breast_cancer-classification modeling.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Classification modeling of breast cancer data "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the [previous blog](https://github.com/saramille/Breast-cancer-Data-analysis/blob/master/breast%20cancer%20data%20analysis%20colinearity%20and%20dimensional%20reduction.ipynb), collinearity issues were addressed by removing some of the variables from the data  and PCA was applied to reduce the dimensionality. The reduced data is used here in this blog for modeling breast cancer. Since, we do not know the best model, first selected classification models will be compared and later the best will be fitted to the data for prediction purposes. \n",
    "\n",
    "The classification models compared are K-nearest Neighbors, Support Vector Classifier, and Logistic Regression.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from matplotlib import pyplot as plt\n",
    "from sklearn.decomposition import PCA\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "# loading the raw data\n",
    "bc=pd.read_csv(\"C:\\data science\\project1\\BC.csv\")\n",
    "bc.head()  \n",
    "P=bc.iloc[:,2:32]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>radius_mean</th>\n",
       "      <th>texture_mean</th>\n",
       "      <th>perimeter_mean</th>\n",
       "      <th>area_mean</th>\n",
       "      <th>smoothness_mean</th>\n",
       "      <th>compactness_mean</th>\n",
       "      <th>concavity_mean</th>\n",
       "      <th>concave points_mean</th>\n",
       "      <th>symmetry_mean</th>\n",
       "      <th>fractal_dimension_mean</th>\n",
       "      <th>...</th>\n",
       "      <th>radius_worst</th>\n",
       "      <th>texture_worst</th>\n",
       "      <th>perimeter_worst</th>\n",
       "      <th>area_worst</th>\n",
       "      <th>smoothness_worst</th>\n",
       "      <th>compactness_worst</th>\n",
       "      <th>concavity_worst</th>\n",
       "      <th>concave points_worst</th>\n",
       "      <th>symmetry_worst</th>\n",
       "      <th>fractal_dimension_worst</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>569.000000</td>\n",
       "      <td>569.000000</td>\n",
       "      <td>569.000000</td>\n",
       "      <td>569.000000</td>\n",
       "      <td>569.000000</td>\n",
       "      <td>569.000000</td>\n",
       "      <td>569.000000</td>\n",
       "      <td>569.000000</td>\n",
       "      <td>569.000000</td>\n",
       "      <td>569.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>569.000000</td>\n",
       "      <td>569.000000</td>\n",
       "      <td>569.000000</td>\n",
       "      <td>569.000000</td>\n",
       "      <td>569.000000</td>\n",
       "      <td>569.000000</td>\n",
       "      <td>569.000000</td>\n",
       "      <td>569.000000</td>\n",
       "      <td>569.000000</td>\n",
       "      <td>569.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>14.127292</td>\n",
       "      <td>19.289649</td>\n",
       "      <td>91.969033</td>\n",
       "      <td>654.889104</td>\n",
       "      <td>0.096360</td>\n",
       "      <td>0.104341</td>\n",
       "      <td>0.088799</td>\n",
       "      <td>0.048919</td>\n",
       "      <td>0.181162</td>\n",
       "      <td>0.062798</td>\n",
       "      <td>...</td>\n",
       "      <td>16.269190</td>\n",
       "      <td>25.677223</td>\n",
       "      <td>107.261213</td>\n",
       "      <td>880.583128</td>\n",
       "      <td>0.132369</td>\n",
       "      <td>0.254265</td>\n",
       "      <td>0.272188</td>\n",
       "      <td>0.114606</td>\n",
       "      <td>0.290076</td>\n",
       "      <td>0.083946</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>3.524049</td>\n",
       "      <td>4.301036</td>\n",
       "      <td>24.298981</td>\n",
       "      <td>351.914129</td>\n",
       "      <td>0.014064</td>\n",
       "      <td>0.052813</td>\n",
       "      <td>0.079720</td>\n",
       "      <td>0.038803</td>\n",
       "      <td>0.027414</td>\n",
       "      <td>0.007060</td>\n",
       "      <td>...</td>\n",
       "      <td>4.833242</td>\n",
       "      <td>6.146258</td>\n",
       "      <td>33.602542</td>\n",
       "      <td>569.356993</td>\n",
       "      <td>0.022832</td>\n",
       "      <td>0.157336</td>\n",
       "      <td>0.208624</td>\n",
       "      <td>0.065732</td>\n",
       "      <td>0.061867</td>\n",
       "      <td>0.018061</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>6.981000</td>\n",
       "      <td>9.710000</td>\n",
       "      <td>43.790000</td>\n",
       "      <td>143.500000</td>\n",
       "      <td>0.052630</td>\n",
       "      <td>0.019380</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.106000</td>\n",
       "      <td>0.049960</td>\n",
       "      <td>...</td>\n",
       "      <td>7.930000</td>\n",
       "      <td>12.020000</td>\n",
       "      <td>50.410000</td>\n",
       "      <td>185.200000</td>\n",
       "      <td>0.071170</td>\n",
       "      <td>0.027290</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.156500</td>\n",
       "      <td>0.055040</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>11.700000</td>\n",
       "      <td>16.170000</td>\n",
       "      <td>75.170000</td>\n",
       "      <td>420.300000</td>\n",
       "      <td>0.086370</td>\n",
       "      <td>0.064920</td>\n",
       "      <td>0.029560</td>\n",
       "      <td>0.020310</td>\n",
       "      <td>0.161900</td>\n",
       "      <td>0.057700</td>\n",
       "      <td>...</td>\n",
       "      <td>13.010000</td>\n",
       "      <td>21.080000</td>\n",
       "      <td>84.110000</td>\n",
       "      <td>515.300000</td>\n",
       "      <td>0.116600</td>\n",
       "      <td>0.147200</td>\n",
       "      <td>0.114500</td>\n",
       "      <td>0.064930</td>\n",
       "      <td>0.250400</td>\n",
       "      <td>0.071460</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>13.370000</td>\n",
       "      <td>18.840000</td>\n",
       "      <td>86.240000</td>\n",
       "      <td>551.100000</td>\n",
       "      <td>0.095870</td>\n",
       "      <td>0.092630</td>\n",
       "      <td>0.061540</td>\n",
       "      <td>0.033500</td>\n",
       "      <td>0.179200</td>\n",
       "      <td>0.061540</td>\n",
       "      <td>...</td>\n",
       "      <td>14.970000</td>\n",
       "      <td>25.410000</td>\n",
       "      <td>97.660000</td>\n",
       "      <td>686.500000</td>\n",
       "      <td>0.131300</td>\n",
       "      <td>0.211900</td>\n",
       "      <td>0.226700</td>\n",
       "      <td>0.099930</td>\n",
       "      <td>0.282200</td>\n",
       "      <td>0.080040</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>15.780000</td>\n",
       "      <td>21.800000</td>\n",
       "      <td>104.100000</td>\n",
       "      <td>782.700000</td>\n",
       "      <td>0.105300</td>\n",
       "      <td>0.130400</td>\n",
       "      <td>0.130700</td>\n",
       "      <td>0.074000</td>\n",
       "      <td>0.195700</td>\n",
       "      <td>0.066120</td>\n",
       "      <td>...</td>\n",
       "      <td>18.790000</td>\n",
       "      <td>29.720000</td>\n",
       "      <td>125.400000</td>\n",
       "      <td>1084.000000</td>\n",
       "      <td>0.146000</td>\n",
       "      <td>0.339100</td>\n",
       "      <td>0.382900</td>\n",
       "      <td>0.161400</td>\n",
       "      <td>0.317900</td>\n",
       "      <td>0.092080</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>28.110000</td>\n",
       "      <td>39.280000</td>\n",
       "      <td>188.500000</td>\n",
       "      <td>2501.000000</td>\n",
       "      <td>0.163400</td>\n",
       "      <td>0.345400</td>\n",
       "      <td>0.426800</td>\n",
       "      <td>0.201200</td>\n",
       "      <td>0.304000</td>\n",
       "      <td>0.097440</td>\n",
       "      <td>...</td>\n",
       "      <td>36.040000</td>\n",
       "      <td>49.540000</td>\n",
       "      <td>251.200000</td>\n",
       "      <td>4254.000000</td>\n",
       "      <td>0.222600</td>\n",
       "      <td>1.058000</td>\n",
       "      <td>1.252000</td>\n",
       "      <td>0.291000</td>\n",
       "      <td>0.663800</td>\n",
       "      <td>0.207500</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 30 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       radius_mean  texture_mean  perimeter_mean    area_mean  \\\n",
       "count   569.000000    569.000000      569.000000   569.000000   \n",
       "mean     14.127292     19.289649       91.969033   654.889104   \n",
       "std       3.524049      4.301036       24.298981   351.914129   \n",
       "min       6.981000      9.710000       43.790000   143.500000   \n",
       "25%      11.700000     16.170000       75.170000   420.300000   \n",
       "50%      13.370000     18.840000       86.240000   551.100000   \n",
       "75%      15.780000     21.800000      104.100000   782.700000   \n",
       "max      28.110000     39.280000      188.500000  2501.000000   \n",
       "\n",
       "       smoothness_mean  compactness_mean  concavity_mean  concave points_mean  \\\n",
       "count       569.000000        569.000000      569.000000           569.000000   \n",
       "mean          0.096360          0.104341        0.088799             0.048919   \n",
       "std           0.014064          0.052813        0.079720             0.038803   \n",
       "min           0.052630          0.019380        0.000000             0.000000   \n",
       "25%           0.086370          0.064920        0.029560             0.020310   \n",
       "50%           0.095870          0.092630        0.061540             0.033500   \n",
       "75%           0.105300          0.130400        0.130700             0.074000   \n",
       "max           0.163400          0.345400        0.426800             0.201200   \n",
       "\n",
       "       symmetry_mean  fractal_dimension_mean           ...             \\\n",
       "count     569.000000              569.000000           ...              \n",
       "mean        0.181162                0.062798           ...              \n",
       "std         0.027414                0.007060           ...              \n",
       "min         0.106000                0.049960           ...              \n",
       "25%         0.161900                0.057700           ...              \n",
       "50%         0.179200                0.061540           ...              \n",
       "75%         0.195700                0.066120           ...              \n",
       "max         0.304000                0.097440           ...              \n",
       "\n",
       "       radius_worst  texture_worst  perimeter_worst   area_worst  \\\n",
       "count    569.000000     569.000000       569.000000   569.000000   \n",
       "mean      16.269190      25.677223       107.261213   880.583128   \n",
       "std        4.833242       6.146258        33.602542   569.356993   \n",
       "min        7.930000      12.020000        50.410000   185.200000   \n",
       "25%       13.010000      21.080000        84.110000   515.300000   \n",
       "50%       14.970000      25.410000        97.660000   686.500000   \n",
       "75%       18.790000      29.720000       125.400000  1084.000000   \n",
       "max       36.040000      49.540000       251.200000  4254.000000   \n",
       "\n",
       "       smoothness_worst  compactness_worst  concavity_worst  \\\n",
       "count        569.000000         569.000000       569.000000   \n",
       "mean           0.132369           0.254265         0.272188   \n",
       "std            0.022832           0.157336         0.208624   \n",
       "min            0.071170           0.027290         0.000000   \n",
       "25%            0.116600           0.147200         0.114500   \n",
       "50%            0.131300           0.211900         0.226700   \n",
       "75%            0.146000           0.339100         0.382900   \n",
       "max            0.222600           1.058000         1.252000   \n",
       "\n",
       "       concave points_worst  symmetry_worst  fractal_dimension_worst  \n",
       "count            569.000000      569.000000               569.000000  \n",
       "mean               0.114606        0.290076                 0.083946  \n",
       "std                0.065732        0.061867                 0.018061  \n",
       "min                0.000000        0.156500                 0.055040  \n",
       "25%                0.064930        0.250400                 0.071460  \n",
       "50%                0.099930        0.282200                 0.080040  \n",
       "75%                0.161400        0.317900                 0.092080  \n",
       "max                0.291000        0.663800                 0.207500  \n",
       "\n",
       "[8 rows x 30 columns]"
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# summary stat of the raw data\n",
    "bc.iloc[:,2:32].describe() "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    1\n",
       "1    1\n",
       "2    1\n",
       "3    1\n",
       "4    1\n",
       "Name: M, dtype: uint8"
      ]
     },
     "execution_count": 112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yd=pd.get_dummies(bc.diagnosis)\n",
    "y=yd.M\n",
    "y.head() # response variable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(569, 11)\n",
      "[ 0.42661046  0.15932139  0.10294428  0.07788731  0.06489774  0.05015242\n",
      "  0.02145044  0.0187846   0.01505759  0.01197751  0.01117206]\n",
      "0.960255820189\n"
     ]
    }
   ],
   "source": [
    "# standardizing and PCA\n",
    "scaler = StandardScaler()\n",
    "P_scaled=scaler.fit_transform(P)\n",
    "P_scaled=pd.DataFrame(P_scaled)\n",
    "PP=P_scaled.drop(P_scaled.columns[[2, 3, 22, 23, 12, 13]], axis=1) \n",
    "pca=PCA(n_components=0.95)\n",
    "P_pca=pca.fit_transform(PP)\n",
    "print (P_pca.shape)\n",
    "print(pca.explained_variance_ratio_) \n",
    "print (pca.explained_variance_ratio_.sum())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZAAAAFgCAYAAACVLS/VAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\noAgBAgAoQoAAAIoQIACAIgQIAKDIJ0Joka1Uc0MmAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xd61aac8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "n=['PC1','PC2','PC3','PC4','PC5','PC6','PC7','PC8','PC9','PC10','PC11','diagnosis']\n",
    "d=bc.iloc[:,1:2]  # diagnosis column\n",
    "#merging the reduced data with diagnosis column\n",
    "Xy=pd.DataFrame(np.hstack([P_pca,d.as_matrix()]),columns=n)\n",
    "# plotting the the first 2 pca components against diagnosis\n",
    "sns.lmplot(\"PC1\", \"PC2\", hue=\"diagnosis\", data=Xy, fit_reg=False,markers=[\"o\", \"x\"],palette=\"Set1\")\n",
    "sns.plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "## Modeling \n",
    "From the above plot, we can see that the positive and negative cases (i.e. B and M) have fairly distinctive regions, this makes it easier to fit linear or non-linear classifiers to model the problem. Below, **k-neighbors,Support Vector MAchine(SVM), and Linear Discriminant Analysis (LDA)** classifiers are fit to the data and later results are compared."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Splitting the data: training and test "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn.cross_validation import train_test_split\n",
    "#predictor X\n",
    "X=(Xy.iloc[:,0:11]).as_matrix()\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1. K-nearest Neighbors classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.cross_validation import cross_val_score\n",
    "knn=KNeighborsClassifier()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Parameter tuning: optimizing the number of neighbors (k)**\n",
    "* For optimization purposes it is suggested using the training data; and\n",
    "* for performance estimation or to figure out the best model, the test data will be used."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0.924  0.874  0.937  0.924  0.943  0.925  0.925  0.912  0.918  0.906\n",
      "  0.906  0.887  0.9    0.887  0.893  0.874  0.893  0.868  0.887  0.868\n",
      "  0.887  0.875  0.887  0.875  0.881  0.874  0.893  0.862  0.862  0.856\n",
      "  0.875  0.855  0.862  0.855  0.862  0.849  0.862  0.843  0.85   0.843\n",
      "  0.843  0.83   0.837  0.83   0.843  0.83   0.837  0.837  0.837]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfUAAAFXCAYAAAC7nNf0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRERchEJdRETERfwPBG2d7i25bDYAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7b12a20>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "k_range=list(range(1,50))\n",
    "k_scores=[]\n",
    "for k in k_range:\n",
    "    knn=KNeighborsClassifier(n_neighbors=k)\n",
    "    scores=cross_val_score(knn, X_train,y_train,cv=10,scoring='recall')\n",
    "    k_scores.append(scores.mean())\n",
    "print(np.round(k_scores,3)) # to display scores to 3 decimal places\n",
    "from matplotlib import pyplot as plt\n",
    "plt.plot(k_range,k_scores)\n",
    "plt.xlabel('k values')\n",
    "plt.ylabel('recall')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* From the above plot the optimal k value is somewhere between 1 and 10. Looking at the scores, at k=5 we have the highest recall/sensitivity.\n",
    "* We can use a more efficient parameter tuning approach called **GridSearchCV** as shown below which basically automates the search process and it also fits the optimal parameter to the training data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\grid_search.py:43: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. This module will be removed in 0.20.\n",
      "  DeprecationWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "# Tuning hyper-parameters for accuracy\n",
      "{'n_neighbors': 6}\n",
      "0.96\n",
      "# Tuning hyper-parameters for recall\n",
      "{'n_neighbors': 5}\n",
      "0.943\n"
     ]
    }
   ],
   "source": [
    "from sklearn.grid_search import GridSearchCV\n",
    "k_range=list(range(1,50))\n",
    "param_grid=dict(n_neighbors=k_range)\n",
    "scores = ['accuracy', 'recall'] # evaluation scores for selecting best parameter\n",
    "for sc in scores:\n",
    "    grid_knn=GridSearchCV(knn,param_grid,cv=10,scoring=sc,n_jobs=-1)\n",
    "    print(\"# Tuning hyper-parameters for %s\" % sc)\n",
    "    grid_knn.fit(X_train,y_train)\n",
    "    print(grid_knn.best_params_)\n",
    "    print(np.round(grid_knn.best_score_,3))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* Here, the optimal n_neighbors according to recall score (sensitivity) is **selected i.e.k=5** instead of k=6. As we are more concerned in predicting cancer cases (M) accurately rather than no-cancer cases (B), we go with results found using recall or sensitivity score."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAETCAYAAADah9Z7AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\njvh/7ID07GLBDgAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7743f98>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# fitting the optimal model (i.e. knn with k=5 based upon recall score) onto the training data\n",
    "knn=KNeighborsClassifier(n_neighbors=5)\n",
    "knn.fit(X_train, y_train)\n",
    "\n",
    "# for display purposes, we fit the model on the first two components i.e. PC1, and PC2\n",
    "knn.fit(X_train[:,0:2], y_train)\n",
    "# Plotting the decision boundary for all data (both train and test)\n",
    "# Create color maps\n",
    "from matplotlib.colors import ListedColormap\n",
    "cmap_light = ListedColormap(['#AAFFAA','#FFAAAA'])\n",
    "cmap_bold = ListedColormap(['#0000FF','#FF0000'])\n",
    "# creating a meshgrid\n",
    "x_min, x_max = X[:, 0].min() - 1, X[:, 0].max() + 1\n",
    "y_min, y_max = X[:, 1].min() - 1, X[:, 1].max() + 1\n",
    "h=0.05\n",
    "xx, yy = np.meshgrid(np.arange(x_min, x_max, h),np.arange(y_min, y_max, h))\n",
    "xy_mesh=np.c_[xx.ravel(), yy.ravel()]\n",
    "Z = knn.predict(xy_mesh)\n",
    "%matplotlib inline\n",
    "Z = Z.reshape(xx.shape)\n",
    "#print(Z)\n",
    "plt.figure()\n",
    "plt.pcolormesh(xx, yy, Z, cmap=cmap_light)\n",
    "ax=plt.scatter(X[:, 0], X[:, 1], c=y, cmap=cmap_bold)\n",
    "plt.xlim(xx.min(), xx.max());plt.ylim(yy.min(), yy.max())\n",
    "plt.xlabel('PC1');plt.ylabel('PC2')\n",
    "plt.title('KNN')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 2. Support Vector Machine (SVM)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "# Tuning hyper-parameters for accuracy\n",
      "{'C': 100, 'gamma': 0.001, 'kernel': 'rbf'}\n",
      "0.979\n",
      "# Tuning hyper-parameters for recall\n",
      "{'C': 100, 'gamma': 0.001, 'kernel': 'rbf'}\n",
      "0.962\n"
     ]
    }
   ],
   "source": [
    "from sklearn.cross_validation import cross_val_score\n",
    "from sklearn.svm import SVC\n",
    "svc=SVC()\n",
    "param_grid = [{'kernel': ['rbf'], 'gamma': [1e-3, 1e-4, 1e-5],'C': [1, 10, 100, 1000]},\n",
    "              {'kernel': ['linear'], 'C': [1, 10, 100, 1000]}]\n",
    "scores = ['accuracy', 'recall']\n",
    "for sc in scores:\n",
    "    grid=GridSearchCV(svc,param_grid,cv=10,scoring=sc,n_jobs=-1)\n",
    "    print(\"# Tuning hyper-parameters for %s\" % sc)\n",
    "    grid.fit(X_train,y_train)\n",
    "    print(grid.best_params_)\n",
    "    print(np.round(grid.best_score_,3))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* Using accuracy and recall as scoring metrics, both give the same model as optimal i.e. {'C': 100, 'gamma': 0.001, 'kernel': 'rbf'}. Here below, we fit this model to the data and do visualization of the results."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAETCAYAAADah9Z7AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nBIIg4v8BoEQxbnIMmSYAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xc3b94e0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn import metrics\n",
    "from sklearn import cross_validation as cv\n",
    "\n",
    "# fitting the optimal model onto the training data\n",
    "svc=SVC(C=100,gamma=0.001,kernel='rbf')\n",
    "svc.fit(X_train, y_train)\n",
    "\n",
    "# for display purposes, we fit the model on the first two components i.e. PC1, and PC2\n",
    "svc.fit(X_train[:,0:2], y_train)\n",
    "\n",
    "# Plotting the decision boundary for all data (both train and test)\n",
    "# Create color maps\n",
    "from matplotlib.colors import ListedColormap\n",
    "cmap_light = ListedColormap(['#AAFFAA','#FFAAAA'])\n",
    "cmap_bold = ListedColormap(['#0000FF','#FF0000'])\n",
    "\n",
    "# creating a meshgrid\n",
    "x_min, x_max = X[:, 0].min() - 1, X[:, 0].max() + 1\n",
    "y_min, y_max = X[:, 1].min() - 1, X[:, 1].max() + 1\n",
    "h=0.05\n",
    "xx, yy = np.meshgrid(np.arange(x_min, x_max, h),np.arange(y_min, y_max, h))\n",
    "xy_mesh=np.c_[xx.ravel(), yy.ravel()]\n",
    "Z = svc.predict(xy_mesh)\n",
    "Z = Z.reshape(xx.shape)\n",
    "\n",
    "#plotting data on decision boundary\n",
    "%matplotlib inline\n",
    "plt.figure()\n",
    "plt.pcolormesh(xx, yy, Z, cmap=cmap_light)\n",
    "plt.scatter(X[:, 0], X[:, 1], c=y, cmap=cmap_bold)\n",
    "plt.xlim(xx.min(), xx.max())\n",
    "plt.ylim(yy.min(), yy.max())\n",
    "plt.xlabel('PC1');plt.ylabel('PC2')\n",
    "plt.title('SVC')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 3. Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "# Tuning hyper-parameters for accuracy\n",
      "{'C': 0.1}\n",
      "0.979\n",
      "# Tuning hyper-parameters for recall\n",
      "{'C': 1}\n",
      "0.962\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.cross_validation import cross_val_score\n",
    "from sklearn.grid_search import GridSearchCV\n",
    "\n",
    "lgr = LogisticRegression()\n",
    "\n",
    "#parameter tuning\n",
    "param_grid = {'C': [0.001, 0.01, 0.1, 1, 10, 100, 1000] }\n",
    "scores = ['accuracy', 'recall']\n",
    "for sc in scores:\n",
    "    grid_lgr=GridSearchCV(lgr,param_grid,cv=10,scoring=sc,n_jobs=-1)\n",
    "    print(\"# Tuning hyper-parameters for %s\" % sc)\n",
    "    grid_lgr.fit(X_train,y_train)\n",
    "    print(grid_lgr.best_params_)\n",
    "    print(np.round(grid_lgr.best_score_,3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAETCAYAAADah9Z7AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzsnXV4FNf+h9/djbsRkhCHZHD30pYWp5S23FJ3995SuRXa\nX93dbr2FKhcoLYUCxd09yBBBIoS4++7+/jgJ2c1KhCg57/PkKZnZmTm7zX7OOV/VGI1GJBKJRNI5\n0Lb1ACQSiUTSekjRl0gkkk6EFH2JRCLpREjRl0gkkk6EFH2JRCLpREjRl0gkkk6EQ1sPQNK5UBQl\nEohTVdWjme43HRivquojdl5zGTBCVdUXGnHfdUAEkF99SAc4A6+qqjq36SNuGRRFCQEWqKo6uq3H\nImnfSNGXdGhUVV0MLK7nZcMAvybc/klVVRfU/KIoylBgs6Ioi1RVLWzC/VoMVVXTACn4knqRoi9p\nNyiK4g18BgwEjMAy4FlVVasURZkKvAXogX3AeGAMMBa4WlXVaYqizABmA4bq1z0JlAP3ATpFUfJV\nVX1OUZRngFuBKiAeuE1V1XzqJxoorr4niqJcXv08J6AEeEJV1a2KorgBXwAjgTzgMICqqrcpinIC\n2A70B54FdgCfAuGAI/CbqqqvK4riAHxS/R4rgCTgdqDMxvEAqndQiqI4Au8D46o/h+3AY6qqFlY/\n/4fqc+HAPFVVn2rAe5ecJ0ibvqQ98TGQDfQDhgIDgCcURfEHfgRuUlV1ILAW6Gbl+neAB1RVHQo8\nD4xVVXU7QoDnVQv+dOA2YJSqqn2B48BDNsbzjqIo+xRFOaEoyhngKmCcqqoViqLEAK8DU1VVHQTc\nA/yuKIp79bMdgJ6IyWlQnfvGqaraS1XVRdXv6ztVVYcAw4HxiqJcA4xCTGj9q88lISYKW8dNmQ2E\nVH9+AxDf83dMznuoqnohYmfwsKIoUTbev+Q8RIq+pD0xBfhUVVWjqqrlCLGeAlwEHFZVdT+Aqqpz\ngAIr1/8GLFIU5RvAF3jbymvGA/NVVc2tvtcsVVVfszGeJ6snmWHAKSBTVdW91ecmAMHAakVR9gE/\nI3YYPYCpwLeqqhpUVS0A5tS570aA6gniYuCV6ntsQ6y+BwIHqV6lK4ryCrBQVdUtdo7X/Ry/UFW1\nUlVVA2JnMMXk/J/V7z0VyKBppi9JB0WKvqQ9UffvUYsweVQBmjrnDHUvVlX1OeACYBdiNb9VUZS6\n96xCmI4AUBTFp9q5bBNVVTOBa4G7FEWZWX1YB6xWVXVgzQ/CnBNnZbz6OrcsMrmHBhhd5x6vq6qa\nR/VOp/r6eYqiPGbreJ372/ocayg1+bcRy89Wch4jRV/SnlgBPKgoikZRFGeEyWQlsBmIVRSlP4Ci\nKP8CfDAXb4dqe7W7qqpfAA8AvaidNGpEbxUwQ1EUr+rfXwRm1TcwVVWTgNeAD6pX6GuAiYqi9Kx+\n/lTgAOACLAVuVxRFW23fv8F0rCb3LECs7mdV38On+r1eoSjKNGA1sEVV1ReBucAAW8etfI73KYri\nWD3pPVj9OUok0pEraRPcFUUpqnNsFPAIwhRxEOEcXQ68Vm1Dvx6YqyiKAbGSr0I4TwGodvb+G/hF\nUZRKxE7gDlVVyxVFWY2wt1eoqvqwoii9EVE4AIeAuxs47ncRDuDnVVV9WlGUe4DfFEXRVI9nuqqq\nxYqivIFwzh5EhHxmmI61DjcAnyqKUvOef1VV9WdFUXQIk0xc9WeVWz3OZBvHTXm1eqz7EN/xHcDD\nDXyPkvMcjSytLGnvVK/KZwMvqqpaoijKYMRqOkRV1Xb3B6woynVAgaqqf1evtBcC/6iq+t82HppE\nIs07kvZPtRmkAthZ7fD8ErimPQp+NXHAc9VjjQPSgG/adkgSiUCu9CUSiaQTIVf6EolE0omQoi+R\nSCSdiHYfvTOf+dL+JJFIJI1kJjOt5l/Ilb5EIpF0IqToSyQSSSdCir5EIpF0IqToSyQSSSdCir5E\nIpF0IqToSyQSSSdCir5EIpF0Ilo8Tl9RlBHAW6qqjlUUpQeiVZsRUZPkweomDxKJRNLpMOjh+D4f\nXD2qCLEoPNsytOhKX1GUpxCFplyqD70PzK5u1aYBrmjJ50skEkl7Zd2cCJ4eMp5nho7nif4TefGS\ni4nf7tviz21p804iMMPk9yHA+up/L0O0rpNIJJJOxaF1Acx5bAAn9vsCGqoqdBxeF8jntw+jrFjX\nos9uUdFXVXUhUGlySGNSDrcQ8G7J50skEkl7ZO33kRTnOlscTz3izT//jW7RZ7e2I9fUfu8J5LXy\n8yUSiaTNyU93sXkuN9W1RZ/d2qK/V1GUsdX/ngJsbOXnSyQSSZvjH15q81xQTMs6dFtb9B8HXlIU\nZSuiH+iCVn6+RCKRtDmT7k/EN9iybXLUkBwuvfNEiz673XfOkqWVJRLJ+cjeZV35612F43t8cHA2\n0POCbG546wDBPYqb5f62Siu3+3r6EolEcj4yaMoZBk05Q1GOIw5OBlw89K3yXCn6EolE0oZ4+FXW\n/6JmRJZhkEgkkk6EFH2JRCLpREjRl0gkkk6EFH2JRCLpREjRl0gkkk6EFH2JRCLpREjRl0gkkk6E\nFH2JRCLpRMjkLIlEIjlPmDnf9Bfrr5ErfYlEIjkPMBN8O8iVvkQikXRgGir2NciVvkQikXRQGiv4\nIFf6EolE0qFoitCbIkVfIpFIOgDnKvY1SPOORCKRtHOaS/ChlVf6iqI4AnOASEAP3K2q6tHWHINE\nIpF0BJpT6E1p7ZX+VMBBVdXRwMvAa638fIlEImn3tJTgQ+vb9I8BDoqiaAEvoHVbxkgkEkk7pSWF\n3pTWFv0ihGnnKBAATGvl50skEkm7orXEvobWNu88BqxQVTUWGADMURTFpZXHIJFYJS/DiT/fjuX3\n13qSnuDW1sORdAJaW/Ch9Vf6udSadHIAR0DXymOQSCxY9ml3Fr3ei7zTrgD89V4sE+5N4oY34tp4\nZJLzjbYQelNae6X/ATBYUZSNwBrgWVVVi1t5DBKJGcmHPfnfC33OCj5Aca4zS96PZdvCkDYcmeR8\no60FH1p5pa+qahFwTWs+UyKpj3XfR1Gc62xxvKpCx47fQxn5r7Q2GJXkfKI9iH0NMiNX0ukpL7Zt\nYSyzc04iqY/2JPY1SNGXdHpiRmbzz3+7AxqLcxH981ttHFvnd2Pt95FknnDHJ6iMUdekMPG+pFZ7\nvqT5aI9iX4MUfUmn58IbT7Hx53AO/BNsdjx6SA7THj/WKmNYNyeC7x4aSFmREwCpR7w5uimAggxn\nrn7hSKuMQXLutGexr0GKvqTTo9XBU39sYeGrvTi6MQC9XkP3obnMeO4I7t5VLf58oxFWfRl9VvBr\n0FfqWD83gsufOIazm77FxyFpOh1B7GuQoi+RAE6uBq5/7VCbPLs4z5HUI55Wz51J9CRptw+9Lsxu\n5VFJGkpHEnyQoi+RtDku7lW4eVdSnGcZQeTiUYl/aEkbjEpij44m9KbI0soSSRvj4GSk77gMq+d6\nX5xJYFRpK49IYo+OLPggV/oSSbvg9o/2UZDpzIGVXaksc0DrYKDnBVnc9fnuth6apJqOLvY1SNGX\nSNoBLh56/rN4C+oWP+K3+RPau4ABk86gsYwilbQi54vQmyJFXyJpRyijc1BG57T1MCScn4IPUvQl\nklYjJ9WZrfPDcPOpZMz1yTg6G9p6SBIrnK9iX4MUfYmkhTEa4een+7Hu+0gKMkUl8cVvKdz41kGG\nTj/dxqOTmHK+Cz5I0ZdIWpz1cyJY8n4shqraYLnUo958/++B9B6biZtXbQJYSb4Dq7+JorzYgSGX\npxE1qPXKQHRWOoPQmyJFXyJpYbYv6mYm+DVkHvdg5RfRXPGUKPWw6ddQfnm6P1mn3AH4822FC244\nxb1f7pEO3Rags4l9DVL0JZIWpjTf0ea5olxReqEwx5Gfn+pPdor72XPlxY6s+Tqa8L4FTH0kocXH\n2VnorGJfg0zOkkhamJCeBVaP18TiA6z+KtpM8GvRsH951xYcXeeisws+yJW+RNLiTJt1jLjVXUlP\nMK+vM3DyaQZflg5AaZHtr2JZsfyaniutLvbl5bB1K2i1MHIkODnVf00r0ep/TYqiPANMB5yAz1VV\n/ba1x9BZMBhg5RfRHFjZFX2llqjBuUx/UsXVU1ZsbE1CYot5fOEW/nxb4eQ+H5xcDfQem8G1rxw6\na6vvP+EMS96NpbLc8isZ3q9zOnNXfhXF1nlh5KW7EBBewiV3HGfUzNRG3aNNVvYrV8LixXDmjPh9\n4UKYMQMuvrgNBmNJq4q+oihjgdHABYAb8ERrPr8zYTTCZ7cOY+NPEdQ0B9mzNIS4NYE8u2yjFP5W\nJqJ/AY/8tNPm+T4XZzF8Riqbf40wOx7aO5/pT6otPbx2x++v9WT+S73RV4rOZSmHvTm6yZ+yIgcu\nuf1kvde3mRknPh5+/hlKTIrknT4Nc+dC9+4QGtpGA6ultW36k4CDwCLgL2BJKz+/07D/n0C2zAuj\nbjcodXMXlrwX2zaDktjlobk7ue71A/S59AwxI7MYf28C//lrE10iOlfBtYpSLet+iDwr+DWUFTmx\n8svuGI32r29Tu/26deaCX0NhIaxa1erDsUZrm3cCgAhgGhAFLFYUpaeqqvX8b5Q0lv0rgi2+NDUk\n7PRr5dFIGoLOwciMZ1RmPNP5VvamnNjnbeH/qCH1qCdFOU54+leYHW83DtriYtvnrE0GbUBrr/Sz\ngRWqqlaoqqoCZUCXVh5Dp0DnaDvF30Gm/0vaMT4hZbh4VFo95+5dibO7eTezdiP4AMHBts9169Z6\n47BDa6/0NwGPKoryPhAMuCMmAkkzc/EtJ1j1ZTQl+XWjBoz0n3CmTcZ0PpOw05f1cyIoLXAkvF8+\nkx5MlC0Om0hgRCm9L85kz9IQi3N9L83AycXQvoTelMsug5074dQp8+NRUTBpUtuMqQ4aY30GsmZG\nUZS3gUsQu4xnVVVdYe/185kvTT9NZNEbCn++1fOs8Ds46Rl9bTIP/LATrczQaDaWfdKd32b3pbSg\ndoLtMSKb/yzehHdghZ0rJbbIOuXCp7eO4OimAAxVWhxdqug3PoO1V2/Hy63l+xafE6dPw4IFkJAA\nGg3ExsK114K/f+uOY+ZMq3ncrS76jUWK/rmRfNiTDXMjqKrQMWjqafqNy5Ap/c1ISb4Ds/pOJMdK\nYtXEBxK467O9bTCq8wOjEfav6EryIS9iRuTw/GlpFGgUNkRfZn2c54T1LuTGN+PaehjnLRt+jLAq\n+ADHtkqH+bmg0cBrhWcg/AzIYqTNhtzkSyTngP2NstxSnQvt1m7fwZGiL5GcAxfedAq/btZD8WJG\nSHNEU5g5Xwp+SyLNOxLJOeDhW8nlT6j874W+lBbWVtPsPiybmS8eavB9Ug57svLLaAqznAmMKmbK\no/F4d+k8TmAp8q2HFH2J5By57N8JxIzMZv3cSErzRcjm5IcTcHFvWMjmlv9147uHB1OQ4XL22LaF\noTw2bysR/a1X6DyfkILfusjoHUm7xWiEuNWBxO/wJTCqhNHXJKO1nmTcYdFXaXh6yDhOHvC1ODd8\nRjJPLNzWBqNqHaTYtzAyekfSkSgt1PHBtaOIWx1IVYUOMLL0gxge+G4nYX0L23p4zcaRDQGcPOBj\n9VzCdn+qKjU4OJ5/6x4p+G2HFH1Ju2TurAHsW2aa0q4hcac/3z4yiBfXbGizcbUqGs6rnAop9O0D\nKfqSdodBD3FrA62ei9/iz/F93kQNPD9qzPe+OJOIAXmc3G9p3okdmY3OoeOv8tuV2MfFwbZtUFkJ\nvXrBRRfR2dLTpehL2h36Ki1lhdb7ylaWO5CT4mpT9LNOubD4nZ6cPOiNs5ueAZPSmfJwQrv9Xmt1\n8K/nj/DtQ4PIT3c9ezy0dz4zXzKP/qks17L0wxjUTSKdX7kgm8sei8exnRbQa1diD/DLL7B0qRB8\ngLVrYccOmDULHDqPFHaedyppEjsXB7N+TiS5p13wDy3h0jtPMHBSyxZsc3Q2ENo3n/w1LhbnAqOK\n6HNJptXrMk+68ua0MSTH1drI9y0L4tRBb+7/ZneLjfdcGfmvVML75bPyi2gKspzpGlXE1H8n4OFb\nW2lSX6Xh3Rmj2Pt3bRGy3Uu6cXhDF576c3O7s/u3O8E/fhyWL68V/Bp27RITwRVXtM242gAp+hKb\nrPoqih+f6E9poSgkFg8cWBnEnZ/u4cIbk1v02VMejufkfm+KsmuF38FJz9jbT9gMhfzz7Z5mgi/Q\nsOXXMCbel0j3oXktOOKmkbDTl6UfxJBy2AtX70oGTU7niv+oFjuTNd9Fmgl+DfuWBbP2+0gm3HO8\nlUZsm3Yn9KZs2QJlZdbPHTkiRV8iMehhxefdzwp+DSV5Tiz/tAdjbkg2czIaDFBa4IiLR1Wz2KGH\nX3kaZ7ftrPoqmozj7nh1KWfUNSlcescJm9ec2O9t9Xh5iSO7lwS3O9GP3+HLB1ePIiu5tnbP0Q1d\nSE/w4P5vzXcm6qYAm/dRNwW0uei3a8EH+/Uy2nnYenMjRb8DUlUp1LYlt/RpxzxshhKe3O9D7mkX\n/ELEymnphz3Y8GMkGcfd8OpSzuCpp7nx7YPnPL4BEzMYMDGjwa93drWdDNUea9v//WGMmeALNGxb\nEMq0WccI61Mbmmqv8Y3Oqe1s+u1e7GsYPlyYdyqsZDnHdq72oe3UvSWxRvIhT967eiQPRU3lwcip\nvDtjFCcPeLXIs9x9qnBxt1633MWj9tyyT7rz83/6c3yPL8W5zpw+5sXSDxW+vn9wi4zLHv3GW58g\n/EOLGXd325s/6pJyyPrOpLTAiT1/m3dgGn5VCg5OlhOXg5OeEVeltsj46qPDCD4IYR83zjJSp18/\nuPzythlTGyFX+h2EolxH3r9mFKmHa4VixyI3Ug578dLGtc1ep8U3uIxeF2VatSP3uigTN+8qjEZR\nWlgkT5mze3EIOWlx+IWUN+u47DH9CZVTB73ZvqAbleXiT9uvWwnXvx5n5hRtL7h42R6Tdxfzz23w\n1DNMfiiBlV9EU14iIpuc3CqZcF8Sgy9Lb9FxmtKhhL4ut90GvXsL521lpZgIJkwAR+uRYucrUvQ7\nCH9/FGMm+DWkqV78/VEM17/a8OJeDeW2D/dRmO1MwnY/RKaQEWV0Frd9uA+AqgotWSet15IvyHQh\nfrs/I65Ka/DzivMdmDe7L+oWfwx6Dd2H5vKv5w/TJaK0QddrdfDITzs4fE8A+1d0xcWjivH3JOHp\n3/4EH2DgpHTUTZYtosP65jHmxlMWx2957wCjrk1m2/xQAEbOTCFmeG6Lj7NDC70pGg2MGCF+OjFt\nIvqKogQCu4EJqqoebYsxdDTOJFkXV4AMO+fOheCYYl7dvIZNv4aTnuBBaK8CRs5MObtDdnAy4N21\njPwMy9BKV68Kwvs1PIGqqlLD65PHEL+tVgRP7vclcZcfz69eh1cjhLv3RVn0viirwa9vK6565ijp\nCR5s/z30bF5CWN88bv94n01/SMzw3FYR+hrOG8GXnKXVRV9RFEfgS6BhyzcJAF4Bts0kXl1azoSi\n1cFFN9WuOotyHVn3QwT6Ki1jrj/FsCvSOHXQm7oNQ/qOyyC4R3GDn7P47Vgzwa/h5H4fln4Q2yI7\nmbZGq4MHf9jFtFnH2LssGO+uZVx4wykcnNo+mkSK/flLW6z03wW+AJ5pg2d3WCbcl8SmX8PNsjYB\nvLuWMuG+pFYZw8ovoln4Ws+z7QH/ekdhwv2JTH00nm0LQslJdcPNp5x+4zO496tdjbr3+rmRNs+d\n2Gvd4Xm+ENG/oF2VUJaCf37TqqKvKMptQKaqqisURZGi3whCYou467M9LHylNyf2iVDKyAF5jH8g\nnvVzIjhz3B0v/wrG3ZPUInVpUo548utzfSjKqTXlFGS6sPgthcfmb+Walw6RtMeHoB5FBITZSIKx\nQ1GOk81zBVnOTRqzKQYD7FjUjVMHvQjqUcwF1yWfF3Vtmgsp9J2H1l7p3wEYFUUZDwwE5iqKMl1V\n1dYLP+jAjJiRxrAr0ji8IQCMGly9K/j4hpGcPlYbtrl1fii3f7SPMTc0b8bs2u8izQS/hspyB7Yt\nCGXo5en0vaThdvTEXb6s+S6S4hxHgnoU4+5bQWGW5f0BYkbkAMLuv+iNnsStCqS8VEfkgHymP3WU\nkFj7ZqT8DCc+uGYURzYGYDRoASPLPu7BQz/uoJtS1OAxn69Iwe9ctKroq6p6Uc2/FUVZB9wnBb9x\naHWcFdd3Z4wyE3yAwiwX/nizJ6OuSWnSStZohL1/B5Gw0xf/sFIuvvkkDk5Gs1aAdbFVHM0Wa76N\n5Kcn+1OUW7uC9/C17pdw9y3nxjfjAPjkpuFs/V/42XNJu/w5utmf//y12a7/4PtHBnF4vWnVTlGm\n+YdHB/Lc8k2NGvv5hBT7zokM2eyg6Ks0JOzws3ru1EEf4tZ0aVQ2K0BJgQPvzxxF3JpADFUiRGf5\nJz24/9tddB+Ww6ovu1u9Lqxvw81JFWVa/nxHMRN8gKJcZzwDysxW+57+5dz87n6c3fQcXN2FnX90\ns7hf2lFvlr4fy12f77X6vLJiHUc2WC9hcHRjABknXAmMbPmYgsyTrix5P5a0o564elcx8l/JjL5W\nJlVJWp82E31VVce21bPPG2w22DCic2x8av7cxwdw4J8gs2Mn9/vy/b8H8n9r1rP5l3Di1nQ1Ox85\nKIdps441+Bm7/gzhtGo9i1jnpOfOz3dzcp8Prl6VjL8nieAYsYKPWxNoNQkM4FScbUdvRamO8mLr\nf+blJY6iEXkLiL6uuIx+z/xCwIYjJOSF8UzOjyQWRp49v+vPEJLjVK59pfWikqTYS0Cu9DssOgcj\nsSOz2bbAzeJc1KC8RsepG41weJ1lyCSAusWfj64fgVeXMmJHZ6Kv0oJRQ/TQHK5+/jDuPtbLNVhD\nq7NtcspLc+P4Hh/u/WqPRccoVzvZq66ets95+lcQ2i+fY5utJEH1ySNiQAs0YzEYGDXjPYL/2Q/A\np7xIIpFmL6mq0LHq6ygmPRSPT9fmzaauixR7iSmy9k4H5rrXDhLez7xypG9IKf964VCjG4gb9BrK\nim1cZNCyfWEYW+ZFcGxLF/QVWmbN38Ldn+/FN7hxOQJDp6cR2suW0GpY8200q76Ksjgz/p7jBIRb\nsdtrjAyeetrm8zQamPpwAu51fAZObpWMvzepRYrWhSzcQcDKuLO/78F6HaL8M65s+S3c6rnmYOZ8\nKfgSS+RKvwMTElvMK5vXsuzjHqQneuAZUM6kBxKaZK7QORhxdm3Yiv3EPl8+u20Y/7dmQ6N7uDo4\nGZnx/GG+uGsoFSVWHMBGDfuWBzHhXvMCaR6+ldz87n5+ebo/Z5I8AJH1e+GNp5j4gP08hdHXpuDu\nU8mabyPJPOWOT1ApF910ipFXN69NvapSw6/P9uXI98MpMc4ihmPczdc4Y3tidPFo+C6pMUixr4eK\nCti6FcrLYcwYcLPcMZ+vaIztvJb0fOa37wG2MEYjrPg8mu0LQynIdCYwuphRM0+RecKDilIHBk4+\nTa8Ls8/pGfoqDb/N7s1f7yoY9A3dIhgZdlUqj83b1qTV8ud3DGHd99FWzw2YdNpmVE1ZkY6130dS\nWuTA8CvTCO1VaPV1bcHntw9l3Q/muxRP8hnDRpYxzeL1wTEFvHNgJU4uzVMaWQp9A9m0CebPh9PV\nO0R/f5gyBaZPb9txNTczZ1pdksmVfjtn3gt9+OPNnmejaZLjfNj9VwgYxf/PJe/HMHJmCg/+sLNJ\nfWCNRvjo+hFsWxDWyCs17FwUyuJ3FGY82/jySSNnprDxpwj0lZaTTORA281OXDz0THk4sdHPa2nO\nJLqx80/LiqSFeFOKKxNYwSrGY0S8X79uJVz3WpxVwffef4LYd//C++Ap9B4unBnXlyOz/4XR0fbX\ntUMKfno6JCdDjx7ga9kYvkXIzIS5cyHP5G8sO1tMAuHhMHBg64yjDZGi344pKXBgw9yIs4J/FmPt\nBF5Z5sDGHyOIGpjLtFkJjX7Gzj9C2P67ZShkQzm8rkuTRH/Q5DOMmJHKlnnmNu3oodlMf0pt8nja\nigOru1Kcaz1z+DjRHAoayuejZ7M+YipuPpVMuDfRqgPX81Ayo/71Hp6JtX2IAzareKppbP/tMYvX\nd0ixLyqCL76AgwehtBS8vGDoULj7btA10hnVWFatMhf8GsrLxQ5Air6kLTmyPoCsUw2poKnhwMqg\nJol+3JrA6ixVO3fX6THaMPukHvVgw0/hDJ2eSkWpA15dyhu049Bo4JGftxM9JJeDawKpKtcSNSiP\nK58+gqefiMapqtCw4acICjKdGTTldLuqT1OX0N4FODjprYaVukcYWbn7HSL9PYnkgN37xL6/xEzw\nawj5cxf+G4+QfWGvjin0pnz5JezYUft7QQGsWQOurnDrrS377CI7GdglJS377HaCFP12jH9YKY7O\nVWcbgtijrKhp/ysdnO23EewzNoMLbz7B1/cOQV9lKWjZyR58evNwHJwM6ByMhPbOZ9y9SYz8VyrL\nPulB1kk3vLuWM+XheItIH60Opj95jOlPWsb5H1zdhR8eHURydXep31/rxYgZKdz/7a5GRyZZoyTf\ngaxkVwIjS3DxOPdWir3GZNNzTJZFHgMY6XtnCZX+ng26j9fhFKvHHcoqCVx9kEvTe53jSNuYrCyx\nwrfGnj1w000tu9qPjLR9LjjY9rnzCBmy2Y6JHJiPckHDnLTJh7x4feoYfn+9J1UVDQ+pufCmU7hY\njXM34hlYSt9xZxh720nG35uERmvL4aihqkJHeYkDibtEeYPH+05g/v/1Ze130fzxRi9mj76Ug2us\n5wHUpapCw/cmgg+i1MP6OVEseqNng9+bNSrLtXx59xAe6z2JJ/pNZlafSXz/6AAMzdBC975vd9F/\nYjqOLiIix7trKZMfSmDGc0cafI8qL1eb5/qe9DjnMbY56em2V9QFBcLM0pJccgn0tPI3FBoK0yyd\n7ecjuhdffLGtx2CXwxx+sa3H0Jb0GJnN8d2+5KS5mNny61JZ5kB6gidxa7qStMeH0demWDWzVJRp\n+eu9GJacxuudAAAgAElEQVR+EMPOP0Nw86oivH8+Sbt9qSw3XWFpqCh25NDaQLJTXLn7i70ERhVT\nku9I5gl37KQDo6/UUVZoXjWzJM+JM4keXHL7iXrDPNfPjWDNN9ZLPugrtVxyxwn7N7BCVYWGX5/r\nyyc3D+fopi5n6wWV5DuRsN2fsmIHBky0NKs0BnefSi66+RSDpqbT55IMbnz7AKNmpjYqrNUpp4ig\nFfssP92gILjnHnCo3dGdyHBlzrpITmW5oYQUNsmR3+q4u8PGjcKWX5fQUJg8mUbHATcGrVb4D0pL\noapKjGfQILjjDggMrP/6jkSfPi9ZOyzNO+2cbkoRL29ay67FISx6QyFxh/U6Mqbs/TuYByOn4Ohi\nILxfAdNmqfS6MJvyEh1vXjaGQ+tq/7g3/xbOuLuSeG37at771yhSDvnUuZuGrfPDmPpoPBffcgp3\n3woOra1rwmgYCTt82fVnEDv+COXEPh+c3fT0uTSDmS8cNmscUpBpu5SyvcJv9vjw+hHs+N12hNKu\nxSFc/1ocjs7nHj4ZPTiP6MG2I5DskfDIFDyOpRHzw6baFXFICNxyC7iIukRGIzzy3UB+2RxGTpEL\nYGRgZB4f3b6Xi3qdW/hui+PuDsOHw/Ll5scdHOCiiywbl7cEnp7CadzeKCmBv/8WoaQeHqKRe3jz\nJ+/ZFX1FUcKAT4AwYBHwhqqq+upzS1RV7Rz7oTZGq4XhV6YRMyKbty4fQ9Ju64XWatGQkyocwGcS\nPUna5cvjC7ewb3mQmeADYNSwYW4EF950ClspG2WFjuxZGkxE/wL6j88gqEch6QkNs1Gboq/U8P2j\ng82c08e2BpB62IvHF249u8AbNPU0C1/tbbV6Z2jvxpdNOLQ2gD1LLMMpTclNdaUoxwnf4Mb3AmhO\nZi7QwNi7IHYq7NwpvvwXXghOtTunD5fG8NmKHhjP7gc07Dvhy4PfDGb3W6twau99Am67TSRD7dol\nImm6dBHvcerUth5Z23H6NLz3Hpwy6Y28aZOY7C++uFkfVd+0+h3wJ3APMAL4S1GUmomi6XF+HRx9\nlYbfnu/D00PH8VD3Kbxx2Rj2LA2q/8JzxDe4nMtmqbj7Nc7umZ3ixrJPehBvoypnZbkDXz84kNQj\n1guhAXj4i/BCJ1cDE+9PxMnNTs9ajXXRcfOptBqNtGdpMAdW1U5G4X0LGTUzGTC/T0B4MZc9Fm/7\nuTY4tL6LzWJttfcuwdNOS8qWxqJkQkgIXHGFWO05mZvKluwONhH8WuKSffhxQ0QLj7QZ0Grhuuvg\nnXfg88/h9dc7t+CDyBMwFXyAwkJYtEiYoZqR+sw7/qqqfg+gKMp04BfgJ+C6Zh1FB+PLe4aw7vva\nzMuMJA/it/vy0NydDJ7aPO0B0lQPivMciRqcdzbjtbTQgd9m96M4p/GdpNLjPfHuansVmxpnOznG\nq0spF99y8uzv02bFExhdxKafI8hJcSE71Y2iHCfKix3p1rOA3mPPsG9ZCJknawVe66CnJM96d6yq\nCh2H13VhwITaUtD3fr2boB5F7FseRGmBI6G9C7jssWN0H9p4s4lnQH0FzYwMn5FCXroLq76KpqzQ\ngZiR2Yy6xrpfpDlpSvhlXqltE9eZvHPvMmaTwkJISBBRLkGNWOQkJ8Nff0FKCjg7Cxv6tGlC/J1s\nd0zrVMTbWMykpYkd0ciRzfao+kS/SlGUPqqqHlJV1agoyi3AckVRvmjAteclacc82GElmako24V/\nPu9+zqJ/Yp83c2YN4Ogmf/SVDngFlnL54ypXPBXP6q+jyDzetAgOF69KtDpb9moj9hyzUYNzLTJH\nh195muFX1hY6yz3tQv4Z5+p4dSMZJ1SWfRxD/A4/Enf4Wc28NcXdx3znoNXCVc+oXPWMSuIuX9Z9\nH8GS92MJUQqZ+mh8oyp7XnrHCZZ/0sOi4QwY8QosY9xdxwmJLeSZYePIP1MdPfOxkfVz0nni9y04\nuTZPmYS6NDXeXgkuZE+S5a7NzamScf0a10OhQRgM8P33IrY+N1fE0/ftC/feKxKr7JGcDG+/DWdM\nnOSHDkFqKtx/f/OPtT2SkQH5+SJc1LEJPilD8/791SfcjwFLFEV5TlXVX1RVraxe8f8M9G7WkXQQ\n9i/vSkm+9dXJ6fjGC/KW+d1Y+HJvMk+6Vyf3aM0iXwoyXPn56QEU5To12cmo0Rk4uc+bgsyacMBa\nkdc66DFYib83Jbxf/UlRvsFlZvbwwMhSbn3/AO/NHFmv4HeJKmL8veZF0wx62PVXMNsWhLJ7cTCl\nhbUr2B2LQnni9810jW5YMo2zm547PtnLj48P4FSccFR7+pcz7KoU7vx0L1UVOp7oN6FW8KG68Fsw\n/3uxDze9ZSOuvAk0R2LVo1Pj2XikCyk55kXCLh96mhExuef+gLrMmwcrVtT+Xloq/A1GIzz1lP1r\n//rLXPBr2LZNmHQiOoA5qqlkZMC338LhwyIUNThYmOus1fiJibH+OYWECMd3M2JX9FVV3QxEKYri\nZHKsGLhSUZRG5ysriuKI8BNEAs7Aq6qqLm7sfdqSwOgSNFqD1SxWd5/G1UXf+HMYn98+zEQUbawC\njBpWfBbDU39swsGliqoy2//bdE5VdI0uIU31BKMG766l6Ks0JoIPNYLffVg2YX3yWfeD9cJnAN6B\npUy4t+m1bvLSrfe9rSE4toAb3zqAm1ftyv3IRn++//dATuzxxdoO5OR+Hxa81JsH5+xq8DgGTMyg\nz55VbP2fKFw3fEYqXcJF2OCqryLItBEDb6vrVlMwE3yDARYvhgMHhCCEhwsxCA4WZpCtW8WqcOxY\n8DGPqBrhrTJv/CE+PDaVg5ld8XSpZEL/M7w483CzjfUsRiPs3m39XFycMD+E2HGS17VT11BaKu57\nvoq+0QiffgpHTUqUnD4Nv/0mdkdjx5q/fuZM8VmZfl6ennDVVWZhus1BvXdTFOV2IA7YWf37G8Cx\nGlt/I7kJyFZV9WZFUfyAfUCHEv3Bl50mZkQOx7bWFQMjgy9rnGln4au96l0F11BW6EjmSXf6js1g\n33LbX7LeF2Xx3IqN7FocTH66CxVlWuY8Zr2eu1Zn5LrX4ti3Ipi805ZJQU6ulcRekIW7nx2nrQmb\nfg1j40/h5J52JSCshLG3nyAgtBTrlXSMXHJnEnd+ss/MfFJRpuWb+4eYJWZZI367f4PGZIqDo5EL\nb7RsGF9RZvv/QVXFuRv1ra7uP/tMxKvXEB8PR45AbCxs314bx750KVx9NUyaJMoBf/EF7N3L6OJi\nRrs9CwMGCDOJi/3JtdFUVMDChcKenGqjBHVZmTDf2BN9e+Nyb0iJkQ7Kzp2gWvnLr6qCzZstRT84\nGF56CZYtqw3ZvPTSFpkU6wvZfBgh1LeYHF4GvKcoiouqqv9t5PPmAwuq/60BWqaYeAuxe2kQG3+K\nwGAAr8BSEU9u1OLhV86Iq1OYMbsRmZcVmuokp4ZTWujI9a8dZN+KYKuJWg5Oeu79ald1iKewt//1\nXozN+50+5olfSDm3fbCPeS/0MbF5GwAtFaWO7FwURtpRLx7731ZKCxzZtzwIJ1c94+9JwtO/djL4\n++Me/PJ0PypKxZ/Uib2+HFrXhfH3JOLhV05RXeezxkh6ggfpCe6E96stj7x+TmS9gi+ur/8lDWX0\ntcksfqsnhdmWTtDoIU03l9g05agqbNlieTwtTfyYkp8vzCv9+omdwSaTktMlJbU7goceavI4LTAY\nhADZci7W4OUlJil79O8vzBt1CQ62FL7zibQ0bMZAWyv4BmISvPrqlhtTNfWt9O8ELlJV9axRV1XV\nDYqiTAFWA40SfVVViwAURfFEiP/sxg237VjxWXd+frovZUW19nYP/3JiR2ZRWaHh+F4f3rlqNBff\nfLJBzTl0jkYcnfVU2jHVmKEx0ufSdML7FBEzIpv4bZZmB+WCLAKjzDMdh12Ryrzn+54VY1OK8xw5\ntDaA0demMPSKNDb8FM6Cl3qTk2I+GaUe8eaNqWMoyHKhsvo+C1/pjU9IKX7dSnB21ZO409/iGaUF\nThzZEMhd/93Nb7P7kR7vwVm1Nmo5sr4rH14/gon3JRHWr4DeF2WRl96w6JPYkc2XhNQlvJRxdyfx\n13uxZjuv8H55XPVM4yuI1mu337AB9I2o+1BUBCtXwv791s/v3w/Fxc23ct60qX7BBxg2rP6SyFde\nKQRw+/baEgtBQSL+3LkFI43amuhoYZaxFm4Z0Hwmw6ZQn+IYTAW/BlVVsxRFaZJXsTrhaxHwuaqq\nvzTlHq1NZbmW5Z/2MBN8gKJsZ/YtDzJrPBK3OpD8DGcm1dPNSaOBXhdmsXtJA9MdjBp2/h5GRN8j\nzPy/w3x59xCy64jz4Q0B3BlwOX0uzWDyQwn0viiboB4leHUtI+uEpc3aaNCy889u9LkkCycXAyGx\nReSkWu8glJ1sXnqhotSBjERPMhLtJ2md2OfNyv92JyfVBWvL85RDPnz38BAcnPT0HJPFyGtOgsYA\nRttmlajBuVzzsmhHWFGmZc23UeSmuRDeL7/JYZY3vBFH5MA8tv/ejdICB8L6FjDt8WP4BjU8dr/B\nTtrEJvhIiorEqt8a+fnip7lEf9Uq++e7dYPBg+GGG+q/l1YrdiGTJ8O+fWKMl156fgs+iJ1Znz6W\nE7Wrq6j/04Y0JGQzUFVVszgwRVG6Ao0uhVd93T/AQ6qqrm7s9c1N5ilXTh3wJmpwLn4htr/cRzf5\nk3rUemha3U5T5cWOrPyyO+PvOY6unszIx+ZvY1afiWQkmayA7bD/n67sWRpEUY4zXbsXoXMykJFU\nK7pGvY7CbB3b5odzeF0g4+5KIn6HP7lptot4ndjvjcFQnf1upG4+lAlNs6dUVeg4tM5e2QbN2dfF\nrelKwg5fu4LfY3gWL6zegIuHnoSdvnxxx9CzETlojPzz30weX7AFry4N80OYMvraFEZfa73KpT0a\nHZHT2FlJoxF13hMThZO3LkFBzbt6tLcL8fAQmaONfQ89eoifzoJGA48+KkJdDx8WPppu3YRvppmj\ncRpLfaL/KfC3oiiPATsQ39ChwHvAV0143rOAL/C8oijPVx+boqpq45u6ngPlJTq+uGsI+5YHUZzr\njKd/OYOnpXHPl3ushkW6eVeic9Q32Ol66oA3acc8COttv5Wfk4uBTxKWs+KzaH6b3ZeSfHurHwPq\n5gBqRDI9wVOsiG1QkOnCn+8o9YZjHl4XyJd3D+H+b3fTc0wWARElZJ1sTgdb4yaLurupugTHFuHi\nocdohLmP968VfACjhiMbApnz+EAenruzKYNtNE0KwbS3Ivf3F52cTBk8GEaNEhUq5883j9vWaOCC\nC5o3yWn0aNvmnbCw1qmPcz7g4QEPPyyc4mVlIhqnJYvJNZD6QjbnKoriDPyIqL8DkAS8q6rql419\nmKqqjwKPNnqUzczX9w9m86+1XvHCbGfWz4nCyVXP3f/da/H66CF59Bieg7q5YaWBXTwq8fRrWPim\nRgOTH0oiabefRX9VM7RGqBsmamdFDNQr+NUjYN2cSPxCS5j5f0cYMDGd1V9br3DZGDRaA1qdscET\nZUMJ7ydMHIm7fInfZj2C5+jGAPRVmnp3Wk1l5nyEnXrBApGhqtWKOOsZM0REzrFjwokXGyts2nVN\nGYMHW7fPBwXB7Nmi6FZSkqgr36uXuK9WK/7r6CicwDk5wp4+cqR4RnMyZYqIy0+vE43m4tLyTU7O\nR5yc2lXmcX3ROyHAFKAI+B54UlXVFsj+aD2Kch3Zv8K6uWHvsmAqSvdbZGBqNHDzO/v5753DSD1S\nE1liO4tVX6njmwcHc8PrBwlR7HTqMeHSu5LYOj+U8mJrsfpGMLRcYwmjXsvCl/uQdtSLQdPSWP1N\nJBjP7XlDr0wlJc7bShZs09E66Dm21Z+sZBeKchxtTigVJQ7NLvoOecVcddcSEb7o5CTCKzMza19Q\n46wsMyl1cfCgiNR59lnzWOvJk0VY3saNwgELYut/yy2ivO9tt1kfhEYjYvmnTxcOwqbEb2/bBuvX\ni8xaPz8RQVPX3KDVwquvCtPEgQNQWQlRUSKWPNp2ToekY6Ax2gorAhRFWQHsBjYA1wJGVVXvaKWx\nATCf+c26XEs+5MnjfSdbPefoUsXH8cvxD7VubSov0bHg5V4s/SC23gJeIKI/Xt2yBhcPPYU5jvz+\nai8Sd/mh1RnpNSaTq547albeYNMvoXzz4GBKzOqnmL59a5OM/RIKjcMobmXUnPN9Jz0YDxpY8ant\nkNGmEjs6i9n/rOc/gyZwOt5yUuk/MZ3ZKzZaubJpuB3P4LJL3oGTJ+t/sTVuv12snuty5owobeDh\nAWPGNC1FvzH88w/8+KN5o5KaFoWXXmr9GqNR/LRHk87u3SISKi9P+DQmTgRFaetRtR9mzrT6Ba5v\nqdBNVdVJAIqirEYkU3VoAqNKCIwqIsNKDZuu0cV2i5I5u+nRaGiQ4AOcOujDsk96MPmhRN6cOob4\n7bXOtsPrAknc5cfTSzah1cGqr6NY+30EleUaNFo9RoMGUQS1PuHV4OFXRkm+Ewb9uX4xNSZzzLlM\nJEZ6XZjF0CvSKMhwZu/fwZQV2RY0rYN4v/X16q3h2BZ/dv7RjYn3J/Lr7L5UlNTe2yeolGmPWbZf\ntMbJA15kLCwnJCqfkFuc0WjN3/NZe/3nC5ou+CBMQNbo2hUuv9z82MGDIiEKREhk375Nf64pBoMI\n+6zbmaq0VJhyxo61LuwaTbuwQ1vwzz/w00/mO6sDB0Q9oKFD225cHYD6RP+sYbq67k7j6gy0Q5zd\n9Iy6Jpk/3+5pluCk1RkYfV3y2YqWtmhs7aP0BA8Wva6YCX4N+5YHsemXcMpLdPzw6MAG9cK1Ruyo\nLB75dTs/Pj6g2h7ftl/SvuPO0GNkFr8915eqci19x53B0VlfXTHT0lndMN+DKRpSj3px3SuHCIgo\nYdMvonl6l8hiJt6fQOxI+xbIijItG6cYuX7D64wwbEOHgUOPDiD1vSmU3NWf1CMe9PnFjeLYbNxd\n9LZFu6E0xJ5rNMJ338Hq1bWx3StXilotd9xx7sKblmZ74kpOFs7jLg3zWbU5VVWiCUtZnQVafj4s\nWSJFvx4aqzLtvDtDw7jhjTic3fRsWxhKbpoL/qGljL7+FFdYadBdl2HT01j2cUyDk6q2/96NcptN\nyzUc3ezPyf0+TRZ8gD1Lu/HiRZcy7u5EdI6GZneeNhZ9uYZXJ4wlPb42nNTVq5IRM5I5vteXk/tr\nIm4shczFo5KyIke8u5bi5Kon00p+AUDX7sJXMmJGGiNmpFGY7ciW/4WRkeRB9JA8s8k7J82Z9XMi\n0WjgoltOsGp2V95YdwMFePMAn5NId/wLspl03wp2vRvIz8dHU1rhSFRgEbdefIL/OxfBdXSEESPq\nf93OnULkTVcVVVViFZ6bC3feWX8ilDXKympt87ZwcxNmno7C8ePWQ1dBTGzNmah2HlKfTb8cME0v\n7Vb9uwZh329xr05z2/RNMRqholSHk6u+UQupr+4bxKovozHtQaPVGZpkXhk8LZWE7f4UZDZD7RSN\n0W4fXfs0s2/Ayr26RBUROzKbnX+GmJlkTBl6RQrTnzxGt54FHNsSwMc3Dqe0Tr/d6KHZvLZ17VlH\n7YKXe/HPF93P1g8K65PP9a8fZOj00/zxpsKS92PPfr6eAWU8V/QyI8s2cRM/kUJtOzoXSriBn5jH\nDRQjJhtHnZ6Pu3/Mfcdm2X/LWq0oOaCqtXVzXFyE07YhSUz//S+sXWv7fNeuIsmpsTbrt96yXTCt\nhhEj4PHHG3fftiQtDZ54wnq2q68vfPJJu4qWaTOaaNOvp7BGx0ajEeaexrB3WVcSd/lRI2qunpX0\nn5hOeoI7J/fX18bQkuN7fXD3LW8e0W+S4AvhFDuE+nwINfNv/X4Ga2Qe96i3H4CrVxU9LxBx6kMu\nP83N7+9nxac9OLnfB2f3KnpdlMVtH+xD52BEV1JOwl05/DFvOhWGWrNR8iFvvntkEOgMLHylF+Um\nE0xhlgu+ZPAuT5gJPkAZbqxkEmsZy318wR6GUqnXMV87k/u6f2Y/k9ZgEJmW114rVu0g4ufDbPfl\nNaO+sgxnzsBHH4n7ubgIoR49Wpw7flxE5Wg0wiEcGiqOq6r9Fb5GAz17CkdzRyIkREx+hw5ZnuvZ\nUwp+PdQXp38O3qvzj7wzTnx93xCzln+lhY7ErQnExb3xGaAAuanu9L44g9MN8z22AEKgG2YSallf\ngUZnYNh084Jj4+86waW3nyBV9cTduxK/bsKO65BbxIWXvcGvW1+jAks/QdZJdxa+1NtM8GtIJpz9\nWK8MnkwECfTgLZ5mAqIcwZliD3jjJWFH/v332oblpmi1YrXu5CQE6cYb7VefrEvv3iISxR5ZWeIH\nRNTP8eNitbt6da19e/ly0ZXq6qtFglWljb9LLy+45x7hLG6Pjtr6uOUW0WrR1E8RGyuOS+zSKbtf\nNYWyIh1vXHah1R6vxbnOaB2a3t1GuSALn+AStswLt0i40ur0FqUezlf8Q0sYdpWwJhqNoj2kk4se\nByejRXZz75cXErA1njx8rN0KgPxM6xnOH/MwflivdKiligCyGM1mQkkmhTCiupQIMZ8+XTgL//rL\n8kKDQQhvWZlY6WdkwGuvNXzVOXasiNrZ1cAeAXq9sPdXVprvEkpK4I8/RO2X4GDb14eFnVs5AKNR\ntE90cmr+ss4NISpK9NZds0ZMhGFhYpfTHkNL2xlS9BuA0QjvzxzF8d22zTee/pUUZjbeGebgpKff\nuDNc/1oWVz1zhO8fHcjpeC8cnIx065nP4XWBVJQ2t+g3p/2++cg65c72haFghOWf9SA5zgtXryr6\njM3gto/24e5da8P12yHKBMRyjFVMsHq/fBsNXArwJdovH3Iszw1lF+NYgx4tLpTi5VrBXeNMiufd\ncIMQ1p07oaBAZM1aM82cPCnCCqdNa9ib12ph1iwh5AsXigJr9VE3eqWGigpRKdOe43fUqIaNyxrb\nt4s6/ydPimzjnj3FCru1q0c6OopaNpJGIUW/Aez9O4gDq+wVDYP+E9LpdVEm2xaEUpzjjEZrwM27\nkqoKDeXFTtgS2vB+efS9VGzZI/oX8eJaUS/9yAZ/PrxuJBV2mmA3nfYn+AAYNWz6OYwjG7tQnCtW\n6cW5okRGXroLzy7bVGuJqP7Hv/mA5UwiCctiXpVljvi5l5NTbL7iH9Mzk/mP7efGT1xYFxeAofpr\n0JMjvMOTaDGyVzuQ4BgfXp60myuGnRYz/5IlwilaXCzKLvToIWzmR2z0UahbxqA+HBzgssuEWejz\nz21X1WwIlZW2dw0ajcgAbgqHDsFXX4lVPgin9bZtoizEyy/LlXYHQIp+A4jf4Yehyt4fsxF1UwAz\nXzzMlU8fZe/fwXSJLGbw1HSyk1347uFB7FocavXKoBjrK7q/P44h97T1Msfti0buGuqJMEo96nVW\n8E2JWxPIgZWBDJgoCr5mj4qly2aVGBK5k294jjet3s/FqYpZY0+wLd4fMDIyJocXrj6Mt3sVK2dv\nZOGv5exblk5gxSnu5hvcKMXo6UmfWy5m/UXrayeZ774z7xN78qSojxMZafu9+jXesc/GjWJyyc8X\n4uzpKX6sda9yc7PuXwAxKVlzdIKYwFJSROnfxrJ6da3gm3LsmBj7xRc3/p62qKgQpqr4ePFZxMSI\nlX19zdgldpGib0Ka6sHyz7qTk+qGb3Apkx5IJLR3IT52snQFGo7v9ePtKy7At1sJt763nyHVrRMD\nws0bhlu7toac086kHfUivF8+aUft16lvPzRu1zD4sjSCuhex7NMYjHVCXL0CS9E6WveN6Ct1JO7y\nPSv6h1+4Gr9dSQSuO8QIdqCl6uyK3eyernreu9V6BItWCzNvdGbmaAOsPALZPcHXF824cbjEmJSP\nyMy03ukqN1eEUnp4WJpjgoJEuGZjUFURU19zL6NRmJDc3ISN/qBJg3ZvbxEptGmTZWcqnU44dG05\naN3cmp7pW+NItkbdrl/nQmUlvPmm6MNbw759wvQVGyvMZm1corijIkW/mgMrA/n8jmHkpNSurrf/\n3o17vtzNJXecYMVnPUg5XF8bPw25qe58dMNIctP2M/XfCSTs8MW3W6nNxiD+ocXMfbwf+1cEkZXi\nSmm+M95dW7XSdKsRHFvAg3N24ulXiYOzkdVfR51d1QeEFzPzxUNsmRdG6iErzlmNkaAetcKq93Rl\nw4rniP5mNVE7EwlemEFqoWW0zNjeGRbHLIiKEpEstti1y/rqFsQq1M1N2JcrK8VMoihw3XXieGNY\ns8a6LT89XTh6x40TE4Orq/h3ly7CefnnnyLGP6faSaHX2y8bERRkad7JyBB2+owMsbMYO1ZEFNXF\nx7bjvFkzev/5x1zwazAYRLPx1NRzm7w6MVL0q/n99Z5mgg+Qd9qNP97syZBp6dz79W6+fWggJ/f7\n1FsjxmjQMu+Fvmz6NZyk3b4YDRpcfSoozXPCdGXsH1bM8k9iLLJx88/UOITbp8O1aRi55qU4PKub\nrN/01kEm3JfI1v+F4eSiZ+ztJ3DzqkKjhUPrAqkqN3dex4zIZuSMZGEaql7BGp0cSHxgEj3e9ia/\n0lJwHLQGrrnAshF6o7HnENXrzSeE0FB4+ummZbjas+Hn5YnSyjWx+TW4uMDUqWLCaCglJWKCqinw\nFh8PH38scgFq2LEDbroJxo83v/bii0VZ6NI6C5PIyObteVtfu8bCQtHhS4p+o5GiD+SlO5O407r9\nNXGHH8f3erP0ox5kHHevFvz6xbisyIHEnbX13kvznAEjbr5lRA3MIzvVlfRjXvXcp77ndJxJwTuo\njFHXmNulu0aVcOV/VLNjY289SemBclb9GENyZhBOrpX0G5DMex5PMjR6PUYHHVkXKERecINoOAI8\nvCaKojJLh3eVQcv/toRxSR87JomGMHy4KCmcZL8FJgCnTsGyZUKgG4u/9f4AgCi5bIsTJ4SpqaGk\np5w5eqcAACAASURBVItV9KBB4vdFi8wFH8TEsHSpEHLTEs5DhsDNNwvz0alTtXkJt97atFLPtmhI\nxVHT0taSBiNFH9DqjGh11qs9aB2M/PRUf+JWB5kcbYjQWnuNhpJcFw6tC6y3AUr105vwjLbE+iTk\n4FzFSxvW1hvYoamoYtgdn3PFkj3o8w1s043EJcaB/ml7cT9V203K43gGbE0T0SKOjhTYiXAqsFHu\nwSoGg7DdHz8uTBwTJwoTglYrsla/+aZh1TZPnKi9344dwmkaHl5/ItTEiSI6KKdOLGlkpDhni5AQ\nUWumpjZ/fWg0tbVpDAbbk1lqqljVDxlifnz8eJF9fOqU+Hy62o9saxLDhgl/hb1MZe/6zK0Sa7Sq\n6CuKogU+BwYA5cBdqqqeYwnDc8erSwWxI3PY/0+QxbmwvvnEb7WzAmsKDRL8tuDcdg4uXhXoHIwU\n59TGx7t6l3Pvl3sIibEuSEYj7P4rmMPruxC2aS/DdpzECRGRcol+HdiqIpCYKCJJJk+mT5hts0iv\n0IKGDb64WPR+NbUjr1oFd90l+tMqCrzxhghXXLfO/r1cXcUq9OOPhQ0eakse/Pvfts1FERHwwAPC\nRl/TOasmu9dekpe/v6j7s3Vrw95r9+4iEqYGnZ08EFvP1emEL6SlGD5cOMJXrhRRPNbGdeGFLff8\n85jWXulfCbioqjpKUZSRiF67V7TyGKxy/esHyTzpRppaGw7WtXshfS89Q9KuZhb9dsu57Rw8/Sq5\n+8tdbPwpktw0V7wDy6gs1/L7a734bXZfIgflcsVTKt2HimzYqkoNH147kl2LQ6qL1Sks4FJe5Tnu\n4Zv6H1hdafGhyYks2BbKzkTz/0+D/I7z6OQ69S1KSsTq1qNODaBffrF0HGZkwM8/C0HVaoX54vbb\nhWPXXvLU6NEiCkc1MV0ZjSKe/9tvRbEwW/TvL36KisTzGprtet994r8HDthf8Ts6ClNMzY6jxvFs\nzVQSFdW0sM7m4tZbhaP6n39E5FJenig70a2b2G2cS4JZJ8Zulc3mRlGU94Edqqr+Vv17qqqqdrNE\nWrLKZl2Kch1Z8l4Mh9YH4ulXzp2f7aG00IlnR4yjrLCFuxp1EOw1iHdw0uPmXYlPcBnDr0rl2CY/\nDqw2LwUQGFXEf/7aRFifQha+0pN5L/SzuE8wqexnAF3ItjhnxtVXwzXXAJCeVML/verAtqK+GIER\n7OBF/o9uoToR111WJgS/qEgIcHQ0XHVVrSNw1izb5Xoff9y8PPLixaKBhzWio+GZZ+CRRyydnSDE\ndtAgkdkbHm553hrFxcLmHhRUf0RQdraoAWSt0JpOB+++axm5k5UljpuaeQICRETTQOs1ihpNjc6c\nS52flBThwO3Ro+W7jJ0PNLHKZnPjBZjuxfWKojioqmqlRmrrs/mXcDb+HHG2hnvKYW9mzD7CwMnp\nbJtft1riuTpRO44T1pTx98RzaG0QKYctQ/eqKnQUZOooyHTh1AFvwDLmPuO4B39/FMO9X+0hbq11\n5+RpuvEdd/If3rY9kJr2eNUErZ/Hl0XLLF9nQ8c5eFD0qZ09W9jErZkQaqi7qp8+XQjxkiXmBc0i\nI4XgFxdbF3wQ4rdnjxDxV16x3HGYUlkpdgZ79ohVrp+faBBy++22TTL+/qL2/nvvCZt7DU5OIhrH\nWiZuQIAYy6pVwo7v5SWSoJrDZn76NMybVxuN06OH6LUbaj1Z0S5NuUZiQWuLfgFgmnWkbS+Cf3Sz\nP78825fSglobZnqCJz8+MYDnVqzHzbuS/Su6UpTjTHCPQjz8yzmyKQB9RVM/wo4n+E5ulVz17DF6\nX5zDd48MIj/dXliiBrAuTGmqELqqCtu+jVLsmDUiI0UcvGnMeFPaGWZliSiUO+4QK/QMKzH9Pj7W\nk4Cuv15EtqxaJXYQYWHC5ODkJIQ8IsL+mFJTRZTPzJm2X/Pdd+ahmDk5wtSh09kvhxwcDC++KKJv\n0tKE03bsWJHUZAtHR+t9fOuSmVlbDXTMGPtO3JISy8knM1N06nrpJeEsl7Q6rS36m4HLgf9V2/QP\n1vP6VmP9nAgzwa+hMMuZ9XMjuO/r3VSUainOc8I7sAytDt65aiQ7/2hgvfQOgNZBj5OrwcSUVbsb\ncfMp59YP9uMXUs6omalED87lny+6k53iyvYFoegb0fLwyIYAVn0dQdTgXNTNlvH1nhQwg9/rHPQU\nNlyDQYhrdJ3+PU2tob5njxDRAQOEeaOu8OflCYGaMMGyuFdwsAhfrItOJ15ftwl5XZKShJ0/JsYy\n3LGkRET+WGP37vqdux4eImO3Llu3iokkI0Os5EeMEHH+DTG7/P67mEhq8hKWLBETRbWJzYK//zYX\n/BpSUsR9rruu/mdKmp3WFv1FwARFUbYg1KTddG8ozrdtI9wwNxLfkDKufOoYTq61JRUCo23UPemg\nGKp0uHqVETEgl9iROUQOzCXrlDtOrnpCexewZV4Y2xeGEhBewpSH47n5nYNUlGqJWxNIQUZjkpF0\nfHPfEF7dtpozf3Vh74naaBatxsAtHovoX2jiVNXpxEp05UphHtm4UYQRPvRQrVj26SPS9BtLTSaq\nk5NoetK/vwjb/P/2zjw8qur849+ZzGTfCRC2EJZwgoAsoiCroLiiqK2KC4pL61Ztte5tLba2tYvW\n2rpUhVr3ij93RWWTRUHZRJBwIKyBECAhZN8mM78/vrnOnZl770y2yYScz/Pkgcy9c+fMncl7znmX\n76vXtNm/nwY8P5+Bz/h4VsRauRvOPptukldeMc8n37iRBrxvXxrPGTq10F27zKuAS0ooz9BcVcuV\nK+ku0lxPhw5RM6e8nDsXK7ZvZz6/fhKrqmKmUW4u75s/VoJzRrsqRVgIq9GXUroB3BLO1wyV5O7m\n+jhVpTF446ERWL4gG40NdiSl1+Oie/PgqrPBGedCQ02w29hS/317+P2tr1l6MAFlh+Mw6aoCTL6a\nDvGVr2ThH1eOR+UxrxDaxo964fb/foOMftWoq2r+18jtjsK6Mz347MrH8deia/DdvhQkxLhw/pgi\n3DC2Efj4Eq4IY2LoJinQVdbW1dEwp6d7m2bsaGUXGk2OeNYsY398fT2wYoX395UrudpOSODqubKS\nK//zz/fq2I8fT4P+yCPG1bZacPPAAQaGu3f3Bk7fest8rLGxzfe3ezycNP3fm9vN933xxdZVxKtX\nG+9aGhr4/o2MvlW8QvWw7TC6fHHWnk0peOXek7H9ywxYGURPox2HJP/QjuwG/n7FRNNz/UnrU43S\ng83/ktujPHA3trXRD349t8uOr9/ui3Nu3Y2v3+mF//x8VIDyZfH+BLz351wkpdehrqplmRSbKwai\n+yuX4C+3VAPX6NPvkrwrz7w8FmEZ8cUXdE80Npqv8vV695oLwyxjraGBGjahZLRVVHAVX1/vDeZ+\n9x3w5ZcM5mp58H37Mmbw3nss+jKjpoaTimb0rc51OJqfvVJXx5W9EUePMr3UKlPHyk1lput/1lmc\nLMr9aiWSkgLlHRRhI1KrhMJCfa0dT193GrYuzYSr1oHmrapDO9fuaMQlD+XBKJMFAGz2RgDGVYct\nabTeVpQdjsFLd43EE5efbih1DFCiYt37LdRlB+CEi8bOSjfmyBHzqszKSq6i33jDvC1g794Mll5+\nOXvMPv64tVukJEiaqJ6qqsDXraxk+qNb93mffjoLux55JFA7R4+2G1i+3DqbqCXNSqKjzdM9Y2KC\ni6UNGtT8Y337MuCs7xPcty/z7/v3t349RbvRpVf67z0msH+LhWpgG5DSow4nTS6B3QG4DfKUPO4o\ncIcRWSmc8an1WPr8AHgsWjU21NuDrPIbYZbBAwD70Rf/xRxct/0t4IkngKFD6dfWBzVHjaIrw0yM\nzOWyFufyeOgfLymh/zw93VoeuC0oLWV3KX3xkN3O91dYaCzTDHj1dcwCuBoXX2x+zO1mzENK7gYm\nT2aapN1OF4yRnz03N3hTlbPO4nvy1+jPzbWWiJg4kbuxzZs5ttGj21ajR9Fsuuzd37K0Oz56wiKF\nrY3oMagS/755TJAmLJFj7AEgPqUeyT3qDJuK+5yX3IC6SuNzklCGC/A+Psd5OAbjVWQNkvA0bsec\nuldhX7uWHZi++44Vq1oeekoKUwM//th8IFqFrVGV7KFDvkVXu3ZZvqc2Y9s244rRqVMpH5Hvpz6S\nluYN5FpV1GZk+BaK6Wlo4E5m40bvY8uXc5K49FLGP8rK6Aqrq/PKPBjJStfW8p7v3csYwrhxwP33\nM4Nnxw5Opjk5vG6wzCmHI1C/R9FhhLUityW0V0Xuw1OmYPsq8xxjR6yryeXTOuyORribkc7YkcTE\nN2DAKcdxzm352Lc5Fe89NtT0XHuUG26PB3Abv7er8V/8Hg/DBuC3+C1exlwYeRNjUY2dGIK+0Clw\njhjBvPKJE7lC9XjYjPyDDwL9wxqnnML0Ss2oOxycDNzGbrWQycho2c5g5kxvkNmfkhLGA6Skoe7f\nH7jwQhrhDRtYOKYPXOu5/35zA/p//8dCKH8SEuheymzSltqzh7GSvn15r/3TNcvLgT//2XcH5XBw\njMGyfBSRQ4RU5EYE21alQ64292EmpNWhpiKYofZ1x8Sn1KG+xgFXfZTuuKfTGHwAGHluEdL71OL9\nvwhUlzkQ5Wg0zb+3ijek4BgexcPIji4C6uvxFzyI93ApyhHoSktCBZLgl5q4ZQt/liyhBEJyMqtg\nnU5q2hgxfjxdGe+8w/xxszaCzcXp5Ov7Tzbdupn7/202a235bt0ovNbQwHhFTAyF3FasoLvKjNGj\ngTFjzI+b9eqtqmK2kZZPP2CAtVjau+8GusxcLhaGTZ/ePqqaXYnyct7PtLTWyVK0kC5n9I/sjccb\nD46Ax0zp0ubG2bfm490/mghN2dzonVuO4dOOorbSgcqSGPQYVIlp1+/Bgp+Nwc6v0+F2RcEW5bb0\nh0ciO9d2Q2mhf7DPO7nZ0Qi3hY9eYww2Ihv7gdMmAbW16LljB6ZUfYWPGs8POHcqViIFJqv3vDyu\niG+/nb/PmAFs2hSYqTN2LA1+cTGNW1sZfMB7LU0XOj6er3fllVw9azLKesaMCU1Xx+nkz6pVDGb7\n77ptNk4QSUnsYnXlldZGwmpXYyVR7I+/60mjqorZOD/6UejXUnjZu5fCflLy8xg4kK43q4m8Hegy\nRn/n12l4/cERkF91C+jKpGfwuBJc/kge1n/QBwVbA1emw6YexW+Xrwx4/B9XjfOpLu1sBt/ucKO0\n0ChP24b+I0rwcMM83LP9EZQieLPvIcinVMKcOVzN1NTgqSPVOPZCMdbs6AYPbLCjERPxJZ7CHdYX\ny8ujMbTZ6GK4917KF2zfzsdOOokSvP/7H/PQrdQvW4J/ALm2ln7wtDQGZo2MfkEBzwtVIXPTJuM0\nUY+HrpwbbwztOoMHG7cYjI1tniKlVeODYE0RFMbU1FBqWx9f2r4deO454Fe/Cms2U5f4BKvLHXj6\nutPw/fKecNVZpWZ6cM6tuxDl8OC8O/IRl+ybNpfSoxYX3BWYKVJREo3vFlt0NuoEOONcMLsvD9b8\nAXO2P4OeOGx4XE+srRaXTSykkJmmGx8XhwH9PVj1u+V4c9ZreKTv81jY++dY0fMK9IoK4i9vaPA1\niE4nXT333ccJ4IILKHX84YetN/gxfqmpRgbO5fJqz5jl0h85Qk2fUFfXVudp7p7KSrpXvvjCPJ3z\n4ouZTaPHbqdLJjs7tLEA5ho9yclt2xKxK/HZZ8YqrseP83MNI11ipf/Z04N8dPLN6H/ycUy+hgG0\ns366B92yqrHiv9koK4pBt6wazLh5F8SEYwHPO14Ug4riEFd1pnggJh1FfHIDNn2SCatUR1NsHiSk\n1jXl1TfHV+jG1RX/QTKq8CTu8nmuDW5MP/wJnHDjfHyC7TAP7qbYynCH+Bxn3jmchrqoiFWeTdWj\n9mVLcPnnr/hWhcbGcrW+c6dxkU92dvDV5ddfW/vCQyU52Vcywcxdop1jlUu/cCEzZ/r25T2ormZa\n5MyZ3mraggKeY1UbMGIEpQ4WLfJ21Hr3XdYeTJrke258PFeNixZR18fpZAeq8eOt37c/P/4xn79F\nJ40VF8dJxapf8IlCXR0zlFJTfWsMWoNVa0f/TmntTJcw+scM3Ra+JKbX4aL7pI99GX3uYYw+N/jq\nNqN/NRLT6lBpUMRks7uDNlIHPDj1kgLc+87X8HiAm3rMREWx8ZhTelTDEeNGyYF4nw5cUc5GXPOX\nb7H4uRxUlTZ3ArJjIWZjH7KQicN4AH/+4UiqowI9a6mT8hgeQBmS8T5moRg9kIhypKIUl+MtxKMW\nV3teQ67cAfxlDA3Zvn00FoMHMz9+7dpAw15byz+yH/+Y0gN6Q5qaytz1Tz5hBoqWPjhjBq+nv0Zr\ncTpD77mqGb7sbPNWgw0NTBfVV8GuX890ynvuYVzirbes0zPHjeP7XbjQ974cOgT89790M/kXVcXE\nWOfxh0JMDPDQQwws79zJ36dMCRS5OxF55x1OxIcP8zuRm8uK6mB1DMGw6n+cHtxl2pZ0CaPfI9u6\nd2iUsxFZw49jzEyTMvUgLPjZaEODb3e4ceqsA9iypCeqy4yrWokNu9Z1g7sRqDruRE2l+ceS2qsO\n/YaXYfVr2T6PNzZEYfNnvULa0fgTg1pci1ewE0PwE9vzeMxzH46jGzJTa3DrjP1o/KI7cLQUTrjw\nIn6K3+FhrMHp6IkiPIKH8Tju817MA6YdalRXGzf00JOfzyYmmZkMFFZUcDdQXMxtsT9Ll1IbfsoU\n/t6vHw1qawh1p2CzefPkL7qIxrs5K7WCAsYftm0zNviZmXw/w4ez6OnZZ413FGVljGFcdVXor90c\noqLoFpo+vX2uH4ksXw68/bb3u9DQwN3OM8+w30Br4hnnnku3YGGh7+MpKWGXpIiaN29eWF+wuWzD\ntnmtvUb2qDJs+LAXyo4Yr4A9bjuO7kvE4fxEnH65WdeNQL54qT+e++kp+HZRJozcKX2HHcfDS1fh\n6N54FObHNwWQjd0uNeVOVJQ4seDOMag+bj5BOJwuFGxNRUNd4MRQXeFotg5OH+zDSkzFXLyMPihE\nHGpxc+pbGHjxCDx1cx7OH3MY0XXlaNy6DVGgbz0JlTgJ27EYM+BAI87Gkma9ZgAeD43Meecx4JiR\nQReFWRZObS396dOnczWWnc2J5fhx3/MSE61dMACDwxMmGEsAa8djY2kAevSgvPKll3LM8+dbVwOb\nUV5uXmGcksKUyKIirjb37DHfgWRlhT3z44Tm9deN9YmOHaObrjWuHqeTO94jR/jZ22z8/aqrvN3b\n2pphwx4xerhLrPRj4hvx8zfW4rUHTsaOr7qhsjQa8AQa3+8W90Tx/lhkZAV3Fyxb0B8L7hiNeouq\nVXdDFF76+SgseX4QgvvYbVj83CC4g2T9lBwwbzxRX2FHt36VKCmwUDf0Yzy+wXBs83ks7fhe/PTg\nI0AKM2uiLpmFffsB11ffoA8OogiZWIwZWIZpeAXXhfxaluizTswUHfUcPQo8+ihX3eecQ5Gzd97h\nriEqitvyqCj6v61wubi9zsw0ligYOpQpo8XFNLJaRs6nn4beiNwfq7TLgwf5A9AdZkUk6dfU1HAH\nVlVFI9aRvXVbitlEDFjLRIfK4MHAb37DScTlomtO5em3PXXVUag8Fo3eohIPfPglju6Lw70jZxi6\nW6rLolG4Izk0oz9/gKXBB4CYxAasei0LoQZVgxn8YAx07cLE2UV49bnJqAmxp+8EmOjAbN/OQKbd\nDthsGDgkGrVf7cdbuAwH0A+nYw1uwfOtGq8PeleHWatBf3bu5M9XX9FP7p/auG8f4wHBJpANGxgn\n+N//fHcGKSkMvKanB/pdjVIjQ2XECPr2rfz5wRACmDat5c8PhsvFe7etaUGQk8OKXCPJhXXrGGPQ\nNPLff58B5Dvu6Fw6O927G6fgOhw02G1FmH34/nSiT6R51NfYseCO0dj8eU+UH41B5uBKTL5mH2bd\ntwOZg6uwe0Og0U/rXY0BY0qDXruhzo7D+cFavXmQObgSu9dbBHDakCg04Or6V3BV4Wr0ev/3WPVK\nFqrKopHRrwpHC+Kw8b3eaHT7ftzdcRi34VnjC9bXM5VQ82PGxSEW9bgWr7XPG9A06IHm90LdvRt4\n8002VdHweBhwnTyZlb1WlJUxc+eSS3itsjKvK0cI4+dYFULFxHCicToDVTgTEhi0Tk9n5XBziqb0\n14+JAZ58kq6w88/3irW1BW43BfD0cZKNG1kz8cADvoa8vp69APRNUVwu7oL69DHvqhUqBQWcvI2C\n1m3NWWfxPfqn/g4bxon6BOGENfrP3DAWX73p3f4WbE3Fm79ORnSsG5Ov2Yf9W1J0kglk3KUHkZRu\nItGrwxHtRlK3OtMYQXxKHSZeWYDTL9uPr94MfaXfHHLxPWywoQg9kYUCXIE38QD+jPL1vXFkWgIq\nj0ejttIBmx246V/fYt9Ne7Hsp0nIOzAAMajDBHyJP3d/ArG10UCFwUo4O5uG89VXudqrr2cmTqir\ncI2EBLpZzDRzAPrezzmHvvq1a2nAsrONV11m6JuoLF/OIGdhIcfco4d1p6bKSuDpp2nMhKA7Rz8J\nGZGT4xuw1nA4gDvv5Pv+5JNAxcyqKk5Qd91FSeJvvuEEUVXlXVUHo67ONzi+fj1X1WYTVHNZudI4\nMP7dd8wpP19XWb1ihblO/5YtLTf6FRUMYm/dyu9FQgIL1W65pf12D6NHAz/5CV13BQX87gwbBsyd\n2yFumPYirEZfCJEC4FUAyQCiAdwtpWyhY9Scovx4fLso8I/W7bLjyzez8Ic1y2CzA6tezcKRvQlI\n7VGLsbMO4YrfhbZlt9mAUecV4UBeYPeivsOO48GPV6N7/xpUlERTmKyNq3OjUYPH8CBm4mNUIAnJ\nKIe9Kch6X/GjeOGnp/yQJrplcSa2fZSO/5x+F245fRcSCrvDmZqApGQbMGAqsCeLQl16t0Z8PI3w\n44+zWlSP3R66iFluLouoYmK4Ci812EVprRCfeoorRG0cvXtTCri83Otmamjw+rv9qa9ndeOWLfS/\nawVdWjC4Xz8aJ6ssHZeL0sHPPcfG4jYbjc/339Pnry9wmjmTBsnfzTNlCl0bNTXm3by2buV1x4/3\n5tAXF1NMzaxFohVHjjCe8eCDzX+uEWYaPkBg4NrKRWXkVtu5k67D3r0ZhDYzps8/7zvxaPpBcXGh\nVyi3hNNP509tLSeXzuSeCpFwv6O7ASyVUj4phBAA3gDQ5ukHO7/uhuoyY7nX4v1x8HiA8+/Mx3l3\n5KOh1g5nrLvZE/lVj21B5bForP+gFyqPxcIZ58JJU4/itgXrkNaLX/akbvUQk4qRt6ItBarcuBP/\nxAUxixBV50YqvMGnbcjF6+WXBdQF7NnZHf/bOQHPaK6Z6Gga0dpauhmGDaNxLSig8ayuBl54ITAb\nBqABzsmhQR84kIFNp5OGbPlyPjcmhquy225jYHX+fGODD9C9YXSssJBG47HHvMVM9fX03RsF1Wpr\nrZuxHDrEiWzZsuC7FSk5eWzaxHhBaSnf49ChlCHu0YP38IEHuCrcuZPGYdQoGv29e2nwzSqEKyr4\nk6RzEWZksFXj22+3rO4gP7950g9WWHXl8jeCp53GrmBGmVb6QHN9PSf2b7/l/+12fo9uuy1wV1VS\n4lsYpmfzZk7O7W2M2+I+RijhNvp/B6BN/w4AbVBVE8jAU0oRm9SAWoNgZnrvmh8MvM0GRMe1THrX\n4fTgtv+sx+E98chb0R39hpdh0NhAIznnr1vwz6vjcGinN3++W78qZI86joN5Sag8Fo0+QytQciAO\nxfuCZ91Msa3EvZNfwtZZV2LIEx8j/iBzxD0AXh5wCyr3GF9jE0Z7f9Gv6o8d4090tO/jZkYa4B/E\nnDm+j518cmDOeH09t+hG5eehUFJCTXftutHRDCa++qqv4TbT0tejqRredx9dEpWVdKcYGSu3mymj\nevdNQwPdG88+Czz8cNOXJ5q5+hpFRZSGKCjw6gUZ0bevsQ/+oos4AT/5JNM1m0NUVNvp4owfzwnc\nPx5htwfKOvfuzbiJfz1Fr16+9+bll31dXW43J9f58ynZoefIEfN03bIyTm5W/XcVlrSb0RdC3Ajg\nLr+Hr5dSrhNCZIJunl+0x2v3ya3EyBlF+Pod37xam92N8ZeZuAdaSM8B1eg5YJ/p8cGnluLRNcvw\nyVM5KCmIR2pmLc67YyfSetXB4wHcjTZEOTzYtKgn5t8+Bkd+MNoexCU3oKbcCS0mMKj/QVz5+3ys\nuGYeYLNh3zWTMeiZxYg+XoXSsQNReGgK9HVSemIQJIMlWD67HqsG2nqWLWu5wdf44gvmxcfG0piu\nX0+DYbPRz5ubS+O0eXPwax06xCIZLZ3wD38wfl58vHlufF4eJwv/lMTGRhovfezCSETNbufPb3/L\n8Y8bB5x5pvf4oEEMGjbX6A8ZEryZSagMH06//aefel00TifHadTARatY3bCBk3G/fnR/aVWsbrd5\ngV5eHndGetdZdrZ5H4PMTPO2j4qQaDejL6WcD2C+/+NCiBEA3gRwj5RyRXu9/q3/WQ9nrBvffd4T\n5cWxyBxcgYmzC3DRvbK9XtKUpG4NuOKRwCCdzQZEOWgYRp93GH/ZtBifPzcQlSUxGDCmFKddehBf\nvpGFg3lJyOhfjek37IUzxhtHqO+Rirx5l8FW70Lv977BjYkv4YvefVBQGLiKnAYL10dzsNvpsw6F\nttAUOX6cWS4XXQT885++wd3KSq4WQ5UHWLqU2Tl3380CqBkz6JrxX1WOHUu/sxHaLsDf6L/2mnWw\nWv98fSHYli2cYGbP9j525pnUE/L370dHU6do715f11tWlu/z24Krr+Z9eP99/n7BBea59zYbJ9Nz\nzzU+7nKZu9QaGrij0xv9uDj61T/80Pdch4Odx5TSZ6sIdyD3JAALAVwhpQxhadZy4pNduPO1ODM0\n9gAAG4RJREFUb1B5zInSQ7HoObCqxa6ccBGf4sLF9/sG/86bth4D85ciamcdji4aikOzTvVxG/R5\new1OmrcQqd9zRf190tt4LPZu/LH2lwAAB+oxEx/hQTzWwkHFe41iUhJzw6dODf48j8dab2TgQBpt\nq6waja1bvdo7/lRU0Kj4u6fM2LOHjVk0Tfybb/Zm+iQm0i8/ezZdOGZj272br6VfWfv3jg2Vxkbu\nZmbO9Los0tOZnlhZ6d0tpKcDt94KjBxJg//pp97U0vPOa3sf9IoVLGzTZAOKirjj8hd5C4XoaLqB\njIqfMjIYK/Hnmmto/Net4/vNyGC85Jxzmv/6Ch/C7dP/E4BYAP9gHBdlUspZ7fmCiekNSAwhDbO9\n8XiAkgNxiI5rRHJGaK6U7PnLMPyhNxB3hH8sOU8twqELxmDNwrvhiXYgrqAYE376ko//PamiEL+P\n/Q1ypqRha8pEnBH/DS5Y+kvYiltwDxwOrqp69mRxytVX848vGGvX0he/Zw99zf656KmpzMCoqeEK\n2UyiWGPXLuvXtdspMvbZZ17DYtTtSqO+ngHa3buBv/6Vq0rNB796Nf3yZllCAN0O27czjqGREpjJ\nFTLHjtFtpckWP/10oJDb8eM0vCNH8v75r+xdLq/rqLXs20cfvH6nceAA8NJLXJE3t44CoLHet893\nV2W3Mx5g5K6x2VjP8OMfe7O3FG1CWI1+exv4SGXNwj746Ikh2PttKqJjGyEmleDax79F7yHm6W7O\nYxUY9tu3fjD4AGBvdKPPB+uR+8d3kTfvMsz8+WLDgKu9tgZz7a8Ac5IAxALTH+XK8Phxrr43bDA3\ntPHx3tRJl4urzcpKGlBNF8eKvDxm/vi7JhwOrtazs+kqyMnh48OH09+7bJm57EBdnbURTk+ncTjr\nLODLL+l/TklhyqkVRUVMV73qKhqZd96h8mWwlFSHwzfzBmD6oX9Hr1Cx2bzKndu2Gefru92cqPxX\nuhs3coItKGDW1EknsTdvQkLLxgKwmM0odbS8nLui669v/jUnTODnsnQpd1DJyYwPmLmE9CiD36ac\neEmoEUbeqm548bYxP+jtN9Q6sPGj3igtjMUf1iyDI9q47/uA+ct/yMzxZ/jC7zF82GXWGSv6/Om0\nNN+G1keOmBv9wYON/dmatsppp5m/JsBzjAyGy0UN+FNOoQ9XSy+02+miCFaYdPw4UwD3+QXNk5K8\ngdDUVE4o2uv16WM9WQCcAK+6iu9vyZLQahBycgKbkkydygnESr/FjEGDvLuGffvMawn8U2i3baMC\npH5Hc/gw7++vftXygiKr3PvWSEecemro8SBFu6Gm0HZm6QsDDRus7NmYjmULzJtT22st3DFaKp2V\n6p9V8+rUwDaQPtc2842H4n+3agjy4YesVv31r5nxcdddNFyLFgUPgpaU0C88ahQni6goTlA33hjY\nLQrganzmzOCZHtq93LjROFvEn/79geuu8zWobjfw97+3zOBnZflWfA4caJ6DXlbG3YjmLvv8c+P7\ntnUr309LsfruZGa2/LqKiECt9NuZkoPm6Y2Hd5lvwQ/+eBzE3z5AdLlB1oOWrXLmmaxS9G9krXVo\nMuPcc+kG8TfQ6ekMlklpvOINxW9tNaHo0xBdLq7C//hHGvBQ2LSJuuYpKZyYgqkUnnkmJ4pnnjFP\ngdSEtJKTeS2jNEuN5GQ2Q/c3ymvWNM+1k55OY3/aadwhaMVQDQ3cLZit9KurKeFQUAD8/OfmaaVu\nN+Mg/jn1oXLBBcypLyjwfTwry1eCQdEpUSv9dia9t3n1Z48BJgUoACqG9kX0pGmB/sz+/b2dkZxO\nFhudcQaNW2YmMHEiUxKtlPwyMqhhIoS3qGfIEGayTJ9urOFitxvnaPszfXrz8qjr60PX86mtpTHS\nOmqF4r4YOhR45BHj1XNMDDNfAKZRWlV52mz0ZRudYya3YMSwYZyEHnqIMQh99esHH4Q2eXzzDXdI\n/nEFPa0RJ0tMZIbThAm8zz16MGvnl79UOfInAGql385Mv2EPNn3cK6CzVvaoUky/0divftnCpv9c\ndx19xxs20OBlZTFfXb/iTk1lKXtzGTmSfuRDh7i67d3ba0RvvRV48UUGZRsaOElMnhxa0G3ECBrH\nTz5himV0NA1Qa4u0NKwkAsz46CPj1fOQIVzp/+531lLJCQlMV5w40fi4f0N1PdruQQuy3nKLeWAy\n1MlDK0abMIFBcP/3lp3Nz6s19O4N/OIX3p3PCSQ41tVRRr+dGTatGDc8vREf/30I9mxKQ3RcI3In\nHcW1j38HZ4yvC+UHY69hs3EVr6XytTU2G/+4/cnMpN+9oIB+7tzc0KtwAbosJk9mjreW5//AA82r\n+jUiPd23ejUU6uuZ621EYSH94mYGPzGRE9i4cdbVrmedxewjf/+6zUYDnJjIz7C1hlhPbCxdccXF\nfO0jR7hry8lhjKCttGmUsT/hUEY/DEy68gAmzj6AI3sSEBPvQmpmoCRCgMGPBPr1C61FnMvFIGNi\nonfVa7d787nT05nSGKwTlJ6oKPqmtZVmSgpwxRXWLo3GRgaFv/+eYxo0iKtzs+BySQljG2ZUVtKI\nBpM36NGDGUALF3pfS1vha1lSu3cz9dSqH+qQIYGqpkakp3uvc+ml9LNv3cp7NHiwMtQKS5TRDxM2\nG9BzYGC6W0Qae4DGZ+lSGrG0NK4qNRlgDY+HVZurVzNQmpLC7Jrrrw90w/zsZzTKGzf6FmvFxgaq\nSjqdNKK9etGAx8TQyFnFKTweqjjqWxhu3kxjmJ5unHnUrZv1ijgqKvRK1+nTvUJly5cH9tytqmL2\n0pQp5pPIRRfRpabXqbHZOIFq96xbN94b/eQXG8sKYz1uN113CQnWwfVIwu3md6moyBvoVjn6bY4y\n+h1AxBp6jZUrgQULfKsnt2xhnrjer//hh77FTMXFzHWvr/ftYgXQ0N17L4O2X3zBa0tpHLiMiWEh\nmLZDCIX166lX48+OHQxMGxn9sWNpPM3cO/37N89gxscz82XRIuPjhw4xCGsmZeB00g22bBnH7XDQ\n8KWlMa4TGxtaoHzJErqt9u7lc4YOpcsnWGOYjqSwEPjXv3wz0YYOZVxBK1xTtAnK6IeRiDf2AFfM\nn30WKEJWV0djMmMGV8AeD10jRqmdmzZxh2CkvRMX582Yufde4zFUVrL61CzttLCQk01+PlfCOTle\nd5ARiYm81rp1THPs1o0G/9pr+b7WrQss+kpIYFpkS7BanQbztTscwNln80fPAPOaDh/WraOEgrZ7\nqq3l51FRwUbykbpyfumlwNTjvDw+fpe/WK+iNSijHyY6hcEH6Jv3d01o7N/P3PqsLGaQmKloVlTQ\nh20luAZYG0CzjJiKCsor6HPIDx9mDr0ZTicN/OzZ3K2kpHivHx9PmeP33uNquqGBdRA/+1nLpYqH\nDDFu9NK3b6AbJhh1ddw57N/PVfvkycYCZRorVhg3YcnPp+tkypTmvX44KC4279aVl9d2zWEUAJTR\nb3c6jbHXiInhatyo1V1srFcJ0umk68OoCjU+3rdrkj8VFXx+bi6LiPzp2dPcOH3ySWDREMDMGYfD\nODVz1Cj+Gx1t3LwkMZGqjrNnU6No+3bKOOfmcsXd3DTR2bM5Rr3URUoK9YGak1VTXs7OYfoV8OrV\nvI6+QYkeq+Y3RhNRJFBZafx9A2jw6+qU0W9DlNFvJzqdsdeIi+NKUh8Q1Rg61BtMtdmYyujvFgFY\nA2BkXDdsYAGSlr8/eDANq17rJzWVWTpmK32zJtwAM420NovaGE8+mXLQwXC5gL/9zVe+4OuvgTfe\nYHxh7tzQRcwyMlgQ9vnn3BklJtIt1lwJg7ffDnR51Nay7mDqVOMK6W7dAvvYamhNTSINLUvMaDLP\nyrLexSmajTL6bUinNfT+zJ1LN8j27d6USS3/W8+PfsSg7Zo1XhfLqFHATTcFXnPnTjYc13YGNTWc\nBAYMoCtlzx6u5s4+2zxwV1MTKDmsR8vS0Yy+x8OAqBaLsGLZMmO9mvp6ukxKSoDf/Cb0dMjYWPPV\neKiYGe/jxxlsv/DCwGPTpjH7xz8mIwQnr0gkKoqf+6uv+q74ExKYOKBSUNsUZfTbiBPG4AM0uvPm\nMSi4fz9XqBMmBAYBbTamD/7oRzS2aWnmvUuXLDF2Be3ZQ8N63XXBx/X00+YuiuRkGkN/FciaGrps\nJk3yFmmNH0+XzZo1nHgaG613EABlD9avb1+VyF272Klq9266gYzUSoMxejTF7D79lLswbec2d27k\nBnEBSkanpACrVnmbpkyfzl2jok1RRr+VnFDGXo/NxnTBYFLKAF0xwYq4rBQste5MVhw4YN5n1emk\nS+j1142PFxQwA0QLPL/9Nico/U4mGG43J6j2MvqFhaE3RE9Nte5eNmUKA76lpfxsWqOt31Lcbu5G\nDh6ky2n69OCB8fHjA2tBFG2OMvqt4IQ1+O2BVb57KN248vONs1IA7i4mT6YxN9N712caFRW1LKgZ\nLBupNSxaFJrBj4uj2yiYn9tmsy5ma09KSig1rdcSWrKEGlGh9jNWtBsdYvSFELkAvgbQU0pp8pcc\nmShD30LOOIOuFH9fc79+1tIEGjk5NHhGipwZGfSh5+YaB6Dbgv79Q+sN3FKsJqGhQylaFxfHyW3I\nkPYbR1vwyiuB4nH797N+YN68DhmSwkvYjb4QIhnA4wBMcrQiE2XsW8mIEfTbL1rE7B2Hg8HFOXNC\ny4fv04f+XX/9Hrvdq345Zw7dSPoAqL6xeyhoBrWggBOMXsQsKoo+97VrfZt1BwsSh4KVC2bkSGrs\n6DlwgFk8Bw5wMhgzJjKCnvX15jn3O3bwvoai56RoN8Jq9IUQNgDPA3gIwPvhfO3WoAx+GzFtGlfL\nWoCxuSmMt99OH/WWLcxhz8zkyldr7JGRQZnklSvpI+/Wjee9/Xbor3H66ZRSqKvzipgNGkRj+sYb\nLOLS4gAlJQy+NjS0vrnIxIkMFPsrkXbvHjip7N/Phu56d9DmzXzPN97YunG0FpfL243M6Fhr2i0q\n2oR2M/pCiBsB+NdP7wPwppRyszBq1BFBKEPfTtjtoUsK+BMTQ8NfU8OCnrS0wGKnqCjfvPyaGhrT\nvXt9z0tO5g5AX8x12mlebaGYGN/OU7W1lJ3wD/y6XEzpPPfc1mXHnHoqcPnlzLrRgt79+7O3sb+y\n6PvvG/v/V6/m5NORGjtaYd733wce69vX26lM0WG0m9GXUs4HMF//mBAiH8CNTRNCJoDPAURcXbgy\n+BFOXFzo+v5xccA991D2WPMzDxnCJu1HjzIG4HIBw4cbp6VqFBSY9wguKuIk1NoioosuYr762rU0\nnmPHGo/HTCajqoqCbrNmtW4creXCC+l20qfoxsVxYmwrnX9FiwnrJyCl/GGaF0LsBXC26ckdgDL2\nnYDqauB//6PfuLGR2SCXXOJtBlNdzRW5ptTpdLI6+PbbA6/VowfbF4ZCRgb97kbuicTE5jWZsSI2\nNnjTHKtOXWZ1EuFkzBhOtEuWcGJNTWXsI1TFVEW7oqZdKGMfUaxdSzdFeTn92Wef7e3Z63ZTKkEv\nhVxQQL/6r37Fwp5PP/WmZ77zDvP326ISNS2NuwEj+eaRIwP1eXbs4OTjcjH7xmoX0VxGjDBurdi7\nd9t252oNQhj3WlZ0ODZPqMUpHcRCLGy3ASpjH2F89BHw5pu+wcyUFPbsHTOGOvzPPGP83LFjWbzl\nHwhNSwP+9Ke2yVmvqODrb93KQG98PGUnbrvNNwPprbfYa0AvKTB2LBvWt4V7w+ViwxhNFRSgSN3c\nub5xCEXX5rLLDFO5uuRKXxn7CKS+Hli8ONBol5UBH39Mo2+lu5Ofb9yDt7SUwmezZ7d+jElJwP33\nMyi8Zw9Xsv49hvfv53j9VSPXr6dCaGv1eABOHHffzWDp99/T7XTmmUqJUhESXc7oK4MfoWzfbq5/\ns28fDbpVLrtVfnpz8vRDITubP0asXm1cQAZQv6ctjL7GsGGhxyQUiiYiWIGpbblsoTL4EU1Skrnr\nIyaGqZgzZhjLOTid1mmgVtr+bY1Z965gxxSKMHFCG33N0Ctj3wnIzmblqxG5uTT66ems6tUXdaWm\nMnvnJz8x1osfOjR4NgzAncTGjQyQtibOdeqp5k1XzN6fQhFGTlj3jjL0nQybjQb92Wd9G7OcdJKv\n7PLEiTSsq1fTbz5hgreZyN13s2J21y6vzMOVV3LCsOLjj9kXuKiIGTY5Oeyk1ZLsEyGoKLl4se/K\n/qST2ta1o1C0kBMue0cZ+05OQwOzdEpK6JYZN659deDXrWMmjH/gtU8ftiq0yok3w+NhY3ctuyYn\nh4VJLe25q1C0hBM5e0cZ+hMIp7NtBMxCRdsx+HPwILB0acs0dWw27kg0ITiFIoLo9D59ZfAVraK8\n3PyYVZPxSKeykqmlZj0IFF2WTrnSV4Ze0WZ0725+rDNKANfVAS++CGzaxAktI4PdqK65JrLbJSrC\nhvoWKLo255xj3Ig9N5d9dTsb//43VT+1HUxxMSud33qrY8eliBg6zUpfre4V7cKgQRRj+/BDZg05\nnTT4117b+VbGx45RV9+Ib76hdHNne0+KNifijb4y9op25+ST+eNy0Sh2VsN48CD1gYwoLaV/Pz4+\nvGNqS2prKcgXF0cto2CpuApDIt7oKxRho7NrvWdlsWZBr2OvofUR7qx8+CFbbWoNZrKyWIOhBOaa\nTSdd0igUigBSUoyNoM1GeenOuoPZuJExCc3gAxS2mz/fOvtKYUgn/RYoFApDbrqJwenu3VlY1rcv\nu4RdcklHj6zlrFplXEtRXMxKakWz6OT7WYVC4YPDweboc+Z4Wzh2dreVVTN11Wi92XTyb4NCoTAk\nOrptGsdEAnqBPX+yssI3jhOEsBp9IUQUgCcAjAUQA2CelPKjcI5BoVB0Ms4/H/j2Wwri6RECmDq1\nY8bUiQm3T38OAKeUciKAWQAGBzlfoVB0dTIzgV/8gpXFPXqwW9n06cC996q0zRYQbvfOOQC2CiE+\nBmADcEeYX1+hUHRGBg6kdLai1bSb0RdC3AjgLr+HjwKoBTATwBQA/2n6V6FQKBRhoN2MvpRyPoD5\n+seEEG8C+EhK6QGwQggxpL1eX6FQKBSBhNunvxrA+QAghBgJYH+YX1+hUCi6NOE2+i8AsAkh1gJ4\nHsAtYX59hUKh6NKENZArpawDcEM4X1OhUCgUXpQMg0KhUHQhlNFXKBSKLoQy+gqFQtGFsHk8no4e\ng0KhUCjChFrpKxQKRRdCGX2FQqHoQiijr1AoFF0IZfQVCoWiC6GMvkKhUHQhlNFXKBSKLoQy+gqF\nQtGF6NI9coUQNgAHAOxsemiNlPJBv3N+AuBmAC4Aj3Zke0chRAqAVwEkA4gGcLeUco3fOf8AMAlA\nRdNDs6SUZWEepx3AMwBGAqgDcJOUMl93/EIAD4P3dIGU8oVwjs8fIYQTwAIA2WAbz0ellB/ojt8F\n4CawHwQA3CyllOEepx4hxEYA5U2/7pFSXq87Fmn3dy6AuU2/xgIYBSBTSnm86XjE3F8hxDgAf5ZS\nniGEGAzgJQAeAFsB3C6ldOvOtfyeRypd2ugDGARgo5TyQqODQohMAHeCPX1jAawWQixuEo7rCO4G\nsFRK+aQQQgB4A8AYv3NOAXCOlLI47KPzcjGAWCnl6UKI8QAeB9tjagb27wBOBVAF4EshxAdSysMd\nNlrgGgAlUso5Qoh0AN8C+EB3/BQA10opN3TI6PwQQsQCsEkpzzA4FnH3V0r5Emg8IYR4GpyIjutO\niYj7K4S4D2zpWtX00BMAfi2l/EII8Rz4HX5X9xTT73kk09XdO6cA6COEWC6E+KTJkOo5DcCXUsq6\nptVyPoCTwz5KL38H8O+m/zvALmQ/0LTyyAHwvBDiSyFERymaTgLwKQBIKdeCk6bGUAD5UspSKWU9\n2GOho7unLQTwm6b/28AVsp5TADwohFgthHgQHc9IAPFCiM+FEMuaDI5GJN5fAIAQYiyAYVLK5/0O\nRcr93QXgUt3vpwBY0fT/RQDO8jvf6nsesXQZoy+EuFEIsVX/A6AIwJ+klNMA/BF0nehJBqB3jVQA\nSOnA8eZIKWuadiCvAvD/A0kA8E9w5XougNuEEB0xSfnft0YhhMPkWNjuqRlSykopZYUQIgnA2wB+\n7XfKm2Dvh+kAJgkhZoZ7jH5UA/gb2HP6FgCvRfL91fEQgEcMHo+I+yul/D8ADbqHbE1d/gDj+2j1\nPY9YIn6AbYVJ+8Z4NK3qpJSrhRC9hRD6D7ocQJLuKUkA9NvSsI4XAIQQI8A/knuklCv8DlcD+IeU\nsrrp3GXgqvC7dh6uP/73zS6ldJkcC9s9tUII0Q/cuj8jpXxd97gNwJNaXEQI8TGA0QA6LLYDYAe4\nmvcA2CGEKAHQC0ABIvf+pgIQUsrlfo9H4v3VcOv+b3Qfrb7nEUuXWemb8FsAvwB+aN9YoDP4APAN\ngMlCiNimIOpQMKDTIQghTgJdEVdJKRcZnDIE9OFGNfl2JwHYGM4xNvElvG0xxwPYojuWByBHCJEu\nhIgGXQ9rAi8RPoQQPQF8DuB+KeUCv8PJALYKIRKbDNR0AB3t278B9B9DCNEbHOOhpmMRd3+bmAJg\nqcHjkXh/NTYJIc5o+v95AFb5Hbf6nkcsXWalb8JjAF4VQlwArvjnAoAQ4m5wJfWBEOIp8MO2A/iV\nlLLW7GJh4E9gQPkfTeGHMinlLL/xvgJgLbhNfVlK+X0HjPNdADOEEF+BPvLrhRBXAUiUUj7fNN7P\nwHu6QEp5sAPGqOchAGkAfiOE0Hz7LwBIaBrvQwCWgxkaS6WUn3TQODXmA3hJCLEazCy5AcDlQohI\nvb8AIADs/uEX3+9DpN1fjV8CeKFp8swDXX8QQrwMugADvucdNdDmoKSVFQqFogvR1d07CoVC0aVQ\nRl+hUCi6EMroKxQKRRdCGX2FQqHoQiijr1AoFF2Irp6yqVAEIITIBgugtoEpkdEACgFcL6U8IIS4\nFsAdAJzgwulFKeVTftf4HQC3lHJeGIeuUARFGX2FwphCKeUo7RchxJ8A/FMIsQiUDLhASnmoqdL0\ncyFElZRyflMR3xMArgTwlw4ZuUJhgXLvKBShsRKseP41KGl9CACa1CKvg7dSexYo1f14RwxSoQiG\nWukrFEFokrS4AsDXYNXl1/rjUso83f9fbnrOvDAOUaEIGWX0FQpjegshvm36fwyow3QPaPRtHTYq\nhaKVKKOvUBjj49PXEELsBnXTV+oemwrgPCnlA2Ecn0LRIpRPX6FoHn8F8HhTTwMIITJA/33Et8lT\nKAC10lcomoWU8rkm1cXFQgg3uHD6t5TyxQ4emkIREkplU6FQKLoQyr2jUCgUXQhl9BUKhaILoYy+\nQqFQdCGU0VcoFIouhDL6CoVC0YVQRl+hUCi6EMroKxQKRRfi/wGkH2g4nCi6MgAAAABJRU5ErkJg\ngg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xc2f7c88>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# fitting the optimal model onto the training data\n",
    "from sklearn import metrics\n",
    "from sklearn import cross_validation as cv\n",
    "\n",
    "# fitting the optimal model onto the training data\n",
    "lgr=LogisticRegression(C=1)\n",
    "lgr.fit(X_train, y_train)\n",
    "\n",
    "# for display purposes, we fit the model on the first two components i.e. PC1, and PC2\n",
    "lgr.fit(X_train[:,0:2], y_train)\n",
    "\n",
    "# Plotting the decision boundary for all data (both train and test)\n",
    "# Create color maps\n",
    "from matplotlib.colors import ListedColormap\n",
    "cmap_light = ListedColormap(['#AAFFAA','#FFAAAA'])\n",
    "cmap_bold = ListedColormap(['#0000FF','#FF0000'])\n",
    "\n",
    "# creating a meshgrid\n",
    "x_min, x_max = X[:, 0].min() - 1, X[:, 0].max() + 1\n",
    "y_min, y_max = X[:, 1].min() - 1, X[:, 1].max() + 1\n",
    "h=0.05\n",
    "xx, yy = np.meshgrid(np.arange(x_min, x_max, h),np.arange(y_min, y_max, h))\n",
    "xy_mesh=np.c_[xx.ravel(), yy.ravel()]\n",
    "Z = lgr.predict(xy_mesh)\n",
    "Z = Z.reshape(xx.shape)\n",
    "\n",
    "%matplotlib inline\n",
    "#print(Z)\n",
    "plt.figure()\n",
    "plt.pcolormesh(xx, yy, Z, cmap=cmap_light)\n",
    "plt.scatter(X[:, 0], X[:, 1], c=y, cmap=cmap_bold)\n",
    "plt.xlim(xx.min(), xx.max())\n",
    "plt.ylim(yy.min(), yy.max())\n",
    "plt.xlabel('PC1');plt.ylabel('PC2')\n",
    "plt.title('Logistc Regression')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "SVC and Logistic regression scored similar performance, but we should be careful since the model has not yet tested on test data or on data independent of the one used for training the models. Below, we fit the three models onto the test data (X_test) and compare the predictions (y_predict) with observations (y_test). For this purpose, various evaluation metrics are implemented:\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Model Selection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "KNN:\n",
      " confusion matrix\n",
      " [[89  1]\n",
      " [ 5 48]] \n",
      "\n",
      "\n",
      "Confusion matrix plot of KNN classifier\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQQAAADtCAYAAABZNAv0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nKv8P/QBda4twvwwAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xd8e59e8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " Classification report \n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "          0       0.95      0.99      0.97        90\n",
      "          1       0.98      0.91      0.94        53\n",
      "\n",
      "avg / total       0.96      0.96      0.96       143\n",
      "\n",
      "#############################################################################\n",
      "SVC:\n",
      " confusion matrix\n",
      " [[89  1]\n",
      " [ 3 50]] \n",
      "\n",
      "\n",
      "Confusion matrix plot of SVC\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQQAAADtCAYAAABZNAv0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAYBdd953rcAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xcb05400>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " Classification report \n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "          0       0.97      0.99      0.98        90\n",
      "          1       0.98      0.94      0.96        53\n",
      "\n",
      "avg / total       0.97      0.97      0.97       143\n",
      "\n",
      "#############################################################################\n",
      "Logistic Regression:\n",
      "confusion matrix\n",
      " [[89  1]\n",
      " [ 3 50]] \n",
      "\n",
      "\n",
      "Confusion matrix plot of Logistic regression\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQQAAADtCAYAAABZNAv0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAYBdd953rcAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xcb3b550>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " Classification report \n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "          0       0.97      0.99      0.98        90\n",
      "          1       0.98      0.94      0.96        53\n",
      "\n",
      "avg / total       0.97      0.97      0.97       143\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "# KNN\n",
    "# fitting the knn model on the training data\n",
    "knn=KNeighborsClassifier(n_neighbors=5)\n",
    "knn.fit(X_train, y_train)\n",
    "y_pred_knn =knn.predict(X_test)\n",
    "# computing and plotting confusion matrix\n",
    "c_m = confusion_matrix(y_test, y_pred_knn)\n",
    "print('KNN:\\n confusion matrix\\n', c_m,'\\n\\n')\n",
    "ax=plt.matshow(c_m,cmap=plt.cm.Reds)\n",
    "print('Confusion matrix plot of KNN classifier')\n",
    "plt.colorbar(ax)\n",
    "plt.xlabel('Predicted')\n",
    "plt.ylabel('True')\n",
    "plt.show()\n",
    "# classification report\n",
    "print('\\n Classification report \\n',classification_report(y_test, y_pred_knn))\n",
    "print ('#############################################################################')\n",
    "\n",
    "# SVC\n",
    "# fitting the SVC model on the training data and predicting for test data\n",
    "svc=SVC(C=100,gamma=0.001,kernel='rbf',probability=True)\n",
    "svc.fit(X_train, y_train)\n",
    "y_pred_svc =svc.predict(X_test)\n",
    "# computing and plotting confusion matrix\n",
    "c_m = confusion_matrix(y_test, y_pred_svc)\n",
    "print('SVC:\\n confusion matrix\\n', c_m,'\\n\\n')\n",
    "ax=plt.matshow(c_m,cmap=plt.cm.Reds)\n",
    "print('Confusion matrix plot of SVC')\n",
    "plt.colorbar(ax)\n",
    "plt.xlabel('Predicted')\n",
    "plt.ylabel('True')\n",
    "plt.show()\n",
    "# classification report\n",
    "print('\\n Classification report \\n',classification_report(y_test, y_pred_svc))\n",
    "print ('#############################################################################')\n",
    "\n",
    "# Logistic Regression\n",
    "# fitting the lgr model on the training data\n",
    "lgr=LogisticRegression(C=1)\n",
    "lgr.fit(X_train, y_train)\n",
    "y_pred_lgr =lgr.predict(X_test)\n",
    "# computing and plotting confusion matrix\n",
    "c_m = confusion_matrix(y_test, y_pred_lgr)\n",
    "print('Logistic Regression:\\nconfusion matrix\\n', c_m,'\\n\\n')\n",
    "ax=plt.matshow(c_m,cmap=plt.cm.Reds)\n",
    "print('Confusion matrix plot of Logistic regression')\n",
    "plt.colorbar(ax)\n",
    "plt.xlabel('Predicted')\n",
    "plt.ylabel('True')\n",
    "plt.show()\n",
    "# classification report\n",
    "print('\\n Classification report \\n',classification_report(y_test, y_pred_lgr))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* Scores on Malignant cases from each model are summarized as below:\n",
    "\n",
    "\n",
    "\n",
    "| Model         | Precision  | Recall | f1-score |\n",
    "| ------------- |:----------:| -----: | --------:|\n",
    "| KNN           | 0.98       |   0.91 | 0.94     |\n",
    "| SVC           | 0.98       |   0.94 | 0.96     |\n",
    "| Logistic      | 0.98       |   0.94 | 0.96     |\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "** Conclusion:**\n",
    "\n",
    "* As results show, both SVC and logistic regression are performing equally good. One may choose either of the models on the basis of other factors such as simplicity in parameter tuning and ease of interpritation and so on.\n",
    "* Threshold of 0.5 is used by default (for binary problems) to convert predicted probabilities into class predictions\n",
    "* Threshold can be adjusted to increase sensitivity or specificity. Specificity explains how often is the prediction correct when the actual value is negative (0). Sensitivity and specificity have an inverse relationship.\n",
    "* For this sepcific case, sensitivity is more important so we focus finding the threshold level which can give us the highest sensitivity. \n",
    "* We can manually change the threshold level and evaluate results. But instead we can plot ROC curve and see the effect of threshold on sensitivity and specificity. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Effect of classification threshold on model performance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "model: KNN\n",
      "thresholds: [ 2.   1.   0.8  0.6  0.4  0.2  0. ]\n",
      "tpr:        [ 0.     0.736  0.868  0.906  0.943  0.962  1.   ]\n",
      "fpr:        [ 0.     0.011  0.011  0.011  0.044  0.222  1.   ]\n",
      "model: SVC\n",
      "thresholds: [ 1.     0.771  0.615  0.5    0.49   0.442  0.411  0.364  0.32   0.307  0.   ]\n",
      "tpr:        [ 0.057  0.925  0.925  0.943  0.943  0.962  0.962  0.981  0.981  1.     1.   ]\n",
      "fpr:        [ 0.     0.     0.011  0.011  0.022  0.022  0.033  0.033  0.044  0.044  1.   ]\n",
      "model: Logistic\n",
      "thresholds: [ 1.     0.737  0.616  0.565  0.461  0.429  0.427  0.375  0.024  0.023  0.   ]\n",
      "tpr:        [ 0.019  0.925  0.925  0.943  0.943  0.962  0.962  0.981  0.981  1.     1.   ]\n",
      "fpr:        [ 0.     0.     0.011  0.011  0.033  0.033  0.044  0.044  0.267  0.267  1.   ]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0xd575c18>"
      ]
     },
     "execution_count": 104,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmYAAAHtCAYAAACzs7IMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nDABgQswAACbEDABgIi1LIycVj1LMAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xd5c1208>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# false positive rate,fpr= FP/(TN+FP) OR fpr=1-specificty, tpr=sensitivity \n",
    "y_pred_knn_p =knn.predict_proba(X_test)[:,1]\n",
    "y_pred_svc_p =svc.predict_proba(X_test)[:,1]\n",
    "y_pred_lgr_p =lgr.predict_proba(X_test)[:,1]\n",
    "\n",
    "models=[y_pred_knn_p,y_pred_svc_p,y_pred_lgr_p]\n",
    "label=['KNN','SVC','Logistic']\n",
    "\n",
    "# plotting ROC curves\n",
    "plt.figure(figsize=(10, 8))\n",
    "m=np.arange(3)\n",
    "for m in m:\n",
    "    fpr, tpr,thresholds= metrics.roc_curve(y_test,models[m])\n",
    "    print('model:',label[m])\n",
    "    print('thresholds:',np.round(thresholds,3))\n",
    "    print('tpr:       ',np.round(tpr,3))\n",
    "    print('fpr:       ',np.round(fpr,3))\n",
    "    plt.plot(fpr,tpr,label=label[m])\n",
    "plt.xlim([0.0,1.0])\n",
    "plt.ylim([0.0,1.0])\n",
    "plt.title('ROC curve for Cancer classifer')\n",
    "plt.xlabel('False positive rate (1-specificity)')\n",
    "plt.ylabel('True positive rate (sensitivity)')\n",
    "plt.legend(loc=4,)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* From the ROC curve, KNN has resulted the least peformance among the three models. SVC and Logistic seem to have the same performance for thresholds ~0.25 and above but for threshold below 0.25 svc performs better than logistic. \n",
    "* From these, we can lean to scv as its overall performace is better under various threshold levels. Therefore, \n",
    "  * **Optimal model: svc with C=100, gamma=0.001, kernel=rbf, and threshold value of 0.44**\n",
    "      * by changing the treshold from its default value of 0.5 to 0.44, the sensitivity increases from 0.94 to 0.962. \n",
    "  "
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}