GPC-RBF.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import warnings\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from imblearn.under_sampling import NearMiss\n",
    "from sklearn.preprocessing import LabelEncoder, StandardScaler\n",
    "from sklearn.model_selection import GridSearchCV, KFold\n",
    "from sklearn.metrics import classification_report, ConfusionMatrixDisplay, multilabel_confusion_matrix\n",
    "from sklearn.gaussian_process import GaussianProcessClassifier\n",
    "from sklearn.gaussian_process.kernels import RBF, DotProduct\n",
    "\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "data = pd.read_csv(\"/Users/Cristian/Documents/EPN/Maestría en Computación/OneDrive - Escuela Politécnica Nacional/2022-B/Tesis/Dataset/20221215_151443_clustering.csv\")\n",
    "#data = pd.read_csv(\"F:/EPN/Maestría en Computación/OneDrive - Escuela Politécnica Nacional/2022-B/Tesis/Dataset/20221215_151443_clustering.csv\")\n",
    "\n",
    "df = data.loc[:, ['steering_angle', 'speed', 'rpm', 'acceleration', 'throttle_position', 'engine_temperature', 'system_voltage', 'barometric_pressure',\n",
    "    'distance_travelled', 'latitude', 'longitude', 'heart_rate', 'accidents_onsite', 'risk_level']]\n",
    "\n",
    "le = LabelEncoder()\n",
    "df['risk_level']= le.fit_transform(df['risk_level'])\n",
    "\n",
    "n = len(pd.unique(df['risk_level']))\n",
    "\n",
    "X = df.iloc[:,0:13]\n",
    "Y = df.iloc[:,-1:]\n",
    "\n",
    "undersample = NearMiss(version = 1, n_neighbors = 4)\n",
    "X_res, Y_res = undersample.fit_resample(X, Y)\n",
    "\n",
    "scaler = StandardScaler()\n",
    "X = scaler.fit_transform(X_res)\n",
    "Y = Y_res"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "Y = np.array(Y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# **Hyperparameter Tuning**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best parameters are {'kernel': 1**2 * RBF(length_scale=1), 'max_iter_predict': 20} \n",
      "Score: 0.7916666666666666%\n"
     ]
    }
   ],
   "source": [
    "model = GaussianProcessClassifier()\n",
    "\n",
    "param_grid = {'kernel': [1*RBF()], 'max_iter_predict':[10, 20, 50, 100]}\n",
    "grid_search = GridSearchCV(model, param_grid, scoring='accuracy', cv = 5)\n",
    "grid = grid_search.fit(X, Y.flatten())\n",
    "\n",
    "print(\n",
    "\t\"Best parameters are {} \\nScore: {}%\".format(\n",
    "\t\tgrid.best_params_, grid.best_score_)\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# **GPC-RBF Implementation**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "rbf = 1 ** 2 * RBF(length_scale=1)\n",
    "gpc_rbf_model = GaussianProcessClassifier(kernel = rbf, max_iter_predict = 20)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# **Cross Validation**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_folds = 5\n",
    "X = np.array(X)\n",
    "Y = np.array(Y)\n",
    "\n",
    "# Define per-fold score containers\n",
    "acc_per_fold = []\n",
    "specifity_per_fold = []\n",
    "sensivity_per_fold = []\n",
    "\n",
    "# Define the K-fold Cross Validator\n",
    "kfold = KFold(n_splits = num_folds, shuffle = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "------------------------------------------------------------------------\n",
      "Training for fold 1 ...\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "     Class 0       0.95      0.98      0.96        94\n",
      "     Class 1       0.83      0.86      0.84       101\n",
      "     Class 2       0.93      0.86      0.90        96\n",
      "     Class 3       0.88      0.88      0.88        93\n",
      "\n",
      "    accuracy                           0.90       384\n",
      "   macro avg       0.90      0.90      0.90       384\n",
      "weighted avg       0.90      0.90      0.90       384\n",
      "\n",
      "\n",
      "Score for fold 1: - Accuracy: 0.8958333333333334 - Specifity: 0.9651300899471033 - Sensivity: 0.8966033265775102\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "------------------------------------------------------------------------\n",
      "Training for fold 2 ...\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "     Class 0       0.99      0.97      0.98        97\n",
      "     Class 1       0.83      0.89      0.86        87\n",
      "     Class 2       0.89      0.91      0.90        95\n",
      "     Class 3       0.91      0.86      0.88       105\n",
      "\n",
      "    accuracy                           0.90       384\n",
      "   macro avg       0.90      0.90      0.90       384\n",
      "weighted avg       0.91      0.90      0.90       384\n",
      "\n",
      "\n",
      "Score for fold 2: - Accuracy: 0.9036458333333334 - Specifity: 0.9680808193293254 - Sensivity: 0.9041339128126039\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "------------------------------------------------------------------------\n",
      "Training for fold 3 ...\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "     Class 0       0.96      0.93      0.94        94\n",
      "     Class 1       0.86      0.87      0.87        94\n",
      "     Class 2       0.90      0.96      0.93        92\n",
      "     Class 3       0.92      0.88      0.90       104\n",
      "\n",
      "    accuracy                           0.91       384\n",
      "   macro avg       0.91      0.91      0.91       384\n",
      "weighted avg       0.91      0.91      0.91       384\n",
      "\n",
      "\n",
      "Score for fold 3: - Accuracy: 0.9088541666666666 - Specifity: 0.9696403266077334 - Sensivity: 0.9097523660428378\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "------------------------------------------------------------------------\n",
      "Training for fold 4 ...\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "     Class 0       0.96      0.97      0.96        94\n",
      "     Class 1       0.92      0.85      0.88       102\n",
      "     Class 2       0.91      0.87      0.89       101\n",
      "     Class 3       0.87      0.97      0.91        87\n",
      "\n",
      "    accuracy                           0.91       384\n",
      "   macro avg       0.91      0.91      0.91       384\n",
      "weighted avg       0.91      0.91      0.91       384\n",
      "\n",
      "\n",
      "Score for fold 4: - Accuracy: 0.9114583333333334 - Specifity: 0.9705662345782741 - Sensivity: 0.9144576632364372\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfIAAAHICAYAAAC1cGgbAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAABLaElEQVR4nO3deVxU5RoH8N+wDYuAAg6IIKLiFu6YkqZoibli3kxzuaiUGS6ZuVwzFUslLdHUK6nd1BZMWzQrNdECMzMB1xRNExEVBBQB2Wfm3D+MqRHUGWY5s/y+n8/53DtnznvOM2fIZ573fc85EkEQBBAREZFZshE7ACIiIqo7JnIiIiIzxkRORERkxpjIiYiIzBgTORERkRljIiciIjJjTORERERmjImciIjIjDGRExERmTEmcjKIwsJCTJ06FQEBAbCzs4NEIsGVK1cMekyJRIKwsDCDHsOShYWFQSKRiB2Gmhs3bmDs2LFo3LgxbGxs6hxfTEwMJBIJkpKSNG7DvycyF0zkFiItLQ1RUVEICgqCi4sLnJyc0Lx5c4wbNw6JiYlGj2f27Nn473//i44dO+KNN97AokWLUL9+faPHIZYrV65AIpFAIpGgcePGUCgUtW535swZ1XatW7fW6Zjjx483yg8mYxo/fjwSEhLQp08fLFiwAIsWLRI7JJ198cUXqu/8888/FzscsgB2YgdAulEqlZg1axZWrVoFOzs79O3bF0OHDoW9vT0uX76M77//Hp9++ineeustLFiwwGhx7dmzB61atcI333xjtGOmp6fD2dnZaMfThJ2dHW7cuIEffvgBAwcOrPH+//73P9jZ2UEul4sQnbqPP/4YpaWlYoehUllZiYMHDyI8PByffvqp2OHoRW5uLqKjo+Hi4oKSkhKxwyELwURu5t58802sWrUKHTt2xJdffonmzZurvV9WVoZ169bh1q1bRo3rxo0b6NWrl1GPqWtFawhPPPEETp06hY8++qhGIq+srMRnn32GgQMHYvfu3SJF+LcmTZqIHYKanJwcKJVK+Pj4iB2K3rzyyitwcXFBZGQkVq5cKXY4ZCHYtW7GLl26hBUrVsDT0xP79u2rkcQBwMnJCbNnz8bixYvV1t+6dQuvvfYaAgMDIZVKIZPJMHLkSJw7d67GPv7ZZbt+/Xq0adMGjo6OCAgIwOLFi6FUKmtsKwgCkpOTVV2I48ePr7Gv+z1oHPOrr75C7969IZPJ4OjoCH9/fzzzzDPYtWuX2nYPGtM01GfVhJOTE0aOHIlvv/0W+fn5au/t3r0b+fn5mDBhQq1tb9y4gUWLFqF79+6QyWSQSqVo2rQpoqOjkZubq7Zt06ZNsXXrVgBAYGCg6rz/83xUv75+/TrGjx8PHx8f2NjYqM73/WPk5eXlaNeuHezt7XH06FG145WVlaFt27ZwcHBASkqKVudEE2FhYQgICAAAbN26VfV5YmJiVNto870+zIcffojg4GDV39acOXNQXl6uz48DAEhISMDXX3+NjRs3ol69enrfP1kvVuRmbMuWLVAoFHj55Zfh7e390G2lUqnq/9+6dQvdu3fHpUuXEBYWhlGjRuHKlSv48ssv8f333yMxMRGhoaE19jF79mwkJSVh8ODBCA8Px65duxATE4PKykosXboUADBs2DA0bdoUixcvRkBAgCqBd+zYsU6fMT4+HtHR0WjUqBGeffZZeHp6Ijs7G8eOHcOuXbswbNiwh7Y35GfV1MSJE7Fx40Z89tlnePXVV1XrP/roI8hkMgwePLjWdocOHcLKlSvx1FNPoVu3brC3t8eJEycQHx+PH374AcePH4e7uzsAYMaMGdiyZQtOnTqFV199VTUfoWnTpjXOR2hoKDw8PDBy5EhUVlbCzc2t1uM7Ojpi27Zt6Nq1K0aPHo2TJ0+qtn3ttdeQnp6O2NhYdO3aVavzoYnx48ejY8eOeP/999GhQwfV91z9w6Su3+v93n77bSxcuBDe3t546aWXYG9vj+3btyM9Pb3W7ZOSktCnTx/07t1bq4lzOTk5mDZtGiZOnIjw8HAcOXJE47ZEjySQ2QoLCxMACAcOHNCq3cSJEwUAwrx589TW79u3TwAgBAUFCQqFQrU+MjJSACAEBgYKN27cUK3Py8sT6tevL7i6ugoVFRVq+wIg9O7du8axq/eVkZFR471FixYJAISffvpJta5z586Cg4ODkJubW2P7/Pz8Rx7TGJ+1NhkZGQIAoX///oIgCMJjjz0mtG/fXvX+tWvXBFtbW+H1119Xxd6qVSu1fdy8eVMoLi6use+tW7cKAIQlS5aorX/Yua0+BgBhwoQJglwur/F+7969hdr+SVi3bp0AQBg9erQgCIKwa9cuAYDQp08ftXOnb9XnMDIyssZ72n6vtf1tXbx4UbCzsxMaN24s3Lx5U7W+sLBQaNWqVa1/Tz/99NMD/7YfZujQoYKvr69QUFCgFs+2bdu02g9Rbdi1bsZycnIAAH5+fhq3qaysxLZt2+Dp6Yk333xT7b3+/fujf//+uHjxYq0Vw4IFC9CoUSPVay8vL0RERKC4uBgXLlyo46d4NHt7e9jb29dY7+np+dB2pvRZJ0yYgNOnTyMtLQ3A370pEydOfGAbmUxWaxfsuHHj4ObmhgMHDmgdh4ODA1asWAFbW1uN20yZMgVDhgxBQkICVqxYgaioKHh4eODjjz+GjY3x/wnR5Xv9p4SEBMjlcsycORMymUy13s3NrcZ+qz3++ONIT0/Hxx9/rHG8H3/8MXbv3o34+HirunKDjIeJ3MqcP38eZWVlePzxx2ud4V3ddXny5Mka73Xu3LnGuuofEXfu3NFnmCrPP/88SkpKEBwcjFmzZuG7777T+Fim9FnHjRsHe3t7fPTRRwDuJfJu3bqhbdu2D2339ddfo3///mjYsKHqenwbGxsUFRXhxo0bWscRGBgILy8vrdt99NFHaNSoEebOnYtbt25h06ZNGv+ATEpKQkxMjNpy//wGbejyvf7TqVOnAABPPvlkjfdqWwcAzs7OaN26tcYTA2/cuIEZM2Zg1KhRGDp0qEZtiLTFMXIz5uPjg/Pnz+P69eto1aqVRm2KiooA4IFj6tUzhAsLC2u8Vz0e+092dvf+hB50nbSu5syZA09PT3zwwQeIi4vDypUrYWdnh4EDB2L16tUIDAx8YFtT+qwymQwDBw7Etm3bMHToUFy6dAmzZs16aJuVK1di1qxZaNiwIcLDw+Hn5wcnJycAwOrVq1FRUaF1HI+aS/EgXl5eePLJJ7Fjxw4EBARolZSSkpJqTLaMjIx85PyGB9Hle/2n6vf/WY1Xq+t5ul90dDRsbW2xdu1aveyPqDasyM1Yjx49AAAHDx7UuE31ZKWbN2/W+n71+gdNgNJVdVdsbddN1/YPr0QiwYsvvojU1FTk5eVh586dGD58OHbv3o1BgwY9NKmK/VnvN3HiRBQUFCAqKgpOTk544YUXHritXC7H22+/DV9fX5w9exafffYZli9fjpiYGCxatAiVlZV1iqGud0b74osvsGPHDnh6eiIzM1OrG7PExMRAEAS1ZcuWLXWKA9Df91r9Y+3+KwAetm9tnTx5Evn5+WjYsKFq5r1EIlH9sHnhhRcgkUiwevVqvRyPrBMTuRkbP348bG1tsXHjRuTl5T102+rqrXXr1nB0dERKSkqtN/9ITk4GUPdZ5o/SoEEDAMD169drvHfixImHtvX09MSwYcOwfft29O3bF+np6bh06dIDtxf7s95v4MCB8PHxwfXr1/Gvf/3roYkmPz8fhYWF6N69Oxo2bKj2XmpqKsrKymq0qR731nfvyNWrVzFp0iTIZDKcPHkSISEheOedd1Tnz9j09b126NABAPDzzz/XeK+2dXUxatQoREVF1Vg6deoEAOjTpw+ioqIQHBysl+ORdWIiN2MtWrTAnDlzkJ+fjwEDBiAjI6PGNuXl5YiLi1Ndf+vg4IAXXngB+fn5iI2NVdv2wIED2Lt3L1q0aKGq9vUtJCQEAGpUZF9++WWtieGHH36oUb1XVVXh9u3bAKDqaq6N2J/1fnZ2dti9ezd27tz5yEvYZDIZnJyccPz4cbVkVVBQgGnTptXaxsPDAwBw7do1vcWsVCoxduxY3LlzB1u2bIGfnx8SEhLg5OSEcePGoaCgQG/H0pS+vtfRo0fD1tYWcXFxalV5UVERlixZUmub0tJSnD9/HlevXtUo1nfeeQcffvhhjaV6aGLSpEn48MMP8fTTT2u0P6LacIzczC1ZsgTl5eVYtWoVWrVqhb59+yI4OBj29vbIyMjAgQMHcOvWLbV/mJYvX47k5GQsWbIER44cQbdu3VTX4Do7O2Pz5s0Gm408bNgwBAYGYsuWLcjKykKnTp2Qnp6OH3/8EQMHDsSePXvUth85ciScnZ3Rs2dPBAQEoKqqComJiTh37hxGjhz5yElHYn7W2nTt2lWj665tbGwQHR2NlStXokOHDhgyZAiKioqwd+9eBAQEwNfXt0abvn374r333sPLL7+MESNGwMXFBU2aNMHo0aPrHO/SpUvx888/Y/r06RgwYAAAICgoCGvWrEFUVBQmTZqEL774os77ryt9fK8tWrTAwoULsWjRIrRv3x7PP/887Ozs8NVXX6Fdu3a1Xp1w7NixOl1HTmRQ4l79RvqSkpIiTJw4UWjRooXg5OQkSKVSoWnTpsILL7wg7N+/v8b2eXl5wvTp04WAgADB3t5e8PLyEp577jnhzJkzNbbV9tpvQXjwdeSCIAiXL18WIiIiBFdXV8HFxUV46qmnhJSUlFr3tX79emHo0KFCQECA4OjoKHh6egrdunUTNmzYIFRVVWl0TEN/1trcfx35o6CW68grKyuFpUuXCkFBQYJUKhWaNGkizJw5UyguLhYCAgKEgICAGvtZsWKFEBQUJNjb29c4Hw/7TgSh5nXkv/76q2BnZycEBwcLZWVlNbZ/7rnnBADCpk2bNPqM2nrYdeSCoN33+rDvbtOmTULbtm0FBwcHwc/PT5g1a5ZQWlqq1+vIHxQPryMnfZAIgiAY+8cDERER6QfHyImIiMwYEzkREZEZYyInIiIyY0zkREREZoyJnIiIyIwxkRMREZkxs74hjFKpxI0bN+Dq6lrne0gTEZF4BEFAcXExfH19DXpzpvLy8jo/o+CfHBwc4OjoqIeI9MesE/mNGzfg7+8vdhhERKSjrKwsjR+Nq63y8nIEBtRDTq7uzyHw8fFBRkaGSSVzs07krq6uAIDM403hVo+jBMbwr07dxA7B6ihLaj4YhMhSyFGFw9ij+vfcECorK5GTq0BmWlO4udY9VxQVKxHQ5QoqKyuZyPWlujvdrZ6NTl8Oac5O4iB2CFZHKakSOwQiw/nr3qLGGB6t5ypBPde6H0cJ0xzCNetETkREpCmFoIRCh5uSKwSl/oLRIyZyIiKyCkoIUKLumVyXtobE/mgiIiIzxoqciIisghJK6NI5rltrw2EiJyIiq6AQBCh0eHK3Lm0NiV3rREREZowVORERWQVLnezGRE5ERFZBCQEKC0zk7FonIiIyY6zIiYjIKrBrnYiIyIxx1joRERGZHFbkRERkFZR/Lbq0N0VM5EREZBUUOs5a16WtITGRExGRVVAI0PHpZ/qLRZ84Rk5ERGTGWJETEZFV4Bg5ERGRGVNCAgUkOrU3RexaJyIiMmOsyImIyCoohXuLLu1NERM5ERFZBYWOXeu6tDUkdq0TERGZMVbkRERkFSy1ImciJyIiq6AUJFAKOsxa16GtIbFrnYiIyIyxIiciIqvArnUiIiIzpoANFDp0RCv0GIs+MZETEZFVEHQcIxc4Rk5ERET6xoqciIisAsfIiYiIzJhCsIFC0GGM3ERv0cqudSIiIjPGipyIiKyCEhIodahflTDNkpyJnIiIrIKljpGza52IiMiMsSInIiKroPtkN3atExERiebeGLkOD01h1zoRERHpGxO5AZXetUH8wsYY17UthjRrjxlDgnDhpJPq/cN73PHGC80w4rFg9PftiD9/d3rI3khbg0bnYP23J/HVid/w1YnfELfjNEJ6FYgdllUYHJmPrUfT8e3l01i37w8EP35X7JAsGs+3ZpR/3Wu9rou2M97lcjnefPNNBAYGwsnJCc2aNcNbb70FpVKp2kYQBMTExMDX1xdOTk4ICwvD2bNntTqO6Il8/fr1CAwMhKOjI7p06YKff/5Z7JD0ZtXr/jh+qB7mrM3EBwfPo0vvYvxnZAvkZ9sDAMpLbdC2awkmvnFD5EgtU36OAza/F4Dpz7bH9Gfb49Sv7lgYfx5NWpSKHZpF6z20AJMX38C2NTJEh7fE77+5YMlnGWjYuFLs0CwSz7fmqsfIdVm0sXz5cnzwwQdYt24d0tPTsWLFCrz77rtYu3atapsVK1YgLi4O69atQ0pKCnx8fNCvXz8UFxdrfBxRE/n27dsxY8YMzJ8/HydOnMCTTz6JAQMG4OrVq2KGpRcVZRIc3lMfL76ZjXbdS9A4sBLjZuXAx78S333sCQB4+rkCjJ15E5168dezIfz2owdSkhvg+hUnXL/ihK2rAlBeaovWHTX/D4S0N3xSPn7Y5oF9CZ7IuuSIDxY1Rt4Newz+9y2xQ7NIPN+aU/5VVeuyaOPXX39FREQEBg0ahKZNm+K5555DeHg4UlNTAdyrxlevXo358+dj+PDhCA4OxtatW1FaWoqEhASNjyNqIo+Li0NUVBRefPFFtGnTBqtXr4a/vz/i4+PFDEsvFAoJlAoJHKRKtfVSJyXOHqsnUlTWy8ZGQO9B+XB0VuD8SVexw7FYdvZKBLUvRVqy+jlOS3ZF25ASkaKyXDzf4igqKlJbKioqat2uZ8+eOHjwIP744w8AwKlTp3D48GEMHDgQAJCRkYGcnByEh4er2kilUvTu3RtHjhzROB7RZq1XVlYiLS0N//nPf9TWh4eHP/ADVFRUqJ2woqIig8aoC+d6SrTpUoKE1T5oEnQF9RvKkbSrAc4fd0bjwNq/dNK/pi1LELfjDBykSpSV2uLt6Na4eslZ7LAslpuHArZ2wJ189X9a7uTZoYFMLlJUlovnWzsKQQKFDo8irW7r7++vtn7RokWIiYmpsf3cuXNRWFiI1q1bw9bWFgqFAkuXLsULL7wAAMjJyQEAeHt7q7Xz9vZGZmamxnGJlsjz8/OhUChq/QDVH+5+sbGxWLx4sTHC04s5azMRN7MJRncOho2tgBbtStHn2QJcOsNEYizXMpwwZWgH1HNToEf/W3h9xUXMGRPMZG5g919uK5EAJnp3S4vA862Z6klrdW9/76RmZWXBzc1NtV4qlda6/fbt2/Hpp58iISEBjz32GE6ePIkZM2bA19cXkZGRqu0kEvUfF4Ig1Fj3MKJfR67NB5g3bx5mzpypel1UVFTjl5Ep8W1aife+voTyUhuUFNvA01uOpS8HwKcJK3JjkVfZIPvqvasBLv5eDy3b3UVEZDbWLmgucmSWqei2LRRyoEFD9WrQ3UuOgjzR/7mxODzf4nBzc1NL5A8ye/Zs/Oc//8GoUaMAAO3atUNmZiZiY2MRGRkJHx8fAPcq80aNGqna5ebm1ihyH0a0MXIvLy/Y2trWqL4f9gGkUqnqBGp6Ik2Bo7MSnt5yFN+xRVqyG0L7m+6QgKWTSAB7B+WjN6Q6kVfZ4OJpZ3TupT6hsHOvYpxLdREpKsvF860dpWCj86KN0tJS2Niot7G1tVVdfhYYGAgfHx8kJiaq3q+srERycjKeeOIJjY8j2k82BwcHdOnSBYmJiXj22WdV6xMTExERESFWWHqVmuQKQQD8m1fgeoYDPny7MfyalyN85L3ZpEUFtsi77oBbN+99DVl/3uueaSCrggfHt3QWOTMTqYcaIC/bAc4uCvQelI923QqxIKqt2KFZtK83emH2miz8cdoJ6akuGDj2FmSNq/D9X1drkH7xfGtOX13rmhoyZAiWLl2KJk2a4LHHHsOJEycQFxeHiRMnArjXIz1jxgwsW7YMQUFBCAoKwrJly+Ds7IzRo0drfBxR+15mzpyJcePGISQkBKGhodi4cSOuXr2KyZMnixmW3pQU2WJzbCPkZ9vDtb4CPQbewYT/ZMPu3mXkOLrfHStfa6LaPvaVpgCAsTNzMG5W7fMESHMNvKow+92L8JBVoqTYFhnnXbAgqi1O/FJf7NAsWvLuBnBtoMCY127CQyZH5gVHvDk2ELnXHcQOzSLxfJuutWvXYsGCBYiOjkZubi58fX3x8ssvY+HChapt5syZg7KyMkRHR6OgoADdunXD/v374eqq+dU1EkEQ9y7w69evx4oVK5CdnY3g4GCsWrUKvXr10qhtUVER3N3dUfBHM7i5in5vG6swIKiH2CFYHWUJLyMiyyUXqpCEb1BYWGiw4dLqXLHheBc41at7/Vp2V46XO6cZNNa6EH02RHR0NKKjo8UOg4iILFxdbupyf3tTZJpRERERkUZEr8iJiIiMQffnkZtm7ctETkREVsFSn0fORE5ERFbBUity04yKiIiINMKKnIiIrILuN4QxzdqXiZyIiKyCUpBAqcPTz3Rpa0im+fOCiIiINMKKnIiIrIJSx651U70hDBM5ERFZhbo8wez+9qbINKMiIiIijbAiJyIiq6CABAodbuqiS1tDYiInIiKrwK51IiIiMjmsyImIyCoooFv3uEJ/oegVEzkREVkFS+1aZyInIiKrwIemEBERkclhRU5ERFZB0PF55AIvPyMiIhIPu9aJiIjI5LAiJyIiq2CpjzFlIiciIqug0PHpZ7q0NSTTjIqIiIg0woqciIisArvWiYiIzJgSNlDq0BGtS1tDMs2oiIiISCOsyImIyCooBAkUOnSP69LWkJjIiYjIKnCMnIiIyIwJOj79TOCd3YiIiEjfWJETEZFVUEAChQ4PPtGlrSExkRMRkVVQCrqNcysFPQajR+xaJyIiMmOsyImIyCoodZzspktbQ2IiJyIiq6CEBEodxrl1aWtIpvnzgoiIiDTCipyIiKwC7+xGRERkxjhGbsKee7wH7CQOYodhFTalfy92CFZnUtfhYodgXRQKsSOwKoKyErgldhTmzSISORER0aMooeO91k10shsTORERWQVBx1nrAhM5ERGReCz16WemOXJPREREGmFFTkREVoGz1omIiMwYu9aJiIjI5LAiJyIiq2Cp91pnIiciIqvArnUiIiIyOazIiYjIKlhqRc5ETkREVsFSEzm71omIiMwYK3IiIrIKllqRM5ETEZFVEKDbJWSC/kLRKyZyIiKyCpZakXOMnIiIyIyxIiciIqtgqRU5EzkREVkFS03k7FonIiIyY6zIiYjIKlhqRc5ETkREVkEQJBB0SMa6tDUkdq0TERGZMVbkRERkFfg8ciIiIjNmqWPk7FonIiIyY6zIiYjIKljqZDcmciIisgqW2rXORE5ERFbBUityjpETEREZyPXr1zF27Fh4enrC2dkZHTt2RFpamup9QRAQExMDX19fODk5ISwsDGfPntXqGEzkRERkFYS/utbrumhbkRcUFKBHjx6wt7fH3r17ce7cOaxcuRL169dXbbNixQrExcVh3bp1SElJgY+PD/r164fi4mKNj8OudSIisgoCAEHQrb02li9fDn9/f2zevFm1rmnTpn/vTxCwevVqzJ8/H8OHDwcAbN26Fd7e3khISMDLL7+s0XFYkRMREWmhqKhIbamoqKh1u927dyMkJAQjRoyATCZDp06dsGnTJtX7GRkZyMnJQXh4uGqdVCpF7969ceTIEY3jYSInIiKrUH1nN10WAPD394e7u7tqiY2NrfV4ly9fRnx8PIKCgvDDDz9g8uTJmD59Oj7++GMAQE5ODgDA29tbrZ23t7fqPU2wa52IiKyCvmatZ2Vlwc3NTbVeKpXWur1SqURISAiWLVsGAOjUqRPOnj2L+Ph4/Pvf/1ZtJ5GoxyQIQo11D8OKnIiISAtubm5qy4MSeaNGjdC2bVu1dW3atMHVq1cBAD4+PgBQo/rOzc2tUaU/DBM5ERFZBV1mrNflZjI9evTAhQsX1Nb98ccfCAgIAAAEBgbCx8cHiYmJqvcrKyuRnJyMJ554QuPjsGudiIisgiDoOGtdy7avvfYannjiCSxbtgzPP/88jh07ho0bN2Ljxo0A7nWpz5gxA8uWLUNQUBCCgoKwbNkyODs7Y/To0Rofh4mciIjIALp27YqdO3di3rx5eOuttxAYGIjVq1djzJgxqm3mzJmDsrIyREdHo6CgAN26dcP+/fvh6uqq8XGYyImIyCqIcYvWwYMHY/DgwQ98XyKRICYmBjExMXWOi4lcRM+/dBXjX7uCXR83xsZ3mosdjtlTyIFvVjXBb7tkKMy1h7usCj1G3MTg6Vmw+Ws2SFSTnrW2HfFGBp6ZfN2I0Vouz4blmPDqRYT0uAUHqQLXrzrj/cWP4VK626Mbk1bGvHIZY165orbudr4Dxvat/e/c2lnqvdZFTeSHDh3Cu+++i7S0NGRnZ2Pnzp0YNmyYmCEZTVBwMZ4ZkY3L513EDsVi7I33Q/KnjTAx7g80blmKK6fr4aNZQXByVaBf1A0AQFzqb2ptziQ1wJbZQegyIF+MkC1OPdcqvLclBadTPLBwaifcue2ARv6luFvMmsFQrlxywfyXOqpeK5SmmWxMgVKQQMKnn+lXSUkJOnTogAkTJuBf//qXmKEYlaOzAnNWnMeaRS0x6uWrYodjMf5Mc0PH8Fvo8FQBAMDLvwK/7W6IK6frqbZxl1WptTmx3xOtQgvRMKD2OzORdp6bcAV5OY5YFfOYal1utpOIEVk+hVyCglu1X/5E1kHURD5gwAAMGDBAzBBEEf3mRRxL9sDJXxswketRUNciJH3mg5zLjvBpVo6scy64lOKGUYsu17p9YZ49zvzYABPjLho5UsvVvXce0o54Yt6KU2jXpQC3ch3x3Q4//LDTT+zQLFbjgFJ8cuAwqqpscOG0G7auaY6c6/zxVBtjz1o3FvZ3GVmvAblo0fYuXn2+s9ihWJwB0ddQVmyLN/t0gY2tAKVCgmdnZ6JbRO3d5ke+lEHqokCXZ9itri8+jcswaMQ17Py0Cbb/LxCtgoswec4FVFXZ4MfvfMUOz+JcOOOOlfPb4nqmM+p7VGLUpCt475M0vPJsNxQX2osdnsm5l8h1GSPXYzB6ZFaJvKKiQu3m9EVFRSJGoz0vn3K8PO9PvPlSO1RV8l48+nbsWy/8ulOGl9ZeQOOWpbh61gWfL26G+t6V6DEit8b2h3d4o/uzebB3NNH/Os2QxEbAxXNu2LouCABw+YIbmjS/i0EjrjGRG0DqYU+11+mn3fG/73/F00OzsfOTJiJFRcZmVok8NjYWixcvFjuMOgt67C4aeFVhzRfHVets7YDgkEIMGX0dER2fhJITVersi6WBGBh9Dd2G3quw/VqX4tZ1R+xZ71cjkf/xmxty/nTG5P9eqG1XVEcF+VJkXVafwJmV4YIeT9X8IUX6V1Fmi8yLLvANKBM7FJPEWesmYN68eZg5c6bqdVFREfz9/UWMSDsnf62PV4Z2UVv32tILuJbhjC8+9GcS11FlmQ0kNurVtY2NAKGW8/rzdm8EtCuGf9sSY4VnFc6drI/GAaVq6xo3KUVutqNIEVkXO3sl/JuV4vfj9cUOxSQJ0P6Z4ve3N0VmlcilUukDb05vDspK7ZB5Sf2Ul5fZouiOPTIv8TI0XXV4+ja+X+sPD9+Kv7rW62H/h43R8/mbatuVFdsi9XsvjHwzQ6RILdfOT5tg5ZYUPD8xAz8neqPVY4UY8K9rWPN220c3Jq1FvX4RvyV5IS/HUTVG7uwix8HdPmKHRkYkaiK/e/cuLl26pHqdkZGBkydPwsPDA02acHyHtDP6rcvY9V4TfPpmcxTn26O+dyV6j8nG0Fez1LY7ttsLEIDHI/JEitRyXTznjiWvd8D4aZcwetJl5Fx3woZ3WyFpbyOxQ7NIXrIKzF1+Fm4NqlB42x4XzrjjtbEhvOTvASy1a10iCOLNw0tKSkKfPn1qrI+MjMSWLVse2b6oqAju7u54qv442EkcDBAh3W/j6e/FDsHqTOo6XOwQrItCIXYEVkWurMTBW5tRWFio9oxvfarOFc22vgFb57oP8yhKy3E5cplBY60LUSvysLAwiPg7goiIrImOFTlMtCLnNVBERERmzKwmuxEREdUV7+xGRERkxix1shu71omIiMwYK3IiIrIOgkS3CWsmWpEzkRMRkVWw1DFydq0TERGZMVbkRERkHSz0ZusaJfI1a9ZovMPp06fXORgiIiJDsdRZ6xol8lWrVmm0M4lEwkRORERkRBol8owMPiWKiIgsgIl2j+uizpPdKisrceHCBcjlcn3GQ0REZBDVXeu6LKZI60ReWlqKqKgoODs747HHHsPVq1cB3Bsbf+edd/QeIBERkV4IelhMkNaJfN68eTh16hSSkpLg6Pj34+CefvppbN++Xa/BERER0cNpffnZrl27sH37dnTv3h0Syd/dDG3btsWff/6p1+CIiIj0R/LXokt706N1Is/Ly4NMJquxvqSkRC2xExERmRQLvY5c6671rl274vvvv1e9rk7emzZtQmhoqP4iIyIiokfSuiKPjY3FM888g3PnzkEul+P999/H2bNn8euvvyI5OdkQMRIREemOFfk9TzzxBH755ReUlpaiefPm2L9/P7y9vfHrr7+iS5cuhoiRiIhId9VPP9NlMUF1utd6u3btsHXrVn3HQkRERFqqUyJXKBTYuXMn0tPTIZFI0KZNG0RERMDOjs9gISIi02SpjzHVOvP+/vvviIiIQE5ODlq1agUA+OOPP9CwYUPs3r0b7dq103uQREREOuMY+T0vvvgiHnvsMVy7dg3Hjx/H8ePHkZWVhfbt22PSpEmGiJGIiIgeQOuK/NSpU0hNTUWDBg1U6xo0aIClS5eia9eueg2OiIhIb3SdsGaik920rshbtWqFmzdv1lifm5uLFi1a6CUoIiIifZMIui+mSKOKvKioSPX/ly1bhunTpyMmJgbdu3cHABw9ehRvvfUWli9fbpgoiYiIdGWhY+QaJfL69eur3X5VEAQ8//zzqnXCX1P5hgwZAoVCYYAwiYiIqDYaJfKffvrJ0HEQEREZloWOkWuUyHv37m3oOIiIiAzLmrvWa1NaWoqrV6+isrJSbX379u11DoqIiIg0U6fHmE6YMAF79+6t9X2OkRMRkUmy0Ipc68vPZsyYgYKCAhw9ehROTk7Yt28ftm7diqCgIOzevdsQMRIREelO0MNigrSuyH/88Ud888036Nq1K2xsbBAQEIB+/frBzc0NsbGxGDRokCHiJCIiolpoXZGXlJRAJpMBADw8PJCXlwfg3hPRjh8/rt/oiIiI9MVCH2Napzu7XbhwAQDQsWNHbNiwAdevX8cHH3yARo0a6T1AIiIifbDqO7v904wZM5CdnQ0AWLRoEfr374/PPvsMDg4O2LJli77jIyIioofQOpGPGTNG9f87deqEK1eu4Pz582jSpAm8vLz0GhwREZHeWOis9TpfR17N2dkZnTt31kcsREREpCWNEvnMmTM13mFcXFydgyEiIjIUCXQb5zbNqW4aJvITJ05otLN/PliFiIiIDM8iHpqiuFMEicRe7DCswqTOEWKHYHU+PrFT7BCsSmTPUWKHYFUkSlvglpEOZs0PTSEiIjJ7FjrZTevryImIiMh0sCInIiLrYKEVORM5ERFZBV3vzmaqd3Zj1zoREZEZq1Mi/+STT9CjRw/4+voiMzMTALB69Wp88803eg2OiIhIbyz0MaZaJ/L4+HjMnDkTAwcOxJ07d6BQKAAA9evXx+rVq/UdHxERkX4wkd+zdu1abNq0CfPnz4etra1qfUhICM6cOaPX4IiIiOjhtJ7slpGRgU6dOtVYL5VKUVJSopegiIiI9I2T3f4SGBiIkydP1li/d+9etG3bVh8xERER6V/1nd10WUyQ1hX57NmzMWXKFJSXl0MQBBw7dgzbtm1DbGwsPvzwQ0PESEREpDteR37PhAkTIJfLMWfOHJSWlmL06NFo3Lgx3n//fYwaxXsUExERGVOdbgjz0ksv4aWXXkJ+fj6USiVkMpm+4yIiItIrSx0j1+nObl5eXvqKg4iIyLDYtX5PYGDgQ587fvnyZZ0CIiIiIs1pnchnzJih9rqqqgonTpzAvn37MHv2bH3FRUREpF86dq1bTEX+6quv1rr+v//9L1JTU3UOiIiIyCAstGtdbw9NGTBgAL766it97Y6IiMhixMbGQiKRqPVqC4KAmJgY+Pr6wsnJCWFhYTh79qzW+9ZbIv/yyy/h4eGhr90RERHpl0j3Wk9JScHGjRvRvn17tfUrVqxAXFwc1q1bh5SUFPj4+KBfv34oLi7Wav9ad6136tRJbbKbIAjIyclBXl4e1q9fr+3uiIiIjEKMy8/u3r2LMWPGYNOmTViyZIlqvSAIWL16NebPn4/hw4cDALZu3Qpvb28kJCTg5Zdf1vgYWifyYcOGqb22sbFBw4YNERYWhtatW2u7OyIiIrNSVFSk9loqlUIqlda67ZQpUzBo0CA8/fTTaok8IyMDOTk5CA8PV9tP7969ceTIEcMlcrlcjqZNm6J///7w8fHRpikREZFF8Pf3V3u9aNEixMTE1Nju888/x/Hjx5GSklLjvZycHACAt7e32npvb29kZmZqFY9WidzOzg6vvPIK0tPTtToIERGR6PQ0az0rKwtubm6q1bVV41lZWXj11Vexf/9+ODo6PnCX99+XRRCEh96rpTZaT3br1q0bTpw4oW0zIiIiUVWPkeuyAICbm5vaUlsiT0tLQ25uLrp06QI7OzvY2dkhOTkZa9asgZ2dnaoSr67Mq+Xm5tao0h9F6zHy6OhovP7667h27Rq6dOkCFxcXtffvn5VHRERkbZ566imcOXNGbd2ECRPQunVrzJ07F82aNYOPjw8SExPRqVMnAEBlZSWSk5OxfPlyrY6lcSKfOHEiVq9ejZEjRwIApk+frnpPIpGougMUCoVWARARERmNkW7q4urqiuDgYLV1Li4u8PT0VK2fMWMGli1bhqCgIAQFBWHZsmVwdnbG6NGjtTqWxol869ateOedd5CRkaHVAYiIiEyCid3Zbc6cOSgrK0N0dDQKCgrQrVs37N+/H66urlrtR+NELgj3PkFAQIB2kRIRERGSkpLUXkskEsTExNQ6410bWo2RazuTjoiIyFTweeQAWrZs+chkfvv2bZ0CIiIiMggT61rXF60S+eLFi+Hu7m6oWIiIiEhLWiXyUaNGQSaTGSoWIiIig7H6rnWOjxMRkVmz0K51je/sVj1rnYiIiEyHxhW5Uqk0ZBxERESGZaEVuda3aCUiIjJHVj9GTkREZNYstCLX+ulnREREZDpYkRMRkXWw0IqciVwEgyPzMeKVPHjIqpD5hyM+WOiL34/VEzssizPmlcsY88oVtXW38x0wtm9PcQKyMAo58FVcExzZ1RB3cu1R37sKvUbkYtj0LNj81ddXXmKDz2ObIvUHD9wtsEND/wr0n5CNp/+d8/Cdk0ZsbJUYE/UHwvpfRwPPChTkO+LAHj98vjkIgsBLhu/HMXLSi95DCzB58Q2se6Mxzh5zwaBxt7Dkswy8FNYKedcdxA7P4ly55IL5L3VUvVYo+Y+bvny73g8HP/XB5FUX4deyFJdP18PG14Pg7CrHM1HZAIBPFgci/Yg7otf8gYZ+FThzqD42z2+O+t6VCOnP2znrasTYPzHg2UysersjMi+7IqjNHcyYfwold+2we0czscMjIxF1jDw2NhZdu3aFq6srZDIZhg0bhgsXLogZksENn5SPH7Z5YF+CJ7IuOeKDRY2Rd8Meg/99S+zQLJJCLkHBLalqKSrgjyV9uXjcFV3Cb6PTUwVo6F+BboNuoV2vAlw+/Xfv0qU0Vzz5XC7ahhahoX8F+o65iSZtS5Bxmj1Q+tC6XQF++9kHKUe8kZvjjF9+8sWJYw0R1LpQ7NBMk6CHxQSJmsiTk5MxZcoUHD16FImJiZDL5QgPD0dJSYmYYRmMnb0SQe1LkZas/qzZtGRXtA2xzM8stsYBpfjkwGF8tPcI5i7/HT6Ny8QOyWK06lqEs7+4I/uyIwAg85wzLqS4oWOfAtU2LR8vxvFED9zOdoAgAGePuCPnshPa9y540G5JC+dOeaBDSD58/e8CAAJbFKFth9tI/ZW30q5Ndde6LospErVrfd++fWqvN2/eDJlMhrS0NPTq1UukqAzHzUMBWzvgTr76ab+TZ4cGMrlIUVmuC2fcsXJ+W1zPdEZ9j0qMmnQF732Shlee7YbiQnuxwzN7Q6Kvo7TYDrPDOsPGVoBSIcGIOZl4Yli+apvIxZfx4dwWmPZ4V9jaKSGxAV5ccQmtHi8WMXLL8cUnzeFcrwobPk+CUimBjY2Ajze0RnJiY7FDIyMyqTHywsJ73UEeHh61vl9RUYGKigrV66KiIqPEpW/33+1WIoHJdtmYs9TDnmqv00+743/f/4qnh2Zj5ydNRIrKchzd7YVfvm6IKWv/QOOWpcg854JPYwLRwLsSvUbkAQB++KgRLh13xesfnYOXXwXO/+aGLfObo4GsEsFPsvtXV72evoE+/a/j3UWdkJnhimZBRZg04yxu50txcI+/2OGZHs5aNyxBEDBz5kz07NkTwcHBtW4TGxuLxYsXGzky/Sm6bQuFHGjQUL36dveSoyDPZL4Ki1VRZovMiy7wDWD3uj4kLG2KIdHXEBpxrwJv0qYU+dek2P1fP/QakYfKMhtsXxGA1zadR6enClTbZJ6th+83NGYi14OJU9PxxSctcOjAvQo88083yHzKMOLfl5jIa2OhidxkbggzdepUnD59Gtu2bXvgNvPmzUNhYaFqycrKMmKEupNX2eDiaWd07qXerdi5VzHOpbqIFJX1sLNXwr9ZKW7nccKbPlSW2aguM6tmYytA+OvKALlcAkWVDSQ2Qo1t+OgG/ZA6KiDcdy6VSglseHGGVTGJMnDatGnYvXs3Dh06BD8/vwduJ5VKIZVKjRiZ/n290Quz12Thj9NOSE91wcCxtyBrXIXvP/Z8dGPSStTrF/FbkhfychxVY+TOLnIc3O0jdmgWodPTt7FrrR88G1fAr2Uprvzugr2bGqP3yJsAAGdXBdp0L8S2JU3h4HgZXo0rkH7UHT9/2RBjF14RN3gLceywN0aOv4S8m07IvOyK5q0K8eyoy0j8jtV4bSR/Lbq0N0WiJnJBEDBt2jTs3LkTSUlJCAwMFDMco0je3QCuDRQY89pNeMjkyLzgiDfHBiKX15DrnZesAnOXn4VbgyoU3rbHhTPueG1sCHKzncQOzSJEvp2BL99rgs3zm6Eo3x4NvCvRd0wOhs/4u6ds6n8vYPs7AVg/rSXu3rGDl18Fnp9zFU+N4w1h9OGDuGCMnXQB0bN+h7tHBW7nOWLvribY9lFLsUMzTRbatS4RRHzQeHR0NBISEvDNN9+gVatWqvXu7u5wcnr0P7ZFRUVwd3dHGCJgJ+EsZGOw9WLPgbF9fGK32CFYlcieo8QOwarIlRU4cHU9CgsL4ebmZpBjVOeKxyYvg63Usc77UVSU4+wHbxg01roQdYw8Pj4ehYWFCAsLQ6NGjVTL9u3bxQyLiIjIbIjetU5ERGQUFtq1bhKT3YiIiIzCRJOxLkzm8jMiIiLSHityIiKyCnyMKRERkTmz0DFydq0TERGZMVbkRERkFdi1TkREZM7YtU5ERESmhhU5ERFZBXatExERmTML7VpnIiciIutgoYmcY+RERERmjBU5ERFZBY6RExERmTN2rRMREZGpYUVORERWQSIIkAh1L6t1aWtITORERGQd2LVOREREpoYVORERWQXOWiciIjJn7FonIiIiU8OKnIiIrAK71omIiMyZhXatM5ETEZFVsNSKnGPkREREZowVORERWQd2rRMREZk3U+0e1wW71omIiMwYK3IiIrIOgnBv0aW9CWIiJyIiq8BZ60RERGRyWJETEZF14Kx1IiIi8yVR3lt0aW+K2LVORERkxliRExGRdWDXOhERkfmy1FnrTORERGQdLPQ6co6RExERmTFW5EREZBXYtU4EQJF/S+wQrM641uFih2BVYn//UuwQrMrdYiUOtDPSwSx0shu71omIiMwYK3IiIrIK7FonIiIyZ5y1TkRERKaGFTkREVkFdq0TERGZM85aJyIiIk3Fxsaia9eucHV1hUwmw7Bhw3DhwgW1bQRBQExMDHx9feHk5ISwsDCcPXtWq+MwkRMRkVWo7lrXZdFGcnIypkyZgqNHjyIxMRFyuRzh4eEoKSlRbbNixQrExcVh3bp1SElJgY+PD/r164fi4mKNj8OudSIisg5K4d6iS3st7Nu3T+315s2bIZPJkJaWhl69ekEQBKxevRrz58/H8OHDAQBbt26Ft7c3EhIS8PLLL2t0HFbkRERkHQQ9LACKiorUloqKCo0OX1hYCADw8PAAAGRkZCAnJwfh4X/fvVEqlaJ37944cuSIxh+LiZyIiEgL/v7+cHd3Vy2xsbGPbCMIAmbOnImePXsiODgYAJCTkwMA8Pb2VtvW29tb9Z4m2LVORERWQQIdLz/763+zsrLg5uamWi+VSh/ZdurUqTh9+jQOHz5cc78SidprQRBqrHsYJnIiIrIOerqzm5ubm1oif5Rp06Zh9+7dOHToEPz8/FTrfXx8ANyrzBs1aqRan5ubW6NKfxh2rRMRERmAIAiYOnUqvv76a/z4448IDAxUez8wMBA+Pj5ITExUrausrERycjKeeOIJjY/DipyIiKyCse/sNmXKFCQkJOCbb76Bq6uratzb3d0dTk5OkEgkmDFjBpYtW4agoCAEBQVh2bJlcHZ2xujRozU+DhM5ERFZByPf2S0+Ph4AEBYWprZ+8+bNGD9+PABgzpw5KCsrQ3R0NAoKCtCtWzfs378frq6uGh+HiZyIiMgABA3G4yUSCWJiYhATE1Pn4zCRExGRVZAIAiQ6THbTpa0hMZETEZF1UP616NLeBHHWOhERkRljRU5ERFaBXetERETmzEKfR85ETkRE1kFPd3YzNRwjJyIiMmOsyImIyCoY+85uxsJETkRE1oFd60RERGRqWJETEZFVkCjvLbq0N0VM5EREZB3YtU5ERESmhhU5ERFZB94QhoiIyHxZ6i1a2bVORERkxliRExGRdbDQyW5M5EREZB0E6PZMcdPM40zkRERkHThGTkRERCaHFTkREVkHATqOkestEr1iIiciIutgoZPd2LVORERkxliRi2BwZD5GvJIHD1kVMv9wxAcLffH7sXpih2WxeL6NZ9DoHAx6IQfefhUAgMyLTkhY54/UQw1EjswyKOTAgdV+OPGNJ4rzHOAmq0SX5/LRd+p12NRSln31RlMc2+aNwQsy8eTEHOMHbGqUACQ6tjdBolbk8fHxaN++Pdzc3ODm5obQ0FDs3btXzJAMrvfQAkxefAPb1sgQHd4Sv//mgiWfZaBh40qxQ7NIPN/GlZ/jgM3vBWD6s+0x/dn2OPWrOxbGn0eTFqVih2YRkj/wxdEEGSIWZ+L1A6cw4D9ZSN7YCEe2etfY9uz+Bsg6WQ9u3vxbr1Y9a12XxRSJmsj9/PzwzjvvIDU1Fampqejbty8iIiJw9uxZMcMyqOGT8vHDNg/sS/BE1iVHfLCoMfJu2GPwv2+JHZpF4vk2rt9+9EBKcgNcv+KE61ecsHVVAMpLbdG6Y7HYoVmEzBP10LZfAdr0vQMPv0q0H3gbLZ8sxLXT6j1MhTn22LWoKUat/hO2dqaZfEh/RE3kQ4YMwcCBA9GyZUu0bNkSS5cuRb169XD06FExwzIYO3slgtqXIi3ZVW19WrIr2oaUiBSV5eL5FpeNjYDeg/Lh6KzA+ZOuj25Aj9Q0pBh//uKOvMuOAIAb55xxJcUVrfvcUW2jVALbZzZH70k34NOyTKRITVT1ZDddFhNkMmPkCoUCX3zxBUpKShAaGip2OAbh5qGArR1wJ1/9tN/Js0MDmVykqCwXz7c4mrYsQdyOM3CQKlFWaou3o1vj6iVnscOyCGGTs1FebIeVT7eHxFaAoJCg/6xr6Dj07x6m5A98YWML9Bh/U8RITZSFzloXPZGfOXMGoaGhKC8vR7169bBz5060bdu21m0rKipQUVGhel1UVGSsMPXq/r8FiQQme32iJeD5Nq5rGU6YMrQD6rkp0KP/Lby+4iLmjAlmMteDU9954MQuT4x6/xK8g8qQfc4F377dBG7elejyr3xcO+OMw5u98ep3v9/7OyerIHoib9WqFU6ePIk7d+7gq6++QmRkJJKTk2tN5rGxsVi8eLEIUepH0W1bKORAg4bq1aC7lxwFeaJ/FRaH51sc8iobZF91AgBc/L0eWra7i4jIbKxd0FzkyMzfntgmCJucjY5DbgMAGrUuQ8F1B/y03hdd/pWPjBQ3lNyyR2yPTqo2SoUE3y9tgl8+8sF/Dp8UKXITwYrcMBwcHNCiRQsAQEhICFJSUvD+++9jw4YNNbadN28eZs6cqXpdVFQEf39/o8WqK3mVDS6edkbnXsU4ss9dtb5zr2L8+oP7Q1pSXfB8mwaJBLB3MNHrdsxMVZkNJDbqycTGFhD+Or2dn81HUI9Ctff/F9kanZ/NR8hzecYK03RZ6OVnoify+wmCoNZ9/k9SqRRSqdTIEenX1xu9MHtNFv447YT0VBcMHHsLssZV+P5jT7FDs0g838YVOTMTqYcaIC/bAc4uCvQelI923QqxIKr24TLSTpun7uDH/zZGfd9KeLcsxY2zLvj5fz4IGXEvSbs0kMOlgXoPlK2dgHoNq9CwebkYIZsUS31oiqiJ/I033sCAAQPg7++P4uJifP7550hKSsK+ffvEDMugknc3gGsDBca8dhMeMjkyLzjizbGByL3uIHZoFonn27gaeFVh9rsX4SGrREmxLTLOu2BBVFuc+KW+2KFZhIiYK/ghzg+7FjTF3Vv2cPOuRLcXcvHU9Otih0YikgiCeD8xoqKicPDgQWRnZ8Pd3R3t27fH3Llz0a9fP43aFxUVwd3dHWGIgJ3E3sDREonDxsVF7BCsSuzvP4odglW5W6xE73bXUVhYCDc3N4McozpXPB30Guxs696rK1dU4MDFVQaNtS5Ercj/97//iXl4IiKyJkoBkOhQuypNs2udD00hIiIyYyY32Y2IiMggePkZERGROdP1NqummcjZtU5ERGTGWJETEZF1YNc6ERGRGVMK0Kl7nLPWiYiISN9YkRMRkXUQlH/fmL6u7U0QEzkREVkHjpETERGZMY6RExERkalhRU5ERNaBXetERERmTICOiVxvkegVu9aJiIjMGCtyIiKyDuxaJyIiMmNKJQAdrgVXmuZ15OxaJyIiMmOsyImIyDqwa52IiMiMWWgiZ9c6ERGRGWNFTkRE1sFCb9HKRE5ERFZBEJQQdHiCmS5tDYmJnIiIrIMg6FZVc4yciIiI9I0VORERWQdBxzFyE63ImciJiMg6KJWARIdxbhMdI2fXOhERkRljRU5ERNaBXetERETmS1AqIejQtW6ql5+xa52IiMiMsSInIiLrwK51IiIiM6YUAInlJXJ2rRMREZkxVuRERGQdBAGALteRm2ZFzkRORERWQVAKEHToWhdMNJGza52IiKyDoNR9qYP169cjMDAQjo6O6NKlC37++We9fiwmciIiIgPZvn07ZsyYgfnz5+PEiRN48sknMWDAAFy9elVvx2AiJyIiqyAoBZ0XbcXFxSEqKgovvvgi2rRpg9WrV8Pf3x/x8fF6+1xM5EREZB2M3LVeWVmJtLQ0hIeHq60PDw/HkSNH9PaxzHqyW/XEAzmqdLrGn8iU2QiVYodgVe4Wm+ZtOC1Vyd1759sYE8l0zRVyVAEAioqK1NZLpVJIpdIa2+fn50OhUMDb21ttvbe3N3JycuoeyH3MOpEXFxcDAA5jj8iREBlQidgBWJcf24kdgXUqLi6Gu7u7Qfbt4OAAHx8fHM7RPVfUq1cP/v7+ausWLVqEmJiYB7aRSCRqrwVBqLFOF2adyH19fZGVlQVXV1e9nhRDKyoqgr+/P7KysuDm5iZ2OFaB59y4eL6Nz1zPuSAIKC4uhq+vr8GO4ejoiIyMDFRW6t67VVsSrq0aBwAvLy/Y2trWqL5zc3NrVOm6MOtEbmNjAz8/P7HDqDM3Nzez+g/OEvCcGxfPt/GZ4zk3VCX+T46OjnB0dDT4cf7JwcEBXbp0QWJiIp599lnV+sTEREREROjtOGadyImIiEzZzJkzMW7cOISEhCA0NBQbN27E1atXMXnyZL0dg4mciIjIQEaOHIlbt27hrbfeQnZ2NoKDg7Fnzx4EBATo7RhM5CKQSqVYtGjRA8dVSP94zo2L59v4eM5NV3R0NKKjow22f4lgqjePJSIiokfiDWGIiIjMGBM5ERGRGWMiJyIiMmNM5ERERGaMiVwEhn42Lf3t0KFDGDJkCHx9fSGRSLBr1y6xQ7JosbGx6Nq1K1xdXSGTyTBs2DBcuHBB7LAsVnx8PNq3b6+6CUxoaCj27t0rdlhkZEzkRmaMZ9PS30pKStChQwesW7dO7FCsQnJyMqZMmYKjR48iMTERcrkc4eHhKCnhDeMNwc/PD++88w5SU1ORmpqKvn37IiIiAmfPnhU7NDIiXn5mZN26dUPnzp3VnkXbpk0bDBs2DLGxsSJGZvkkEgl27tyJYcOGiR2K1cjLy4NMJkNycjJ69eoldjhWwcPDA++++y6ioqLEDoWMhBW5ERnr2bREpqKwsBDAveRChqVQKPD555+jpKQEoaGhYodDRsQ7uxmRsZ5NS2QKBEHAzJkz0bNnTwQHB4sdjsU6c+YMQkNDUV5ejnr16mHnzp1o27at2GGRETGRi8DQz6YlMgVTp07F6dOncfjwYbFDsWitWrXCyZMncefOHXz11VeIjIxEcnIyk7kVYSI3ImM9m5ZIbNOmTcPu3btx6NAhs37UsDlwcHBAixYtAAAhISFISUnB+++/jw0bNogcGRkLx8iN6J/Ppv2nxMREPPHEEyJFRaQ/giBg6tSp+Prrr/Hjjz8iMDBQ7JCsjiAIqKioEDsMMiJW5EZmjGfT0t/u3r2LS5cuqV5nZGTg5MmT8PDwQJMmTUSMzDJNmTIFCQkJ+Oabb+Dq6qrqfXJ3d4eTk5PI0VmeN954AwMGDIC/vz+Ki4vx+eefIykpCfv27RM7NDIiXn4mgvXr12PFihWqZ9OuWrWKl+YYSFJSEvr06VNjfWRkJLZs2WL8gCzcg+Z6bN68GePHjzduMFYgKioKBw8eRHZ2Ntzd3dG+fXvMnTsX/fr1Ezs0MiImciIiIjPGMXIiIiIzxkRORERkxpjIiYiIzBgTORERkRljIiciIjJjTORERERmjImciIjIjDGRE+koJiYGHTt2VL0eP368KM88v3LlCiQSCU6ePPnAbZo2bYrVq1drvM8tW7agfv36OscmkUiwa9cunfdDRDUxkZNFGj9+PCQSCSQSCezt7dGsWTPMmjULJSUlBj/2+++/r/Fd4zRJvkRED8N7rZPFeuaZZ7B582ZUVVXh559/xosvvoiSkhLEx8fX2Laqqgr29vZ6Oa67u7te9kNEpAlW5GSxpFIpfHx84O/vj9GjR2PMmDGq7t3q7vCPPvoIzZo1g1QqhSAIKCwsxKRJkyCTyeDm5oa+ffvi1KlTavt955134O3tDVdXV0RFRaG8vFzt/fu71pVKJZYvX44WLVpAKpWiSZMmWLp0KQCong7WqVMnSCQShIWFqdpt3rwZbdq0gaOjI1q3bo3169erHefYsWPo1KkTHB0dERISghMnTmh9juLi4tCuXTu4uLjA398f0dHRuHv3bo3tdu3ahZYtW8LR0RH9+vVDVlaW2vvffvstunTpAkdHRzRr1gyLFy+GXC7XOh4i0h4TOVkNJycnVFVVqV5funQJO3bswFdffaXq2h40aBBycnKwZ88epKWloXPnznjqqadw+/ZtAMCOHTuwaNEiLF26FKmpqWjUqFGNBHu/efPmYfny5ViwYAHOnTuHhIQE1fPnjx07BgA4cOAAsrOz8fXXXwMANm3ahPnz52Pp0qVIT0/HsmXLsGDBAmzduhUAUFJSgsGDB6NVq1ZIS0tDTEwMZs2apfU5sbGxwZo1a/D7779j69at+PHHHzFnzhy1bUpLS7F06VJs3boVv/zyC4qKijBq1CjV+z/88APGjh2L6dOn49y5c9iwYQO2bNmi+rFCRAYmEFmgyMhIISIiQvX6t99+Ezw9PYXnn39eEARBWLRokWBvby/k5uaqtjl48KDg5uYmlJeXq+2refPmwoYNGwRBEITQ0FBh8uTJau9369ZN6NChQ63HLioqEqRSqbBp06Za48zIyBAACCdOnFBb7+/vLyQkJKite/vtt4XQ0FBBEARhw4YNgoeHh1BSUqJ6Pz4+vtZ9/VNAQICwatWqB76/Y8cOwdPTU/V68+bNAgDh6NGjqnXp6ekCAOG3334TBEEQnnzySWHZsmVq+/nkk0+ERo0aqV4DEHbu3PnA4xJR3XGMnCzWd999h3r16kEul6OqqgoRERFYu3at6v2AgAA0bNhQ9TotLQ13796Fp6en2n7Kysrw559/AgDS09NrPDs+NDQUP/30U60xpKeno6KiAk899ZTGcefl5SErKwtRUVF46aWXVOvlcrlq/D09PR0dOnSAs7OzWhza+umnn7Bs2TKcO3cORUVFkMvlKC8vR0lJCVxcXAAAdnZ2CAkJUbVp3bo16tevj/T0dDz++ONIS0tDSkqKWgWuUChQXl6O0tJStRiJSP+YyMli9enTB/Hx8bC3t4evr2+NyWzViaqaUqlEo0aNkJSUVGNfdb0Ey8nJSes2SqUSwL3u9W7duqm9Z2trCwAQ9PD04czMTAwcOBCTJ0/G22+/DQ8PDxw+fBhRUVFqQxBA7c8Zr16nVCqxePFiDB8+vMY2jo6OOsdJRA/HRE4Wy8XFBS1atNB4+86dOyMnJwd2dnZo2rRprdu0adMGR48exb///W/VuqNHjz5wn0FBQXBycsLBgwfx4osv1njfwcEBwL0Ktpq3tzcaN26My5cvY8yYMbXut23btvjkk09QVlam+rHwsDhqk5qaCrlcjpUrV8LG5t50mR07dtTYTi6XIzU1FY8//jgA4MKFC7hz5w5at24N4N55u3Dhglbnmoj0h4mc6C9PP/00QkNDMWzYMCxfvhytWrXCjRs3sGfPHgwbNgwhISF49dVXERkZiZCQEPTs2ROfffYZzp49i2bNmtW6T0dHR8ydOxdz5syBg4MDevTogby8PJw9exZRUVGQyWRwcnLCvn374OfnB0dHR7i7uyMmJgbTp0+Hm5sbBgwYgIqKCqSmpqKgoAAzZ87E6NGjMX/+fERFReHNN9/ElStX8N5772n1eZs3bw65XI61a9diyJAh+OWXX/DBBx/U2M7e3h7Tpk3DmjVrYG9vj6lTp6J79+6qxL5w4UIMHjwY/v7+GDFiBGxsbHD69GmcOXMGS5Ys0f6LICKtcNY60V8kEgn27NmDXr16YeLEiWjZsiVGjRqFK1euqGaZjxw5EgsXLsTcuXPRpUsXZGZm4pVXXnnofhcsWIDXX38dCxcuRJs2bTBy5Ejk5uYCuDf+vGbNGmzYsAG+vr6IiIgAALz44ov48MMPsWXLFrRr1w69e/fGli1bVJer1atXD99++y3OnTuHTp06Yf78+Vi+fLlWn7djx46Ii4vD8uXLERwcjM8++wyxsbE1tnN2dsbcuXMxevRohIaGwsnJCZ9//rnq/f79++O7775DYmIiunbtiu7duyMuLg4BAQFaxUNEdSMR9DHYRkRERKJgRU5ERGTGmMiJiIjMGBM5ERGRGWMiJyIiMmNM5ERERGaMiZyIiMiMMZETERGZMSZyIiIiM8ZETkREZMaYyImIiMwYEzkREZEZYyInIiIyY/8HnY1rVyGgsckAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "------------------------------------------------------------------------\n",
      "Training for fold 5 ...\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "     Class 0       0.96      0.89      0.92       101\n",
      "     Class 1       0.84      0.84      0.84        96\n",
      "     Class 2       0.89      0.92      0.90        96\n",
      "     Class 3       0.86      0.90      0.88        91\n",
      "\n",
      "    accuracy                           0.89       384\n",
      "   macro avg       0.89      0.89      0.89       384\n",
      "weighted avg       0.89      0.89      0.89       384\n",
      "\n",
      "\n",
      "Score for fold 5: - Accuracy: 0.8880208333333334 - Specifity: 0.9628048364803135 - Sensivity: 0.8881511691691146\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# K-fold Cross Validation model evaluation\n",
    "fold_no = 1\n",
    "for train, test in kfold.split(X, Y):\n",
    "    \n",
    "    # Generate a print\n",
    "    print('------------------------------------------------------------------------')\n",
    "    print(f'Training for fold {fold_no} ...\\n')\n",
    "\n",
    "    # Fit data to model\n",
    "    gpc_rbf_model.fit(X[train], Y[train].flatten())\n",
    "    \n",
    "    # Predict the labels of test set\n",
    "    y_pred = gpc_rbf_model.predict(X[test])\n",
    "\n",
    "    target_names = [\"Class {}\".format(i) for i in range(n)]\n",
    "    print(classification_report(Y[test], y_pred, target_names = target_names, zero_division = 0))\n",
    "\n",
    "    # Confusion Matrix\n",
    "    mcm = multilabel_confusion_matrix(Y[test], y_pred)\n",
    "\n",
    "    tn = mcm[:, 0, 0]\n",
    "    tp = mcm[:, 1, 1]\n",
    "    fn = mcm[:, 1, 0]\n",
    "    fp = mcm[:, 0, 1]\n",
    "\n",
    "    specifity = np.mean(tn / (tn + fp))\n",
    "    sensivity = np.mean(tp / (tp + fn))\n",
    "\n",
    "    # Generate generalization metrics of evaluation\n",
    "    scores = gpc_rbf_model.score(X[test], Y[test])\n",
    "    print(f'\\nScore for fold {fold_no}: - Accuracy: {scores} - Specifity: {specifity} - Sensivity: {sensivity}')\n",
    "\n",
    "    # Plot Confusion Matrix of each fold\n",
    "    ConfusionMatrixDisplay.from_predictions(Y[test], y_pred)\n",
    "    plt.title(f'Confusion Matrix - fold: {fold_no}', fontsize = 14)\n",
    "    plt.show()\n",
    "    \n",
    "    acc_per_fold.append(scores)\n",
    "    specifity_per_fold.append(specifity)\n",
    "    sensivity_per_fold.append(sensivity)\n",
    "\n",
    "    # Increase fold number\n",
    "    fold_no = fold_no + 1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# **Resume Evaluation Metric scores**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Resume scores per fold\n",
      "------------------------------------------------------------------------\n",
      "> Fold 1 - Accuracy: 0.8958333333333334 - Specificity: 0.9651300899471033 - Sensitivity: 0.8966033265775102\n",
      "------------------------------------------------------------------------\n",
      "> Fold 2 - Accuracy: 0.9036458333333334 - Specificity: 0.9680808193293254 - Sensitivity: 0.9041339128126039\n",
      "------------------------------------------------------------------------\n",
      "> Fold 3 - Accuracy: 0.9088541666666666 - Specificity: 0.9696403266077334 - Sensitivity: 0.9097523660428378\n",
      "------------------------------------------------------------------------\n",
      "> Fold 4 - Accuracy: 0.9114583333333334 - Specificity: 0.9705662345782741 - Sensitivity: 0.9144576632364372\n",
      "------------------------------------------------------------------------\n",
      "> Fold 5 - Accuracy: 0.8880208333333334 - Specificity: 0.9628048364803135 - Sensitivity: 0.8881511691691146\n",
      "------------------------------------------------------------------------\n",
      "Average scores for all folds:\n",
      "> Accuracy: 0.9015625\n",
      "> Specificity: 0.9672444613885499\n",
      "> Sensitivity: 0.9026196875677008\n"
     ]
    }
   ],
   "source": [
    "# == Provide average scores ==\n",
    "print('Resume scores per fold')\n",
    "for i in range(0, num_folds):\n",
    "  print('------------------------------------------------------------------------')\n",
    "  print(f'> Fold {i+1} - Accuracy: {acc_per_fold[i]} - Specificity: {specifity_per_fold[i]} - Sensitivity: {sensivity_per_fold[i]}')\n",
    "print('------------------------------------------------------------------------')\n",
    "print('Average scores for all folds:')\n",
    "print(f'> Accuracy: {np.mean(acc_per_fold)}')\n",
    "print(f'> Specificity: {np.mean(specifity_per_fold)}')\n",
    "print(f'> Sensitivity: {np.mean(sensivity_per_fold)}')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "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": "3.11.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}