Cluster analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sb\n",
    "from matplotlib import rcParams\n",
    "import sklearn\n",
    "from sklearn.cluster import KMeans\n",
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "from sklearn.preprocessing import scale\n",
    "import sklearn.metrics as sm\n",
    "from sklearn import datasets\n",
    "from sklearn.metrics import confusion_matrix, classification_report\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "rcParams['figure.figsize']=7,4\n",
    "sb.set_style('dark')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-0.90068117,  1.03205722, -1.3412724 , -1.31297673],\n",
       "       [-1.14301691, -0.1249576 , -1.3412724 , -1.31297673],\n",
       "       [-1.38535265,  0.33784833, -1.39813811, -1.31297673],\n",
       "       [-1.50652052,  0.10644536, -1.2844067 , -1.31297673],\n",
       "       [-1.02184904,  1.26346019, -1.3412724 , -1.31297673],\n",
       "       [-0.53717756,  1.95766909, -1.17067529, -1.05003079],\n",
       "       [-1.50652052,  0.80065426, -1.3412724 , -1.18150376],\n",
       "       [-1.02184904,  0.80065426, -1.2844067 , -1.31297673],\n",
       "       [-1.74885626, -0.35636057, -1.3412724 , -1.31297673],\n",
       "       [-1.14301691,  0.10644536, -1.2844067 , -1.4444497 ]])"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iris=datasets.load_iris()\n",
    "X=scale(iris.data)\n",
    "y=pd.DataFrame(iris.target)\n",
    "var_name=iris.feature_names\n",
    "X[:10,]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KMeans(algorithm='auto', copy_x=True, init='k-means++', max_iter=300,\n",
       "    n_clusters=3, n_init=10, n_jobs=1, precompute_distances='auto',\n",
       "    random_state=5, tol=0.0001, verbose=0)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clustering=KMeans(n_clusters=3,random_state=5)\n",
    "clustering.fit(X)c"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>sepal_length</th>\n",
       "      <th>sepal_width</th>\n",
       "      <th>petal_length</th>\n",
       "      <th>petal_width</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5.1</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.7</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4.6</td>\n",
       "      <td>3.1</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   sepal_length  sepal_width  petal_length  petal_width\n",
       "0           5.1          3.5           1.4          0.2\n",
       "1           4.9          3.0           1.4          0.2\n",
       "2           4.7          3.2           1.3          0.2\n",
       "3           4.6          3.1           1.5          0.2\n",
       "4           5.0          3.6           1.4          0.2"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iris_df=pd.DataFrame(iris.data)\n",
    "iris_df.columns=['sepal_length','sepal_width','petal_length','petal_width']\n",
    "y.columns=['Target']\n",
    "iris_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7fb9c8735810>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAagAAAEFCAYAAAC2MHPwAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nS5KghBBC+NL/B7HsiQKmZf2aAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fb9c89ccd10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "colortheme=np.array(['darkgray','lightsalmon','powderblue'])\n",
    "plt.subplot(1,2,1)\n",
    "plt.scatter(x=iris_df.petal_length,y=iris_df.petal_width,c=colortheme[iris.target],s=50)#maker_SIZE=50\n",
    "plt.title(\"Ground Truth Classification\")\n",
    "\n",
    "plt.subplot(1,2,2)\n",
    "plt.scatter(x=iris_df.petal_length,y=iris_df.petal_width,c=colortheme[clustering.labels_],s=50)#maker_SIZE=50\n",
    "plt.title(\"KMeans Classification\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7fb9c857ce10>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAagAAAEFCAYAAAC2MHPwAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nrkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fb9c869ce10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "relabel=np.choose(clustering.labels_,[2,0,1]).astype(np.int64)\n",
    "colortheme=np.array(['darkgray','lightsalmon','powderblue'])\n",
    "plt.subplot(1,2,1)\n",
    "plt.scatter(x=iris_df.petal_length,y=iris_df.petal_width,c=colortheme[iris.target],s=50)#maker_SIZE=50\n",
    "plt.title(\"Ground Truth Classification\")\n",
    "\n",
    "plt.subplot(1,2,2)\n",
    "plt.scatter(x=iris_df.petal_length,y=iris_df.petal_width,c=colortheme[relabel],s=50)#maker_SIZE=50\n",
    "plt.title(\"KMeans Classification\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "             precision    recall  f1-score   support\n",
      "\n",
      "          0       1.00      1.00      1.00        50\n",
      "          1       0.74      0.78      0.76        50\n",
      "          2       0.77      0.72      0.74        50\n",
      "\n",
      "avg / total       0.83      0.83      0.83       150\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y,relabel)) #precision measures relvancy and recall measures completeness\n",
    "#high precision +high recall =high accurate model resuts"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Hierarchical Methods"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from scipy.cluster.hierarchy import dendrogram, linkage\n",
    "from scipy.cluster.hierarchy import fcluster, cophenet\n",
    "from scipy.spatial.distance import pdist\n",
    "from sklearn.cluster import AgglomerativeClustering\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "np.set_printoptions(precision=4,suppress=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "cars=pd.read_csv('mtcars.csv')\n",
    "x=cars.iloc[:,[1,3,4]]\n",
    "y=cars.am"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAb8AAAEiCAYAAABp69zvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3XlUFFf+NvCnaUBQjAqyCGFwGRRHIKKIoriBioIKrpHR\nqDjGZeK+TKJGE+MaM1Ff4zjKYDSLJooSFFGjoHHDLQaDRM0ElYAijeLGvjT3/cMfNaCoBOguoJ7P\nOZ5jV1fX/XKrup+61VXVKiGEABERkYIYyF0AERGRvjH8iIhIcRh+RESkOAw/IiJSHIYfEREpDsOP\niIgUh+FHNVKPHj1w/vz5Sr9+0aJF2Lx5czVWVHkv+1vOnz8Pf39/nbbxKrGxsfD29q5yDdVhwoQJ\n2L9/v9xlVIugoCCEh4fLXQa9gKHcBVDFubm5Sf/Pzc2FsbEx1Go1AGDp0qUYPHiwXKWVERsbi/ff\nfx/Hjh3TyfLnzZsHBwcHTJ8+XZr2+++/o1+/fvj1118BACtWrNBJ29Wtc+fOiIqK0nk7ly9fxmef\nfYbLly9DrVbDwcEBo0ePRmBgYLW10aNHD3zyySfo3LlzpZfx+eefV1s9pcXGxmLChAkwNTUFADRs\n2BAdOnTAxIkT4ezsrJM2qWbjyK8WiYuLk/7Z2tpi8+bN0uPygq+oqEiGKmu3yvRZTe/nH3/8EcHB\nwfD09ER0dDTOnz+PJUuW4OTJk3KXJikuLkZxcbFO27C1tUVcXBx++ukn7Nq1Cw4ODggKCqrSEQY5\n1PTtrbZg+NUh69atw6xZszBnzhy4ublh//79mDdvHj777DNpnmcPcfXo0QPbtm3DoEGD0LFjR8yZ\nMwcFBQXS80eOHEFAQAA6dOiAvn374vTp0wCAsLAwDBgwAG5ubujTpw/CwsIAAJmZmZgyZQpSU1Ph\n5uYGNzc3ZGRkoLi4GJs3b0afPn3QuXNnzJ49G48fP5baCQ8PR+/evdG5c2eEhIRUuS+e/btjYmIw\nePBguLu7IygoCP/973/L9EFoaCgGDRqE9u3bAwA2bdoEHx8fuLm5wd/fHzExMdL8YWFhGD16NJYv\nXw4PDw/8+9//BgB8++236N+/P9zc3DBw4EBcv35des3Vq1fL7eNn18edO3fw97//HV26dEHnzp2x\nfPlyAEBSUhLeeusteHh4oHPnzpg/fz4yMzMr1Bdr1qzBsGHDMHHiRDRp0gQqlQouLi5Yu3btc/MW\nFRWhTZs2uH37drl9mZGRgbfffhvu7u7w8PDA6NGjAQBz5sxBeno63n77bbi5uWHbtm0AgEuXLmHk\nyJFwd3dHQEAALl68KC03KCgI69evx5tvvon27dsjNTW1zKHCsLAwjBkzBitXroS7uzt8fHyk7Q8A\nkpOTERQUBDc3N0yYMAEffPAB3nvvvVf2h0qlQrNmzTB79mwMGTIEn376qfRcYmIixo8fDw8PD/Tv\n3x/ff/99mX5Yvnw5Jk6cCDc3N7z55ptISUmRnj958iR8fX3RsWNHLF++HKVvnlVcXIyNGzeid+/e\n8PT0xHvvvYesrCzp+b1796JXr17o3LkzNm/eXOYwdnnv67i4OKlfvby8sHz5chQWFpZZhzt37kSf\nPn3g5uaGzz77DElJSRg5ciQ6dOiAOXPmSPMrlqBaqXfv3uLMmTNlpq1du1a0a9dOxMTECK1WK3Jz\nc8XcuXPFhg0bpHnOnDkjevfuLT3u3r27GDFihEhPTxcPHjwQ/fr1E7t37xZCCHHp0iXRsWNHERsb\nK7RarUhNTRU3btwQQggRExMjkpOTRXFxsYiNjRUuLi7i2rVr5bYhhBBbt24Vb775pkhLSxN5eXli\n4cKFYt68eUIIIa5fvy7at28vfvzxR5Gfny+WLVsm2rZtK86dO1fu3/7s3ySEEElJSaJ169blzvPz\nzz8LT09P8fPPP4uioiIRFhYmfHx8RH5+vtQHQ4YMEXfv3hW5ublCCCGioqKERqMRWq1W7N+/X7Rv\n317cu3dPCCHE7t27Rdu2bcWOHTtEUVGRyM3NFZGRkaJHjx7iypUrori4WNy8eVOkpqa+so9L91Vh\nYaHw9/cXq1evFtnZ2SI3N1f8+OOPQgghbt68KWJjY0V+fr64f/++ePPNN8Xq1avLrMfy+iszM1O0\nadNGXLx4sdy+LK+G1q1bi5SUlHL7cvXq1WLp0qWioKBA5OfniwsXLrywhtTUVOHh4SFOnjwptFqt\nOHHihPDw8BAPHjwQQggxatQo0bt3b5GYmCgKCgpEYWGhGDVqlNi7d6/Uz3/5y19EWFiYKCoqEl9+\n+aXo0aOHtPyhQ4eKNWvWSHW0b99evPvuu6/8G0s7deqUcHJyEnl5eSIrK0t4eXmJ7777ThQWFoor\nV64IDw8PaZufO3eu8PDwEPHx8aKgoEDMnDlTzJ07VwghxP3790X79u3FkSNHREFBgfjPf/4j2rZt\nK/0t3377rejXr59ITk4WmZmZYsqUKeK9994TQvxv+7906ZLIz88XK1asKLP9l/e+/vnnn8Xly5dF\nYWGhSE5OFv369RNfffVVmXX4zjvviMzMTHHt2jXRrl07MXbsWJGSkiIePXokfH19xb59+164TSgB\nR351TIcOHeDt7Q0DAwOYmJhU6DXjxo2DpaUlmjRpgl69euHatWsAgD179mDEiBHw9PSEgYEBmjVr\nhpYtWwIAvL29YW9vD5VKBU9PT3h6euLSpUsvbOPbb7/FnDlzYG1tjXr16mHatGk4fPgwiouLcfjw\nYfj4+KBjx44wNjbG7Nmzy+w1lyckJATu7u7SvyFDhrxw3t27d+Ovf/0rXF1doVarMXz4cADAlStX\npHnGjh0LGxsbqc/8/PxgZWUFAwMDDBo0CHZ2dkhISJDmb9asGf76179CrVbDxMQEYWFhmDRpEpyd\nnaFSqdCiRQs0a9bslX1c2uXLl/Hw4UPMmzcP9evXh4mJCTp27AgAaNGiBTw9PWFsbAwLCwuMHz8e\nFy5ceGkfAcDjx48hhIClpeUr560IIyMjpKen4+7duzA2NkanTp1eOG9ERAS8vb3RvXt3GBgYoEeP\nHnBycsKpU6ekeYYOHYpWrVrByMgIhobPn4Jgb2+P4cOHQ61WY8iQIUhLS8ODBw+QkpKCX3/9FdOn\nT5fq6NWr1x/+e6ysrFBcXIzMzEzExMSgefPmCAwMhKGhIZydneHj41Nm9Ofr6wsXFxcYGRlh0KBB\n0uj++PHjcHJyQt++fWFkZIQJEybA3Nxcel1kZCQmTJgAe3t7mJmZYc6cOYiMjJS2/z59+qBDhw4w\nNjbGrFmznqvz2fe1q6sr3njjDRgaGsLe3h4jR458bnt4++23YWZmBicnJ7Rs2RI9e/bE66+/jkaN\nGsHLy6vcbVBJeMJLHVP6A7eimjZtKv3fxMREOhyZlpYGFxeXcl9z/PhxbNq0Cb///juKi4uRl5f3\nwnkBIDU1FVOmTIGBQdn9rYyMDKSnp5epu0GDBmjUqNFLa540aVK5J7y8qO3IyEhs375dmlZYWAiN\nRiM9trGxKfOa8PBwbN++HampqQCAnJwcPHz4UHr+2X5OS0uDvb39C+t9UR+XdvfuXdjZ2UknMZV2\n7949LF++HD/99BOys7MhhECTJk1e2F6JRo0aQaVS4d69e3BwcHjl/K8yadIkbNiwAePHj4eBgQFG\njRqFiRMnljtvamoqDhw4gKNHj0rTioqK0L17d+nxq7bX0qFdsmOSk5MDjUaDxo0bl9nBs7GxKbOO\nKkKj0cDAwAANGzZEamoq4uLi4O7uLj2v1WrL7FiVrsfU1BQ5OTkA8Nw2bGBgAGtra+lxeno6bG1t\npce2trYoLCzEgwcPnntt/fr18dprr5Wp89l+unHjBj7++GP88ssvyM3NhVarhaura5l5LCwspP+b\nmJg89/jJkyev6J26jeFXx6hUqjKP69evj9zcXOnxvXv3KrwsGxsbJCcnPzc9Ly8PM2bMwLp169Cz\nZ08YGRlh8uTJ0mjt2RpKlvXpp5/ijTfeeO45S0vLMt8xZWdnlxsOlWVjY4N33nkHkyZNeuE8pWtO\nSUnBhx9+iO3bt+ONN96AWq3GwIEDy4xGn/0bbWxsynz/UxnNmjVDamoqtFrtcwH4z3/+E8bGxoiM\njETjxo1x+PBhrFmz5pXLNDMzg6urK77//vsyH+ovYmhoCGNjY+Tl5UnT7t+/LwWnmZkZFi5ciIUL\nF+LXX3/F2LFj4erqCg8Pj3L7ZOjQoVi6dOkL2ytvW6kIKysrPHr0CPn5+ahXrx6ApzsgJf+vqKNH\nj8LFxQX16tWDjY0NunTpgtDQ0D9cj6WlZZkRbXFxcZmdKysrK2lHCni6o2NkZARzc3NYWlqWeS4n\nJ+e5YHq2nz744AO88cYbWLduHRo0aICtW7fihx9++MN1KxkPe9ZxTk5OOHHiBB4/foz09HR89dVX\nFX7t8OHDsWfPHpw7dw7FxcVIS0vDzZs3UVBQgMLCQjRp0gRqtRrHjx/H2bNnpddZWFjg4cOHZb7Q\nHzVqFNauXSu9yTMyMqSTSPr374+YmBjExcWhoKAA69evr/SHYnlGjhyJnTt3Ij4+HkIIZGdn49ix\nY9Je+7Oys7OhUqlgbm4OIQR2796NmzdvvrSNESNGIDQ0FFevXoUQAklJSbh79+4fqrN9+/Zo3Lgx\n1q5di9zcXOTl5UmHkrOzs2FqaoqGDRvi7t27f+iSgH/84x/Ys2cPtm3bhkePHgF4egLO3Llzy53f\nyckJkZGR0Gq1+OGHH8oczj527BiSk5MhhEDDhg2hVqul0byFhUWZHYCAgABER0fjzJkz0Gq1yM/P\nx7lz58qEQmX96U9/QuvWrbFx40YUFBTgxx9/rPCHvxACaWlp2LBhA7777jvMnj0bAODj44PExERE\nRkaisLAQhYWFiI+Pf+W6B4DevXvj+vXriI6ORmFhIbZv344HDx5Iz/v7+2Pbtm24ffs2srKysG7d\nOgwcOBAGBgbo378/oqOjcfnyZRQUFGDDhg2vbC87OxsNGzZE/fr1cePGDezatatCfzv9D8Ovjiv5\nTqV3796YOHHiH7qgukOHDli2bBlWrFiBjh07YuzYsUhLS8Nrr72GBQsWYNq0afDw8MD3339f5vuW\n1q1bo1+/fvDx8YG7uzsyMjIQHByM7t27Y/z48XBzc8OoUaOk79ycnJywaNEizJo1C927d4elpWWZ\nw4RV1b59e3z44Yf48MMP0alTJ/j6+r70QmonJyeMGTMGI0aMgJeXF27duvXcIaVnDRw4EBMnTsTs\n2bPRoUMHTJ8+/Q+PXg0NDbF582bcuHEDvXr1Qq9evaTvm6ZPn44rV67A3d0dU6dOfeEh3vK4u7tj\n27ZtOH36NLy9veHh4YEPP/wQPXr0KHf+999/H0ePHoW7uzsOHz5c5mzUW7duYdy4cXBzc0NQUBDG\njh0rjSinTJmCzz77DO7u7ti+fTtef/11bNy4EZs2bYKnpyd69eqFbdu2vfL73Ipau3YtLl68iM6d\nO+Nf//oX/Pz8YGxs/ML5S5+BPGLECCQmJmLHjh3w9PQE8PTav61bt2L//v3w8vKCl5cXPv300zJn\nP79I06ZNsW7dOnzyySfo0qULUlNTy2wzI0eOhJ+fH0aPHo0+ffqgQYMGWLRoEYCn29uCBQswc+ZM\ndO/eHY0bN0bjxo1f+re8++67+O6779ChQwcsWbIEAwYMqGi30f9RieraEomIZDR9+nQ4OTnhnXfe\nkbuUKsnKyoK7uzuOHz9eqe/wqWI48iOiWik+Ph4pKSkoLi7GiRMncPz4cfTp00fusiolJiYGubm5\nyM7OxurVq/GXv/yFwadjPOGFiGql9PR06fCytbU1li9fjjZt2shdVqVER0fj3XffBYAX3oCAqpfO\nDnsuWLAAP/zwAywsLHDgwAEAwKNHjzB79mzcuXMHdnZ2WL9+vXRK+5YtW7Bnzx4YGBjg/fffl06H\nTkhIwIIFC5CXl4eePXti0aJF1XoyBBERKY/ODnsOHTr0uVOGQ0JC4OnpiSNHjsDT01O6jVViYiKi\noqIQFRWF0NBQLF26FFqtFgDw4YcfYtmyZThy5AiSkpJq1P0IiYiodtLZYc9OnTqVuXYLeHpcu+RU\n+8DAQLz11luYP38+YmJi4O/vD2NjY9jb28PBwQHx8fGws7NDVlaWdL/FwMBAxMTEoGfPns+1d+9e\nxe5zSEREymFp2bDc6Xo94SUjIwNWVlb/V5AlMjIyADy9y0LpO2xYW1tDo9E8N93GxqZarhEiIiJl\nk+1sT5VKxe/uiIhIFnoNPwsLC6SnpwN4eqZWyY1fra2tkZaWJs2n0WhgbW393PS0tLQy98sjIiKq\nDL2Gn7e3NyIiIgA8veO7j4+PND0qKgoFBQVISUlBUlISXF1dYWVlBTMzM1y+fBlCiDKvISIiqiyd\nXeowZ84cXLhwAQ8fPoSFhQWmT5+OPn36YNasWbh79y5sbW2xfv16NG7cGADw73//G3v37oVarcbC\nhQulk1quXLkiXerQo0cPLF68uNzDpTzhhYiInvWiE17qzO3NGH5ERPSsGnG2JxERUU3A8CMiIsVh\n+BERkeLwxta1xJdfGiE8nKuLqLKGDi3C2LGFcpdBNQRHfrVEeLghEhLUcpdBVCslJKi580hlcGuo\nRZydtYiIyJW7DKJaJzDQVO4SqIbhyI+IiBSH4UdERIrD8CMiIsVh+BERkeIw/IiISHEYfkREpDgM\nPyIiUhyGHxERKQ7Dj4iIFIfhR0REisPwIyIixWH4ERGR4jD8iIhIcRh+RESkOAw/IiJSHIYfEREp\nDsOPiIgUh+FHRESKw/AjIiLFYfgREZHiMPyIiEhxGH5ERKQ4DD8iIlIchh8RESkOw4+IiBSH4UdE\nRIrD8CMiIsVh+BERkeLIEn5btmyBn58fBg4ciDlz5iA/Px+PHj1CcHAw+vXrh+DgYDx+/LjM/H37\n9oWvry9OnTolR8lERFSH6D38bt++jV27diE8PBwHDhyAVqtFVFQUQkJC4OnpiSNHjsDT0xMhISEA\ngMTERERFRSEqKgqhoaFYunQptFqtvssmIqI6RO/hZ2ZmBiMjI+Tl5aGoqAh5eXmwsrJCTEwMAgMD\nAQCBgYGIjo4GAMTExMDf3x/Gxsawt7eHg4MD4uPj9V02ERHVIXoPv8aNG2PChAno3bs3vLy8YGZm\nBi8vL2RkZMDKygoAYGlpiYyMDACARqOBjY2N9Hpra2toNBp9l01ERHWI3sMvOTkZ27dvR0xMDE6d\nOoXc3Fzs27evzDwqlQoqlUrfpRERkULoPfwSEhLg5uYGc3NzGBkZoV+/foiLi4OFhQXS09MBAOnp\n6TA3NwfwdKSXlpYmvV6j0cDa2lrfZRMRUR2i9/Br0aIFfv75Z+Tm5kIIgbNnz6JVq1bw9vZGREQE\nACAiIgI+Pj4AAG9vb0RFRaGgoAApKSlISkqCq6urvssmIqI6xFDfDbZt2xYBAQEYNmwYDAwM0LZt\nW7z55pvIzs7GrFmzsGfPHtja2mL9+vUAAEdHRwwYMAB+fn5Qq9VYsmQJ1Gq1vssmIqI6RCWEEHIX\nUR3u3cuUuwSdCgw0BQBEROTKXAlR7cP3j3JZWjYsdzrv8EJERIrD8CMiIsVh+BERkeIw/IiISHEY\nfkREpDgMPyIiUhyGHxERKQ7Dj4iIFIfhR0REisPwIyIixWH4ERGR4jD8iIhIcRh+RESkOAw/IiJS\nHIYfEREpDsOPiIgUh+FHRESKw/AjIiLFYfgREZHiMPyIiEhxGH5ERKQ4DD8iIlIchh8RESkOw4+I\niBSH4UdERIrD8CMiIsVh+BERkeIw/IiISHEYfkREpDgMPyIiUhyGHxERKQ7Dj4iIFIfhR0REisPw\nIyIixWH4ERGR4sgSfk+ePMGMGTPQv39/DBgwAHFxcXj06BGCg4PRr18/BAcH4/Hjx9L8W7ZsQd++\nfeHr64tTp07JUTIREdUhsoTfihUr0L17dxw+fBj79u1Dq1atEBISAk9PTxw5cgSenp4ICQkBACQm\nJiIqKgpRUVEIDQ3F0qVLodVq5SibiIjqCL2HX2ZmJi5evIjhw4cDAIyNjfHaa68hJiYGgYGBAIDA\nwEBER0cDAGJiYuDv7w9jY2PY29vDwcEB8fHx+i6biIjqEL2H3+3bt2Fubo4FCxYgMDAQixYtQk5O\nDjIyMmBlZQUAsLS0REZGBgBAo9HAxsZGer21tTU0Go2+yyYiojpE7+FXVFSEq1evIigoCBERETA1\nNZUOcZZQqVRQqVT6Lo2IiBSiwuF3584dxMbGAgDy8vKQlZVVqQZtbGxgY2ODN954AwDQv39/XL16\nFRYWFkhPTwcApKenw9zcHMDTkV5aWpr0eo1GA2tr60q1TUREBFQw/Hbv3o0ZM2ZgyZIlAIC0tDS8\n8847lWrQ0tISNjY2uHnzJgDg7NmzaNWqFby9vREREQEAiIiIgI+PDwDA29sbUVFRKCgoQEpKCpKS\nkuDq6lqptomIiADAsCIz7dixA2FhYRg5ciQAoHnz5njw4EGlG128eDHmzZuHwsJC2NvbY9WqVSgu\nLsasWbOwZ88e2NraYv369QAAR0dHDBgwAH5+flCr1ViyZAnUanWl2yYiIqpQ+BkbG8PY2Fh6XFRU\nVKVG27Zti/Dw8Oemf/HFF+XOP3XqVEydOrVKbRIREZWoUPh16tQJmzdvRl5eHs6cOYOdO3fC29tb\n17URERHpRIW+85s3bx7Mzc3RunVr7Nq1Cz179sSsWbN0XRsREZFOVGjkl5eXh2HDhknf+Wm1WuTl\n5cHU1FSnxREREelChUZ+48ePR15envQ4Ly8PwcHBOiuKiIhIlyoUfvn5+WjQoIH0uEGDBsjNzdVZ\nUURERLpUofAzNTXFL7/8Ij1OSEiAiYmJzooiIiLSpQp957dw4ULMnDkTVlZWEELg/v37WLduna5r\nIyIi0okKhZ+rqysOHTqEW7duAQBatGgBIyMjnRZGRESkKxUKPwC4cuUK7ty5A61Wi6tXrwKA9BNE\nREREtUmFwm/+/PlISUmBk5OTdGsxlUrF8CMiolqpQuGXkJCAgwcP8meGiIioTqjQ2Z6Ojo64d++e\nrmshIiLSiwqN/B4+fAh/f3+4urqWOdFl8+bNOiuMiIhIVyoUftOnT9d1HURERHqjEkIIuYuoDkM3\nnpa7BJ1KSHh6hNrZuVjmSohqH75/lCt8mle50ys08rt8+TKWLVuGmzdvorCwEFqtFqampvjpp5+q\ntUgiIiJ9qNDIb+jQoVi3bh1mzpyJvXv3IiIiAklJSZg7d64+aqyQe/cy5S5BpwIDn/6CRkQE76lK\n9Efx/aNclpYNy51eobM9AcDBwQFarRZqtRrDhg3DqVOnqq04IiIifarQYU9TU1MUFBSgbdu2WLNm\nDaysrFBczGPnRERUO1Vo5LdmzRoIIbBkyRLUr18fd+/excaNG3VdGxERkU5UKPyio6NRr149mJmZ\nYdq0aViwYAGOHz+u69qIiIh0okLhFxER8dy07777rtqLISIi0oeXfud34MABHDhwALdv38aUKVOk\n6dnZ2WjUqJHOiyMiItKFl4afm5sbLC0t8fDhQ0yYMEGa3qBBA7Rp00bnxREREenCS8PPzs4OdnZ2\n2LZtG0xMTGBgYIBbt27h5s2baN26tb5qJCIiqlYV+s5vzJgxyM/Ph0ajwd/+9jfs27cP7733nq5r\nIyIi0okKhZ8QAqampjhy5AiCgoKwYcMGJCYm6ro2IiIinahw+MXFxSEyMhK9evUCAF7kTkREtVaF\nwm/hwoXYsmUL+vTpA0dHR6SkpKBz5866ro2IiEgnKnR7Mw8PD3h4eEiP7e3t8f777+usKCIiIl16\nafitWLECixYtKnONX2n8JXciIqqNXhp+AQEBAFDmGj8iIqLa7qXh5+zsDODpYc8HDx4AAMzNzXVf\nFRERkQ698ju/zz77DF9//TWEEBBCQK1WY8yYMZg2bZo+6iMiIqp2Lz3bc9u2bfjpp5+wZ88eXLhw\nARcvXkRYWBji4uKwffv2KjWs1WoRGBiIyZMnAwAePXqE4OBg9OvXD8HBwXj8+LE075YtW9C3b1/4\n+vryR3SJiKjKXhp++/btw6effgp7e3tpmr29PT755JNyf+nhj/jyyy/RqlUr6XFISAg8PT1x5MgR\neHp6IiQkBACQmJiIqKgoREVFITQ0FEuXLoVWq61S20REpGwvDb+ioqJyv+MzNzdHUVFRpRtNS0vD\nDz/8gOHDh0vTYmJiEBgYCAAIDAxEdHS0NN3f3x/Gxsawt7eHg4MD4uPjK902ERHRS8PPyMioUs+9\nysqVKzF//nwYGPyv+YyMDFhZWQEALC0tkZGRAQDQaDSwsbGR5rO2toZGo6l020RERC894eX69evo\n0KHDc9OFECgoKKhUg8ePH4e5uTmcnZ1x/vz5cudRqVRQqVSVWj4RvZzJl9tQLzxM7jL0yjBhAwCg\nUeAMmSvRr/yhI5A3NljuMmqkl4bftWvXqr3Bn376CceOHcPJkyeRn5+PrKwszJs3DxYWFkhPT4eV\nlRXS09Olw63W1tZIS0uTXq/RaGBtbV3tdREpRb3wMBgmXEGRs4vcpejNUWdlhR4AGCZcAQCG3wtU\n6PZm1Wnu3LmYO3cuAOD8+fP4/PPP8c9//hMff/wxIiIiMGnSJERERMDHxwcA4O3tjblz5yI4OBga\njQZJSUlwdXXVd9lEdUqRswseRxyUuwzSoUaBfnKXUKPpPfxeZNKkSZg1axb27NkDW1tbrF+/HgDg\n6OiIAQMGwM/PD2q1GkuWLIFarZa5WiIiqs1UQgghdxHV4d69TLlL0KnAQFMAQERErsyVUG1XMiLg\nyK9u43p+ytKyYbnTK/STRkRERHUJw4+IiBSH4UdERIrD8CMiIsVh+BERkeIw/IiISHEYfkREpDgM\nPyIiUhyGHxERKQ7Dj4iIFIfhR0REisPwIyIixWH4ERGR4jD8iIhIcRh+RESkOAw/IiJSHIYfEREp\nDsOPiIgUh+FHRESKw/AjIiLFYfgREZHiMPyIiEhxGH5ERKQ4DD8iIlIchh8RESkOw4+IiBSH4UdE\nRIrD8CP5MYZsAAAbZElEQVQiIsUxlLsAIqK6xuTLbagXHiZrDYYJ8QCARoF+stWQP3QE8sYGy9b+\ny3DkR0RUzeqFh8Ew4YqsNRQ5u6LI2VW29g0Trsi+A/AyHPkREelAkbMLHkcclLsM2cg54qwIjvyI\niEhxGH5ERKQ4DD8iIlIcvYff3bt38dZbb8HPzw/+/v744osvAACPHj1CcHAw+vXrh+DgYDx+/Fh6\nzZYtW9C3b1/4+vri1KlT+i6ZiIjqGL2Hn1qtxnvvvYeDBw9i165d2LlzJxITExESEgJPT08cOXIE\nnp6eCAkJAQAkJiYiKioKUVFRCA0NxdKlS6HVavVdNhER1SF6Dz8rKyu0a9cOAGBmZoaWLVtCo9Eg\nJiYGgYGBAIDAwEBER0cDAGJiYuDv7w9jY2PY29vDwcEB8fHx+i6biIjqEFm/87t9+zauXbuGN954\nAxkZGbCysgIAWFpaIiMjAwCg0WhgY2Mjvcba2hoajUaWeomIqG6QLfyys7MxY8YMLFy4EGZmZmWe\nU6lUUKlUMlVGRER1nSwXuRcWFmLGjBkYNGgQ+vXrBwCwsLBAeno6rKyskJ6eDnNzcwBPR3ppaWnS\nazUaDaytrau1ni9/2Ybw32runQgAIOH+BgBAYMQMmSt5uaGOIzC2Xc28nRERUQm9j/yEEFi0aBFa\ntmyJ4OD/fUh6e3sjIiICABAREQEfHx9pelRUFAoKCpCSkoKkpCS4ulbvLXvCfwtDwn15b0X0Ks7z\nZsB5Xs0OvoT7V2r8TgQRESDDyO/SpUvYt28fWrdujYCAAADAnDlzMGnSJMyaNQt79uyBra0t1q9f\nDwBwdHTEgAED4OfnB7VajSVLlkCtVld7Xc5NXRARqNxbEVWHwIiafTsjIqISeg8/d3d3/Prrr+U+\nV3LN37OmTp2KqVOn6rIsIiJSEN7hhYiIFIfhR0REisPwIyIixWH4ERGR4qiEEELuIqrD0I2nK/3a\nksscnJu6VFc5isR+rB1KfmG8yJnrSVfYxzWnD8KneZU7nSM/IiJSnDoz8rt3L7PSry25Po3X+VUN\n+7F2aBT4dD09juB60hX2cc3pA0vLhuVOl+X2ZkREVLOZfLkN9cIrf8cmw4Snv75TEoKVlT90BPLG\nVv8tExl+RDVIVT9wKqK6PpReRlcfWKQ/9cLDYJhwpdLf2RU5V/02lCXfGzL8iOq4qn7gVER1fCi9\njC4/sEi/ipxdZD1sqcsdNIZfHVETfpki4f7TEYWc9/isC78qIfcHTlXp8gOLqLow/OqIkl+mkPMy\nA+emuh1RvErJpRa1PfxIfjXh+y4eOtYthl8dovRfpuCvSlB1kfv7Lh461j2GHxFROeQ8/MxDx7rH\ni9yJiEhxGH5ERKQ4DD8iIlIchh8RESkOw4+IiBSH4UdERIrDSx2oxqjqXWqq6w4zdeEuMUT0cgw/\nqjGqepea6rjDDO8Sw7ubkDIw/KhGkfsuNbxLDO9uQsrA8KsheMiPahLe3YTqOp7wUkOUHPKrLOem\nrlU+7Jdw/4rsvwxBRKQPHPnVIDzkR0SkHxz5ERGR4jD8iIhIcRh+RESkOAw/IiJSHIYfEREpDsOP\niIgUh+FHRESKw/AjIiLFYfgREZHi1JrwO3nyJHx9fdG3b1+EhITIXQ4REdVitSL8tFotPvroI4SG\nhiIqKgoHDhxAYmKi3GUREVEtVSvCLz4+Hg4ODrC3t4exsTH8/f0RExMjd1lERFRLqYQQQu4iXuXw\n4cM4deoUVqxYAQCIiIhAfHw8lixZInNlRERUG9WKkR8REVF1qhXhZ21tjbS0NOmxRqOBtbW1jBUR\nEVFtVivCz8XFBUlJSUhJSUFBQQGioqLg7e0td1lERFRL1YofszU0NMSSJUswceJEaLVaDBs2DI6O\njnKXRUREtVStOOGlriooKEBWVhbMzc3lLoWISFFqxWHPuig7OxuDBw/mBfvl4P4Y1QTcDus2hl85\niouLdbr8rKwsDBs2DElJSYiMjER8fLxO26sMXffBy6hUKgDyf/jI2Qc1pQa529d3DaW3uZqyHQLy\nrwe529dFDeoPP/zww2pdYi20cuVK5OTkoLi4GBYWFmU2+pL/V5esrCwMHjwYf/rTnzB79mwcOXIE\nVlZWcHd3h1arhYGBPPsj+uyDZ2VnZ2PlypUICwvD/v37odVq0aRJE5iZmaG4uFjn7ZeQsw9qSg1y\nty9nDdnZ2Vi8eDG+/fZb7N69Gzk5OTA3N0ejRo30uh0C8q8HudvXRw2KD7/U1FTExcXhu+++Q3R0\nNOLj49GyZUuYmZlBrVZX68ouHXxr1qxB+/btcefOHYSHhyMwMBBmZmbV0s4fpc8+eFZOTg6GDh2K\nhw8fwtraGhkZGYiMjMT58+fh4uICS0tLvXzwyNkHJe7evStrDXK3L2cNubm5GDZsGHJzc+Hs7Izs\n7GycOHEChw8fhrOzM2xsbPQWgHKvB7nb11sNgoQQQiQnJ4ujR4+K3r17C39/f7Fp0yaRmZkphBCi\nuLi4ysvXarVizJgxYty4cUKj0UjLPHfunOjWrZtYs2aNKCgoqHI7VaHrPijPunXrxODBg8Xdu3el\naVu2bBHt27cXnTt3FteuXRNCPO0/fZCjD2paDXK3L0cNYWFhYtCgQSI1NVWadvToUTFu3Djh4uIi\nzp8/L4TQ33YohPzrQe72dV2DYkd+jx8/homJCYqKimBgYIBGjRqhZcuWGDlyJH755RecOXMGV69e\nhaenJ+rVq1flPQ2VSoVOnTph0KBBaNq0qbQsW1tbXLx4EXFxcQgMDISJiYne9jBjYmJQVFQECwsL\nANB5H5Tn4MGDyM7ORlBQEFQqFVQqFTp27IjMzEycPXsW33//Pbp16wYrKyudtK/v7aAi5K5B7vbl\nqOHixYuIjY3FW2+9BVNTUwBAy5Yt4eTkhNTUVPzrX/9Ct27d9DoClHs9yN2+rmtQZPj97W9/Q1hY\nGHr06IHXXntN+q5Nq9XC2NgYPXv2xJMnTxAbG4sbN27A09MTxsbGVW63UaNG0hsLePoFroGBAdq0\naYMvvvgCQgh06dJFL2+syZMn48SJEzA3N0fr1q2hVqsBQOd9UKLkA+T48eO4desWxo4dC5VKhYKC\nAqjVajRo0ACxsbGwsbHBuXPn0KNHD9SvX7/a2gfk2w5KO378OE6ePImkpCRotVpYWVkBeHoZjImJ\nic5r2L17N/bv34+ff/4ZWq0W9vb2AIDCwkLUq1evzvdByTq/efMmYmNj0a9fP1hYWKCwsBBqtRpN\nmzaFo6Mjbt26hb1796JHjx5o0qRJtbRdmtzrQe72Af1vB4oLv+TkZKxZswZCCMTFxcHd3V36Qlut\nVqO4uBiGhoZwc3NDRkYGzpw5gwYNGuAvf/lLtYdSyfKMjY2RlJSEH3/8Ed26dUPjxo2rtZ1nzZw5\nE7///jvee+89eHl5SaEihICBgQEKCwthbGys0z4o+XBp1qwZvvjiC9y8eRN9+/aVQjghIQGXLl1C\nQEAAzp8/jw4dOsDW1rba9jZrwnbw97//HREREbh48SK+//57XL58GU2aNEGrVq30UsOkSZMQHR2N\n9PR0xMXFITY2Fvb29mjevHmd74OSna+cnBwYGxvDyckJ33zzDa5evYrBgwdDrVaXCcDGjRvj7Nmz\nAIBOnTpV66hH7vUgd/uAPNuB4i51aNiwIaysrNCmTRsUFBTg/fffx927d6U9fgMDAxQXF8PIyAjv\nvPMOmjVrhm+++QZ5eXkAdHPac8OGDTF8+HD89ttvuHDhgs7aAYDffvsNN2/exPz58+Hu7o6GDRsi\nKysLWVlZePToEQDAyMgIWq222vsgJycHISEhmDt3Lj744AMcPHgQTk5OmDNnDo4cOYIpU6bg4MGD\niIiIwPz589GjRw8EBwejadOmOHr0KABU25tN7u1g7ty5SElJwccff4zjx48jNDQUxsbG2LVrF548\neSLtiOiqhhkzZuDOnTv45z//iX379iEkJASmpqZlLrsp2RGqa32QnZ2NVatWYdy4cZgyZQp27NgB\nlUqFBQsW4OLFi1iwYAGAp++DwsJCAEDPnj3Rtm3bat8O5V4PcrcPyLcdKGrkJ4SAqakpMjIy8Prr\nr6NDhw44c+YMzp07h06dOqFRo0YA/rdhq9VqeHl5ITQ0FMXFxejUqZPODkna29vj1q1bOHDgAPr0\n6YPXXntNJ+3897//RUREBObOnYsGDRrg+vXrePfdd7F161ZERUUhLS0NnTp1gqGhoRSA1dEHWVlZ\nGDVqFH7//XcUFRUhLS0NX331FaytrREUFITXX38dx44dQ2RkJI4dO4YhQ4Zg3rx5MDQ0xKFDh9Cw\nYUP07NmzWvpA7u3g2LFjOHDgAGbNmoWuXbtCrVbDzs4OJiYm2Lp1K/z8/GBpaSnVUJ3rAQD27t2L\nU6dOYd68efDw8IBKpYKFhQV++uknmJqaomXLlsjOzpbOrCsoKICxsXGd6IOsrCyMGDECT548QbNm\nzWBkZIRt27bB3t4e/fv3h5GREfbs2YNbt27Bx8dHOhIBPD0j+Nq1axgyZAgMDat+Z0i514Pc7QPy\nvhdqxb09qyovLw8mJiZSB9WrVw8nTpzArl27oNVqsWPHDixevBj/7//9P8TGxsLKygpubm7S9SVj\nx47FlStXkJ+fDyMjI51di+fr64uoqCjExsZi+PDh1Rq0JX1gbm4OrVYLjUaD3NxcvPXWW/D09ETH\njh2RnZ2Nzz//HMnJyfj4449hbGyMwsLCKvdBQUEBZs+eDUtLS3zwwQdwcHBAamoqNm3ahH//+9/w\n9PTEwIEDMXDgQCQlJaG4uBgtW7YEAKnO5s2bA6jaNT41YTsoLCxEUVER7t27B2dnZ2mP1sDAAJ07\nd4alpSVu3ryJtm3bStNLDvtURw1arRYODg7o0qUL3Nzcyrz+xo0biI+PxxdffAEbGxv4+/tjxowZ\n1bYdyN0H+fn5mDx5MqytrbFs2TLY29sjMzMTRUVFOHv2LAICAjBy5EgIIbBp0yY8ePAAixYtgqWl\nJYqKinD58mWYm5tXy/tS7vUgd/uA/O+FOn+pw+rVq8W5c+eEEEIUFhYKIYRITU0VI0eOFA8fPhRC\nCPH111+LgIAA0bt3b9GmTRtx+fJlaV4hhIiPjxfu7u7i+vXrOq93+vTpIjExsVqXWboPkpOTxcCB\nA8WGDRvEtm3bxPjx48X9+/eFEE9P4/7hhx+Eq6urWLNmTZllVKUPfvjhBxEQECCOHj1a5lTxU6dO\nCRcXF3HmzJlyX5eUlCQWLFggunbtKpKTk/9wu6XVhO1g9erV0inzN2/eFEKUPV07Pz9f+Pj4iM2b\nN79wGVWpoXT7T548KfPcmDFjhJeXl/jmm29EZGSkWLlypWjTpo34z3/+U23tP1uDvvvg4MGD4q9/\n/as4efJkmTaXL18uFi9eLM6dOyfi4uJEVlaWOHTokOjTp4/o1q2b8PPzE6NHj662zwC514Pc7T9b\ngxzvBSGEqPMjv8TERDx69AidO3eWDlWYmZkhPT0dhw4dQlBQEEaPHo1jx44hMTERLVu2hLm5uXTY\nT61Ww8XFBUOHDq32sw1LE/83otmwYUO1L7t0H9jb22PgwIHYuHEjHB0d4eDgIF3qIIRA9+7dMX78\neBw7dgwTJkxAkyZNYGBgUKU+sLe3R6NGjeDp6Vlm787T0xM2NjZISEhA165dpf4GgAsXLmD79u24\nfv06QkNDpbPPqqMP5NoOEhMT8fDhQ3h4eKBFixYAnr+FVoMGDZCVlSW9pqioCImJiXBycgKAKtVQ\nuv2GDRtK09PT02Fvb493330X7dq1g0qlgqenJ9LS0nD8+HG8+eabMDMzg0qlqtV90KlTJ2RnZ6Nz\n585Sm1lZWThx4gTy8vJw6NAhZGdnIyAgAPPnz0dUVBR27tyJBw8eoEGDBli+fLl0BKIq5F4Pcrf/\nbA1yvBdKGqqTSvYiDhw4IMaMGSNSUlKEEP/b61+1apX45JNPhBBCrF27VrRr107MmzdPBAYGilGj\nRok7d+6UWV5+fr4eq68ez/bB77//Lj23ZMkS0aZNG9GnT5/nRpqff/658PLykkZEJarSBzk5OUKI\n5y8SHjhw4HOjzBIHDhyQ1ltl1YTt4Nkabt++XWZ6aRMmTBCLFy8WQgiRm5srNm/eLFxdXUVycrI0\n/x+t4UV9ULr98m6wsGTJEjFgwACRl5dXZnpt7INn69BqtaKoqEh4eXmJYcOGiQsXLojExERx6NAh\n4eLiIpYtW1ap5VekbbnWg9ztl1eDXNuBEELU2bM9S/YiOnTogJs3b2Lfvn0AIO31Ozg44PTp01i1\nahW2bt2KlStXYvXq1RgyZAiePHmC+/fvl1ledV/Tog/P9kFkZKT03NSpUzFy5EikpKTgq6++wo0b\nNwA83RNOT0+HnZ3dczeSrUoflFzfWHJcXqvVAng6+io5Y6uk/ejoaACAv78/Xn/99Uq3CdSM7eDZ\nGiIiIspML83Q0BCPHz8GAGzfvh3r16/HypUry4x8/2gNL+oDlUol7WUbGRmVec2DBw/w+PFj6eQn\nUepsutrYB8/WUfL9UVBQEDZs2IBOnTqhVatW6N+/PyZMmIDo6GikpaVJ7wFRDWdfy70e5G6/vBrk\n2g4A1N2RX2lfffWVaN++vTh+/Lg0LTExUfj6+or27duLsLCwMnsQJXsjdUlJHxw7dkya9vjxY7F6\n9WrRtm1bMWTIEDFr1iwxffp00a5dO3H06FG91DV58mQxc+ZMIcTT7x8WLFgg2rRpI9LT06u9rZqw\nHZRXgxD/GxFPmTJFTJ06VezYsUO0adNGREVFCSGe7hlXxy2lXtR+6WU/efJEbNq0SXTt2lWcPXu2\nym1WtAZ99UHptsqzcuVKERgYKIqKiqqlrfLIvR7kbv9lNehrO1DEpQ4ODg64du0aTp8+jRYtWsDW\n1haNGzdGgwYNMHToUPj4+KBevXrSha+6usxATiV9cObMGTRv3hy2traoV68evLy88Oc//xlGRkb4\n7bff0KxZM8yYMQO9e/fWy+2Ljhw5gpycHPj6+mLVqlU4dOgQvvnmG+l7gOpUE7aD8moA/vedb1xc\nHPbv34/Tp0/jk08+gb+/v7S3XR3r4kXtlyz722+/xa5du7B371589NFH6NWrV7VvB3L3wbPLKf33\naTQaHDhwAA4ODujevTsMDAx08h6Qez3I3f7LatDXdqCI8DMxMUHbtm1x+PBh/Pzzz7CxsYG9vT2c\nnJzw+uuvS0Nnff5kib492we2traws7MDAPz5z39Gt27dEBAQAB8fHzRv3rzaP2yeVRIwJ06cQGZm\nJq5cuYJ9+/Zh586daNeunU7arAnbwbM12NnZwc7OTmrz119/xdmzZ7Fp0yb4+vpW+3p4UfvA01PP\nz58/j8TERCxcuFBnO0By90Fppf++5ORk/Otf/8L58+fx0UcflbkHb3WTez3I3f7LatDbdlDlsWMt\n8uuvv4qRI0eKgQMHisjIyDLPlT6lvS4r3QcHDhwo85wuD/O8yKeffiratGkjOnbsKBISEvTSZk3Y\nDl62HkpO3a7Ow3x/pP179+7pvP1X1aCPPiht7dq14m9/+5vo2bOn9Esi+iD3epC7/VfVoMvtQBEj\nvxIWFhbo0qULkpKSsGHDBty/fx8GBgZwcHCQ7Udk9a2m9UGjRo0QHR2NnTt3ok2bNnppsyb0wbM1\n3Lt3DwYGBmjevDmaNm0qzaerkcfL2i85dbzkVzZ0Re4+KK1Ro0a4f/8+Fi9ejD//+c86b6+E3OtB\n7vZfVYMutwOVEDq6iWQNVlBQgDNnzmDLli0oKCiASqXCxIkT4evrq5gQrEl9UHLnFX2rCX0gdw1y\nt19TagBQ5jpTfZO7D+RuX44aFBl+JTIzM/Hw4UPs3bsXw4cPr/KF1LUR+6Bm9IHcNcjdfk2pQW5y\n94Hc7euzBkWHHxERKZMyjvERERGVwvAjIiLFYfgREZHiMPyIiEhxGH5ERKQ4DD8iHbt37x5mz56N\nPn36YOjQoXj77bdx69Yt3L59GwMHDqzUMsPDw6HRaKpU1/379zF58mQMHjwYfn5+ePvttwE8vb/l\njBkzqrRsopquzv+YLZGchBCYNm0aAgMDsW7dOgDA9evXkZGRARsbm0ov97vvvoOjoyOsra0r/Jqi\noiLpp5wAYMOGDejatSvGjRsn1QUA1tbWOvlRZaKahOFHpEPnzp2DoaEhgoKCpGklv0R9+/ZtaVp4\neDgSEhKwZMkSAMDkyZMxYcIEuLu7Y9GiRUhISIBKpcKwYcNgY2ODhIQEzJs3DyYmJti1axcSExOx\nevVq5OTkoEmTJli1ahWsrKzw1ltvwcnJCZcuXcLAgQMxYcIEqc309HR069at3LqmTJmCAwcOSG0D\nT0eEY8aMwbRp0xAaGopDhw6hoKAAffv25UiRah2GH5EO/fbbb1X6lYpr165JP7MDAE+ePMFrr72G\nHTt24B//+AdcXFxQWFiI5cuXY9OmTTA3N8fBgwexbt06rFq1CsDTu/SHh4c/t+zRo0dj9uzZ+Prr\nr9G1a1cMHTr0uZHkihUrAAB37tzBxIkTMWTIEJw+fRq///479uzZAyEEpk6diosXL6JTp06V/juJ\n9I3hR1SD2dvbIyUlBcuWLUPPnj3h5eX13Dy3bt3Cf//7XwQHBwN4+nNRlpaW0vN+fn7lLrt79+6I\njo7GqVOncPLkSQwZMkQK2dLy8/Mxc+ZMLF68GHZ2dvj6669x5swZBAYGAgBycnKQlJTE8KNaheFH\npEOOjo74/vvvXzmfWq1GcXGx9Dg/Px/A018b2LdvH06fPo1vv/0Whw4dkkZ0JYQQcHR0xK5du8pd\ntqmp6Qvbbdy4MQYNGoRBgwZh8uTJuHjx4nMj1Q8++AD9+vVD165dpfYmTZqEUaNGvfLvIqqpeLYn\nkQ516dIFBQUFZYLp+vXr+PHHH8vMZ2dnh+vXr6O4uBh3795FfHw8AODBgwcQQsDX1xezZs3C1atX\nAQANGjRAdnY2AKBFixZ48OAB4uLiADw9zPnbb7+9srazZ88iNzcXAJCVlYXk5GQ0a9aszDw7duxA\ndnY2Jk2aJE3z8vLC3r17pfY1Gg0yMjL+UL8QyY0jPyIdUqlU2LhxI1auXIn//Oc/qFevHuzs7LBw\n4cIy83Xs2BF2dnbw8/NDq1atpNFXeno6FixYII0K58yZAwAYMmQIPvjgA+mElw0bNmD58uXIzMyE\nVqvFuHHj4Ojo+NLafvnlFyxbtgxqtRpCCIwYMQKurq5lTsTZunUrjIyMEBAQAAAYNWoUgoKCcOPG\nDWnkV79+fXzyySewsLConk4j0gP+qgMRESkOD3sSEZHiMPyIiEhxGH5ERKQ4DD8iIlIchh8RESkO\nw4+IiBSH4UdERIrz/wGmjSczoH4kyAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fb9c81d7510>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "z=linkage(x,'ward')\n",
    "dendrogram(z,truncate_mode='lastp',p=12,leaf_rotation=45.,leaf_font_size=15.)\n",
    "plt.title('Truncated Hierarchical Clustering Dendogram')\n",
    "plt.xlabel('Cluster Size')\n",
    "plt.ylabel('Distance')\n",
    "plt.axhline(y=500)\n",
    "plt.axhline(y=150)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.78125"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "k=2\n",
    "Hclustering=AgglomerativeClustering(n_clusters=k,affinity='euclidean',linkage='ward')\n",
    "Hclustering.fit(x)\n",
    "sm.accuracy_score(y,Hclustering.labels_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.4375"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Hclustering=AgglomerativeClustering(n_clusters=k,affinity='euclidean',linkage='complete')\n",
    "Hclustering.fit(x)\n",
    "sm.accuracy_score(y,Hclustering.labels_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.78125"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Hclustering=AgglomerativeClustering(n_clusters=k,affinity='euclidean',linkage='average')\n",
    "Hclustering.fit(x)\n",
    "sm.accuracy_score(y,Hclustering.labels_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.71875"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Hclustering=AgglomerativeClustering(n_clusters=k,affinity='manhattan',linkage='average')\n",
    "Hclustering.fit(x)\n",
    "sm.accuracy_score(y,Hclustering.labels_)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# K nears neighbour algorithm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import urllib\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn import neighbors\n",
    "from sklearn import preprocessing\n",
    "from sklearn.model_selection import train_test_split\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "rcParams['figure.figsize']=7,4\n",
    "sb.set_style('whitegrid')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "x_prime=cars.iloc[:,[1,3,4,6]]\n",
    "y_tar=cars.iloc[:,9]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "x_scaled=preprocessing.scale(x_prime)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "x_train,x_test,y_train,y_test=train_test_split(x_scaled,y_tar,test_size=.20,random_state=17)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski',\n",
      "           metric_params=None, n_jobs=1, n_neighbors=5, p=2,\n",
      "           weights='uniform')\n"
     ]
    }
   ],
   "source": [
    "clf=neighbors.KNeighborsClassifier()\n",
    "clf.fit(x_train,y_train)\n",
    "print clf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "             precision    recall  f1-score   support\n",
      "\n",
      "          0       0.80      1.00      0.89         4\n",
      "          1       1.00      0.67      0.80         3\n",
      "\n",
      "avg / total       0.89      0.86      0.85         7\n",
      "\n"
     ]
    }
   ],
   "source": [
    "y_pred=clf.predict(x_test)\n",
    "print sm.classification_report(y_test,y_pred)\n",
    "#High precision + low Recall= Few results were returned but many of the label predictions returned were correct"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}