cluster 2\n", + "Example n. 6045 = (50.0, 305.0): cluster 2\n", + "Example n. 6046 = (50.0, 304.0): cluster 2\n", + "Example n. 6047 = (50.0, 303.0): cluster 2\n", + "Example n. 6048 = (50.0, 302.0): cluster 2\n", + "Example n. 6049 = (50.0, 301.0): cluster 2\n", + "Example n. 6050 = (50.0, 300.0): cluster 2\n", + "Example n. 6051 = (50.0, 299.0): cluster 2\n", + "Example n. 6052 = (50.0, 296.0): cluster 2\n", + "Example n. 6053 = (50.0, 293.0): cluster 2\n", + "Example n. 6054 = (50.0, 292.0): cluster 2\n", + "Example n. 6055 = (50.0, 291.0): cluster 2\n", + "Example n. 6056 = (50.0, 290.0): cluster 2\n", + "Example n. 6057 = (50.0, 289.0): cluster 2\n", + "Example n. 6058 = (50.0, 288.0): cluster 2\n", + "Example n. 6059 = (50.0, 287.0): cluster 2\n", + "Example n. 6060 = (50.0, 286.0): cluster 2\n", + "Example n. 6061 = (50.0, 285.0): cluster 2\n", + "Example n. 6062 = (50.0, 284.0): cluster 2\n", + "Example n. 6063 = (50.0, 283.0): cluster 2\n", + "Example n. 6064 = (50.0, 282.0): cluster 2\n", + "Example n. 6065 = (50.0, 281.0): cluster 2\n", + "Example n. 6066 = (50.0, 280.0): cluster 2\n", + "Example n. 6067 = (50.0, 279.0): cluster 2\n", + "Example n. 6068 = (50.0, 278.0): cluster 2\n", + "Example n. 6069 = (50.0, 277.0): cluster 2\n", + "Example n. 6070 = (50.0, 275.0): cluster 2\n", + "Example n. 6071 = (50.0, 274.0): cluster 2\n", + "Example n. 6072 = (50.0, 273.0): cluster 2\n", + "Example n. 6073 = (50.0, 272.0): cluster 2\n", + "Example n. 6074 = (50.0, 271.0): cluster 2\n", + "Example n. 6075 = (50.0, 268.0): cluster 2\n", + "Example n. 6076 = (50.0, 267.0): cluster 2\n", + "Example n. 6077 = (50.0, 266.0): cluster 2\n", + "Example n. 6078 = (50.0, 265.0): cluster 2\n", + "Example n. 6079 = (50.0, 264.0): cluster 2\n", + "Example n. 6080 = (50.0, 263.0): cluster 2\n", + "Example n. 6081 = (50.0, 262.0): cluster 2\n", + "Example n. 6082 = (50.0, 261.0): cluster 2\n", + "Example n. 6083 = (50.0, 260.0): cluster 2\n", + "Example n. 6084 = (50.0, 259.0): cluster 2\n", + "Example n. 6085 = (50.0, 258.0): cluster 2\n", + "Example n. 6086 = (50.0, 257.0): cluster 2\n", + "Example n. 6087 = (50.0, 256.0): cluster 2\n", + "Example n. 6088 = (50.0, 255.0): cluster 2\n", + "Example n. 6089 = (50.0, 254.0): cluster 2\n", + "Example n. 6090 = (50.0, 253.0): cluster 2\n", + "Example n. 6091 = (50.0, 252.0): cluster 2\n", + "Example n. 6092 = (50.0, 251.0): cluster 2\n", + "Example n. 6093 = (50.0, 250.0): cluster 2\n", + "Example n. 6094 = (50.0, 249.0): cluster 2\n", + "Example n. 6095 = (50.0, 248.0): cluster 2\n", + "Example n. 6096 = (50.0, 247.0): cluster 2\n", + "Example n. 6097 = (50.0, 246.0): cluster 2\n", + "Example n. 6098 = (50.0, 245.0): cluster 2\n", + "Example n. 6099 = (50.0, 244.0): cluster 2\n", + "Example n. 6100 = (50.0, 243.0): cluster 2\n", + "Example n. 6101 = (50.0, 242.0): cluster 2\n", + "Example n. 6102 = (50.0, 241.0): cluster 2\n", + "Example n. 6103 = (50.0, 240.0): cluster 2\n", + "Example n. 6104 = (50.0, 239.0): cluster 2\n", + "Example n. 6105 = (50.0, 238.0): cluster 2\n", + "Example n. 6106 = (50.0, 237.0): cluster 2\n", + "Example n. 6107 = (50.0, 236.0): cluster 2\n", + "Example n. 6108 = (50.0, 230.0): cluster 2\n", + "Example n. 6109 = (50.0, 229.0): cluster 2\n", + "Example n. 6110 = (50.0, 228.0): cluster 2\n", + "Example n. 6111 = (50.0, 227.0): cluster 2\n", + "Example n. 6112 = (50.0, 226.0): cluster 2\n", + "Example n. 6113 = (50.0, 225.0): cluster 2\n", + "Example n. 6114 = (50.0, 224.0): cluster 2\n", + "Example n. 6115 = (50.0, 215.0): cluster 2\n", + "Example n. 6116 = (50.0, 214.0): cluster 2\n", + "Example n. 6117 = (50.0, 213.0): cluster 2\n" + ] + }, + { + "ename": "IndexError", + "evalue": "index 6118 is out of bounds for axis 0 with size 6118", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mIndexError\u001b[0m Traceback (most recent call last)", + "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m\u001b[0m\n\u001b[1;32m 7\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 8\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mn_samples3\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 9\u001b[0;31m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'Example n. {i} = ({(data3[i,0])}, {data3[i,1]}): cluster {km2.labels_[i]}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", + "\u001b[0;31mIndexError\u001b[0m: index 6118 is out of bounds for axis 0 with size 6118" + ] + } + ], + "source": [ + "data3, feature_names3, n_samples3, n_features2 = load_data(file_path, file_name3)\n", + "\n", + "np.random.seed(5)\n", + "\n", + "k=4\n", + "km3 = KMeans(n_clusters=k, random_state=0).fit(data3)\n", + "\n", + "for i in range(n_samples3):\n", + " print(f'Example n. {i} = ({(data3[i,0])}, {data3[i,1]}): cluster {km2.labels_[i]}')" + ] }, { "cell_type": "markdown", @@ -689,10 +13035,25 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 10, "metadata": {}, - "outputs": [], - "source": [] + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "plot(data3, n_samples3, km3, 'Clustered points in dataset n. 2')" + ] }, { "cell_type": "markdown", @@ -706,12 +13067,68 @@ }, { "cell_type": "code", - "execution_count": null, - "metadata": { - "collapsed": true - }, + "execution_count": 11, + "metadata": {}, "outputs": [], - "source": [] + "source": [ + "from numpy.linalg import norm\n", + "from pandas import DataFrame\n", + "from functools import partial\n", + "from multiprocessing import Pool\n", + "\n", + "def sse(X, Y):\n", + " return norm(X - Y)**2\n", + "\n", + "def wss(clusters, means):\n", + " wss = 0\n", + " for mean, cluster in zip(means, clusters):\n", + " wss += sum(map(lambda example: sse(example, mean), cluster))\n", + " return wss\n", + "\n", + "def compute_clusters(labels, data):\n", + " df = DataFrame(zip(labels, data)) # tuple \n", + " return map(lambda nd: nd[1]. # list of lists\n", + " drop(0, axis=1). # cluster is the index, drop it from tuple\n", + " values.flatten(), # flatten list of single values\n", + " df.groupby(0)) # group by cluster \n", + "\n", + "def wss_from_data(n_clusters, data, seed):\n", + " km = KMeans(n_clusters=n_clusters, random_state=seed).fit(data)\n", + " means = km.cluster_centers_\n", + "\n", + " clusters = list(compute_clusters(km.labels_, data))\n", + " #return min(map(lambda clu: wss(clu, means), clusters))\n", + " #print(seed, clusters)\n", + " return wss(clusters, means)\n", + "\n", + "\n", + "def eval_kmeans(n_clusters, data, n_iter = 10):\n", + " for seed in range(n_iter):\n", + " yield wss_from_data(n_clusters, data, seed)" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "data1 evaluation: 62525.0\n", + "data2 evaluation: 2586285.25177\n", + "data3 evaluation: 490062743.13532\n" + ] + } + ], + "source": [ + "print(\"data1 evaluation:\", round(min(eval_kmeans(3, data1, 10)), 5))\n", + "print(\"data2 evaluation:\", round(min(eval_kmeans(4, data2, 10)), 5))\n", + "print(\"data3 evaluation:\", round(min(eval_kmeans(5, data3, 2)), 5))" + ] }, { "cell_type": "markdown", @@ -722,10 +13139,302 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 13, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[540169.52, 225889.10891089108, 62525.00000000001, 45950.0, 30325.0, 15625.0, 12565.0, 9660.0, 6953.0]\n" + ] + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "evaluations = list(min(eval_kmeans(k, data1, 3)) for k in range(1, 10))\n", + "print(evaluations)\n", + "plt.plot([k for k in range(1,10)], evaluations)\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[11537928.01912388, 6012536.807920576, 3327557.113231734, 2586444.7018824834, 2055550.8429586363, 1660581.039399704, 1290968.9205037882, 1108531.0806234784, 943472.771559587]\n" + ] + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "evaluations = list(min(eval_kmeans(k, data2, 3)) for k in range(1, 10))\n", + "print(evaluations)\n", + "plt.plot([k for k in range(1,10)], evaluations)\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 15, "metadata": { - "collapsed": true + "scrolled": true }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[2418979043.1137347, 1425203644.7839656, 916173975.2419217, 632521252.3875252, 490062743.1353237, 384262182.9810209, 314207574.83201766, 275304694.96816456, 240376762.14969328]\n" + ] + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "evaluations = list(min(eval_kmeans(k, data3, 1)) for k in range(1, 10))\n", + "print(evaluations)\n", + "plt.plot([k for k in range(1,10)], evaluations)\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# DBSCAN" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "from sklearn.cluster import DBSCAN\n", + "cls = DBSCAN(eps = 3)\n", + "cls.fit(data1)\n", + "\n", + "plot(data1, n_samples1, cls, 'DBScan 1')" + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfQAAAI1CAYAAAAgiggGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOydd5yU1fX/P3cXWDoINsQCarBiDBaISSzRxGhsifqzRTSJLSb5ijH2BhrAqFExURNsiEoRFhQQQXovAktbWLb33ndnd2Z2Zs7vj888ewYEBKW4y3m/XvPizDzPmadd7p29n3POdSICwzAMwzBaNnEH+wQMwzAMw/ju2IBuGIZhGK0AG9ANwzAMoxVgA7phGIZhtAJsQDcMwzCMVoAN6IZhGIbRCrAB3TCM74xzbqhz7qODfR6GcShjA7ph7Eecc9nOuUbnXJ1zrto5t9w5d59zLi5mnzHOuaBzrj6631rn3EUx2491ziU658qdczXOuc3OuTtjtreLDqhpzjlf9JjvOef67HAuY5xzIedcrx0+H+qcE+fc/4v5rE30s+2+Yx/dkzHOuX/s6+/d2+M45450zo13zhVG7+sy59zA/X1ehrG/sAHdMPY/V4tIFwAnAHgBwKMA3t1hnxdFpDOArgDeAjDFORcf3fYhgLyof08AtwMoifGdDOAaALcC6AbghwDWArjU28E51wnA9QBqAPxuJ+dYCWBYzDEPBToD+ArAOQB6APgAwOfOuc4H9awM41tiA7phHCBEpEZEpgG4CcAdzrkzd7KPABgHDjBHRT8+D8AYEfGJSEhEkkTkCwBwzl0G4BcArhWRr6Lba0TkDRGJ/dFwPYBqAM8BuGMnpzcLQBA7H+y/hnOur3NuUXRGYQ6Aw3fYPsk5Vxz9y3exc+6M6Of3ALgNwCPRGYnp0c8fc85lRL9vi3PuNzHfdXL0WDXRWYqJMdtOdc7Ncc5VOue2ebMMuzrODvc6U0ReEZEiEQmLyGgA7QCcsif3wDC+b9iAbhgHGBFZDSAfwM923Bb9C3kwgCzoX+ErAbzhnLvZOXf8Di6XAVgtInnfcNg7AIwHMAHAqc65c3Y8LQBPA3jWOdd2Dy5jHDgLcDiA5/H1HwlfAPgBgCMBrAPwMQBEB82PEZ2REJGro/tngPejG4BhAD6KkQaeB/AlgMMAHAvg30DzrMOc6LkcCeBmAG86507fzXF2iXPubHBAT9+D6zeM7x02oBvGwaEQ/Cvc4+/OuWoA9QBeA/C0iISj224EsAQccLOcc+udc+dFt/UEULS7A0V/BFwCYJyIlACYB/5o2I7o7EEZgLv24PvOi55jQEQWA9juL2AReU9E6kQkAGAogB8657rt6jtFZJKIFIpIREQmAkgDcH50cxMoNxwjIn4RWRr9/CoA2SLyvjdzASARvF97hXOuKyhtDBORmr31N4zvAzagG8bBoTeoW3u8LCLdAXQEcC6Al5xzVwCAiFSJyGMicgY4Db8ewKfOOQegAkAv7J7bAWwVkfXR9x8DuHUXf4k/BeBJAO13833HAKgSEV/MZzme4ZyLd869EJ1CrwWQHd203bR8LM65wdEfKtXRHzZnxuz/CAAHYLVzLtk594fo5ycAGOj5RP1uA3D0bs59Z8fuAP4gWSkiI/fG1zC+T9iAbhgHmOhf170BLN1xm5DNAJYB+PVOtpcDeBkcVHsAmAvgfOfcsbs55GAAJ0Y17WIAr4CD5ZU7+f454JTz/bv5viIAh0WnvD1ipYBbAVwLygHdAPSJfu68w8R+mXPuBABvA/gLgJ7RHzabvf1FpFhE7haRYwDcC06rnwwGCi4Ske4xr84i8qedHWdnOOcSAHwKSiD3ftP+hvF9xgZ0wzhAOOe6OueuAnXsj0Rk0y72OxXATwEkR9//0zl3ZjSVrAuAPwFIF5EKEZkL6shTnXPnePs4psb9wTn3YwAngdPXZ0dfZ4K689em3aM8Cf5VvFNEJAfAGjAqvp1z7qcAYjXqLgAC4OxBRwAjdviKEgAnxrzvBA6+ZdHr/X30HL37cWPMD5aq6L4RADMA9HPO3e6caxt9neecO20Xx9mO6AzFZACNAO4Qkciu9jWMloAN6Iax/5nunKsD/6J8EvwL+fc77ONFY/vAALD3Afwvuq0jgKlglHomONV8TYzvDQBmApgIpqVtBqft54LBap+JyKboX7rFIlIMYBSAq5xzsTo+AEBElgFY/Q3XdCuAgaBs8CyAsTHbxoJT8AUAtoBBfbG8C+D06DT5pyKyBcC/AKwAB+H+4AyFx3kAVjnn6gFMA/BANEK9DsAvwWC4QgDFAP4JIGFnx9nJNVwA6vC/BFAdvf/1zrmvBSsaRkvAMUvGMAzDMIyWjP2FbhiGYRitABvQDcMwDKMVYAO6YRiGYbQCbEA3DMMwjFaADeiGYRiG0Qpoc7BP4Ltw+OGHS58+fQ72aRiGYRjGAWHt2rXlInLEzra16AG9T58+WLNmzcE+DcMwDMM4IDjncna1zabcDcMwDKMVYAO6YRiGYbQCbEA3DMMwjFaADeiGYRiG0QqwAd0wDMMwWgE2oBuGYRhGK8AGdMMwDMNoBdiAbhiGYRitABvQDcMwDKMVYAO6YRiGYbQCbEA3DMMwjFaADeiGYRiG0QqwAd0wDMMwWgE2oBuGYRhGK8AGdMMwDMNoBdiAbhiGYRitABvQDcMwDKMVYAO6YRiGYbQCbEA3DMMwjFaADeiGYRiG0QqwAd1oFbz7LnDeebS/+go4/HCgrg6oraW9Zs3BPT/DMIz9TZuDfQKGsS+4/HLguONon3EG8MYbQJcufP/GG8Dppx+8czMMwzgQOBE52OfwrTn33HNljf3pZRiGYRwiOOfWisi5O9tmU+6GYRiG0QqwAd1osTz5JPDII7QnTACuuIL2xo2cdq+v5+uMM/gZwH0mTDg452sYhrE/MQ3daLH88pdAOEx7wAAgEqF9/PHAvfcCHTvy/b338jMA+N3vuK9hGEZrwzR0wzAMw2ghmIZuGIZhGK0cG9CNFstLLwEjRtCePh0YPJh2Sgpw6aVAQwNfl14KbNvGbYMHc1/DMIzWhmnoRoulXz/V0Hv3Bs45h/ZhhwEDBwLt2vH9wIFA9+60BwwAjj32wJ+rYRjG/sY0dMMwDMNoIZiGbhiGYRitHBvQjRbL228Db71Fe/584KGHaGdlAbfeCgQCfN16Kz8DuM/8+QfnfA3DMPYnpqEbLZZ27YD4eNpt2wIdOtCOj6ftHN936AC0aaN227YH/lwNwzD2N6ahG4ZhGEYLwTR0wzAMw2jl2IBufGumTQPGjKG9YQPw3HO0y8uBIUMAv5/lWP/+dyAnh9tefRVYsoT2lCnARx/RXrsWGD6cdkkJ/QMBpqU99BCQn89tL78MrFhBe9Ikrcu+ejXwwgu0CwuBBx8Empr4evBBoKiI2154gfsCwPjx/A4AWL6c3w0AeXk8ZjjMcxgyBCgt5bbhw4F162h/+CEwdSrtJUuA116jnZ3Na45EgMZG4IEHeE8AYNgwrStvGIaxLzEN3fjWFBfrQFdVBaSn025oAFJTgWCQ2nVaGlBXx23Z2UCfPrSLirh4CgBUVgIZGeqflgaEQtTBU1PVPysLOOUU2oWF3AcAKiqAzEzaPp/6A7S942RmAmefTbugQPX0igoNnKuv5zHDYf4gSEvjdwI8x8pK9e/alXZZmfrX1dEnEuE9SEvjwA7wHlVV7e2dNgzD+GZMQzcMwzCMFoJp6IZhGIbRyrEB3dgrxo0DvEmRRYuoowOcVvZywmtrgZEjOV0twprrxcXc9uGHQFIS7fnzgc8/p52SAoweTbuqiv6hEKetX3yRU9oANXtPg54zB5g1i3ZyMvDuu7QrKqiVh8N8vfACPwO4T3Iy7S++4HcA/E4vHqCsjMcU4TmMHAlUV3Pb6NE8VwCYMUNz2pOSeG0Ar/Wll+gfDLLefG0tt731lkoTn33Ge2gYhrEvsAHd2Cu+/FIH1NWrdUBKS+MABXBAnDyZWnQoBCQmalDbrFnA5s20V67UALlt23TRlJIS+jQ0cECcPJl6NQDMnKkD8vLlwNKltLdu1R8HRUX09/v5SkzUHxQzZuiAvGyZBtglJ3OAB3isyZMZENfQQP+SEm6bNo36OsBzX7WK9qZNwOzZtPPy6BMK8R4kJuoPkk8/5b0CeO+++mrv7r9hGMau2K8aunMuG0AdgDCAkIic65zrAWAigD4AsgH8PxGpcs45AKMAXAmgAcCdIrJud99vGrphGIZxKHGwNfRLROTsmBN4DMA8EfkBgHnR9wBwBYAfRF/3AHjrAJybYRiGYbQKDsaU+7UAPojaHwC4LubzsUJWAujunOt1EM7P2A2zZ+va4klJOmVeUMCpZYApWu+9R/0bAMaOVQ165kydcl67ltPeAJCbqzndPh/w/vvq/8EHqkHPmKHpaatXc9oeYDqcp+fX1VEPF+FrzBhNW5s2jfsC9PVy0jMz+d0AUFPDYwI8h/ff17S1qVM5pQ7w3NeupZ2WplP2VVWqp4fDvBde2lpiosoHixcD69fv+l4bhmHsDft7QBcAXzrn1jrn7ol+dpSIRMt8oBjAUVG7N4C8GN/86Gfb4Zy7xzm3xjm3pswTJo0Dxn/+o4FoU6ZwsAOoRY8YQTs7G3jmGeZrB4O0Pd171Chg7lzan3zCwR6gFv7Pf9LOzASefpqDeGMj/bdu5bZXXtFAtAkTtDDNokUMRAM4uD79NAfx+nranu790kscSAH6eoVp5s1j0RuAGvszz/DYNTX0935EvPCC6vYffKCFaebM4bUBvNann2ZQYGUlba+wzvDh+iPk/fd5Dw3DMPYF+1tD7y0iBc65IwHMAfBXANNEpHvMPlUicphzbgaAF0RkafTzeQAeFZFdiuSmoRuGYRiHEgdNQxeRgui/pQCmAjgfQIk3lR79N1prDAUAjotxPzb6mWEYhmEY38B+G9Cdc52cc108G8AvAWwGMA3AHdHd7gAQTXbCNACDHRkEoCZmat74nvDVV6oBZ2QwXQvg1LI3lR0KaQoZQN3d05BXrdK66mlpOhVfXq5T2U1N1No9Zs1i+hnAqX0vBW3bNp2KLy1lGhvAdDNPzwZoBwK0ly3TFLKtWzUeoLhYp8L9fpUVAJ5LUxPtJUs0pz05WeMBCgtVj29oYHqfx+efaxnaxYu19OumTVru1jAM4zsjIvvlBeBEABuir2QAT0Y/7wlGt6cBmAugR/RzB+ANABkANgE495uOcc4554hxYLnwQpHhw2kPGSJy/fW0x44V6duX9saNIu3bi5SUiAQCIl26iMybx22DBom89BLt++8XueUW2u+8I3LKKbTXrBHp0EGkokLE5xPp3Flk8WJuGzBAZNQo2nfdJXLHHbTffFOkf3/aK1aIdOwoUlPDV8eOIitXctuZZ4q89RbtwYNF7r6b9muviXjNafFiHtPnEykv57msXctt/fqJvPsu7ZtvFvnzn2m/+KLIj39Me+5cXnMgIFJczHuxcSO39ekj8uGHtK+/XuTBB/fq9huGcYgDYI3sYky0Wu7GXhEOA/HxtEUYBe69D4W4GMue2l4Uelzct/P3ouAPpr9zfO14b/bEPxzmsT1/wzCMb2J3GrqttmbsFd6ABXAgin3fps3e2bGD4bfxj9tBMDrY/nt7L2L3NwzD+K5Y6ddDmLIyzclubNSSrCJarx3gWueehhy7/GdJCfPHAerGW7aov5efDTDXOnYp05oa2sXFWhLW51M9PBLZ3j8piX/NAkw/83LSCwtVz6+r05KukYiuWQ7Q9v4aT0nRpVgLClTPr63V1LZwWOvNAzwXz3/rVs1Jz8tTPb+6WvX0UGj7/PI1a3hPAOruDQ20c3O3X37W09ObmnjPY/0NwzC+kV3NxbeEl2no340hQ0R+9SvaY8eKHHkk7c2bRZwTycsTCYWoQU+bxm2DBokMHUr7/vtFrrmG9jvviBx7LO1160Ti4qgfBwIiCQkis2Zx24ABIiNG0L7rLpEbbqD95psiJ55Ie+VK+peXU8du21Y1+P79VYO/4w7V4EeNUg1+yRKRNm1UQ2/Thp+JcB9Pg7/lFtXgX3pJNfh580TatVMNPS5OZNUqbuvbVzX4669XDX74cNXgv/iCunkgIFJURP+kJG7r3Vs1+KuvVg3+2WdVg//sM97zUEgkN5fPIjl5Fw/RMIxDCpiGbuyMQICFX7p04V+gVVVAz57cVlYGHHEE7fJy4PDDadfUAB07Am3bMho8FAI6d6Z/dTXQo8fX/WPt6mqgUyf6NzbSr1Mn/lVcU7Nn/p07c+q6oYF/+Xr+tbXAYYft3r+qCujaldPdPh+n/Dt21IVUunffvX9lJdCtG/3r6/lvhw78q9rn2zP/7t05XV9fz+to357PobGR373jPY/1Nwzj0GZ3GroN6IZhGIbRQjjYi7MY3yPKyzUn3OfjX4wA/8ItLNT98mKK8ObnqwZcVqY54fX1qqeHQqpH785/VzQ1qR69o3+sXVqqOeW1tarHB4O6xOnu/EtKuC9AX0+PDwRUz96df3GxxhNUV6se7/drfruIxgbs6F9UpPEEVVVaY76xkc/mm/wLCzWewDAMIxYb0A8xfvtb4PnnaT/5JHDrrbQ//hg4+2zamzcDxx/PgSQcBvr107XKr76a9cwB4JFHgDuiJYLGjAHOO4/2unXACSfo4HnSSbpW+K54+23gggtor1oF9OnDAi4NDfT36rdfeinw73/T/utfgT/9ifZbbwEXXUR7yRKgb18O1rW1tL2iNRdeCPz3v7Tvuw/4v/+j/frrwGWX0Z43j8dsaOA5nHCCFo358Y+Bd9+l/cc/Ag89RPtf/wKuuIL2rFnAySfz2ouL6e8FyZ17rtavHzwYeCy61uALL/DeAlxApl8/3vu8PD4LrwDPD38IjBu3+3tpGMYhyq7E9ZbwsqC4vScvj4FiIiKVlSIFBbQDAZG0NN0vNggrJYUBWiIM0qqtpV1RwaAvERG/XyQ9fef+W7eKhMO7P6/GRpGMDNqRyK79s7JE6utpl5WxeI0IA9gyM9V/yxb137KFn4lwH5+PdkkJv0OE35mVRTsc5jFjr8Xzz8jguYow6K+8nHZdnUhODu1Q6Ov+HunpvFciIoWFvIcivKe5ueqfkrJz/7Q0PivDMA5NYEFxhmEYhtHyMQ3daCYQ0JzqcFj1aEDzo3dn+/3qHwqpHr07nz1BRLX93X1XY6Pq8U1NqmcfDP9gUPXwSERjC3bnH2t/V3/DMIxYbEA/xLjiCmDoUNqPPgpcfz3tceOo+wIsENOtGwOzwmHgqKN0sZTLLtN1z//2N+Dmm2mPGQOcfjrtDRuYmhUbpPZNvP02cNZZtNesYfqZp6H37Mn1zgFq2K+9Rvvee4E//IH2G29Qnwa4SEuPHqqh9+jBRV0A4JxzgDffpP3731NHB7gWuqfhL1zIY3oa+mGHaXGX/v2Bd96h/bvfUccHgBdfVA3/yy+ZZuZp6N27Axs3ctupp3IddQC46SbV4IcPB37xC9qffw4cfTTvfX4+n4VXdOekk4Dx4/f8vhqGcQixq7n4lvAyDX3v2bSJ2q0I9XRPa66t5aImItSQvUIuIizK4unGGzZQOxahZuxpvTU1WnwlFBKZP3/vzquqSuSrr2g3NW3vv2iR6sbr1qlunZmpun1FBRd1EREJBkUWLFD/BQv4mQj38XTr9HTV3cvK+N0iPNaiReo/fz7PSYTnWFVFOzVVJDubdmmpyPr1tP1+XUxGhPfSi0FYtUpjGFJSVDcvLtYFXBoatBCO5+/FEKxYQb3eMIxDE5iGbhiGYRgtH9PQDcMwDKOVYwP6Icavf60a+NNPAzfeSHvKFNXQU1OpO3tFTHr1AubO5bZf/hJ4+WXajz4K3HYb7QkTVENPTqYGXVZGHfmoo4DFi3d/XmPHMscaYM52z54svNLQABx5pGrgF1zAnHMA+MtfgHvuof3uu5oH/9VXLJtaV0cN/fDDVQM/7zzNI7/nHn4HwO/8yU9oL1/OYzY2svBOz566WMoPfwh8+CHtO+9kHAEAjBoFXHIJ7UWLeM3BIIvV9OihC9ecdhowcSLtW2/VPPSXX+a9BXive/XSYj89eujCMSefzGdlGIbxNXY1F98SXqah7z2LF1P7FaF+vnw57dJSXYAlEBAZN05128mTRaqraS9YoPnimzdzIRURasCff07b76e/l7s9aZLmru+KggIuaiJCvX78ePWfOFF14zlzNN87KUlk7VraeXkis2fT9vlEJkzQ754wQXPPZ88Wyc+nvXat6t45OfxuER5r4kTakQjPxYshmDlTYxBWr1bdOytLdf+aGl6zCO/huHGaez5jhsYgrFzJeyhCPX/hQtpVVSKJidv7ezEA06bxWRmGcWgC09ANwzAMo+VjGrphGIZhtHJsQD/A3Hcf8M9/0n7jDeYyA9RdBwxg0ZKiImqtubncNnAgc5sB1g9/5RXar73GXGqAuqunIefl0d9bbOWcc4AFC2gPHqy10F96STXoL77QPOysLPqXllLHPftsrYV+883A//5He/hw1aCnTdM87NRU6umVldSRzzqL9dkB5r2/9x7toUOBBx+kPXmy1lLfsoX+NTXUsfv3Z314ALjmGtWwn3yS9eQBavheLfWNG4EzzuDCJ/X1tL088Cuu4L4AfZ98kvbYscC119Jeu5bH9Pt5Dqefrnngl10GJCbSHjIEGDaM9rvvAjfcQHvlSl5zUxPz2E87DUhL47YLL9S6+H/+s8Yz/Pe/wC230F66lPc8EmEu/2mn8ZkAfEazZtG+5x4+Q4DPdPBg2gsW8JkDQEEB/b3FXs47j7XqAcYAjBpF+1//Au66C4ZhtGDih3pVRlogo0ePHnqPNyK1EOLiOEAceyzXBO/dm8VGEhK4zvd55wHt2rEzv+gi7uMcC6p463ifcQZwzDHqf8op9O/alR15QgJ/GFx0EdfbBjgQdOnC4/fvz6Crtm2B447jQiAJCSxgMmAAbeeAn/2M+4gAP/0pzy8ujoPV0Udz2wknMFCrXTsWYDn77K/7RyL079SJ/mefzaCzNm24cMpJJ9G/Z08GnSUk8Dp/+lP6h8McCDt04PcOGMDCLW3a0LdvX/ofcQSvLSGBfj/5iR7/wgu57rhzLEDTsyeP8YMfcCEYz//MM+nfrh3vWay/d13nncdAtfh43rvjj+f+Rx3FZ5OQwHMdNGj7Z9muHf0HDmSxmfh4PvvjjuNxevXi4JuQwHt1/vn08Z5l27Z8loMG8Vl5bal376+3pS5deJ0JCXr82LbQtSv9zzxT25LXFgzD+P4ybNiwoqFDh47e2TbT0A3DMAyjhWAaumEYhmG0cmxAPwD89reqYT/xhOZRjx0LPPAA7VWrqO+KMH/74otVA7/mGtWwH35Ya4m/957WAl+2DLjqKtrFxfQvLeX7K6+krgtQ9/Vqif/vf5oHvXAh8Jvf0M7Pp39FBadrL79c87j/8hddj/s//2EuO0AN38tpz85mTnZNDXXkX/xC87jvvRf45BPar74KPPcc7S++0LXZ09OBn/+ceeSBAHXrzZu57Y9/1Dzsl15SDXr6dNWQU1K4bnpDA1+XXgps28Ztgwerhj1ihGrQU6bwuwEe67LLeOy6Op5LRga33XKLatjDhmld+YkTtS78+vXMKQ+FgOpq3oucHG674QbVsJ96inEUANej9+IR1qwBfvUr3vuKCj4LTwO/7jqta//YYxrPMGaMxiOsWMFnDlCDv/hitgmAbWT5ctoPPaTxDG+/rfEIS5awzRmG0bJoc7BP4FDg/POpOQPUWI84gnafPrra15FHUht1DujYkXbnztw2cCD1WYCa53HH0e7blzoowO0DB9Lu1In+nTqp/5FH0j7rLB4XAE48Ufc5+mieJ0D9ddAgnkdcHG3vnM86i7o5QO28Z0/avXppUF7XrvRp35667aBBut/ZZ1NzBqhfeyucHXOMBnJ1785zbt9ej9+jB7f96Ed6/f36UV8HqB97/ocdRv927fT6u3enPWAA4xcA6s2ernzccfxugMcaNEjjFwYOpGYNUJc+5hjap52m9+/443ltAK914EBq5O3b0+7aldt2bAvedZ1wgq6kdvjh9ImL07bQpcvO/Xv1ot23r66C57UlYPdtoX9/fZYnnki9Hdi+LRmG0XIwDd0wDMMwWgimoRuGYRhGK8cG9APAffdpHvbLL6sGPXWqasgbNzIvGGAN85tuopYOAHffzdxoABg5UjXoSZNUQ163TjXg8nL6V1by/e9/rxr288+rBj1+PNfxBoDVq6lvA9Rdb7qJGrgIdefkZG579lnmnAPMB/dy4pcvZ141QO3/5puZAx4OM9c+JYXbnnySejkAvP++5sQvXqzxBHl51KobGihJ3HYbdXWA9eO9nPy339Z4hPnzNZ4gK4t6fCDA1623ah73Qw9xX4Dror/9Nu0vv+R3AzzWbbfx2A0NPJe8PG574AFqzADw+uu8BoDrxXs57SkpwO2389rr6ngvvHiI++/XuvSvvAJ89BHtadN4bwFq+HfcwXtfU8Nn4a0tf++9rFUP8Nl5OfVTpgD/+Aft9eu1PkFlJf0rKvj+j38EkpJoDx/O/H+AMQAvvEB7zRq2OcMwWhamoR8AOnZUrdbLcQb4b/v2tNu04X6Aaqfx8XzfoYP6t2+/5/6evr6n/h067Nw/9ly+zfFjr799e82nbtdOddu2bbc/focO/NeLKYg9fqy/93msf3y85qzveP0dOqi/l6++M/+OHenvHD/f2bNISNj5+cce37sWz39XbSHWf2fPYlfHj/Xfl23B8zcMo+VgGrphGIZhtBBMQzcMwzCMVo4N6AeAZ58FNm2i/d57wOef0543T/OQ09I0J7y+nlptTQ3fP/WU1hIfPVrzoL/8UvOQU1JUw62poX9dHd8/9piup/3WW7q2+cyZujb45s3AM8/Qrqqiv89HHfeRRzQP+9//Zs46QN13zBjaGzZoPEB5OfPd/X6mUv3975qH/eqrqkFPmaIa8tq11HQB6sVDhlD/Doepe3t52C+/rBr0pEmqIa9erRpwYSFzspua+HrwQdbHB7jP6tW0x4/ndwCMAfDWec/L4zHDYZ7DkCGa02XBen0AACAASURBVD98uNaV//BDxkEAvCYvJz07m9cciTAt74EHeE8A5q57deXHjNGc+AULNJ4gI0Nzwn0++ldV8f0zz2g8wzvv8BkCwJw5Gk+Qmgo8/jjtujr619by/RNPaDzDf/+r8QizZmk8wZYtbHOGYbQsTEM/AKSns8AIwMHCUzlKS3Wgq61lRyzCQSQ1lYNBt24c7D3/3FzVbUtKdAGXmhodtP1+2n4/85fT0vTHQU6O5lQXF6t/dbUuINLYSDsYpOaalqY/DrKzNQ+6uFgHuqoqDVxraODxg0HqsTv6e3nwRUX88QIweMv70dDQQJ9QiDp0aqr6Z2Wxdj3AgTsUol1RAWRm0vb51B+g7R0nM1PzxQsKVEOvqNDAufp6HjMc5g+CtDR+J8Bz9IINCwo0v7ysTP3r6ugTifAepKVpvn16ug7OeXlah6CsbPu24PkHAup/2GHbt4W8PNXWY9vSrtpC167bt4XcXK11sGNb8tqCYRgtB9PQDcMwDKOFYBq6YRiGYbRybED/DqxdyxrcAKePPQ02EGB+uDdN/MYbOp08dapqyKtXqwacn099GeD06IgROs37+uucqgaYN+zV4l6xQjXgnBxd27qhgf7eNO+rr2oe9cSJmhO/dKnmpGdmsjY7wCnnESN4HQDXyvbyqMeN07ruixZpTnpammq4tbXMl29qooTw0ktaS/zDDzUPev58jSdISWF8AMAp6ZEjOWUeiTDf2svJHzNGNeg5czSeIDlZ4wEqKqiVh8N8vfCC5mG/+65q0F98we8A+J1ePEBZGY8pwnMYOVKnuUePVg16xgzNaU9K0nXai4t5zSKcch8xQjXst95SaeKzz7Qu+5o1Wp+goID3HNC25EkG//mPSgtTpmiN/1WrtD5Bbq7q+Y2N9PfKyo4apVPzkyZpjf9ly3Sd96ws1fONlo3fn4+8PBaLCIf9yMkZgXCYHUt+/utobMwGAJSWTkZNDTuWmpoVKC2dFPXPQX7+qAN/4sa3wgb078CmTcDs2bRzc9khesVEEhM1EOrTT7UTX7hQB8T16zVALTubg7VXTGTyZNVap0zRTnz+fA3KSkrShT6ysrRDrqqivzcIJSbqD4K5c3lcgOfhBbhlZGiAV0UFfWpreT2JiaqvfvmlDqirV+uAlJbGAQrggDh5MgehUIj+XlDbrFm60MrKlfrjZts2DRArKaFPQwMHxMmTOcgBDALzBuTly3VA27pVfxwUFdHf7+crMVF/UMyYoQPysmUaYJecrAVvCgp4zECA55CYqIVdpk1TfXrJEv1xFNsW8vLoEwrxHiQm6g+STz9VfXrRIi0Ss2GD/rjw2lIkwmeQmKg/SKZO1R+HCxZowaGkpK+3JYBtILYtJSaq1j9vnraldev0x0lmpv7QM1o2gUAOysoSISIIh2tQVjYZTU0MAikrmwK/nx1LdfV81NWxMdTXJ6Gqih1LY2MWysoSD87JG3uNaeiGYRiG0UIwDd0wDMMwWjk2oO8lS5fqNGVqqmq4lZWaUx0OM9/c7+f7SZM0D3rRIq2rvnWrTrOWl6uG2tRErTcY5PuJE3XKd/58nbJOTtYp99JS1eODQfp7KVHjx+uU79y5zDMGOHXuTbkXFakGGwjQ30v7+ugjTdWaPVvXFk9K0inzggKd8m9s5PV7y3mOHavT/zNn6pTz2rWc9gY4zexN+ft8rJHu+X/wgWrQM2ao/LB6tWrA2dmq59fVUQ8X4WvMGNWgp01T+WHlSs1Jz8zkdwOUPLw14yMRnosXzzB1qsYjLFumU95paTplX1WlerrXFrx4hsRElQ8WL1b5Y9s2nbKvqNDYjFCIz8KLZ/jkE5UPFi5U+WPLFp1yLyvjMwe+3pYmTNBUw3nzVL7YtIlT+ADb2sSJMFooPt9WVFayYwkGy1FSwo4lEgmhqOhdRCJsDKWlExEMsmOpqpqP+vrNUf/k5in3YLAUJSUTov7BqH/TAb0eYy8QkRb7Ouecc+RA84c/iDzxBO3XXxe58kraCxeK9Okj0tQkUlwscswxIqmp3HbWWSJTp9K+/XaRoUNpv/yyyHXX0Z4zR+Skk0QiEZH8fPpnZnLbqaeKzJhB+6abRIYPpz1ypMiNN9KeOVOkXz/a2dn0z83l+5NPFpk9m/Zvfyvy0ku0hw0Tue022p9+KnLmmbTT0uhfVCQSCon07SuyYAG3XXWVyGuv0X7qKZHf/572pEkiAwbQ3rJFpHdvkbIykUBA5IQTRJYu5bZf/lLkzTdpP/KIyD330P74Y5GBA2lv3Ej/qiqRhgaR448XWbmS2y65ROTtt2k/+KDIn/9Me8wYkZ/+lPbatSLHHitSW8vXscfyMxHu88EHtP/8Z36HiMjo0SI//zntlSt5zIYGkcpKnsvGjdx2/vki48bRvvtukUcfpf3GGyKXX057yRJeczAoUlrKe7l1K7f96EcikyfTvvNOkaefpv3aayJXX017/nze83BYpLCQ/mlp3HbGGSKffUb71ltFnnuO9osv8tmK8FmffDLt3Fz6Z2fzfb9+Il98QfvGG0VeeIH2P/4hcvPNtKdPFzntNDFaKLm5/5JNm9ixVFTMkRUrTpRIJCJ+f74sW3aMNDSwY1m16lQpL2fHsnnzTZKdzY4lO3ukbN7MjqW8fKasXMmOpbExW5YtO0YaG3MP9CUZMQBYI7sYE01DNwzDMIwWgmnohmEYhtHKsQF9L9m8WVPQCgo07cjnUz1chClUnga9cKGmDW3cqBpwXp5qsPX1qoGKUM8Nh/l+wQIt15mUpHnEOTma011bq2lHkQj9PQ163jzNiV+7VjXgrCzV86urVU8Ph+nvTd7MmaMa8ldfqQackaE16isrqQkDvG4vhQygNuxpyKtWaTxBWppquOXlmoLW1KQ1ygHGKXjxCCtWqIa8bZvWuC8t1fz8QED1bIC2p0EvW6bxBFu3ajxAcbHq8X6/xkYAPBcvHmHJEk0hS07WeIDCQtXjGxq0RjqwfVtYvFjbwqZNmoKWn6/pjDu2pdi2sHChxiOsX68paLm5GttRV6exFTu2hfnzNR5h3TpNR8zOVj2/pkb1dKPl4ffnoa6OHUsoVI/KSnYsIoLy8hkQYWOqqlqAUIgdS11dEvz+nKh/DurqkqL+taiqmh/1j0T9Iwf0eoy9YFdz8S3hdTA09BtvFPm//6M9cqTqtrNni3TtSg29oECkfXuR5GRuO+441V2vvVbk73+nPWyY6rbTp4scdhg19Jwc+nsa/NFHq+565ZUijz9O+6mnVLedMkXkiCNop6fTPyuL73v0UN31sstEnn2W9iOPqG47YQK1YhHqve3bU8sPhUS6dVPd9cILVcMfMkTk+utpjx1L3VeEenP79iIlJdTQu3QRmTeP2wYNUg3//vtFbrmF9jvviJxyCu01a0Q6dBCpqBDx+UQ6dxZZvJjbBgwQGTWK9l13idxxB+033xTp35/2ihUiHTuK1NTw1bGjavBnniny1lu0Bw+mDi5CDdtrTosX85g+n0h5Oc/F0+D79RN5913aN9+sGv6LL4r8+Me0587lNQcCjKdo3141+D59RD78kPb116uGP3w4760I73W3brz3eXn09zT43r35rET47B55hPazz/LZivBZ9+xJOyuL/unpfH/EEWwrImw7Tz1F+/HHNR5k0iS2OaNlkpX1nCQlXSIiIuXlM2TJksMkEolIY2OOLFrUXnw+dizLlh0tJSWTRERkw4YrJSODHUtm5lOyfj07ltLSKbJ0KTuWhoZ0WbSovTQ0ZB3gKzJigWno+45IhAuGOMf34bAukBEKcTGS3dnhMBAXt2/8RXg+B9rf2//b+MfaXhR6XNy38/f+6jyY/odCWzBaHiJhOMeHGYmEEBfXZrc2/2qPg3MOHBMi3+hvHBx2p6HbgG4YhmEYLQQLituH5OVpTnh1terpTU2qQQLUQ73fSps3q4ack6MabuySocGg6tmev8emTaohZ2VpSdnYJT8DAdWzd/TfuFHzkDMzNae8rExzshsbNb9dZHv/DRtUQ45d/jN2yc2GBs1vF9HYAID3JXYpUy8eoLhYS8L6fKqHRyLb+yclqYacmqoacGGh6vl1dVrSNRJRPRmg7f01npKi8QQFBarne8vXAjyWF5sA8Fw8/61bNZ4gL0/1/NjlZ0OhXbeF5GStq56bu/3ys15baGradVvYvFnbQna2tqXY5WMDAc1P39F/40aNJ8jK0niA8nJtC37/9m3JaFk0NVWisZGNKRIJor5eG1NtrTaG+vpNCIfZmBobsxAMlkf9K9DYmBX1D6C+ftNO/Y3vIbuai28Jr4OhoV93nch999F+7jnmJYswT7xDB2ro+fkizqluevTRzJMWoU7pafBPPaUa/JQp1G0jEeqezomkpHBbjx6qwV92mWrwjzyiGvyECSLdu9NOTaV/Rgbfd+miGvyFF2oe/ZAhIr/6Fe2xY0WOPJL25s30z8ujjtuxo8i0adw2aJDm0d9/v8g119B+5x3me4uIrFsnEhdH/TgQEElIEJk1i9sGDBAZMYL2XXeJ3HAD7TffFDnxRNorV9K/vJw6dtu2qsH3768a/B13qAY/apRq8EuWiLRpoxp6mzb8TIT7eBr8LbeoBv/SS6rBz5sn0q6dauhxcSKrVnFb376qwV9/vWrww4erBv/FF9StAwHm8sfFiSQlcVvv3qrBX321avDPPqsa/Gef8Z6HQswjd07jMY44QjX4yy9XDf7xx0Uuuoj2pEl85iJsA85pPEa3biITJ9K+5BLNo3/oIZFf/IL2xx+rBm+0PDIzn5J169ixlJZOkUWLOkkkEpGGhixZsMBJfT0DMpYs6SHFxexY1q+/TNLT2bGkpz8iSUnsWEpKJsiSJexYfL5UWbDASUNDxoG+JCMG7EZDP+iD8nd5HYwBva5OpLGRdjAoUl2t28rK1C4tVbu8nEVCRFjoxPMPBPbcPxKhXVMj4vfT9vv5XoTbd+VfVra9fyCg/rW1tMNhHmdX/h7V1bxuEV5HXZ36V1Ts3D/WrqpS/4YGkfp62qHQnvs3NdH2+bb3r6z8Zv/KSu4rQl+fj3ZTE7/7m/wrKtS/ro7XIMJr2lN/ry3EtqVv0xZqa7UtBALfri3EtiWvLUQi27cFo2URDgekqUkbUyBQFmNrYwgGyyUSbQxNTTUSDvuj/n5pamJjikQiu/Q3Dg67G9BNQzcMwzCMFoJp6PuQqiqtC+73q4YponowoLneALVeTwOurFQNtrFR9fBIRPXg3flXVKgG29CgGuju/AsKVAMuL1c93+dTPT0c1jXPd/TPz1cNuKxMNdz6etXTQyHVo3f0b+nEXktxscYTVFerHr+nbaGoSOMJYttSbFv4Nm3p27SF2Lb0bdpCbFsyDgx+vz6MQKAQXk55U1Nl8zrn4XBjsx4uEkEgULAH/hUIhxui/g1oaqrYhX/u/rgsY1+xqz/dW8LrYEy5X3utaujDhqmGPn06dVNPQwdUQz/qKNXQr7hCNfQnn1QNPTFRpFMn1dABzT0+7DDV0C+9lHqniMjDD1MHFREZP576qIjItm309zT0zp2pq4qI/Oxnmsf+wAOax/7BB5rHvmkT/XNzOb3coYPmsQ8cqHnsf/qT5rG//bbmsa9dS922uHhv7+73j/JyXounoffpoxr6b3+rGvo//qEa+syZjBvwNHTnVEM/5hjV0K+6SjX0Z55hfIII6+p36KAaOsC4BhGRww9nvIMI6+IPGUL7scc0j/2TT1RDT0+nf6yG7uWxX3yx5rH/7W+ax/7RR4zbEGFdfoC14CMRavteHvsFF2geu7H/8fsLZMECSF3dBhERWbr0KCksfF9ERDZsuEJSU/8qIiIZGU/K2rU/ERGR0tLEHTR0xGjoh0lx8cciIpKUdKmkpbFjSU9/uDmPvbh4vCxezI7F59smCxbANPSDDExD33cUFanWW1vLIjAi7Hy9IDYRDWIS4cIanlZZWKhab00NA89E+ENg2zb18TpwEXbGnu6cn69abXU134twu9dp7+i/bZvqznl5qrVWVrIIjggHH28BkB3PPyVFdePcXNVaKyp4P0R4fV7xkh39WzrJyao7Z2So7l1crFpzXd32bcH7Meb5e6Snb98WYtuSt5jON7UlLwaioGD7trAnbWnbtu3bkqfbV1VpW9hdW0pJ0baUm6ttyTgw1NdrY/D50pp1b7+/UIJBdixNTTXS2MjGEA43ic+3LcZ/c4x/qoTDwah/vgSDVVH/avH786P+weZCNDv6GweH3Q3opqEbhmEYRgvBNPR9SDCoGmgkonoyoHrk7uxgUDXQb+MfCKh/OKw5xXvj72mo38bf71f/UEjz23fn09KJvZbGRtWQ90Vb+C7++6ItfBf/3bUFY9/hadu7syORYLMeLhJpzi/fc/9AjH8YkUjgG32M7x82oO8lt9wCDBlCe+RI4Oc/p/3FF8BRR7FjKywEunXThUd+8APgo49oX3898PDDtIcNA371K9qffQb06sXBIieH/l6hkz59gE8+oX3ttcCTT9J++mng6qtpT54MHH887fR0+ntFZ3r3Bj79lPYVVwBDh9J+9FGeDwCMGwecfDLtLVvon5/Pjv6oo3SxlMsuA0aMoP23vwE330x7zBjg9NNpb9gAdO+uBXhaMhUVwGGHaXGW/v2Bd96h/bvfAX/9K+0XXwQuuoj2l18CRxzBAa64mPfCK/Ry6qnABx/Qvukm4KGHaA8fDvziF7Q//xw4+mje+/x8Pguv6M5JJwHjx9P+7W+Bxx6jPXQocOWVtKdO5TMHWGymWzctgHTccUBiIu2rrgKeeYb2E08A111He+JEoG9f2tu20T8nh22zVy9g+nRuu/xy4PnnaT/8MHDjjXt9e41voKJiFpYtOxKRSAiBQCGWLu2G+npWgFq1qh+Ki9mxJCffgIwMdizZ2c9h40Z2LOXl07B8eS+ICPz+HCxd2g0NDexYVq7sg9JSdiybNl2LzEx2LFlZT2PTJnYspaWTsWIFO5aGhnQsXdqtueiM8T1kV3PxLeF1MDT0bdtUKy0pEdnA+BRpbNTiJSIsTuLlG69cqbrz1q2qdRYVMQBNhPnMS5du7+/ptsuXa751crJqnQUFqq/W13M/EfrF+i9bpvnWmzZRuxXheWzZQru2louaiPC8vUIuIrwuTzfesEGD3XJyVOutqdHAsVBIZP78Xd/Dlsb8+aobf/WV6tapqQwWE2Gu9/r1tP1+XUxGhPfSi0FYtUp155QU1c2LizWIsqFh121pxQrN/d+yRWMoCgtV6/b5+MxFdt4WvLa0ebO2pfx8bUt1dV9vSx5Ll2ru/caNGkORm7t93ICxbwiFGqWqShtDZeU8iUTYGGpqVkpTEzuW+vqtzbq5318kdXWbov4NUl29dAd/Nobq6uXS1FQX9U8Wv78g6l/QrNWHQvVSXc3GEIlEtvM3Dg4wDd0wDMMwWj6moRuGYRhGK8cG9L3k9ttVA3/lFWrKADB/PnXPUIi6aY8eulhIv37UuAHqpp4G/sILqnvOnk3dU4SFPHr00MU6+valxg4Av/kNtXeA+uW119KePh044QTaWVn09xZOOfZYavwA8Otfqwb+9NOqe06Zohp6air9vSImvXoBc+dy2y9/Cbz8Mu1HHwVuu432hAmqoScnAz17aqGVlkxlJa/FWyzlhz8EPvyQ9p13Mo4AAEaNAi65hPaiRYw7CAa5+EqPHrpwzWmnUaMGgFtvVQ385Zd5bwHe6169tMBLjx4aT3HyyXxWAHDDDXyGAJ/pVVfRnjmTzxyg9t2jh8ZTHH88MGMG7WuuUQ186FBq8gDjLTwNPSOD/l5BmWOOYYwAwHiMf/6T9hNPaDyFsX8IBIqxdGkP+HzsWFat6ofS0kkAgOTkm5o18Nzcf2LjxisAAJWVs7F8+TFRDT0fS5f2aF64ZeXKvigvZ8eyefNvkJ3NjiU7+3ls2sSOpbx8OlasYMfS2JiFpUt7WHGZ7zO7motvCa+DoaGvWqW6d0aGyIIFtKurdQGUcJiFYLx84enTqbeLUAP1dOu0NJFFi2hXVmrBjlCIC2R4+cKffab5zkuXqm69bZtqreXlLEgiQr33449Vt50yRXPfFy/WHOMtW1QrLS3VBVgCAZ6/p9tOnqz5ygsWaMGazZsZHyBCDfjzz2n7/fRvDVJbJMKiPV4MwcyZGoOwerXq3llZGjdQU6OFfLy24OWez5ihMQgrV6runZ4usnAh7aoqFhqK9ffawrRpWpt92TLVrVNTVbevqBCZOpW21xa8GIBPP9Xc9yVLNF89JUV197IyLSQUDG7flhITNYZg0SKtPZCcrDEYxv4hEglLcfE4CYfZsZSVTZdAgB1LdfUKqa9nx+LzpUlVFTuWYLBSSkunRP1DUlz8cXPueVnZZxIMlkf9l4rPlxL139as2weD5VJWxo4lHG6S4uKPJRIJHYjLNXYBTEM3DMMwjJaPaeiGYRiG0cqxAX0v+etfgX/8g/bo0dTEAWD5cuqroRC149NOUw38pz9lbjEA3Hef6o5vvMFcZoC664AB1CmLiujvaeADB6pu+cc/UrsHgNdeA37/e9pz5wLnnUc7L4/+3gIb55wDLFhAe/Bg4N//pv3SS8A999D+4gvgggtoZ2XRv7SUOu7ZZwNLl3LbzTcD//sf7eHDgb/8hfa0aZqH3Zq57DLN4x4yROMZ3n2XmjYArFwJnHUWF3GpqOC9TEvjtgsv1DzuP/9Z4xn++1/WOAB4r88+m0VbSkro72ngF1wAzJpF+557+AwBPtPBg2kvWMBnDnAxltNO08VezjsPmDeP9p13UvsHgH/9C7jrLtqzZwODBtHOyaF/cTHb5o9+BCxezG233Qa8+SbtF14A/vSnb3VLjd1QU7McX331Q0QiIQSDZVi9+rRmDXzdup+iooIdy7Zt9yE3lx1LQcEb2LKFHUt19SKsWfMjiAgCgSKsXn0a/P4cAMDatQNRWcmOJSXlj8jLY8eSl/caUlLYsVRWzsXatexY/P48rF59GgKBmJV7jO8V8UO9KiMtkNGjRw+9xxuRDhDx8SwOctxxQNu2DF46/XQgIQHo2JGDb7t27Iwvuog2wA6ye3cgLo77H3ss/Xv35vclJACdO7PDjfVv2xZwDvjxj4GuXXn8M85gcJLnf8op9O/alR15QgI734suAtq04fEvuADo0oXH79+f5922La+jXz/6dOvGHxUJCTzmz37GfUT4o6RzZ/qfdRYDANu2ZSDeySfznA87jANRa8Y5PqMePfgs+vVjoFm7dgyEO+MM3r8OHfjMd2wLzrGNdO++87Z02mn079QJOP98+njPsm1bnsOgQXxWXlvq3fvrbalLF+Dcc2l7x49tC1270v/MM7Ut7aot7NiWfvKTr7eFNm3YFn7wg4PzXForcXEJiI/viG7dBsK5dhCJoHv3ixAXx46la9dBaNOmO5yLR6dOpyEh4Vg41w4JCcegU6dT4VwC4uM7o2vX8xAXF+vfFoBD164/Rps2XQHEo1OnM5CQcAzi4toiIaE3OnY8JXr8rujS5Rw4lwBA0L37RXCuzcG8LYc0w4YNKxo6dOjonW0zDd0wDMMwWgimoRuGYRhGK8cG9J1QW8ucYk+3vOkm1bCffRZ4/XXa48cD999Pe9061rYOh5m7fMkl1LIB5vd6GvYTTwBvvUV77FjggQdor1rFvF4RavAXX6wa+DXXqIb98MNaS/y997QW+LJlmodcXEz/0lK+v/JK6roAdV+vlvj//qd50AsXMscdoN568cXUfyMRXpdNhOw569czpzwUAqqr2RZycr6+31NPMY4CAD7+WOMR1qxhjf9IhM/g4otVA7/uOsZbAHx2XjzDmDHAgw/SXrFC6xuUlNC/uJjvr7qK8R4A285779F++23gkUdoL1nCNgewDV58MVBezrZ5xRXA6tXc9n//pzn5b76p9RXmz9c1AnJzef1VVfy/cfnl/L9i7Bl1deuwYcPlEAmjqakK69df0pwHvnnzb1FVxY4lM/MJFBSwYykuHou0NHYstbWrsXHjFRARBINlSEq6uFkD37TpGlRXLwEAZGQ8jMJCdixFRe8hPZ0dS03NMmzcyI4lEChGUtLFCAbZsWzceCVqalYAANLTH0RR0RgAQGHh/5CRwY6lunoRNm3iIgF+fz6Ski5GU1MFRCLYsOFXqK1lx5Ka+heUlHy8P27hIYUJITshIYE6ZdeufH/uudQZAWqc3brRPuEEDv4AcPjh1Ebj4qifDhxIHROgFnr00bTPOIMLdwBcdKWpifaRR/KYzlGLHzSIOiXA7zrqKNpnnkmtE2Dxj7joT7KjjuJ+APXXQYP4r+d/5JG0zzqLxwWAE0/UfY4+mucJ8LwHDeJ5xMXR9s7Z+GZ69uQ9j48H2ren7bWlWM44g1o8wLbkrWQW25a8trCrttSrF+2+fXXlM68tAbtvC/37azGiE09kuwe2b0udO2tb8PR/ry3EtqWTT9b/Fzu2pYED+X8iLo724Yfv9S09ZGnb9nB07ToQQBzi4jqgS5eBiI9nY+rS5Xy0a8fG0KnTGWjblg+mffs+EGmK+h+BLl0GwjmH+PiO6Np1EOLj2bF07ToQ7dodFfU/EwkJx0X9+8L7W69t26Oixwfi4ztF/TtF/QehXbsjo/790aFD36j/SYiL6wgAaNfuaHTtysbQpk0XdO06CHFxHeFcHLp2HYi2bdkYOnf+IRISTtgPd/DQwjR0wzAMw2ghmIZuGIZhGK0cG9CjpKYyJzwUAnw+5lsXFHDbX/5CjRpg7renQc+YoetJb9nCPOBIhNPwN9+suuV991EjB1ize9w42lOnAs89R3vjRuYFA9Qbb7pJa6HffTewdi3tkSN1bfRJkzSPed065qgD1DtvuolaPsBcda8W+fPPay3w8eO5jjdAXfTee2mXlNC/poa66eDBura7sQekpLDofzgM1NWxMXgBEfffT5EbWJxPngAAIABJREFUrCfwEZezxrRpjM8AgM2bgTvu4L2vqeGz8NaWv/de4KuvaL/4ImvoA3ymXn2E9eu1PkFlJf0rKvj+j38EkpJoDx+uawxMnMhccoAa/t130y4ro391Nc/nzjuBTZu4bdgw1n0HGAPg1fhfuVJz0ouKePm1tfy/MXiw1rV/+mmtK2/sHJ9vC7ZuHQyRCEKhWiQn34xAgB1LauqfUFvLjiU39+VmDbqs7FNkZ7Njqa/fiK1b7wQANDVVITn5JgSD7Fi2bbsbdXXsWHJyRjavjV5aOgk5OexY6urWISWFHUswWI7k5JvQ1MSOJSXl96irWw8AyM7+B8rK2LGUlExAbi47ltrar7Bt271R/xIkJ9+EUKgGIoKtW+9oXts9K+tZlJdPAwAUF3/YnBNv7B37fUB3zsU755KcczOi7y91zq1zzq13zi11zp0c/TzBOTfROZfunFvlnOuzv88tlvh41Qk9HdvTpzt21BzchATVGtu2pUYKcHuHDvT1dPT4+J37e7np7dpt79+RslOzdur5d+ig/u3b77m/d/576t+hw879Y8/F2APi4/eoMcS2hbZt9f7v7FnsrC3s6L8v28LO/J3b+7bk3YpdHd/LrTd2jnNtEBcXvZmIQ3x8BzjHmxkX16E5HzwuLiGaJw7ExbVFXFz7Zv/4+A5ROw7x8R134d8ezrWL2u128O+4nb83bFALjz1+2xj/hJ2cf/wO/rHH7xDjn9Dsb+wd+11Dd879DcC5ALqKyFXOuVQA14rIVufc/QDOF5E7o/ZZInKfc+5mAL8RkZt2992moRuGYRiHEgdNQ3fOHQvg1wDeiflYAHgxv90AeHUErwUQnczGZACXOufc/jw/wzAMw2gt7O8p99cAPAIgEvPZXQBmOufyAdwOIKrcoTeAPAAQkRCAGgA99/P5NZOby7zcSATw+5mv7WnYzz9PXRJg7ri3NvmiRZqTnpnJHHERph898IBq2M8+q7rje+9pXfd58zQPOS1Nc8Lr6+lfU8P3Tz0FbN1Ke/RoreX95Zeah5ySonnANTX0r6vj+8ce0/W033pL1zafOZM1yAHqtl48QFUV/X0+Xs8jj2hdemMvaWzkzSwv3+1uCxZojf2MDM0J9/noXlXF9888o/EM77zDZwgAc+ZofYPUVODxx2nX1dHfS6984gm2FYD14736CrNmMRcdoMb91FO0q6vpX1/P948+CqSn037jDeacA9TC33+f9qZNXF8doHb/wAP8PyHC/yNefYdRozSn3tg5jY1ZyMh4GCKCcLgBaWkPNGvYWVnPor6eHUtR0XsoL2dAQlXVPBQUsGNpaEhDRsajAIBQqD7qXw0AyMx8Cj4fO5bCwtGoqGDHUln5JQoL2bH4fCnN66yHQjVIS3sAoVBd1P9xNDSwYykoeAuVlexYKipmNue0+3zJyMpix9LUVIW0tAcQDvsAABkZjzTXpc/P/3dzTn15+bTmnHZj79hvA7pz7ioApSKydodNDwK4UkSOBfA+gL2KfnDO3eOcW+OcW1Pmjbj7gLo6DqrhMHPDU1M1Lzg9XQfnvDwG+gAc8LOzt/ePRIBgkHZjo/pXV6u/FyxXWqoFR2preUwRIBCg7fmnpal/bq4GSJWU6AIuNTU6aPv9tP1+9fd+HOTkaMGZ4mL1r67WBUQaG2kHgzyftDT9cWDsJTs2hl1QVrZ9W/DaUiCwvXtsW8jL07YQ25b2tC3k5mpb2LEteW3B8w8EtC14Pw6ys7dvS14hpaoqHfRj21Ik8nX/ffhfuFUSDteioSENQASRSBCNjWkIh9kxNTamIxRiYwgE8tDUxMYQDJY1L8ASDteisTEtul52AI2NqYhE/FH/tGZ/vz83xr+kuXhNOFyDxkY2pkjEv51/Q0MqQqGaqH/Odv6BABtDKFSNxsa0qH8jGhvTEIkEIRKJHr82xr8s6l/c7G/sHftNQ3fOjQT/Ag8BaA9Osy8AcKqInBTd53gAs0TkdOfcbABDRWSFY6REMYAjZDcnaBq6YRiGcShxUDR0EXlcRI4VkT4AbgYwH9TJuznn+kV3+wWA6GQypgG4I2rfAGD+7gZzwzAMwzCUA5qHHtXG7waQ6JzbAP4F/3B087sAejrn0gH8DcBjB/LcSkqY1yvCKfeRI3Vq8r//1SnM6dNZ9xxgbvjH0fLDRUWahxsIMD/cm6Z+4w3VoKdOZa1sgLnfXh5xfj7w6qu0/X76+yg14fXXdWp/8mStxb1iBXPRAU63emtbNzTQ35umffVVnQ6dOFFz4pcu1Zz0zEzgP/+hXV9P/0Bgb+/iN/Phhg+RVMRE6PlZ8/F5KgMKUspTMHotVwSsaqzCyCUjEYqEEJEIXlz2Isp8nI4bs34MNpZs3PcndpBZs0brExQUcH1yQNuSp2H/5z98VgCfnVfjf9UqrU+Qm8t6CQDbwIgRKh+NGqVT85MmaY3/Zct0nfesLNXzfT76e1P2r7yideUnTNCc+MWLNSc9PV3XSa+ro38wyPcvv6yS1ccfa32FQ5GCgv82a9Dl5dNRVbUQAFBXt7Y5pzwQKEJuLjuWSCSInJwRzRp2QcEbzRp0WdmnzXXZa2tXo6SEHYvfn9+c0x0O+5GTM6JZw87Pfx2NjdkAgNLSyaipYcdSU7MCpaWTov45yM8fFfVviPqzY8nLew1+f17Uf2JzTnx19dLmnPTGxkzk57NjCYXqkZMzApFIIOr/LwQCLPZRUjKuua57dfUilJd/9t1u7iHKARnQRWShiFwVtaeKSH8R+aGIXCwimdHP/SJyo4icLCLne58fKAoK2KE1NbETmzxZ9cFp03RAX7RIF6fYtAmYPZt2bi79vVoiiYkaB/Xpp6opLlyoC52sX68BatnZPKZXTGTyZA2EmjJFO/H583Vxi6QkBtYB7IS9Drmqiv6e1pqYqD8I5s7VAL81a/THSUYGf2wADGRKTFStc18yK2MWNpeymMTK/JVYkstOaFv5NkxPnQ4AKPGVIHFrIhqaGhAMBzF5y2QU1PE//sy0mUgubX1VbjZsYGAboG3JK1KUmKiFYaZO1R+HCxbogJiU9PW2BLANxLalxEQNSps3T9vSunUa4JaZqT/0Kivp7xUZSkzUHwRz5uy8LaWn6+BeXk6fujoWbUpMVK1+9mwNFj0UqaiY1jyg19QsRl0dB8T6+k2orGTHEgjkorw8EZFICOFwHcrKEtHUxI6lvPxTNDayY6muXoC6uq+i/htQVTUn6p+DsrLEaFBdDcrKJjcH1ZWVTYHfnxn1n4+6unVR/yRUVbFjaWzMQlkZO5ZQqAplZZObdfeyssnw+9mYqqrmoa4uKeq/FtXVC6L+GSgvnxr1r0BZWSJCoVqIRFBWltis1VdVzYHPx+pXtbVfobraoiW/DVbL3TAMwzBaCFbL3TAMwzBaOTagR6muZo45wKnO995T3XHKFNUNly7VacrUVM0Jr6zUutzhMP093XHSJNUNFy3Suupbt+o0a3m5aqhNTcwP93THiRM1PWn+fOaMA8xH9qbcS0tVjw8G6e8tzTp+vKYHzZ2rtbQPFGM3jEW1n9N0M9NmIq2CaSxrC9diWS6L5OfW5GLqVk7N+YI+vJ/0PiLC8gUfrP8AtQHO/89InYHMKk4Tri5YjZX5Kw/otewvtm1T+aaiQmMzQiE+Sy+e4ZNPNO1x4UKuAQDwmXpT7mVlfObA19vShAkqJc2bpzntmzZxCh9gW5s4kXYgQP9QiO/HjdPp/y+/1Jz2DRs0p7ywUKf8/X7+XwiH+f6jjzQFdNYslbIOBUQiKCp6rzntjFPe7Fiqq5c2T3k3NKQ254Q3NVWiuPijqH846s+OpbR0EgKBoqj/ItTXs2Px+baispIdSzBYjpISdiyRSAhFRe8iEglG/SciGGTHUlU1v7muus+X3DzlHgyWNuvxkUgw6s+OpaRkfHNd+MrKufD52LHU129sjgcIBIqba8RHIoGofyjq/zGamiqi/rPR0LANAFBXtx7V1Yu/y60+dGF+Yst8nXPOObKvWLZM5IQTRPx+kfJykd69RZKTue2cc0QmTqT9hz+IPPEE7ddfF7nyStoLF4r06SPS1CRSXCxyzDEiqancdtZZIlOn0r79dpGhQ2m//LLIddfR/v/sXXecXWW1XffOTGYmZdIoCZGmFOmIQiLwgCDKAws8FSkq6sOHqKAgAioECArEgBSRIjUGCcE0IRFSSCYzmUmdTC+Z3uvtvZ1z1vtj3/t9k0BiEkJMwlm/3/zYw7kr555z99nnzll77W/VKvIznyEti+zpEX5bm2z77GfJZcskvvZa8uGHJX70UfKaayR+913ypJMk7ugQfleX/H7CCeSKFRJ/85vkY4/tk1O2W0gYCR775LEs6SwhSX7l9a/wuc3PkSTvXnk3b37nZpLkG9VvcOpLU0mS1QPVnPKnKfTFfIwmozzmyWO4sXsjSXL6nOl8aetLJMk7lt/Bn//r5/vvYD5GPPUU+fWvS7xmDXn88aRpkn198lk2N8u2004j335b4htuIB96SOLZs+WzJeWzPuEEibu6hN/RIb+fdBL53nsSX3MNOWuWxH/4A3nddRIvXUqecorEra3C7+2V3Pz0p8n335dtV11F/ulPEj/wAHnjjRIvXkyedZbEjY3CHxyUa+O448iiItl25ZXkM8985FN30CCZ9LC0dArD4VqSZFnZFzg4KIWloeEmtrZKYenu/jOrqqSw+HxF3LDhOJpmionEAEtLj2Ik0kiS3Lz5TA4NLSZJ1td/n21tD5Aku7r+xJoaKSwezypu2PBpWpbFeLyHpaVHMRqVwrJp02fpdkthqa29lh0dUlg6Oh5lba0UFrf7XW7cKIUlFutgaelRjMWksGzceAI9HiksNTXfZGfnbJJke/tDrKu7gSTpcr3NTZtOI0lGIs0sLT2K8XgfLcvghg3H0+tdQ5Ksrv46u7ufIkm2td3HhoYf7pNzfigCQBl3ck+0NXQbNmzYsGHjIIGtoduwYcOGDRuHOOwbehqxmNYwAZm3ntGgi4u17ldbqy1ovb3ahxuJaD2cFH5Gd1y7VtuGqqu1Ba27W9uOwmGtgZIyGzujOxYWak98RYW2DXV26rWtg0FtO7Is4VvpCfqrV2tP/Nat2pO+v7CiZQViKfGuburZhP6Q6H7NnmZlQXNH3SjpElN1ykzh3eZ3FX95y3LEDdENN3RvwEBYRORGdyMaXA04FNDTo+2MO+bS8FxYu1bbESsrtQWtq0v3doRCurdix1xYs0bbEcvLtYWso0Nb0AIBraebpvAzD/Lef1974svKdG9JW5vW830+racbhl67ABDdPTNfYcsWuYY+SfB4/qU0aL+/WFnIwuFaRKNSWBKJXgSDUlhMM6L0cJJpvhQWn28tUilfml+NWEwKSzzerdY5N4ywmrFOEm73MpBmml+oRreGQhVqXGw83qksaIYRhM+3Js230nwrzV+tPPGhULmyoMVi7UrPT6X8Sk8nzTRfksnrXaU88cFgmeoniMVa1Yx6G3uInT2LPxh+9qWGXlhIjh5NxmLk0BCZn09WVsq2z3yGnDNH4muuIX/xC4kffZS88EKJV6wgCwpEJ+ztJfPytAZ/9NHkvHkSX3UV+etfSzxzJnnppRIvXUqOHy86ZWen8DMa/KRJ5MKFEl95Jfnb30p8333k5ZdLvHgxefjhEre0CL+9XX6fMEHrrpddJnrn/kLCSHDMI2O4um01SXLay9P4WKmI+D9b9jNev/B6kuTLW1/myc+cTJIs6y1j/h/y6Yl6GElGOPqR0SzuKCZJnvPXc/j0xqdJkj9++8f8wZIf7L+D+Rjx8MPkRRdJ/N575NixpGGQ3d3yWTY0yLYpU8j58yX++tfJu++W+IEH5LMl5bOeOFHi9nbht7TI74cfLrlCSu7cd5/Ev/2t7gdZsEByjhQNPC9PctKyyHHjdD/H9Olaw7/zTt0P8sYb5DHHSFxbK/y+Prk2CgrIlStl2wUXaA3/k4BEwsWionwGgxUkRYPu63uNJFlb+x02Nd1GkuzsnMWtWy8gSXo8K1hcXEDTTDEe72NRUR7DYSks69cfw4GBN0iSNTVXs7n5TpKiYVdUTCdJut3LuG7deFqWxVisk0VFeYxEpLCUlk7i4OACkmRV1ZVsbZXC0tZ2HysrpbAMDS1mSYkUlmi0hUVFeYxG20mS69ZNpMslhaWy8jKl4be03M3qamkIGRycz9LSKSTJcLiBRUV5jMW6aVkGi4vH0u2Who7y8ouUht/cfDtrar61D874oQnYGvruwTCA7Oxdx5YFOBzyA8hfMFlZu883TcDp3Dd8Ut7PR9n//oBhGch2Zn8gJgmCcDqcu3zdzuJMF3yGf7DjP5kLe8vPvP6j8j8psCwDzkz+Dovlr14HMitGkyYcjqxdcrbnmwCc+4Qv9wRrv/Mzr9+Rb2N77EpDt2/oNmzYsGHDxkECuyluN2AYWo8GRGvOfNepr9ee9OFLVvr9Wk9PpbQGCYi+mOHX1uq56p2d2hPu9eoxnsmk9qdn+BnU1GhPe3u7Hinr8WgNNZHYfozmcH51tfYht7XpfoD9hcqBShhp3a/Z04xAXHS7gfAAeoKim0WSEaWHW7SwtU8P+a7or4Bpie7X5GlSnvS+UB96g4eGCOvz6VxIpXaeC7W1OheGLz/q8ejejERC69k78qurtae9vV17yt1uPR44Ht95LlVV6VxqbdW5NHz51lhM+9vJ7fmVlbo3Zfiywp8UhEJblYYcidQrT3o83q084amUX+nplpVCKKQLSzBYpvjhcK2aqx6PdypPeCrlVTPeLSup9OwMP4NwuEZ52mOxdiST7jTfg1isPc1PbKdnb8+vVnPZY7E25SlPJl1qRrxpxpW//YP8KtVPEIu1qn6AZHJI6fE29hA7exZ/MPzsSw195UoyN1c09MFB0ukky8pk2zHHkC++KPHVV5O33CLxQw+R550n8bJlorunUuIjdzjI6mrZNmmS1uCvvFJr8PfdpzX4xYtFw7cs0T0dDnLbNtk2YYLW4C+7TGvwd9+tNfj580XfJEV7dzjEQ0ySY8ZoDf6ii7SPfn8gYSSY+/tcLm9eTlI08EeKHyEpGvi3//FtkuRzm5/jp5/+NElyY/dGOmc66Y64GUlGmPNQjtLgz3juDKXB/2DJD5QGf7DjgQfIL35R4rffJkeOFA29q0s+y0w/xuGHk6+/LvHll5N33CHxb39LXnyxxAsWyGdOSg44HLofY+xYPVNh+nTynnskvvNO8stflviNN7QG39Ag/I4Oyc1Ro/RMhQsuIGfMkPi228ivflXi114jJ0+WuKpK+L29cm3k52sN/rzztAb/SUAiMcTCQieDQSks69cfw95eKSw1NVezsVEKS3v7Qywrk8Lidi9jUVF+WkPvZWGhg6GQFJbS0kns75fCUlV1JZuapLC0td3H8nIpLENDi1lUNIqWZTEabWdhoYPhsDRkrFs3gQMDUlgqKy9jS4sUlpaWu1lRIYVlcHA+162TwhKJNLGw0MFoVApLcfEYpcGXl1+sNPjm5juUBt/f/7rS4MPhOhYWOhiLddGyDBYVjVQa/NatX1QafGPjz5UGb+ODwC409P/4Tfmj/OzLGzopzXAfFns8MuSDJEMhuemTZDJJ+v36dS7Xh/Pdbs0PBjU/kdh9vmVJHAjI8BtS/hsISGxZO+e7XNvzE4kPHvvHiaGwfjO+mI9JI0mSjCajDCfCJEnDNOiJej6UsyM/ZaZIkpFkRPEPduxNLgSDOhcSib3LheG5FAxqvtu9c34Gfr/OpVhM801z5/zhsd8v19AnCYmEPgHJpIeWJYUhlQrRMKQwmGaSqZR/GMe1E757GD84jJ/YA76V5gdomvE0P85USpLJsqyd8hMJ1w78xDB+UPGTSfdO+RmkUn6apiSDYcSYSoVo48Oxqxu6raHbsGHDhg0bBwlsDX03kfHUAtt7tfv7tafc59M+3Hhca5jkzvl9fdpH7PVqH24spvVwy9rek7szvsej9fxoVGugu+L39mofstut9fyPinAyDF9MdC/DMpS/HAC6A/oN9AR7lO43FBlCwhDdLZQIqRnvKTOl/OU78ofHw/nBRFDp8UkzicHw4L/lD4YHkTRFBA7EA0qPTxgJDEWGPpSzv7E3ubQ3uTA8lyIRrYebpuznw/g9Pbo3ZHguhcN61oJh6LULAO1135Hvcul+gE8KMuuHA7LWecZTnkr5YBhSWEwzrvRwksqf/UF+HzKe8lTKqzzdphlTejhpqTXHd833KD3fNKNKD/8gv2sYvxcZT3oy6VZ6vmlGlL+eNJFI9A3jdw+LdV0QviSDYYSVnm5ZhppXvyu+jTR29qf7wfCzLx+5r1hBjhihNXSHQ2voRx+tNfSrrtIa+syZWkNfulT8thkNHdAa+pFHag39iiu0hn7vvVpDX7RI9MmMhg5o7/H48VpD/9KXRO8kybvuEh2UJN98U/RRUrzDgNbQR48WXZUk/+u/tI/9o+Kny37Kr88TreulrS9xyp/Eb7q1bysdDzo4EBpgwkhwxO9H8L1m8Zt+7oXP8eFi8Zve9PZN/NZb4jd9dvOzPP6p40nuWkM//bnTlYZ+45IblYb+1IanlI+9uKOYWTOzGIgHGIgHmDUzi+s615EkT3rmJOVjv27hdcrHPrtkNs947gyS5Put7zPnoRxGkpF9c6L2EP/8p2jNGQ0dED83SR52GDl3rsRf+Qp5++0S/+Y32sf+j39oDb2lRfjDNfSMj/2SS7SP/Ve/0j72v/9d+jZIsr5e+BkNfeRI7WM//3ztY7/1Vu1jf/VV7WOvrBR+RkPPy9Ma+rnnfhI1dAcDgS0kyfXrj2Zv719JktXVV3Hbtp+QJNvbZ7Ks7FySpMu1lEVFecM0dDAUqiJJlpQcqXzsVVVXKB97a+u9ysc+NLRoBw0dwzT08crHXlHxJeVjb2m5S/nYBwbeZHGxFJZIpJGFhRimoY8epqFfNExDv32Yhj53mIZey8JCDNPQ84dp6NOGaeg/Uxp6X9/LysceDJazsNDBeLyfppng2rW5ysf+SQJsDf3fwzT1DZTUTUikFMWM7t3fL5o6KZphZ6fEhqGb2HbkNzdrrbKvj/R6JQ4EZHAIKcWusVFzMgWclGKc0Rp7ekifT2K/X34nZXumaO/Ib2yUf5+U/WW01o8KT9TD/lA/STKeirPF06K21Q3pE9DgaqCZ1vo6fB0MJUQfc0fcHAgNkCRjqRhbvVIoLMvaKb/d1650c1fExcHwIEnR09u8bYpfP1Sv+PVD9Urra/O2qRv1YHiQrojoeOFEmO2+dpKkaZlscA1Lhv2Mf5dLGd26t3f7XNidXGps3D6XMrq9zyf/HrnrXNq2TedSV5fOJa9XcpuUXM8sJrPj+9+2TY6PlGsno7t/UpAZCkPKoJaM7h2P9zOZlMKSSgUZi0lhsSyDkci2D+VHIs1K947H+5hMetP8AGMxSQbTTKnFXIRfO4zfpHTreLyHyaQvzfczHu9J85NqEM0H+Y000/0ssVi30u2TSR/j8d40P8FIpHkYf/j730bLMtL8LqW7J5MexuN9aX6c0WjLh/LD4QbF/yRhVzd0W0O3YcOGDRs2DhLYGvpuIqMn7ipOJrWeblnba4C7y89ooHvDTyQ03zS1p3hP+BkNdW8QN+JqQpthGUqPBoBoKvpv453xSap577vix1IxpZulzBRSZmq/8g8k7Itc+Cj8eFznkmFof/re8D8pyOjUO8aWlVR6OmkpPXlXnB35GT187/gJxd+bYxnOJ03lT9/d/ZtmHBk93rIMtWb73vA/ybBv6GmsXg0cdphudBs3Tg+aOe004NVXJb7+euD22yV+9FHg0kslfu894MgjpbD19QFjx+rhGieeCPz97xJ/61vAXXdJPHMm8N//LfHbbwOTJ0vDUGen8JuaZNtxxwH/+IfEV10F3HuvxDNmAF//usQLFwLHHCNxS4vwM0NnpkwB/vlPia+4Anjwwb0/T5fNvQyPrHsEAPCrFb/CdQuvAwDMqZyDU589FQBQNVCFcbPGqQa0wx87HKtaZYGJi167CI+vfxwAcOu7t+LGJTcCAF4qfwlnvnAmAKCsrwzj/zgenqgH0VQUE2dPRFGHrPbxxVe+iKc2PgUA+Mmyn+B/3/lfAMCzW57FF16SL63ru9djwh8nIJgIIpgIYsIfJ2BD9wYAwOdf/Dye2/IcAOBHb/8Ityy7BQDw5MYncf6r5wMA1nasxcTZExFNReGJejD+j+NR1nfgPQn65jeB3/xG4gcfBK68UuIlS+QzB2TYzNixegDS0UcDixZJ/LWvAfffL/HvfgdcfbXEb70FHH+8xI2Nwu/slNycPBlYulS2XX458PvfS3zXXcA110j8+uvAySdLXFsr11Jfn1wbRxwBLF8u26ZPB2bN2qen5IBGMulCSck4tfDJli2no79fCkt9/Q1oaZHC0tU1C1VVUlg8nuUoLT0i3RzWh5KSsWpQy6ZNJ2FgQApLXd230doqhaWj4yFUV0thcbvfwfr1k0ES8XgnSkrGIhqVwrJx43EYGpLCUlNzFdra7t2j49mwYQpcriUAgOrqK9DR8SAAoLX1HtTWfhMAMDg4D5s2fQYAEIk0oKRkbLqZzcT69UfC45GVe6qqLkNn58Np/q9QX38tAKC/fw42bz4FgAyiKSkZh2RyEJaVRGnp4WrhmsrKi9Dd/dgevf9DEjt7Fn8w/OxLDT0eJ4uK9O9r1mitb/NmrTU2NmrdfHBQBmeQorGvW6f5q1dr7/nGjVorbGjQWmd/P1lTI3E0SpaUbM/P+IXXryfDabt1XZ3WOnt7tT4ZDsvrSOEN55eWkpF0f1dNjdY69wZVA1VK9+70d3KbS/S9QDzATT2bSIqnfE3bGsUp7ihmPCVaX2V/pfKVd/g62OQWfc4X83FLrzQLpczUdvyijiImDBGOy/vK6Y6Ir7XN26Z0e0+GIQadAAAgAElEQVTUw7Je6WJMGkkWthcqfmF7ofK+l/WWKb97i6dF6e6uiIvlfeUkZRhOUYdOhjVta5T3/UBCfb3uoejr01p3JCKfOfnhuZDJpdpanUs9PTqXQqEP5lIGJSWSq6Q0ffZLCwW7unQPSiAgOU/KNTCcX1ys+1GqquQa+iTB612jdN9AYLPSnSORRqWbJxKDqvHNMGL0+dYN469W3vNAYKPSncPhBqWbx+P9DIVq0vwo/f6SHfiSDH7/euX3DofrlO69u/D7S2kYUlhCoRqle8di3QyHpYcllQrR799AkrQsk16vTgafbx0NI5rmVzORSPfTxLpU30AqFWAgsCnNN+j1rhnGL1Y9BKFQ5XYe90MZsDV0GzZs2LBh4+CHraHbsGHDhg0bhzjsG3oa69aJvhePy8CMiRP1AhWnnw7Mmyfx97+vNfAnngAuu0ziNWuASZNEJxwYACZMALZtk20nnSQaNwBce63WwGfN0rrnihWie5IyyGPCBL1Yx/HHi8YOAP/zP6K9A6JfXnWVxEuXAsceK3F7u/AzAz0+9SnR+AHgq18FHnlk78/TV17/itLA71l1D767+LsAgPm185WGXjdUh4mzJ8IVcSFpJnHk40eiuLMYAHDJnEvw501/BgDcsfwO/OjtHwEA5lbNxVkvnAVAFnOZOHsifDEfoqkojnjsCKWBn//K+Xh+y/MARIO/eenNAIBXyl/BuS+dCwDY0rsFh80+DKFECMFEEIfNPkxp4Oe+dC5eKX8FAHDz0ptx67u3AgCe3/I8Lnj1AgCiwR/x2BGIpWLwxryYOHsiqgaGrZZygODb35Y+CkA+0699TeJ335XPHBDte8IE3U9xzDHAsmUSf+MbWgN/8EHR5AHpt8ho6K2tws8MhDnqKGDlStl2xRXAH/8o8e9+B1wn7RRYsEBr6Nu2CX9gQK6NSZOAwkLZ9qUvAU8+uU9PyQGNZNKNkpKJarGTzZtPx+CgFJaGhu8rDby7+0lUVn4JAODzFaK0dFJaQx9ASckERCJSWDZtOglDQwsAAHV11yoNvKvrj6iuvgIA4PWuwPr1R6U19B6UlExQC7ds3Hg83G4pLLW1/4OOjpl7dDzr138KHs+7AIDq6q+hs1MKS3v7DNTWfhsA4HItxsaNJwAAotEmlJRMUANt1q+fDK/3fQBAVdVX0NUldaW19Teor78BADA4OF9p6JFIPUpKJiCZHEpr6EfC75femoqKS9DT8+c9ev+HJHb2LP5g+NmXGnowKAM5SNEN583T3vF//UtrhZs2ad27tZUsLJTY79cLoJim8DN+4aVLtVa4YYNon6R4dTO6vderB3YYhiyQkfELv/22no1dUqI9yo2NWrd3u2UgCSk+4Tfe0D0Aixdr73tx8fYe4z1FYXuh8ovXDtZyY7eIpQOhAf6r6V8kxZM+r3qe0uoW1C1gMC5a3+q21crvXT1QrXTz3mCvGj4TS8X4Zs2biv9W7VvKu76qdRU7/aI1VvRXcGvfVpJkd6CbK1pWkBRP+vya+eo9z6+Zr7znK1pWsCcgwvPWvq2s7K8kKf0Aq1pXkSRDiRDfqpUVTCzL4ps1bzKWiu39SfuYUFqqdeumJvlsSZmTkFlAJZMLGe/4P/+p5yisW6f96tu2ad3d5ZKcIyUHh+fSokXa+15UJDMaSNHfN4hUysFByXlSroF583Q/ycKFuh+lsJBsa9snp+KggGVZHBiYp7znbve/GI9LYQkENindOxpto9dbSFI84UNDC9N8kwMD89TMdJdrKRMJKSx+/walW0cizfT5pLAkk14ODS1O8w0ODLyhvOcu19tqzrrfX7Kd3313MDS0RHnnfb5i5VcPhxvo90sTRiIxRJfrHZLiaR8YmKd6AIaGFqkeAp9vrfKbh8O1DAQ2pvkDdLuXpfnxNF/qwuDgAtVD4PWuZjTavkfv/2AFbA3dhg0bNmzYOPhha+g2bNiwYcPGIQ77hp7Gli3AGWfIcA2fDzj1VPHgAuKXzfi4b7sN+MMfJH7xRdHEAWD9euCss0QndLmAU07RGviFFwL/ErslbrlF647PPgt873sSFxUB55wjOmV/v/AzGvjUqVq3vOkm0e4B4KmngB+JBI333wfOFQkZ3d3Czyyw8fnPa93yxhuBZ57Z+/N03cLr8NeyvwIAHi5+WGnQ7zS+g4vnXAwAaPI04dRnT4U35kXSTOLM58/Epp5NAIBv/eNbeLVCvLcPrn0Qdyy/AwCwsH4hLpsrDQn1rnqc+uypCMQDiKViOOP5M1DeXw4A+Mab38DrVa8DAO5dfS/uXnU3ANHwr3hDdMPqwWqc9txpCCfDCCfDOO2501A9WA0AuOKNKzC/dj4A4O5Vd+Pe1aI7zq2ai6vmS0PC1r6tOOP5MxA34gjEAzj12VPR4GrY+5P2MeHmm4HH0tbbZ56RzxaQz/rzn5e4t1dyIbPYy7nnyswFAPjhD4Gnn5b4T38CfvxjiVesAKZNk7izU/gDA5Kbn/scUCztEPjud4HnxNKPWbOAn/5U4mXLgP/6L4lbW4Xvcsm1cdZZcq0AwHe+A7z00j49JQc0UikfNm8+FdGoFJbKyulwuaSwNDf/Ah0d0tDQ1/cS6uq+AwAIBNZjy5azYFkGkkkXNm8+RWng5eUXKh93Y+Mt6OqSwtLb+yzq66Ww+P1FKCv7HEgikejH5s2nIB7vBABs3ToVXq8Ulm3bbkJ39xN7dDxlZZ+HzyeFpaHhRvT0SGHp6noMjY3S2+LxvIfycpnvEIu1Y/PmU5BMDoK0sGXL2fD7SwAA9fXXo7f3BQBAZ+cjaGr6OQDx0VdUSF2JRpuxefMpSKU8sKwUtmw5E8Gg1JXa2m8rT/8nGVkPfpQpI/9hvPjiiw/efPPN++Tfys0F8vKA888HcnJkctbFF8v/dziA884Dxo8HsrKAz35WBnTk5MigjVNPldeNHCk33xEjZALWxRdLDEiBHDcOcDrl9Z/6lPCnTJF/LzcXGD1aCu5wfk6O7P+LXwQKCmT/p50mzUkZ/sknC7+gQAp5bq4U34svBrKzZf/nnw+MGSP7P+MMed97A6fDiTOPPBOTRk9CjjMHx447FidMOAEjskZgfP54nD3pbORm58LhcOC/jv0v5GTlwKKFC4+5EKNGjIITTpw96WwcMeoIZDuzcfy44/GZCZ/BiKwRmJg/EWdNOgu52bnIcmThwmMuRE5WDkzLxEXHXoT8nHw4HA6cM/kcHD7qcGQ7s/GZCZ/B8eOPx4isETh85OE448gzkJudixxnDi445gK1/4uOvQh52XlwwIEvHPUFTBw5EVnOLJw44UQcN+444Y86HKcfcTpys3IxImsEzj/6fM0/7iLkZufuk1zbV8jk0pQpH8ylMWOAL3xB4kwuDc+FggLhn366zqWjj5YGztxcGSZzzjkfnksXXCC56nQCZ54pjW7Z2dKUeeKJkr/jx8vNP3P9XHSR7IOUL7ijRgn/rLNkINMnAQ5HDkgT48ZdDKczF4ADBQXnISdnPByOLIwc+Vnk5R0Nh2MEcnMnY9SoU+F05iIrayTGjp0Kh2MESCvNl8JSUDAN2dnj4HBkYdSoU5Cb+6k0/yiMGvVZOBy5yMoajYKCc+F0DufnpPf/RWRnFwDIwqhRpyE396g9OqaxY89HdvYYOBxOjBp1OnJzJ8PhyEFe3tEYOfIkOJ25yM4eizFjzlHHPG7cRXA6swEQ48ZdiKys0QCcGD36TIwYMQkORzby8o7FyJEnwOEYgezs8Rgz5mw4HLkAnBg79qL0+7cwduyFyMoaBcCB0aPPxogRR+zDT+zAxMyZM/sffPDBFz9sm62h27Bhw4YNGwcJbA3dhg0bNmzYOMRh39DTqK4GvvxlIJUCgkHRzTPe3Wuv1Rr2Aw8Af07bHd98E/jZzyQuL5fZ1qYJeL3C7+6Wbd/8ptawf/c74HmxUWPuXOCXv5R40ybx9ZKiN15yidbAv/ENoESkJtx1F/DyyxK/+ipw550Sl5ZqH/LAgPCHhuT3K68ENm7cN+fp1ndvxbwa8c7+ZfNfMGONGKHfb3sf1yyQYd4d/g5M/9t0BOIBpMwUvvz6l5WP+ydLf4J/1Mn86Cc3PImHih4CALzX/B5uWCTe0xZvCy7926UIJUJIGAlcNvcy1A7J/Oqb3r4JixsWAwAeK31MzZVf2rhUzYXf5t6GL839EqKpKKKpKL4090todItueeOSG7G0UYaRP7LuETxWKiL04obFuOntmwAAtUO1uGzuZUgYCYQSIVz6t0vR6m3dNydwH+I3vwH+Ku0MmDMHuEPaEbBhg55vMDgouTAwIL9/7Wtaw77zTr1GwUsvAXdLOwLWrZOcAyQHL7lEZjOQkqObN8u2X/xC5rYDoqVn5iusWSNrFgDSBzJ9uvSlmKZcI+XSDoGf/UyuIUCuqQcekHjlSt2bcjBiaGghGhv/DwAQDlejqurLsKwkDCOIysrpiMXaAIh3PKNht7c/oHzUg4NvoqlJCksoVI6qqstBmkilfKisnI54XJpramu/qTTstrbfobdXCsvAwFw0N0thCQY3o7r6CpBEMulCRcUlSCSksNTUfAN+/zoAQGvrXejre3mPjrO6+koEAjIfoqXlDvT3zwEA9PX9Fa2tssiA31+EmhpZJCAe70FFxSVIpTwgLVRV/TeCQXnC2tR0KwYH3wAgPQDt7VJXvN73UVcndSUW60Bl5XSkUn5YloGqqq8gFKoEADQ2/kTNpe/ufkp56j2e5aivvx4AEI22oLLyUhhGaI+O82BC9n/6DRwomDhRdO6sLNH9pk0TnREQLfKotLR0yimiLwKiGQaDEh92mOjnTieQny/xmDGy7bzzRGcERP8+/HCJjztOvkAAMtRm2jTRG0eOlHj0aNk2darWGU8/XbROQIZ/ONNfyY48Ul4HiD45bZr8N8M/Yh9JS2ceeSaOHSsTbE6YcAIm5k8EAEwePRnnHiVdeQW5BZg2ZRrysvOQ7czGtCnTMHGkvO7sSWfjmLGyisyJE09UK5wdNeYofH6ydHKNyxuHqVOmIi87D06HE9M+NQ0T8icAAD43+XM4ukBOwEkTT4KZXuFpSsEUxR+fNx5Tp0zFiCzRGadOmYpxeeMAAOdMPgefKpCpK5897LPIdsolcHTB0fjc5M8BACbkT8C0T01DTlYOHA4Hpk6ZirF5Y/fNCdyHOO003Qtx/PF65bJMLgG7zoUzztDDiD79acl7YPtcGj1a+CNHSm5Onarz98wzJYcB4IQT9HUxaZLkPCDXwNSpck04nRIfdphsO+ssvf8TT9Q5PnmybvA8GJGXdzRGj5ZcysmZiIKCaXA4suF05qU1bzlRY8Z8ASNGSGEZOfIU9f/z8o6FaQbT/MNQUDAVgBNOZz7GjJmKrKyCNP88jBghhWXUqNOQk3N4mn8cyFSafzjGjJkKh8OBrKyRKCiYltasgYKCqRgx4sg0/3Tk5h69R8dZUDBNadajRp2B/Pzj0/v/DJzOkQCAESMmoaBAkiE7ewwKCqbB6RwJh8OJgoKpyMmRZBg9+izk5h6r+NnZcr3n5k7GmDHnpvkF6fefD4cjawf+2cjNPSZ9Lk+EaUbS/KMwZswX0vxxKCiYCqczb4+O82CCraHbsGHDhg0bBwlsDd2GDRs2bNg4xGHf0NNoahJPuGEAkYjMpe7tlW233ioaNSDe77/9TeJly/R60vX14gO2LHkMf911Wre85RbRyAHg8cf1XPglS4CHREJGdbX4ggHRG6+9VrR0APi//wO2bpX40Uf12ugLFny0uex7gwcKH8A7je8AAF6veh1PbBDv6vru9fj5v8Q72hfqw3ULr0M4GYZpmfje4u9hm1vmT9+7+l681yyD5V+reA3PbBLvanFnMX75nuh+3YFuXL/oekRTUaTMFL67+Lto8cqC3vesugcrW0V3fGnrS2qu+5r2NbhzhTQUtPvaccOiG5AwEkgYCdyw6Aa0+6Qh4s4Vd2JN+xoAwHNbnsNLW8UIvbJ1Je5ZdQ8A0fC/u/i7SJkpRFNRXL/oenQHuj+W8/lRMHs2MF8s9Vi8WM9HqKzU8wm8Xsklj0d+v+kmoEKW48bDD+s1Bt56S69NXlYmOQdIDl57LeD3i4b+wx/qNQ5mztTzGd54Q3IbkH6NjCe9v1+uhWBQro0bb5RrBZA59Jm58n/7m1xbgPSL3HabxD09wo9EPvLp+ljR0/MX9PfLGgEez3K0tf0WgMwvr6//HizLgGlGUFd3HeJxGQrQ3HwbAgEpLN3dT2FgQAqL271MaciRSD0aGm4EacEwgqiruw6JhBSWpqafKh92V9fjSoN2uf6Jjg4pLOFwNRoafghAfPB1ddcimZTC0tj4fwiFpLB0dj6qNOjdxbZtP1IadkfHH+BySW/L4OB8dHXNBgAEg1vQ2PgTAEAyOYi6umthGAGQREPDD9Ta7u3tD8DtlroyMPC68sQHAhuUJz2R6ENd3XUwjDBIEw0N31dz7dva7lVz5fv7X1P9CH5/seoniMe7UV9/PUwzCstKob7+u4hGW/bomA902Bp6GllZWifM6NgZfXrkSO3Bzc3VWmNOjnjXAdmeny/cjI6elfXh/Iw3fcSI7fkjRXaC0ylxhp+fr/l5eR/O31/Iy85T2vSIrBHIy5Y3kO3MxsgcOQCnw4mROSPhdMgJHJkzUmnVedl5yMnKUfyMtzvHmYP8nHzFz8/Oh9PhhMPhwMjskchyZGm+U/OznFkf4Gc5s5CfLZ51AMjPzlf7z8/JV/zcrFz1XrbjO7IwMnskHA4HHHAgPztf7edAwvBc2jEXd8ylTC7vTi7tjO9w7D4/X04lsrK0fv5h+8/J0ccy/Loazh/+/g9UOJ25aZ804HTmwOmUA3A4spCVJblEOtJxVvp1+XA4sv8NP1vFgDOtHzt3g5+n+FlZmX/Lmd7/h/Hz4HCM2MNjHrnD/nPS8Yi053zH9y/nIvN35Pb7zx/Gz/1Q/vBz+e/4ZDLNGX4unenYCWD7z+JQga2h27Bhw4YNGwcJbA3dhg0bNmzYOMRh39DT6OoSX65lyZrot9+uNezf/150SUC845m1yYuKtCe9rU084iQQjYq/3OuVbQ88oHXHV1/Vc91Xr5Z57gDQ3Cy+YgAIh4UfCMjv990HNKRHib/4IrB8ucQrV2of8v7CM5uewdqOtQBkfvucyjkAgKqBKuUpd0fduH357YgbcVi08OuVv0anX+ZHP7nhSazrFO/r4obF+Hv13wHI/PSHix8GAAyGB3H78tuRMBIwLRN3rrgTPUHRHR9f/7haG31B3QI1l31z72bMKhERuC/UhzuW34GUmULKTOGO5XegP9QPAJhVMgube8VI/WbNm1hQJ+tJr+9er9Z57w50484Vd8K0TCSMBG5ffjuGIkMfw9n8aHj5ZVn7HABWrdLzDZqagN+KhItQSHIpY6/83e9kjXIAeOEFPV9h+XI9V72+XnIOEO38l7+UnASAe+4BWtKy47PPiuccEC38tdckrqmR9dUB0e5/+Uu5Jki5RjLzHZ5+Wq4hQK6puXMlrqjQ67S7XHItxuMf6VR97BgcnAeXaxEAIBAoVRpwPN6FlpY7QVowzTiam29XGnZHx+8RCklDw8DAXLU2ud9fhJ4eGbIfi7WjtfUukIRpRtHc/EukUlJY2tsfUGur9/e/CrdbGhJ8vtXo7ZXCEo02o7VVekMMI5zm+wEAbW33IRKRwtLX9yI8nuV7dMxtbb9FNNoEAOjtfV6tbe7xvKs87ZFIHdrbpdEolfKhufmXylLW2nq3mkvf0/OM8tS73e8oT3s4XKU85cmkO82PgbTQ0vJrxGIdAKQHIeOpd7mWYGBA6kooVI7OzofT/EE0N98Oy0qANNHScqfqZzhUYGvoaYRCclM1TfGGNzVJEQKkgGVuzt3d+v+7XEBHx/Z8ywKSSYljMc33+zU/o3IMDcniF4AU3KYm2ZZISByLibe3uVnzu7q01jg4qBdw2V/o8Hdg0mjxvg6EB9SNzhf3qca1aCqKJk8TkmYS2c5sNHubEUqGFP+4cccBAPpD/Qgn5U7hjXnR6mtV/GZvMwzLgMPhQJO3CaGE8Nt97Th54skA5MZtWAYAwBP1oM0nAzsiyYjiA0Czt1ntp83XhrMnnQ0A6A31Kj3dE/WoxrlwMowmbxNMmkhZKTR7mxFJRoBR+/x0fiR0d+s+i+G5FAhI/gByI2xqkv8WFEguZb4odnXpWQfDcykQkNcN5ycS4mVvbtZfDjo6ZH4DIA2gmUFIPp++6cdiwkkmJW935E+ZInF/v77GfD69sFE0KvtPpfZ/v8ieIJHoUd7rVMqNeFxyyTRDiMWaQZogU4jFmmBZUkBisVYYhk/xTVP+fzLpQjzekeYHEY02A7BgWUnEYs0wzShyciYgFmuBYfjT/G4AHMbvVHzZP0Em0vuPp/ffrPjxeJfSrXcX0WgTDCOQ5nemZ8LLjVPeD2AYfsRikkyWFUMs1gzLSsLpzE/vP6j4GU98MjmAZHIwfS59iMVatuOTKZA56XMRSvPbkZd3bJrfr/7dVMqrvjSYZjTNNwA4EIs1Kf6hAltDt2HDhg0bNg4S2Bq6DRs2bNiwcYjDvqGnMTgovl5SHu89+qh+NPnCC/oR5tKlwNq1Em/dKv5bQB4ZZny4iYT4w0PppznPPqsfIS5ZIrOyAZmJnfER9/QATz4pcTwu/Iz39s9/1o/2Fy7Us7g3bBAv+v7EvJp5KOuTpyJFHUXKk97saVae8GAiiEfXPYqUmQJJPFb6GAbC4p19vep1VPSLbrimfQ3+1SQNBdvc2/DiVlkR0Bfz4dF1j8KwDFi0MLt0NlwR0R3nVM5Ra5uval2F5S2i+9UN1eGV8rQPOOrBrJJZMC0TpmViVskseKJixH6l/BXUDdUBkPnxq1pXAZA11DP9AK6IC7NLZ+Ngenq1aZOeT9DVpT3dsZjkUkYmevpp/Wh+wQI947+0FFgkEjDa22V9dUBy8JFHtIb9xBN6bfX584EtWyQuLtae9JYWvU56KCT8pLiI8Pjjcq0Acu1k5isUFsq1Bci19oIsjY1AQK7FzIjkAxUez7tKQw6HqzAwIA0ByeQguroklywrhc7OR9Vj6t7eF5QG7XYvhc+3FgAQCm1VnvJEoh9dXVJYLCuJzs5H1Czy3t5n1eNkl+ufSkMOBjdjcFAKSzzeo/R804yjs/MRpWH39PxZadAfFUNDbylPvN9fojzpsVgbenr+AkA0/M7OR2BZCQBAd/efkEjIsI/BwXlqrrvfX6T6CaLRZjWj3jCCab7Ula6ux5Qnf2DgddWP4POtUf0Ekcg29PVJXUmlfOjsfBSWZYC00NU1W/UzHCqwb+hp9PZKQUulpIgtXKgXN3nnHX1DLyrSi1PU1AArVkjc1SV805QitmiRLGgBSKHLaIpr18rgDkAa7d6XGoCODtknKUVs4ULREgEZGtIm8jDWrNGLW1RUSGPd/sTK1pXqhrq5dzOKOqSrqdnbjLcb5SJ0RVxY2LAQ4WQYhmVgUcMi1dS2vHW5WmhlY89GrOuSItTobsTSJqnog5FBLGpYhGgqiqSZxML6hegNyYX/bvO76oa8vns9Srpk1ZoGdwP+1SxfDvrD/VjUsAhxI464EceihkXqC8Wy5mVqyE1pdyk29EiDXd1QHd5rkYE3vaFeLKxfiISZ+FjO4ceBiooP5hIgvRfDc2nRIt2Utnq1zqXyct3g1tYmOQeIrr1woeQkKfzMF4JVq3SzaFmZ/qLb0qJv7m63cEIhGdq0aJHW6les0M2imzfLlwJArrV35HsihoZk/wf6YJlAoATBoHw7ikRq4fVKLiUSvXC5FoFMwTQjcLkWIpmUwuLxvKNu6IFAMUIhuSGGwzXwelek+V1wuxelB9OE4HItQiolhcXt/qfSl/3+QoRCW9L8Kvh8q9L8zvT+CdMMwOVaqJrqXK7FiMfb9snx+3yr1Q01HN4Kv18a3GKxVrjdSwAAhuGBy7UIhhEEacHlWqQWmvH5ViESkQWcgsEt8PuL0vxmdXNPpVxwuRbBNEMgDbhci5BISF3xelcgEqlJ8zchEFiX5jfC41ma5g/B5VoEy4rCspJwuRaqLxSHCmwN3YYNGzZs2DhIYGvoNmzYsGHDxiEO+4aeht+vfbCWJX7xjO64eLHWDUtK9GPKpibtCfd6gb+L9RGmKfyM7rhggdYNi4qAKnmyhIYGeWwJyKPJzIz3VAp45RWtO771lmj8gDwWrZUn1qir2/+P3Fe0rFBri1f0VyhPeW+wF4vqRYSNpWJ4teJVWJT1POdWzYU/LvaYd5vfRbNHbCxb+7aitEtmWXcFurCkQR7NRZIRvFbxmuL/rfJvCCbEhrKsaZmyp23u3YyNPfKYs8PfofT8UCKEOZVz0lYdYk7lHGVbe6fxHXT4Oz6ms3PgYcdcmj9fS0mrV0sOAfLou1CekmJwUHIOkH6QV16Rx+WA5GhmLvzKldrTXlWlPeV9ffqRfzwu14Ipq9zi73/X9rTly7WUVV4u1xYg11rmkX80KvzM0rAHKgKB9UoDjkZb1FxxwwgoPZ200N//qrKnySNvKSx+fwlCofI0v0l5wlMpr/JUk2aaL4VlaGgBEon+NL8I4bAUlkikAV7vqo/9mPcUlpVAf/8rsNJ20sHBN5BKSTJ5vSsQjUpdCYUq4feL/pKRLADANGPo738VTC+ZPDDwuvLUezzvpe190oOQmZEfj3fD5VqS5kfQ3/8amK4rAwN/U/a2QwX2DT2N+npZaCWREL3x/vt1I9ojj+hGtNde081DK1bo5qGaGllswjDk5jxjhviEAVk0I7M4yyuvaH3x3Xd181BlpeyTlIJ7//16cZgHH9S6+4svan1x6dL9P1jmL1v+ohrRFjcsxmuVMk1kQ88GPFIiK8V0+Dtwf+H98Ma8SJpJ3F94v9K9n970NN5vE7H3H3X/wNwqKXYlXSX4Y+kfAbS5HGUAACAASURBVIhXfEbhDAQTQcRSMdy/9n40uGQAxhMbnlCLq8yvna8G0xR1FOGx9Y8BED1/RuEMhJNhhJNhzCicgSaP3DkeW/8YijuLP96TdABhYEByKfOF8oEH9BfSv/5VL47yzjuSW4A0us2UWR7o7RX+0JDk5owZWjd/7jk92GbJErnxApLrmYViurqE4/HItTFjhv5C+swzerDNwoXAnDkSl5ZKIxwgev/99+s5DAcqBgdfh8sl34L8/tXo6ZGuxEikHu3t98OyEjAMP9rb71ce9a6uRxEMSmEZGJijblxe7wr09j6T5teio2MGLMtAKuVGe/sMJBKiO3d2/kHp9v39r6gbl9f7Hvr6nttPR777iMe70d5+P1IpF0gT7e0zEA5LP05v77PweqWuuN2LMDAgdSUQ2KAGw8TjnWm+D5aVRHv7DEQikkw9PU+rvoGhoQXqS1QgUIKuLqkrsVgb2ttnwDCCMM0Y2tvvRzTasP9OwH6AraHbsGHDhg0bBwlsDd2GDRs2bNg4xGHf0NOIxbQFDZB56xnva3Gx1v1qa7UFrbdX+3AjEa2Hk8LP6I5r12rbUHW1tqB1d2sfbjisbUekPArN6I6FhdoTX1GhbUOdnXpt6/2FLb1b0BsULaDV24qaQbGKeGNe9SjbsAzlLwdEd4+lZA7upp5Naq56s6dZPYp3R93KgpYyU3i3+V3FX96yHHFDdMMN3RuUBa3R3agexQ9FhrC+Wx5fJoyEWnMdEL95whALWmlXqfK0fxJgWZJLGQ16zRo9erW8XFvIOjr0o/RAQOvppin8zIO899/Xc93LynRvSVub5DYguZ7R0w1Dr10AyCP2jAVtyxYtK7W06EfxHo+e1ZBKbc8/UBGJ1CsLWiLRrzzZphlTFjQA8Hj+BcuSwuL3FysLWThcq9bmTiR6EQxuSfMjSg8nmeZLYfH51iKV8qX51YjFpLDE491qnfMDCaQJt3uZmu/g9a5SnvhgsEz1E8RirWpGfSrlVXq6ZRnweHQyeL0rVT9CMLgJiYTMHo5GmxGJSF1JJt3w+0vS/JTqbQBk3fpMP8Ihg0zj0MH48/nPf577CoWF5OjRZCxGDg2R+flkZaVs+8xnyDlzJL7mGvIXv5D40UfJCy+UeMUKsqCATKXI3l4yL4+sq5NtRx9Nzpsn8VVXkb/+tcQzZ5KXXirx0qXk+PGkZZGdncJvapJtkyaRCxdKfOWV5G9/K/F995GXX77PTsFu4aLXLuLDxQ+TJG9/73Z+661vkSTnVs7l8U8dT5KsHqhm3h/yOBgeZMJIcMwjY7i6bTVJctrL0/hY6WMkyZ8t+xmvX3g9SfLlrS/z5GdOJkmW9ZYx/w/59EQ9jCQjHP3IaBZ3FJMkz/nrOXx649MkyR+//WP+YMkPSJLPbX6OZzx3BklyQ/cGjnx4JAPxAAPxAEc+PJIbuzeSJE9/7nQ+v+X5j/UcHUhob5dcammR3w8/nFy8WOLLL5ccIiWnrrxS4gULJOdIsrFR+J2dkpvjxpHLlsm26dPJhx6S+M47yauvlviNN8hjjpG4tlb4fX1ybRQUkCtXyrYLLiBnzZL4ttvI73xH4tdeI084QeKKCrkWXa59eVb2Perrv8fGxltIkl1dT7Cs7DySpNdbyOLi0TSMGBMJF4uK8hkMVpAkN248gX19r5Eka2u/w6am20iSnZ2zuHXrBSRJj2cFi4sLaJopxuN9LCrKYzgshWX9+mM4MPAGSbKm5mo2N99Jkmxvf4gVFdP3z4HvAcLhBhYV5TEW66ZlGSwuHku3+z2SZHn5RezokLrS3Hw7a2qkrvT3z+WGDVJXQqFqFhXlMZEYoGkmWFw8hl7v+yTJrVunsbNT6kpj489YVyd1pa/vZW7cKHUlGNzKoqJ8JpMeGkaExcWj6fMV778TsI8AoIw7uSfaGvowGAaQnb3r2LIAh0N+APkLJrNAxu7wTRNwOvcNn5T3k+HvD5iWiSyn7JAkLFrqd8MykO3M3u2YJAjC6XDuFT/TBb+3/E8K9iaXdjcXM6/fF3xS3s+uOAcqMp3TjnQuWpYBZyZPdyMWvgOOdGEgTTgcWXvANwE4P5R/IGFX7z/zfuWeZO3R8W//bxEA9+izOJiwKw3dvqHbsGHDhg0bBwnsprjdgGFsr0dv3ap1w/p67Unv7taecL9f6+mplNYgAdEXM/zaWr2UamenXmfd69Uz3pNJ7U/P8DOoqdGe9vZ2PVLW49FjPPcXWrwt8MVEtxsMD6IrICJsNBVFvasegHxD3tqnNbzKgUq9lKmnGYG4NAQMhAfUSNhIMqL0cIvWdvyK/gqYljQUNHmalCe9L9Sn9PxQIqRGulq0UN5frvjl/eXqr/lt7m1qKdZPCobnUnW1WDMByZ2Mp9zt1jbNeFyPZN2RX1WlPe2trbq3ZPjyrbGY9reT2/MrK3VvyvBlhQcHtc0zEpFrLsPfeuDJwR9APN6jPOGGEVB6umUZaiQqIB7pzB9RkUi90oDj8e5hS4b6lZ5uWSmEQrqwBINlih8O18I0Y2l+p5pLPnzJ0AMNGa8+ICNqM/0EsVir6gdIJofUSFjTjCo9nOR2/FCoUvUTRKPNypOeSAwgHu9O8yNqzXfS2q63IBSqUJ72QwY7exZ/MPzsSw195UoyN1c09MFB0ukky8pk2zHHkC++KPHVV5O3iFTGhx4izxOpjMuWidaXSpE9PaTDQVZXy7ZJk7QGf+WVWoO/7z6twS9eLBq+ZYnu6XCQ27bJtgkTtAZ/2WVag7/7bq3B7y9Me3kaHyx8kKRo4N948xskRQP/1BOfIkmW95XTOdPJgdAAE0aCub/P5fLm5SRFA3+k+BGSooF/+x/fJika+Kef/jRJcmP3RjpnOumOuBlJRpjzUI7S4M947gylwf9gyQ+UBv/0xqeVBr+ucx2zH8pWGnr2Q9lc17mOJHnyMycrDf6TgNZWyaVMP8bYseRbb0k8fTp5zz0S33kn+eUvS/zGG+TEiRI3NAi/o0Nyc9QocskS2XbBBeSMGRLfdhv51a9K/Npr5OTJEldVCb+3V66N/HytwZ93ntbgb7lFa/Avvqg1+LIyuRaHhvbpadnnqK39Dhsa/pekaOBbtpxNkvR4VnLt2ty0hj7EwkIng0EpLOvXH8PeXiksNTVXKw2+vf0hpcG73ctYVJSf1tB7WVjoYCgkhaW0dBL7+6WwVFVdyaYmKSxtbfexvPzC/XTku49wuI6FhQ7GYl20LINFRSPpcr1Nkty69Ytsa3uAJNnY+HNWV3+dJNnX9wpLS6eQJIPBChYWOhmP99M0E1y7Nldp8Fu2fF5p8Nu2/Z/S4Ht6nlcafCCwiYWFTiaTbhpGhGvX5tDrXb3fjn9fAbvQ0P/jN+WP8rMvb+jk9kVjeOzxkKYpcSgkN32STCZJv1+/bnjjznC+2635waDmJxK7z7csiQMBMh6XOB6X3/cn/DE/k0aSJBlLxRhKhEiSpmXSE/Wo1w2Fhz409sV8ih9NRhlOhEmShmnsNj9lpkiSkWRkO7436v23fG/US8M09vi4D2YMzyWXa+e5FAxKbFmSczvjZ+D3Sw6TktMZvmnunD889vvlGiLJaFSurQzf4/lwzoEKwwjTMKIkSdNMMZn0qW2JxNCHxsmkh5YlhSGVCtEwYml+kqmUfxjHtRO+exg/OIyf2I5/IGH7c6GPK5Xy0zQlGQwjxlRKksGyTCaTnp3wh58L3zB+lIYRTvON3eIfTNjVDd3W0G3YsGHDho2DBLaGvpvIeGoBrecBMjYz4yn3+bQPNx7Xeji5c35fn/aUe73ahxuLaT3csrQnd1d8j0fr+dGo1kD3F1wRl/KEh5NhpacblqH85QDQHdAH0BPsUbrfUGRIecJDiZCa8Z4yU8pfviN/eDycH0wElR6fNJMYDA/+W/4nEcNzqbdXe9KH51IkovVw05Sc+zB+T4/uDXG7dW9IOKxnLRiGHjULaK/7jnyXS/eG7IrffRB8fKmUX61TblkJtUQqAOWvllgfTCLRrzTgVMoHw5DCYppxpYeT3AW/T2nAqZRXeboPFsTjui4kk27lCTeMsNLTLctQvQnC6d4J36XWWTeMkNLTLSul1kzfkX9IYmd/uh8MP/vykfuKFeSIEVpDdzi0hn700VpDv+oqraHPnKk19KVLxW+b0dABraEfeaTW0K+4Qmvo996rNfRFi0SfzGjogOiXpPjTMxr6l74keidJ3nWX6KD7E1NfmsoHCh8gSf502U/59Xmidb209SVO+ZNoXVv7ttLxoENp6CN+P4LvNYvW9bkXPqd87De9fZPysT+7+VnlY9+Vhn76c6crDf3GJTcqDf2pDU8pDb24o5hZM7OUhp41M0tp6J80tLRILg3X0OfPl/iSS6QPgyR/9SvpzyDJv/9d+jZIsr5e+BkNfeRI7WM//3ztY7/1Vu1jf/VV7WOvrBR+RkPPy9Ma+rnnag39Jz/RGvpf/6o19C1b5Fo80B+719ZeozT0jo5Hh2noK7h27YhhGrqDgcAWkuT69Uezt/evJMnq6qu4bdtPSJLt7TNZVnYuSdLlWsqiorxhGjoYClWRJEtKjlQ+9qqqK5SP/WCAaOj5wzT0acM09J8N09BfHqahl7Ow0LETDf2cYRr6j4dp6M8N09A3Kg39YAZsDf3fwzT1DZTUQ2FIKYoZ3bu/X+t7waAM3CBJw9BNbDvym5u1VtnXR3rTUm8gQHZ3S5xKyRCPDGprddzUpLXGnh7Sl5bn/H75fX+iy9/FYFzEUk/Uw/5QP0kynoqzxdOiXlc3pE9Ag6uBZlrr6/B1KN3dHXFzIDRAUvT4Vm8rSdKyrJ3y233tSjd3RVwcDA+SFD29zdum+PVD9YpfP1RPKyMcfwIxPJcaG7fPpUwPh88nN11Stme+AOzI37ZNcpUku7p0D4fXK7lNSq43N2vO8Gth2za5Vki5djK6u8cj1xYp11pLy4fzD1QkEgNKEzaMMGOxDpKiAYfDurBkhsKQZDTaonTveLxfab2pVJCxWGeabzAS2fah/EikmaYZT/P7mEzqHpKDAZHINlqWJEMs1sVUSpIhmfQwHpdkMs04o1GdDMOPPxxuGMbvVLp7MulmIiF1xTBijEZ1XRnOP1ixqxu6raHbsGHDhg0bBwlsDX03kdETdxUnk1pPtyytAe4JP6OH7w0/kdB809Se4v2FuBFXnm7DMpA0k2pbNBX9t/HO+CTVvPdd8WOpmNLNUmYKKTO118cynL83+0+aSeWvP5CxN7m0O7kYj2s93jC0P/3j5B+osKyU8lST1nYzwjNe813FlpVUevqOfBs2dhf2DT2N1auBww7TjW7jxulBM6edptd6vv564PbbJX70UeDSSyV+7z3gyCOlMPX1AWPH6uEaJ54I/F2W7ca3vgXcdZfEM2cC//3fEr/9NjB5sjQMdXYKv0lmU+C444B//EPiq64C7r1X4hkzgK9//WM5HTvFZXMvwyPrZN3zX634Fa5beB0AYE7lHJz67KkAgKqBKoybNQ6D4UEkzSQOf+xwrGqVBSYueu0iPL7+cQDAre/eihuX3AgAeKn8JZz5wpkAgLK+Moz/43h4oh5EU1FMnD0RRR2y2scXX/kintooa03/ZNlP8L/v/O9eH8uP3v4Rbll2CwDgyY1P4vxXzwcArO1Yi4mzJyKaisIT9WD8H8ejrE+eBJ3x/Bl4ufxlAMD3Fn8Pt717217vf3+grU1yKTMA6eijgUWy7Da+9jVZaxwAfvc74OqrJX7rLeD44yVubBR+Z6fk5uTJwNKlsu3yy4Hf/17iu+4CrrlG4tdfB04+WeLaWrmW+vrk2jjiCGC5LHuN6dOBWbMk/uUvgRtukPiVV4DTT5e4okL4rgN8PZ1t236A5uafAwC6u59ARcWFAACfbw1KSw9TjW4lJePUoJktW05Hf78Ulvr6G9DSIoWlq2sWqqou/Q8chY2DHjt7Fn8w/OxLDT0eJ4uK9O9r1mitb/NmrTU2NmrdfHBQBmeQovutG9Z3tXq19p5v3Ki1woYGrZv395M1NRJHo2RJyfb8jOy7fj0ZFtmYdXVa6+zt3f/6YtVAldK9O/2d3OYSfS8QD3BTzyaS4glf07ZGcYo7ihlPidZX2V+pfOEdvg42uUWs9cV83NIrzUIpM7Udv6ijiAlDDM/lfeV0R6Sppc3btp1uv6do8bQo3d0VcbG8r5wkmTASLOrQybCmbY3yvm/p3UJfTJoYmtxN7PB17PX+9wcsa/tcKi3VuVRbq3Opp0fnUigkOTecn0FJieQqKU2fGd27q0v3oAQCkvOkXAPD+cXFuh+lqkquIVKuqUwPid8v1xwp1+AanQoHLCKRZkaj7STFX51ZgMU04/T5dC55vWuU7hsIbFZ+8UikUenmicSganyzYWNHwNbQbdiwYcOGjYMftoZuw4YNGzZsHOKwb+hprFsn+l48LgMzJk7UC1Scfjowb57E3/++1sCfeAK47DKJ16wBJk0SnXBgAJgwAdgma4XgpJOAhQslvvZarYHPmgVceaXEK1YAU6aITtndLfzMwi3HHy8aOwD8z/+I9g6IfnnVVR/P+dgZvvL6V5QGfs+qe/Ddxd8FAMyvna809LqhOkycPRGuiAtJM4kjHz8SxZ3FAIBL5lyCP2/6MwDgjuV34Edv/wgAMLdqLs564SwAspjLxNkT4Yv5EE1FccRjR2BD9wYAwPmvnI/ntzwPQDT4m5fevEfv/9yXzsUr5a8AAG5eejNuffdWAMDzW57HBa9eAABY370eRzx2BGKpGLwxLybOnoiqAVk556wXzsLrVa8D+H/23jw+qvLs/3/PTEIWVkGsuKBUrVWLuBaq1rr12/rUx/bnUteqrS1aW6uWqk+1KLQuKLYKVmxVRNEqCMEFkE32sIYsLEnISkL2PZPJ7Mv1++M6c05UQKyAgvf79eLFFSafzHDmPnOfnM+1wG3v3cYfF/0RgEnrJ3Hx6xd/rtdyIKiu1rWUHOIzdCjMm6fxlVc6Hvi4cXDVVRq/957joVdUqD7ZEOaoo2DxYn3s8svhqac0fughuF7TKZg1y/HQt29XfWOjnhtHHgnLl+tjl14Kzz6r8f3367kFeq4lPfStW/VcTDZgOhiorf0n+fkXAtDZmc2aNUdYHnor2dmD6O7WD5aNG79DU5N+sBQX/4KKCv1gqal5loKCSwHo6FjOmjVH2glzBsMe2d29+IPhz7700Lu6RN55R+NEQhu5JGvH5893vMINGxzfu6JCZPlyjTs7RWbP1jgeV32yz/XcuY5XuG6dNusQ0VrdpG/f3u407IjFdEBGsl74/fed3tjZ2U69e0nJx337A8HyHcvtevFtTdtkfY2apY2+RplfOl9EtCb9rS1v2bXfswpn2bXrSyuXyo6OHSIisqVxi+2b13XV2c1ngtGgvL31bVs/c9tMu3Z9ScUSqe5UrzG/IV9y63M/1+tfVL5Iar1avJ9bnysFDQUiovkASyqWiIiIL+yTmdt0gkkikZC3t74twagavx+Wfij1XVoju7F2o2xp1O5BOzp2fMz3/6oQjepaStaOv/ee00dh9WrHt96+Xf11Ee3X/r72+5BIRPXJfJKsLKcPwsqVTr14YaGubRFd63PnahwO67mQzCeZPdvJR1m+XKRSUxhk61Y9t0T0XJuvS0mCQdUfTG0EgsEqaW//SES0prypST9YEomENDa+Zdeet7bOl1BIP1i83g3i8+kHSyBQKe3tyy19pzQ3zz7A/wPDVxmMh24wGAwGw8GP8dANBoPBYDjEMRu6RU4ODB+uzTU6OuDUU7UGF7Re9r33NL77bnjsMY1fekk9cYC1a2HECPUJW1rglFMcD/yCC2D+fI3vvNPxHV94AW6+WeOVK+Gss9SnbGhQfXKoxciRjm95++3q3QM89xz88pf753jsjutnX8+/N/0bgMdXPW570B+UfMAPXvsBAKVtpZz6wqm0B9uJxCOc/uLpbKjdAMDV71zNq/laeztuxTjuW3gfALOLZnPZdE1IKGop4tQXTsUb8hKMBhn+4nDyGvIAuPLtK20P++GlD/PAkgc+1+u//D+XM2PbDAAeWPIADy/VhIbpm6fz0xmakJBbn8vwF4cTioXwhryc+sKpFLcUA1qHn1Wkhdz3LryX8Ss0oWFq3lSueeeaz/VaDgR1dbqWkoODzj1Xey4A3HYbTJqk8d//Dr/+tcaLFsGoURpXV6u+sVHX5plnwipNh+Cmm2DKFI0nTIDf/lbjefPg+9/XuKJC9S0tem6MGKHnCsDPfw4vv6zx3/4Gf/iDxu+9p+cc6Dl46qnO4JaDja6uHHJyhpNIhIlGO9i48VQCgZJPfV9Z2R+oqtKEhvr6lyks/DkAXu9acnJGGA/dsFek7O8ncLlcHmATUCciV7hcLhfwGHAtEAdeFJHJ1r9PAv4HCAC3iUje/n59SYYNgzvugLQ08Hhg9GhNAAL94Dtde55wxRXQt6/Go0Zpwg7AiSeqPiVFG2HccYcmAAH86ldOks+VVzqa887TRDjQxLnRo8Hl0iSi0aNh8GB97De/0Q810KS45M/9/vfhm9/cL4djt/z8tJ9zwmEnAHDJsEvwhnXa2fAjhvPLM/Tq4qi+RzH67NH0S+tHijuFO86+g28epi/0+tOu55TBpwBw2Tcvs7uznXHkGdw64lYAjul3DHecfQd9evXB7XIz+qzRHD/geABuHH4jI76hyXP/74T/R9yaNrW33Dz8Zs4achYAPz7xx3hcHgDOPepcenl6AXD8gOMZfdZo0jxppLpTuePsOzi6n75Rt464lTOOPAOAn5z0EzJSMwAYecxI+qX1+1yv5UBw+OG6lg4/XL/+zW/g29/W+KqrnPV34YVOItuppzqb+xFHqH7gQF2bd9yhaxXgmmuc5LmLLnI23e98R9c86Fq94w49J1JSND7xRH3suuucn3Xppc4Uw9NP13MO9BwcPdo55w420tOHMWTIHbhcvfB4PAwZMppevY761PcNGnQFHk8fAPr1G0Vq6kAAMjJO5Kij7sDt3u8f1YZDgP3uobtcrj8C5wD9rA39l8DF6IadcLlcR4hIs8vl+h/gbnRDHwlMEpGRe/rZxkM3GAwGw9eJL81Dd7lcxwA/AV7p8c+/Bf4qog29RSQ5OPinwHQrkW89MMDlcg3Zn6/PYDAYDIZDhf3toT8HPAAkevzbCcB1Lpdrk8vlWuByuU6y/v1ooOf0+Vrr3w44XV3q4SVrd6+7zvGwH30UJmsZNW+/DXfdpXFenva2jsehvV31Ndb/5qqrnNrbhx6CF7WMmunTtYc1wIYNWtcron7jRRdp/2vQ2/TZ2Rrffz+8Yl0evfoqjBmzXw7Bbvn9h7/nra1aO/vPjf9k7LKxAHxU+RHXztJm3lWdVVz8+sV4Q16i8Sg/fOOHdh33HXPv4J1CbUz/7Lpn+evKvwKwoGwBN2ZpM+/y9nIuef0SfGEf4ViYy6ZfxrbmbQDc/v7tzCmeA8DENRPtvvJ7yy3v3sLcEm1G/sTqJ5i4ZiIAc4rncPv7twOwrXkbl02/jHAsjC/s45LXL6GiXRMibsi6gYXl2ox8/Irxdl/5mdtm2n3hv0o0NelaamzUr6+4wvGwx4xxZhS8/DI8YKUjrF6taw50DV50kdaBi+ga3bhRH/vDH7RvO6iXnuyvsGyZziwAzQO5+GK9HR+P6zmSZxlpd92l5xDoOfXooxovXuzkplRWqr6ra18ela82TU1vU1p615f9MgwHIfvNmHG5XFcAzSKS63K5LurxUBoQEpFzXC7XVcCrwPc/x88dDYwGGDp06D58xT1eYJr64/0sS/Sccxw//ZRTdFgFwHHHOR80hx+uyWtuN2RkaJz0/b77Xcf3Pu00xxs//niIWsPCjjhCn9PlgsxMjfuopcbIkTr4BdSfPPZYjYcN0+c7kJz+jdM5rv9xAJw48EQGZWhCwJA+Qzj3qHMB6JfWj1FHjyI9JZ0Udwqjjh7FoEz9vjOOPIOh/fV9O2nQSbaHflTfozh7yNkADEgfwMijR5Keko7b5WbUMaMYmKGe4plDzuTYfnoAvjXoW5/bQz9ryFkc0+8YAL59+LdJsbzJY/sdy5lDzgRgYMZARh0zilRPKi6Xi5FHj6R/ur7p5ww5h6P66mI4ZfAp9E7tDcDQ/kNtb/2rRO/eupZ668tk5Ehda6BJoMfpW8k3v6nrHnStjbTMrj59VJ+ZqWtz5Ehn/Z5+uq5hUF88eV4ceaSuedBzYORIPSfcbo2Tfv6IEc7zn3SSs8aHDNHkPdCfOWoUpKfv08PylSY9/Tji8a/RFYxhn7HfPHSXy/Uk8AsgBqQD/YA5qJ9+uYjssBLhOkWkv8vl+jewQkTetvQlwEUi0rC75zAeusFgMBi+TnwpHrqI/FlEjhGR44HrgWUicjPwHpoUB/ADwBoSygfALS5lFODd02ZuMBgMBoPB4cuoQ58AXO1yubYCTwJWgQwfApVAOfAy8JU0kZ57Dl5/XeN585x50kVFcMstkEjobfjrr3d8yzvvVI8c4JlnnL7w774Lf1ULmS1bnFKdjg71EJMzoH/zG8jN1fjJJ53Z6LNmwROfz0L+wjy6/FE+KPkAgDc2v8E/1mlR/Nqatfxuvs6DrvfVc/3s6+mOdBNPxLl5zs1sb9XG9g8vfZgFZQsAmJY/jec3PA/AqupV3LNAEwpqvDXckHUDgWiAaDzKTXNuorxdB3o/uORBFldoQsPLuS/bfd2X7VjGmEWaULCjYwc3Zt1IOBYmHAtzY9aN7OjQhIgxi8awbMcyAKbkTOHlXC2EXlyxmAeXPAioh3/TnJuIxqMEogFuyLqBGq8mRNyz4B5WV68GYPKGyUzLnwbAh2Uf2jXt21u384t3f0E8EccX9nH97Oup92lCxF3z77L70h8I2tt1LbW16de3364zxgEef9yZMTBzpjObfNMmXXOga/C666CzUz30225zczmsiQAAIABJREFUZhyMH+/0Z/jPf3RtA6xf79SkNzToudDVpefGLbfouQIwdqzTV/711/XcAs0XudsaM19bq3q/f58eFoPhkOSAFDeKyApghRV3opnvn/weAX53IF7PFyEtzfEaU1Mdby8lRX1Cl8vx0T1a4kxmpj6e1PfScmd69fq4PjNTY7db46Q+I8PRp6fvWn+gSE9Jt+u1e3l6kZ6iLyDFnUJmqv4H3C43mamZuF16vZiZmml71ekp6aR6Um19WooezFR3ql3T7Xa5yUjJwO1y43K5yEzJtOvF01PSSXU7eo/b8ym9x+0hIyUDdXQgIyXDfv6M1Axbn+ZJs1/Lx/QuD5kpmbhcLly4yEjJsJ8nI9X5WWmetF2+fo/LeX63y01Gaob9+nseiwNBci0lcy32Zi3tai263bq291afoYcCj8fxz3f1/Kl6+D91XvXU93z9BoNh95he7gaDwWAwHCSYXu4Gg8FgMBzimA39czJ9ujObfOVKpya9slJrxEUgEND68vZ2fezRRx3f8dVXnb7uS5dqP3eAsjL4v//TuLtb9V7tqspf/gLF2kqcl16ChVoGzeLF8O9/77//6654fsPzrKhaAWj/9tcKXgNgc+Nmu6a8NdDKvQvvJRQLkZAEf1r8J6o7qwGtPU960HOK5/DmljcB7Z/++KrHAWjqbuLehfcSjoWJJ+KMWTSG2i5tRv7M2mdsD3pW4Sy7L/vGuo1MyFYTuN5Xz30L7yMajxKNR7lv4X00+DS/ckL2BDbWaSH121vfZlbhLEBzAJJz3mu8NYxZNIZ4Ik44FubehffS7Nf+R4+vetzuK//G5jd4t/hdAFZXr7Zr0qs6q/jT4j+RkATBaJB7FtxDa0AHeo9fMZ4tTVsAeK3gNbsmfn/h8+laSpZXPvSQzigH+Ne/nP4KCxc6fdWLinTNgXrn99zjtGV98EEo13QGXnhBa85BvfBpmk7A1q06Xx3Uu7/nHj0nRPQcSfZ3mDRJzyHQc2r6dI3z85057S0tcO+9EArts0NiMByymAbBn5OaGv1wAv2wqarS2OfTTTmRgEhE46CWWFNerh+MSX3S5Whu1uEXoB+4paX6WDiscTCodbhlZY5+507Ha2xqcga4HCiqOqs4so8W1Td2N9obXUeow05cC0QDlLaVEolHSHGnUNZehi/is/XJvuwNvga6I7pTtAfbqeiosPVl7WXEEjFcLhel7aX4wqrf0bGDkwdp0/F6Xz0xa2hFW6CNyo5KAPwRv60HKGsvs5+nsqPSrhev89XZfnpboM1OnOuOdFPaXkpc4kQTUcray/BH/NAbKjoqaA+22/pk//aWQIut94V9lLWXkZAEkXiEsvYyu96+vKOcjqA2Pa/x1hCNR7/4m7IHQiFdS6GQ9lUoK3MuFHfudHod9FxLXq9+X099OKy17GVlzsVBVZUzl6Cx0WmE1NHhbPrBoGoiEV23n9Qne8k3NDgXwB0dzmCjQECfPxr9etWiGwz/DcZDNxgMBoPhIMF46AaDwWAwHOKYDf1zMncurFihcW6u1t+C3jJM1uGGw1of7tO7xLzwgnML8d13tVc2aE/sGWoBU1sLzz6rcSik+mTt7eTJzq392bOdXtzr1mkt+oHkra1vsale74qsrFpp16SXtZXZNeFd4S6eXP0k0XgUEWHimok0dmtR/hub3yC/QQuhl+1YxvxSTSjY3rqdl3JfAqAj2MGTq58kloiRkARPr3maFr8W5b9W8JrtQS+pWGL3VS9sLmRq3lRAb59PyJ5APBEnnogzIXsCbQEtxJ6aN5XC5kJA+8cvqVgCwJamLXY+QIu/hafXPI2IEEvEeHL1k3SG1PN4Kfclu6Z+Xuk8u6Y9vyHfntPe2N3IxDUTEREi8QhPrH6CrrDeZ34x50Xbmnh/+/usrFr5xd+UPRAM6lpK2kSTJjk2z6xZWjMOsGYNZOmYd3bsgOe1PQB+v+qTHvY//uHMVp8xA3JyNF61yqlJLy935qT7fKqPRPTrZ57RcwX03En2V1i+XM8t0Fvs//qXxl6v9l6I7l9nwmA4JDAb+udk5UpnOMXWrbBokcY7d+oHYjyuH2JZWTrQAvSDLukprlihjTsACgrgo480rqrSzVpEP8Rmz3bmS8+Zo0l3oElIyeEW+fmaWHcgWVyx2N5QN9ZttDeksvYy3i/RbMEWfwuzi2fTHekmloiRVZxlJ7UtrFhoD1pZX7ue1Tv16qaktYS5pfqJ3uRvIqs4i0A0QCQeYXbRbOp8dYA2cEluyGtr1pK9U6fWFLcWM79MLw4auhvIKs4iFAsRioXIKs6yLyjmlc2zN+Q1NWtYV6sJdoXNhSwo14Y3db46ZhfNJhwPE4gGyCrOoqm7CdBEwNI2bW64uno1G2q1Y9DW5q0sqtDFUOOtIas4i1giRnekm6ziLPuC5L2S9yhrU4N6ZfVKcupz9sG7sns6Oz++lrKynKS0pUudtZSX5yS4VVbqmgP1tWfP1jUpovrkBcGSJbqGQdd08kK3vNzZ3FtbVePzQSymcdKrX7TISRbduFEvCkA39A/0OpHmZn1+01jGYPhsjIduMBgMBsNBgvHQDQaDwWA4xDEb+hegtNSpCW9vhze1pJp4XOvNk77jrFmOb7hyJWzW0eAUF+ttS9Bbk8ke79EoTJ3q+I4zZ2pZEeht0W3b9u//a08sKl9ESWsJoL5xsqa8rquOrCI1YYPRIK/mv0pCEgBM3zzd9qA/LPvQvuWcW5/Lmp1rANjp3WnXdPsjfqblT7P1rxe8bnvQ80rn2eVpG+s2sr5WTeCqzirbz/eFfbxW8BoigojwWsFrdtnaByUfUNVZBegt/2RNemVHJfNKtbG4N+Tl9QJt2J+QBNPyp2nZGvBu8bt2X/c1O9eQW68mcFlbmd2jviPYYfvp8UScV/NftcvWsoqyqOtS+2BV9SoKGgv+uzdiL/nkWpoxQ29jg95yL1T3gq1b1ccGXWszZ2ocDqs+phWAvPWW0xd+8WKnpn3zZqemvL7e6REfCum5ELem3L75plOetnChnkOgt/yz1T2htta55R8IqD6R2DfHw2A4lDEb+hdg0SIneWjrVh02EYvp5jx2rNacAzz2mDOcZepUx1/88EMneaigQAe9iOgH7iOPQJ1+7jNunOO7v/SS4y9+Gfwz5592Itqc4jlMK9BuIutq1/FEtk6Kqeqs4pHlj9AebCcSj/DI8kds33vShkl8VKmJA+8UvsP0zdpNJHtnNk+teQrQzXXs8rF0hbsIRoM8suIRilu0s84/1v3DTkSbsW2G3ZhmZdVKJq6dCKifP3b5WLoj3XRHuhm7fKzte09cO5FV1WrWvrnlTbsxzdLKpTy7XrMSt7du55EVjxCMBvGGvIxdPta+iJiwZoLt27+++XVmFWlW4pLKJUzaMAmAwpZCxi4fSzQepT3YztjlY6n2qvH8+OrH7YuQaQXTmFM8Zx+8K7unsVHXUvKC8tFHHd/83/92hqN88IGuLdBEt/HjNa6rU31zs67NsWMd33zKFF3DoMmer76q8YYNuuZB/fKxY/UiIBbTOHlB+vzzTmOb2bPhtdc0XrNGE+FA/f5HHnH6MBgMht1jPHSDwWAwGA4SjIduMBgMBsMhjtnQDZ+LnLoc2wOuaK9ga5PWHbUH2+1b2bFEzK4vB/Xdkx7yhtoNdl/1srYy+1Z8a6DVvpUdjUf5sOxDW7+wfCGhmCYkrKtZZ5eglbSW2Lfim/3NrK3RAv1wLGz72aD15uFYGFDfO1lCVtxSbOcDNHY32rfCQ7GQbSuA+v7JFq2rq1fbNe2FzYV2PkC9r9724wPRgD2zHWB+6Xy7De2q6lV269etTVupaK/4rEP+hUgk9LZ60oNetsxpvZqX55SQVVU5t9K9XsdPj8dVn7yR99FHTl/3TZucmvTKStii1Yx0dDh+eizmzC4AvcWeLEHLyXFspfJy51Z8W5vTqyEa/bjeYPgsygIBCq1F1hqJkG35NdFEgg+TCSDAwrY2QlZyxzqvl8awfkaUBAIUW/rmSIS1Vq/kcCLBgh76BW1thK0Ta43XS4uVqFLs91OSbPxwoEkmDh2Mf84++2wxHFgunHahPL7qcRERuXfBvXL1zKtFRGR6wXQZ9twwERHZ0rhF0h9Ll6buJgnHwtL3ib6ytHKpiIiMemWUTFwzUURE7pp3l9ww+wYREXkl9xU5+fmTRURkU90myXgsQ9oCbeKP+KXPE31kVdUqERE5699nyaT1k0RE5Nfv/1puffdWERGZsnGKDJ8yXERE1tWsk8zHM8Ub8oo35JXMxzNlfc16ERH5zpTvyIs5L4qIyC3v3iK/+eA3IiLy3Lrn5Ox/63paVbVK+jzRR/wRv7T6WyXjsQzJrc8VEZFvPf8tmZo3VURErp99vfxu/u9EROTp7Kfle698T0REPqr4SPo+0VfCsbA0+hol/bF02dK4RUREjn/ueHlj8xsiInL1zKvlvoX3ffE3ZQ/s2CGSni5SXq5fDx4sMmeOxj/6kchf/qLxn/8s8j//o/GsWSJHHqlxSYnqq6tFEgmRAQNE5s3Txy6+WOSvf9V4zBiRn/1M4//8R2ToUI23bVN9fb1INCrSr5/I4sX62Pnni0yYoPHdd4v8/OcaT5smcuKJGufni2RkiLS07MujYjiUuaukRG4oLBQRkVfq6+Xk9Xru53Z1ScbKldIWiYg/FpM+q1bJqo4OERE5KydHJtXUiIjIr7dvl1uLikRE5MXaWhm+caOIiKz3eiVz5UrxRqPijUYlc+VKWdfZKSIiwzdulCm1tSIicmtRkfx6+/b99v8DNslu9kTjoRs+F/FEHI/bA+jFYEIS9texRIwUd8pexyKCILhd7v9Kn8yC/zL1Lly4XK5PHZu90ccTcdwut63fX8RikJLy6TgeB7cbXC79DTyRAI9nz5pP6pPfvy/0Ivp69qQxGD4L/VwBt3VexRIJUtzuvY4T1p64r/T7mj156GZDNxgMBoPhIMEkxRn2GeXtzvjPpu4mdnrVhA1EAxS1FAF6hZyszwYoaCxwRpm2leENqSfV2N1ot4T1R/y2H56QxMf0+Q35xBPqdZW2ldo16fW+etvP94V9dkvXhCTsmeUAeQ159m/j21u326NY67rqbD+/K9xll7bFE3G73zxovXxSX9xSbNek13hrbD+/M9Rp++mxROxj9eWb6jeRvHAubC4kEFV/bad3pz1+dn/S85p3yxatLQctCUtagq2tzryAUMhpyfpJ/ebNTk17RYVTU95zFHAw6NS3i3xcX1Dg9GXvOVa4qckp8/T7dSZ7Up/rLAWDYZeUBQJ0WgurMRymxmoC4o/HbT88IUJucsAGkO/zEbfOy9JAgC6r2UJ9OEyddZL4YjG299Dn9dDn+Xz2b+Pb/X58lr4uHKbB0nfFYpQeSD99d/fiD4Y/xkM/8Ix6ZZSMWz5ORNQDv/LtK0VEPfBj/nGMiIjk1eeJe7xbGn2NEo6FJe1vabKwbKGIqAf+xKonREQ98GveuUZE1AP/5qRviojI+pr14h7vllZ/q/gjfkn9a6rtwQ+fMtz24G9991bbg5+0fpLtwa+uXi0pf02xPfSUv6bI6urVIiJy8vMn2x78DbNvsD34iWsm2h780sql0utvvWwP3T3eLRtqN4iIyLDnhtke/NUzr7Y9+MdXPW578AvKFkj6Y+kSjoWlwdcg7vFuyW/IFxGRo/9+tO3B/+9b/2t78PuLigoRl0uktFS/7t9fZOZMjS++WOTBBzUeM0bkhz/U+D//ERk0SOPiYtVXVamH3ru3yLvv6mPnny8ydqzGd98t8pOfaDxtmsiQIRpv3qz6ujr10DMyHA/+u991PPg773Q8+Jdecjz4TZtE3G6R5uZ9elgMhxhn5+TI41VVIiLym+3b5eqtW0VEPfBh69aJiMgGr1fcy5dLq+Whp65YIUvb20VEPfCJ1dUioh540oOfVFNje/CrOzokZcUK20NPWbFCVlse/Mnr19se/A2FhbYHP7G62vbg9xUYD92wr/CGvGSmZpLqSSUUCxFLxOjTqw8JSdAZ6mRgxkBAB7QM7j34U3FnqJPeqb1J9aQSjAZJSILevXoTT8Txhr17pe/Tqw8p7hQC0QAiYuu7wl0clnHYHvUdwQ76pfXD4/bgj/hxuVxkpmbag1QGpA/Yo7492E7/tP543B66I914XB4yUjOIxqP4o/690g9IH4Db5aY70k2KO4X0lPT98l4laWmBwfr0tLbCoEHqm3d1QVqa/gmH9Tfvvn31t+L2dv2+XekPP1xjrxcyMqBXL/2tPhpVfSKhme670veMvV7IzITUVP2tPh6HPn1U39kJAwd+WmMw7IrOaJTeHg+pbjfBeJwE0NvjIS6CNxZjYGoqAC2RCIN79fpU3BmN0sfjIcXtJhCPIz30XbEYh32GviMapV9KCh6XC388jgvI9HiIJRJ0x+MMsPT7AuOhGwwGg8FwCGA8dMM+o8XfYteEd0e6bT89lojZfjRg9zsHqO2qtT3kZn+zXRPuC/vsHu/ReNT2oz+p7xn31HeFu2w/PhKP2CNO96Rv6m4iElcT2Bvy2n58OBb+mJ+9O31jd6Ndk94Z6rT9+FAsZNe3i4idG/BJfYOvwc4n6Ah22D3m9yc1ztNTV+fUpLe1OXPS/X7HD4/HtR/7rvS1tU5Nemur/mYNWpueHNEaizmtZsGpdf+kvqXFmXewJ33P5zcYdkVLJGLXhPtiMdtPjyYSdn05YHvrn4ybe+i7YjG8lh8eSSRoSiaN7EHfFIkQsfTeWMz248OJBM099Psbs6EbPhf/+/b/MiF7AgAPLHmAW9+7FYDXCl7j3JfPBTQJ7bjnjrM3zxMmn2DPCv/xmz/m7+v+DsB9i+7j1x/8GoCX817mvKnnAdp85vhJx9MWaCMQDXDC5BPs/u2XTr+U5zdqA/27F9zNb+f/FoAXc17kB6/9ANDmL8MmDaMr3EVXuIthk4bZTWsufO1C/rXpXwDcOf9O/rDgDwBM3jCZy6ZfBmhf9xMmn0AgGqAt0MZxzx1nN4353tTvMTV/KgC3f3A7YxaPAeDva//O5f+5HNBGOCdOPpFIPEJjdyPHPXecnSR3zsvn2P3rb3nvFv7vo//bB+/K7qmogKFDoUzz9TjtNB0WBHDNNU7P9kcegeuu03jGDBg+XOPiYtVXV+tGfPLJziyCn/4UntD2/fz5z3DzzRq/8QacdZbGmzfDccfpBUIsBied5PR//8lPYKK23+dPf4Jf/UrjV1+FUaM03rRJ9S0t+/a4GA4tfrxlC3+3rvz+WFHBr0u0YdQrDQ2cl68Jrhu6ujh+/XraolEC8TgnbNjAMusq8tLNm3ne6pJ0d1kZv7WmBr1YX88PLP3qzk6GrV9Pl7VhD1u/ntVWVueF+fm8aF0F31layt3WCTe5tpZLk9O4DgS7M9cPhj8mKe7As7Nzp3SFukREpC3QJg2+BhERCUVDUt5Wbn9fYXOhHRe3FEs8ERcRkaqOKvGFfSIi0upvlUZfo4iIBKNBqWivEBGRRCKxW/2Ojh3SHe4WEZEWf4s0dTeJiIg/4pfK9kpbX9RcZOuLmoskkUiIiEhle6X4I34REWnqbpIWv3Ys6Q53y46OHSIiEk/Epbil+GP/l6S+or1CgtGgiIg0+hql1d8qIiK+sE+qOzWpJhaPfUqfpLytXELRkIiI1HfVS1ugbTdHet+xbZsTl5SIRCIa19aKWH0xpKNDE9dE9PFkEt0n9du3a3KbiMjOnSJer8bt7do8RkQkFBIpK3M0hYUf18diGldXi3TpUpK2NpEGXUoSDDqNcD6pNxh2RXUwKD5rYbZGItIYDouISDAWk4pAQESsz5XubltT3N0tceu83hEISLe1MFvCYWmy9P5YTCp76It66Iu6u53PlUBA/Ja+KRyWFkvfHYvJDku/r8AkxRkMBoPBcPBjPHTDfiGWiNl+NGDXV+8pDsVCdk13T72I2P3e96QPRoO2Hx+NR20/+8vQR+IR2w9PSMLOLdjbY3Gg6FkG2zMOh5055fG4U5++J03POBRy/PhYzKlP3596g2FXhOJxuyY8lkjYfraIEEwuciCwmzgYjzufC4kE0f2k39+YDd3wX/PHRX/k+tnXA+qhn/rCqQBsbtzMgAkDbA998MTBLKlYAsCF0y7kmbXPAPD7D3/PLe/eAqiHfvq/Tge0EcthTx1me+iDnh7Eyiqd9vG9qd/jufXPAXDHvDv41QdqvL6Q8wLnvKwXrWtr1jLwqYG2hz7wqYGsq1kHwNkvnc2UHB1C/8v3f8md8+4E4Nn1z3Leq+rhr6hawaCnB9ke+mFPHcamer0TNPzF4byS9woAN8+5mbs/vBuAp9c8bXv4iysWM3jiYNtDHzBhAFuadHLJt//5bV4veH2fvQefRWUl9O+vTVwAjj0WsrI0vuIK9c4BHnoIfvYzjWfOhGHDNC4pUX3SQx8yBObO1cd+9CP42980vv9+uPZajd94Q7120IErAwY4HvoRR8BCa+7NxRfDBE3H4J574MYbNZ46Fb7zHY3z81VvPHTDnriwoICJlof++7IyflGsTapebmjgdOsu7qauLg7LzrY99EFr1rDS8sC/l5fHc5aHfkdpKb+yPPgp9fWcY3U2Wuf1MjA72/bQB2Zn24NbzsnNZYrlof+qpIQ7LQ/+udpa28M/IOzuXvzB8Md46F8u1Z3Vsr1FhxB4Q167+UosHpNllcvs71tVtcr2jQsaCqS5W7uEVHVUSWmrmrUdwQ7JqcsREZFoPPox/cqqlRKOqSeVV59n+9aV7ZW2b98WaJNNdZtERCQSi8jyHctt/fIdyyUSU+N4U90m27cubyu3ffcWf4vk1eeJiEg4FpaVVStt/bLKZRKNqz+XU5cjHUFtJlHaWipVHdrMorm7WQoaCkRE8wmSw2REtFFNLK7+2obaDeINeffq+O4LEgmRpUv1bxGRNWtEkjbgtm2Ob15b63jVPp/I2rUf1yfJzhZJWoJbtji+986d2oRGRH11qxeHxOMf169apR65iDadadIUCKmuVn9fRH39ZC+OWExkmbMUDIZdUuDzSbPlW1cFg1Lq1zyZjkhEcqxEj2g8LsusRjIiIis7OiQc19ycvK4uabWSSyoDASm3FnlbJCKbrESPSDwuy3vol7e3S8TSb+rqkjZLXx4I2L57SzgseclEkX0ExkM3GAwGg+Hgx3joBoPBYDAc4pgN3bBficQjfOOZb7CqehUAF712EZM3TAbgvoX38cv3fwnA9M3TGfGvEYAOcxn09CA6gh0EogGOmHiE7YGfN/U8Xsx5EVAPfvTc0QBMzZtq18Hn1OVw+NOH4wv76Ap3cfjTh9se+Lkvn8vUPK0jHz13NL//8PeA1rGf/+r5gHrwR0w8gmA0SHuwnUFPD2Jzo9aSjvjXCN7Y/AYAt713G39c9EcAJq2fxMWvX7xfjuEXobpaW6ju2KFfDx0K8+ZpfOWVjgc+bhxcdZXG773neOgVFapPNoQ56ihYvFgfu/xyeOopjR96CK7XdApmzXI89O3bVd/YqB76kUfC8uX62KWXwrPPanz//fCLX2j81luOh751q7aQbW3dp4fFcIhxUX4+ky0P/I/l5dxmeehvNDYyIicHgM3d3QzKzqbD8tCPWLOGdZYHfl5eHi/W6aCn35eWMtry0Kc2NHCu5aFv6uricMtD98ViHJ6dTU6XNqY6NzeXqVY3pNElJfw+WcdeV8d5ec6gqP3O7u7FHwx/jId+cDCrcJZdu760cqld772lcYvtm9d11cmCsgUiojXpb299267xnLltpl27vqRiiV3vnd+QL7n1uSIiUuOtkUXli0REa9JnbJ1hP/+MrTPs2vNF5Yuk1lsrIiK59bm2713dWS1LKpaIiNaUz9ymE0wSiYS8vfVtu/b8w9IPpb5LC6431m6ULY1bRETr43v6/l8VolEdtpKsHX/vPa35FhFZvdrxrbdvV39dRKSlReT99zWORFSfrB3PytKadRGRlSudevHCQhFrBoY0NYnMnatxOCzy1lvqpYuIzJ7t1L4vXy5SqSkMsnWryAZNwZCGBpH58zUOBlWfzAEwGHbF0vZ2u957i88nGy3fvD4UkgWtmnMTjMXk7cZG53OlqcmuXV/S1ibVVnJHfleX5Fq+d00wKIusE8Yfi8mMZNKHiMxoarJrzxe1tUmNpc/t6pJ8S18dDMqStn3bawLjoRsMBoPBcPDzhTx0l8t1t8vlOmzfvyyDwWAwGAz7ir3x0L8B5LhcrndcLtePXS6Xa3+/KMOhQyQe4fQXT2dD7QYArn7nal7NfxWAcSvGcd/C+wCYXTTb7qVe1FLEqS+cijfkJRgNMvzF4eQ1qA915dtX2h72w0sf5oElDwAwY9sMu5f6lqYtnDblNLoj3XRHujltyml2Hfjl/7mcGdtmANqL/uGlDwPq4f90xk8ByK3PZfiLwwnFQnhDXk594VSKW9STu2z6ZWQVaSH3vQvvZfwKbYY+NW8q17xzzX45hl+Eujo45RT1wAHOPReWLtX4tttg0iSN//53+LW21WfRIqeXenW16hsb1UM/80xYpekQ3HQTTNGSfiZMgN9qW33mzYPvf1/jigrVt7Sohz5iBKxdq4/9/Ofw8ssa/+1v8Adtq89772mNOmgd/KmnOoNbDIZdcc22bbxqedjjduzgPqvxQlZLC5cV6ByFIr+fUzduxBuLEYzHGZ6TQ55PhytduXUrbzTqcKiHKyt5oKICgBlNTVy+RT87tnR3c9rGjXTHYnTHYpy2cSNbunW40uVbtjCjSYdDPVBRwcOVlYB6+Fdu3XogDgEAnnHjxu3xG8aNG7ds/PjxzwNdwC+BCePHjz9y/PjxVePGjftST7OXXnpp3OjRo7/Ml2D4DDxuDwlJcMHQC+jdqzdu3Jxx5Bkc0fsIUtwpDBswjBMGnkAvTy8GZQxixJEjSEtJw+PycMFm8RqVAAAgAElEQVTQC0j1pBJPxLnwuAvJSM3A5XJx1pCzGNx7MCnuFE4YeALDDhtGL08vBmcOZvg3hpOWkkaqO5Xzh55PqieVhCS48LgLSU9Jx4WLc446h0GZg/C4PZw08CSOH3C86nsP5jtHfIc0Txq9PL0479jzHP3xF5KWkoYLF+cefS4DMwbicXn41uHfYmj/ofTy9OIbvb/BaUec9mUf8o+Rlqbd2H7wA0hJ0X877zzo1w/cbk0+O+oonUl+7LHwrW+ppn9/HbCSlqYbeU/9+efr3HK3G04/XRPdUlJ0iMpJJ+l89MMO080/LU1nr194oT6HCFxwAfTurfoRI+Ab39DHjj8eTjhB9QMHwhlnqN7t1gsEt0nhNewGF3BGnz4c0asXKS4Xw9LTOSEjg14uF4NSUxnRpw9pbjcel4sL+vcn1eUiLsKFAwaQ4fHgAs7q25fBlv6EjAyGZWTQy+1mcGoqw/v0Ic3lItXt5vz+/Ul1u0kAF/bvT7rHgxs4u29fBqWm4nG5OCkzk+PT0+nlcnFEairf6dNnn/1fx48f3zBu3LiXdnkc9tZDd7lcI9AN/cfAcmAUsEREHthXL/TzYjx0g8FgMHyd2JOHnrIX4nuAW4BW4BXgfhGJulwuN1AGfGkbusFgMBgMBmVvbmINBK4SkR+JyCwRiQKISAK4Yr++OsNBTzQe5Ydv/NCu475j7h28U/gOAM+ue5a/rvwrAAvKFnBjljbzLm8v55LXL8EX9hGOhbls+mVsa94GwO3v386c4jkATFwzkSdW60DuuSVz7b7w21u3c+n0SwlEAwSiAS6dfiklrVpXesu7tzC3RJuRP7H6CSau0YHcc4rncPv7twOwrXkbl02/jHAsjC/s45LXL6GiXT21G7JuYGG5NiMfv2K83Vd+5raZdl/4rxJNTXDRReqBg/ZvT3rYY8bo7HFQL/sB69J89WqtUQftwX7RRVoHLqK15xt1NDx/+IP2bQf10h/WdASWLYOrr9Z45071wzs6dADMj34EybLcu+6Ct9/WePJkePRRjRcvdmazV1aq3ir3NRh2yR0lJbzT3AzAczU1jK+qAmBhWxs3FBUBUB4IcElBAb5YjHAiwWUFBWyzPPDbt29njjUwYOLOnTxRXQ3A3NZWbrFq2ksCAS4tKCAQjxOIx7m0oIDtfj8AtxQXM9dqlvBEdTUTd+4EYE5LC7dv334AjoDymb+hi8ije3iseN++HMOhRoo7hVFHj2JQ5iAAzjjyDIb2HwrASYNOsiecHdX3KM4ecjYAA9IHMPLokaSnpON2uRl1zCgGZgwE4MwhZ3Jsv2MB+NagbxEXnWR0dL+jbf1h6Ycx8uiR9PL0AmDk0SMZkD4AgLOGnMUx/Y4B4NuHf5sUt54Cx/Y7ljOHnAnAwIyBjDpmFKmeVFwuFyOPHkn/9P4AnDPkHI7qexQApww+hd6pvQEY2n8oZxx5xr4/gF+Q3r01wa23vkxGjtQBKQDDh6vvDfDNb6pfDeppjxypcZ8+qs/MVC985EgYPFgfO/109b0BTjxRfXdQT/2739W4b1/VZGSoBz5yJBx+uD42YoTz/CedpM8LOgDmXO0RRP/++vzp6fv0sBgOMc7o04eh1gI+KTMTvzXh7Ki0NM7p2xeAASkpjOzXj3S3G7fLxah+/RiYmgrAmX36cKyl/1ZmJnHLij4mLY2zP6HvZeWFj+zXj8Ms/dl9+3K0pf92ZiYe63uOTUvjzH3on38Wpg7dYDAYDIaDBNPL3WAwGAyGQxyzoRsOKA8vfZgFZQsAmJY/jec3PA/AqupV3LPgHgBqvDXckHUDgWiAaDzKTXNuorxd60ofXPIgiyu0mfjLuS/bfd2X7VjGmEVjANjRsYMbs24kHAsTjoW5MetGdnRoM/Mxi8awbMcyAKbkTOHlXC2EXlyxmAeXPAioh3/TnJuIxqMEogFuyLqBGq/OWr5nwT2srl4NwOQNk5mWP20/Hq29Z84ceOwxjQsK4JfaIp/2dvWj29r069tv1xnjAI8/DrNnazxzpjObfNMm+M1vNG5pUX1np3rot92m/dUBxo/XmnGA//wHntEx96xf79SkNzRoj/euLi2fu+UWsCxNxo51+sq//jo8p+kIZGfD3Tpmntpa1VtWpcGwSx6urORDa5FPa2iw+7qv6uzknrIyAGpCIW4oKiIQjxNNJLipqIjyQACABysqWNzeDsDL9fV2X/dlHR2MsWradwSD3FhURDiRIJxIcGNRETuCahmOKS9nmdUs4cW6Ol62ZqMvbm/nQaum/UDwmR66wbAvSU9JJ9WjvlMvTy/SUtR3SnWnkpGaAYDb5SYjJQO3y43L5SIzJROPy+Po3Y7e4/Z8Su9xe8hI0Zp1gIyUDNsrz0jNsPVpnjT7tXxM7/KQmZKJy+XChYuMlAz7eTJSnZ+V5kmzX/+XTWqq4zOnpKjnDepbZ2Y6NdwZGU49eXq61nyD/v1Zepdr7/UZeijxeBz/fFfPb1mQpKU5Hn5q6sf1PV+/wbArMtxuUq3zPc3tJmJZyakuFxnW4nFbsRutW8/0eGyvO72HvpfbjWcXeo8Vu3o8Z1Kf8Ql9Mu6pPxAYD91gMBgMhoME46EbDAaDwXCIYzZ0g+Eg5ZVX4MMPNV6yBF7UdAJKS+HPf9bY54N77nHquB96SGeUA/zrX85s84ULnb7qRUXwl79o3NmpeqtclwcfBMtS5IUXtOYc1AufZqUTbN2q89VBvft77oFAQD34++93ZrNPmgQrV2r8/vswfbrG+fnOnPaWFrj3XgiFvtChMhziPFdTw+rOTgDebWnhTavxQp7Px+NWTXlTJMK9ZWWEEwniIowpL6fWWljP7Nxpz0af1dxs92Xf2NXFBEvfEA5zX3k50USCaCLBfeXl1IfDAEyormajdZK93dTELKsmfq3XyzNWTfqBwHjoBsNBSk2NeswAzc06SAXA69VNHXQjLC3Vv/v1g7IyfRy06UuyRLapSb9O6q08IlsfDmste1mZc3FQVQWDtL0AjY3ahAa0iUxy0w8GVROJqEf+Sf3RR2vc0KAJfEl9Mo8oENDnj0ZNLbph9+wIhTjOWiANkQhdsRgA7dEoFVbiWiAepywYJCaCCygNBvFZ9eo7QiFOthJH6iMRYpYV3R6NUmlt+t3xOGWBgP1YWSBg17tXhkKMiEZtfYrlobdFo+w4gFejxkM3GAwGg+EgwXjoBoPBYDAc4pgN3WA4SJkzR2u2ATZsgHe0RT47dzo13cEgPPGE3roG9a2Tt+ZnzdKacYA1ayBLx7yzYwc8r+0B8PtVn7xr+I9/OLPVZ8yAnByNV61yatLLy5056T6f6iMR/fqZZ/T2Omjtem6uxsuXw1xtsU9pqfr7oLf/n3xSb7kbDLvjjcZG8q3Z5ss6Ophn9VXf7vfzkuUFdUSjPFldTSyRICHC0zt30mItzNcaGuzZ5kva21lo1bQX+v1MtRZsWzTKhOpq4iLERZhQXU2btTCnNjRQaDVLWNjWxhLLP9rS3c1ryQV/ADAbusFwkLJ8ubMh5ufDRx9pXFXlNIzp7NTY6nlBVpaTlLZ0qTMoJS/PSXCrrNSLBVBfe/Zs3VhFVJ+8IFiyRJvYgDajWbFC4/JyZ3NvbVWNzwexmMZJr37RIqdJzcaNelEAuqF/8IHGzc36/KaxjGFPLGpvZ6u1SDZ0dbHaShQpCQaZa23OzdEoWS0tBBIJIokEs1taqLOS2j5sb7c35LVdXWRb+mK/n/mWvjESIau1lVAiQSiRIKu1lQZLP7+tjWJLn+31ss5KFCn0+1mQTA45ABgP3WAwGAyGgwTjoRsMBoPBcIhjNnSD4SBlxQrYskXjoiLnlntLizNnPBqFqVMdD3vGDL2NDXrLvbBQ461b9RY+aAnbzJkah8Oqt6qAeOstpy/84sVOTfvmzU5NeX29c8s/FNKZ61Z1D2++6ZSnLVzolNfl5Tn5ALW1zi3/QED1icR/fZgMXwMWtLVRZiWK5Pp8rLFumdeEQrxrzTn3x+NMa2ggYd2Vfr2x0S5vm9faSqVV3raxq4v1lr4qGOQDy4/vjsV4raEBEUFEeK2hAZ+l/6C1lSpLv97rtWvSK4NB288/EJgN3WA4SHn5ZW3IAjB/vtNYJj8fHnlE48ZGjZN5OY8+6vjm//63Mxzlgw/gpZc0zsnRwSsAdXWqb25WD33sWMc3nzLFaWzz7ru68YIm6CUHxezcqZq2Nr0oGDsWtm3Tx55/3mlsM3s2vPaaxmvWaCIcqN//yCOaC2Aw7I5JtbUssRJFZjU3M91qLJPt9fKUlbRRGQwydscOumIxgvE4j+zYQbF1EfCP2lp7uMqM5mbetBrLrPR6mVijg5lKg0HGVlXRHY/THY8ztqqKMmsTn1hTw0rrIuDNpiZmWFfNyzo6+Ecyi/QAYDx0g8FgMBgOEoyHbjAYDAbDIY7Z0A2Gg5SCAqcEbedO51a6z6f+OKj3PG+e40EvW+a0Xs3Lc0rIqqqcW+ler+Onx+OqT97I++gjp6/7pk1OTXplpePnd3Q4fnospnZAksWLnRK0nBy9pQ9a6pa8Fd/WBqtX/9eHZZ8SisdZkEwaAD5sayNiHczszk5areSEIr+fUuv2bUM4zAbrIAfjcRb1KFua39ZG1NKv6uyk3apj3tbdbc/mrguHybH0/njcrmkWEea3tRGz9Cs6Ouiw9Fu6u20PuCYUIteqye6Oxfioh35eaytx681c3tGBN5kccQhRFgjYJWitkQjZll8TTSTsmemg9eIhK7ljnddLo1WCVhII2CVozZEIa61b6eFE4mNrYUFbG2HrvVjj9do17cV+PyXWe9kYDtt+/IHAbOgGw0HKI484DVhefVUHp4Buxtddp/HOnXDttc7Gf/31zmb/0EPOQJaXXoKHH9Z4yRK48UaNKypUX1Ojm/q11zqb9QMPOANZpkxRfx5gwQK45RaNS0rgmmvUw4/FVL92rT52332aJAcwebIzkGXuXPjVr/bNMfqi5HZ3c21hId2xGJ3RKNcWFrLZuqIZXVrKu1bC05M7d/KsdXUzo7mZP1jN8Dd0dXFNYSGheJzWSIRrCwvtzeb2khI74epv1dVMtq5u3mxq4j6rGf4ar5drCguJJRI0RiJcU1hIqbVx37p9u13j/GhVFVMs/WuNjdxvNcNf6fXy86IiRISacJhri4rsjf/G4mL7YuFQ4rnaWnsgy/ttbfy6pASArX4/1xQW0h6NEojHubaoiBzrwuf3ZWW8YyXPPVNTY/vuc1pauNPK3Czo7uaawkK6YjG6YjGuKSy0m9n8trSU2Zb+qZ07ecby3d9paeF3ycEIBwDjoRsMBynxOLjd4HLpZptIOMNaYjFISdlzvC/0ye/fF3oRfT2f/L4vm1giQYr1wnYXJzOn3dZQjr3RfFLvAlyWPi6C53P8rLgI7i+gP5QQEYQv9l6wn/T7gj156F+RU8ZgMHxekpsh6Kbc8+uem+Hu4q+a3vr8+9T3fdn0/DDeXezu+eL3UrMnvafH13uj93xB/aGEy+Wi59H4ou/FvtTvbw7Nd9Rg+BpQVaU156C+c2WlxuGw42eDet1JtmzRx0FvwyctwdZW/XmgtePJlqyf1G/e7NS0V1Q4NeU9x7cGg059u8jH9QUFTl/28vKvZjmaLxazPdC4CHnWbVXQGufkb2DFfj/dlgddGwrZbUC9sZjtp8cSCfu2bFKfvCta5PcTsDzcmlCIJuvAdkajtp8eTSQo6KHf1NVl67d1dxO09NWhkO3h9hwZGkkkbIsgqU+ytbvb9pB3BIN2PsDXhXyfz84nKA0E7Jr0+nDYbgnri8XYblkkiU+shbwea2G732/XpNeFw/Za6OqxFg4EZkM3GA5S7rzTqdf++98d33nuXLjgAo0rK+G733Xmm194oVO7fvvtMHGixhMmwOjRGs+ZAxdfrPH27aqvrtbN+fzzndrzW291hsA89hj87ncaz5wJP/yhxlu3qr6+Xm+jn3eeU3t+003OEJivEq80NHCVlaGX7fUyKi+P7liMjmiUkbm5dsLbldu28bpVrzymooK/WIkK/6qv57qiIgCWd3byvbw8QvE4LZEI383NJc/aYC/fsoX/WPo/lJczzrqier6ujpuKiwFY3NHBefn5xBIJ6sNhvpuXxzZrg/nhli2273tXaSmPWVdUz9bWcpvV8Wd+Wxvn5+UhIlQFg3w3L8/eoC4qKLBzAEaXlvKU5ft+HQjE44zMy2OldUV5TWGhPcTlocpKOwdhWmMjP7PWwlqvl5F5ebaHPjIvz06Y+9m2bUyzat/vr6jgz9bV9Uv19VyTvLo9ABgP3WA4SPH5oFcvSEvT35pDIejXTzfetjY4/HD9vpYWGDxY49ZWGDRIb293dak2LU1/a49EoG9f1be36/ftSp/8uV4vZGToawiF9Dfvvn3VS+/o2LW+Z+z1QmYmpKbu/2P1eYglEvjicQ6zXlhLJMLgXr0+FXdEo/RLScHjcuGPx3EDGR4PsUSC7nicAZ+hb49GGZCSgtvlojsWI8XlIt3jIZpIEEgk6G/5Dq2RCIfvQt8WjXKYpffFYqRa+kgiQXAv9QNTUnC5XHTFYqS53aQdorfhd0XPY9EZjdLH4yHF7SYQjyNAb4+HuAhdsdjnXgsuIHMXa2FfsCcP3WzoBoPBYDAcJJjGMgbDIUh7u1PTHQg4fngi4dR3g5acJamrc2rS29qcOel+v+OHx+N6i3xX+tpapya9tVX9ctDa9OSI1ljMaTULTq37J/UtLc6c9a8SkUTC9rNB/e1dxY3hsF1T3hmN2h5qOJGguYe+djf6hnDYrinviEZtPz55ex40Y3t3+vpw2PaA26NR/JYfHrRK5EB936QfvCd9m1XK9XWi57FojkTsmvKuWMyuz9/btdAUidj9CbzWLXn49FrY35gN3WA4SLnhBvjLXzT+29/gqqs0zsqCU07RuKIChg51PPTTToNZszS+5hqnZ/sjjzi16zNmwPDhGhcXqz7poZ98sjPr/Kc/hSee0PjPf4abb9b4jTfgrLM03rwZjjvO8dBPOsnx4H/yE8fD/yrxz7o6LrG67Kzo6OCbGzbYHvrx69ezzvJNL8jP5yXrymV0aSn3WrXjz9bU8CMrK3FxezsnbNhgb9LHrV9vJ6aNzMvjVct3/eX27fzJ8m2frqnhJ1ZW4vy2Nk7auNH20IeuX88Wy4M/a9Mm3rD0NxcX277t49XVtu/7XmsrJ2/YYHvoQ9evtz307+TkMNPqOX5dYSGPJJsVfA0IxOOcsGGD3b/90s2bed7qI3B3WRm/tWrPX6yv5wf5+QCs7uxk2Pr1toc+bP16Vlse/IX5+bxoXQXfWVrK3dYJN7m2lks3bz5w/7Hk5JiD8c/ZZ58tBsPXlbo6kY4OjTs7RWpqNI5GRUpKnO/bts2JS0pEIhGNa2tVJ6I/p65O40hEpLR01/rt2/Xni4js3Cni9Wrc3i5SX69xKCRSVuZoCgs/ro/FNK6uFunq+nz/5wOBPxaTHYGAiIgkEgkp6u62Hyvq7pZEIiEiIpWBgASs/0xjOCwt4bCIiHTHYlIVDIqISDyRkOIe+sIecXkgIEFL3xAKSZv1xnRFo1Jt6WOJhGz3+3epL/P7JRSPi4hIfSgk7ZbeG41KjaWPxuNS0kO/rYe+1O+XiKWvDYWkI7kwviYUd3dL3HovdwQC0m29Fy3hsDRZ76U/FpPKvVwLfkvf9Im1kFxL+wpgk+xmTzQeusFgMBgMBwnGQzcYDkHCYWfOeDzu1JeD443vKf6i+lDI8eNjMac+fW/1X1VExK7vBj7mLfeMg/G4XRMeTSRsPz0hYtd370nfM44kEraf/t/qk374f6MP99B/Xdib9/KLroVP6vc3ZkM3GA5SrroK/u//NB43Dv7nfzR+9104+miNKyuhf39t4gJw7LHqsQNccYUzN/2hh+BnP9N45kwYNkzjkhLVJz30IUO0zh3gRz9y+q/ff7/2aQf10E8+WeNt22DAAMdDP+IIWLhwnx+KfcrkujpGWpNuVnd2MjA72/bQD8vOZqPlgZ+xaRP/tnzTW7dvt3t2/6Omhgss33VZRweHr1lje+gDsrPtRjPfycnhVcuDv7GoyPbgJ+zcySWW77qwrY0j1qyxPfT+2dlsszz0b23YwJuWh35NYaFdO/3Xqip+bHn4H7S2MmTtWkSE6lCI/tnZdqOT49ev5x3LQ//p1q08nOxM9DUgEI8zaM0auw79e3l5PGd56HeUlvIrq//7lPp6zsnNBXSAy8DsbNtDH5idbdehn5ObyxRrLfyqpMTu//5cbS3nWWvhgLC7e/H76g/gAfKBeZ/498lAd4+v04CZQDmwATj+s3628dANX2eKitQHF1H/Oul1+/0ia9ZonEiILF2qf4vovydtwG3bHN+8ttbxun0+kbVrP65Pkp0tkrQEt2wRaWjQeOdOkeJijb1ekfXrNY7HP65ftUrEsne/srRGIpJrmfuReFxWJBMVRGR5e7tELd85x+u1fesyv9/2SlvCYcm39KF4XFb20C9rb5eY9WZs9Hql00pIKPH7bd+8KRyWzT6fiIgEYzFZ3UO/tL3d9n3Xe73SZemLu7tt37whFJKtlj4Qi0l2MlHC0id937WdneKz9IXd3VIXCv3Xx+xgZGVHh4St9zKvq0tarfeyMhCQcuu9bItEZFOPtbC8vd3WL29vt3MQNnV12TkQ5YGA7bu3hMOSt48TRfgyPXSXy/VH4Bygn4hcYf3bOcA9wP8nIn2sf7sLOF1E7nS5XNdbj123p59tPHSDwWAwfJ340jx0l8t1DP8/e2ceZ1Vd///XvTPDLMCALCpoLmma+5ZC+s20/Fb2zfKb/lLLtDK1b5uZZVmCWoEkZpoLpSKoiZiMCyKKgjDDDAyzMAPMvt3Z771z930757x/f7zf53MuKGgqBHpej8d9+Hbufd975pw393Pm83wvwP8AeDzvZwUAFgK4dZeXfwPAk2KvAPBFh2OXLve2bNmyZcuWrXfU3mbo94MXbiPvZz8FsJKI3Lu89jAAQwBARBqACICpe/n4bNnaqxobA6ZMAaS1N044gRk1wDPHTQZ+773Al77E9tq1zKrNBi9TpgCC5HDssdxrHeA68jlz2J4/n5k4wHXehx/O9sAA+5slxkccAaxaxfbXv24x8DvvtOrYX3rJYui9vexvNoSZOdPqxX7xxcCf/8z2737Hs9b3Z8U1DdNralQv9tmNjap/9/91dana40dHRzFbuOmWaBTTa2qQ0HWEczlMq65WAzrObGjAUmHg13V04CZh6A8ND+N84abV4TAOFobuz2YxtboaO4SBn1xXh2XSy/277e2Kgf91aAhflDr49aEQDhWG7slkMKW6WtWRH7dlC57PqyM3GfifBwdxsTD0NcEgZgpDH06nMaW6Wg1uObq2Fi9LL/f/bWnBXdJL/o/9/fiG1MG/4vfjyM2bAfAAlynV1RiUhiqHb9qE1dLN6Gvbt2O+9JKf43LhcqmD/0/olz09+J70wn/a48Fp9fUAgG3xOKZWVyMkTXQOrqlRPQXO3boVi6Qb00+7unCDMPTFbjfOllhoiEYxTRh6TNMwrboa9RJLZzc2YrHZk6CzEz8169hHRnCu5GPsE+1uL/6DPgB8DcAjYl8AYBWAmQCqARTKz/MZeguAw/P+vxfAtHd43xsANABoOOKIIz5UNmHL1octXSdatoxrs4mIVq0i8njYrq21uHdPD9GGDWyHQkQVFTv7myXCK1cSjY2xXVNjceuuLubTRESBANGLL7KdyxE984xVO/7SS/w8EdHGjVa9ekeHxd19PqKXX2Y7m2V/s3a8osKqfa+s5OMmYv6+efP7P0/7Ssu9XlVv/EYgQIPCnbdGo4p1DqZS9IacpLim0XNer/J/1uNRteev+f2KOzdEo4p796dStFZYazSXo3+Jv2EYtMzjUbXnr/r95Bb/LZGI4t59yaRiteFcjlbIBdfF3+S+r/h8ql56czisaqS7EwnF7YPZLL0g/pph0DMej+K+L/t8ihtXh8Oq3r0zkVDc3p/N0ks+HxFxTfszHo/KAXhxbExx46pQiLrEvz0ep0153H5fa3ssRnXSIGE0nabX/H4i4nyEZz0elUPwnNercgjeDARUDkNTNKpyKIZSKVojsZDQNFqeFwvLvV5Ve74mEFA5DI3RqMqhGEil6E3zH9yHJPwnGLrD4bgbwHcBaABKAJQDyMjD7Jl3BIA+IjrW4XCsAXAnEW12OByFADwAptMeDtBm6LZs2bJl6+Ok/whDJ6LbiOhwIjoKwJUA3iKig4joUCI6Sn6eJKJjxWUlgGvFvlxe//EqjLRly5YtW7bep/anOvTFAKY6HI4eAL8E8Nv/8PHYsvW+VFsLnHoqjxMNBJib588jN+u4f/ITqxf63//OvdkBoLoaOP10btri9bK/ycDPPdeq477hBqsX+oMPAtdcw/b69cBZZ7E9MsL+UmKLs88G1q1j+3vfAx54gO2//AX44Q/ZXrMGmD2b7YEB9vd4mKGfcQZQVcXPfec7wCOPsL1gAfB///eBT92HrpZ4HCfV1SGmaUjqOk6uq0OzMPCv5c0j/23eDOtnvF58TRh0UyyGk+vqkNR1RDUNJ9XVoVUY9pe2bVMM+5aeHtULfanbjf8VhlwfjeKU+npkDAOhXA4n1tWhU+rAL2xuxksyz/zn3d34ozDsx0ZH8S2Zob0pEsFp9fXQDAO+bBYn1NUpBv5fW7fiVWHYP+rsxJ9lCs7DIyO4WpI2KsNhnNHQACKCO5PBCXV1GBAGPquxEW/IRJ7rOjpwn0zhuX9oCN+Xeeprg0HFkIfSaZxQV4dR6UB0VkMD1ksv9Gva21Uv9IWDg4pBvxYI7BOGfHlLi6rpv9Plws1S01/h8+EiyUdoSyRwYl0dIpqGlK7jlPp6lQ/x9R07VF/83/f14VbJZ1ju9ap8hO0SS3FNQ1xiYX+O2FkAACAASURBVHvebPvlEku39vaqfIanPR58XfIR9oUK98WHENEGABve4ecT8uw0gP+3L47Hlq29qWOOAW68ked8T5rE9owZ/NwPfmANPrnkEn4eAD77WW66AvAAkxtvBJxO4KCD2D7kEH7uuuuAE09k+xvfsGaLn3ceJ7wB3NTlhhvYnjaNbXOG+fXXA5/+NNvf/KbVgOb8861mMCeeaC3uBx/M/lOm8Az1G28EjjuOn7v8cit57oILrGlr+5M+UVKCG2bOxPiCAjgA3DBzJo4sKQEAfOeQQ3DGBP4K+vKUKTBLas6aMEH9pXNkSQlunDkTpU4nDPE/vLgYAHDNIYfgdPH/6tSpGCdFOWeXl6OsoAAAcHRJCW6cMQPjHA4UFBTghhkzMFNmaH/v0ENxqvh/bepUTBCf2eXlmCLzs48tLcWNM2ei0OnE5MJC3DhjBg4V/x/MmIGTx48HAHxj2jTlc155OQ6T1xxXWoobZ8yAw+HAlKIi3DBjBg6W110/YwZOLCsDAPzvtGnqfT83eTI+WVoKADhh/HjcIME7Xfyniv8NM2fiePG/bPp0fELOy+cnT8aJclwnjR+P68zg34u64uCDcYIcy0UHHYSUdGo7bfx4XHvooQCAw4uLcePMmZhQUAAngBtmzMBREgvfPvhgnCbX4ktTpqiueWdOnKgyuo8Qf/Pa3jhzJo6Q3/m7hxyCMydOBAB8ZcoUFEgsfGbiRBTtw2Itu5e7LVu2bNmydYDI7uVuy5YtW7ZsfcRlL+i2bH3Iam7mmnJNA8Jh4MILmUUDvE1tMuzbbwcefpjtZ54BfvpTthsagK98hRl6IMDb2SYDv/RSoLKS7d/+FvjHP9heuhS4+Wa2N2+2+rp7vewveBBf+xqwaRPbt9wCPPEE2489BtwqrZ42buQadYDr4C+4APD7maFffDFQV8fP/fzn3LcdYJb++99/sPO2N9SVTOILzc1I6DrSuo4vNjejXRj499rbVR32goEBxaBf9vsVQ25LJPDF5makdR0JXccXmpvRLQz8O21tqg77T/39ikGvGBvD9cKQt8fj+O9t25A1DEQ1DRc2N6NPGPgVra2KYd/hcuFvcpGf9XrxY6lj3hqL4cvbtkEnQiiXw4XNzaoO/JstLYph/66vT9VRP+XxqJr4umgUF2/fDiKCL5vFBU1NioF/fccONc/71729eFxq8p9wu3GLMOiaSETlE3gyGVzQ1IQxmcLz1e3bVR33zT09qib/H6Oj+K0w6MpwGJfuA4Z8Y2en6kt//9CQqql/PRDAVZJP0COxENM0ZAwDFzU3q77413V04AXJZ1g4OKhq6l/x+3GN1LR3JpP4YnMzkrqOpMSS2RPgmvZ2vCKxNH9gAAslll7w+XCdxNK+0D5h6LZsfZw0dSowaxZQUACUlLBdXs7PnXMOIEgPJ53EbBoAjjzSmkQ2bRr7OJ1AWRknqAmee5u/iSePPtqafHbwwVZS2/jxbAvSxKxZFqs/5RT+XAD45CcBwYE45BB+HQBMmMD+ZWXM0GfNsrj9qacCRx3F9rHHWvkA+5MmFxZidnk5xjkccDocmF1ejoMK+WvvzIkTFQ8/vqxMMfTDi4sVW58i/kVOJwqIMLu8HJPF/zMTJyoe/umyMhQ7+e+jT5SUKP+pRUWYXV6OQocDJU4nZpeXY9I7+J9QVqZ+fmRJCaIyoWtaURFmlZfDCaDU6cSsiRNRLgz3nIkTFfc+afx4TBe2fVRJCXKCUqcXFWHWxIlwOBwoKyjA7PJyxepnlZfjEPE/efx4xcCPLilRf+kdIp8PAOPFf3we6z9Y/E8ZPx5HC3c/pqQEZXIuDh03DueYwb8XdfqECYpnf6qsDAk5fzOLi/EZ+cczubAQs8rLUeJ0qlgw8w7OmDBB/f7HlZUphn54cTHO2sXfzJWYVV6Og8T/rIkTcZj4f7qsTDH0T+TF0r6QzdBt2bJly5atA0Q2Q7dly5YtW7Y+4rIXdFu2PmR1dADf/S73Yo/FuMe54En8+MfMuAHgvvuAf/6T7ZUrgTvuYLulBbj2WmbWkQhwxRXMwgEuG5PW1LjnHmD5crZfeAH405/Ybm4Gvv99toNB9hfUi+uuA8zxzPPmAStWsP3cc1xLDjDDv/56tn0+9g+H+Xi+9z3ARKJ33cV93/dnDaTTuKqtDWldR9Yw8O22NsWwf93bi7XCsP8xOqpmm68NBlVf9d5UCt9ua0PWMJDSdVzV1qbquG/u6VHztB8aHla9vF8PBFRNe1cyiavb2qAZBhK6jitbWzEs/j/r7kaNMOj7h4bwpCQ6rPL7MUdq2tsSCVzT3g6DCFFNw5WtrfAIA/+/ri7Vl/7ewUFVU/+Sz4c/CEPeHo+rvuahXA5XtLbCJwz8+s5ONEod9t0DA4pBPz82phjy1lhMMWB/NosrWlsRzOUAAN/v6FA1/X/q71cMernXi3uEIddHo7hR8gn2pn7f16fyGZa43SofoSocVvkEQxILSV1HzjDwnbY29Ajn+k1vr8pneGx0VOUjvBUKqXwCl8RCxjCQkVhySSzd0tODtySfYdHICB6TWHojGMRvJJb2hWyGbsvWh6yCAqC0lJmz08m2YEeUlQGCSlFcDAiCRFERvw7g503b5Oimf2np7v2lpBaFheyT7y9Icyf/khLLf9y4d/d3OHbvv7+qAECZ0wmHwwGH2CbfLHU6USQnZpw8DwBFTidK5efKH7D85XWlTicK5b2KnU4Ui72Tv7Brh8MBBxHKCgp2+vx38y90OJTtBFBaUADne/AvyfeX4HHKsbyTf4nTqdjwuF38y3b1l9+/bJfPL8rzL36H49+bKs37/GKnE1lByUX5509sJ+Ra5l2Lkl2Ov+Ad/AvEduR95k6xlOdv2kX76Pc3ZTN0W7Zs2bJl6wCRzdBt2bJly5atj7jsBd2WrX9D8+YBZmvqp58GXnyR7Y0bgfvvZ7u/H/jVr7iMLJUCbrqJ67gB5s5S1oulS62+7uvXcz92gGeQmzXhiQT7m21V584FpM03Hn+cZ58DwJtvAosWsd3VBdx2G9uxGPsLasXvfseMH+D+8eZs89df51p0gGe33377BzlL+0b3DQ0pBl3h86nZ4g3RKO4WBuzJZHBzTw+yhgHNMPDLnh6MCIO+Z3AQteL/3NgYnhOGXBuJKAY8ksnglz090AwDWcPAzT09imHPHxhAg5zYZV4vKoQh10QiqiZ9MJ3GLT09MIiQ1nX8ortbMew/9vejSRj0Ux6PqomvDIfxgDBgVyqFX/f2goiQ1HXc1N2tGPYdLpearf6E241V4r8uFMLDwoC7k0nFcOOahpu6uxEW/9v7+lRN/qOjo3hdGPQbwaDKJ+hIJFRf8oj4xzQNAHBbXx+6hEEvGhlR+QirAwFV096aSKge96FcDjd1d6uSslt7e1Vf+geHh1VN/fvR/UNDqqb+RZ8P/5R8hK2xGOZJLHizWfyiuxsZw4BOhFt6elQ+w72Dg6qm/vmxMdWXvS4axQLxd0ss5QwDOYkFs6Z/wcAA6iQWnvV6VY//TZEI7pVY2heyGbotW/+Gens50QzgwSdmia3PZw1QicV4GIthANks2/K9hZ4ea3EeGuIBLqa/2XwmGrX8MxnL/6CD2JbvLQwNWWx9bMzyj0R4UQeAdJrtdJqPtbubnweAwUGuMwc46c783olErGEy+7Nc6TSOkWSD0UxG9e8OappaKBKGga5kEhoRnOAktbgsKH2pFE6SAn3zixkAApoGl3zRx3Ud3akUdAA6EbqSSSTkc/pSKQSlRnk4k1G11/5cTvnHTH8i5IjQlUohKf69qRRCsjgOZzJIynH5sln0i39U19GdTMIAkDUMdKdSSOo6phQVoSeVQlj8hzIZmPDUl82qxL2ofD4RISOfn5bP787zH0ynFff2ZrOqeU1E19El5zJtGMp/opzLiPgPpNMol+QKbzaLITmfYU1Dt/in5PizhoFSpxPdqRSief6HfICEDFc6rXr0u7NZ9b7BXE7FQlLOhUYEB4CuVAoxOeeudFr1pR/NZqEJig7mcujLjwWJJYBvlsybk750GqfJP+bRbFblFgTyYmFfyGbotmzZsmXL1gEim6HbsmXLli1bH3HZC7otW3uQpgF3321tcz/6qMWgV60C3nqL7aYmq6+5x8Nzyol4y33+fIthL1rE2+4A8PLLVl/2hgZg2TK2R0Z4PjnAW+7z5wOCSvHQQ4AgTbzwAs9OB4AtW4B//YvtwUGL56dS7G+2lX3gAWtr/vnneXY7ANTUABUVbLtcFs/fn0REuGdwEF5h0E95PNgmJ2ZtMKjqkNsTCcVwg7kcFgwMQCeCQYQ/Dw7CL/5PuN2ql/eaYBBrhKW0xONYIjXl/mwWfx4chEEEnQgLBgYUw35sdFQx6NWBgGLI2+JxPCUM15vN4p7BQRARcoaBuwcG1Db130dGFIN+xe/HBmExjbGYqil3ZzKKwWYNA/MHBhTDfnhkRG0nv+TzKYZcF40qBjycTiuen9Z1zB8YUNvEfxseRr/4rxgbwyZhMZsjEcWAB9JpxfOT4p8S//uHhjAk28nPjY2pmvjqcFjVpPelUnhI/OOahvkDA8jIlv9fhoZUPsMyr1flI1SGwyqf4L3qaY9H5SO8FQqpfIKORAKPSiyEcjncPTAAzTBgSCyZ+QxL3W412/zNYFDlE7QmEqq/QCAvlsxYCEgsLHa70Sqx8HoggDclFrbH46rH/b6QvaDbsrUHJZO80JmNXVautPj0xo28kALcbGXNGraHhthH03ghrqhgRg5wIxaTT1dWWk1itm3jxDaAF+SKCmbo0SjbZmOYF19kjg9wIl1jI9tNTcDatWz391sNY8Jhtk1uX1Fhsf5166wEv61brZuTvj6+WdjflCNChc+nFoHXgkG0yJdobTSKalmQOpJJvCInzJvNYoXPh5SuI20YWOHzYVS+xFcHAmiTBXVTJKIWtLZkEq+K/2g2iwqfDxnDQFLXscLnUzcUqwIBdIh/dSSCWlmQWhIJvCZf6COZDCp8PuSIkBB/c7jJykBALehVkQi2yIK0Ix5XNxeDmQwq/H5ohoGYrqPC54NfFpGX/H70yIK8PhxGvfhvi8fxplzwAfl8IkJEPt+8IXnB51N8+K1wGFvFvykexzrxd6VSKtkvpGlY4fMp7r7C51N8eF0opBbUxngc6+XmojeVwouyuAY0DRU+H6KaBkOupcnq3wyFsE2uZX00qhr2vFetCQaxQ/y3RKPYKNeyM5VSsTCWy6HC50NSEhxX5MXS6mBQLcib8mKpPZFQseDJZlHh9yNtGEgbBir8frjF/9VAQN3cVUci2Cyx0JoXC/tCNkO3ZcuWLVu2DhDZDN2WLVu2bNn6iMte0G3Z2oMMA1iyhOvBAd7yFiSJmhpry7u7G3jtNbZDIYun6zrPHDfL1ioqmJEDQFUV910HgM5Oa8s+EOD56ABv2y9ezCwdYE5uzjbfsMGqaW9rs7bcfT7g2WfZzuXYX3Z5sXw5l7gBvOVu1rTv2MFb+ADjheeee1+na6/rKY9HMejVgYDqxd0Qjaot84F0Gi/JNnFc07DE7QYRgYiw1O1WDPoVv1/14t4SjSoG7Eql1GzrqKYpBkpEWOJ2Kwb9Ut6W8aZIRDHgnmRS8fyIpimebhDhCbdblae94POpOujqvC3vrmRSMdxgLqdqqnXxT4v/82Njasu3MhxW+QTtiYRiuP5sVtXna4aBxW43ssKwnxsbU/jgrVBI5RO0JhJqy30sm1U8Piv+OfF/1utVDHptMIg2+UeyPR5X+QCeTEb1iM+Ivyb+z3i9ikGvCQbRKdeyORZD1b+55f5aIKDm1DfGYqo/wVA6jRclFhK6jiVuNwzZlX7S41Hlbav8ftXjvy4aVf0J+lMprJRYiEssvFMsrfT7VT5CbSSiatL7UinF8/eF7AXdlq09KBIB5syxEtEWLLAS0Z58khPLAObfDzzAdmsr++RyXLM+Z46ViDZvnpWItmSJxapffx14+GG2t29nH8PgxXnuXOsm4g9/AOrq2H7sMU6sA4BXX7UayzQ1sQ/Ai//cuYCZl3PHHRY3/8c/OLEP4NyARx9lu76eG+Dsb0rrOua6XGrh+OvwMNbJF/9zPh/+KQvPxnAY98gJ60mlMLe/HzFdR9IwMLe/Xy0c9w4NYYP4L/N61cK3PhzGX8S/M5nE3P5+JHQdUV3H3P5+xa3vGRpSC8/TXi+ek4VjXTiM+yURrE0aq2QMA2FNw1yXS3HnuwcHsUm++Jd6PIpVrwkG8aDc9bUkEpjT3w/NMODP5TDH5cKgLOJ/GhhQ3H6x260WrteCQTwiiWDN8TjmuFwgInhzOcx1uRQ3vrO/Hw1yE/Ho6ChWyk3EK36/aizTGIvhDhn04s5mMdflgkcW8bn9/WiSm4BFo6OKNb/s9+MxCbi6WEwNihlKpzHX5YIvl4NOhDkul0pEe3hkBK/LTUiF348l5l3re9QDw8Mqb+D5sTF1E1UdieDPklTYl0phjsuFqKYhJbHULrFw3/CwGq6yfGxMxVJlJIKFEgtdqRTm9PcjruuI6zrm9PerGvuFQ0OolJuAf3q9WC43MW+FQrhPYmFfyGbotmzZsmXL1gEim6HbsmXLli1bH3HZC7otW++i1autFq0bN1olZK2tVgna6Ki1FZ5MWj3SAd4OF9SGqiqrhGzHDqsEbXiYa9EB5vVmCRsRb4sLNsWGDVZNfHOzVYI2OGhtpcdizMcB3rZftYr/C3BpmlkTv3Wr1e61v9/i+fuz1gSDqg56SzSqGHJXMqm24n3ZLKrlJGUNQ/FsgFmrWQe9KRJRJWQdiQQ6xN+bzaq+3mldx2t5/qsDAcWgq8NhVdPelkioEjR3JqN4fErXVQkawOVNJoOuCodVCVlLPK7yAUYyGdSLf0LXFQ8nIrwaCCgGvSEUQkj8t8fjigEPpdNqznlc01R9PBFhld8PXXZl14dCKh+hKRZT7WIH0mlVghbVNLUVbYi/yaDXhUKKIW+NxVQ+gSuVUjw/nMspnq6Lv7kr/GYwqPIRGqJRlU/Qm0qpHvXvVVuiUdW+tzuZVCVo/rxYyO0SC68HAiofYXMkonr0dyaTqgRtLJtVuRkZw9gpFvJjqSYSUfkE7YmEwjqeTEbx+H0iE/AfiI+zzjqLbNnam/L7iUpLiRob+f+PO45o8WK2r7yS6Cc/Yfuee4g++1m2164lmjiRKJMh8niISkqItm/n5446iujpp9m+7DKim29me948ovPPZ/u114gmTSLSNKKhIfZvb+fnDjuMaPlyti+5hOjWW9m+4w6iiy5i++WXiaZOZdvlYv+eHv7/6dOJXniB7S9/mej229m+7Tair371g56tvauUptGEqipaHwwSEdHZDQ103+AgERHd2NFBV7e1ERHRP0ZG6MQtW4iIqC4SobLKSgplsxTL5Wh8ZSXVhMNERHRaXR09NDxMRETfb2+nH8hJfnBoiE6vrycioupwmMZXVlIsl6NQNktllZVUF4kQEdEJW7bQoyMjRER0dVsb/aizk4iI7hscpHMaGoiIaH0wSBOqqiilaeTLZKi0spKaolEiIjq2tpaWjI4SEdG3WlroZ11dRES0YGCAzpOAWxMIUHlVFeV0nUbTaSqprKTWeJyIiI7YtIme8XiIiOjSHTvolu5uIiL6g8tFFzY1ERHRKr+fDtq4kQzDoIFUikoqK6krkSAiokNrauh5r5eIiL66bRvd1ttLRES39/XRl5ubiYjohbExml5dTUREPckklVRWkiuZJCKiqRs30ss+HxERXdTcTHf09RER0a09PXSJBPxyr5cOq6khIqL2eJxKKitpKJUizTBoUlUVveb3ExHR+Vu30rz+fiIi+kV3N122Y8e7B0SeZjc20sKBASIi+nFnJ13V2kpERI+PjtLxtbVERNQYjVJpZSUFsllKSCxVhUJERHRmfT09MDREREQ/7OigayWWFg0P0yl1dUREVCuxFMnlKJLLUVllJW2WWDqlro4ekVi6tq2NftjRQUREDwwN0ZkSSx+WADTQbtZEm6HbsvUu0jRA5k7sZBsG4HDwA+C/os1hKbvzybd1HXA6Pxx/Ij6eD8t/f5VmGCiUISL5tvlXo1NO5u5etz/7OwA4xF8nQsG/8V66DJ/5MPyJCAawz/3N1+/q/16U/15EBPqA1wJ7yf/D0J4Yur2g27Jly5YtWweI7KQ4W7beQSMjVjlXNGq1dNV1Lv0y1dhoMej2dqsmfWjIqgkPhy2ermk78+iGBv4LGGDubvZVHxy0asJDIYun53LcCjbf31RLC49CBZh7my1lAwGrtC6TserTd/XfF0rpumKYRKTqswGuMTYZck8yqWZze7NZ1Rc8oeuKhxOR4sEAs16TIeeP73RnMorBxjVNMVCDSNV3A8x6TYbcmUwqBjyayew0QvVAVv7I0KxhKJ4NYKdrsSMeVwzZlUqpfIBALqfq8zOGsRPPzvffHo8rhtyXSqmacl82q2qy07qu6tt39d8Wj6tY6E2lVD7AWN741uS7xJIZC915seTJZHaKpfxY2DWWzFjoSiZVTfpoJqNK+2KapnIr3imWzL/GOxIJFUsjmYzK7Yhqmsqt2Cfa3V78gfCwGbqtD6KrriK69lq2Fy4kOuUUttetIxo3jiiRYIbudBIJkqWjjyZatIjtyy4juv56tufNIzLD8bXXmFtnMkRuN/sL0qTDDrMY/CWXWAz+jjssBv/yy0RlZczQBweJHA4iQYI0fbrF4L/8ZYvB33Yb0ec/z/bzzzPDJyLq7WV/wbP7REtGR2mGcNNtsRg51q+nkXSacrpOpZWVtEq46TkNDfQHl4uIiH7U2UmXCjd9dGSEjti0iYiIGqJRcq5fT2OZDKU0jcZt2EBvBAJERHR6fT0tEG76g/Z2+lZLCxERPTQ8TMcKN90UDlPB+vWKoRdu2EAbhJueuGWLYvBXt7UpBn+g6/a+PvqvrVuJiBn4+MpKMgyDXMkkOdavp3Zh8FM2bqRlwuAvam6mX0mixa09PfQFCdjlXi9N3riRiIi6EglyrF9PvcLQJ1ZVKQb/+a1bFYO/ubtbMfin3W7F4FvjcXKsX0+DwtDLKisVg/9sY6Ni8D/p7FQMfvHoqGLwTRIL7nSaMrpOxRs2KAZ/Vn29YvDXd3QoBr9oeJiO3ryZiIi2RCLkXL+e/MLQizZsoHWSj3FKXZ1i8Ne2tSkG/8DQkGLwG0MhKtywQTH0wg0baKPE0vG1tYrBX9Xaqhj8woEBxeA/LMFm6LZsvV2JBPPjsjJrkMrkyfyczwdMn/52OxgEJk1i1hyP839LS/mv6kTivflPnszsOh5nnl1Swp3cUil+bwDw+4Fp097uHwgAU6bwccdiwLhxQHEx+6fTQHk57wYEAu/svy9kECGkaZhaVMSfn81i+rhxb7MjmoYypxNFTidSug6dCBMKC2EQIaxpmPIu/uFcDhMKClAo/gaA8QUF0IkQ1TQc9C7+oVwOE8Xf7N5Wtr8nEbwHZQ0DKcPAJEmW8GezmPYOv38gl8OUwkI4HA5ENQ3FTieKnU5kDAMZw0B5YSGICIFc7h39/dksphYVKf8SpxPjxD9rGJgo/sHdxEL+ceXHQlrXob2PWBhfUKBiKT8WIv9mLCV1HYR/P5bKCwtR4HAgoetwgGNJMwzEdR2Txf/DkM3QbdmyZcuWrY+AbIZuy5bI67X6mkciVk12JmPxbMBqtbqr7fFYNenhMP+VDPBfxybPJuK68nfyd7utmvRQyJpznkrxX+Xv5j86atWkB4MWz08mrfp4w7D6xe/qv7cU1zTFQDXDUAwRgOKhAM/nNv+I8GWziuHuyX8ozz/fHstmFcONapri6TnDUDXFe/L3ZrOqpjyqaYqhHuhK6bri4QaR4sHAzr//aCajGHIgl1O7FEldVzx8V//8azmSySiG7M9mVX+AhK6r+nqdaKfchKHdxIL/fcTCrrFkxkJM0xRPf6+xsLtYyhqG6ne/J//8WIrkxVLGMFSvg30he0G39bHS+ecDf/872z/6EfDzn7P9t78BF13E9rp1wDHHWIvkkUdaTWM++1kedgIA110H3HIL23/5C3DxxWy//jpw7LF84+DxsL+ZJPeZzwBPPcX2NdcAv/0t2wsWAJdcwvbKlcBxx/HCPTQEHHGENUTltNOAZcvYvuoq4Pbb2f7jH4FvfpPtigrghBPY7u1lfzNhb2/pNpcLV7e3A+C+5mfK1Jpt8TiOrK3FaCYDzTDwqbo6rJZGJ/+zY4fqk/2r3l78oLMTAPCEx4PZ0iWnIRrFkbW1avE/ZssW1WjlS9u2qZ7pv+jpwQ3i/4/RUXxOTvimSARH1dYinMshrmn45JYtatb2hc3NeFjufH7c1YWf7O2TtI80b2AAl7a0AOCZ6cdv2QIiQn8qhSNqa1WS18n19XhO7mKvaG3FXOlSdGd/P/6fBNy/xsZwkgR/VzKJI2trVQObT9fV4QW5i72stRV/lIEFt7tc+HZbGwDukX+67KK2JhI4orYWQ+k0dCIcV1enZpVf0tKCBdLl6Dd9fbi2owMA8KTXi7MllppiMRxZWwtPJoOsYeDYLVtU056vbN+u+u//srcXP5RYeNztxrmS4bolGsVRtbXq5uWYLVtU05wvbtuGByWWftbdjf+TDNlFo6P4vPhvDIdxdG2tuvk7urYWGyWWzm9qwiLpf/+jri78TGLpb8PD+GJ+huve1u7g+oHwsJPibP276uvjZDciIq+XSHJyKB7nJixERLpuNXIh4oQ0w2C7t5colWLb4+GkOSKiWIxIcmpI097ub6qnhyidZnt0lEjyuyga5QQ401/6UrzNv7ubk+2IiEZGiCQnh8JhbkJDRJTLEUmPEyIiklyxvapgNkuj8ouldZ26zZNMpBqhEBF1JBKkyckcSKUomssREVEgmyW3+Kc0jXok8WpX//Z4nHTx70+lKK5pRETky2TIIycmpYxKpwAAIABJREFUqWnUJ/6GYVBbnn9bPE6G+LuSSUqI/1gmQ2PmiT3AFcnlaEiCNKfr1Jl3LVryzkVXIkFZXSciouF0mkLZLBERhXM5GpZrkdV11YhmV//ORIJy4j+USlFYrmUom6UR8c+8x1gY3CUW8mNpT7GQH0sx8fdnsyoWUpqmkvgMw9htLLmSyZ1iySv+ifcYS315seTNZMgn/nFNU414PizBToqzZcuWLVu2DnzZDN2WLVEqZdWE53IWDyeyZpYDVq34rna+fzZr8XDDsOrD9+Sfb39Q/0zG4um6bs1M35PP3pImmc3qM80D24Od1nXFYD+of84wVE0zESmeuyf/lK4rBpvvf6DLIFI8Gnhv5zJjGIqn60SKJ+8r/w8zFvL9P2gsfNj+e1v2gm7rY6WzzgIeeYTt73+fOToA/PWvwLnnsr1hAzB1qsXQDzrIas5yyinA44+zffXVwM9+xvY99wCf/zzbb7zBZWImQ5882Wr08ulP8xx1ALjiCovBz5sH/Pd/s/3qq8Chh/IiPTzMpWyCp3HMMcCzz7L9zW9aDP7OO4GvfpXtF18EDjuM7b4+9u/p+aBnbs/6dV+f4q5Pezw4XrhrSzyOydXViqEfXFOD14WbXrhtm+KmN/X0KO662O3GyfX1AJibTq6uVgx9Wk2N4p7nNTXhPuGmP+7uVtz176OjOEMu2JZoFAdVVyuGPqW6Wg3rOGfrVjV3/IednbheuOuBrj/09+MrEnAr/X7M2LQJRISBdBqTqqtVo5OjamvxL2Ho39ixA7+XzkRzXC5csmMHAGDF2BiO2LwZADcCmlRdrZrOHLZ5s5rBfvH27bhT5p7/prcX3xSGv8zrxTFbtgDgoSWTqqsxLAz9kE2b1Az1i7Ztwzxh8L/s7cUVEgtL3W6cILG0TWLJTECbXlOj8inOb25W+Rg/7e7Gd+UfzGNuN06VWGiQWDAZ+tSaGpVP8dmtW1U+xo1dXSqf45HRUXxGGP7mSARTqqsVQ59SXa0Gt3ymsVHNoP9BZyd+JAz+/uFhxfD3iXa3F38gPGyGbuvfVUODxa17epipEzFLl14clMkQVVZaPm+9xVyaiKi+3uLWXV1E0suCxsaIpJcGpdNEVVWW/7p1zMWJuEGNzPagjg6Lm3s81gCXZJJIenkof0GVtHkz83oiorY2IpkHQaOjFitPJIikFwcZBvubOQB7S4OplGpYEsnlqFZ+Sd0wVPMOIqKqUIhScjK2xWKKVQ6kUor1hnM5NQBFMwx6K8+/MhSitJyMpmhUsUpXMqlYbTCbpQYZgJLTdTXMhYhoQyikuHFjNEp+4ca9yaRirQe63Ok07ZAgSWoaVcsAESKidcGg4r6bwmHFnVvjccW9R9JpxZrjmkabxN+Qa2n614TDihvviMUU9x5KpRRrjuVyaoDJrrGwMRSipPhvj8UU9x5MpahDrmUkl6Mtu4mFqrxYaI7FVA5EfyqluH8om6V68c/p+ttiKSP+W/NioS+ZVNw+kBdL2V1iaX0wqGKpIRqlgPj3JJOKu/syGdoq/h+WYDN0W7Zs2bJl68CXzdBt2bJly5atj7jsBd3WR0KLFwNnn812fT23PY3FuHHMtGkWAz/7bKuO/IYbgJ/+lO1Fi4DzzmN70ybg4IM5AS4YZJ5ulpKedhrw9NNsf+97wC9/yfYDDwAXXsh2ZSVwyCHM0MfGuFWrIEGccALw3HNsf/vbFgO/917gS19ie+1aYMYMZuijo+xvDo459ljghRfYvvxyYM4ctufPB772NbZXrwYOP/wDnU6l3lQKU4R7EhFmbtqEN4RbXrx9O/4sDPx3fX24Uhj682NjOF64aUcigSnV1fAIQz+0pgbrzdrf5mb8Vbjnr3t7Ffdc5vXiZOGmO+JxTK2uVo1HDq6pUQz8c01NeEi4503d3bhOGPpStxtnygXfGothWnU1IpqGuKZhek0Ntkg3odmNjXhUuOdHVWuCQcwUhj6cTmNKdbUa3HJ0bS1elm5G/9vSgruEgf+xvx/fEIb+it+PI4WhuyQWzOYyh2/ahNXCwL+2fTvmCwOf43LhcmHoL/h8OLa2FgDXsU+RfAqdCDM2bcLavJ4C90os/ba3V+VTLPd6FUNvk1gaE4Z+SB4Dv6CpCX+TWPhlTw++Z/ZE8HhwmuRjbJNYCglDP7imBpuFgZ+7dSsWST7FT7u6VE+DxW63qoNviEYxTRh6TNMwrboa9RJLZzc2YrFMerqhsxM/Nf/B7mvtbi/+QHjYDN2WqaEhojVr2E4kiJYvt55bvtyqPV+zxuLOjY0W9x4YIHrzTbZjMaLnnmPbMIiefdaqPV+9mnk1EVFdncW9XS5m7UTMyJ9/nm1dJ1q2zKo9X7WKeTkRUW2txb17eog2bGA7FCKqqNjZX/AcrVzJvJ6IOblZ797VZXH7QIDoxRff86nbo7K6Ts94PKret2JsTNUrV4ZCijW2xuOKlXozGXpFCvwzuk7LPB5V77tibEzVK68PBhVr3BGLKVbqTqfpVSnwT2kaLfN4FLf9l9er6pXXBoPULxemORZTrHMknabXJVEiqWn0rHnCiYeNmPXGbwQCNGhe2I+ogtksvSABoxkGPePxKO77ss+nuHF1OKy4dWcioYaO+LNZekmuZW6XWHhxbExx46pQSHHr9nhccfexTIZWin92l1ioyIuFDXmx1BKPqxwMTyajhvmkxd+MhefzYmFdMKjqvbfHYioHYzSdVgNcUhILpv9zXq/KIXgzEKABiYWmaJQaJZaGUilaI7GU0DRaLsNoiDiWzByCNYGAqv1vjEap6UPm5vmCzdBt2bJly5atA182Q7dly5YtW7Y+4rIXdFsHrH7/e+DWW9levtzqpb59O3DSSTz4JB5n26wDv/hifi3Avr//PdtPPQV84xtsNzZyvXk6zQNcTjzRqgO/6CLulQ4Av/gFcNddbC9ezEwbAGprgVNP5aY1gQBzc7NN+PnnA6+8wvZPfsLsG+D+8lddxXZ1NXD66dxsxutlf2mzjXPP5V7xe1NrgkHMFm44kE7jhLo6eDIZEBHOaGhAlXDL77S14RHhjgsGBlT/61V+Pz4ntbe9qRROqKuDL5uFZhg4rb5e1e5+q7UVjwnD/mN/P34uJ+klnw8XSi/2zmQSJ9bVIZTLIWMYOKW+XnHL/21pwVLhlnNdLtwixfb/GhvDl/dl/+yPqc5qaFD5ENe0t6te6AsHBxWDfi0QwLnSl98lseDNZmEQ4fT6epUPcVVbG/4usTR/YAA/kVha6ferXurdySROqKtDIJdDzjBwan29yoe4vKUFT0gs3Oly4WaJhQqfDxdJLLUlEjixrg4RTUNK13FKfT22ynSlr+/Ygac9HgDA7/v6cGtvLwBm+BfLl8f2eBwn1dUhvh8P8Sm48847/9PH8L716KOP3nnDDTf8pw/D1n9IhYXcaOXoo3ku+PTpvBAXFwNFRZzkVlTEC+P55/PccYeDB6RMncqzzD/1KeCooyz/k09m/3HjePHM9y8uZv+zz+ZEtYICHqJyxBH8+kMO4ZuH4mKekT57Nv/cMLjpzLhx7D9rFjebKSjgRjOf+AR/zowZvHgXFwPjxwPnnMM+ROxvjlSePduam743VOxwYFJhIc6cOBHFDgcIwOcnT0ah0wkQ4bxJkzChsBBOAKeOH49Di4tR6HDgyJISfKqsDOMcDhxUWIgzJk5EsdMJB4DzJ09GkdMJAvBfkyZhfEEBnA4HTpswAYeMG4cihwNHlZTgmNJSjHM6MaWoCKdPmIBihwNOAJ+bNAlFDgcMInxu0iSUFRTACeCMiRMxfdw4FDoc+GRpKT5ZWopxDgemFhXh1AkT9t5JsgUAOHfSJEwsLITT4cDJ48djRnExihwOfKKkBMeVlaHY6bRiKS8WCs1YmDwZE+RanjphAg6Va3lkSQmOzYul0yUWnQDOnzQJRU4nDFix5ABw+oQJOFj8jzZjSWLhtAkTUOx0osDhwH9JLOlEOH/yZJSK/5l5sXRMaSmOllicXlSEUyQWi5xOnDdpEhwOx3/snN91113uO++889F3es5m6LZs2bJly9YBIpuh27Jly5YtWx9x2Qu6rQNWCxdaDPqVV3i+OAB0dABf/CL3Yk8m2TbbdF9zjcWw58/n9wC4tvu669huaWFWnslwLfsXvsBzxQHm3CbDvusu4P772X7uOasvfHMz15RrGhAOc326lOji8st53jrAs8wffpjtZ56xauIbGoCvfIW36gMB4IILuKc7AFx6Kde5701tDIfxdalDHs1kcEFTE/zZLIgIF2/fjjrhlj/v7lbc8ZGREdUL/K1QCJdJHfJgOo0Lm5sRyuWgE+HL27Ypbvnjri486/UC4LnRd0iiwBvBIK6Qmva+VAoXNjcjqmnIGgb+e9s2bI/HAQDXd3ZihfQi/8vQEP4kddSvBgK42iz8t7XX9NXt21Ud9809PSqf4R+jo/it/IOpDIdxqcTScDqNC5qaEMjlYBDhK9u2oUFi6addXXhGYuHhkRHMkVhYGwyqGQH9EgvhXA6aYeBL27ahWWLpxs5O1Zf+/qEhVVP/eiCAqyQWepJJfKG5GTFNQ8YwcFFzM1oklq7r6FCz3RcODqqa+lf8flwjCTSdySS+2Ny801CW/U2F/+kDsGXr/eq446xpY4cdxoNXAB6mMmsW82fAYtYAcOaZVtOVT3+aOTzAHPuMM9ieMoU5dVGRxbxNZv2ZzwAzZ7J9wgnMugHm6KefzvbUqexTUMDcftYsoLycnzvnHB68AjBvnzKF7SOPtKaiTZvGPk4nUFbGxzJx4tv995YOGTcOs+SAJxQUYHZ5OcoKCuBwODBr4kRMF5h/6vjxOKqkBABwbGkpJsnJPHTcOJwj/hMLCjBr4kSUOp1wAphVXo5p4n/ahAk4Uvw/VVqKQ+SCzRg3DmfLLzypsBCzy8tR4nSi0OHA7PJyTBX/MyZMwCfE/7jSUjXh67Bx43CWecJs7TXNLi/HwXLNThk/HkeXlgIAjikpQZmT/1bcKRbkWpY5nXA6HG+PheJi5T9FYmlGcbGKhXLxLy0oQMEu/qdPmIAjxP9TZWVIyBfDzOJifEb8JxcWYpbEklNiaUp+LIn/cWVlanLc4cXFKpZM/3H/QX7+brIZui1btmzZsnWAyGbotmzZsmXL1kdc9oJu64DVY49xD3YAeOsta7a4y8V90jMZfnz721Yd9y238GsBnov+2GNsv/EG8JvfsN3TA3znO1xHnkwyN5eW47jpJmDjRrb/9jdgyRK2V6+2ato7OoDvfpdxQCwGXHkl92QHgB//GJDW2LjvPuCf/2R75UrgjjvYbmkBrr2Wy9UiEZ6bLngRN97Iver3phqiUTUb3JfN4orWVoRzORARvtfejh3CHe/q78dLwh2f8XpVL+7aSETVpLszGVzZ2oqopsEgwjXt7WhLJABwz+9V0kv8SY8H98tJrg6H8TOpSR9Op3FlaysSug7NMHB1W5ua531bX5+arb7Y7VZ93TeEQqoO2dbe0/c7OhTD/lN/v2LQy71e3COxUB+N4kaJJa/EUkTTQES4tr1dMew7XC6slFh42uNRc+43RyKqJn1UYimuadCJ8N32dnRILP2+r0/1lV/idqu+7lXhMG6SWBpKp3FVWxuSuo6cYeA7bW3okVj6TW+vmlHw2Oio6uv+Viik+hu4Uil8u61NoZ39UTZDt3XAatw45tQA825BeCgoYNtEXaWlFisvLbXquc169XfyLytjf4eDf25+Tv57FRfzY0+f73Tu7F9WtrO/yfnz/QsLLdvk6O/0+XtLhQ6HYqBOh4Nrvh0OOBwOlBYUoFBObInTiXHyunEOB0rELnQ4UCp2gfiYfzmUCgs3/YvkdcUOB4rFLnI6d/I3a84dYheIf2me/ziHA/QO/rb2nsryrmWx04kiscc5nepa7hQLgLqWwM6xUJrnX7w7f7n+jjyf3flnBSUX5fk7xXYCcMixFOTHYt7xF7yDf4HY+y9Btxm6LVu2bNmydcDIZui2bNmyZcvWR1z2gm5rv5PXy33SMxnm0LfcYtVh33uvxaCff97qy15XByxYwPboKHDzzczAczm2pUQWCxbwawHg2Wf5PQCegX7vvWwPDfFn6jofwy9+wXPNAWDePEBaU+Ppp4EXX2R740arJr2/H/jVr7iOPJVi7i54EHfdZfWVX7rUqolfvx548EG2e3utHvWJBPtLy+zd6s1gUHG/rmQSt0lNeEzTcFN3N6LSf/p3fX2KO/59ZERxw9cDAdVXvS2RwO3iH87lcFN3t+pf/ZveXsUdHx4ZwVtyYKv8fiyRk7wjHsedkrQQEP+kroOI8OveXrhkHvcDw8NqnvXLfj+ekpr2plgMf5Q6Yl82i190dyOt6zCIcEtPj5rHfd/QEGqkDrrC58MyM9FgN9I0nl8vpwn33MN99wHuI2DOqa+t5ecAfu0vf8m+2SzHkhwm5s/nngEAsGyZ1eO/pobzIwBgcJBjyTB4NsAvfgEIasYf/whIm3I89RTw8stsV1YCDzywx19lv9OikRE123x1IIDHJZZaEwnMlVgISSyYJWW39vaq2ewPDg+rvvAr/X5V074tHlc15f5sFjd1dyMlsfCrnh70i//9Q0PYKLH0os+Hf8pF2hqLYZ7UlHslljKGAV1iaVhi6d7BQVVT//zYGJZLLNVFo1gg/u5MBjf39CBnM3Rbtt67kkkeZqJpzKG7uji5DODktuOPZ3t0lF8DcAMWWYOQSFj+ANuSe4O+PqtefGTEYuiBgJU4F4/zZ+o63xB0d/N7ArzYyvcWRkas+nKfz/KPxdjHMHgR6O7mhR3ghDtzcR4a4vc3/c3mM9Go5Z/JWP4HHbT7czaWy2FAvpwimqYSx9KGga5UCmnDQDmA7lQKEflCHcxkMEHgvDeXw2Amo/y75YBN/wwRxhOhO5VCVPz702lMFaDvyWYxms0CAEKahh7xT+k6ulMpZA0DxU4nupPJnfwPkyQCdyaDoFywkKapL/qkfH6OCAXy+THxd6XTOEaSDUYzGaTe5YtW0/i65sfCSSexbSYtAm+Phe5ujgVdZ38zFvr6rFgYHuZcB4Bv3naNBTOWurqsfgO9vVYsDA9bP/f5+KbwQNJAOo1yiQVvNoshiaVwXiylDEPFQqnTybEk13wgnVZ9CDzZLLxmLOVyViyJf44IRe8QC2ZPA3c2q943mMtZsSSxqBHBAaBrF//j5QKOZrPQBEUHczn0yb+ruK6jO5mERoSivXESPwTZDN2WLVu2bNk6QGQzdFu2bNmyZesjLntBt7VfqKMDeFQGAoZCwN138xapYTDPNLnj0qUWg37zTauvemsrzyQHeMt0wQJrm3TBAv4ZwK+R1tB47TV+D4Dfc+lStn0+/kwiPoa77+ae7AAfY0cH26tWWTXtTU3M1AFmrAsXsn82y6xVWlZj0SLedgeYmX6Qvuwv+HxqnvSWaFT1sh5Mp1VNd0rXMX9gQPWffmB4WG3NPz82hlrhhjWRCCrkJLtSKTXbOiH+afG/b2hIccflXq+aTV4VDqua9J5kUs1Jj2ka5g8MICvb4fcODsIt27HPeL1oFJayPhTCK5Jo0JVMqtnYEU3D3QMD75tbBoNWLBgG8Oc/W/kMTzzBNf8AsGYNPwD+mdlfwO9nH8OwYsncZn/sMUDafGP1amDtWra3bWMmDnA+iBlLuRzHkpxy/P3vvAUPcC7Fhg1sNzZyb3+Acz/M3I6Po7qTSZUbEpVYyhkGiAgLBwfhkVh62uNBk8TSW6GQ6m/QkUjgUeEpoVwOdw8MQDMMGES4Z3AQPtnaX+p2qxkBbwaDqr9BayKBxcLzA7kcFgwMqLaw+6PsBd3WfqHOTitBzOvlBKNkkhfEFSusRKbVq60FedMmoLqa7fZ24NVX2Xa72T+d5kdFhZXItGqVtSDX1FgJdq2tvMAD/FkrVjC/TibZ38y3WrnS+hLeuBHYsoXtHTusBWFoiH00jRlsRYV1Q/LSS8xUAV7MP0iTmPXhMBrlS6gpFsNaAbL96TRWyAeGNQ0rfD6EhClW+HwqKW1dKISt4r81FlMJbn3pNF6QL8RgLocVPh8iktRW4fNhQL5E3wyF0Cz+DbEYNsjNRU8qhZfE35/LocLnQ0waw1T4/YrVrwkGVZOaulgMVbLSdSWTWClfqGPZLFb4fCqR6t+V18vXMpXiWFixwuLlq1cD5gyXTZv4AfDPzFgaHeXrZ8bCihVWLOTHUnW1lWDX0rJzLFVU8GKeSLC/mWCZH0tVVe8cS4ODVix9HNWdSuFliSVffixJLA7nx5IkN2yJRrFRYqkzlcIrZiyJf9IwkDUMrPD5MCL+q4NBtIr/pmgU1eLfnkjgVfH3ZLOo8PuR3o+T4myGbsuWLVu2bB0gshm6LVu2bNmy9RGXvaDb2i80OGjVdCcSzDDNna0nn7QY9KpVVnlaXZ21zdnfz1uYAJcKLV3K3JKIbbNUaeVKqySottaqSe/r4/cGmHE++STbhsHHYpYqvfii1de9poZ5J8Db6OY2ayhk8XRdZ1Zrlq1VVFj4oKqKZ6e/X20IhRT3a0skVB2wL5tVc8ZzhoHFbrdi2Mu9XowJN1wXCqltxh3xuKoD9mazeE72hTPir4n/Mq8XAam1eyMYVDXt2+JxVVM+msmoOeVpXccTbrfijv/0eBAU/9cDAVVetzUWU/kAw+m06gueFH/jfe4kxuN8/fJjwSyBfOUVq7xsyxZry9vlsvBPNGrlVhDtHAsvvcRxC/B2vblZ2NPD2/kAx5LJ0w2DY8EsT3vhBau/QnW11d+gq8vKDQkGrX7/H0eNZDIqtyO1Syw97fEgLLH0WiCAbjmxjbGY6k8wlE7jRfFP6DqW5MXSkx6PKm9b5fejT/6R1kWjKrekP5VSPebjmoalbjf2511te0G3tV+oupqTjwBeXOfM4S/TVAqYO9dKPrrvPisRbfly68uuspIT0QBeXOfM4S/zeJxtk1UuXMgLKcC+ZmOadeuAv/6V7Y4O/sxUir+Q58yxbiIWLLC4/ZNPWo1p3nzTagbS2so+uRx/Ic+ZY9WYz5tn3YQsWcJf6u9Xj7ndii++GghgkcDhpngcc+WuxZPNYq7LBbcs4nf092OrrGj/GB3FKuGDKwMBPCrJP/XRqGrmMZLJYK7LhTEZzjLH5VLc/JGREayWm4gXfT48If5bolH8SX7hwUwGc1wuBHI5aIaBOf39aJEV8cG8xjYrfD4slUSHmmgUd8tK6UqnMdflQvh9QuSeHr6WsRgvpHPncr4GwMlmZiLasmX8ALjJz1/+wnZnJ/skEhyPc+daSY333GPF0tNPW41p1q2zmgy1tbFPJsOJlXPnWjcRd99tcfulS63GNGvWWE2GWlo4fj6uDH1zJKIawwxILIRyOWQNA3NcLhVLDwwP4025IX1+bEw1KaqORPBniaW+VApzXC5ENQ0pXcdclwvtchNw3/CwyiFZPjaGf8oNcWUkgoVyB9+VSmFOfz/i7zOfY1/IZui2bNmyZcvWASKboduyZcuWLVsfcdkLuq39Qn6/tZWdy1kMEmCeKKXP2LzZKkHr7LS24sfGrO3LTMbi2QDbUp2CmhqrhKy93dp+9XisrfB02mKYAB+L2aJ140arpr211SpBGx21eHwyyfPVTb36qrVlWlVltfvcsYPbf75fNcdiqgRtMJ1WW+kxTcM6+RCDCKv8fsUN3wqFFDfcGoupvuj9qZSabR3RNMXTdfE3d/LWBoOqr3tDNKpq0vtSKcXzQ7mc4umaYaiyH4C5u1mCVh+NqrKhnmRSzcYO5HKqL3duF/9/V9nszrGUHwubNlklZB0dVgma12uVM6bTO8fS6tX8ngDHq1nT3tZmYR232+LxqZRVggZwLJixVFVl1bS3tFhb+SMjVjljImH1Svg4KpjLoWoPsWT2V9gSjWJULmx3MqlyQ/zZrMrNyBmGmpkOcA6H2V9hcySiato7k0m0i/9YNotNZuOAA0FEdMA+zjrrLLL10dDjjxMdfzzbDQ1EpaVEgQBRIkE0YQJRVRU/d+aZRA88wPYPf0h07bVsP/II0SmnsL15M1FZGVEkwo+yMqLaWn7u5JOJFi1i+5priK6/nu377ycyw6mqij8zkSDy+/lYGhv5ueOOI1q8mO0rryT6yU/Yvuceos9+lu21a4kmTiTKZIg8HqKSEqLt2/m5o44ievppti+7jOjmm9//Obtk+3a6taeHiIju6Ouji5qbiYjoZZ+Ppm7cSERErmSSSiorqSeZJCKi6dXV9MLYGBERfbm5mW7v6yMiott6e+mr27YREdHzXi8dWlNDRESdiQSVVFbSQCpFhmHQ5I0baZXfT0REFzY10R9cLiIiuqW7my7dsYOIiJ7xeOiITZuIiKglHqeSykoaTacpp+tUXlVFbwQCRER0XmMjLRgYICKin3V10bdaWoiIaMnoKB0rF6wpGqXSykryZTLv6xzV1fH1D4WIYjGi8eOJ5Fej004jeughtr//faIf/IDtBx8kOv10tqur2ScW4/coK+P3JCI64QSiRx9l++qriX70I7bvu4/onHPYXr+eYymVIvL5OJaamvi5Y48lWrKE7W99i+hnP2N7wQKi885je80aovJyolzuff36B7yecrvp6M2biYhoeyxGJZWV5MlkKKPrNLGqitYGg0RENLuxkRZKLP24s5Ouam0lIqLHR0fpeImlRomlQDZLCU2jCVVVVBUKERHRmfX19MDQEBER/bCjg65tayMiokXDw3SKecH3EwFooN2siTZDt7XfSNMAme/wnmwzC97p/M/5Oxz8ADijXWadvCd/XefPNv3/XelEcAJwOBwgIhgACuTNNMNAofxiu7M/DH/z9R+GPwFwvov/+9FHKZY+jnovsZRvv9dY2p1t7ma9k//+oD0xdHtBt2XLli1btg4Q2UlxtvYbdXdbvaw9HqsON5GweLhhWPXdAPdJNyt4oXjBAAAgAElEQVRFurqsmvTRUaumOxazGKhhWDW9ANvmX1AdHVYd8siINSc9GrUYqK5bc6oBPhbTv73dqkMeGrJ4fjhs8XRN27m+vKGBa5gB5u5mHfIHVX8qpXpRB3I5VUebMQzFswFm3aa2x+PIyC/jSqVUTbk/m1WzpdO6rlqy7uq/LR5XNe29qZSqKR/LZlWP+JSuK4ZJRDv5N8diqi97TzKp6oi92SyGxD+h62gzT/Ie1NVlxZLbbcVSPL5zLO0aC2YsdXZasTA6arWEjcWs3Apd39l/11gwT9PwsBVLkYgVS5r29lgyY6GtzYqFoSGrpWw4bPH0XO6D9Sr4KOmdYsnsj9CdF0ueTGanWDJ5uEGkZgcA3C7ZrGnvSiZVbsloJqNyOw447W4v/kB42Az9wNOZZxLNn8/2D39IdPnlbD/yCNEnP8l2bS2R08n8OpEgKioiWreOnzvlFKKFC9m+9lqiq65i+4EHLAa/cSNRYaHF0AsL+WdE/BqTwV91lcXgFy60GPy6dUTjxlkM3ekk2rKFnzv6aIvBX3aZxeDnzbMY/GuvMTfPZIjcbvY3uelhh1kM/oPqy83NdHN3NxExA//81q1ExAx8oiQd9CaT5Fi/nroSCSIimlRVRc95vUTEDPw3wuBv6e6m/xYG/4zHoxh8ezxOjvXrqV8Y+vjKSnpRGPx5jY00Rxj8z7q66H+EwS8ZHaUZAqq3xWLkWL+eRoShl1ZWKgZ/TkODYvA/6uxUDP7RkRHF4Pek009n3kzE/Ptb32L7oYeYTxMRbdpEVFBgMfTCQqING/i5E09k3k3EDPzqq9m+7z5+johfW1TEvsEgv5d5aMceazH4b33LYvALFlgM/o03iIqLmaGPjXEsNDTwc0ccYTH4Sy+1GPwf/mAx+FWrmLt/XBl6vpqiUXKuX0/udJoyuk7FGzbQaxJLZ9XX07z+fiIiur6jgy6TWFo0PKwY/JZIhJzr15NfGHrRhg20Thj8KXV1isFf29amGPz+KOyBoe/1RRdAAYAmAKvk/58B0AmgBcATAIrk5w4AfwPQA2A7gDPf7b3tBf3AUyhElM2ynUwSxeNsaxonwZmSNeNtdihkfbklEjv7y7/NPfoHg/xaIvaVdY5yuf/f3nmHR1Vmf/x7k5CEohQBRZBiR0EsINgQ++ryU9dVF8ti711cV3RV3LWwWFFEBVFEUVBAUARBpBMpobcQQiCQhPQ+yaTN+f3xvTdniIDAEkIm5/M883hm7px733vvkXdyv+ecl/v+I//sbPUvLOQ5iPCc9tW/slIOCgXl5eJ3d1ZaWSn57oUJBAK7JJFlBNmZpaUSCARERCQ/yN9fWSkFQf5Z3k3ajb9HXnm5lLr+JRUVVf6Ve/HPqOZf5voXV1RIYZB/dpD/ngiOheqxtK+xEBxLwbFwILHkxcL+xJIXC4WFnPRFGEt5efq9zMzfn3t9ZU+xlFtWtkssFbk3pqJaLO3Nv9z19wX5H47sbUKvcQ3dcZynAXQHcKSI9HUc5xoAXiHI1wDmi8hH7uePAbgGQE8AQ0Wk5972bRq6YRiGUZ+oNQ3dcZx2AP4M4FPvMxGZFvRLYymAdu6m6wCMcTctBtDMcZw2NTk+49CQnKy6YUaG1gEXFuo64+XlqkcD2i+9uh3sX1CgGmpZmWqQe/NPT9c64vx81eNLS7UmeW/+aWlaR5yXpxqs36/17YeKnPLyqpru4srKKj08ILKLBujpiQBbuXpZvNnl5VV1vL7Kyio9vFKkqqa3un+y3+89eUNWWRlKXP+iigrkuv4VgUDVmucAqmrdq/tnlpVV1QHvK/sSC/saS8GxUFCgsbCvsbSnWKgeS562X91/507tT5Cbq3p8cCyJ7OpvGHujppPi3gPwLIDfLSDrOE4DAH8H4LXwaAsgKNyR7H5m1GHKyoATTtDmGn/6k/bJfuop4N57aY8cCZx/Pu0lS4COHdnApbiY/l7/9ssu0z7Xjz0GPPQQ7Y8+Ai6+mPaCBUCnTvqPdKdO2rSmd2/g449pP/gg8PjjtN9/H7j8ctq//spjFhdzDB06aNOY884DRo2ifc89wIABtN9+G7j66oNyyfaZWzZswL/cxuD/SUrCDevWAeCa553dAW8pKUH7xYurFq44felSfOfOFjeuX1/Vs/2lrVvxN3dx8HEZGejqdjbZ6POh/eLFSHIn4lOWLq1a6/y6devwutsne+DWrbjdzUT7Mj0dZ7tZjauLitBh8WKklpaiIhDASUuXVvV///PatVV9svcFv5/3xWu0cuWV2jP9ySeB+++n/cknwEUX0Y6JYSzl5XHCPP549v0HgEsuAT78kPbDDwOPPEJ72DDg0ktpz51Ln6IiTrodO2rTmQsvBEaMoH3//RwDwDUBrrqK9syZHLM3SXfooIu49OzJxVoA4K67gGeeoT1kCPDnP9P+6SfgpJPqby93Yz/Z07P4//UFoC+A4a7dB66GHrR9JID3gt5PBXBh0PtfAXTfzX7vBxALILZ9+/YHR5QwapSNG1Ur3LaNeqEIE87S0miXlIhs2UI7EBAJzkkJ9t+6VbXSzEwRN79LfD4RNz9LAgERty+EiNB2ZWNJTFStND1d9cmiIu5bhMfauFH9169X/y1bVOtMS+M5iPCc3JyaQ0aK3y+5rj6YV14uO9yBlVdWyibvJIXNXTw2+XxVWmOy3y95roicW1YmKX6/iIiUVVZWJdFV94/z+aq0xu0lJVW6fU5ZmaS6/v7KStkc5L++mn+FezGTSkqqdPd9pXosBceCF0vFxfsWC1u3aixkZKi+7fNpLPxRLHm6eVrarrHk5mftNpY8EhI0lnbu1BySggKNpYoKkbi4fbkyRn0BtaGhO47zBvgXeAWAaABHApgkIrc7jvMygLMA3CAiAff7nwCYKyLfuO83AegjIjv3dAzT0A3DMIz6RK1o6CIyUETaiUhHAP0AzHYn83sBXAXgFm8yd/kBQH+H9AKQv7fJ3Kg7BNdd+/1ax1tRoRqmiK4ZXt0n2C4pUT2+vFw1zNrwLyvTR6GBgPabP1SUBgJVdbSVIlX15QCqtPG92f+rv7+yskqPrwgEqurTD8R/X9lTLNV2LAT7V4+FPfkH23uLpYPVt8AIfWqjsczHAI4G8JvjOKscx3nJ/XwagESwbG0kgIdrYWzGQaasDGjVSnXP3r25DjUAPPoo0L8/7ZEjgTPOoB0bCzRvrhr6UUep7nneeaqbPvAAcPfdtD/8EOju/maNiQFatFANvUUL1T3POQcYPpz2XXdRRweoe3oa/ty5PKanoTdvrrpn167Ap26K5+23U8cHqHt6Gv6h4oZ16/Ccu1D7oG3bcM2aNQC4Nnlb94QTS0rQdOFCJLizwnG//YaJrobed+1avORq8M8nJuJ6V4Mfn5GBTu5KNZuKi9F04cIqDb1NTAx+dDX0q9aswX/ctar/kZiIm9avBwB8mZaGU1wNf11REZotXFilobdetAg/uwtkXLJ6NQa7Gvy+4PcDLVtqPsUFFwDvvEP74YeBO+6g/fHHwFln0V6yhPfP09BbtNB8inPP1XyMe+8F7ruP9vvvU98GmI/RooVq6M2baz7FmWdSrwd4bE+Df+cd6usAx9qypWrozZppo5kuXVRDv/VW1eAHD1YN/+efgdatTUM39pE9PYuvCy+rQ68bzJ8v4sqrsmqVapXbtonEx9POzRVZtox2ebnI7NnqP28em7SIiKxYobp1YiJ1SBHqj17DjrIyLorhMWeO1r7HxqpWmZCgWmtmJvctwmPNm6f+s2drvfKyZVpjHB+vWmlGBs/tULKhqEiS3Qub6vdXad2+igpZ5BYyBwIB+TUnp6r2fFFeXlWN7bqioirdPNnvr9K6C8vLJaaav8fCvDwpdv3XFBbKTtd/e0mJbHT988vLZXF+voiwpjzYf35urpS4/qsLCyV9PxddmTdPY2nlStWtt24VcXvsSE6OxkJ5+a6xMHeuxsLy5RpLW7ZoDkdWli7GU1amjWhEuK/gWPBObfNm1d0zM7WRkN//+1jySpyXLtV6802bVDdPTxdxe/RISYk2RTIMkVquQ69JTEM3DMMw6hPWy90wDMMwQhyb0I0apawMOPpoYP58vu/ThxolwDr0u+6iPWYM0K0b7VWrqGHn5lLHbt1aNfDzz2fNOUAN3qs9HjUK6NGD9rJl1C0LC6mht2ypGniPHlpHfv/93AfAfV5wAe2YGB6zpATIyeFYVq/mtm7dgC+/pH3nncDTT9MeOpR1zYeSG9etw4uuBv56UhL6uhr6tOxstIuJAQAk+f1osXAhtrpZXu1/+w1TXQ382rVr8R+3Dn3Q1q1VdeyTMzOrNPQtJSVosXBhVUOYY2NiMNOtI796zRr819XAn09MRD9XQ/8uIwOnLFkCAIjz+dBi4UKkuRr6MYsWYU5uLgDgslWr8O5+1qG3bq0a+EUXsWYcAJ54gn0BAGD0aODss2mvWMH7n59PHbxVK+rqANCrl9aRP/SQ9jQYMYLbAH63VSsuyJOXx315i7WcfTaPBfDYTzxBe9gw5ooAHGvr1hx7VhZjae1abuvSBfj6a9p//zvwj3/Qfvdd9lsAgDlzgGOOMQ3d2Ef29Cy+LrxMQ68bfPcda2tFuPCJpzWuWaO6eUoKFzURoW74zTda7zt+vNau//KLao0rV6rWuWOHyIwZtH0+kXHj9Pjjxmm98YwZIsnJtJcvV907KYn7FuGxxo+nHQhwLF698LRpIqmptJcu5TmI8JyCdf9DwaK8vCrdOt7nk/muuJ9dVla1gEp5ZaWMTUurqh2fnJlZ1dt6QW5uVb16nM9XpbtnlpbKFFecLnP9vdrxiRkZVbXv83JzJcEtxF5fVCS/uf7ppaXyo+tfWlkpX6elSaXrPyEjo6r2fU5OjiR6hdz7yLffaizNmqU5DKtWqW6ekiLy88+0i4t5/zzGjdPa9ZkzRbZvp71iheZQbN/ObSL8rhcLItyXN+Tp03ksER7b0723bePYRDjWb7+lHQiIfP21xtJPP7H+XISL/7jriUhiour+eXkiEybs1yUyQhyYhm4YhmEYdR/T0A3DMAwjxLEJ3ahRyspYX+7pln/9q9beDhpEHR0AJkzQXuobNgCnnUbds6SEtd+ebnnttaphv/AC8OyztMeN017qa9YAp59OzbSoiLYrL+Pqq/ldgL4vvEB7zBjguutoL1/OY/r9HMNppwFum3JcfjkwcSLtJ58EXnnloF2q/4kPkpPR3x3knNxcnOM+uUopLUXnpUuR7HYq6bF8OX51New7N27EUHflj7d37MC9cXEAgBk5Oejl9mJP8vvReelSpJWWQkRwVmws5rsr6ty2YQOGp6QAAAYnJeGh+HgAwNSsLFzkFltvKSlB56VLkVlWhopAAN2WLUOMu4rK+pvXI3VkKgDgP//RvvqTJ2s+wqZNvP65uVz0pGtX5kgAwF/+ohr2Sy9pX/1vv9Ve6uvW8f4XFjIfo0sX5mgAQN++wNixtJ97Dhg4kPbYsdwGsGa8Sxf6FhRwX26qAK68EvjuO9oDBnAMAMf0l7/QXraMYy4t5TmcdhrPCeA5Tp5M+/HHeQ0A9mS4+WbaMTHM2zAN3dgXwgcNGlTbYzhgRowYMeh+LyvKOCwJD2fnqwsvBBo3BsLC2JCjdWsgIoILp5xwAhAZyYShbt2AqCj6XXgh0KABUFnJJKOGDQHHYTJSq1b0P+EE7iMykp917Ur/Bg2Y5NagAY/fuzcQHU3/7t15rPBwLnzRsaP6d+lC/8hIJuAF+0dF0b9HDzYbCQ8HTj4ZaN++tq8y0MBx0DYqCqc2aoSosDAcERGB7kccgaiwMAREcHGzZogI4+/385s2xZEREQhzHHRp3BjHRkWhgePguKgonNyoEaIcB00jInD2EUcgynEggPqL4IKmTdEkIgJhAM5o3BjHREUhwnHQIToaJzVqhEjHQfOICJzlHt8B0LtZMzQIC4MAuLBpUzQOD4cT5qBJtyaIPDoSDRrwPnix0KIF4yQqijFz0UV6Ly66CGjUiJ+fdZbGwvHH8+XF0hln7FsstWzJ75x44q6x4MVSZKTGkudfPZYiIhgLHTrwe0cfzck/KorfPe889b/4Yo2lc89ls5rwcODUU4HjjuPx2rTh5B8VxXPtudeFpI36xCuvvLJz0KBBI3a3zTR0wzAMw6gjmIZuGIZhGCGOTehGjVJeDlxxhdZxP/AANU6A9bb//jft6dPZzxoAEhLYy7qwkNrj5ZdTCwVY7ztpEu033wRef532jz9qX/i4ONbxFhfzddllqlv278/vAvR9803akyZpHfO6dTxmaSnHcOmlwJYtB//aHArSy8rQZ+VKpJWWAgD6rllTpWEPSEjAZzu5/tHI1FQ8657kgrw8XOsVS9cgDz8MfPMN7fffB15+mfbMmcDf/kY7MZFac0EB8zGuuELzIe67j7kXANejf/VV2j/9xD77ABAfz/vn8zEn4rLLNB/izjuBKVNoDx4M/Pe/tKdM0f4IGzbQx+/nPi69FNi8mdtuuw2YNo32q69qX/kJE7Qv/Jo1HHNZGc/hkkt4TgDPceZM2i+/rP0ZvvmG1wZg7shVV/FRvWH8ERG1PQAjtImIYJOOo47i+zPPVM35pJN0Vatjj+XCKQAXsOjZk9pjWBj9W7TgtrPOos4IULP0/qFr21b9mzenf2Qk3/fsyX0C1EzbtaN96qkcH8B9egt6tGjBYzZoQJ2zZ0+gadODe10OFY3DwtDryCPRODwcANDzyCPRukEDAEDXxo3RIToaAHB8w4aIcjX2oyMj0fPII2t8bN26UXMGGAtHH027TRttEtS0Ke9FdPTvY6l6LLi/WXaJhWbN6BMZqbHUvDm3BcfCKafwXgP8bHexEB5O24ul7t0ZtwBjKSqKdnAsHXUUfSIieA69emksBft37qyfd+jAyR+gvt+zJ8duGH+EaeiGYRiGUUcwDd0wDMMwQhyb0I0apbKSeqZb4owXXqBeDgCff67rUc+fr72wd+wAbrmF+nd5ObXKhARu++c/VXccOVL7us+erXXIW7dSjy8t5evWW/kZwO9462kPH859ANznP/9JOyGBxywv5xhuuYVjqovklJfjb+vXI7u8HABwT1wcVhYWAgBeS0rChIwMAFwDfbC7tnlsQQHu85IOapAXXwSmTqX9xRe6zv3ChbrOfHIy0K8f9euKCsaSW+6OgQO5XjjA/vxeX/e5c7W/QVIS75/fTx371ltVw/7HP4BZs2h/8omubT5rlvZV37KFPmVllIduuYX7BHiMefNoDxumawT8/LPWtMfHc8wVFTyHfv14TgDPcdEi2u+9x2sA8Jq8+CLtDRuY9xEIHNAlNuoZpqEbNU6jRqpVR0dTjwSoa3q6Y4MGrA0GqBc2bMj/Og79XQn4d/7e58H+4eFaZwzQ9o7fsKH6e/Xqu/Nv1Ij+jsPPvePUNcIANAoPr/rl3jAsDBHuhYkOC0OkK85GOg6iXTvCcdDoEIi2wfcyKmr3seDdi93FQvC9jIwEPPVwb/dyb/5evOyPvxdXfzT+A/WPiFDbMP4I09ANwzAMo45gGrphGIZhhDg2oRs1SiAAPPOM6o7vvgssWEB70iTgq69oL18OvPYa7fR09kkvLaUGP2CA6o5vvaVro3/3nfZlX7qUtcQAkJpKfbO8nK+nngLccmsMHszvAqz39Xpxx8Rw3wD18gEDQrP29/nERMT5fACAj1NSqtY2/zk7GyNTUw/pWIYOVQ16yhT20wfYP93ra56ZyVjw+xlLAwYA7hLseOcd1aAnTtS1xWNjgTfeoJ2WxvtfVkYd++mnAbf9PIYMAdxl3zF+PF8APxsyhHZKCn0qKriPp57iPgH2MfAeEH79tfb4X7RIa9K3b+eYAwGew5NP8pwAnqPb8h5jxmhN/Lx5vDaGsb+Yhm7UKIEAG3G4eVjYto39sgFOskVFtHNytHlLcTF9KiqoO8bHq//WrawZBjhxe4tWZGdrspPPp/4Abe84iYmshQf4j7WnoWZna+JcURGPWVlZd7XzPbG5pAT57i+V7aWlaOKeYHp5ObZ7hdyHiG3bWDMOMBbc3xbIzd01FuLj+cMsPHzXWNq6lb3fAcaC19MgOJZ8PvpXVFCHj4/fNRZOP139ParHwubNjIXKSvq7v4eQmKhjTk6mPg4AWVnqX1io/uXl9C8u5rYtW3iunr/3eWYmr41h7C+moRuGYRhGHcE0dMMwDMMIcWxCN2oUEfZL93THL79U3XD2bPbdBlinPsJdEDA3lxpoRQUf2Q8Zorrj6NHay/uXX7QOef16rQPOzqZW7j0mHTyYnwH8jreetVEzJCSwxh/gI+fXX6f+DDBPwctnGDuWuRMAMGeO9tiPjwc+/ph2fj5jobycsTRkCHMsAOrO3hoBs2ZpX/WNG4FPP6Wdk6OxEAiwX3tWFrd99pmuETBjBl8AP/v8c9pZWfQJBDSWvMfsI0dqX/hp07SmffVqzQdIT+eYRXgOb7zBczKMmsAmdKNGqahgspCX1Pbzz/qP6OLFmiC3aZP+g56eTp/iYk4EEyZoItO0aTohx8SwCQnAf1i9Hwc7d9Lf7+dr4kT9QTF1qja5MWqGhARg8mTaWVm8/oWFGgteUtuMGYC3BszSpWwuBHBC/+EH2hkZvP8+HyfEiRM1FqZP3zWWvFiIi9s1liZMoL7u99P29PJp09i4BWAsxcTQ3rBBYyk1lccsLWU8TpigPyiCY2nhQk2wW7dOmyelpNC/vJznMGECz8kwagLT0A3DMAyjjmAaumEYhmGEODahGzXOmDFAXh7tadN0Penly7WOePt24Pvvaft81DC9/tVffKHLSU6dquVpS5fqY85t2/QxbWEhtXYRvkaP1lKlH36wkqCaYPVqrSlPTdV1yv1+atVeTf9XX6kG/fPP2pd9xQp9ZJ6crGveFxfT34uFMWNUg542TXv8x8bqI/OkJH3kX1TEWAqOBa/s7ccftbxsyRK+AH7mPbIvKKAPQP/PP9eytcmTVT6IidGa9IQE1fPz81VPDwR4Ll55mmEcdESkzr7OOeccMQ5vSktFOnQQWbiQ76+8UmT4cNrPPity//20x44V6dmT9po1Im3biuTmihQXi7RvL7J4MbddconIyJG0n3pK5JFHaI8eLXLhhbSXLxdp106koICvdu34mQi/88UXNXrK9ZKXXxbp35/2pEki3brR3rRJ5NhjRdLTRcrLRTp2FJk3j9uuuUbkgw9oDxwocs89tMeNE+nenfa6dYyF7GyRkhLGUkwMt11+ucjHH9N+5hmRhx6i/eWXIuedR3vlSt7//HyRoiKR444TWbaM23r3FvnsM9qPP86XiMioUSIXX0x76VL6FBWJ5OVxX6tWcdt55/FYIiIPPsgxiHBMV1xBOyaGY/b7eQ5t2/KcDONAARAre5gTTUM3DMMwjDqCaeiGYRiGEeLYhG7UODNmaFvOJUu0DnnzZi1By8pSDbW8XDVIgFqr30/7t9+0BG3TJq0DzshQDbW0VMuGANpeV9NFi7Sm3Th4JCZqf4DcXNXTKyq0BAzguvOeBr1smZagJSRoCVp2tpYzlpfv6r+nWIqP1xK0zEyNpbKyXWMpOBZiYrSELC5OS9DS03W9AL9/11iaNk1r6hcu1Jr2DRs0H2DnTtXjS0q0vh3gubhL0xvGwWdPz+Lrwss09MOf0lKRI44Q+fVXvu/VS+TNN2k//LDILbfQ/vRTkVNOoR0bK9KwITVHn0+kSROR+fO57eyzRYYOpX3vvSJ33EF7+HCRrl1p//abSKNG1E3z82l7GnyXLiIffVSjp1wvGTBA5PrraY8dy7wHEerF0dEiqanU0I88UmTmTG674AKRwYNpP/aYyM030/78c5ETT6S9ciVjITOTGnqTJiJz5nBbjx4i77xD+4EHRG6/nfYnn4icdhrtpUt5/3NzRQoLRRo3Flm0iNu6dRMZNoz2XXeJ3H037Q8+EDnzTNoLF9KnsJD7aNSI+xQR6dxZZMQI2rffTh1dhGM691zac+ZwzCUlPIeGDXlOhnGgwDR0ozapqAAiIn5ve5nHYWF7/96ebC/z+UD9jYOHCO+Ht5jNvlz/4MVvajsWasPfMA6EvWnoFlpGjRP8D1iw7Th8/dH39mSHVROM9tffOHg4zq4r0+3L9Q/+fm3HQm34G8bBxjR045CyebPWEaelaUtYn0/18EBAe3wD7P3u1THHx2tNemqqarCFhaqBBgKsa/ZYsUL/goqL0zpk4+CRkaFr3peUaG6EiNZnA8CqVaohJyRof4L0dK5DDzAWPD1c5Pex4C2LGx+vsbRzp8ZSUdGusVQ9FrxY2rRJYyE1VVvCFhZyG8DvBvsvX66xtHGj9jdITlY9Pz9f9fSKCl27wPOvww9FjcOdPT2Lrwsv09DrHmefLfL667TvvVfkxhtpDx8ucvzxtBcvFgkLE8nKoobeoIFq8F27qgZ/xx2qwQ8dqhr8ggUiERGqoUdE8DMRfsfT4I2Dx2OPifz5z7Q//1ykTRvaq1eLOI5ISgo19IYNRaZO5bZzzxX5979pP/igavAjRqgGHxvLWMjIoA4dGaka/JlnqgZ/992qwQ8bphp8TIxIeLhq6BERInPncttpp6kGf/vtqsG/845q8HPnMv4KC0Vycrgvrw7+xBNVg7/5ZtXgBw9WDX7mTJGoKI49I4PnEht7wJfZMExDNw4f8vKAxo2BBg34l1wgwPeVlfzLpkULfi8zE2jV6vd2Xh7QpAkfXRYX868dz7+gAGjefO/+ubnAkUfu+rjX+N/x+/mX9xFH8J7m5gJHHcVte7oX+flAo0YaC5WVvLeBAO/z/sRC9Vja11g44giNJYDjqajgX+n7E0s+Hx+7N2xI/6IioFmzvfsbxoGwNw3dJnTDMAzDqCNYYxnjsKe8XOvLAdVTq9sZGVpHXFCgGmpZmS5ruTf/9HStI87PVz2+tHTXZfVWjJ0AACAASURBVC335J+WphpwXp5qsH6/1reLqJ5b3X/nTtWADybFxbrmeyCguQXVj5+Sohpwdrb+ZerzaY/1ykrVk6v7JyerBpyVpTXhRUX8ixXg+Xl6MqD9zqv7Z2Zqf4G9+e9vLOxrLAXHQkGBxsK+xtKeYqF6LO0pFgzjoLOnZ/F14WUaeujw4YcinTrR3puG3qWLauj9+6uG/t57qqHPn0+t09PQw8NVQz/5ZNXQ+/XTOvYhQ7SOfdYsHtPn4xgcR2TJEm7r2FHr2G+4QeS++2i/+qqIF47TplE3LS0V2bmT/l7t8bHHslf4wea559ibXETk229Z+y8ikpDAgrD4eL5v2pS90kVE+vRhP30RkaefZm90EZGvvhJp0YL2hg3037ZNJBBgHfakSdx2/vki//oX7UcfZW92EfZHP+YY2qtW0d/T0KOjVUPv0UM19AceUA39k09UQ1+2jNfP09AbNFANvVs31dDvuks19A8+UA190SLG0u409M6dVUO/7TbV0N9+WzX0OXPo42noYWGqoZ9wgmroN92kGvobb6iGPmMGdX9PQ3cc7SVvGAcC9qKh1/qk/L+8bEIPHUpKRLZsoR0IiKxfr9s2bhSprKS9dSsXyhBho470dNo+n0hiovpv2KD+GzbwMxF+x+ejnZ7OfYhwn1u30q6s5DE91q9X/y1bOFYRkbQ0Tvgi/Ac/KYl2RcXv/T0SErhQx8EmL09kxw7a5eVcFMUjeDGQTZtEyspoJyfTT4QTXkoK7bIy/QFQ3T8ujvsXEdm+nT+YRDjZpabS9vtFNm9Wn+Dzj4vj9RHh9SoooJ2dzR8/Iry+CQm79w+OhW3bdo2FtDTaxcX7Fgtbt2osZGTwJcLPvFj4o1gqLqadlrZrLG3bRnt3sWQY/wt7m9BNQzcMwzCMOoJp6MZhj4jqscCua0YH2yUlqsGWl6uGWRv+ZWWqhwcCqgfvzT/YDvY/EEpLtaa6slL15H09/v/q7/erHl9RoXp0bfjXdiwE++9rLBjGwcYmdOOwYORI4IwzaMfGsmTIS9o66ihd7OO884D33qP9wAPA3XfT/vBDoLv7mzUmhiVPXqJTixa62MY55wDDh9O+6y7gwQdpv/sucP75tOfO5TG9RLPmzbU5SteuwKef0r79duCxx2gPGQJcfDHtmTNZmlRWxsSpZs104ZJTTwW++IL23/4GDBhw4NfshhuA556jPWgQcM01tL//HmjblnZiItC0KZu4AMBxxwETJ9Lu2xd46SXazz8PXH897fHjgU6daG/aRP+kJE5ebdoAP/7IbVddBfznP7T/8Q/gpptof/klcMoptNet4/mnpnLSbt2ai+0AwCWXAIMH037iCeDWW2mPGgV06UJ75Ur6ewl0LVsCs2dz2wUXAO+8Q/vhh4E77qD98cfAWWfRXrKE9y8vj4l3LVrowi3nngt88AHte+8F7ruP9vvvAz170l6wgD5e0l7z5sDSpdx25pnAJ5/QvuMO4JFHaL/zDnDhhbRnz+aYvaTJZs12bTRjGAeVPT2Lrwsv09BDh9xcTRYqLxeZPVu3zZvHBDMRkRUrVLdOTFStNTtbG3aUlekCHiK0Pd04NpbfFaGvp7VmZnLfIjzWvHnqP3u26sbLlnGsItSZPa00I4MJYCLUkL3FZESY0OfpxkuWqO4cF0cd+kDZsIE6uAj1a0/r9vl0AZJAgMf3dN9Fi1R3XrdOdfPkZNV3Cws18cvz91i4UHXjNWtU996+XbXi/HxdDKeyclf/+fM1B2H1as2BSEpS3T8vTxdAqaj4fSx4OQgrV6puvXWr6vY5ORoL5eW7xsLcuRoLy5drLG3ZojkcWVncJsLvekl0ItxXcCzk5NDevFl198xMTYL0+38fS14sGMaBANPQDcMwDKPuYxq6YRiGYYQ4NqEbhwVjxgDdutFetYoadm4udezWrVUDP/984KOPaD/6KHD//bRHjQJ69KC9bBl1y90twtKjB78L0PfRR2l/9BE1WYAafOvWTHrKyeFYVq/mtm7dqBEDwJ13Ak8/TXvoUGrCAPX+o4+mhp6RQQ3WW2ykc2dq1AA1Y08DPxBuvBF48UXar79OTRwApk0D2rWjnZTE42/dyvft2wNTp9K+9lrVwAcNoiYPAJMnq4a+ZQv9vYYwxx7LHAEAuPpq4L//pf3880C/frS/+0419Lg4+qelUUM/5hhgzhxuu+wy5i4A1OD//nfaX3+tGvratbz+WVnUoVu3Vg38oouAYcNoP/EEcM89tEePBs4+m/aKFYyF/Hzq4K1aUVcHgF69gBEjaD/0EF8AP+vVi/aSJfTx+ajDt2ypi7WcfTaPBfDYTzxBe9gwoHdv2gsXcsx+P8/hqKN4ToZRI+zpWXxdeJmGHjqkpIhMn067pETkm29U9x0/nrquiMgvv2i998qVqnXu2MEmHiLUkL3mKdWZMUN15+XLVfdOSuK+RXis8eNpBwIci6f7Tpum9dZLl1JHFqF+6mm9+fki331Hu7JS5OuvVfedOlXrpRcv3rXGe39ZtEh16/h41e2zs0W+/552ebnI2LGq+06erDkECxaobh0Xp7p7ZqbIlCm0y8ro7+m+EydqDsG8eZrDsH69yG+/0U5PF/nxR9qlpTx/r3Z8wgStfZ8zR3MY1q7V5j07d4r89BPtkhL6e7Hw7bdauz5rluYwrFqlunlKisjPP9MuLub98xg3TnMIZs7UHIYVKzSHYvt2bV5TVKSxIMJ9eTkE06drDkJsLHMCRDimWbNoFxRwzCI8h6+/1lgyjAMBpqEbhmEYRt3HNHTDMAzDCHFsQjcOCyZMAC6/nPaGDcBpp1H3LClh7benW157rWrYL7wAPPss7XHjqOkCrPk+/XRqpkVFtL068Kuv5ncB+r7wAu0xY4DrrqO9fDmP6fdzDKedBmzcyG2XX6513E8+CbzyCu1Ro6hpA8DixaypLy9nHXvnzsDmzdzWu7fWcT/yCLXvA+X++4E336T9wQdA//6058xhvT3AxVg6d9YFQnr0AH79lfadd1L7B4C332YtNgDMmKEaclIS/dPSqKGfdRYwfz633Xab1vQPHqwa9NSp1LcBavCdO7MGu6KCOQgxMdx2883sPwBQy3/8cdqTJ2s+wqZNvP65uWx807UrcyQA4C9/UQ37pZe0pv/bb1kjD7AO/vTTmU9RXExtftUqbuvbFxg7lvZzzwEDB9IeO1bzEVaupE9xMXsanH46sH49t115JfMFAB7bq+kfPZpjAzjWrl13bdpjGDVF+KBBg2p7DAfMiBEjBt3vZUUZdZrISCYMdesGREVxjekLL+Ra2ZWVnAgbNgQch8lIrVpxHesTTmACV2QkP+valf4NGjDJrUEDdu7q3RuIjqZ/9+48Vng4cNJJQMeO6t+lC/0jI5mAF+wfFUX/Hj2Y6BUeDpx8MhPNIiOZCHf66fxew4acFCMj6X/xxbQdh01LmjWj/6mnstnLgRAWxsmubVuOs21b7i8qiut8d+9O2zt+RAT9zj+f63iHhfF8jz2W/scdx/OJimIzmbPPpi2yq/8FF3Ad8rAw/nA55hhu69CB1zMykg1YzjpLr1nv3jyGCO9r48b079aN161BA96HE06gf4sWbNwSFcXvXXSR3ouLLuK65WFhPIYXC8cfz5cXS2ecsW+x1LIlv3PiibvGghdLkZEaS55/9ViKiOC169CB3wuOhehoNkRynIPwP4pR73nllVd2Dho0aMTutpmGbhiGYRh1BNPQDcMwDCPEsQndOCyYPl17eSckAJdeSt2ztJS69bp13HbPPcCkSbTffFM16B9/VA05Lo41zsXFfF12GbVYgN/xNOzXX1cNetIkrWNet47HLC3lGC69lFowANxyi/Yif+UV7Ss/frz2hV+1ivpqRQVrly+5hFo0QJ3d07D/9S/2oAeo23o18bGxwJ/+xMfL2dlAnz6qgV9/vfa1f+457SU+ejTw1FO0f/tN+7qnp9M/LY3v+/ZVDXvAAOCzz2iPHKn5CAsWMFcBYA/2Pn1YQy3CHASvl/njj2s+w/Dhmo8wezbw17/S3r6d55+by8fVV12l+RAPPwx88w3t998HXn6Z9syZ7HMPsBf9JZdQvy4rA664QvMh7ruPuRcAcwBefZX2Tz+xzz4AxMfz/vl8zIm47DLNh7jzTmDKFNqDB2tN/ZQp7PMPMJ/jssvo6/NxX14+xG23sebfMA4XImp7AIYBUMf1ErmaNaPOHB1NnbRXL2qqADVTT3M++WRdLaxtW/Vv3pz+kZF872nWADVTr+nKqaeqLnzccbqgR4sWPGaDBqp5N23Kbd27c6wAk70aN6bdvj01X4Caas+e1GWjo2kfeSS3nXsuNWeAGqt3Xh066EpcLVvSJyyMWnGvXtTEd+ffpg3tTp105bHWrTWprXFj2t44e/bkdoAacYcOtI8/nnovQP3XW5ykSRP6N2qk16JVK2474wxqzgD1Z+8aHXMMxwlw3D17UrMOC6PdsiW3deumxz/pJB4X4Dl5TYKaNuXxo6N5r3r14vUFfh8LXuJZcCw0a6a5DF4sNW/ObcGxcMopqnG3a7f7WAgPp+3FUnAsGMbhgGnohmEYhlFHMA3dMAzDMEIcm9CNw4L587UX9o4d1KqLi1nLfdttup73P/+pvcRHjtS+7rNnax3y1q3U40tL+br1Vu1lPmCArqc9fLjWQc+cyX0DPNZtt/HYxcUcy44d3PbEE9SYAeq+n39Oe9o01ZDj4tiXvLKSGny/ftSiAerGXl/6d94BvvqK9g8/qIa8bh3X1xZhHfzf/kYtHOAa8F4d9pAhWlM/aZJqyKtWqQack0P/7Gy+v+ceXY/7tddUgx4/Xtcmj43VtcEzM+mfl8fx3Hmn9iJ/5RXWjAPMAXjrLdqLF2tN+s6dPP+CAkoC/ftrX/sXX9S+8l98ofkICxfqOvPJyfT3+ZiTcPvt1MUB1o17+QyjRmlf97lzNZ8gKYn3z++nBn/rrdTlAfaPnzWL9iefaD7CrFncBjB34tZb6VtSwn15+RBPPaX5DIZxOGAaunFY0KABdVaAWqenuToO9dvwcG6LjuZ3Aeqi3ufB/uHhWmcM0Pa08oYN1d+rV9+dv6cZOw4/944TvK+oKNWd93R871w8/0aNdvX3dP5g/4iIXa9F8PlXP36wf3S0+jdqtKt/WNjv/aOj1T8y8o/9vWuxL/7Vr8Wejh98L/Z0LYOPX/1aBMeCpx7u7V7uzd+Ll/3xj7B/QY3DCNPQDcMwDKOOYBq6YRiGYYQ4NqEbhwXLl1PTBagXP/kk9e/KSureXh32W2+pBv3dd6ohL12qGnBqKvXN8nK+nnqKWi7A73h11N98o724Y2JUA96xg8esrOQYnnyS65oDHKNXR/3ll8D339NesEA14G3bgGeeoWZcUkLdPSuL2155ReuoR4/Wmvg5c9iPHaBu69WE+3z0z83l+5de0l7in36qddC//KL5BPHx2pe8sJD+BQV8//zz1PgB4OOPNR/h5581n2DDBtbIA9TOn3iCPfEB5hl4+Qwffqj5CFOnaj7B2rVcXx2gdv/EE8xFEKE27eUzDB2qGvSUKeynD1Dj99Zpz8zk9ff7eT0HDGBtO8AchEWLaE+cyHXUAeYAvPEG7bQ03v+yMmrwTz/N/vYAcxAWL6Y9fryuU794MbcB/O7TT9O3rIz78mr6X3+dxzKMwwVTgIzDgpwcbd5SXMzmHRUV1C3j4zkxAZwMTjmFdmoqvwNw4vCSnXw+9QdoexNSYqLWi6ekqIaana0TTVERj1lZyR8EmzdznwDHmJOj/l59eWam+hcW0icQ4CSweTMndoCToTc579jB/Xv+XrJVQYH6l5aqf/PmtPPy1N/TczMy1D8/XxPH/H7afj/HunkztwOcGJs0oZ2erhNlfr42T/H8S0tZy755s/442LZNa8LT0jTxLzdXJ/2SEvqUlVEjr+7fti3tnTv1uubm7hoL8fG8TuHh9A+OhRNOoJ2aqtc4OJZ8PvpXVFCHj4/fNRZOP139ParHwubNjIXKSvp7sZCYqGM2jMMB09ANwzAMo45gGrphGIZhhDg2oRuHHbm51EArKvjYecgQPpIGqDt7GvQvv2gd8vr1rEUG+Mh08GB9TDp4sNZhjxqlGvT06dwHwH16a2tnZvKYIhzDG2/oY+4RI1SDnjpVNeSVK7WveVoae8SL8FHz66/rY+aPPtLH0VOmqIYcG6sacEoKe5MDfNT9+uv6mHjYMJUWJk1izTYALFnCdcABPjr39PySEvp7bWWHDtVH8999pxryokW6zvvWrarn+3z09/v5/p13NJ9h3DitiZ8/X2vSExJ0nfTCQvqXlfH9W29pPsPYscydAJhD4OUTxMdT3wf4+P+NN/jIXYT3xavJHzMGWL2a9qxZmk+wcSPzCwA+EvdiIRBgv3Yvn+Gzz3SNgBkz+AL4mZcPkJVFn0BAY8kesxuHKzahG4cd6emcXIqLORFMmKCJTNOm6YQcE6MT2saNXJQD4IQxcSInIb+ftpfINHWqTsiLFmmC3fr1nOABHmvCBE6mxcX09yaRH35QfXrBAk6kABPBvAlhxw76VFRwIp44UX+QTJ6s+vS8eTohrl6tPy62b6dPIMAfAhMn6g+S779XfXjOHJ0QV67UJinbtmnDmLw82p5uP3Gi6sO//qoJfitW6I+TxERdACcnh/75+ZxQJ07UHwS//MImNgB/kMydSzshQSf3rCz6FBbyekycqFr9jBnapGbpUv4oAHh9f/iBdkYGj+/zcVKfOFFjYfp0nZAXL9ZYiIvTHwfp6fQvKWEsTJigevm0adrkJiZGF63ZsEFjKTWVx/RiYcIEjQXDONwwDd0wDMMw6gimoRuGYRhGiGMTulFnWbpUNeBt2/QxbWEh9XARvkaPVg36hx/4XYC+Xk16YqL2Fc/PZ29xgI+9P/9cS5W+/177ui9apI+8N2/WR/a5uaqnV1ZSq/VKqoIfGc+fr4+sN23SR/bZ2dSXAT6mHjVKlwb99luVD+bO1XyCDRv0kXtmpq4zXl5Of0/DHjdOa+p//VXli7Vr+Qgf4CNlrya7tJT+Xgng11/r4/+ZM1W+WL1a8wFSU/WRv9/P8/eWuf3qK9Wgf/5Z5YsVK/SReXKyPvIvLqa/tzTsmDFadjdtmuYjxMbqI/OkJH3kbxj1ChGps69zzjlHjPrLU0+JPPII7dGjRS68kPby5SLt2okUFPDVrh0/E+F3vviC9iOPcB8iIiNGiFx6Ke3Fi0XatxcpLhbJyRFp21ZkzRpuO/dcka+/pn3ffSL//CftDz8Uueoq2gsWiHToIFJWJpKRIXLssSIbN3LbWWeJTJhA+847RV58kfZ774n83//Rnj1bpFMnkcpKkdRU+m/ezG2nny4yZQrtW28V+fe/aQ8ZInLDDbRnzBA58UTa27fTf9s2vj/5ZJHp02nfdJPI4MG0X31VpF8/2j/+KNK5M+0tW+ifkiISCIgcf7zIrFncdt11Im+/Tfvll0X696c9aZJIt260N22if3q6SHm5SMeOIvPmcds114h88AHtgQNF7rmH9rhxIt270163jtc/O1ukpITXNSaG2y6/XOTjj2k/84zIQw/R/vJLkfPOE8MISQDEyh7mxBrX0B3HCQcQCyBFRPo6jtMJwDgARwFYDuDvIlLmOE4UgDEAzgGQDeBvIrJtb/s2Dd0wDMOoT9S2hv4EgI1B7/8L4F0RORFALoB73M/vAZDrfv6u+z3DMAzDMPaBGp3QHcdpB+DPAD513zsALgXgKmz4AsD1rn2d+x7u9svc7xvGbtm0ieVqAHVhT0MtLVU9G6DtadCLFmkJ2caN3AdAXdrT4/1+rW8HqNV6LVoXLFANef16LUFLTVU9vrhYe6QDLIHyNOj587WEbO1aLUFLTta+4D6flrCJUNv3NOi5c7UmftUqLUHbvl1L0AoLqY8D1J6nTlUNevZsrYlfsUJLyLZtUz0/P1/19MpK+nsP8mbN0nyE2FitSU9MVD0/N1f19IoKLQEDeF28fIRlyzSfICFBS9Cys3XN+fLyXf1nzNB8hCVLtKY9Pl5L0DIzVY83jHrFnp7FH4wXODGfA6APgKkAWgJICNp+HIB1rr0OQLugbVsAtNzNPu8HH+HHtm/fvkY0CqNucO+9InfcQXv4cJGuXWn/9ptIo0Yi+fl8NWpEXVxEpEsXkY8+ot2/P3VwEWrYXkrG/PkiTZqI+HwiWVkiDRuqBn/yySKjRtHu1081/CFDVLedNUvkiCNESktF0tJEoqNVg+/YkRqviMhf/6oa/muvifTuTXv6dJGmTUUqKkR27KC/p8G3bUuNWYSa+7PP0n75ZWrKItTYjzqK9tat9E9I4PtWrahxi1Dz/9e/aA8cSE1bROS770SOOYb2pk30T0qiht6smcjUqdx2ySWq4Q8YIHL99bTHjmUOggg18Oho5gKUl4sceaTIzJncdsEFquE/9pjIzTfT/vxzzQFYuZLXPzOTGnqTJiJz5nBbjx4i77xD+4EHRG6/nfYnn4icdpoYRkiC2tDQHcfpC+AaEXnYcZw+AJ4BcCeAxcLH6nAc5zgA00Wki+M46wD8SUSS3W1bAPQUkaw9HcM09PqN91dnmPucqaICiIjYd/tg+DsOXwD/mvUWS9kX/8pKHvtg+ItwPIfa3/v+wfAXqZl7aRihxN409JoM+QsAXOs4zjUAogEcCWAogGaO40SISAWAdgDch25IAf9iT3YcJwJAUzA5zjB2S1g1wSj4H/B9sQ+2vzc57at/8Pf/V3/Hqfv+wQLbwbyXhlFfqDENXUQGikg7EekIoB+A2SJyG4A5AG50v3YHgCmu/YP7Hu722VJTjw+MkCA1VTXYwkKtiQ4EVE8GaHt/wcXF6fKbKSmqwRYUaE10ZSVbqXosX67+GzeqBrxjh9aE5+Wpnl5RoXo0QK3Zi+T167Wv+vbtWhMevGRoebn2KPf8Pdat077q27ZpPkDw8rGlpapnV/dfs0bzCbZu1XyArCytz/f7tSVrdf/Vq7WmPXgp2eDlW0tKtL5dZFf/Vas0HyEhQfMB0tO1vt/nUz1cRGv9Ad4XLx8hPl5r0nfuVD2/qEhzKwyjXrGnZ/EH8wVXQ3ft4wEsBZAA4DsAUe7n0e77BHf78X+0X6tDr9/ccYfILbfQHjpU5JRTaC9YIBIRoRp6RAQ/E+F3hg6lfcstqsG/+aZq8L/+KhIZqRp6WJjIkiXc1qmTavB//atq8K+9phr89OnUjUtLRXbupP/KldzWtq1q8P/3f6rBv/yyavBTplD3r6hgHbnjiKxfz22tWqkGf9VVqsEPHChy8cW0v/uOGr4I68gdRyQ+nu+bNhUZP572JZdoHf2AASJXXEF77FjV4DdupP+2bdTQGzcW+f57brvgAq2jf+wxkT//mfbnn4u0aUN79Wr6p6RQQ2/YUDX4c89VDf7BB1WDHzFCNfjYWF6/jAxq6JGRqsGfeaZq8HffrRr8sGGqwRtGqIHarEOvSUxDr98UF/MvuMaN+Vd1QQHQvDm3ZWYCrVr93s7NBY48ko97fT4+5m3USBdSadZs7/45OUDTpvQvKuJ/GzbkX50+3775N2vGR8RFRXw0HB3Nv3pLSrhvgH8xt2z5e//sbKBFC467sBCIjASioujv9/PcRPi93flnZQFHHUX/ggL6RkXxr/ayMuCII+ifk8Pv7c7f229+Ps89MpLHLi+nfyDA67w7/2A7P5/XvkEDnntlJdCkCf3z8niee/PPy+P3IyLoHwjsPhYMI5TYm4ZuE7phGIZh1BFqu7GMYdQ4ZWW7Lmvp6bHV7fR01YDz87Umu7RU9ey9+aelqQacl6d6vN+veraI6rnV/XfuVA04N1druktKdJ3uvfmnpmpNek6O6vnFxaqHBwKaW1DdPyVF8wGys1XP9/lUD6+s1CVGq/snJ2s+QFaW1oQXFWl9fUWF5iYAWute3T8zU/MB9ua/p3uRkaH5AAUFqqcbRr1lT8/i68LLNHTD4733VEOfP18kPFw19PBw1dBPPlk19H79VEMfMkQ19FmzRBo0UA3dcVRD79hRNfQbblAN/dVXVUOfNk0kKko1dMdRDf3YY1VD79tXNfSXXhLp1Yv25MnUmj0NHWA9t4hIy5YiY8bQvvJKkSefpP3cc1rH/u23qqEnJNA/WEP36tj79NE69qef1jr2r74SadGC9oYN9Pc09EaNtI79/PO1jv3RR7WO/bPPtI591Sr6exp6dLRq6D16qIb+wAOqoX/yiWroy5bx+nkaeoMGqqF366Ya+l13qYZuGKEM9qKh1/qk/L+8bEI3PHw+kcRE2oEAJyKPDRv4mQi/4/PRTk9nwxIRkaIiNmER4aIoXiMXESakef5btnBiEWHTmKws2oWFbL4iwom4ur9HQoKI3087NZWLjohwEZnt29U/Lm73/ps384eCCCfJ3FzaeXlsQiPCiXPTJvXxfgyI8POyMtrJyfQT4X5SUmiXlekPgOr+cXHcvwjHm59POyeH5yPC8/MWk6k+/rg4np8Ir1dBAe3sbP74EeH19RrhVPffuJH3R4Q/MoqKaGdm8n4YRqiztwndNHTDMAzDqCOYhm4YhzmBgOrJgGrbe7NLS1VPr6xUPflQ+fv9qsdXVGhugmEYtYNN6IZxGPDaa8AVV9D+6SfgmGM4ySYns5TNa5RywgnAN9/QvuEG4LnnaA8aBFxzDe3vvwfatqWdmEj/hAS+P+44YOJE2n37Ai+9RPv554Hr3WWSxo8HOnWivWkT/ZOSmMzWpg3w44/cdtVVwH/+Q/sf/wBuuumgXhLDMPYTe+RuGIcB6enM2u7alZnjy5cDF17IbbNnA336sHZ98WKgSxfWX2/cyLrztm2ZFZ6TA5x+Ov+KXrUKOP98TsJz5gCXXMLa85gYoFs31muvX89a7WOPZfZ7fj5w2mnMOF+7FjjvPPW/9FKOZdEi4OyzpXtBmQAACatJREFUWX++di1rwo85htnnPh9w6qm1dgkNo15gdeiGYRiGEQKYhm4YhmEYIY5N6IZxGPDWW8CVV9KeNYtatdfgpUULXTjmxBOBSZNo33gj8OKLtF9/nZo4AEybBrRrRzspif5bt/J9+/bA1Km0r71WNfBBg6jJA8Dkyaqhb9lCf68hzLHHAjNnctvVVwP//S/t558H+vU7qJfEMIz9xBYZNIzDgL/8BejRg3b37sCHH7JP/DHH0PYm2HffBXr1ov3009rv/KabgIsuot2rFzBsGO22bWkfdxzff/ABtXUAePZZoHVr2rfcoj8oLrwQGDqUdvv29G/Thhr8sGHAuedy28CBmnx3++3adc8wjNrBNHTDMAzDqCOYhm4YhmEYIY5N6IZxGPDxx3zsDQALFwJnnsmmLenpQOfOqoGffz7w88+0778fePNN2h98APTvT3vOHOCcc2inpNDfW+ylRw/g119p33mnPlp/+23g3ntpz5ihj/WTkuiflkYN/ayzgPnzue2224Dhw2kPHgw89NBBvSSGYewnpqEbxmHAeeepnn3SScADD7DuvHlz2kcfzW333MNacQC47jpdG/yCC6h3A8App3CyB7h2+f336xrm992nteI33KAaeO/e9AO4f29yb92a/t4a7A88AJx8MrfdeKNq+3366GpphmHUDqahG4ZhGEYdwTR0wzAMwwhxbEI3jMOM2FjgT3+ihp6dzcfZngZ+/fXAvHm0n3sO+OQT2qNHA089Rfu337Sve3o6/dPS+L5vX7Z/BYABA4DPPqM9ciTL2ABgwQLWqAOsg+/TB8jKooZ+9dXA0qXc9vjjwJdfHvzzNwzjwDAN3TAOM1q2BHr2pIbeqBET1I44gtvOPZe16QD7trdpQ7tTJ135rHVrTWpr3Jh248Z837OnavVduwIdOtA+/nggKor20UfzewB7xvfqxXE4Dj/3dPszzgA6dqyRS2AYxgFgGrphGIZh1BFMQzcMwzCMEMcmdMMwDMMIAWxCNwzDMIwQwCZ0wzAMwwgBbEI3DMMwjBDAJnTDMAzDCAFsQjcMwzCMEMAmdMMwDMMIAWxCNwzDMIwQwCZ0wzAMwwgBbEI3DMMwjBDAJnTDMAzDCAFsQjcMwzCMEMAmdMMwDMMIAWxCNwzDMIwQwCZ0wzAMwwgBbEI3DMMwjBDAJnTDMAzDCAFsQjcMwzCMEMAmdMMwDMMIAWxCNwzDMIwQwBGR2h7DAeM4TiaApKCPWgLIqqXhHA7Y+dv52/nXX+z868f5dxCRVrvbUKcn9Oo4jhMrIt1rexy1hZ2/nb+dv51/bY+jtqjv5w/YI3fDMAzDCAlsQjcMwzCMECDUJvQRtT2AWsbOv35j51+/sfOv54SUhm4YhmEY9ZVQ+wvdMAzDMOolITOhO47zJ8dxNjmOk+A4znO1PZ6axnGc4xzHmeM4zgbHcdY7jvOE+3kLx3F+cRxns/vf5rU91prCcZxwx3FWOo4z1X3fyXGcJW4MjHccJ7K2x1iTOI7TzHGcCY7jxDmOs9FxnPPqy/13HOcpN+7XOY7zjeM40aF+/x3H+cxxnAzHcdYFfbbb++2Q991rscZxnLNrb+T/O3s49zfd2F/jOM73juM0C9o20D33TY7jXFU7oz70hMSE7jhOOIAPAVwN4DQAtziOc1rtjqrGqQAwQEROA9ALwCPuOT8H4FcROQnAr+77UOUJABuD3v8XwLsiciKAXAD31MqoDh1DAfwsIqcC6AZei5C//47jtAXwOIDuItIFQDiAfgj9+z8awJ+qfban+301gJPc1/0APjpEY6wpRuP35/4LgC4icgaAeAADAcD9d7AfgNNdn+HuHBHyhMSEDuBcAAkikigiZQDGAbiulsdUo4jIThFZ4dqF4D/mbcHz/sL92hcArq+dEdYsjuO0A/BnAJ+67x0AlwKY4H4lZM8dABzHaQqgN4BRACAiZSKSh3py/wFEAGjoOE4EgEYAdiLE77+IzAeQU+3jPd3v6wCMEbIYQDPHcdocmpEefHZ37iIyU0Qq3LeLAbRz7esAjBORUhHZCiABnCNCnlCZ0NsC2BH0Ptn9rF7gOE5HAGcBWALgaBHZ6W5KA3B0LQ2rpnkPwLMAAu77owDkBf0PHuox0AlAJoDPXdnhU8dxGqMe3H8RSQHwFoDt4ESeD2A56tf999jT/a5v/ybeDWC6a9e3c68iVCb0eovjOE0ATATwpIgUBG8TljCEXBmD4zh9AWSIyPLaHkstEgHgbAAfichZAHyo9ng9hO9/c/CvsE4AjgXQGL9/HFvvCNX7/Uc4jvMCKEGOre2x1DahMqGnADgu6H0797OQxnGcBuBkPlZEJrkfp3uP1tz/ZtTW+GqQCwBc6zjONlBeuRTUk5u5j2CB0I+BZADJIrLEfT8BnODrw/2/HMBWEckUkXIAk8CYqE/332NP97te/JvoOM6dAPoCuE20BrtenPvuCJUJfRmAk9ws10gwIeKHWh5TjeJqxqMAbBSRd4I2/QDgDte+A8CUQz22mkZEBopIOxHpCN7r2SJyG4A5AG50vxaS5+4hImkAdjiOc4r70WUANqAe3H/wUXsvx3Eauf8feOdeb+5/EHu63z8A6O9mu/cCkB/0aD4kcBznT6Dsdq2IFAdt+gFAP8dxohzH6QQmBi6tjTEeakKmsYzjONeAumo4gM9E5LVaHlKN4jjOhQAWAFgL1ZGfB3X0bwG0B1eiu1lEqifShAyO4/QB8IyI9HUc53jwL/YWAFYCuF1ESmtzfDWJ4zhngkmBkQASAdwF/kgP+fvvOM4rAP4GPmpdCeBeUCcN2fvvOM43APqAq4qlA3gZwGTs5n67P3SGgVJEMYC7RCS2NsZ9MNjDuQ8EEAUg2/3aYhF50P3+C6CuXgHKkdOr7zMUCZkJ3TAMwzDqM6HyyN0wDMMw6jU2oRuGYRhGCGATumEYhmGEADahG4ZhGEYIYBO6YRiGYYQANqEbhmEYRghgE7phGIZhhAA2oRuGsc84jtPDXX862nGcxu6a5F1qe1yGYVhjGcMw9hPHcV4FEA2gIdhP/o1aHpJhGLAJ3TCM/cRdL2EZAD+A80WkspaHZBgG7JG7YRj7z1EAmgA4AvxL3TCMwwD7C90wjP3CcZwfwEVQOgFoIyKP1vKQDMMAEPHHXzEMwyCO4/QHUC4iXzuOEw4gxnGcS0Vkdm2PzTDqO/YXumEYhmGEAKahG4ZhGEYIYBO6YRiGYYQANqEbhmEYRghgE7phGIZhhAA2oRuGYRhGCGATumEYhmGEADahG4ZhGEYIYBO6YRiGYYQA/w+eXeKcfcruqgAAAABJRU5ErkJggg==\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "DBScan2 = DBSCAN(eps = 1.4, min_samples = 5)\n", + "DBScan2.fit(data2)\n", + "\n", + "fig = plt.figure(figsize=(8,8))\n", + "\n", + "ax = fig.add_subplot(111)\n", + "fig.subplots_adjust(top=1)\n", + "ax.set_title('DBSCAN dataset 2')\n", + "\n", + "ax.set_xlabel('x')\n", + "ax.set_ylabel('y')\n", + "\n", + "for clu in range(-1, max(DBScan2.labels_) + 1):\n", + " data_list_x = [data2[i,0] for i in range(n_samples2) if DBScan2.labels_[i]==clu]\n", + " data_list_y = [data2[i,1] for i in range(n_samples2) if DBScan2.labels_[i]==clu]\n", + " plt.scatter(data_list_x, data_list_y, s=2, edgecolors='none', c=color[clu])" + ] + }, + { + "cell_type": "code", + "execution_count": 18, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "DBScan3 = DBSCAN(eps = 4, min_samples = 20)\n", + "DBScan3.fit(data3)\n", + "\n", + "fig = plt.figure(figsize=(8,8))\n", + "\n", + "ax = fig.add_subplot(111)\n", + "fig.subplots_adjust(top=1)\n", + "ax.set_title('DBSCAN 3')\n", + "\n", + "ax.set_xlabel('x')\n", + "ax.set_ylabel('y')\n", + "\n", + "for clu in range(-1, max(DBScan3.labels_) + 1):\n", + " data_list_x = [data3[i,0] for i in range(n_samples3) if DBScan3.labels_[i]==clu]\n", + " data_list_y = [data3[i,1] for i in range(n_samples3) if DBScan3.labels_[i]==clu]\n", + " plt.scatter(data_list_x, data_list_y, s=2, edgecolors='none', c=color[clu % len(color)])" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Reachability Distance" + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "metadata": {}, + "outputs": [], + "source": [ + "from sklearn.neighbors import LocalOutlierFactor\n", + "def reachability_distance(x, n):\n", + " lof = LocalOutlierFactor(n_neighbors = n)\n", + " lof.fit(x)\n", + " kneighbors = lof.kneighbors()[0]\n", + " distances = list(max(kneighbor) for kneighbor in kneighbors)\n", + " distances = sorted(distances)\n", + " return distances" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "[]" + ] + }, + "execution_count": 20, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "plt.plot(reachability_distance(data2, 5))" + ] + }, + { + "cell_type": "code", + "execution_count": 21, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "[]" + ] + }, + "execution_count": 21, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD4CAYAAAD8Zh1EAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAATBUlEQVR4nO3de5CddX3H8fc3mxsXwy0rRkJY0BRLreWypThohwEvqTLQGXEmbVVodZix44i1HTV1yijTdsZOR63FqY3QDl5asNFKZKQWC2i1JXaD4RoukYtAwSwBEqAESM63f5xnN8d1l+zlnD3n/J73a+bMPud5fnvOd588+8lvf8/veU5kJpKk/reg2wVIktrDQJekQhjoklQIA12SCmGgS1IhFnbrjZcvX55DQ0PdentJ6kubN29+PDMHJ9vWtUAfGhpiZGSkW28vSX0pIh6captDLpJUCANdkgphoEtSIaYd6BExEBE/johrJtl2QUSMRsSW6vG+9pYpSdqfmZwUvQjYCiybYvtVmfmBuZckSZqNafXQI2Il8Hbgss6WI0marekOuXwW+AjQeIk274iIWyNiQ0QcPVmDiLgwIkYiYmR0dHSmtUqSXsJ+Az0izga2Z+bml2j2LWAoM18HXAdcMVmjzFyfmcOZOTw4OOm8eEkq1tVbHuHT/34337+nMx3a6fTQTwfOiYgHgCuBMyPiK60NMnNHZj5fPb0MOKWtVUpSAS66cgufu34b/33fjo68/n4DPTPXZebKzBwC1gLXZ+a7WttExIqWp+fQPHkqSZrgg2et5qNrXtOR1571pf8RcQkwkpkbgQ9GxDnAHuAJ4IL2lCdJZRj7dLjo4HvMKNAz80bgxmr54pb164B17SxMkkoy9mmfC6Jzke6VopI0DxpjPfQOdtENdEmaB1UHnQUGuiT1t309dIdcJKmvjY2hO+QiSX1uPNA7OM/FQJekeZB4UlSSirBv2mLn3sNAl6R5MH5S1CEXSepvY9MWHXKRpD6X1c3HnbYoSX1u7KSoY+iS1Oca49MWO8dAl6R5kF4pKkllaDhtUZLKMDaG3slpLrP+gAtJ0v7t2v0i27Y/w5PPvgB0toduoEtSB334qi18d+v28ecHLe5c7BroktRBO597kV9esYyPrjmexQML+PVjD+/YexnoktRBexvJEQct5ozjX97x9/KkqCR10N6EBZ0cOG9hoEtSB2UmA/OT5wa6JHXS3kayoJN35GphoEtSB+1tpEMuklSCTBiwhy5J/W9vJgvmKWkNdEnqoMY8jqE7D12S2mDro7v4++/9hL358+t/tms3v7rykHmpwUCXpDa49rZH+eaW/+W45Qf93Pojly3l9Fctn5caDHRJaoM9jWThguD6PzmjazU4hi5JbdCYxytCp2KgS1IbNDI7emvc6TDQJakN9jZy3uabT8VAl6Q2aOT8XRE6FQNdktpgPuebT8VAl6Q22JvJQL/00CNiICJ+HBHXTLJtSURcFRHbImJTRAy1s0hJ6nWNpK966BcBW6fY9l7gycx8NfAZ4FNzLUyS+klzyKW7NUzrwqKIWAm8HfgL4MOTNDkX+ES1vAG4NCIiM3OStpLU1+792dP82dW382LLdf4P7niWRQPdHcWe7rt/FvgI0Jhi+1HAQwCZuQfYCRwxsVFEXBgRIxExMjo6OotyJan7Rh58kpvue4IFAQcsGuCARQO85hXL+N1TV3W1rv320CPibGB7Zm6OiDPm8maZuR5YDzA8PGzvXVJfGht7+NvfOZlXHLK0u8W0mE4P/XTgnIh4ALgSODMivjKhzSPA0QARsRA4BNjRxjolqWckzUTv8jnQX7DfQM/MdZm5MjOHgLXA9Zn5rgnNNgLnV8vnVW3sgUsqUq+m26zvthgRlwAjmbkRuBz4ckRsA56gGfySVKSxPO+xDvrMAj0zbwRurJYvblm/G3hnOwuTpJ7XY4nulaKSNFPVmEv0WKIb6JI0Q+NDLr2V5wa6JM1Ur54UNdAlaYZyfMiltxjokjRD+4ZceivSDXRJmqXeinMDXZJmbGwMvcc66Aa6JM3UvguLeivRDXRJmqFevbOJgS5Js9VbHXQDXZJmyzF0Sepz4ydFu1vGLzDQJWmG9t0Pvbci3UCXpBnq0XOiBrokzVSv3g/dQJekGfLCIkkqjBcWSVKf69sPiZYk/TxPikqSOspAl6QZGv+AC4dcJKkMnhSVpD7ntEVJKoQXFklSIfb10Hsr0g10SZqhpDfnLRrokjRD3j5XkgrTYyMuLOx2AZI037bv2s3aL97EM7v3zOr7n32++X29NoZuoEuqnYee/D/uG32WM44fZMUhS2f1Gq8aPLjNVc2dgS6pdsbGwN/7hmN54+rB7hbTRo6hS6qdffPIe2vIZK4MdEm106t3S5wrA11SbfXYOc05M9Al1c743RK7XEe77TfQI2JpRPwoIm6JiDsi4pOTtLkgIkYjYkv1eF9nypWkuSt0xGVas1yeB87MzGciYhHwg4i4NjNvmtDuqsz8QPtLlKQOKayLvt9Az+bfJs9UTxdVj1L/g5NUA/su3S8r0ac1hh4RAxGxBdgOXJeZmyZp9o6IuDUiNkTE0VO8zoURMRIRI6Ojo3MoW5Jmr1c/5HmuphXombk3M08EVgKnRsRrJzT5FjCUma8DrgOumOJ11mfmcGYODw6WM5lfUp8pdIxhRrNcMvMp4AZgzYT1OzLz+erpZcAp7SlPkjqnsA76tGa5DEbEodXyAcCbgbsmtFnR8vQcYGs7i5Skdhq/UrSwMZfpzHJZAVwREQM0/wP4WmZeExGXACOZuRH4YEScA+wBngAu6FTBkjRXvfqZoHM1nVkutwInTbL+4pbldcC69pYmSZ3Rq584NFdeKSqptgrroBvokuqn1CEXA11S7ZQ54GKgS6q1srroBrqk2hm/22JZeW6gS6qffZ9YVBYDXVL9FDqIbqBLqq3SrhQ10CXVzvjdFrtcR7sZ6JJqx3noklSIdAxdkspSy08skqSS7Lt9blfLaDsDXVLtZKFjLga6JBXCQJdUOw65SFIhxqctelJUkvqdY+iSVBSHXCSpz3mlqCQVYt/tc8tKdANdUu0UOg3dQJdUX6UNuSzsdgGSyvdfP3mcL37/vp6ZW7J91/NAebfPNdAlddy/3f4Y37/3cV77ymXdLgWARQPBb/7SIEcddkC3S2krA11SxzUyOezARVz9gTd0u5SiOYYuqeMaWd7HvfUiA11Sx2UmC8zzjjPQJXVcowEL7KF3nIEuqeMamQb6PDDQJXVccwy921WUz0CX1HFpD31eGOiSOq7hSdF5YaBL6rhGelJ0PhjokjqukekY+jzYb6BHxNKI+FFE3BIRd0TEJydpsyQiroqIbRGxKSKGOlGspP6U9tDnxXR66M8DZ2bmrwEnAmsi4rQJbd4LPJmZrwY+A3yqvWVK6mdOW5wf+72XS2Ym8Ez1dFH1mHjTtHOBT1TLG4BLIyKq75XUJT/c9jiP7dzd7TJ4+MnnHHKZB9O6OVdEDACbgVcDn8/MTROaHAU8BJCZeyJiJ3AE8PiE17kQuBBg1apVc6tc0kt6eveLvOvyTT3zYQ5vXL282yUUb1qBnpl7gRMj4lDgXyPitZl5+0zfLDPXA+sBhoeHe+Qwk8r0wp4GmXDRWat5x8kru10OL1+2pNslFG9Gt8/NzKci4gZgDdAa6I8ARwMPR8RC4BBgR9uqlDRjjarLtPzgxaw64sDuFqN5MZ1ZLoNVz5yIOAB4M3DXhGYbgfOr5fOA6x0/l7orq1Nd3ra2PqbTQ18BXFGNoy8AvpaZ10TEJcBIZm4ELge+HBHbgCeAtR2rWNK0jHWpnF1SH9OZ5XIrcNIk6y9uWd4NvLO9pUmai0aV6F5yXx9eKSoVqmEPvXYMdKlQjcbYGHqXC9G8MdClQjmGXj8GulSo8TF0f8trw39qqVD7ToraQ68LA10q1NiFIM5Drw8DXSpUOm2xdgx0qVBj0xYDE70uZnQvF6mf7dnb4K7Hnu6Zuw922gM7ngXsodeJga7a+Icf3s9ffnvibYjKd+ASf83rwn9p1cau5/awIGD9u4e7Xcq8WbpogNe/6ohul6F5YqCrNpLmx6C96YQju12K1BGeFJWkQhjoqo1M72uishnoqo3EKXwqm4EuSYUw0FUb2eyiS8Uy0CWpEAa6aiNJO+gqmoEuSYUw0FUfTltU4Qx0SSqEga7acB66SmegqzayLvfNVW0Z6KoVx9BVMgNdtWEHXaUz0FUrdtBVMgNdtZFAOOaighnoklQIA121kemQi8pmoKs2Es+KqmwGuurFLroKZqCrNpy2qNIZ6KoVO+gqmYEuSYXYb6BHxNERcUNE3BkRd0TERZO0OSMidkbElupxcWfKlebGeegq2cJptNkD/HFm3hwRLwM2R8R1mXnnhHb/mZlnt79EqT28OZdKt98eemY+mpk3V8tPA1uBozpdmNQJdtBVshmNoUfEEHASsGmSza+PiFsi4tqI+JU21Ca1VfN+6FK5pjPkAkBEHAx8HfhQZu6asPlm4JjMfCYi3gZ8E1g9yWtcCFwIsGrVqlkXLc2GIy4q3bR66BGxiGaYfzUzvzFxe2buysxnquVvA4siYvkk7dZn5nBmDg8ODs6xdGnmPCmqkk1nlksAlwNbM/PTU7R5RdWOiDi1et0d7SxUmisv/VfppjPkcjrwbuC2iNhSrftTYBVAZn4BOA94f0TsAZ4D1qZTCtSD7J+rZPsN9Mz8Afv5PcjMS4FL21WU1Al2MVQ6rxRVrTiErpIZ6KoNO+gqnYGu2mgOudhFV7kMdEkqhIGuGknH0FU0A12SCmGgqzb8kGiVzkBXrTjkopIZ6KoNLyxS6Qx01Uo46KKCGeiqDW/OpdIZ6KqNTMfQVTYDXZIKYaCrNvwIOpXOQJekQhjoqo3mGLp9dJXLQJekQkznI+h6yvfuGeXPr7mz22WoDz22azfLli7qdhlSx/RdoB+8ZCGrjzy422WoD60+8mB+49gjul2G1DF9F+inHHMYpxxzSrfLkKSe4xi6JBXCQJekQhjoklQIA12SCmGgS1IhDHRJKoSBLkmFMNAlqRCRXfqgxYgYBR6c5bcvBx5vYzklcd9Mzv0yOffL1Hp13xyTmYOTbehaoM9FRIxk5nC36+hF7pvJuV8m536ZWj/uG4dcJKkQBrokFaJfA319twvoYe6byblfJud+mVrf7Zu+HEOXJP2ifu2hS5ImMNAlqRB9F+gRsSYi7o6IbRHxsW7X0wkRcXRE3BARd0bEHRFxUbX+8Ii4LiLurb4eVq2PiPhctU9ujYiTW17r/Kr9vRFxfsv6UyLitup7Phd99OnJETEQET+OiGuq58dGxKbqZ7kqIhZX65dUz7dV24daXmNdtf7uiHhry/q+PL4i4tCI2BARd0XE1oh4vcdLU0T8UfV7dHtE/HNELC32mMnMvnkAA8BPgOOAxcAtwAndrqsDP+cK4ORq+WXAPcAJwF8BH6vWfwz4VLX8NuBaIIDTgE3V+sOB+6qvh1XLh1XbflS1jep7f6vbP/cM9s+HgX8Crqmefw1YWy1/AXh/tfyHwBeq5bXAVdXyCdWxswQ4tjqmBvr5+AKuAN5XLS8GDvV4SYCjgPuBA1qOlQtKPWb6rYd+KrAtM+/LzBeAK4Fzu1xT22Xmo5l5c7X8NLCV5oF5Ls1fXKqvv10tnwt8KZtuAg6NiBXAW4HrMvOJzHwSuA5YU21blpk3ZfNo/VLLa/W0iFgJvB24rHoewJnAhqrJxP0ytr82AGdV7c8FrszM5zPzfmAbzWOrL4+viDgE+E3gcoDMfCEzn8LjZcxC4ICIWAgcCDxKocdMvwX6UcBDLc8frtYVq/qT7yRgE3BkZj5abXoMOLJanmq/vNT6hydZ3w8+C3wEaFTPjwCeysw91fPWn2X856+276zaz3R/9bpjgVHgH6uhqMsi4iA8XsjMR4C/Bn5KM8h3Apsp9Jjpt0CvlYg4GPg68KHM3NW6reop1WrOaUScDWzPzM3drqXHLAROBv4uM08CnqU5xDKujscLQHXe4Fya/+m9EjgIWNPVojqo3wL9EeDolucrq3XFiYhFNMP8q5n5jWr1z6o/f6m+bq/WT7VfXmr9yknW97rTgXMi4gGaf9qeCfwNzSGDhVWb1p9l/Oevth8C7GDm+6vXPQw8nJmbqucbaAZ83Y8XgDcB92fmaGa+CHyD5nFU5DHTb4H+P8Dq6gz1YponLTZ2uaa2q8bsLge2ZuanWzZtBMZmHpwPXN2y/j3V7IXTgJ3Vn9rfAd4SEYdVPZW3AN+ptu2KiNOq93pPy2v1rMxcl5krM3OI5r/99Zn5e8ANwHlVs4n7ZWx/nVe1z2r92mpGw7HAapon/fry+MrMx4CHIuL4atVZwJ3U/Hip/BQ4LSIOrGof2zdlHjPdOhs72wfNM/T30Dyz/PFu19Ohn/ENNP88vhXYUj3eRnMs7z+Ae4HvAodX7QP4fLVPbgOGW17rD2iewNkG/H7L+mHg9up7LqW6arhfHsAZ7JvlchzNX65twL8AS6r1S6vn26rtx7V8/8ern/1uWmZs9OvxBZwIjFTHzDdpzlLxeGnW/kngrqr+L9OcqVLkMeOl/5JUiH4bcpEkTcFAl6RCGOiSVAgDXZIKYaBLUiEMdEkqhIEuSYX4f+YoQJpP2ObuAAAAAElFTkSuQmCC\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "plt.plot(reachability_distance(data3, 20))" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, "outputs": [], "source": [] } diff --git a/anno3/apprendimento_automatico/esercizi/meo/dataset-DBSCAN.csv b/anno3/apprendimento_automatico/esercizi/meo/dataset-DBSCAN.csv new file mode 120000 index 0000000..36a0b6a --- /dev/null +++ b/anno3/apprendimento_automatico/esercizi/meo/dataset-DBSCAN.csv @@ -0,0 +1 @@ +Datasets/dataset-DBSCAN.csv \ No newline at end of file