FINAL_PREDICT_TREATMENT.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# BREAD CANCER DATA ANALYSIS AND PREDICTION"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "from sklearn.decomposition import PCA, KernelPCA\n",
    "\n",
    "import seaborn as sns\n",
    "from sklearn import model_selection\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.kernel_ridge import KernelRidge\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "from sklearn.svm import LinearSVC\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.gaussian_process import GaussianProcessClassifier\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from sklearn.metrics import classification_report\n",
    "from sklearn.linear_model import RidgeCV\n",
    "from sklearn import metrics\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# I-EXPLORE THE DATASET"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I-1 IMPORT THE DATA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(184, 4654) (184, 1) (92, 4654)\n"
     ]
    }
   ],
   "source": [
    "#TRAINING SET\n",
    "Xtrain = pd.read_csv(\"./dataset/xtrain.txt\", sep=\"\\t\", header=None).T\n",
    "header = Xtrain.iloc[0]\n",
    "Xtrain = Xtrain[1:]\n",
    "Xtrain = Xtrain.rename(columns = header)\n",
    "\n",
    "#TRAINING LABEL SET\n",
    "Ytrain = pd.read_csv(\"./dataset/ytrain.txt\", sep=\"\\t\", header=None)\n",
    "#TEST SET\n",
    "Xtest = pd.read_csv(\"./dataset/xtest.txt\", sep=\"\\t\", header=None).T\n",
    "header1 = Xtest.iloc[0]\n",
    "Xtest = Xtest[1:]\n",
    "Xtest = Xtest.rename(columns = header)\n",
    "\n",
    "#transform the dataframe into array\n",
    "X_t= np.array(Xtrain)\n",
    "Y_train= np.array(Ytrain)\n",
    "X_te= np.array(Xtest)\n",
    "\n",
    "print(X_t.shape, Y_train.shape, X_te.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I-VISUALIZATION OF THE DATA AND STANDARDIZATION OF THE DATA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "First Class precentage: %64.130435\n",
      "Second Class precentage: %35.869565\n"
     ]
    }
   ],
   "source": [
    "#CHECK IF THE DATA ARE BALANCED\n",
    "N,D = Y_train.shape\n",
    "\n",
    "label, counts = np.unique(Y_train, return_counts=True)\n",
    "dictionary = dict(zip(label, counts))\n",
    "\n",
    "first_pct = dictionary.get(-1)/N\n",
    "second_pct = dictionary.get(1)/N\n",
    "\n",
    "print('First Class precentage: %{:2f}'.format(100*first_pct))\n",
    "print('Second Class precentage: %{:2f}'.format(100*second_pct))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "sc= MinMaxScaler()\n",
    "sc.fit(X)\n",
    "X=sc.transform(X)\n",
    " \n",
    "sc1= MinMaxScaler()\n",
    "sc1.fit(Y)\n",
    "Y=sc1.transform(Y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Standardize the dataset, X_train and X_test\n",
    "sc= MinMaxScaler()\n",
    "sc.fit(X_t)\n",
    "X_train=sc.transform(X_t)\n",
    " \n",
    "sc1= MinMaxScaler()\n",
    "sc1.fit(X_te)\n",
    "X_test=sc1.transform(X_te)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Comments\n",
    "\n",
    "#### We noticed the dataset is unbalanced, hence the Accuracy metrics is not suitable to score our prediction cause it will be unfair with respect to the second class. So we will use f1 score to measure the performance of our model and the ROC CURVE to present our result."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1MAAAIuCAYAAABac1I3AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\newAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def colors(y):\n",
    "    colors=[]\n",
    "    for i in range(y.shape[0]):\n",
    "        if (y[i]== -1):\n",
    "            colors.append('red')\n",
    "        if (y[i]==1):\n",
    "            colors.append('blue')\n",
    "    return colors\n",
    "\n",
    "np.random.seed(0)\n",
    "kpca = KernelPCA(n_components =3)\n",
    "X_kpca = kpca.fit_transform(X_train)\n",
    "fig=plt.figure(figsize=[15,10])\n",
    "ax = fig.add_subplot(111,projection='3d')\n",
    "ax.scatter(X_kpca[:,0],X_kpca[:,1],X_kpca[:,2],c=colors(Y_train),marker='o')\n",
    "plt.title(\"Visualization of The dataset\")\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I-3 SPLIT THE TRAINING SET INTO TRAIN AND TEST AND THE FONCTION OF PREDICTION ERROR"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "xtrain,xtest,ytrain,ytest= train_test_split(X_train,Y_train,test_size=0.3)\n",
    "#Prediction error\n",
    "def error(ypred, ytrue):\n",
    "    e = (ypred != ytrue).mean()\n",
    "    return (e)\n",
    "def ConfusionmatrixPlot(Ytrue,Ypred):\n",
    "    \n",
    "    cm = metrics.confusion_matrix(Ytrue,Ypred)\n",
    "    from IPython.display import Image, display\n",
    "\n",
    "    fig, ax = plt.subplots(figsize=(5, 5))\n",
    "    ax.matshow(cm, cmap=plt.cm.Reds, alpha=0.3)\n",
    "    for i in range(cm.shape[0]):\n",
    "         for j in range(cm.shape[1]):\n",
    "             ax.text(x=j, y=i,\n",
    "                    s=cm[i, j], \n",
    "                    va='center', ha='center')\n",
    "    plt.xlabel('Predicted Values', )\n",
    "    plt.ylabel('Actual Values')\n",
    "    plt.show()\n",
    "def ROCCURVE(predict,ytrue,ypred):\n",
    "    fpr,tpr, thresolds=metrics.roc_curve(ytest,predict)\n",
    "    roc_auc=metrics.auc(fpr,tpr)\n",
    "    print(\"the Area under the ROC curve:%f\" %roc_auc)\n",
    "\n",
    "    #plotting the ROC curve\n",
    "    plt.plot(fpr,tpr,label=\"AUC = %0.2f\" %roc_auc)\n",
    "    plt.xlabel('false Positive rate')\n",
    "    plt.ylabel('True Positive rate')\n",
    "    plt.legend(loc=\"lower right\")\n",
    "    plt.title('Receiver operating characteristics')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# II-PREDICTION WITH BRUT DATA"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "II-1 LOGISTIC REGRESSION"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Acuracy of the model is: 0.5892857142857143\n",
      "The error of prediction is: 0.41709183673469385\n",
      "[[28 10]\n",
      " [13  5]]\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "         -1       0.68      0.74      0.71        38\n",
      "          1       0.33      0.28      0.30        18\n",
      "\n",
      "avg / total       0.57      0.59      0.58        56\n",
      "\n"
     ]
    }
   ],
   "source": [
    "model = LogisticRegression()\n",
    "model.fit(xtrain, ytrain)\n",
    "Ypred_Logistic = model.predict(xtest)\n",
    "accuracy = accuracy_score(ytest.flatten(),Ypred_Logistic)\n",
    "parameters = model.coef_\n",
    "\n",
    "print(\"The Acuracy of the model is:\",accuracy)\n",
    "print(\"The error of prediction is:\", error(ytest,Ypred_Logistic))\n",
    "print(confusion_matrix(ytest,Ypred_Logistic))\n",
    "print(classification_report(ytest,Ypred_Logistic))\n",
    "PredictProbLR=model.predict_proba(xtest)[:,1]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "the Area under the ROC curve:0.548246\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe8AAAFnCAYAAACPasF4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAATgAAAE9CAYAAACfsn7ZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ROCCURVE(PredictProbLR,ytest,Ypred_Logistic)\n",
    "ConfusionmatrixPlot(ytest, Ypred_Logistic)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "II-2 GAUSSIAN PROCESS CLASSIFIER"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Acuracy of the model is: 0.6964285714285714\n",
      "The error of prediction is: 0.40433673469387754\n",
      "[[32  6]\n",
      " [11  7]]\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "         -1       0.74      0.84      0.79        38\n",
      "          1       0.54      0.39      0.45        18\n",
      "\n",
      "avg / total       0.68      0.70      0.68        56\n",
      "\n"
     ]
    }
   ],
   "source": [
    "model = GaussianProcessClassifier()\n",
    "model.fit(xtrain, ytrain)\n",
    "Ypred_GPC = model.predict(xtest)\n",
    "accuracy = accuracy_score(ytest.flatten(),Ypred_GPC)\n",
    "\n",
    "print(\"The Acuracy of the model is:\",accuracy)\n",
    "print(\"The error of prediction is:\", error(ytest,Ypred_GPC))\n",
    "print(confusion_matrix(ytest,Ypred_GPC))\n",
    "print(classification_report(ytest,Ypred_GPC))\n",
    "PredictProbLGPC=model.predict_proba(xtest)[:,1]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "the Area under the ROC curve:0.500000\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe8AAAFnCAYAAACPasF4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/gc0zYsMZEFSiwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAATgAAAE9CAYAAACfsn7ZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nATAWgQNgrP8Fotp1v0lfshUAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#compute the Area under the ROC Curve\n",
    "ROCCURVE(PredictProbLGPC,ytest,Ypred_GPC)\n",
    "ConfusionmatrixPlot(ytest, Ypred_GPC)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "II-3 DECISION TREE CLASSIFIER"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Acuracy of the model is: 0.5178571428571429\n",
      "The error of prediction is: 0.44260204081632654\n",
      "[[24 14]\n",
      " [13  5]]\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "         -1       0.65      0.63      0.64        38\n",
      "          1       0.26      0.28      0.27        18\n",
      "\n",
      "avg / total       0.52      0.52      0.52        56\n",
      "\n"
     ]
    }
   ],
   "source": [
    "model = DecisionTreeClassifier()\n",
    "model.fit(xtrain, ytrain)\n",
    "Ypred_DTC = model.predict(xtest)\n",
    "accuracy = accuracy_score(ytest.flatten(),Ypred_DTC)\n",
    "\n",
    "print(\"The Acuracy of the model is:\",accuracy)\n",
    "print(\"The error of prediction is:\", error(ytest,Ypred_DTC))\n",
    "print(confusion_matrix(ytest,Ypred_DTC))\n",
    "print(classification_report(ytest,Ypred_DTC))\n",
    "PredictProbLDTC=model.predict_proba(xtest)[:,1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "the Area under the ROC curve:0.454678\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe8AAAFnCAYAAACPasF4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nbyIiIpVheBMREanM/wAKuYbUYO4X9gAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAATgAAAE9CAYAAACfsn7ZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngQNgLAIHwFgEDoCxCBwAYxE4AMb6f0p1apPbDvhWAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ROCCURVE(PredictProbLDTC,ytest,Ypred_DTC)\n",
    "ConfusionmatrixPlot(ytest, Ypred_DTC)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "II-4 KNN CLASSIFIER"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Acuracy of the model is: 0.6964285714285714\n",
      "The error of prediction is: 0.3405612244897959\n",
      "[[37  1]\n",
      " [16  2]]\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "         -1       0.70      0.97      0.81        38\n",
      "          1       0.67      0.11      0.19        18\n",
      "\n",
      "avg / total       0.69      0.70      0.61        56\n",
      "\n"
     ]
    }
   ],
   "source": [
    "model = KNeighborsClassifier()\n",
    "model.fit(xtrain, ytrain)\n",
    "Ypred_KNN= model.predict(xtest)\n",
    "accuracy = accuracy_score(ytest.flatten(),Ypred_KNN)\n",
    "\n",
    "print(\"The Acuracy of the model is:\",accuracy)\n",
    "print(\"The error of prediction is:\", error(ytest,Ypred_KNN))\n",
    "print(confusion_matrix(ytest,Ypred_KNN))\n",
    "print(classification_report(ytest,Ypred_KNN))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "the Area under the ROC curve:0.559211\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe8AAAFnCAYAAACPasF4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nmzHGGEsxHN6MMcZYiuHwZowxxlLMX2vJAX2GOdD9AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAATgAAAE9CAYAAACfsn7ZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#compute the Area under the ROC Curve\n",
    "PredictProbLKNN=model.predict_proba(xtest)[:,1]\n",
    "ROCCURVE(PredictProbLKNN,ytest,Ypred_KNN)\n",
    "ConfusionmatrixPlot(ytest, Ypred_KNN)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "II-5 LDA ANALYSIS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Acuracy of the model is: 0.5892857142857143\n",
      "The error of prediction is: 0.40433673469387754\n",
      "[[29  9]\n",
      " [14  4]]\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "         -1       0.67      0.76      0.72        38\n",
      "          1       0.31      0.22      0.26        18\n",
      "\n",
      "avg / total       0.56      0.59      0.57        56\n",
      "\n"
     ]
    }
   ],
   "source": [
    "model = LinearDiscriminantAnalysis()\n",
    "model.fit(xtrain, ytrain)\n",
    "Ypred_LDA= model.predict(xtest)\n",
    "accuracy = accuracy_score(ytest.flatten(),Ypred_LDA)\n",
    "\n",
    "print(\"The Acuracy of the model is:\",accuracy)\n",
    "print(\"The error of prediction is:\", error(ytest,Ypred_LDA))\n",
    "print(confusion_matrix(ytest,Ypred_LDA))\n",
    "print(classification_report(ytest,Ypred_LDA))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "the Area under the ROC curve:0.543860\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe8AAAFnCAYAAACPasF4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzt3X1YVHX+//HXwACmEMEGmVBptt6E\n6WrmZlomQiBJbmYBJVq6ld1Y3mQqmlgKaqaVpmV9227UNV1lq+0+My1N08o0KcubFdFSblUQRYTP\n7w9/zkYCg8IwHHw+rmuvizPnzOe85y3bi8+ZmfOxGWOMAACAZXi4uwAAAHBmCG8AACyG8AYAwGII\nbwAALIbwBgDAYghvAAAshvCGpbVu3VqRkZGKjo5WdHS0IiMjlZSUpKKiolo/16effqpx48bV+rju\ntnnzZm3btk2StHDhQj333HMuP2fr1q21f/9+l5/nj3bt2qWNGzee8fNmzpypxYsXV3nMl19+qV9/\n/bXaxwM1YeN73rCy1q1ba/Xq1WratKkk6fjx4xoxYoSuuOIKjRgxws3VWcPEiRN19dVXq2/fvnV2\nzj/+u9WVl19+WSdOnNCDDz5Y62MPGTJEDzzwgDp37lzrYwN/xMwbDYq3t7euv/56/fTTT5JOhvmU\nKVMUFRWl8PBwvfTSS45jt27dqn79+ikqKkoDBgxQZmamJGnHjh0aMGCAoqKiFBsbqx9++EGSlJaW\nprvvvlurV69WbGxsufP27dtXX3zxhQ4fPqzRo0crKipKvXr10vLlyx3HtG7dWvPnz1dUVJRKS0vL\nPb+4uFgTJ05UVFSUevfurWnTpjmOad26td5880317dtXXbt2LTejW7JkiaKjoxUeHq6RI0fq2LFj\nkqSxY8dq6tSpio2N1YcffqijR49q+PDhjj5Mnz5dkrR48WK98847mjFjhl577TXNmTNH48ePlyQl\nJibqtddeU0JCgq6//nqNHDlSp/7WT0tLU7du3XTLLbcoLS1NrVu3rvDf44svvtDNN9+sqKgo3X//\n/Tp48KBj3+rVq9WvXz91795d//jHPxyPz507V1FRUYqIiND999+vw4cPS5LmzJmjCRMmqH///nr9\n9ddVVlamJ5980vGaRo8erZKSEklSXl6ehg4dql69eik2NlZr1qzRypUrNX/+fL355puaNm3aGfVv\n7NixmjdvnqSTVyd69+6t6Oho9e/fX9u3b9dzzz2n9evXa/To0frggw/KHV/Z71lF4wDVZgALa9Wq\nlfntt98c2wcPHjR33XWXmTdvnjHGmBdeeMEMGjTIFBcXmyNHjpi//e1vZuXKlcYYYyIjI82qVauM\nMca89tpr5t577zWlpaXmpptuMkuXLjXGGPPNN9+Y7t27m5KSErN8+XLHWJ07dzZ79uwxxhizZ88e\n06VLF1NSUmLGjRtnHn/8cVNaWmpyc3NNjx49zM8//+yo9cUXX6zwdcyfP9/ce++9pqSkxBw9etTc\ndttt5u2333Y876mnnjLGGLNz507Trl07k5eXZzZu3Gi6du1q9u/fb4wx5oknnjDTpk0zxhgzZswY\nExsba44dO2aMMebVV181f//7301ZWZk5ePCg6dKli9m4caMxxpgBAwY4zjV79myTlJTkeHzAgAHm\n6NGj5siRI6Zr167mm2++Mfn5+aZ9+/bm559/NqWlpWbEiBGmVatWp72mI0eOmC5dujhe/5QpU8yk\nSZMcr2nmzJnGGGO2bNlirrrqKnP8+HHzww8/mK5du5qCggJTWlpq7r77bjN37lxHbd27dze5ubnG\nGGM++ugj06dPH3P8+HFz7Ngx07t3b8frSEpKMk8//bQxxpj09HTTpUsXU1xcbMaMGeMY70z6d+p5\nBQUFpnPnzqagoMAYY8wHH3xgXn75ZWOMMT179nT09Pfnqej3rKpxgOpg5g3LS0xMVHR0tHr16qVe\nvXrp2muv1b333itJ+vzzz3XnnXfK29tbjRs3Vt++ffXJJ5/ov//9r/Lz89WjRw9J0oABAzRnzhzt\n2rVLubm56t+/vyTp6quvVmBgoDZt2uQ4n7e3t3r27KmVK1dKklasWKGIiAjZ7XZ9/vnnGjhwoDw8\nPBQYGKjIyEh98sknjufeeOONFb6GVatW6Y477pDdblejRo0UGxurtWvXOvbfdtttkqTLL79cLVq0\n0JYtW7Ry5UrFxMTooosukiQlJCSUO1fXrl3l4+MjSRo8eLDmzZsnm80mf39//fnPf9bevXud9jY6\nOlqNGjVS48aN1bx5c/3222/avHmzmjdvrlatWsnDw0MJCQkVPve7775T06ZN1apVK0nS6NGjy31m\n4JZbbpEkXXnllSouLlZ+fr7atWunVatWydfXVx4eHurYsaNjpipJHTp0UGBgoCQpKipKy5cvl5eX\nl3x8fHTVVVc5jl29erX69OnjGP+zzz6Tt7d3ufrOpH+n+Pj4yGazadmyZcrJyVHv3r0dv2sVqez3\n7EzHAf7I7u4CgJpasGCBmjZtqry8PEVHRysmJkZ2+8lf7YKCAk2dOlWzZs2SdPIyevv27ZWfny8/\nPz/HGHa7XXa7XYcPH9axY8fUu3dvx77CwsJyl3ulk8Hx5ptvatCgQVqxYoXjPdSCggINHz5cnp6e\nkk5eDo+OjnY874ILLqjwNeTl5cnf39+x7e/vr9zc3HLbv//58OHDKigo0Keffqo1a9ZIkowxjsvG\nf3zO7t27NW3aNO3atUseHh7av3+/+vXrV2VfJcnX19fxs6enp0pLS3X48OFyY58Kvz/Kz8/X+eef\n79j+Y3ieGvtUr8rKynT06FFNnTpVX3/9tSTp0KFD5f7g+f158/LyNHnyZP3444+y2WzKycnRoEGD\nJEkHDx4s9+/7+9dxypn07xQvLy+9/vrreumllzRnzhy1bt1aycnJlb5tUNnvmaQzGgf4I8IbDUZg\nYKASExM1Y8YMvfjii5Kk4OBgDR48WD179ix37H//+18dPHhQZWVl8vDwUElJiQ4cOKDg4GA1adJE\nH3300Wnjp6WlOX6+/vrrlZSUpN27d2v37t269tprHeebO3euY7ZZXRdeeGG5PxAOHjyoCy+80LGd\nn5+vkJAQxz5/f38FBwfr1ltv1ZgxY5yO/9RTTyksLExz586Vp6en4uPjz6i+3/P19S33af6srKwK\njwsICFB+fr5j++jRozp06FCVH1J74403tHv3bqWlpalJkyZ69tlndeDAgQqPffbZZ2W32/Wf//xH\n3t7eGjVqlGPfBRdcoPz8fIWGhkqS9u7de9ofGWfSv9+78sorNXv2bB0/flz/93//p+TkZL311lsV\nHhsQEFDh71loaOgZjQP8EZfN0aDcc8892rRpkzZs2CBJ6tWrl/71r3+ptLRUxhjNmzdPX3zxhZo3\nb66mTZs6LpMuW7ZMEydOVEhIiJo2beoI77y8PI0cOfK0r555e3ure/fumjFjhnr16uWYPYaHhzv+\nA3zixAmlpqYqPT3dad033nijli1bptLSUhUVFemdd95xXGqVpPfff1+StHPnTmVkZKhDhw4KDw/X\nJ598ory8PEknL9+//PLLFY6fm5urtm3bytPTU2vXrlVGRobjNdntdhUUFFSvwZLCwsL0888/KyMj\nQ2VlZVq2bFmFx1199dXKzs7Wli1bJEnz5s3T3Llzqxw7NzdXl19+uZo0aaJ9+/Zp9erVlX7tLzc3\nV61atZK3t7e2bdumTZs2OY4NDw/Xv//9b0knP4DYr18/lZaWlnutZ9K/U37++Wc98sgjOn78uLy9\nvdWuXTvZbDZJFfexst+zqsYBqoOZNxoUX19f3XfffZo+fbqWLVumO++8U3v37tXNN98sY4zatWun\nQYMGyWaz6fnnn9fo0aM1a9YsBQUFaerUqbLZbJo1a5YmTZqk5557Th4eHrrnnnvUuHHj084VFRWl\nYcOG6fXXX3c8Nnz4cMcnoKWTM/TqXApNTExUZmambr75ZtlsNkVHR5e7dB8YGKi+ffvqwIEDmjBh\ngvz9/eXv76+hQ4cqMTFRZWVl+tOf/qQnn3yywvEfeOABTZ06VfPmzVOvXr308MMPa/bs2Wrbtq0i\nIiI0Y8YMZWZmVnh5+Y+Cg4M1cuRIDRw4UBdeeKHi4+MdQfl75513nubMmaPRo0dLki677DLHp7wr\nEx8fr0ceeURRUVFq3bq1xo4de1qPTxk8eLDGjBmjtLQ0de7cWWPGjNH48ePVvn17jR49WmPGjFF4\neLiaNGmiZ555Ro0aNVLPnj312GOPad++fZo9e3a1+3dKq1atFBoaqj59+sjLy0tNmjTRxIkTJZ38\nfRg5cqQeeeQRx/GV/Z5dcskllY4DVAff8wbqOXd9J7oqxhjHTHH79u268847z+rmJwDODpfNAZyR\nEydO6Prrr9fmzZslSR988IH+8pe/uLkq4NzCZXMAZ8Rutys5OVljxoyRMUZBQUFKSUlxd1nAOYXL\n5gAAWAyXzQEAsBjCGwAAi7HMe97Z2dX/Hmp1BAQ0Vn5+7S8bea6hjzVHD2uOHtYcPaw5V/QwKMiv\nwsfP2Zm33e7p7hIaBPpYc/Sw5uhhzdHDmqvLHp6z4Q0AgFUR3gAAWAzhDQCAxRDeAABYDOENAIDF\nEN4AAFgM4Q0AgMUQ3gAAWIxLw/uXX35RRESEFi5ceNq+r776Sv3791dcXJzmzp3ryjIAAGhQXBbe\nRUVFmjx5srp27Vrh/ilTpmjOnDlavHix1q5dqx07driqFAAAGhSX3dvc29tbr7zyil555ZXT9mVm\nZsrf318XX3yxJKlHjx5at26drrjiCleVAwD4/5au3KGN27LKPebpaVNpKStE18QNnUIVe+2ldXIu\nl4W33W6X3V7x8NnZ2QoMDHRsBwYGKjMzs8rxAgIa1/p9Yyu74TvODH2sOXpYc/Sw+r7bnq38wmJd\n6N+o3OOenjY3VdRw1NXvoWVWFXPFSi21vVLZuYg+1hw9rDl6eGZKS40CfH007f7/va1JD2vOFT2s\nV6uKBQcHKycnx7F94MABBQcHu6MUAAAsxy3hHRoaqsLCQu3du1cnTpzQ559/rm7durmjFAAALMdl\nl823bt2q6dOna9++fbLb7fr4448VHh6u0NBQRUZGatKkSRo1apQkKSYmRi1atHBVKQAANCguC+92\n7dppwYIFle6/5pprtGTJEledHgCABos7rAEAYDGENwAAFkN4AwBgMYQ3AAAWQ3gDAGAxhDcAABZD\neAMAYDGENwAAFkN4AwBgMYQ3AAAWQ3gDAGAxhDcAABZDeAMAYDGENwAAFkN4AwBgMYQ3AAAWQ3gD\nAGAxhDcAABZDeAMAYDGENwAAFkN4AwBgMYQ3AAAWQ3gDAGAxhDcAABZjd3cBAIDKLV25Qxu3ZdXq\nmPkFxQrw86nVMVG3mHkDQD22cVuW8guKa3XMAD8fXdMmuFbHRN1i5g0A9VyAn49mPHidu8tAPcLM\nGwAAiyG8AQCwGMIbAACLIbwBALAYwhsAAIshvAEAsBjCGwAAiyG8AQCwGMIbAACLIbwBALAYwhsA\nAIshvAEAsBgWJgFQ52p7mUtPT5tKS02tjVefsHwnKsLMG0Cdc8Uylw0Vy3eiIsy8AbhFbS5zGRTk\np+zsgloZC7ACZt4AAFgM4Q0AgMUQ3gAAWIxL3/NOTU3V5s2bZbPZlJSUpPbt2zv2LVq0SO+++648\nPDzUrl07jR8/3pWlAADQYLhs5r1hwwZlZGRoyZIlSklJUUpKimNfYWGhXn31VS1atEiLFy/Wzp07\n9f3337uqFAAAGhSXhfe6desUEREhSWrZsqUOHTqkwsJCSZKXl5e8vLxUVFSkEydO6OjRo/L393dV\nKQAANCguC++cnBwFBAQ4tgMDA5WdnS1J8vHx0UMPPaSIiAj17NlTHTp0UIsWLVxVCgAADUqdfc/b\nmP/d/aiwsFDz58/XRx99JF9fXw0aNEjbtm1TmzZtKn1+QEBj2e2etVpTUJBfrY53rqKPNXeu9dDT\n0yapdl/3udZDV6CHNVdXPXRZeAcHBysnJ8exnZWVpaCgIEnSzp07dckllygwMFCS1LlzZ23durXK\n8M7PL6rV+ripQ+2gjzV3Lvbw1K1Ma+t1n4s9rG30sOZc0cPK/hhw2WXzbt266eOPP5YkpaenKzg4\nWL6+vpKkkJAQ7dy5U8eOHZMkbd26Vc2bN3dVKQAANCgum3l36tRJYWFhio+Pl81mU3JystLS0uTn\n56fIyEgNGTJEAwcOlKenpzp27KjOnTu7qhQAABoUl77n/dhjj5Xb/v1l8fj4eMXHx7vy9AAANEjc\nYQ0AAIshvAEAsBjCGwAAiyG8AQCwGMIbAACLIbwBALAYwhsAAIshvAEAsBjCGwAAiyG8AQCwGMIb\nAACLIbwBALAYwhsAAIshvAEAsBjCGwAAiyG8AQCwGMIbAACLIbwBALAYwhsAAIshvAEAsBjCGwAA\niyG8AQCwGMIbAACLIbwBALAYwhsAAIshvAEAsBjCGwAAiyG8AQCwGMIbAACLIbwBALAYu7sLANAw\nLF25Qxu3ZVXr2PyCYgX4+bi4IqDhYuYNoFZs3Jal/ILiah0b4Oeja9oEu7gioOFi5g2g1gT4+WjG\ng9e5uwygwWPmDQCAxVQrvMvKypSdne3qWgAAQDU4De9169YpIiJCiYmJkqTU1FR9/vnnLi8MAABU\nzGl4P/vss1q6dKmCgoIkSUOHDtWLL77o8sIAAEDFnIZ348aNdeGFFzq2AwMD5eXl5dKiAABA5Zx+\n2rxRo0basGGDJOnQoUN6//335ePD9zMBAHAXpzPv5ORkvfrqq/rhhx9000036csvv9TkyZProjYA\nAFABpzPvPXv2aP78+eUeW7FihUJCQlxWFAAAqFyl4b13715lZmZq+vTpGjt2rIwxkqQTJ04oNTVV\nERERdVYkAAD4n0rDOzs7Wx988IH27dunuXPnOh738PBQfHx8nRQHAABOV2l4d+zYUR07dlSPHj1O\nm2V/9913Li8MAABUzOl73tdee60WLVqk/Px8SVJJSYmWL1+uNWvWuLw4AABwOqefNh8+fLh+/vln\npaWl6ciRI/r88881adKkOigNAABUxGl4FxcX66mnnlJISIjGjBmjN998Ux9++GG1Bk9NTVVcXJzi\n4+O1ZcuWcvt+++03JSQkqH///po4ceLZVQ8AwDnIaXiXlJSoqKhIZWVlys/P1wUXXKDMzEynA2/Y\nsEEZGRlasmSJUlJSlJKSUm7/tGnTNHjwYC1btkyenp769ddfz/5VAABwDnH6nnffvn21dOlS3X77\n7YqJiVFgYKAuu+wypwOfWtBEklq2bKlDhw6psLBQvr6+Kisr07fffqtZs2ZJOnkjGAAAUD1Owzs+\nPl42m02S1LVrV+Xm5qpt27ZOB87JyVFYWJhjOzAwUNnZ2fL19VVeXp6aNGmiqVOnKj09XZ07d9ao\nUaOqHC8goLHsdk+n5z0TQUF+tTreuYo+1lxD6KGn58n/TrjrtTSEHrobPay5uuqh0/AeOHCgFixY\nIEm66KKLdNFFF53ViU7d5OXUzwcOHNDAgQMVEhKi++67T6tWrdKNN95Y6fPz84vO6ryVCQryU3Z2\nQa2OeS6ijzXXUHpYWnry/+PueC0NpYfuRA9rzhU9rOyPAafh3bZtWz3//PPq2LFjudXEunbtWuXz\ngoODlZOT49jOyspyLCsaEBCgZs2a6dJLL3WMtX379irDGwAAnOQ0vH/66SdJ0jfffON4zGazOQ3v\nbt26ac6cOYqPj1d6erqCg4Pl6+t78qR2uy655BLt3r1bzZs3V3p6um6++eaavA4AAM4ZTsP71CXz\nM9WpUyeFhYU53jNPTk5WWlqa/Pz8FBkZqaSkJMc901u1aqXw8PCzOg8AAOcap+FdE4899li57TZt\n2jh+vuyyy7R48WJXnh4AgAbJ6fe8AQBA/UJ4AwBgMU7De9u2berXr5+io6MlSXPnztXmzZtdXhgA\nAKiY0/B+6qmnlJqa6viaV0xMjKZOnerywgAAQMWchrfdbi/3QbMWLVrIbnfp59wAAEAVnKaw3W5X\nZmam4xapq1evLne3NAAN19KVO7RxW1a1js0vKFaAn4+LKwIgVSO8x4wZowcffFD//e9/dfXVVysk\nJERPP/10XdQGwM02bsuqdigH+PnomjbBdVAVAKfh7eXlpf/85z/Ky8uTt7e34y5pAM4NAX4+mvHg\nde4uA8DvOH3P+4EHHlD//v313nvv6fjx43VREwAAqILTmffHH3+srVu36sMPP1R8fLxatGihvn37\nKiYmpi7qAwAAf1Ctm7S0a9dOo0eP1qJFi9SsWTM9/vjjrq4LAABUwunMOysrS5988ok++ugj5eXl\nKSYmRu+//35d1AYAACrgNLxvu+02xcTEaMyYMbrqqqvqoiYAAFCFSsM7KytLwcHBevPNNx03ZcnM\nzHTsv+SSS1xfHQAAOE2l4T19+nTNnDlTQ4YMkc1mK3djFpvNps8++6xOCgQAAOVVGt4zZ86UJL3y\nyitq2bJluX2bNm1ybVUAAKBSlX7a/PDhw9qzZ4+SkpKUmZnp+N+uXbs0duzYuqwRAAD8TqUz702b\nNumNN97QTz/9pEGDBjke9/DwUPfu3eukOAAAcLpKw7tHjx7q0aOHFi9erISEhLqsCQAAVKHS8F6+\nfLluu+02HThwQM8///xp+x999FGXFgYAACpWaXh7eJx8O5y1uwEAqF8qTeZbb71VkvTwww+rsLBQ\nvr6+ysnJ0e7du9WpU6c6KxAAAJTn9N7mkydP1ocffqiDBw8qPj5eCxcu1KRJk+qgNAAAUBGn4f3j\njz/q9ttv14cffqhbb71Vzz33nDIyMuqiNgAAUAGn4X3qzmqrVq1SeHi4JLGuNwAAbuQ0vFu0aKGY\nmBgdOXJEbdu21dtvvy1/f/+6qA0AAFTA6UfJp0yZol9++cVxi9QrrrhCTz/9tMsLAwAAFXMa3seO\nHdPKlSv1/PPPy2az6S9/+YuuuOKKuqgNAABUwOll8yeeeEKFhYWKj4/XHXfcoZycHE2YMKEuagMA\nABVwOvPOycnRrFmzHNs9e/ZUYmKiS4sCAACVczrzPnr0qI4ePerYLioqUnFxsUuLAgAAlXM6846L\ni1Pv3r3Vrl07SVJ6ejr3NQcAwI2chnf//v3VrVs3paeny2az6YknntBFF11UF7UBAIAKVBneq1ev\n1q5du3T11VcrIiKirmoCAABVqPQ97zlz5ujFF19UVlaWJkyYoHfffbcu6wIAAJWodOa9Zs0aLVq0\nSHa7XQUFBRo2bJhuueWWuqwNOOctXblDG7dlue38+QXFCvDzcdv5AVSs0pm3t7e3Yy1vPz8/lZaW\n1llRAE7auC1L+QXu+3ZHgJ+PrmkT7LbzA6hYpTNvm81W5TaAuhHg56MZD17n7jIA1COVhvfOnTv1\n+OOPV7rN/c0BAHCPSsP7scceK7fdtWtXlxcDAACcqzS8b7311rqsAwAAVJPT26MCAID6hfAGAMBi\nqhXe+fn5+uGHHyRJZWVlLi0IAABUzWl4v/fee4qLi9O4ceMkSZMnT9a//vUvlxcGAAAq5jS8X3vt\nNb3zzjsKCAiQJI0ZM0ZLly6t1uCpqamKi4tTfHy8tmzZUuExM2fOZH1wAADOgNPw9vPz03nnnefY\nbtSokby8vJwOvGHDBmVkZGjJkiVKSUlRSkrKacfs2LFDGzduPMOSAQA4tzkN74CAAP373/9WcXGx\n0tPTNWPGDAUGBjodeN26dY6VyFq2bKlDhw6psLCw3DHTpk3TiBEjzrJ0AADOTU7X837yySf13HPP\n6ciRI5owYYKuvvpqTZkyxenAOTk5CgsLc2wHBgYqOztbvr6+kqS0tDR16dJFISEh1So0IKCx7HbP\nah1bXUFBfrU63rmKPtZcZT309LRVuR//Q49qjh7WXF310Gl4n3/++Zo4cWKNT2SMcfx88OBBpaWl\n6bXXXtOBAweq9fz8/KIa1/B7QUF+ys4uqNUxz0X0seaq6mFp6cn/39DjqvF7WHP0sOZc0cPK/hhw\nGt49evSocFGSVatWVfm84OBg5eTkOLazsrIUFBQkSVq/fr3y8vJ011136fjx49qzZ49SU1OVlJTk\nrBwAAM55TsP7n//8p+PnkpISrVu3TsXFzpco7Natm+bMmaP4+Hilp6crODjYcck8Ojpa0dHRkqS9\ne/dq3LhxBDcAANXkNLz/+J508+bNNWTIEN19991VPq9Tp04KCwtTfHy8bDabkpOTlZaWJj8/P0VG\nRtaoaAAAzmVOw3vdunXltvfv3689e/ZUa/A/rkzWpk2b044JDQ3VggULqjUeAACoRnjPmzfP8bPN\nZpOvr6+efPJJlxYFAAAq5zS8x44dW+4rXwAAwL2c3qRl+vTpdVEHAACoJqcz72bNmikxMVEdOnQo\nd1vURx991KWFAQCAijkN79DQUIWGhtZFLQAAoBoqDe93331Xt9xyix5++OG6rAcAADhR6Xvey5Yt\nq8s6AABANTn9wBoAAKhfKr1svmnTJt14442nPW6Mkc1mc3pvcwAA4BqVhveVV16pWbNm1WUtAACg\nGioNb29v72qvtQ0AAOpOpe95t2/fvi7rAAAA1VRpeI8ePbou6wAAANXEp80BALAYwhsAAIshvAEA\nsBjCGwAAiyG8AQCwGMIbAACLIbwBALAYwhsAAIshvAEAsBjCGwAAiyG8AQCwGMIbAACLIbwBALAY\nwhsAAIuxu7sAoCpLV+7Qxm1Z7i7DpTw9bSotNRXuyy8oVoCfTx1XBKC+Y+aNem3jtizlFxS7uwy3\nCfDz0TVtgt1dBoB6hpk36r0APx/NePA6d5fhMkFBfsrOLnB3GQAshJk3AAAWQ3gDAGAxhDcAABZD\neAMAYDGENwAAFkN4AwBgMYQ3AAAWQ3gDAGAxhDcAABZDeAMAYDGENwAAFkN4AwBgMSxMAreo7lKf\nLIkJAKdj5g23qO5SnyyJCQCnY+YNt2noS30CgKu4NLxTU1O1efNm2Ww2JSUlqX379o5969ev16xZ\ns+Th4aEWLVooJSVFHh5cCAAAwBmXpeWGDRuUkZGhJUuWKCUlRSkpKeX2T5w4UbNnz9Zbb72lI0eO\n6Msvv3RVKQAANCguC+9169bI8W0/AAATMElEQVQpIiJCktSyZUsdOnRIhYWFjv1paWlq2rSpJCkw\nMFD5+fmuKgUAgAbFZeGdk5OjgIAAx3ZgYKCys7Md276+vpKkrKwsrV27Vj169HBVKQAANCh19oE1\nY8xpj+Xm5mro0KFKTk4uF/QVCQhoLLvds1ZrCgryq9XxzlVn00dPT9tZP7chog81Rw9rjh7WXF31\n0GXhHRwcrJycHMd2VlaWgoKCHNuFhYW69957NXz4cHXv3t3pePn5RbVaX1CQn7KzC2p1zHPR2fax\ntPTkH3P8G/C7WBvoYc3Rw5pzRQ8r+2PAZZfNu3Xrpo8//liSlJ6eruDgYMelckmaNm2aBg0apBtu\nuMFVJQAA0CC5bObdqVMnhYWFKT4+XjabTcnJyUpLS5Ofn5+6d++ut99+WxkZGVq2bJkkqU+fPoqL\ni3NVOQAANBgufc/7scceK7fdpk0bx89bt2515akBAGiwuCsKAAAWQ3gDAGAxhDcAABbDwiSokrOl\nOz09bY6vfZ0JlvoEgLPHzBtVqu7SnWeKpT4B4Owx84ZTVS3dyY0dAKDuMfMGAMBiCG8AACyG8AYA\nwGIIbwAALIbwBgDAYghvAAAshvAGAMBiCG8AACyG8AYAwGIIbwAALIbwBgDAYghvAAAshoVJzlHO\nlvo8haU7AaD+YeZ9jqruUp8s3QkA9Q8z73NYVUt9AgDqL2beAABYDOENAIDFEN4AAFgM4Q0AgMUQ\n3gAAWAzhDQCAxRDeAABYDOENAIDFEN4AAFgM4Q0AgMUQ3gAAWAzhDQCAxbAwiZtUd0lOV2GpTwCw\nLmbeblLdJTldhaU+AcC6mHm7EUtyAgDOBjNvAAAshvAGAMBiCG8AACyG8AYAwGIIbwAALIbwBgDA\nYghvAAAshvAGAMBiCG8AgGV8+ulH6tHjrzp48KDjsZSUSVq79styx/XvH6uioiJJ0k8/pWvYsPt1\n//33aPDgAfrHP16WMeaMz719+y8aOnSwHnhgsJ55Zupp+9PS0tSv3816+OH79PDD9+mNN14tt//t\nt5erf//YMz5vRQhvAIBlfPrpxwoJCdWqVSuqdfyRI4V68sknNGLEaM2f/5pefvl1bd/+i/7zn7fP\n+NyzZ8/Uo4+O0osv/kOFhYVat27taceEh0fqhRde1gsvvKxBg4Y4Hs/Pz9MXX3x+xuesjEtvj5qa\nmqrNmzfLZrMpKSlJ7du3d+z76quvNGvWLHl6euqGG27QQw895MpSAAAWd/jwIf30U7rGjZuof/7z\nTf3tb/2dPufTTz/SDTf00OWXXyFJstvteuKJJ+Xj06jccW+88ao2bvy63GOjRo1VixaXS5JKSkr0\n22+/qm3bMElSt27X65tvNqhr127Vqn3evNkaMmSokpPHVet4Z1wW3hs2bFBGRoaWLFminTt3Kikp\nSUuWLHHsnzJlil599VVddNFFGjBggKKionTFFVe4qhwAQC1xxaqI17QJ1h3hVWfAypUrdN113fXX\nv3bV9OlTlJ2dpaCgqhdYysjIcATuKY0bNzntuEGDhpSbKf/RoUMH5efn59gOCAhUbm7Oacd9//13\nGjlymEpLT+ihhx5Vq1Zt9N1338jHx0dhYe2qrPVMuCy8161bp4iICElSy5YtdejQIRUWFsrX11eZ\nmZny9/fXxRdfLEnq0aOH1q1bV2fhvXTlDn23PVulpWf+nkdtYUlOADgzK1Z8rEGDhsjT01M9e/bS\nZ599ovj4AZUeb7PZZLNJZWWltV5LRe+Zd+jQQYMHn6frruuurVu3aMqUZL366kK9+up8TZ06s1bP\n77LwzsnJUVjY//7aCQwMVHZ2tnx9fZWdna3AwMBy+zIzM6scLyCgsex2z1qp7bzG3pIkT09brYx3\nNi68oJG6dQhRUJCf84PruYbwGtyNHtYcPay56vbwobiOLq7kdPv379ePP27VSy/N1vz5Nh07dkx+\nfn4aNuwBXXxxsGy2knL1l5WV6tJLgxUW1kbbt28vty8vL09Hjx5VSEiI47EXX3xRX331VblzJicn\nOyaVF1zQSIWFBY5xiosLdOml5f8bHhTkp5YtW0qSevbspgkTDik7O1OHDuVr3LgRkqTc3Bylpk7U\ns88+W6N+1NmSoGfzyb7fy88vqqVKpNhrL9Xg2DBlZxfU2phnqz7UUBNBQX6Wfw3uRg9rjh7WXH3v\n4ZIly3Xrrbdr2LCTIWiMUXz8rfr++5905ZV/0fLlS9Slyw2y2+369NOP1K5dB2VnF6hr15566aWX\ndcMNEbryynYqKSnRk0+O17XXXqc+ff7mGL9//wHq3//0WfzvexIaeqlWrPhSHTr8Re+994Fuuy2u\n3P63335LTZpcoMjIaO3atUPnn++vZs0u18KFy353nlglJT1V7V5X9geVy8I7ODhYOTn/ez8gKytL\nQUFBFe47cOCAgoOrft8CAHDuWrHiY02Y8KRj22azqXfvPo5L6bt379JDD90rLy8v/elPf9KIEY9L\nkho3bqyZM5/X00+nqri4WJ6enoqMjC4X3NX1yCOjNGNGqowp05VXttM11/xVkjR27EhNmzZLsbGx\nGj58pN55J02lpSc0duwTtfPiK2AzNZ0SV+K7777TnDlz9Nprryk9PV1TpkzR4sWLHftvvvlmzZ8/\nX02bNlVcXJyeeeYZtWjRotLxavsvwvr+V6ZV0Meao4c1Rw9rjh7WnCt6WOcz706dOiksLEzx8fGy\n2WxKTk5WWlqa/Pz8FBkZqUmTJmnUqFGSpJiYmCqDGwAA/I/LZt61jZl3/UQfa44e1hw9rDl6WHN1\nOfPmDmsAAFgM4Q0AgMUQ3gAAWAzhDQCAxRDeAABYDOENAIDFEN4AAFgM4Q0AgMVY5iYtAADgJGbe\nAABYDOENAIDFEN4AAFgM4Q0AgMUQ3gAAWAzhDQCAxZwT4Z2amqq4uDjFx8dry5Yt5fZ99dVX6t+/\nv+Li4jR37lw3VVj/VdXD9evX64477lB8fLzGjRunsrIyN1VZv1XVw1NmzpypxMTEOq7MOqrq4W+/\n/aaEhAT1799fEydOdFOF1lBVHxctWqS4uDglJCQoJSXFTRXWf7/88osiIiK0cOHC0/bVSa6YBu7r\nr7829913nzHGmB07dpg77rij3P7evXubX3/91ZSWlpqEhASzfft2d5RZrznrYWRkpPntt9+MMcYM\nGzbMrFq1qs5rrO+c9dAYY7Zv327i4uLMgAED6ro8S3DWw0ceecR88sknxhhjJk2aZPbt21fnNVpB\nVX0sKCgwPXv2NCUlJcYYY+655x6zadMmt9RZnx05csQMGDDATJgwwSxYsOC0/XWRKw1+5r1u3TpF\nRERIklq2bKlDhw6psLBQkpSZmSl/f39dfPHF8vDwUI8ePbRu3Tp3llsvVdVDSUpLS1PTpk0lSYGB\ngcrPz3dLnfWZsx5K0rRp0zRixAh3lGcJVfWwrKxM3377rcLDwyVJycnJatasmdtqrc+q6qOXl5e8\nvLxUVFSkEydO6OjRo/L393dnufWSt7e3XnnlFQUHB5+2r65ypcGHd05OjgICAhzbgYGBys7OliRl\nZ2crMDCwwn34n6p6KEm+vr6SpKysLK1du1Y9evSo8xrrO2c9TEtLU5cuXRQSEuKO8iyhqh7m5eWp\nSZMmmjp1qhISEjRz5kx3lVnvVdVHHx8fPfTQQ4qIiFDPnj3VoUMHtWjRwl2l1lt2u12NGjWqcF9d\n5UqDD+8/MtwNtsYq6mFubq6GDh2q5OTkcv9hQMV+38ODBw8qLS1N99xzjxsrsp7f99AYowMHDmjg\nwIFauHChfvzxR61atcp9xVnI7/tYWFio+fPn66OPPtJnn32mzZs3a9u2bW6sDpVp8OEdHBysnJwc\nx3ZWVpaCgoIq3HfgwIEKL4Oc66rqoXTy//D33nuvhg8fru7du7ujxHqvqh6uX79eeXl5uuuuu/Tw\nww8rPT1dqamp7iq13qqqhwEBAWrWrJkuvfRSeXp6qmvXrtq+fbu7Sq3Xqurjzp07dckllygwMFDe\n3t7q3Lmztm7d6q5SLamucqXBh3e3bt308ccfS5LS09MVHBzsuMwbGhqqwsJC7d27VydOnNDnn3+u\nbt26ubPceqmqHkon36sdNGiQbrjhBneVWO9V1cPo6Gh98MEHWrp0qV544QWFhYUpKSnJneXWS1X1\n0G6365JLLtHu3bsd+7ncW7Gq+hgSEqKdO3fq2LFjkqStW7eqefPm7irVkuoqV86JVcWeeeYZffPN\nN7LZbEpOTtaPP/4oPz8/RUZGauPGjXrmmWckSTfddJOGDBni5mrrp8p62L17d11zzTXq2LGj49g+\nffooLi7OjdXWT1X9Hp6yd+9ejRs3TgsWLHBjpfVXVT3MyMjQ2LFjZYxRq1atNGnSJHl4NPj5yVmp\nqo9vvfWW0tLS5OnpqY4dO+rxxx93d7n1ztatWzV9+nTt27dPdrtdF110kcLDwxUaGlpnuXJOhDcA\nAA0Jf5YCAGAxhDcAABZDeAMAYDGENwAAFkN4AwBgMYQ34CaPPvqobr31Vu3fv7/SY1q3bq0TJ07U\n6DyJiYm65ZZblJiYqAEDBighIUEbN248q7FGjBihAwcOSJLeeecdSdJPP/2kyZMn16jGM3HqvMC5\njK+KAW7Stm1bbdq0qdJ7JEsnwzs9PV12u/2sz5OYmKgHHnhA1113naSTSxnec889WrNmjWw221mN\neeDAAQ0fPlyLFy8+67rORmlpqWJiYhw3GQHOVcy8ATcYP368ysrK9Pe//12//vqrnn/+ecXHxys+\nPl7Dhw9XSUlJuePXr1+v22+/XYmJiYqLi3Oswbx+/XrHjPruu+9WZmam03O3atVKJ06cUH5+voqK\nijRq1CjdddddiouL0z//+U9JJwM+Pj5eiYmJuv322x33CQ8PD1dGRoZGjRqlX375RY8//ri+/vpr\nJSQkaNWqVRo8eLDjPN98841uv/12SdKCBQt05513Ki4uTuPGjXPcweuUvXv3KjY2VqNGjdJLL72k\noqIiDRs2TImJierXr59efvllSVJSUpL27dvnOM8HH3ygO++8UwkJCXrooYdY0Q7njlpfZBRAtbRq\n1cqUlJSYkpISM3/+fFNaWmqMMWbw4MFm5cqV5Y4ZOnSoef/9940xxuzcudOsWLHCFBUVmZtuusnk\n5+cbY4z59NNPzcMPP3zaeQYMGGDWrl3r2P7qq69MdHS0McaYl156yUyaNMkYY8zRo0dNz549zZ49\ne8zkyZPN/PnzjTHG5OTkmH//+9/GGGN69uxpdu/ebdavX2/i4+ONMcbxc0lJienWrZujnqeeesos\nWLDAbN682SQmJpqysjJjjDEpKSnmzTffLFdjZmamadu2rdm5c6cxxpg9e/Y4zllcXGw6depkCgoK\nTGZmprn++uuNMcb8+uuvJjY21hQXFxtjjHn99dfN1KlTz+afArCcs78WB6BW2O12eXh46M4775Td\nbteuXbtOm0HGxsZq1qxZ2rJli3r16qVevXppy5Ytys7O1rBhwySdvKRc2WXwadOmyd/fX8YYBQYG\nat68eZKkzZs3q1+/fpKkRo0aqV27dkpPT1dUVJTGjh2rX3/9VT179lTfvn2r9ToiIyO1YsUK9evX\nT5999pnS0tK0fPly7dmzRwMHDpQkFRUVVfg2gL+/vy6//HJJ0p/+9Cd9++23euutt+Tl5aXi4mId\nPHiw3PGbNm1Sdna249aTx48fV2hoqNM6gYaA8Abc7Ntvv9Xy5cu1fPlyNW7cWI888shpx8TExKh7\n9+5as2aN5s6dq/bt2ysmJkbNmjWr1n3Qx44d63jP+/f+GPbGGNlsNl1zzTV67733tG7dOqWlpend\nd9+t1hrZffr00UsvvaTQ0FC1adPGsTpVeHi4Jk6cWOVzvby8HD+/8cYbOn78uBYvXiybzaa//vWv\npx3v7e2t9u3ba/78+U7rAhoa3vMG3Cw3N1chISFq3Lix9u3bp++//17Hjx8vd8zs2bMdH9YaP368\nNm3apObNmys/P1+//PKLJGnjxo1asmTJGZ27Q4cO+vLLLyWdnBGnp6crLCxMCxYs0P79+xUeHq6U\nlBRt3ry53PM8PDwq/BR8p06dlJmZqXfffVe33HKL47EvvvhCR44ckSQtWrRImzZtctqTli1bymaz\n6bPPPtOxY8d0/Pjxcue96qqrHFcfJOnDDz/UihUrzuj1A1bFzBtws27duukf//iHEhIS9Oc//1nD\nhg3T3Llzy802L7vsMg0ePFjnn3++ysrKNGzYMDVq1EgzZszQ+PHj5ePjI0l66qmnzujciYmJeuKJ\nJ3TXXXfp+PHjevDBBxUaGqrLL79co0aNUpMmTVRWVqZRo0aVe94VV1yh3Nxc3XPPPRo6dKjjcZvN\npqioKL311ltKTk6WdDJk77rrLiUmJsrHx0fBwcGOS/WVue222zRy5EitWbNGvXr1UmxsrB577DEt\nXbpUF154ofr166eFCxdq/Pjxuv/++3XeeeepUaNGmj59+hm9fsCq+KoYAAAWw2VzAAAshvAGAMBi\nCG8AACyG8AYAwGIIbwAALIbwBgDAYghvAAAshvAGAMBi/h+xs4XY/95b7gAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAATgAAAE9CAYAAACfsn7ZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nOADGInAAjEXgABiLwAEwFoEDYKz/A2ZDZMe13sOlAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#compute the Area under the ROC Curve\n",
    "PredictProbLDA=model.predict_proba(xtest)[:,1]\n",
    "ROCCURVE(PredictProbLDA,ytest,Ypred_LDA)\n",
    "ConfusionmatrixPlot(ytest, Ypred_LDA)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "II-6 GAUSSIAN NAIVE BAYE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Acuracy of the model is: 0.5535714285714286\n",
      "The error of prediction is: 0.48086734693877553\n",
      "[[22 16]\n",
      " [ 9  9]]\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "         -1       0.71      0.58      0.64        38\n",
      "          1       0.36      0.50      0.42        18\n",
      "\n",
      "avg / total       0.60      0.55      0.57        56\n",
      "\n"
     ]
    }
   ],
   "source": [
    "model = GaussianNB()\n",
    "model.fit(xtrain, ytrain)\n",
    "Ypred_GNB= model.predict(xtest)\n",
    "accuracy = accuracy_score(ytest.flatten(),Ypred_GNB)\n",
    "\n",
    "print(\"The Acuracy of the model is:\",accuracy)\n",
    "print(\"The error of prediction is:\", error(ytest,Ypred_GNB))\n",
    "print(confusion_matrix(ytest,Ypred_GNB))\n",
    "print(classification_report(ytest,Ypred_GNB))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "the Area under the ROC curve:0.586257\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe8AAAFnCAYAAACPasF4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzt3XlYVGX/BvB7YFhkERkFF3BBFBUQ\nc8lXwyUQAlHErcAFTS3TzHJNxQVX1EorfV17y9JM8VVSK3Pf0jC1FAV3VMSNHWRfZp7fH/6cNxIY\nFGaGA/fnurwuzpwz53zny8g9zzMz58iEEAJEREQkGQb6LoCIiIheDMObiIhIYhjeREREEsPwJiIi\nkhiGNxERkcQwvImIiCSG4U2S1qpVK3h7e8PX1xe+vr7w9vZGSEgIcnJyKv1Yhw4dwqxZsyp9v/oW\nFRWFa9euAQC+//57fPHFF1o/ZqtWrfD48WOtH+efbt++jXPnzr3w/VasWIFt27aVuc1vv/2Ghw8f\nlnt7ooqQ8XveJGWtWrXCiRMn0KBBAwBAQUEBJk+ejBYtWmDy5Ml6rk4a5s2bh44dOyIgIEBnx/zn\n701XNm7ciKKiIrz//vuVvu8xY8Zg/Pjx6NSpU6Xvm+ifOPKmasXY2Bjdu3fH1atXATwN88WLF8PH\nxweenp5Yv369etvo6GgMHDgQPj4+GD58OOLj4wEAt27dwvDhw+Hj4wN/f39cvnwZABAREYG3334b\nJ06cgL+/f7HjBgQE4OTJk3jy5AmmT58OHx8f9OrVC7t27VJv06pVK2zYsAE+Pj5QKpXF7p+fn495\n8+bBx8cHvXv3xrJly9TbtGrVCps3b0ZAQAC6du1abEQXHh4OX19feHp6YsqUKcjLywMAzJw5E0uX\nLoW/vz9+/fVX5ObmYtKkSeo+LF++HACwbds27NmzB59++ik2bdqE1atXY/bs2QCA4OBgbNq0CUOG\nDEH37t0xZcoUPHutHxERAXd3d/Tr1w8RERFo1apVib+PkydPok+fPvDx8cF7772H9PR09boTJ05g\n4MCB6NatG7755hv17WvWrIGPjw+8vLzw3nvv4cmTJwCA1atXY86cORg8eDC+/fZbqFQqLFiwQP2Y\npk+fjsLCQgBAamoqxo0bh169esHf3x+nTp3C0aNHsWHDBmzevBnLli17of7NnDkTa9euBfB0dqJ3\n797w9fXF4MGDcfPmTXzxxRc4c+YMpk+fjn379hXbvrTnWUn7ISo3QSRhTk5O4tGjR+rl9PR0MWzY\nMLF27VohhBD//ve/xciRI0V+fr7Izs4W/fv3F0ePHhVCCOHt7S2OHz8uhBBi06ZN4t133xVKpVK8\n8cYbYseOHUIIIc6fPy+6desmCgsLxa5du9T76tSpk7h3754QQoh79+6Jzp07i8LCQjFr1izx8ccf\nC6VSKVJSUkTPnj3F9evX1bWuW7euxMexYcMG8e6774rCwkKRm5srBg0aJHbv3q2+38KFC4UQQsTG\nxgpXV1eRmpoqzp07J7p27SoeP34shBBi7ty5YtmyZUIIIWbMmCH8/f1FXl6eEEKIr7/+WrzzzjtC\npVKJ9PR00blzZ3Hu3DkhhBDDhw9XH2vVqlUiJCREffvw4cNFbm6uyM7OFl27dhXnz58XaWlpws3N\nTVy/fl0olUoxefJk4eTk9Nxjys7OFp07d1Y//sWLF4v58+erH9OKFSuEEEJcunRJtG3bVhQUFIjL\nly+Lrl27iszMTKFUKsXbb78t1qxZo66tW7duIiUlRQghxP79+0Xfvn1FQUGByMvLE71791Y/jpCQ\nEPHJJ58IIYSIiYkRnTt3Fvn5+WLGjBnq/b1I/57dLzMzU3Tq1ElkZmYKIYTYt2+f2LhxoxBCCA8P\nD3VP/36ckp5nZe2HqDw48ibJCw4Ohq+vL3r16oVevXqhS5cuePfddwEAx44dw9ChQ2FsbAwzMzME\nBATg4MGDuHPnDtLS0tCzZ08AwPDhw7F69Wrcvn0bKSkpGDx4MACgY8eOUCgUuHDhgvp4xsbG8PDw\nwNGjRwEAhw8fhpeXF+RyOY4dO4YRI0bAwMAACoUC3t7eOHjwoPq+r7/+eomP4fjx43jrrbcgl8th\namoKf39/nD59Wr1+0KBBAIDmzZvDwcEBly5dwtGjR+Hn54f69esDAIYMGVLsWF27doWJiQkAYPTo\n0Vi7di1kMhmsrKzQsmVL3L9/X2NvfX19YWpqCjMzMzRr1gyPHj1CVFQUmjVrBicnJxgYGGDIkCEl\n3vevv/5CgwYN4OTkBACYPn16sc8M9OvXDwDg7OyM/Px8pKWlwdXVFcePH4eFhQUMDAzQvn179UgV\nANq1aweFQgEA8PHxwa5du2BkZAQTExO0bdtWve2JEyfQt29f9f6PHDkCY2PjYvW9SP+eMTExgUwm\nw86dO5GcnIzevXurn2slKe159qL7Ifonub4LIKqoLVu2oEGDBkhNTYWvry/8/Pwglz99amdmZmLp\n0qVYuXIlgKfT6G5ubkhLS4OlpaV6H3K5HHK5HE+ePEFeXh569+6tXpeVlVVsuhd4GhybN2/GyJEj\ncfjwYfV7qJmZmZg0aRIMDQ0BPJ0O9/X1Vd+vTp06JT6G1NRUWFlZqZetrKyQkpJSbPnvPz958gSZ\nmZk4dOgQTp06BQAQQqinjf95n7t372LZsmW4ffs2DAwM8PjxYwwcOLDMvgKAhYWF+mdDQ0MolUo8\nefKk2L6fhd8/paWloXbt2urlf4bns30/65VKpUJubi6WLl2KP/74AwCQkZFR7AXP34+bmpqKRYsW\n4cqVK5DJZEhOTsbIkSMBAOnp6cV+v39/HM+8SP+eMTIywrfffov169dj9erVaNWqFUJDQ0t926C0\n5xmAF9oP0T8xvKnaUCgUCA4Oxqeffop169YBAGxtbTF69Gh4eHgU2/bOnTtIT0+HSqWCgYEBCgsL\nkZCQAFtbW5ibm2P//v3P7T8iIkL9c/fu3RESEoK7d+/i7t276NKli/p4a9asUY82y6tevXrFXiCk\np6ejXr166uW0tDTY2dmp11lZWcHW1hYDBgzAjBkzNO5/4cKFcHFxwZo1a2BoaIigoKAXqu/vLCws\nin2aPzExscTtrK2tkZaWpl7Ozc1FRkZGmR9S++6773D37l1ERETA3Nwcn3/+ORISEkrc9vPPP4dc\nLsdPP/0EY2NjTJ06Vb2uTp06SEtLg729PQDg/v37z73IeJH+/Z2zszNWrVqFgoIC/Oc//0FoaCi2\nb99e4rbW1tYlPs/s7e1faD9E/8Rpc6pWRo0ahQsXLuDs2bMAgF69euG///0vlEolhBBYu3YtTp48\niWbNmqFBgwbqadKdO3di3rx5sLOzQ4MGDdThnZqaiilTpjz31TNjY2N069YNn376KXr16qUePXp6\neqr/ABcVFSEsLAwxMTEa63799dexc+dOKJVK5OTkYM+ePeqpVgD45ZdfAACxsbGIi4tDu3bt4Onp\niYMHDyI1NRXA0+n7jRs3lrj/lJQUtGnTBoaGhjh9+jTi4uLUj0kulyMzM7N8DQbg4uKC69evIy4u\nDiqVCjt37ixxu44dOyIpKQmXLl0CAKxduxZr1qwpc98pKSlo3rw5zM3N8eDBA5w4caLUr/2lpKTA\nyckJxsbGuHbtGi5cuKDe1tPTEz/++COApx9AHDhwIJRKZbHH+iL9e+b69ev48MMPUVBQAGNjY7i6\nukImkwEouY+lPc/K2g9ReXDkTdWKhYUFxo4di+XLl2Pnzp0YOnQo7t+/jz59+kAIAVdXV4wcORIy\nmQxffvklpk+fjpUrV8LGxgZLly6FTCbDypUrMX/+fHzxxRcwMDDAqFGjYGZm9tyxfHx8MHHiRHz7\n7bfq2yZNmqT+BDTwdIRenqnQ4OBgxMfHo0+fPpDJZPD19S02da9QKBAQEICEhATMmTMHVlZWsLKy\nwrhx4xAcHAyVSoW6detiwYIFJe5//PjxWLp0KdauXYtevXrhgw8+wKpVq9CmTRt4eXnh008/RXx8\nfInTy/9ka2uLKVOmYMSIEahXrx6CgoLUQfl3tWrVwurVqzF9+nQAQNOmTdWf8i5NUFAQPvzwQ/j4\n+KBVq1aYOXPmcz1+ZvTo0ZgxYwYiIiLQqVMnzJgxA7Nnz4abmxumT5+OGTNmwNPTE+bm5vjss89g\namoKDw8PTJs2DQ8ePMCqVavK3b9nnJycYG9vj759+8LIyAjm5uaYN28egKfPhylTpuDDDz9Ub1/a\n86xx48al7oeoPPg9b6IqTl/fiS6LEEI9Urx58yaGDh36Uic/IaKXw2lzInohRUVF6N69O6KiogAA\n+/btwyuvvKLnqohqFk6bE9ELkcvlCA0NxYwZMyCEgI2NDZYsWaLvsohqFE6bExERSQynzYmIiCSG\n4U1ERCQxknnPOymp/N9DLQ9razOkpVX+ZSNrGvax4tjDimMPK449rDht9NDGxrLE22vsyFsuN9R3\nCdUC+1hx7GHFsYcVxx5WnC57WGPDm4iISKoY3kRERBLD8CYiIpIYhjcREZHEMLyJiIgkhuFNREQk\nMQxvIiIiiWF4ExERSYxWw/vGjRvw8vLC999//9y633//HYMHD0ZgYCDWrFmjzTKIiIiqFa2Fd05O\nDhYtWoSuXbuWuH7x4sVYvXo1tm3bhtOnT+PWrVvaKoWIiKha0dq5zY2NjfHVV1/hq6++em5dfHw8\nrKys0LBhQwBAz549ERkZiRYtWmirHCKiai0xLQd/Xk/Cy17j2dzcBNnZ+ZVaU03TuKEVXJtYQSaT\naf1YWgtvuVwOubzk3SclJUGhUKiXFQoF4uPjy9yftbVZpZ83trQTvtOLYR8rjj2suJrew28PXMfJ\nCw/0XUaN9/0CX1hZmGj9OJK5qpg2rtRS2Vcqq4nYx4pjDyuOPQSysgsAAOMCXGBq/OJ/2q2saiEj\nI7eyy6pRHBpboyC3AEm5BZW2z9JelOolvG1tbZGcnKxeTkhIgK2trT5KISKqVpybKWBRy+iF78cX\nQBWnyx7q5ati9vb2yMrKwv3791FUVIRjx47B3d1dH6UQERFJjtZG3tHR0Vi+fDkePHgAuVyOAwcO\nwNPTE/b29vD29sb8+fMxdepUAICfnx8cHBy0VQoREVG1orXwdnV1xZYtW0pd/+qrryI8PFxbhyci\nIqq2eIY1IiIiiWF4ExERSQzDm4iISGIY3kRERBLD8CYiIpIYhjcREZHEMLyJiIgkhuFNREQkMQxv\nIiIiiWF4ExERSQzDm4iISGIY3kRERBLD8CYiIpIYhjcREZHEMLyJiIgkhuFNREQkMQxvIiIiiWF4\nExERSQzDm4iISGIY3kRERBLD8CYiIpIYhjcREZHEMLyJiIgkhuFNREQkMQxvIiIiiWF4ExERSQzD\nm4iISGIY3kRERBLD8CYiIpIYhjcREZHEMLyJiIgkhuFNREQkMQxvIiIiiWF4ExERSQzDm4iISGIY\n3kRERBLD8CYiIpIYhjcREZHEyPVdABFRdaBUqZCQmguhp+Pn5hfp6cikDwxvIqJKsHn/dfx26ZG+\ny4CBTKbvEkgHGN5ERJUgLTMfANCjXSMYGugnQBvVM4eZKf+s1wT8LRMRVaJh3i1hJDfUdxlUzfED\na0RERBLD8CYiIpIYhjcREZHEaPU977CwMERFRUEmkyEkJARubm7qdVu3bsXevXthYGAAV1dXzJ49\nW5ulEBERVRtaG3mfPXsWcXFxCA8Px5IlS7BkyRL1uqysLHz99dfYunUrtm3bhtjYWFy8eFFbpRAR\nEVUrWgvvyMhIeHl5AQAcHR2RkZGBrKwsAICRkRGMjIyQk5ODoqIi5ObmwsrKSlulEBERVStaC+/k\n5GRYW1urlxUKBZKSkgAAJiYmmDBhAry8vODh4YF27drBwcFBW6UQERFVKzr7nrcQ/ztpYFZWFjZs\n2ID9+/fDwsICI0eOxLVr19C6detS729tbQZ5JX930sbGslL3V1OxjxXHHlacvntobPz0z2m9epYw\nNpLm97z13cPqQFc91Fp429raIjk5Wb2cmJgIGxsbAEBsbCwaN24MhUIBAOjUqROio6PLDO+0tJxK\nrc/GxhJJSZmVus+aiH2sOPaw4qpCDwsKnp5bPDk5U5InaakKPZQ6bfSwtBcDWps2d3d3x4EDBwAA\nMTExsLW1hYWFBQDAzs4OsbGxyMvLAwBER0ejWbNm2iqFiIioWtHayLtDhw5wcXFBUFAQZDIZQkND\nERERAUtLS3h7e2PMmDEYMWIEDA0N0b59e3Tq1ElbpRAREVUrWn3Pe9q0acWW/z4tHhQUhKCgIG0e\nnoiowvILlChSqTRup1Tp62KgVBPxwiRERKW4FpeGFeEXXzCYeUlO0j6GNxFRKR6n5UCpEmjawBIK\nSxON2zdtYAkjOc86TdrH8CYi0sDn1cbo4tJA32UQqfElIhERkcQwvImIiCSG4U1ERCQxDG8iIiKJ\nYXgTERFJDMObiIhIYhjeREREEsPwJiIikhiGNxERkcQwvImIiCSG4U1ERCQxDG8iIiKJ4YVJiKhU\nMXdSsW53NAqVmq9nrU8yANq4mraK1+imKorhTUSluvPoCXLyi1BfYQYzk6r758LIyACFhdp5gVHL\nxBBOjetoZd9EL6vq/m8koipjmHdLuDrU1XcZpbKxsURSUqa+yyDSGb7nTUREJDEMbyIiIolheBMR\nEUkMw5uIiEhiGN5EREQSw/AmIiKSGIY3ERGRxDC8iYiIJIbhTUREJDEMbyIiIolheBMREUkMw5uI\niEhieGESohqmsEiJleFRSM3M07htTl6RDioiohfF8CaqYZIz8nA9Ph3GRgYwNzUqc1tjI0Moapui\nsY2FjqojovJgeBPVUK+5NMAI39b6LoOIXgLf8yYiIpKYcoW3SqVCUlKStmshIiKictAY3pGRkfDy\n8kJwcDAAICwsDMeOHdN6YURERFQyjeH9+eefY8eOHbCxsQEAjBs3DuvWrdN6YURERFQyjeFtZmaG\nevXqqZcVCgWMjMr+hCoRERFpj8ZPm5uamuLs2bMAgIyMDPzyyy8wMTHRemFERERUMo0j79DQUHz9\n9de4fPky3njjDfz2229YtGiRLmojIiKiEmgced+7dw8bNmwodtvhw4dhZ2entaKIiIiodKWG9/37\n9xEfH4/ly5dj5syZEEIAAIqKihAWFgYvLy+dFUlERET/U2p4JyUlYd++fXjw4AHWrFmjvt3AwABB\nQUE6KY6IiIieV2p4t2/fHu3bt0fPnj2fG2X/9ddfWi+MiIiISqbxPe8uXbpg69atSEtLAwAUFhZi\n165dOHXqlNaLIyIioudpDO9JkyahUaNGOHXqFHx8fHD69GnMnz9fB6UR0Ys48ud9/HE1QeN2BYVK\nHVRDRNqkMbzz8/OxcOFCBAcHY8aMGUhPT8eiRYvK9YG1sLAwREVFQSaTISQkBG5ubup1jx49wpQp\nU1BYWAhnZ2csXLiwYo+EqIY7duEBHiZnw0Am07it3NAADo1q66AqItIGjeFdWFiInJwcqFQqpKWl\nwdraGvHx8Rp3fPbsWcTFxSE8PByxsbEICQlBeHi4ev2yZcswevRoeHt7Y8GCBXj48CEaNWpUsUdD\nVMNZ1DLCqo+667sMItIyjeEdEBCAHTt24M0334Sfnx8UCgWaNm2qccfPLmgCAI6OjsjIyEBWVhYs\nLCygUqnw559/YuXKlQCengiGiIiIykdjeAcFBUH2/9NwXbt2RUpKCtq0aaNxx8nJyXBxcVEvKxQK\nJCUlwcLCAqmpqTA3N8fSpUsRExODTp06YerUqWXuz9raDHK5ocbjvggbG8tK3V9NxT5WXGX00NDQ\nAAYGshr7+6ipj7sysYcVp6seagzvESNGYMuWLQCA+vXro379+i91oGcneXn2c0JCAkaMGAE7OzuM\nHTsWx48fx+uvv17q/dPScl7quKWxsbFEUlJmpe6zJmIfK66yeqhUqqBSiRr5++DzsOLYw4rTRg9L\nezGgMbzbtGmDL7/8Eu3bty92NbGuXbuWeT9bW1skJyerlxMTE9WXFbW2tkajRo3QpEkT9b5u3rxZ\nZngTERHRUxrD++rVqwCA8+fPq2+TyWQaw9vd3R2rV69GUFAQYmJiYGtrCwsLi6cHlcvRuHFj3L17\nF82aNUNMTAz69OlTkcdBRERUY2gM72dT5i+qQ4cOcHFxUb9nHhoaioiICFhaWsLb2xshISHqc6Y7\nOTnB09PzpY5DRERU02gM74qYNm1aseXWrVurf27atCm2bdumzcMTERFVSxqv501ERERVC8ObiIhI\nYjSG97Vr1zBw4ED4+voCANasWYOoqCitF0ZEREQl0xjeCxcuRFhYmPprXn5+fli6dKnWCyMiIqKS\naQxvuVxe7INmDg4OkMu1+jk3IiIiKoPGFJbL5YiPj1efIvXEiRPFzpZGRNrzKCUbP/1+F0VKzf/n\nUp/kQW7Ij7EQ1QQaw3vGjBl4//33cefOHXTs2BF2dnb45JNPdFEbUY33x5UEnInRfI3uZ1raW2ix\nGiKqKjSGt5GREX766SekpqbC2NhYfZY0ItI+1f8PuCcOagvHRlYatzevxbe0iGoCjf/Tx48fD0tL\nS/Tr1w99+/bVRU1E9A/mpkaobW6s7zKIqIrQGN4HDhxAdHQ0fv31VwQFBcHBwQEBAQHw8/PTRX1E\nRET0D+X6dIurqyumT5+OrVu3olGjRvj444+1XRcRERGVQuPIOzExEQcPHsT+/fuRmpoKPz8//PLL\nL7qojYiIiEqgMbwHDRoEPz8/zJgxA23bttVFTURERFSGUsM7MTERtra22Lx5s/qkLPHx8er1jRs3\n1n51RERE9JxSw3v58uVYsWIFxowZA5lMVuzELDKZDEeOHNFJgURERFRcqeG9YsUKAMBXX30FR0fH\nYusuXLig3aqIiIioVKV+2vzJkye4d+8eQkJCEB8fr/53+/ZtzJw5U5c1EhER0d+UOvK+cOECvvvu\nO1y9ehUjR45U325gYIBu3brppDgiIiJ6Xqnh3bNnT/Ts2RPbtm3DkCFDdFkTERERlaHU8N61axcG\nDRqEhIQEfPnll8+t/+ijj7RaGBEREZWs1PA2MHj6djiv3U1UuVRC4NczccjIKgAA1DIzRm5OQYnb\nxj7M0GVpRCQRpSbzgAEDAAAffPABsrKyYGFhgeTkZNy9excdOnTQWYFE1c3DpGzsOnG73NvLAFjx\noiRE9Dcah9WLFi1C69at4e3tjaCgILi6umLv3r1YuHChLuojqnaU/3+dzy4u9eHbuQkUCnOkpmaX\nur1FLSMoapvqqjwikgCN4X3lyhXMnTsX27Ztw4ABAzBhwoRinz4nopdjWcsYTepbwsbGEhZG5bpG\nEBERgHJcVezZmdWOHz8OT09PAEBBQcnvzxEREZH2aQxvBwcH+Pn5ITs7G23atMHu3bthZWWli9qI\niIioBBqnzRcvXowbN26oT5HaokULfPLJJ1ovjIiIiEqmMbzz8vJw9OhRfPnll5DJZHjllVfQokUL\nXdRGREREJdA4bT537lxkZWUhKCgIb731FpKTkzFnzhxd1EZEREQl0DjyTk5OxsqVK9XLHh4eCA4O\n1mpRREREVDqNI+/c3Fzk5uaql3NycpCfn6/VooiIiKh0GkfegYGB6N27N1xdXQEAMTExPK85ERGR\nHmkM78GDB8Pd3R0xMTGQyWSYO3cu6tevr4vaiIiIqARlhveJEydw+/ZtdOzYEV5eXrqqiYiIiMpQ\n6nveq1evxrp165CYmIg5c+Zg7969uqyLiIiISlHqyPvUqVPYunUr5HI5MjMzMXHiRPTr10+XtRER\nEVEJSh15Gxsbq6/lbWlpCaVSqbOiiKqrlIw8bDl4HQBQ29xIz9UQkVSVOvKWyWRlLhPRi7l4Kxlf\n/3wF2XlF6OJcH14dG+u7JCKSqFLDOzY2Fh9//HGpyzy/OVH5FClViDh5G/v/uAe5oQFG+LZCz3aN\n+IKYiF5aqeE9bdq0Ystdu3bVejFE1U3qkzys3xuDW/czYGtdC+/3d0WT+pb6LouIJK7U8B4wYIAu\n6yCqdi7fTsFXP11BVm4hXm1ti7d7t0YtE42nViAi0oh/SYgqmVKlwu7f7uCXyDjIDWUY/oYTPNrb\ncZqciCoNw5uoEqVl5mPD3hjciE+HTR1TjO/vimYNauu7LCKqZsoV3mlpabh//z7atm0LlUoFAwON\n1zMhqnFi7qRi408xyMwpREcnG4zyaw0zU34djIgqn8bw/vnnn7Fq1SoYGxvj559/xqJFi+Ds7Iw3\n33xTF/URVXkqlcDe03fw0+m7MDCQYYhXS3h1tOc0ORFpjcYh9KZNm7Bnzx5YW1sDAGbMmIEdO3aU\na+dhYWEIDAxEUFAQLl26VOI2K1as4PXBSbIysvKxIvwi9p6+C0VtU8wa3hHenRozuIlIqzSOvC0t\nLVGrVi31sqmpKYyMNE8Fnj17FnFxcQgPD0dsbCxCQkIQHh5ebJtbt27h3Llz5dofUVVzNS4NG/bG\n4El2AV5pUQ9j+raBOafJiUgHNI68ra2t8eOPPyI/Px8xMTH49NNPoVAoNO44MjJSfSUyR0dHZGRk\nICsrq9g2y5Ytw+TJk1+ydCL9UImn0+Sfbb+A7NxCBHq2wMRBbRncRKQzGkfeCxYswBdffIHs7GzM\nmTMHHTt2xOLFizXuODk5GS4uLuplhUKBpKQkWFhYAAAiIiLQuXNn2NnZlatQa2szyOWG5dq2vGxs\neLKMylCT+piemY8VP/yJizeSUK9OLcwI7oTWzTS/mNWkJvVQW9jDimMPK05XPdQY3rVr18a8efMq\nfCAhhPrn9PR0REREYNOmTUhISCjX/dPScipcw9/Z2FgiKSmzUvdZE9WkPl6/93SaPD2rAG6OdfFO\nX2dY1DKq8OOvST3UFvaw4tjDitNGD0t7MaAxvHv27Fnih2+OHz9e5v1sbW2RnJysXk5MTISNjQ0A\n4MyZM0hNTcWwYcNQUFCAe/fuISwsDCEhIZrKIdI5lRD49Uwcfjx5BwDw5uuO8PlXExjwQ2lEpCca\nw/uHH35Q/1xYWIjIyEjk5+dr3LG7uztWr16NoKAgxMTEwNbWVj1l7uvrC19fXwDA/fv3MWvWLAY3\nVUmZOQX4z89Xcfl2CqwtTfB/XoMhAAAfVklEQVRePxc4Na6j77KIqIbTGN7/fE+6WbNmGDNmDN5+\n++0y79ehQwe4uLggKCgIMpkMoaGhiIiIgKWlJby9vStUNJEu3LyfjvV7YpCWmQ9XBwXe8XdGbTNj\nfZdFRKQ5vCMjI4stP378GPfu3SvXzv95ZbLWrVs/t429vT22bNlSrv0R6YIQAgfOxmPn8VgICAzo\n0Rx9ujblNDkRVRkaw3vt2rXqn2UyGSwsLLBgwQKtFkWkL1m5hfjml6u4eCsZVubGeK+fC1o3tdZ3\nWURExWgM75kzZxb7yhdRdRX7MAPrd0cj5Uk+2jS1xth+LrAy5zQ5EVU9Gk/Ssnz5cl3UQaQ3Qggc\nPBePZd//hdQn+Qjo5oCpga8wuImoytI48m7UqBGCg4PRrl27Yqcx/eijj7RaGJEu5OQV4pt91/DX\njSTUNjPC2H4ucK6Ek64QEWmTxvC2t7eHvb29Lmoh0qk7j55g3e5oJGfkoXWTOhjbzwV1LEz0XRYR\nkUalhvfevXvRr18/fPDBB7qsh0jrhBA4+tcDhB+9CaVSoO9rzRDQrRkMeZ16IpKIUv9a7dy5U5d1\nEOlETl4R1u2JwdZDN2BqLMfkt9phYI/mDG4ikhSN0+ZE1UXc40ys2x2NxPRctLS3wrgAV1hbcpqc\niKSn1PC+cOECXn/99eduF0JAJpNpPLc5UVUhhMDxiw+x7fBNFClV8OvSFAN6OHC0TUSSVWp4Ozs7\nY+XKlbqshajS5eYX4bv913D2aiLMTeX4YKAr3Bzr6bssIqIKKTW8jY2Ny32tbaKqKD4xC2t3RyMh\nNQeOdrUxrp8r6lqZ6rssIqIKKzW83dzcdFkHUaURQuC3S4+w9dANFBap4NO5MQb1dITckNPkRFQ9\nlBre06dP12UdRJUir6AIWw5cR2RMAsxM5BgX4IL2LW30XRYRUaXip82p2niQ9HSa/FFKDhwa1sb4\nABfUq1NL32UREVU6hjdVC6cvP8KWA9dRUKSCVyd7vOXRgtPkRFRtMbxJ0vILldh68AZOXX6EWiZy\nTPB3RsdWtvoui4hIqxjeJFmPUrKxdnc0HiRlo2kDS4zv7wpbTpMTUQ3A8CZJiox5jM37ryO/UAnP\nDnYI9GwJIzmnyYmoZmB4k6QUFCrxw+GbOBn1EKbGhhgX4ILOberruywiIp1ieJNkPE7Nwbrd0YhP\nzEITWwuM7++K+gozfZdFRKRzDG+ShLNXE7Dp12vIL1Di9VcaYYhXSxjJDfVdFhGRXjC8qUorLFJi\n+5FbOHbhAUyMDDHW3xldXBrouywiIr1ieFOVlZiWg7W7o3EvIQt2NuZ4v78rGtY113dZRER6x/Cm\nKun8tURs+vUqcvOV6O7WEEO9nWBixGlyIiKA4U1VTGGRCjuO3cKRP+/D2MgAY/q0gXvbhvoui4io\nSmF4U5WRlJ6L9XuicedRJhrVM8f4/q6wq8dpciKif2J4U5Vw4UYSvv7lKnLyi/CaawMEv9EKJsac\nJiciKgnDm/SqSKnCzuOxOHguHkZyA4zq3Rrd3BpCJpPpuzQioiqL4U16k5KRh3V7onH74RM0UJjh\n/f6usLe10HdZRERVHsOb9OLirWR8/fMVZOcVoYtzfQT7tEItEz4diYjKg38tSaeKlCpEnLyN/X/c\ng9zQACN8W6Fnu0acJiciegEMb9KZ1Cd5WL83BrfuZ8DWuhbe7++KJvUt9V0WEZHkMLxJJy7fTsFX\nP11BVm4hXm1ti7d7t+Y0ORHRS+JfT9IqpUqF3b/dwS+RcZAbyjD8DSd4tLfjNDkRUQUwvElr0jLz\nsXFvDK7Hp8OmjinG93dFswa19V0WEZHkMbxJK2LupmLj3hhk5hSio5MNRvm1hpmpkb7LIiKqFhje\nVKlUKoG9p+/gp9N3YWAgwxCvlvDqaM9pciKiSsTwpkqTkZWPjT9dwdW4NNSt/XSavHkjTpMTEVU2\nhjdViqtxadi4NwYZ2QV4pUU9jOnbBuacJici0gqGN1XIs2nyPafuwEAmQ6BnC7zxamNOkxMRaRHD\nm17ak+wCrIqIxMUbSVDUNsG4AFe0sLPSd1lERNUew5teyvV7adiwNwbpWQVwc6yLd/o6w6IWp8mJ\niHSB4U0vRCUEfj0Thx9P3gEAjOzjjO6u9WHAaXIiIp1heFO5ZeYU4D8/X8Xl2ymoY2GMcQGucO/Q\nGElJmfoujYioRmF4U7ncvJ+O9XtikJaZD1cHBd7xd0ZtM2N9l0VEVCNpNbzDwsIQFRUFmUyGkJAQ\nuLm5qdedOXMGK1euhIGBARwcHLBkyRIYGBhosxx6CUIIHDgbj53HYyEgMKBHc/Tp2pTT5EREeqS1\ntDx79izi4uIQHh6OJUuWYMmSJcXWz5s3D6tWrcL27duRnZ2N3377TVul0EvKyi3E6l2XsePYLVia\nGWF6UHv4v9aMwU1EpGdaG3lHRkbCy8sLAODo6IiMjAxkZWXBwsICABAREaH+WaFQIC0tTVul0EuI\nfZiB9bujkfIkH22aWmNsPxdYmXOanIioKtDayDs5ORnW1tbqZYVCgaSkJPXys+BOTEzE6dOn0bNn\nT22VQi9ACIGD5+Kx7Pu/kPokHwHdHDA18BUGNxFRFaKzD6wJIZ67LSUlBePGjUNoaGixoC+JtbUZ\n5HLDSq3JxsayUvcndVm5hVgVfgGRlx+hjoUJpg3riHZONhrvxz5WHHtYcexhxbGHFaerHmotvG1t\nbZGcnKxeTkxMhI3N/4IgKysL7777LiZNmoRu3bpp3F9aWk6l1mdjY8mvOP3NnUdPsG53NJIz8tC6\nSR2M7eeCOhYmGnvEPlYce1hx7GHFsYcVp40elvZiQGvT5u7u7jhw4AAAICYmBra2tuqpcgBYtmwZ\nRo4ciR49emirBCoHIQSO/HkfS7//EykZeej7WjNMDXoFdSxM9F0aERGVQmsj7w4dOsDFxQVBQUGQ\nyWQIDQ1FREQELC0t0a1bN+zevRtxcXHYuXMnAKBv374IDAzUVjlUgpy8Iny7/xrOX0uERS0jjPV3\nhmvzuvoui4iINNDqe97Tpk0rtty6dWv1z9HR0do8NGkQ9zgT63ZHIzE9Fy3trTAuwBXWlhxtExFJ\nAc+wVsMIIXD84kNsO3wTRUoV/Lo0xYAeDjDkCXKIiCSD4V2D5OYXYfOB6/jjSgLMTeX4YKAr3Bzr\n6bssIiJ6QQzvGiI+MQtrd0cjITUHjna1Ma6fK+pameq7LCIiegkM72pOCIHfLj3C1kM3UFikgk/n\nxhjU0xFyQ06TExFJFcO7GssvUGLzgeuIjHkMMxM5xgW4oH1LzSddISKiqo3hXU09SHo6Tf4oJQcO\nDWtjfIAL6tWppe+yiIioEjC8q6HTlx9hy4HrKChSwauTPd7yaMFpciKiaoThXY3kFyqx9eANnLr8\nCLVM5Jjg74yOrWz1XRYREVUyhnc18SglG2t3R+NBUjaa1rfE+AGusOU0ORFRtcTwrgYiYx5j8/7r\nyC9UwrODHQI9W8JIzmlyIqLqiuEtYQWFSvxw+CZORj2EqbEhxgW4oHOb+voui4iItIzhLVGPU3Ow\nbnc04hOz0MTWAuP7u6K+wkzfZRERkQ4wvCXo7NUEbPr1GvILlHj9lUYY4tUSRnJDfZdFREQ6wvCW\nkMIiJbYfuYVjFx7AxMgQY/2d0cWlgb7LIiIiHWN4S0RiWg7W7o7GvYQs2NmY4/3+rmhY11zfZRER\nkR4wvCXg/LVEbPr1KnLzleju1hBDvZ1gYsRpciKimorhXYUVFqmw49gtHPnzPoyNDDCmTxu4t22o\n77KIiEjPGN5VVFJ6LtbvicadR5loVM8c4/u7wq4ep8mJiIjhXSVduJGEr3+5ipz8Irzm2gDBb7SC\niTGnyYmI6CmGdxVSpFRh5/FYHDwXDyO5AUb1bo1ubg0hk8n0XRoREVUhDO8qIiUjD+v3RCP24RM0\nUJjh/f6usLe10HdZRERUBTG8q4CoW8n4z89XkJ1XhC7O9RHs0wq1TPirISKikjEh9KhIqcKPJ2/j\n1z/uQW5ogBG+rdCzXSNOkxMRUZkY3nqS+iQP6/fG4Nb9DNha18L7/V3RpL6lvssiIiIJYHjrweXb\nKfjqpyvIyi3Eq61t8Xbv1pwmJyKicmNi6JBSpcLu3+7gl8g4yA1lGP6GEzza23GanIiIXgjDW0fS\nMvOxcW8Mrsenw6aOKcb3d0WzBrX1XRYREUkQw1sHYu6mYuPeGGTmFKKjkw1G+bWGmamRvssiIiKJ\nYnhrkUolsPf0Hfx0+i4MDGQY4tUSXh3tOU1OREQVwvDWkoysfGz86QquxqWhbu2n0+TNG3GanIiI\nKo7hrQVX49KwcW8MMrIL8EqLehjdpw0sanGanIiIKgfDuxKphMDPv9/FnlN3YCCT4S2PFvDp3JjT\n5EREVKkY3pXkSXYBvvopBjF306CobYJxAa5oYWel77KIiKgaYnhXguv30rBhbwzSswrg5lgX7/R1\n5jQ5ERFpDcO7AlRC4Nczcfjx5B0AwODXHeH7ryYw4DQ5ERFpEcP7JWXmFOA/P1/F5dspqGNhjHEB\nrnBqXEffZRERUQ3A8H4JN++nY/2eGKRl5sPVQYF3/J1R28xY32UREVENwfB+AUIIHDgbj53HYyEg\nMKBHc/Tp2pTT5EREpFMM73LKyi3EN79cxcVbybAyN8Z7/VzQuqm1vssiIqIaiOFdDrEPM7B+dzRS\nnuSjTVNrjO3nAitzTpMTEZF+MLzLIITAofP38d9jt6BSCQR0c4D/a81gYMBpciIi0h+Gdyly8grx\nzb5r+OtGEmqbGWFsPxc4N1PouywiIiKGd0nuPHqCdbujkZyRh9ZN6mBsPxfUsTDRd1lEREQAGN7F\nCCFw9K8HCD96E0qlQN/XmiGgWzMYGhjouzQiIiI1hvf/y8krwrf7r+H8tURY1DLCWH9nuDavq++y\niIjobw4d2o/Fi0OxZ88B1Knz9MRYS5bMx+uv94K7e3f1doMH+2Pz5nCYmZnh6tUYrF27CgUFBSgs\nLES3bj0watS7L3zRqJs3b2DFimWQyQBHx5aYNm1WsfURERH4/PMv0KiRHQDg1Vf/hZEjx+Dy5Sis\nXv05jIyM4Ob2Ct57b0IFuwBwSAkg7nEmFn57DuevJaKlvRUWjO7M4CYiqoIOHToAOzt7HD9+uFzb\nZ2dnYcGCuZg8eTo2bNiEjRu/xc2bN/DTT7tf+NirVq3ARx9Nxbp13yArKwuRkaef28bT0xv//vdG\n/PvfGzFy5BgAwGefLcOsWfOwZs1XSE1NweXLUS987H/S6sg7LCwMUVFRkMlkCAkJgZubm3rd77//\njpUrV8LQ0BA9evTAhAkVfyXyooQQOH7xIbYdvokipQp+XZpiQA8HTpMTEVVBT55k4OrVGMyaNQ8/\n/LAZ/fsP1nifQ4f2o0ePnmjevAUAQC6XY+7cBTAxMS223XfffY1z5/4odtvUqTPh4NAcAFBYWIhH\njx6iTRsXAIC7e3ecP38WXbu6a6whJSVZvZ/Onbvi7NkzaNu2neYHXAathffZs2cRFxeH8PBwxMbG\nIiQkBOHh4er1ixcvxtdff4369etj+PDh8PHxQYsWLbRVznNy8gqx8acr+ONKAsxN5fhgoCvcHOvp\n7PhERFK14+gtnLuWWKn7fLW1Ld7yLDsDjh49jNde64Z//asrli9fjKSkRNjY2JZ5n7i4OHXgPmNm\nZv7cdiNHjlGPlEuSkZEOS0tL9bK1tQIpKcnPbXfx4l+YMmUilMoiTJjwEZycWqNhw0a4ePEvtGvX\nHufP/wFDQ8Myay4PrYV3ZGQkvLy8AACOjo7IyMhAVlYWLCwsEB8fDysrKzRs2BAA0LNnT0RGRuos\nvDOyCzD36z/wICkbjna1Ma6fK+pamWq+IxER6c3hwwcwcuQYGBoawsOjF44cOYigoOGlbi+TySCT\nASqVstJrEUI8d1u7du0wenQtvPZaN0RHX8LixaHYvDkcs2bNxZdfroCBgQFatnRCdnZ2hY+vtfBO\nTk6Gi8v/Xu0oFAokJSXBwsICSUlJUCgUxdbFx8eXuT9razPI5RV/tQIAiZkpSEjNRf+ejhjZxxly\nQ06TV4SNjaXmjahM7GHFsYcVV94eTghsr+VKnvf48WNcuRKN9etXYcMGGfLy8mBpaYmJE8ejYUNb\nyGSFxepXqZRo0sQWLi6tcfPmzWLrUlNTkZubCzs7O/Vt69atw++//17smKGhoepBZZ06psjKylTv\nJz8/E02a2BXbr42NJRwdHQEAHh7umDMnAwqFGWxs2uOHH74HAGzfvh1Pnjyp8PNVZ582L+lVyotI\nS8uppEoAW0tj/HdpH6SlZiMtteKvgGoyGxtLJCVl6rsMSWMPK449rLiq3sPw8F0YMOBNTJw4GcDT\nTAkKGoCLF6/C2fkV7NoVjs6de0Aul+PQof1wdW2HpKRMdO3qgfXrN6JHDy84O7uisLAQCxbMRpcu\nr6Fv3/7q/Q8ePByDBz8/iv97T+ztm+Dw4d/Qrt0r+PnnfRg0KLDY+t27t8PcvA68vX1x+/Yt1K5t\nhdTUHISFLcBbbw2Fg0Nz7NwZgenTZ5W716WFvNbC29bWFsnJ/3s/IDExETY2NiWuS0hIgK1t2e9b\nVDaOtomIpOPw4QOYM2eBelkmk6F3777qqfS7d29jwoR3YWRkhLp162Ly5I8BAGZmZlix4kt88kkY\n8vPzYWhoCG9v32LBXV4ffjgVn34aBiFUcHZ2xauv/gsAMHPmFCxbthL+/v6YNGkK9uyJgFJZhJkz\n5wIA+vYNQFjYfACAl5ev+sNzFSETFR0Sl+Kvv/7C6tWrsWnTJsTExGDx4sXYtm2ben2fPn2wYcMG\nNGjQAIGBgfjss8/g4OBQ6v4q+xVhVX+VKRXsY8WxhxXHHlYce1hx2uihzkfeHTp0gIuLC4KCgiCT\nyRAaGoqIiAhYWlrC29sb8+fPx9SpUwEAfn5+ZQY3ERER/Y/WRt6VjSPvqol9rDj2sOLYw4pjDytO\nlyNvvvFLREQkMQxvIiIiiWF4ExERSQzDm4iISGIY3kRERBLD8CYiIpIYhjcREZHEMLyJiIgkRjIn\naSEiIqKnOPImIiKSGIY3ERGRxDC8iYiIJIbhTUREJDEMbyIiIolheBMREUlMjQjvsLAwBAYGIigo\nCJcuXSq27vfff8fgwYMRGBiINWvW6KnCqq+sHp45cwZvvfUWgoKCMGvWLKhUKj1VWbWV1cNnVqxY\ngeDgYB1XJh1l9fDRo0cYMmQIBg8ejHnz5umpQmkoq49bt25FYGAghgwZgiVLluipwqrvxo0b8PLy\nwvfff//cOp3kiqjm/vjjDzF27FghhBC3bt0Sb731VrH1vXv3Fg8fPhRKpVIMGTJE3Lx5Ux9lVmma\neujt7S0ePXokhBBi4sSJ4vjx4zqvsarT1EMhhLh586YIDAwUw4cP13V5kqCphx9++KE4ePCgEEKI\n+fPniwcPHui8Rikoq4+ZmZnCw8NDFBYWCiGEGDVqlLhw4YJe6qzKsrOzxfDhw8WcOXPEli1bnluv\ni1yp9iPvyMhIeHl5AQAcHR2RkZGBrKwsAEB8fDysrKzQsGFDGBgYoGfPnoiMjNRnuVVSWT0EgIiI\nCDRo0AAAoFAokJaWppc6qzJNPQSAZcuWYfLkyfooTxLK6qFKpcKff/4JT09PAEBoaCgaNWqkt1qr\nsrL6aGRkBCMjI+Tk5KCoqAi5ubmwsrLSZ7lVkrGxMb766ivY2to+t05XuVLtwzs5ORnW1tbqZYVC\ngaSkJABAUlISFApFievof8rqIQBYWFgAABITE3H69Gn07NlT5zVWdZp6GBERgc6dO8POzk4f5UlC\nWT1MTU2Fubk5li5diiFDhmDFihX6KrPKK6uPJiYmmDBhAry8vODh4YF27drBwcFBX6VWWXK5HKam\npiWu01WuVPvw/ifBs8FWWEk9TElJwbhx4xAaGlrsDwOV7O89TE9PR0REBEaNGqXHiqTn7z0UQiAh\nIQEjRozA999/jytXruD48eP6K05C/t7HrKwsbNiwAfv378eRI0cQFRWFa9eu6bE6Kk21D29bW1sk\nJyerlxMTE2FjY1PiuoSEhBKnQWq6snoIPP0P/+6772LSpEno1q2bPkqs8srq4ZkzZ5Camophw4bh\ngw8+QExMDMLCwvRVapVVVg+tra3RqFEjNGnSBIaGhujatStu3rypr1KrtLL6GBsbi8aNG0OhUMDY\n2BidOnVCdHS0vkqVJF3lSrUPb3d3dxw4cAAAEBMTA1tbW/U0r729PbKysnD//n0UFRXh2LFjcHd3\n12e5VVJZPQSevlc7cuRI9OjRQ18lVnll9dDX1xf79u3Djh078O9//xsuLi4ICQnRZ7lVUlk9lMvl\naNy4Me7evatez+nekpXVRzs7O8TGxiIvLw8AEB0djWbNmumrVEnSVa7UiKuKffbZZzh//jxkMhlC\nQ0Nx5coVWFpawtvbG+fOncNnn30GAHjjjTcwZswYPVdbNZXWw27duuHVV19F+/bt1dv27dsXgYGB\neqy2airrefjM/fv3MWvWLGzZskWPlVZdZfUwLi4OM2fOhBACTk5OmD9/PgwMqv345KWU1cft27cj\nIiIChoaGaN++PT7++GN9l1vlREdHY/ny5Xjw4AHkcjnq168PT09P2Nvb6yxXakR4ExERVSd8WUpE\nRCQxDG8iIiKJYXgTERFJDMObiIhIYhjeREREEsPwJtKTjz76CAMGDMDjx49L3aZVq1YoKiqq0HGC\ng4PRr18/BAcHY/jw4RgyZAjOnTv3UvuaPHkyEhISAAB79uwBAFy9ehWLFi2qUI0v4tlxiWoyflWM\nSE/atGmDCxculHqOZOBpeMfExEAul7/0cYKDgzF+/Hi89tprAJ5eynDUqFE4deoUZDLZS+0zISEB\nkyZNwrZt2166rpehVCrh5+enPskIUU3FkTeRHsyePRsqlQrvvPMOHj58iC+//BJBQUEICgrCpEmT\nUFhYWGz7M2fO4M0330RwcDACAwPV12A+c+aMekT99ttvIz4+XuOxnZycUFRUhLS0NOTk5GDq1KkY\nNmwYAgMD8cMPPwB4GvBBQUEIDg7Gm2++qT5PuKenJ+Li4jB16lTcuHEDH3/8Mf744w8MGTIEx48f\nx+jRo9XHOX/+PN58800AwJYtWzB06FAEBgZi1qxZ6jN4PXP//n34+/tj6tSpWL9+PXJycjBx4kQE\nBwdj4MCB2LhxIwAgJCQEDx48UB9n3759GDp0KIYMGYIJEybwinZUc1T6RUaJqFycnJxEYWGhKCws\nFBs2bBBKpVIIIcTo0aPF0aNHi20zbtw48csvvwghhIiNjRWHDx8WOTk54o033hBpaWlCCCEOHTok\nPvjgg+eOM3z4cHH69Gn18u+//y58fX2FEEKsX79ezJ8/XwghRG5urvDw8BD37t0TixYtEhs2bBBC\nCJGcnCx+/PFHIYQQHh4e4u7du+LMmTMiKChICCHUPxcWFgp3d3d1PQsXLhRbtmwRUVFRIjg4WKhU\nKiGEEEuWLBGbN28uVmN8fLxo06aNiI2NFUIIce/ePfUx8/PzRYcOHURmZqaIj48X3bt3F0II8fDh\nQ+Hv7y/y8/OFEEJ8++23YunSpS/zqyCSnJefiyOiSiGXy2FgYIChQ4dCLpfj9u3bz40g/f39sXLl\nSly6dAm9evVCr169cOnSJSQlJWHixIkAnk4plzYNvmzZMlhZWUEIAYVCgbVr1wIAoqKiMHDgQACA\nqakpXF1dERMTAx8fH8ycORMPHz6Eh4cHAgICyvU4vL29cfjwYQwcOBBHjhxBREQEdu3ahXv37mHE\niBEAgJycnBLfBrCyskLz5s0BAHXr1sWff/6J7du3w8jICPn5+UhPTy+2/YULF5CUlKQ+9WRBQQHs\n7e011klUHTC8ifTszz//xK5du7Br1y6YmZnhww8/fG4bPz8/dOvWDadOncKaNWvg5uYGPz8/NGrU\nqFznQZ85c6b6Pe+/+2fYCyEgk8nw6quv4ueff0ZkZCQiIiKwd+/ecl0ju2/fvli/fj3s7e3RunVr\n9dWpPD09MW/evDLva2RkpP75u+++Q0FBAbZt2waZTIZ//etfz21vbGwMNzc3bNiwQWNdRNUN3/Mm\n0rOUlBTY2dnBzMwMDx48wMWLF1FQUFBsm1WrVqk/rDV79mxcuHABzZo1Q1paGm7cuAEAOHfuHMLD\nw1/o2O3atcNvv/0G4OmIOCYmBi4uLtiyZQseP34MT09PLFmyBFFRUcXuZ2BgUOKn4Dt06ID4+Hjs\n3bsX/fr1U9928uRJZGdnAwC2bt2KCxcuaOyJo6MjZDIZjhw5gry8PBQUFBQ7btu2bdWzDwDw66+/\n4vDhwy/0+ImkiiNvIj1zd3fHN998gyFDhqBly5aYOHEi1qxZU2y02bRpU4wePRq1a9eGSqXCxIkT\nYWpqik8//RSzZ8+GiYkJAGDhwoUvdOzg4GDMnTsXw4YNQ0FBAd5//33Y29ujefPmmDp1KszNzaFS\nqTB16tRi92vRogVSUlIwatQojBs3Tn27TCaDj48Ptm/fjtDQUABPQ3bYsGEIDg6GiYkJbG1t1VP1\npRk0aBCmTJmCU6dOoVevXvD398e0adOwY8cO1KtXDwMHDsT333+P2bNn47333kOtWrVgamqK5cuX\nv9DjJ5IqflWMiIhIYjhtTkREJDEMbyIiIolheBMREUkMw5uIiEhiGN5EREQSw/AmIiKSGIY3ERGR\nxDC8iYiIJOb/AG8y7yjdD2SZAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAATgAAAE9CAYAAACfsn7ZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\njEXgABiLwAEwFoEDYCwCB8BY/wMZx7U1YxacrwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#compute the Area under the ROC Curve\n",
    "PredictProbLGNB=model.predict_proba(xtest)[:,1]\n",
    "ROCCURVE(PredictProbLGNB,ytest,Ypred_GNB)\n",
    "ConfusionmatrixPlot(ytest, Ypred_GNB)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "II-7 RANDOM FOREST CLASSIFIER"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Acuracy of the model is: 0.625\n",
      "The error of prediction is: 0.36607142857142855\n",
      "[[33  5]\n",
      " [16  2]]\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "         -1       0.67      0.87      0.76        38\n",
      "          1       0.29      0.11      0.16        18\n",
      "\n",
      "avg / total       0.55      0.62      0.57        56\n",
      "\n"
     ]
    }
   ],
   "source": [
    "model = RandomForestClassifier()\n",
    "model.fit(xtrain, ytrain)\n",
    "Ypred_RFC=model.predict(xtest)\n",
    "accuracy = accuracy_score(ytest.flatten(),Ypred_RFC)\n",
    "\n",
    "print(\"The Acuracy of the model is:\",accuracy)\n",
    "print(\"The error of prediction is:\", error(ytest,Ypred_RFC))\n",
    "print(confusion_matrix(ytest,Ypred_RFC))\n",
    "print(classification_report(ytest,Ypred_RFC))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "the Area under the ROC curve:0.707602\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe8AAAFnCAYAAACPasF4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nofAmhBBCJIbCmxBCCJEYCm9CCCFEYn4AiR4+pJ7YFREAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAATgAAAE9CAYAAACfsn7ZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#compute the Area under the ROC Curve\n",
    "PredictProbLRFC=model.predict_proba(xtest)[:,1]\n",
    "ROCCURVE(PredictProbLRFC,ytest,Ypred_RFC)\n",
    "ConfusionmatrixPlot(ytest, Ypred_RFC)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "II-8 KERNEL LOGISTIC REGRESSION"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Acuracy of the model is: 0.5714285714285714\n",
      "The error of prediction is: 0.42346938775510207\n",
      "[[27 11]\n",
      " [13  5]]\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "         -1       0.68      0.71      0.69        38\n",
      "          1       0.31      0.28      0.29        18\n",
      "\n",
      "avg / total       0.56      0.57      0.56        56\n",
      "\n"
     ]
    }
   ],
   "source": [
    "model = KernelRidge()\n",
    "model.fit(xtrain, ytrain)\n",
    "Prob=model.predict(xtest)\n",
    "N,D= Prob.shape\n",
    "Ypred_KRR= np.ones(N)\n",
    "#print(Prob.shape, Ypred_KRR.shape)\n",
    "for i in range(N):\n",
    "    if(Prob[i,0]<0):\n",
    "       \n",
    "        Ypred_KRR[i,]=-1\n",
    "    else:\n",
    "        Ypred_KRR[i,]=1\n",
    "accuracy = accuracy_score(ytest.flatten(),Ypred_KRR)\n",
    "\n",
    "print(\"The Acuracy of the model is:\",accuracy)\n",
    "print(\"The error of prediction is:\", error(ytest,Ypred_KRR))\n",
    "print(confusion_matrix(ytest,Ypred_KRR))\n",
    "print(classification_report(ytest,Ypred_KRR))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "the Area under the ROC curve:0.540936\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe8AAAFnCAYAAACPasF4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nIbwBALAYwhsAAIshvAEAsBjCGwAAi/l//pd2re5bm1oAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAATgAAAE9CAYAAACfsn7ZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ROCCURVE(Prob,ytest,Ypred_KRR)\n",
    "ConfusionmatrixPlot(ytest,Ypred_KRR)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# II-9 LINEAR SVM\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Acuracy of the model is: 0.5892857142857143\n",
      "The error of prediction is: 0.41709183673469385\n",
      "[[28 10]\n",
      " [13  5]]\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "         -1       0.68      0.74      0.71        38\n",
      "          1       0.33      0.28      0.30        18\n",
      "\n",
      "avg / total       0.57      0.59      0.58        56\n",
      "\n"
     ]
    }
   ],
   "source": [
    "model = LinearSVC(penalty='l2',loss='hinge',C=0.1)\n",
    "model.fit(xtrain, ytrain)\n",
    "Ypred_LSVC=model.predict(xtest)\n",
    "accuracy = accuracy_score(ytest.flatten(),Ypred_LSVC)\n",
    "\n",
    "print(\"The Acuracy of the model is:\",accuracy)\n",
    "print(\"The error of prediction is:\", error(ytest,Ypred_LSVC))\n",
    "print(confusion_matrix(ytest,Ypred_LSVC))\n",
    "print(classification_report(ytest,Ypred_LSVC))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAATgAAAE9CAYAAACfsn7ZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#compute the Area under the ROC Curve\n",
    "#PredictProbLSVM=model.predict_proba(xtest)[:,1]\n",
    "#ROCCURVE(PredictProbLSVM,ytest,Ypred_LSVC)\n",
    "ConfusionmatrixPlot(ytest, Ypred_LSVC)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "II-10 NON LINEAR SVM WITH RBF KERNEL"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Acuracy of the model is: 0.6785714285714286\n",
      "The error of prediction is: 0.32142857142857145\n",
      "[[38  0]\n",
      " [18  0]]\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "         -1       0.68      1.00      0.81        38\n",
      "          1       0.00      0.00      0.00        18\n",
      "\n",
      "avg / total       0.46      0.68      0.55        56\n",
      "\n"
     ]
    }
   ],
   "source": [
    "model = SVC(kernel='rbf')\n",
    "model.fit(xtrain, ytrain)\n",
    "Ypred_SVCRBF=model.predict(xtest)\n",
    "accuracy = accuracy_score(ytest.flatten(),Ypred_SVCRBF)\n",
    "\n",
    "print(\"The Acuracy of the model is:\",accuracy)\n",
    "print(\"The error of prediction is:\", error(ytest,Ypred_SVCRBF))\n",
    "print(confusion_matrix(ytest,Ypred_SVCRBF))\n",
    "print(classification_report(ytest,Ypred_SVCRBF))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAATgAAAE9CAYAAACfsn7ZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\noqWhAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#PredictProbRBF=model.predict_proba(xtest)[:,1]\n",
    "#ROCCURVE(PredictProbRBF,ytest,Ypred_SVCRBF)\n",
    "ConfusionmatrixPlot(ytest,Ypred_SVCRBF)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "II-11 NON LINEAR SVM WITH LINEAR KERNEL"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Acuracy of the model is: 0.5892857142857143\n",
      "The error of prediction is: 0.41709183673469385\n",
      "[[28 10]\n",
      " [13  5]]\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "         -1       0.68      0.74      0.71        38\n",
      "          1       0.33      0.28      0.30        18\n",
      "\n",
      "avg / total       0.57      0.59      0.58        56\n",
      "\n"
     ]
    }
   ],
   "source": [
    "model = SVC(kernel='linear')\n",
    "model.fit(xtrain, ytrain)\n",
    "Ypred_SVCLIN=model.predict(xtest)\n",
    "accuracy = accuracy_score(ytest.flatten(),Ypred_SVCLIN)\n",
    "\n",
    "print(\"The Acuracy of the model is:\",accuracy)\n",
    "print(\"The error of prediction is:\", error(ytest,Ypred_SVCLIN))\n",
    "print(confusion_matrix(ytest,Ypred_SVCLIN))\n",
    "print(classification_report(ytest,Ypred_SVCLIN))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAATgAAAE9CAYAAACfsn7ZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\noqWhAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#PredictProbLSSM=model.predict_proba(xtest)[:,1]\n",
    "#ROCCURVE(PredictProbLSSM,ytest,Ypred_SVCLIN)\n",
    "ConfusionmatrixPlot(ytest,Ypred_SVCRBF)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "II-12 NON LINEAR SVM WITH POLYNOMIAL KERNEL"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Acuracy of the model is: 0.6785714285714286\n",
      "The error of prediction is: 0.32142857142857145\n",
      "[[38  0]\n",
      " [18  0]]\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "         -1       0.68      1.00      0.81        38\n",
      "          1       0.00      0.00      0.00        18\n",
      "\n",
      "avg / total       0.46      0.68      0.55        56\n",
      "\n"
     ]
    }
   ],
   "source": [
    "model = SVC(kernel='poly')\n",
    "model.fit(xtrain, ytrain)\n",
    "Ypred_SVCPOLY=model.predict(xtest)\n",
    "accuracy = accuracy_score(ytest.flatten(),Ypred_SVCPOLY)\n",
    "\n",
    "print(\"The Acuracy of the model is:\",accuracy)\n",
    "print(\"The error of prediction is:\", error(ytest,Ypred_SVCPOLY))\n",
    "print(confusion_matrix(ytest,Ypred_SVCPOLY))\n",
    "print(classification_report(ytest,Ypred_SVCPOLY))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAATgAAAE9CAYAAACfsn7ZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\noqWhAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#PredictProbLPOL=model.predict_proba(xtest)[:,1]\n",
    "#ROCCURVE(PredictProbLPOL,ytest,Ypred_SVCPOLY)\n",
    "ConfusionmatrixPlot(ytest,Ypred_SVCPOLY)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "II-13 NON LINEAR SVM WITH SIGMOID KERNEL"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Acuracy of the model is: 0.6785714285714286\n",
      "The error of prediction is: 0.32142857142857145\n",
      "[[38  0]\n",
      " [18  0]]\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "         -1       0.68      1.00      0.81        38\n",
      "          1       0.00      0.00      0.00        18\n",
      "\n",
      "avg / total       0.46      0.68      0.55        56\n",
      "\n"
     ]
    }
   ],
   "source": [
    "model = SVC(kernel='sigmoid')\n",
    "model.fit(xtrain, ytrain)\n",
    "Ypred_SVCSIG=model.predict(xtest)\n",
    "accuracy = accuracy_score(ytest.flatten(),Ypred_SVCSIG)\n",
    "\n",
    "print(\"The Acuracy of the model is:\",accuracy)\n",
    "print(\"The error of prediction is:\", error(ytest,Ypred_SVCSIG))\n",
    "print(confusion_matrix(ytest,Ypred_SVCSIG))\n",
    "print(classification_report(ytest,Ypred_SVCSIG))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAATgAAAE9CAYAAACfsn7ZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\noqWhAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#PredictProbLSIG=model.predict_proba(xtest)[:,1]\n",
    "#ROCCURVE(PredictProbLSIG,ytest,Ypred_SVCSIG)\n",
    "ConfusionmatrixPlot(ytest,Ypred_SVCSIG)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# III- CROSS VALIDATION (MODEL SELECTION)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LR: 0.609357 (0.090805)\n",
      "GPC: 0.593275 (0.089699)\n",
      "DTC: 0.623977 (0.142198)\n",
      "KNN: 0.621053 (0.152520)\n",
      "LDA: 0.636842 (0.124445)\n",
      "GNB: 0.565205 (0.073077)\n",
      "RFC: 0.620760 (0.132885)\n",
      "LSVM: 0.559942 (0.091332)\n",
      "RBFK: 0.642982 (0.158858)\n",
      "PLOKYK: 0.642982 (0.158858)\n",
      "SIGK: 0.642982 (0.158858)\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA2gAAAO6CAYAAADq4HQZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
      "text/plain": [
       "<Figure size 1080x1080 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# prepare configuration for cross validation test harness\n",
    "seed = 100\n",
    "# prepare models\n",
    "models = []\n",
    "models.append(('LR', LogisticRegression()))\n",
    "models.append(('GPC', GaussianProcessClassifier()))\n",
    "models.append(('DTC', DecisionTreeClassifier()))\n",
    "models.append(('KNN', KNeighborsClassifier()))\n",
    "models.append(('LDA', LinearDiscriminantAnalysis()))\n",
    "models.append(('GNB', GaussianNB()))\n",
    "models.append(('RFC', RandomForestClassifier(n_estimators=14)))\n",
    "#models.append(('KRR', KernelRidge()))\n",
    "models.append(('LSVM',LinearSVC()))\n",
    "models.append(('RBFK',SVC(kernel='rbf')))\n",
    "models.append(('PLOKYK',SVC(kernel='poly')))\n",
    "models.append(('SIGK',SVC(kernel='sigmoid')))\n",
    "\n",
    "# evaluate each model in turn\n",
    "results = []\n",
    "names = []\n",
    "scoring = 'accuracy'\n",
    "for name, model in models:\n",
    "    kfold = model_selection.KFold(n_splits=10, random_state=seed)\n",
    "    cv_results = model_selection.cross_val_score(model, X_train, Y_train, cv=kfold, scoring=scoring)\n",
    "    results.append(cv_results)\n",
    "    names.append(name)\n",
    "    msg = \"%s: %f (%f)\" % (name, cv_results.mean(), cv_results.std())\n",
    "    print(msg)\n",
    "    \n",
    "# boxplot algorithm comparison\n",
    "fig = plt.figure(figsize=[15,15])\n",
    "fig.suptitle('Algorithm Comparison')\n",
    "ax = fig.add_subplot(111)\n",
    "plt.boxplot(results)\n",
    "ax.set_xticklabels(names)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Acuracy of the model is: 0.5714285714285714\n",
      "The error of prediction is: 0.42346938775510207\n",
      "[[27 11]\n",
      " [13  5]]\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "         -1       0.68      0.71      0.69        38\n",
      "          1       0.31      0.28      0.29        18\n",
      "\n",
      "avg / total       0.56      0.57      0.56        56\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Cross validation for kernel ridge regression\n",
    "alpha= 10**np.linspace(-3,10,100)\n",
    "model = RidgeCV(alphas=alpha).fit(xtrain, ytrain)\n",
    "Prob=model.predict(xtest)\n",
    "N,D= Prob.shape\n",
    "Ypred_CVKRR= np.ones(N)\n",
    "for i in range(N):\n",
    "    if(Prob[i,0]<0):\n",
    "       \n",
    "        Ypred_CVKRR[i,]=-1\n",
    "    else:\n",
    "        Ypred_CVKRR[i,]=1\n",
    "accuracy = accuracy_score(ytest.flatten(),Ypred_CVKRR)\n",
    "\n",
    "print(\"The Acuracy of the model is:\",accuracy)\n",
    "print(\"The error of prediction is:\", error(ytest,Ypred_CVKRR))\n",
    "print(confusion_matrix(ytest,Ypred_CVKRR))\n",
    "print(classification_report(ytest,Ypred_CVKRR))\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "the Area under the ROC curve:0.542398\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe8AAAFnCAYAAACPasF4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAATgAAAE9CAYAAACfsn7ZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAFvlJREFUeJzt3XtUVXXex/HPgSOJkiAK+ISmpXhB\nqwcJ1Hy6LGV5KWdWhRfUsYbHyVLHJhsy0Xki81KaY5NkalZWoIEJOt7QzMxycqw0Qm0MzQmVvKEC\n3pDbef5weSaW0rHgsA8/36+1Wsu9OezzlVrvfvvsczY2h8PhEAAYyMvqAQDAXQgcAGMROADGInAA\njEXgABiLwAEwFoHzMDNmzNCQIUMUFxennJwcq8eBB8rNzVVMTIxSU1OtHsXj2a0eAP/xxRdfKC8v\nT+np6fr+++81adIkpaenWz0WPMj58+c1depU9ejRw+pR6gVWcB5k27ZtiomJkSS1bdtWRUVFOnv2\nrMVTwZP4+Pho0aJFCg4OtnqUeoHAeZCCggI1bdrUuR0YGKgTJ05YOBE8jd1uV8OGDa0eo94gcB6M\nT9EBNUPgPEhwcLAKCgqc28ePH1dQUJCFEwH1G4HzID179tSGDRskSXv27FFwcLD8/Pwsngqov2zc\nTcSzzJ49W1999ZVsNpuSkpLUsWNHq0eCB9m9e7dmzpyp/Px82e12hYSEKDk5WQEBAVaP5pEIHABj\ncYoKwFgEDoCxCBwAYxE4AMYicACMReAAGIvAATAWgQNgLI+5H9zWJZlWj+BRIh7ora/XbrJ6DI/R\nJSrc6hE8yo2tb9GZvH9bPYbHCGh/9U/8sILzUI0D/K0eAR7M+4YbrB6hXiBwAIxF4AAYi8ABMBaB\nA2AsAgfAWAQOgLEIHABjETgAxiJwAIxF4AAYi8ABMBaBA2AsAgfAWAQOgLEIHABjETgAxiJwAIxF\n4AAYi8ABMBaBA2AsAgfAWAQOgLEIHABjETgAxiJwAIxF4AAYi8ABMBaBA2AsAgfAWAQOgLEIHABj\nETgAxiJwAIxF4AAYi8ABMBaBA2AsAgfAWAQOgLEIHABjETgAxiJwAIxF4AAYi8ABMBaBA2AsAgfA\nWAQOgLEIHABjETgAxiJwAIxF4AAYi8ABMBaBA2AsAgfAWAQOgLEIHABjETgAxiJwAIxF4AAYi8AB\nMBaBA2Asu9UDQMr+7lut/OQjlVeUq7FvI4144EHNmjVLa1eucj6mtKxUNzZurOceG2fhpLBCeXm5\n5r37npau/LtWLX5LIc2bS5JOFRYq6a9z9OOxY8p4Y6HFU3omAmex08VFeuvvHygx/gndFBSij7/c\nppQ1K7V204e6K7Sd83Ep61bqpubBFk4KqyRMm6HwsHZV9hUWFmp04mT1iOyqH48ds2gyz+fWU9QZ\nM2ZoyJAhiouLU05Ojjufqt7y9vbWqIfjdFNQiCQp7OY2yj9R9T/Yw8ePKjfv37rvzm5WjAiLjYwb\nrFHDh1XZZ7PZNGtyou7uFm3RVPWD2wL3xRdfKC8vT+np6Zo+fbqmT5/urqeq15o09tNt7To4t3ft\nz9Wtoa2qPGbVlk3qd9c98vbyruvx4AFu69jxin3+/v5q3bKlBdPUL24L3LZt2xQTEyNJatu2rYqK\ninT27Fl3PZ0Rvj2wXxv/uVVxfR9w7jt2qkAH8g+pW5f/tnAyoH5yW+AKCgrUtGlT53ZgYKBOnDjh\nrqer93bu3aO3Vy3Xn4Y+6jxdlaQv9+xS147hsnuzegN+qTq7yOBwOH726xEP9FbjAP86msazfP75\n51rxZrKWpL2vtm3bOvf/z/CHlbwqXWPHjtX/3HOPhRPCU/jf2k4BLVpIkgLad5Tf6SJ5NfBRQPsr\nT2PhxsAFBweroKDAuX38+HEFBQVV+/iv125y1yge7WJZqSbPm6M/Dh6hI//8Rkf++Y2kS3HbuiRT\n3+7Zo1O79mnroQIXRzJbl6hwq0fwCEUH9qthcaEC2ndUYe5enT18UJVlpSrM3Wv1aJaqLvBuC1zP\nnj2VnJysuLg47dmzR8HBwfLz83PX09Vb2d99qzPnzmnRivQq+zP63qOzF86rtKxM/n43WjQdrHby\ndKFGJ05ybo+ZNFneXt4aPW6c5icnq+TiRZ0sLNTgJ8YoqFkzzZs+1cJpPY/N4ercsQZmz56tr776\nSjabTUlJSep4latBl21dkumuMeqlyys4XMIKrqrLKzhcUucrOElKSEhw5+EB4GfxWVQAxiJwAIxF\n4AAYi8ABMBaBA2AsAgfAWAQOgLEIHABjETgAxiJwAIxF4AAYi8ABMBaBA2AsAgfAWAQOgLEIHABj\nETgAxiJwAIxF4AAYi8ABMBaBA2AsAgfAWAQOgLEIHABjETgAxiJwAIxF4AAYi8ABMBaBA2AsAgfA\nWL84cJWVle6YAwBqncvAZWZmasmSJSovL9fQoUPVu3dvLV26tC5mA4AacRm49PR0DRo0SB999JHC\nwsK0adMmZWVl1cVsAFAjLgN3ww03yMfHR1u2bFH//v3l5cXLdgDqh2uq1ZQpU7Rz505FR0fr66+/\nVmlpqbvnAoAacxm42bNnq3Xr1lqwYIG8vb2Vn5+vKVOm1MVsAFAjLgMXHBys1q1b6x//+Ick6fbb\nb1eHDh3cPhgA1JTLwL388svKyMhQZmamJGn16tWaNm2a2wcDgJpyGbgvv/xSr732mho3bixJGjt2\nrPbs2eP2wQCgpq7pKqok2Ww2SVJFRYUqKircOxUA1AK7qwd07dpViYmJOn78uBYvXqwPP/xQ0dHR\ndTEbANSIy8CNHz9e69evV8OGDXX06FHFx8erT58+dTEbANSIy8AdOnRInTt3VufOnavsa9WqlVsH\nA4Cachm4Rx991Pn6W2lpqU6dOqWwsDCtXLnS7cMBQE24DNzHH39cZXvfvn1avny52wYCgNryiz9Y\nGhYWxttEANQLLldwr776apXto0ePqri42G0DAUBtcbmC8/b2rvJPhw4dtGjRorqYDQBqpNoV3OU7\n944ZM6bOhgGA2lRt4MLDw51XT3/K4XDIZrPpX//6l1sHA4CaqjZwe/furfabfvjhB3fMAgC1yuVF\nhoqKCm3dulWnT5+WdOm9cAsWLLji7SMA4GlcBu6ZZ55RUVGRvvvuO3Xt2lXffPONxo0bVxezAUCN\nuLyKevToUb311lu65ZZbNHfuXC1dulS7du2qi9kAoEau+Y2+5eXlunjxokJDQ7V//353zgQAtcLl\nKWr37t21aNEixcTE6OGHH1ZoaCi//BlAveAycE8++aQqKirk7e2tiIgInTx5Uj179qyL2QCgRqoN\nXL9+/TRo0CA9+OCDatasmaRLN78EgPqi2tfgZsyYoby8PA0YMEBjx47V5s2bOTUFUK9UG7iuXbvq\nhRde0JYtWzRgwAClpaWpd+/emjNnjvLy8upyRgD4VVxeRfXx8VH//v21cOFCLVu2TPn5+erXr19d\nzAYANeLyIoMk7d+/XytWrFBWVpY6deqk119/3d1zAUCNVRu4wsJCrVmzRpmZmSopKVFsbKyWLVum\n5s2b1+V8APCrVRu4Pn36KCYmRpMnT1ZkZKTbB+nSpbXbn6O+4WfyH7bmN1k9gsfhZ+JatYHbvHmz\n87fZA0B9VO1FBuIGoL77xb90BgDqC5e3LK+OlxdtBODZrumW5Q6HQ5Jks9m4ZTmAeoNblgMwFrcs\nB2AsblkOwFjcshyAsbhlOQBj/aJblj/00ENq2bIl94UDUC9wy3IAxnIZuOXLl1+xb926dRo4cKBb\nBgKA2uIycDt27HD+ubS0VDk5OeratSuBA+DxXAbuxRdfrLJ94cIFJSYmum0gAKgtv/gDpb6+vjp4\n8KA7ZgGAWuVyBTds2DDnZ1Il6dixY2rfvr1bhwKA2uAycE899ZTzzzabTX5+furUqZNbhwKA2uDy\nFDUzM1PR0dGKjo5WVFSUOnXqpJEjR9bFbABQI9Wu4FatWqW0tDTt27dPw4cPd+4vKytTQUFBnQwH\nADVRbeB++9vfqlu3bkpISKjy4XovLy+1a9euToYDgJr42VPUkJAQLViwQAUFBc7T1H379qlBgwZ1\nNR8A/GouX4ObOHFilVPSkpISTZgwwa1DAUBtcBm4wsJCPfLII87t+Ph4FRcXu3UoAKgNLgNXVlam\n77//3rm9e/dulZWVuXUoAKgNLt8Hl5iYqDFjxujMmTOqqKhQYGCgZs2aVRezAUCNuAzcHXfcoQ0b\nNuj06dOy2WwKCAjQjz/+WBezAUCNXPNnURs1aqRPP/1Ujz76qAYPHuzOmQCgVrhcwWVnZysjI0NZ\nWVmqrKzUCy+8oL59+9bFbABQI9Wu4BYtWqT7779f48ePV7NmzZSRkaGbb75ZAwYM4H1wAOqFaldw\nf/vb39SuXTs999xz6t69uyRVuasIAHi6agP3ySefaMWKFUpKSlJlZaUeeugh3h4CoF6p9hQ1KChI\no0aN0oYNGzRjxgwdPHhQ+fn5euKJJ7Rly5a6nBEAfpVruooaFRWll156SZ999pnuu+8+zZs3z91z\nAUCN2RwOh8PqISSp8Jsdrh90HQm4I5KfyU/YWoVZPYJH8Q9soqJTfGTyMv/AJlfd/4t/JwMA1BcE\nDoCxCBwAYxE4AMYicACMReAAGIvAATAWgQNgLAIHwFgEDoCxCBwAYxE4AMYicACMReAAGIvAATAW\ngQNgLAIHwFgEDoCxCBwAYxE4AMYicACMReAAGIvAATAWgQNgLAIHwFgEDoCx7FYPAKm8vFzzlqZp\n6Zp1WjU/WSHNmkmS3li2XB99vk2VDoc6tGmjiaNG6sbGjS2eFlb58ciPih30sFqGtpSXt5cqKyoV\nHt5ZU5KmWD2axyJwHiDh5b8qvG3bKvvWrFmjL3J26b1ZL8rHbtekV+bqnRV/17jfDbNoSniC4KBg\nfZC+XP6BTVR0qtjqcTyeW09Rc3NzFRMTo9TUVHc+Tb03MvYhjRo8sMq+du3aacIf/lcNfXzk5eWl\nyM6dlPfjEYsmBOont63gzp8/r6lTp6pHjx7uegpj3Na+/RX7OnbsqMKL5yRJZ8+f16Zt23X/vXfX\n9WjwMOfOnVPCswk6dOigQoJDNP6pp3VLm1usHstjuW0F5+Pjo0WLFik4ONhdT3Fd+L9XX9P9o8ao\nZYsWuv8eAnc9a9Sokfr26aunn3pa69atU7fobkqYkKDy8nKrR/NYbguc3W5Xw4YN3XX468bUP/1R\nG99+Q7433KCk5NetHgcWCvAP0DMJE3TTf90kLy8vDRs6XKdOndTBQwetHs1jecxFhhvbh8vb19fq\nMSznH367Alq00LZt29S8eXOFhYVJkn7XJFDDhw9XwB2RFk8IqxQVFam4uFitWrWSJPkHNpHD4VDT\nZgHyD2xi8XSeyWMCdyb3W6tH8AhF3+ao4bF87dixQ198+olmT0iQT4MGyspcqbYtQ1X4zQ6rR7SE\nrVWY1SNY7p/bt+vFmS/qnbfeUZu2rfXu4ncVEhyiJo39r/srqtUF3mMCd706WVik0c+/4Nwe8/w0\neXt7KSUtXfl7v9XwhIlyyKGQZs00+fHHLJwUVuverbsGPjxQf3j8D7LbvdUssJleenGmvL29rR7N\nY9kcDofDHQfevXu3Zs6cqfz8fNntdoWEhCg5OVkBAQFXffz1ujKpTsAdkfxMfoIVXFW8D66qOl/B\ndenSRSkpKe46PAC4xGdRARiLwAEwFoEDYCwCB8BYBA6AsQgcAGMROADGInAAjEXgABiLwAEwFoED\nYCwCB8BYBA6AsQgcAGMROADGInAAjEXgABiLwAEwFoEDYCwCB8BYBA6AsQgcAGMROADGInAAjEXg\nABiLwAEwFoEDYCwCB8BYBA6AsQgcAGMROADGInAAjEXgABiLwAEwFoEDYCwCB8BYBA6AsQgcAGMR\nOADGInAAjEXgABiLwAEwFoEDYCwCB8BYBA6AsQgcAGMROADGInAAjEXgABiLwAEwFoEDYCwCB8BY\nBA6AsQgcAGMROADGInAAjEXgABiLwAEwFoEDYCwCB8BYBA6AsQgcAGMROADGsjkcDofVQwCAO7CC\nA2AsAgfAWAQOgLEI3HXu8OHD6tKli0aMGKERI0YoLi5Of/7zn1VcXPyrj/nBBx9o4sSJkqTx48fr\n2LFj1T52586dOnTo0DUfu7y8XB06dKiy7+zZs4qKitKpU6eq7N+xY4f69u37s8fr0KGDysvLr/n5\nUb8QOCgwMFApKSlKSUlRWlqagoODNX/+/Fo59iuvvKKQkJBqv56ZmfmLAnc1fn5+iomJ0Zo1a6rs\nX7lypWJjY2t0bNRvdqsHgOeJiopSenq6JKlXr17q37+/Dh06pLlz52rdunVKTU2Vw+FQYGCgpk2b\npqZNm2rJkiV6//331aJFCwUHBzuP1atXLy1evFitWrXStGnTtHv3bklSfHy87Ha71q9fr5ycHCUm\nJqp169aaMmWKLly4oPPnz+vpp5/WXXfdpQMHDuiZZ56Rr6+vunXrdtWZY2NjNWPGDD3yyCOSpIsX\nL2rjxo1avXq1JOnVV1/Vtm3bJEktWrTQyy+/rAYNGji/Pzk5WeXl5Ro/fnyVuVu3bq05c+Zo586d\nKikpUVRUlCZMmKDjx48rISFBklRSUqIhQ4Zo4MCBtfmvAbWAFRyqqKio0MaNGxUZGenc16ZNG82d\nO1dHjhzRggUL9M477+j9999XdHS0Fi5cqDNnzmju3LlKSUnRm2++qdOnT19x3FWrVqmgoEDLli3T\nm2++qRUrVqhXr17q1KmTJk6cqB49euj5559XfHy83nvvPc2fP19/+ctfVF5ernnz5ik2NlapqalX\nnJ5eduedd+r8+fPKzc2VJG3atEkREREKCgpSeXm5fH19tXTpUqWlpenMmTPaunXrNf08srKydOzY\nMaWmpmr58uU6ePCgNm/erKysLN16661KSUlRamqqSkpKfsVPG+7GCg46deqURowYIUmqrKzUnXfe\nqd///vfOr0dEREiSvv76a504cUIjR46UJJWWlqply5bKy8tTaGiomjZtKknq1q2b9u7dW+U5cnJy\nnKuvJk2a6I033rhiju3bt+vcuXOaN2+eJMlut+vkyZPKzc3VqFGjJEndu3ev9u8RGxurFStW6Nln\nn9XKlSs1ZMgQ53G8vLw0bNgw2e12HThw4KoRvprt27crOzvb+fM5c+aMDh8+rLvvvltLly7VxIkT\nde+99zqfC56FwMH5Glx1Lp/K+fj46Pbbb9fChQurfH3Xrl2y2WzO7crKyiuOYbPZrrr/p3x8fJSc\nnKzAwMAq+x0Oh7y8Lp1sVFRUVPv9Dz74oAYNGqT4+Hh99913uvfeeyVdutiQkZGhjIwMNWrUSE8+\n+eRV5/up0tJS50yDBw92Rv2n1q5dqy+//FLr16/Xu+++q7S0tJ/9+6HucYqKa3bbbbcpJydHJ06c\nkHTp9O2jjz7SzTffrMOHD6u4uFgOh8P5WtdPRURE6LPPPpN06arnoEGDVFpaKpvNprKyMklSZGSk\nsrKyJF1aVU6fPl2S1LZtW2VnZ0vSVY99WVBQkMLDwzVz5kz95je/kd1+6f/fJ0+eVGhoqBo1aqT8\n/HxlZ2c7A3aZn5+fjh49Kknat2+f84psZGSkNm7c6LzS+tprr+mHH37Q6tWrtWvXLt11111KSkrS\nkSNHuBrrgVjB4ZqFhIRo8uTJevzxx+Xr66uGDRtq5syZ8vf31xNPPKHhw4crNDRUoaGhV7wm1b9/\nf+3cuVNxcXGqqKhQfHy8fHx81LNnTyUlJWnSpEmaPHmynnvuOa1du1alpaUaPXq0JGns2LF69tln\ntX79ekVERDjDdTUDBw7U6NGjtX79eue+nj176u2339bQoUMVFhamcePGad68eVUuWPTr108ZGRka\nNmyYunTponbt2kmS+vTpo+zsbMXFxcnb21vh4eFq1aqVLly4oKSkJPn4+MjhcOixxx772blgDT6L\nCsBYnKICMBaBA2AsAgfAWAQOgLEIHABjETgAxiJwAIxF4AAY6/8Bb7BoNSBrPdgAAAAASUVORK5C\nYII=\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ROCCURVE(Prob,ytest,Ypred_CVKRR)\n",
    "ConfusionmatrixPlot(ytest,Ypred_CVKRR)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.5.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}