swell-kw-stress-detection.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "acfab5d2",
   "metadata": {
    "papermill": {
     "duration": 0.005672,
     "end_time": "2024-02-01T18:31:50.073835",
     "exception": false,
     "start_time": "2024-02-01T18:31:50.068163",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "8d73f011",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-01T18:31:50.086513Z",
     "iopub.status.busy": "2024-02-01T18:31:50.085726Z",
     "iopub.status.idle": "2024-02-01T18:31:53.849490Z",
     "shell.execute_reply": "2024-02-01T18:31:53.848697Z"
    },
    "papermill": {
     "duration": 3.772555,
     "end_time": "2024-02-01T18:31:53.851853",
     "exception": false,
     "start_time": "2024-02-01T18:31:50.079298",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.10/site-packages/scipy/__init__.py:146: UserWarning: A NumPy version >=1.16.5 and <1.23.0 is required for this version of SciPy (detected version 1.24.3\n",
      "  warnings.warn(f\"A NumPy version >={np_minversion} and <{np_maxversion}\"\n"
     ]
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.metrics import classification_report, confusion_matrix, silhouette_samples, silhouette_score\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "import seaborn as sns\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3f14e9f1",
   "metadata": {
    "papermill": {
     "duration": 0.005245,
     "end_time": "2024-02-01T18:31:53.863072",
     "exception": false,
     "start_time": "2024-02-01T18:31:53.857827",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Analyze Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "fb832991",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-01T18:31:53.875521Z",
     "iopub.status.busy": "2024-02-01T18:31:53.874638Z",
     "iopub.status.idle": "2024-02-01T18:32:02.484229Z",
     "shell.execute_reply": "2024-02-01T18:32:02.483281Z"
    },
    "papermill": {
     "duration": 8.61778,
     "end_time": "2024-02-01T18:32:02.486313",
     "exception": false,
     "start_time": "2024-02-01T18:31:53.868533",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              MEAN_RR      MEDIAN_RR           SDRR          RMSSD  \\\n",
      "count   369289.000000  369289.000000  369289.000000  369289.000000   \n",
      "unique            NaN            NaN            NaN            NaN   \n",
      "top               NaN            NaN            NaN            NaN   \n",
      "freq              NaN            NaN            NaN            NaN   \n",
      "mean       846.650104     841.965890     109.352531      14.977498   \n",
      "std        124.603984     132.321005      77.117025       4.120766   \n",
      "min        547.492221     517.293295      27.233947       5.529742   \n",
      "25%        760.228533     755.750735      64.205641      11.830959   \n",
      "50%        822.951438     819.689595      82.608243      14.415918   \n",
      "75%        924.117422     916.821570     118.237002      17.927144   \n",
      "max       1322.016957    1653.122250     563.486949      26.629477   \n",
      "\n",
      "                 SDSD     SDRR_RMSSD             HR          pNN25  \\\n",
      "count   369289.000000  369289.000000  369289.000000  369289.000000   \n",
      "unique            NaN            NaN            NaN            NaN   \n",
      "top               NaN            NaN            NaN            NaN   \n",
      "freq              NaN            NaN            NaN            NaN   \n",
      "mean        14.976767       7.396597      73.941824       9.841143   \n",
      "std          4.120768       5.143834      10.337453       8.195574   \n",
      "min          5.529630       2.660381      48.737243       0.000000   \n",
      "25%         11.830671       4.541896      66.715776       3.666667   \n",
      "50%         14.415388       5.952112      74.217809       7.600000   \n",
      "75%         17.924839       7.919841      80.334937      13.333333   \n",
      "max         26.629392      54.523950     113.752309      39.400000   \n",
      "\n",
      "                pNN50            SD1            SD2           KURT  \\\n",
      "count   369289.000000  369289.000000  369289.000000  369289.000000   \n",
      "unique            NaN            NaN            NaN            NaN   \n",
      "top               NaN            NaN            NaN            NaN   \n",
      "freq              NaN            NaN            NaN            NaN   \n",
      "mean         0.866001      10.593708     154.178997       0.523235   \n",
      "std          0.990189       2.914795     109.170222       1.790348   \n",
      "min          0.000000       3.911344      38.307745      -1.894820   \n",
      "25%          0.000000       8.368340      90.326864      -0.352783   \n",
      "50%          0.466667      10.196621     116.221063       0.040736   \n",
      "75%          1.466667      12.679005     166.764850       0.722833   \n",
      "max          5.466667      18.836107     796.852945      64.088107   \n",
      "\n",
      "                 SKEW   MEAN_REL_RR  MEDIAN_REL_RR    SDRR_REL_RR  \\\n",
      "count   369289.000000  3.692890e+05  369289.000000  369289.000000   \n",
      "unique            NaN           NaN            NaN            NaN   \n",
      "top               NaN           NaN            NaN            NaN   \n",
      "freq              NaN           NaN            NaN            NaN   \n",
      "mean         0.041628 -1.756587e-06      -0.000465       0.018571   \n",
      "std          0.699522  1.630256e-04       0.000868       0.005455   \n",
      "min         -2.136278 -1.233914e-03      -0.004425       0.008987   \n",
      "25%         -0.359291 -7.281695e-05      -0.000917       0.014261   \n",
      "50%         -0.060966 -9.330777e-07      -0.000312       0.017318   \n",
      "75%          0.282417  6.911667e-05       0.000131       0.021827   \n",
      "max          6.777800  1.244098e-03       0.002095       0.036571   \n",
      "\n",
      "         RMSSD_REL_RR    SDSD_REL_RR  SDRR_RMSSD_REL_RR    KURT_REL_RR  \\\n",
      "count   369289.000000  369289.000000      369289.000000  369289.000000   \n",
      "unique            NaN            NaN                NaN            NaN   \n",
      "top               NaN            NaN                NaN            NaN   \n",
      "freq              NaN            NaN                NaN            NaN   \n",
      "mean         0.009701       0.009701           2.006817       0.523235   \n",
      "std          0.003897       0.003897           0.375845       1.790348   \n",
      "min          0.003220       0.003220           1.169342      -1.894820   \n",
      "25%          0.006984       0.006984           1.749801      -0.352783   \n",
      "50%          0.008691       0.008691           1.934416       0.040736   \n",
      "75%          0.011460       0.011460           2.221232       0.722833   \n",
      "max          0.026955       0.026955           3.724134      64.088107   \n",
      "\n",
      "          SKEW_REL_RR            VLF        VLF_PCT             LF  \\\n",
      "count   369289.000000  369289.000000  369289.000000  369289.000000   \n",
      "unique            NaN            NaN            NaN            NaN   \n",
      "top               NaN            NaN            NaN            NaN   \n",
      "freq              NaN            NaN            NaN            NaN   \n",
      "mean         0.041628    2199.580170      64.289242     946.530252   \n",
      "std          0.699522    1815.773422      16.774844     574.171780   \n",
      "min         -2.136278     159.480176      19.031219      90.048557   \n",
      "25%         -0.359291    1001.189280      52.909877     545.449386   \n",
      "50%         -0.060966    1667.903111      66.350237     782.716291   \n",
      "75%          0.282417    2654.121052      76.825032    1201.432256   \n",
      "max          6.777800   12617.977191      97.738848    3291.548112   \n",
      "\n",
      "               LF_PCT          LF_NU             HF         HF_PCT  \\\n",
      "count   369289.000000  369289.000000  369289.000000  369289.000000   \n",
      "unique            NaN            NaN            NaN            NaN   \n",
      "top               NaN            NaN            NaN            NaN   \n",
      "freq              NaN            NaN            NaN            NaN   \n",
      "mean        34.095182      95.566718      39.245603       1.615576   \n",
      "std         16.040290       4.123365      45.398869       1.761073   \n",
      "min          2.165119      69.879083       0.061783       0.002150   \n",
      "25%         22.305936      93.645734      10.720312       0.346803   \n",
      "50%         32.047025      96.643140      24.841938       1.039513   \n",
      "75%         44.647115      98.771946      45.272368       2.245115   \n",
      "max         77.928847      99.987175     364.486936      13.095664   \n",
      "\n",
      "                HF_NU             TP          LF_HF          HF_LF  \\\n",
      "count   369289.000000  369289.000000  369289.000000  369289.000000   \n",
      "unique            NaN            NaN            NaN            NaN   \n",
      "top               NaN            NaN            NaN            NaN   \n",
      "freq              NaN            NaN            NaN            NaN   \n",
      "mean         4.433282    3185.356025     115.977200       0.048506   \n",
      "std          4.123365    1923.227187     360.855129       0.049238   \n",
      "min          0.012825     377.692795       2.319952       0.000128   \n",
      "25%          1.228054    1828.147788      14.737458       0.012433   \n",
      "50%          3.356860    2796.856587      28.789747       0.034735   \n",
      "75%          6.354266    4052.260157      80.429614       0.067854   \n",
      "max         30.120917   13390.684098    7796.443096       0.431043   \n",
      "\n",
      "               sampen         higuci  datasetId  condition  \n",
      "count   369289.000000  369289.000000   369289.0     369289  \n",
      "unique            NaN            NaN        NaN          3  \n",
      "top               NaN            NaN        NaN  no stress  \n",
      "freq              NaN            NaN        NaN     200082  \n",
      "mean         2.062471       1.182292        2.0        NaN  \n",
      "std          0.206999       0.062192        0.0        NaN  \n",
      "min          0.434576       1.033984        2.0        NaN  \n",
      "25%          2.032977       1.139929        2.0        NaN  \n",
      "50%          2.134214       1.174293        2.0        NaN  \n",
      "75%          2.181929       1.223621        2.0        NaN  \n",
      "max          2.234841       1.361219        2.0        NaN  \n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def analyze_data(file_path):\n",
    "    df = pd.read_csv(file_path)\n",
    "    pd.set_option('display.max_columns', None)\n",
    "    print(df.describe(include='all'))\n",
    "\n",
    "    numeric_data = df.select_dtypes(include=['number'])\n",
    "    correlation_matrix = numeric_data.corr()\n",
    "    sns.heatmap(correlation_matrix, cmap='coolwarm')\n",
    "    plt.title('correlation_matrix')\n",
    "    plt.show()\n",
    "\n",
    "    condition_counts = df['condition'].value_counts()\n",
    "    labels = condition_counts.index\n",
    "    sizes = condition_counts.values\n",
    "    plt.pie(sizes, labels=labels, autopct='%1.1f%%')\n",
    "    plt.title('condition_counts')\n",
    "    plt.show()\n",
    "    \n",
    "analyze_data('../input/swell-heart-rate-variability-hrv/hrv dataset/data/final/train.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3e84a80c",
   "metadata": {
    "papermill": {
     "duration": 0.006761,
     "end_time": "2024-02-01T18:32:02.500165",
     "exception": false,
     "start_time": "2024-02-01T18:32:02.493404",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Supervised Learning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "66baa540",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-01T18:32:02.516288Z",
     "iopub.status.busy": "2024-02-01T18:32:02.514828Z",
     "iopub.status.idle": "2024-02-01T18:32:02.522221Z",
     "shell.execute_reply": "2024-02-01T18:32:02.521400Z"
    },
    "papermill": {
     "duration": 0.01722,
     "end_time": "2024-02-01T18:32:02.524204",
     "exception": false,
     "start_time": "2024-02-01T18:32:02.506984",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def preprocess_data_supervised(file_path):\n",
    "    df = pd.read_csv(file_path)\n",
    "\n",
    "    condition_encoding = {'no stress': 0, 'time pressure': 1, 'interruption': 2}\n",
    "    df['condition'] = df['condition'].map(condition_encoding)\n",
    "\n",
    "    features = df.iloc[:, :-2]\n",
    "    target = df.iloc[:, -1]\n",
    "\n",
    "    scaler = StandardScaler()\n",
    "    standardized_features = scaler.fit_transform(features)\n",
    "\n",
    "    pca = PCA(n_components=0.95, random_state=42)\n",
    "    pca_features = pca.fit_transform(standardized_features)\n",
    "\n",
    "    return pca_features, target"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "a9e19924",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-01T18:32:02.542774Z",
     "iopub.status.busy": "2024-02-01T18:32:02.541359Z",
     "iopub.status.idle": "2024-02-01T18:32:07.302114Z",
     "shell.execute_reply": "2024-02-01T18:32:07.300697Z"
    },
    "papermill": {
     "duration": 4.775048,
     "end_time": "2024-02-01T18:32:07.306207",
     "exception": false,
     "start_time": "2024-02-01T18:32:02.531159",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "X_train, y_train = preprocess_data_supervised('../input/swell-heart-rate-variability-hrv/hrv dataset/data/final/train.csv')\n",
    "X_test, y_test = preprocess_data_supervised('../input/swell-heart-rate-variability-hrv/hrv dataset/data/final/test.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "a413e6f8",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-01T18:32:07.339186Z",
     "iopub.status.busy": "2024-02-01T18:32:07.338604Z",
     "iopub.status.idle": "2024-02-01T18:32:07.349528Z",
     "shell.execute_reply": "2024-02-01T18:32:07.348422Z"
    },
    "papermill": {
     "duration": 0.030861,
     "end_time": "2024-02-01T18:32:07.352720",
     "exception": false,
     "start_time": "2024-02-01T18:32:07.321859",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def evaluate_classification(model, X_train, y_train, X_test, y_test, cv):\n",
    "    y_pred = model.predict(X_test)\n",
    "    report = classification_report(y_test, y_pred)\n",
    "    print(report)\n",
    "    \n",
    "    cv_scores = cross_val_score(model, X_train, y_train, cv=cv, scoring='accuracy')\n",
    "    print('cv_scores.mean:', f'{cv_scores.mean():.4f}')\n",
    "\n",
    "    conf_matrix = confusion_matrix(y_test, y_pred)\n",
    "    sns.heatmap(conf_matrix, annot=True, cmap='Blues')\n",
    "    plt.title('confusion_matrix')\n",
    "    plt.xlabel('Predicted Label')\n",
    "    plt.ylabel('True Label')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b1184b8f",
   "metadata": {
    "papermill": {
     "duration": 0.014244,
     "end_time": "2024-02-01T18:32:07.381961",
     "exception": false,
     "start_time": "2024-02-01T18:32:07.367717",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# (1) Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "e32ddb49",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-01T18:32:07.414747Z",
     "iopub.status.busy": "2024-02-01T18:32:07.413392Z",
     "iopub.status.idle": "2024-02-01T18:32:36.325811Z",
     "shell.execute_reply": "2024-02-01T18:32:36.324796Z"
    },
    "papermill": {
     "duration": 28.931232,
     "end_time": "2024-02-01T18:32:36.328058",
     "exception": false,
     "start_time": "2024-02-01T18:32:07.396826",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.63      0.87      0.73     22158\n",
      "           1       0.60      0.29      0.39      7093\n",
      "           2       0.52      0.30      0.38     11782\n",
      "\n",
      "    accuracy                           0.61     41033\n",
      "   macro avg       0.58      0.49      0.50     41033\n",
      "weighted avg       0.59      0.61      0.57     41033\n",
      "\n",
      "cv_scores.mean: 0.6018\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiwAAAHHCAYAAACcHAM1AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8WgzjOAAAACXBIWXMAAA9hAAAPYQGoP6dpAABspUlEQVR4nO3dd1QUVxsG8GdpS+9dqqAiEcESEbvRiCVGjDGxo8EONqzErlGMJfYaCxbUWCKxRcSuEbtYo1GDnaLSBOk73x9+rm4oArsbNvD8zpkT5t47M+9sEF5umREJgiCAiIiISIWplXcARERERB/DhIWIiIhUHhMWIiIiUnlMWIiIiEjlMWEhIiIilceEhYiIiFQeExYiIiJSeUxYiIiISOUxYSEiIiKVx4SFSIE2b94MNzc3aGpqwtjYWOHnnzZtGkQikcLPW5E8fPgQIpEIYWFh5R0KESkQExYiBblz5w769u0LFxcX/Pzzz1izZk15h/Sfd/DgQUybNq28wyAiFSDiu4SIFGPVqlUYMmQI7t27B1dXV6VcIy8vD3l5edDW1lbK+VVNUFAQli9fjtL8mBIEAdnZ2dDU1IS6uroSoyOif5NGeQdAVFEkJiYCgFKGgt7R0NCAhgb/2RYmLy8PEokEWlpalSahI6pMOCREFdazZ88QEBAAW1tbiMViODs7Y8iQIcjJyQEA/P333+jatStMTU2hq6uLhg0b4sCBAzLnOHHiBEQiEXbs2IFZs2bBzs4O2traaNWqFe7fvy9t5+TkhKlTpwIALCwsIBKJpEMZH379IScnJ/Tt21e6n5ubi+nTp6NatWrQ1taGmZkZmjRpgqioKGmbwuaw5OXlYebMmXBxcYFYLIaTkxO+//57ZGdnF7jeF198gTNnzqBBgwbQ1tZG1apVsWnTplJ/tiKRCEFBQdi5cyfc3d2ho6MDHx8f3LhxAwCwevVquLq6QltbGy1atMDDhw9ljj99+jS6du0KBwcHiMVi2NvbY9SoUcjMzJS26du3L5YvXy693rsNeD9PZf78+Vi0aJH03m/fvl1gDktiYiIsLCzQokULmZ6a+/fvQ09PD99++22p75+I/n38U40qpOfPn6NBgwZISUnBwIED4ebmhmfPnmHXrl148+YNkpOT0ahRI7x58wbDhw+HmZkZNm7ciC+//BK7du1C586dZc43Z84cqKmpYcyYMUhNTcXcuXPRs2dPnD9/HgCwaNEibNq0CXv27MHKlSuhr6+P2rVrlyrmadOmITQ0FP3790eDBg2QlpaGS5cu4cqVK/j888+LPK5///7YuHEjvv76a4wePRrnz59HaGgo/vzzT+zZs0em7f379/H1118jICAA/v7+WL9+Pfr27Yt69erhk08+KVW8p0+fxt69exEYGAgACA0NxRdffIFx48ZhxYoVGDp0KJKTkzF37lx89913OHbsmPTYnTt34s2bNxgyZAjMzMxw4cIFLF26FE+fPsXOnTsBAIMGDcLz588RFRWFzZs3FxrDhg0bkJWVhYEDB0IsFsPU1BQSiUSmjaWlJVauXImuXbti6dKlGD58OCQSCfr27QsDAwOsWLGiVPdNROVEIKqA+vTpI6ipqQkXL14sUCeRSISRI0cKAITTp09Ly1+/fi04OzsLTk5OQn5+viAIgnD8+HEBgFCzZk0hOztb2nbx4sUCAOHGjRvSsqlTpwoAhBcvXshcD4AwderUAnE4OjoK/v7+0n1PT0+hQ4cOxd7Xu2u8ExMTIwAQ+vfvL9NuzJgxAgDh2LFjMtcDIJw6dUpalpiYKIjFYmH06NHFXvefAAhisViIjY2Vlq1evVoAIFhbWwtpaWnS8pCQEAGATNs3b94UOGdoaKggEomER48eScsCAwOFwn5MxcbGCgAEQ0NDITExsdC6DRs2yJR3795d0NXVFf766y9h3rx5AgAhIiKiVPdNROWHQ0JU4UgkEkRERKBjx46oX79+gXqRSISDBw+iQYMGaNKkibRcX18fAwcOxMOHD3H79m2ZY/r16wctLS3pftOmTQG8HVZSFGNjY9y6dQv37t0r8TEHDx4EAAQHB8uUjx49GgAKDHG5u7tLYwfeDl/VqFGjTPfRqlUrODk5Sfe9vb0BAF26dIGBgUGB8g+voaOjI/06IyMDL1++RKNGjSAIAq5evVriGLp06QILC4sStV22bBmMjIzw9ddfY/Lkyejduzc6depU4msRUfliwkIVzosXL5CWloZatWoV2ebRo0eoUaNGgfKaNWtK6z/k4OAgs29iYgIASE5OljdcqRkzZiAlJQXVq1eHh4cHxo4di+vXrxd7zKNHj6CmplZgVZK1tTWMjY0/eh/A23spy33881xGRkYAAHt7+0LLP7zG48eP0bdvX5iamkJfXx8WFhZo3rw5ACA1NbXEMTg7O5e4rampKZYsWYLr16/DyMgIS5YsKfGxRFT+mLAQlUBRy2MFOZ4KkJ+fL7PfrFkzPHjwAOvXr0etWrWwdu1a1K1bF2vXrv3ouUr6MDlF3kdR5/rYNfLz8/H555/jwIEDGD9+PCIiIhAVFSWdJPvPOSjF+bCnpiQiIyMBvE2enj59Wqpjiah8MWGhCsfCwgKGhoa4efNmkW0cHR1x9+7dAuV37tyR1iuKiYkJUlJSZMpycnIQFxdXoK2pqSn69euHbdu24cmTJ6hdu3axD05zdHSERCIpMIyUkJCAlJQUhd6Hoty4cQN//fUXFixYgPHjx6NTp05o3bo1bG1tC7RV5FN9Dx06hLVr12LcuHGwsLCAv78/8vLyFHZ+IlIuJixU4aipqcHPzw/79u3DpUuXCtQLgoD27dvjwoULiI6OlpZnZGRgzZo1cHJygru7u8LicXFxwalTp2TK1qxZU6CH5dWrVzL7+vr6cHV1LbA8+UPt27cH8HaV0od++uknAECHDh3KGrbSvOuB+bBXRxAELF68uEBbPT09ACiQ8JVWSkqKdPXV7NmzsXbtWly5cgWzZ8+W67xE9O/hsmaqkGbPno3Dhw+jefPmGDhwIGrWrIm4uDjs3LkTZ86cwYQJE7Bt2za0a9cOw4cPh6mpKTZu3IjY2Fjs3r0bamqKy+X79++PwYMHo0uXLvj8889x7do1REZGwtzcXKadu7s7WrRogXr16sHU1BSXLl3Crl27EBQUVOS5PT094e/vjzVr1iAlJQXNmzfHhQsXsHHjRvj5+aFly5YKuw9FcXNzg4uLC8aMGYNnz57B0NAQu3fvLnQeTb169QAAw4cPh6+vL9TV1dGtW7dSX3PEiBF49eoVjhw5AnV1dbRt2xb9+/fHDz/8gE6dOsHT01Pu+yIi5WLCQhVSlSpVcP78eUyePBnh4eFIS0tDlSpV0K5dO+jq6sLY2Bhnz57F+PHjsXTpUmRlZaF27drYt2+fwnslBgwYgNjYWKxbtw6HDh1C06ZNERUVhVatWsm0Gz58OPbu3YvDhw8jOzsbjo6O+OGHHzB27Nhiz7927VpUrVoVYWFh2LNnD6ytrRESEiJ9kJ2q0dTUxL59+zB8+HCEhoZCW1sbnTt3RlBQUIHE4auvvsKwYcOwfft2bNmyBYIglDph2bt3LzZt2oQFCxbAzc1NWv7TTz8hKioK/v7+uHjxIjQ1NRVyf0SkHHyXEBEREak8zmEhIiIilcchISKSio+PL7ZeR0dH+lwVIqJ/E4eEiEjqY8uI/f39pc9LISL6N7GHhYikPnwzdGEKe1YKEdG/gT0sREREpPI46ZaIiIhUHhMWIiIiUnkVcg6LTp2inwxKlVPyxWXlHQKpkMS0ol93QJWPg6lY6ddQ1O+lzKuV92cZe1iIiIhI5VXIHhYiIiKVImL/gLyYsBARESnbR55xRB/HhIWIiEjZ2MMiN36CREREpPLYw0JERKRsHBKSGxMWIiIiZeOQkNz4CRIREZHKYw8LERGRsnFISG5MWIiIiJSNQ0Jy4ydIREREKo89LERERMrGISG5MWEhIiJSNg4JyY2fIBEREak89rAQEREpG4eE5MaEhYiISNk4JCQ3JixERETKxh4WuTHlIyIiIpXHHhYiIiJl45CQ3JiwEBERKRsTFrnxEyQiIiKVxx4WIiIiZVPjpFt5MWEhIiJSNg4JyY2fIBEREak89rAQEREpG5/DIjcmLERERMrGISG58RMkIiIilcceFiIiImXjkJDcmLAQEREpG4eE5MZPkIiISNlEIsVspXTq1Cl07NgRtra2EIlEiIiI+EdYokK3efPmSds4OTkVqJ8zZ47Mea5fv46mTZtCW1sb9vb2mDt3boFYdu7cCTc3N2hra8PDwwMHDx4s1b0wYSEiIqqgMjIy4OnpieXLlxdaHxcXJ7OtX78eIpEIXbp0kWk3Y8YMmXbDhg2T1qWlpaFNmzZwdHTE5cuXMW/ePEybNg1r1qyRtjl79iy6d++OgIAAXL16FX5+fvDz88PNmzdLfC8cEiIiIlK2choSateuHdq1a1dkvbW1tcz+b7/9hpYtW6Jq1aoy5QYGBgXavhMeHo6cnBysX78eWlpa+OSTTxATE4OffvoJAwcOBAAsXrwYbdu2xdixYwEAM2fORFRUFJYtW4ZVq1aV6F7Yw0JERKRsChoSys7ORlpamsyWnZ2tkBATEhJw4MABBAQEFKibM2cOzMzMUKdOHcybNw95eXnSuujoaDRr1gxaWlrSMl9fX9y9exfJycnSNq1bt5Y5p6+vL6Kjo0scHxMWIiKi/4jQ0FAYGRnJbKGhoQo598aNG2FgYICvvvpKpnz48OHYvn07jh8/jkGDBmH27NkYN26ctD4+Ph5WVlYyx7zbj4+PL7bNu/qS4JAQERGRsiloSCgkJATBwcEyZWKxWCHnXr9+PXr27AltbW2Z8g+vV7t2bWhpaWHQoEEIDQ1V2LVLggkLERGRsinoOSxisVgpScLp06dx9+5d/PLLLx9t6+3tjby8PDx8+BA1atSAtbU1EhISZNq8238376WoNkXNiykMh4SIiIgquXXr1qFevXrw9PT8aNuYmBioqanB0tISAODj44NTp04hNzdX2iYqKgo1atSAiYmJtM3Ro0dlzhMVFQUfH58Sx8geFiIiImUrp1VC6enpuH//vnQ/NjYWMTExMDU1hYODA4C3y5J37tyJBQsWFDg+Ojoa58+fR8uWLWFgYIDo6GiMGjUKvXr1kiYjPXr0wPTp0xEQEIDx48fj5s2bWLx4MRYuXCg9z4gRI9C8eXMsWLAAHTp0wPbt23Hp0iWZpc8fw4SFiIhI2copYbl06RJatmwp3X83H8Xf3x9hYWEAgO3bt0MQBHTv3r3A8WKxGNu3b8e0adOQnZ0NZ2dnjBo1SmZei5GREQ4fPozAwEDUq1cP5ubmmDJlinRJMwA0atQIW7duxaRJk/D999+jWrVqiIiIQK1atUp8LyJBEITSfgCqTqdOUHmHQCom+eKy8g6BVEhimmKWgVLF4GCq/ImjOh1XKOQ8mfuGKuQ8/0XsYSEiIlI2vvxQbkxYiIiIlI0vP5QbExYiIiJlYw+L3JjyERERkcpjDwsREZGycUhIbkxYiIiIlI1DQnJjykdEREQqjz0sRERESiZiD4vcmLAQEREpGRMW+XFIiIiIiFQee1iIiIiUjR0scmPCQkREpGQcEpIfh4SIiIhI5bGHhYiISMnYwyI/JixERERKxoRFfkxY/kWN67pgVJ/WqOvuABsLI3wzag32nbhe7DGDvmmGwd82g6OtKZ7EJ+PHdZHYuv+CUuNsWq8afhz9FdxdrPE0PgVz1h7Cln3nC207pt/nmDm8E5aFH8fY+buVGldFs2P7Vuz4ZRueP3sGAHBxrYZBQ4aiSdPmhbbPzc3Fup9XY9/eCCQmJMDJyRkjg8egcdNmSo3zcOTvWL50MZ4/ewYHRyeMDB6Dps2aS2NatmQRzpw+hadPn8BAXx/ePo0wYtRoWFpaKTWuimbbxrU4c/IonjyKhVgshruHF/oPHQl7R+cij4k88Bvm/zBZpkxTSwsHT15Saqy/7dqOneFhSEp6CRfX6ggMDoHbJx7S+kVzZuDKpXN49eIFdHR14e7hif5DR8HBqeh7qeiYsMiPc1j+RXo6Ytz46xlGhv5SovYDujbBjGEdMWv1QdT9ehZ+WHUQiyZ8g/bNapU5BgcbU2ReXVZkvaOtGfYsHYxTl/6Cd7c5WLb1OFZO6YHWPjULtK3n7oCALo1x/a+nZY6nMrO0ssaIUWOwbeev2LpjNxp4N8SIoEDcv3+v0PbLlizCrp2/YML3k7Fn70F0/bYbRo0Iwp9/3i5zDBcvnEe7zz8rsj7m6hVMGDsanb/6Gr/sikDLz1ph5LBA3Lv3FwAgKysLd/68jYGDh+CXnb/ip8XL8DA2FiOChpQ5psrq+tVL+LJLNyz5eQvmLF6DvLw8TBg5GJmZb4o9TldPH7/sPybdwvdEyhVH5IHfMHrod0XWnzhyCKuXzEOvgMFYGfYLqlargZBRg5Gc9ErappqbO8ZMnIF12yMQumglBEHAhJGDkJ+fL1dsVLkxYfkXHf7jNqav2I+9x4vvVXmnR4cGWLf7D+w6fAUPn73CzsjLWP/rHxjd93OZdn07++Dq7klIPrcQMb9OwsCuTcsc44Cvm+Dhs1eY8NMe3I1NwKpfTmHP0RgM69lSpp2ejhY2zO6LoTO3ISUts8zXq8xatPwMTZs1h6OjE5ycnDFsxCjo6uri+rWYQtsf2Pcb+g8YjKbNmsPO3h7fdOuBJk2bY1PYemkbiUSCdT+vRrs2n6FB3dro2vlLREUeKnOM4Vs2oVGTpuj7XX9UdXFB0PCRqOnuju1btwAADAwMsHrtBvi2bQ8n56qo7emFkImTcfvWLcQ9f17m61ZGoYtWwbdDJzhVdYVLtRoYO2kmEuPjcO9O8QmpSCSCqZm5dDMxNZOpz8nJweol89GtY2t0bNkAwwJ64NqVi2WOc/e2TWj3ZRe0/cIPjs4uGDFuMsRiHUTuj5C26eD3NWrXqQ9rmyqoVsMd/QYNw4uEeCTEVeLvCZGCtkqsXIeEXr58ifXr1yM6Ohrx8fEAAGtrazRq1Ah9+/aFhYVFeYZX7rQ0NZCVkytTlpmVi/q1HKGhoYa8PAm6tauPKUO+wKg5OxBz5ym83OywfHIPZGTlILyIYZzieHs64/j5uzJlUWf/xLwxXWTKFoV8i0Onb+L4+buY0L9t6W+OZOTn5+Nw5CFkZr6Bp2edQtvk5ORCS6wlUybWFiPmyhXp/rqfV+PA/r2YNGU6HB2dcPnSRXw/YSxMTE1R/9MGpY7rekwMevv3lSlr1LgJjh89UuQx6enpEIlEMDA0LPX16L2M9HQAgIGhUbHtMjPfoGdnXwgSCVxr1MR3g4fDqaqrtH7Zgtl4FPs3Js78EWbmlvjj5FGEjBqCNVt2w87esVQx5ebm4q+7f6Jbn/7SMjU1NdT91Bu3b14rMr7I/RGwtq0CCyvrUl2vIuGQkPzKLWG5ePEifH19oauri9atW6N69eoAgISEBCxZsgRz5sxBZGQk6tevX14hlrsj0X+ir18j7Dt+HVf/fIK67g7o27kRtDQ1YG6sj/iXaZg0uAMm/PQrfjv29ofFo+ev4FbVGv27NC5TwmJlZoiEpNcyZYlJaTAy0IG2WBNZ2bno6lsPXm72aNJrrkLuszK799dd9O7RDTk52dDV1cXCJcvh4upaaNtGjZtg88Yw1Kv/KeztHXD+XDSOHYmSdrPn5ORg7c+rsWbtBnh6vU167OztcfXqZeza8UuZEpaXL1/CzMxcpszMzAwvX70stH12djYW/TQf7dp3gL6+fqmvR29JJBKsXDQXn9SuA2eXakW2s3dwwpjvp8PZtToy0tOxa2sYRgzsg7Vbf4WFpTUS4+MQeeA3hO+JhLmFJQCga8++uHjuD0Tuj0DAkBGliis1JRmS/PwCvTgmpmZ48ihWpmzv7u34eflCZGVmwt7BCT8uXgNNTc1SXY/oQ+WWsAwbNgxdu3bFqlWrCmSegiBg8ODBGDZsGKKjo4s9T3Z2NrKzs2WPl+RDpKau8Jj/baE/H4KVmSFObhwDkQhITHqN8H3nMbrf55BIBOhqa8HFwQIrp/TE8sk9pMdpqKshNf39MM3lXRPhYGMK4P0bzl/8sUBa/8fV+/ALWlmimOysjDFvbBd8MWQZsnPyFHCXlZuTkzN27I5AevprRB2OxOTvx2Nd2JZCk5ZxIRMxY+ok+H3RDiKRCHb29ujk9xUi9ryd7Pz48SNkZWZiUH/Z+Qe5ublwq/l+DlLD+u97cCSSfOTk5MiUdejYEZOnzij1veTm5mJs8AgIgoCJU6aX+nh6b+n8WXj4930sXB1WbDt3D0+4e3hK9z+p7YmAbn44sGcX+g4KQuyDe5Dk56Pftx1ljsvNyYWhkTEAIDE+DgE9/KR1+fn5yM/LQ8fPvKVl3fv0R4++A0p1D618O6BuAx8kvXyBnVs34odJY7Bo9SZoicWlOk9FwR4W+ZVbwnLt2jWEhYUV+j9RJBJh1KhRqFOn8K7xD4WGhmL6dNkfjupWn0LTpvR/TaqarOxcDJ4ejqBZ22Blaoi4l6kI6NIYaemZeJGcDguTt3/BBs7cigs3H8ocm58vSL/uPGwFNDTeJnC2lsaIWjsS3t1C318n6/2wU8KrNFiZGsicy9LUEKmvM5GVnYs6NR1gZWaI6K3jpfUaGupoUtcFg79tBiPvkZBIBFDJaGppwcHxbbe8+ye1cOvmDYRv2YQp0womDKampli0dAWys7ORkpICS0tLLPppPqrY2QMA3rx5Ozlz2crVBVboaGm9H0rasTtC+vWNG9ew6Kf5WLdhs7RM74OeEXNzc7z6R2/Kq1evYP6PXpfc3FyMHT0Scc+f4+cNG9m7Ioel82fj/B+nsGDlBlhYlm4IRUNDEy7V3fDs2WMAb4dj1NTVsWLDdqj94484HV1dAICZuQVWbdwpLT9z8gjOHD+CCdPmSMveDUsZGZtATV1dZoItACQnvYLJP74n9PQNoKdvADt7R9Ss5Ymv2jTGmZNH8Vmb9qW6p4qCCYv8yi1hsba2xoULF+Dm5lZo/YULF2Bl9fFlkSEhIQgODpYps2w6vojW/015eRI8S0wBAHT1rYffT9+CIAhITHqN54kpcLIzx/bfi17G+DguWeZcAPD3k8K79M9fi4Vvk09kylo1dMP562+7e49fuIt6X8+SqV8zvRfuxiZgQVgUkxU5SSQS5ObkFNtGLBbDysoKubm5OBp1GG3atgMAuLi4QEtLC3Fxz4sd/nmXIAFAQkI8NNQ1ZMo+VNvLC+fPnUOvPn2lZeeiz6K2l5d0/12y8vjRI6zdsAnGxiYluFP6J0EQsGxBKP44eQzzV6yDja1dqc+Rn5+Phw/uoUGjtxPvXau7QZKfj5TkJHh41Sv0GHUNDVSxd5DuG5uYQUusLVP2jqamJqrXqImrl86jcfO3q8skEgmuXjqPTl93L/beBOHt9wpRWZVbwjJmzBgMHDgQly9fRqtWraTJSUJCAo4ePYqff/4Z8+fP/+h5xGIxxP/oYlTV4SA9HS242L+fSOxUxQy1q1dBctobPIlPxoxhX8LW0gj9J7/9a9fVwRL1azni4s2HMDHQxfDen8HdxVZaDwAzVx3AgrFdkZaeicN//AmxlgbqujvAxFAXS7YcK3WMP+86g8HdmmHWiE7Y+Ns5tPi0Orp8Xgedh68CAKS/ycbtB3Eyx2Rk5iApNaNAORVv8cIFaNK0GaxtbPAmIwMHD+zHpYsXsHLNOgDAxJBxsLS0wohRowEA169fQ2JCAtzcaiIxMQErly+FRJCg73dvJ0Dq6enDv+93mP9jKASJgDp16yE9/TWuXr0CfT19fOnXudQx9uzVBwF9e2Nj2Ho0a9Ych34/iFs3b2Ly/3uAcnNzMWbUcPz5520sXb4akvx8vHzxAgBgZGQETS2t4k5PH1g6fxaOHf4d039cDF1dPST9v2dLT08fYm1tAMCP07+HuYUVAoa+nXuyed0q1KxVG1XsHJCenoYd4WFIiI9Duy+/AgDYOTihlW8HzJ0xEQOHj4FrdTekJifj6qXzqOpaHd6NS/8Mny7d+2DuzEmo7uaOGp94YM/2LcjKyoTvF34AgLhnT3HiyCHU824EY2MTvEhMwPbN66AlFqOBTxMFfFL/TexhkV+5JSyBgYEwNzfHwoULsWLFCunEQXV1ddSrVw9hYWH45ptvyis8pajr7ojDa99Pcpv7/5U3m/eew8CpW2Btbgh7a1Npvbq6CCN6f4bqjlbIzcvHqUt/oWXfBXgclyRtE7YnGpmZuRjp3wqzR/ohIzMHt+4/x7Lw42WK8dHzV+g8bBXmjvkKgT1a4FlCCobM2Ioj0X+W8a6pKElJrzApZDxevEiEvoEBqlevgZVr1sGnUWMAQHxcHNRE7588kJOdjeVLFuHp0yfQ1dVFk2bNMWvOXBh+sBoncPhImJiaYt3a1Xj65CkMDA1Qs6Y7+g8cXKYYverURejc+Vi2ZBGWLvoJDo5OWLR0OapVeztJPjExASeOv02Mv+nSSebYtRs24dMG3gXOSYXb9+sOAMCYQNk5SGMmzYRvh7efbWJCPERq778n0l+nYeGc6Uh+9RL6Boao5uaOxWs2wdHZ5YPjZyB8wxqsWTIfL18kwtDYBDU/qY2GZUhWAKBF67ZISU7GxrUrkPzqJVyq1cDshSulE3E1tbRw49oV/PrLFqS/ToOJqRk8vOph8ZpNBSbrVirMV+QmEgSh3Pvwc3Nz8fLl278mzM3N5Z5JrlMnSBFhUQWSfLHoh+VR5ZOYlv3xRlRpOJgqfyKwmf82hZzn1caih94qOpV4NL+mpiZsbGzKOwwiIiKl4JCQ/FQiYSEiIqrImLDIjwkLERGRkjFhkR/fJUREREQqjz0sREREysYOFrkxYSEiIlIyDgnJj0NCREREpPLYw0JERKRk7GGRHxMWIiIiJWPCIj8OCREREZHKYw8LERGRkrGHRX5MWIiIiJSN+YrcOCRERERUQZ06dQodO3aEra0tRCIRIiIiZOr79u0LkUgks7Vt21amTVJSEnr27AlDQ0MYGxsjICAA6enpMm2uX7+Opk2bQltbG/b29pg7d26BWHbu3Ak3Nzdoa2vDw8MDBw8eLNW9MGEhIiJSsn8mBWXdSisjIwOenp5Yvnx5kW3atm2LuLg46bZtm+ybpXv27Ilbt24hKioK+/fvx6lTpzBw4EBpfVpaGtq0aQNHR0dcvnwZ8+bNw7Rp07BmzRppm7Nnz6J79+4ICAjA1atX4efnBz8/P9y8ebPE9yISBEEoxb3/J+jUCSrvEEjFJF9cVt4hkApJTMsu7xBIhTiYipV+DbuhEQo5z9MVfmU+ViQSYc+ePfDze3+Ovn37IiUlpUDPyzt//vkn3N3dcfHiRdSvXx8AcOjQIbRv3x5Pnz6Fra0tVq5ciYkTJyI+Ph5aWloAgAkTJiAiIgJ37twBAHz77bfIyMjA/v37pedu2LAhvLy8sGrVqhLFzx4WIiIiJVNUD0t2djbS0tJktuxs+RLwEydOwNLSEjVq1MCQIUPw6tUraV10dDSMjY2lyQoAtG7dGmpqajh//ry0TbNmzaTJCgD4+vri7t27SE5OlrZp3bq1zHV9fX0RHR1d4jiZsBAREf1HhIaGwsjISGYLDQ0t8/natm2LTZs24ejRo/jxxx9x8uRJtGvXDvn5+QCA+Ph4WFpayhyjoaEBU1NTxMfHS9tYWVnJtHm3/7E27+pLgquEiIiIlE1Bq4RCQkIQHBwsUyYWl31Iq1u3btKvPTw8ULt2bbi4uODEiRNo1apVmc+rDExYiIiIlExRz2ERi8VyJSgfU7VqVZibm+P+/fto1aoVrK2tkZiYKNMmLy8PSUlJsLa2BgBYW1sjISFBps27/Y+1eVdfEhwSIiIiIgDA06dP8erVK9jY2AAAfHx8kJKSgsuXL0vbHDt2DBKJBN7e3tI2p06dQm5urrRNVFQUatSoARMTE2mbo0ePylwrKioKPj4+JY6NCQsREZGSldey5vT0dMTExCAmJgYAEBsbi5iYGDx+/Bjp6ekYO3Yszp07h4cPH+Lo0aPo1KkTXF1d4evrCwCoWbMm2rZtiwEDBuDChQv4448/EBQUhG7dusHW1hYA0KNHD2hpaSEgIAC3bt3CL7/8gsWLF8sMXY0YMQKHDh3CggULcOfOHUybNg2XLl1CUFDJV/UyYSEiIlKy8kpYLl26hDp16qBOnToAgODgYNSpUwdTpkyBuro6rl+/ji+//BLVq1dHQEAA6tWrh9OnT8sMO4WHh8PNzQ2tWrVC+/bt0aRJE5lnrBgZGeHw4cOIjY1FvXr1MHr0aEyZMkXmWS2NGjXC1q1bsWbNGnh6emLXrl2IiIhArVq1Sv4Z8jksVBnwOSz0IT6HhT70bzyHxWnE/o83KoGHi79QyHn+izjploiISMn48kP5MWEhIiJSNuYrcuMcFiIiIlJ57GEhIiJSMg4JyY8JCxERkZIxYZEfExYiIiIlY74iP85hISIiIpXHHhYiIiIl45CQ/JiwEBERKRnzFflxSIiIiIhUHntYiIiIlIxDQvJjwkJERKRkzFfkxyEhIiIiUnnsYSEiIlIyNTV2sciLCQsREZGScUhIfhwSIiIiIpXHHhYiIiIl4yoh+TFhISIiUjLmK/JjwkJERKRk7GGRH+ewEBERkcpjDwsREZGSsYdFfkxYiIiIlIz5ivw4JEREREQqjz0sRERESsYhIfkxYSEiIlIy5ivy45AQERERqTz2sBARESkZh4Tkx4SFiIhIyZivyI9DQkRERKTy2MNCRESkZBwSkh8TFiIiIiVjviI/JixERERKxh4W+XEOCxEREam8CtnDEv1baHmHQComOSOnvEMgFWKqp1XeIVAlww4W+VXIhIWIiEiVcEhIfhwSIiIiIpXHHhYiIiIlYweL/JiwEBERKRmHhOTHISEiIqIK6tSpU+jYsSNsbW0hEokQEREhrcvNzcX48ePh4eEBPT092Nraok+fPnj+/LnMOZycnCASiWS2OXPmyLS5fv06mjZtCm1tbdjb22Pu3LkFYtm5cyfc3Nygra0NDw8PHDx4sFT3woSFiIhIyUQixWyllZGRAU9PTyxfvrxA3Zs3b3DlyhVMnjwZV65cwa+//oq7d+/iyy+/LNB2xowZiIuLk27Dhg2T1qWlpaFNmzZwdHTE5cuXMW/ePEybNg1r1qyRtjl79iy6d++OgIAAXL16FX5+fvDz88PNmzdLfC8iQRCEUt6/yot5/Lq8QyAVY2UkLu8QSIUYaGuWdwikQvTFyh+uabrgjELOc3p0kzIfKxKJsGfPHvj5+RXZ5uLFi2jQoAEePXoEBwcHAG97WEaOHImRI0cWeszKlSsxceJExMfHQ0vr7SMDJkyYgIiICNy5cwcA8O233yIjIwP79++XHtewYUN4eXlh1apVJYqfPSxERET/EdnZ2UhLS5PZsrOzFXb+1NRUiEQiGBsby5TPmTMHZmZmqFOnDubNm4e8vDxpXXR0NJo1ayZNVgDA19cXd+/eRXJysrRN69atZc7p6+uL6OjoEsfGhIWIiEjJ/jkHpKxbaGgojIyMZLbQUMU8LDUrKwvjx49H9+7dYWhoKC0fPnw4tm/fjuPHj2PQoEGYPXs2xo0bJ62Pj4+HlZWVzLne7cfHxxfb5l19SXCVEBERkZIpapFQSEgIgoODZcrEYvmHvHNzc/HNN99AEASsXLlSpu7D69WuXRtaWloYNGgQQkNDFXLtkmLCQkREpGSKWtYsFosVniS8S1YePXqEY8eOyfSuFMbb2xt5eXl4+PAhatSoAWtrayQkJMi0ebdvbW0t/W9hbd7VlwSHhIiIiCqpd8nKvXv3cOTIEZiZmX30mJiYGKipqcHS0hIA4OPjg1OnTiE3N1faJioqCjVq1ICJiYm0zdGjR2XOExUVBR8fnxLHyh4WIiIiJSuv58alp6fj/v370v3Y2FjExMTA1NQUNjY2+Prrr3HlyhXs378f+fn50jklpqam0NLSQnR0NM6fP4+WLVvCwMAA0dHRGDVqFHr16iVNRnr06IHp06cjICAA48ePx82bN7F48WIsXLhQet0RI0agefPmWLBgATp06IDt27fj0qVLMkufP4bLmqlS4LJm+hCXNdOH/o1lzZ8tKflqmOIcG17yHgkAOHHiBFq2bFmg3N/fH9OmTYOzs3Ohxx0/fhwtWrTAlStXMHToUNy5cwfZ2dlwdnZG7969ERwcLDM0df36dQQGBuLixYswNzfHsGHDMH78eJlz7ty5E5MmTcLDhw9RrVo1zJ07F+3bty/xvTBhoUqBCQt9iAkLfagiJywVCYeEiIiIlIyvEpIfExYiIiIlU2PGIjeuEiIiIiKVxx4WIiIiJWMHi/yYsBARESmZoh4cV5kxYSEiIlIyNeYrcuMcFiIiIlJ57GEhIiJSMg4JyY8JCxERkZIxX5Efh4SIiIhI5bGHhYiISMlEYBeLvJiwEBERKRlXCcmvRAnL9evXS3zC2rVrlzkYIiIiosKUKGHx8vKCSCRCUS92flcnEomQn5+v0ACJiIj+67hKSH4lSlhiY2OVHQcREVGFxXxFfiVKWBwdHZUdBxEREVGRyrSsefPmzWjcuDFsbW3x6NEjAMCiRYvw22+/KTQ4IiKiikBNJFLIVpmVOmFZuXIlgoOD0b59e6SkpEjnrBgbG2PRokWKjo+IiOg/TyRSzFaZlTphWbp0KX7++WdMnDgR6urq0vL69evjxo0bCg2OiIioIhCJRArZKrNSJyyxsbGoU6dOgXKxWIyMjAyFBEVERET0oVInLM7OzoiJiSlQfujQIdSsWVMRMREREVUoHBKSX6mfdBscHIzAwEBkZWVBEARcuHAB27ZtQ2hoKNauXauMGImIiP7TKvuEWUUodcLSv39/6OjoYNKkSXjz5g169OgBW1tbLF68GN26dVNGjERERFTJleldQj179kTPnj3x5s0bpKenw9LSUtFxERERVRjsX5FfmV9+mJiYiLt37wJ4O/vZwsJCYUERERFVJJV9hY8ilHrS7evXr9G7d2/Y2tqiefPmaN68OWxtbdGrVy+kpqYqI0YiIiKq5EqdsPTv3x/nz5/HgQMHkJKSgpSUFOzfvx+XLl3CoEGDlBEjERHRf5qaSDFbZVbqIaH9+/cjMjISTZo0kZb5+vri559/Rtu2bRUaHBERUUXAISH5lbqHxczMDEZGRgXKjYyMYGJiopCgiIiIiD5U6oRl0qRJCA4ORnx8vLQsPj4eY8eOxeTJkxUaHBERUUXAB8fJr0RDQnXq1JHpzrp37x4cHBzg4OAAAHj8+DHEYjFevHjBeSxERET/wCEh+ZUoYfHz81NyGERERBVXZZ8wqwglSlimTp2q7DiIiIiIilTmB8cRERFRyXBISH6lTljy8/OxcOFC7NixA48fP0ZOTo5MfVJSksKCIyIiqgiYrsiv1KuEpk+fjp9++gnffvstUlNTERwcjK+++gpqamqYNm2aEkIkIiKiyq7UCUt4eDh+/vlnjB49GhoaGujevTvWrl2LKVOm4Ny5c8qIkYiI6D9NTSRSyFaZlTphiY+Ph4eHBwBAX19f+v6gL774AgcOHFBsdERERBUAn8Miv1InLHZ2doiLiwMAuLi44PDhwwCAixcvQiwWKzY6IiIiKrNTp06hY8eOsLW1hUgkQkREhEy9IAiYMmUKbGxsoKOjg9atW+PevXsybZKSktCzZ08YGhrC2NgYAQEBSE9Pl2lz/fp1NG3aFNra2rC3t8fcuXMLxLJz5064ublBW1sbHh4eOHjwYKnupdQJS+fOnXH06FEAwLBhwzB58mRUq1YNffr0wXfffVfa0xEREVV4IpFIIVtpZWRkwNPTE8uXLy+0fu7cuViyZAlWrVqF8+fPQ09PD76+vsjKypK26dmzJ27duoWoqCjs378fp06dwsCBA6X1aWlpaNOmDRwdHXH58mXMmzcP06ZNw5o1a6Rtzp49i+7duyMgIABXr16Fn58f/Pz8cPPmzZJ/hoIgCKX+BD5w7tw5nD17FtWqVUPHjh3lOZXCxDx+Xd4hfFTE9jBsW7cM7Tp3R9+howttc/70MURs24D450+Qn58Ha1sHfPF1TzT7vINSY4v8bQf27dyMlKRXcHSphn6BY+HqVktav2bRLNy8cgFJr15CW0cHNdxro0f/4aji4KTUuORhZaR6vX/hYWtx6vgRPH4UC7FYG594eGLQsFFwcHQu8pjYB/exYc1y3L1zGwlxzxE4ahy6du+t9FhPHInEutXLEB/3HHb2DhgUNAoNGzeT1m9YswLHon7Hi4QEaGhqoLqbO/oPGQ73WrWVHltZGGhrlncIBXzR9jPEPX9eoLzrtz0wYeKUQo95nZaG5UsX4djRKKSlpsDG1hajx32PJk2bKy3OqMOHsHLZYsQ9fwZ7B0cMHzVGer3c3FysXLYYZ06fxLOnT6FvoA9v70YYNjIYFpZWSotJXvpi5Y+1DNp1SyHnWf31J2U+ViQSYc+ePdKHwQqCAFtbW4wePRpjxowBAKSmpsLKygphYWHo1q0b/vzzT7i7u+PixYuoX78+AODQoUNo3749nj59CltbW6xcuRITJ05EfHw8tLS0AAATJkxAREQE7ty5AwD49ttvkZGRgf3790vjadiwIby8vLBq1aoSxV/qHpZ/atiwIYKDg+Ht7Y3Zs2fLe7pK4f7dWzhy4Fc4VK1WbDt9Q0N07vEdZi7egLmrt6OFb0esnD8DMRejy3ztE5H7MH30wCLrz544jE2rF6JLrwGYs3ILHKtWx+yQYUhNfr9cvWq1mhg8Zip+WrcT34cugyAImDUhEJL8/DLHVRnFXLkEv67dsGJdOOYvXYP8/DyMHTYImZlvijwmOzsLNlXsMDBwJEzNzBUSx9XLF/FtJ98i629ej8GMyePR4cuvsHbzTjRp/hkmjR2Bvx+87za2d3DEiLHfY/223Vi6ZhOsbapg7LBBSEnmYw5KavPWXYg8dlq6rVizHgDQuk3h/29yc3MwdNB3eP78GeYuWIxf9/6OSVNnwlKOxODSxfP4ou1nRdZfi7mCieNHw6/z19i6Yw9afNYao0cE4f69vwAAWVlZuPPnbfQfNBThv+zG/J+W4uHDWIwaPrTMMZGs7OxspKWlyWzZ2dllOldsbCzi4+PRunVraZmRkRG8vb0RHf3290x0dDSMjY2lyQoAtG7dGmpqajh//ry0TbNmzaTJCgD4+vri7t27SE5Olrb58Drv2ry7TknInbC8ExcXx5cflkBW5hssC52MgaMmQl/foNi2n3jWR4MmLWHn6AxrWzu0/6o7HKq64u6tGGmb3JwcbF69CIO7tUOfjk0wcZg/bl27VOb4DuwOR6t2fmjZ9kvYOVZF/xEh0BJr43jkXmmb1h2+gnvturC0tkXVam74tt9QvHqRgMSEuDJftzKat2QV2n3hB2cXV7hWr4EJU35AQnwc/vrzdpHHuLnXwpDho9GqTTtofvDD4UMSiQThYWvRrVNbtGlaHwE9uuDE0cNljnP39i1o0LAxuvXuB0fnqggYPAzV3NyxZ8c2aZvWbTugfgMf2Faxh7OLKwJHjkVGRjoe/P8XGX2ciakpzM0tpNvpkydgZ++AevUbFNr+tz2/IjU1FQsWLYNXnbqwrWKHevUboHoNN2kbiUSC9WtXo2PbVmj0qSe6fd0JRw4fKnOM28I3w6dxE/TpFwDnqi4YGjQCbjXdsWN7OADAwMAAK9asRxvfdnByrgoPTy+M/34y/rx9C3FxBXuPKhNFrRIKDQ2FkZGRzBYaGlqmmN69xNjKSjbJtbKyktbFx8fD0tJSpl5DQwOmpqYybQo7x4fXKKrNhy9S/hiFJSxUMuuW/og63o1Ru653qY4TBAE3rlxA3NNHqOlRR1q+ftlc/HX7OkZMnI25q7ejYbPWCA0Zjrinj0sdW15uLv7+6w48PohNTU0NHnUb4N7t64Uek5WZiRORe2FpXQXmFqrb5ftf8G4Sm4GRkVznCQ9bi8iDexE8YTLCtu/B1z16Y9bUEMRcuVim8926cQ31GjSUKWvQsBFu37hWaPvc3Fzsi9gFPX0DuFSvUaZrVna5uTk4eGAvOvl9VeS8hVMnjqG2pxd+nD0Dn7dojG86d8T6n1ch/4Oezg1r1+DAvt8QMnkaduzZjx69/TH5+3G4fOlCmeK6fi0G3t6NZMp8GjXG9WsxRR6Tnv4aIpEIBgaGZbpmRaGoVUIhISFITU2V2UJCQsr79v4VfDT/v+iP45GIvXcHs5dvKvExbzLSMbhbO+Tl5kBNTR0Bw8ejdr23vzxeJsbjROQ+LA/fD1NzCwBAx669EXMxGici96F7QGCp4ktLTYFEkg8jE1OZciMTUzx/8lCmLHLvToT/vATZWZmwtXfExB+XQ0NT9eYF/FdIJBIs++lH1PKsg6ouxQ8VFicnJwfhYWuxYNkafFLbCwBgW8UeN2KuYt+vO+FV99NSnzPp1UuYmprJlJmYmiEp6aVM2dnTJzFj0lhkZ2XBzNwCC5atgbGxSZnvpTI7fuwo0l+/RsdOnYts8/TpE8RdOId2HTpiyYrVePL4MebMmo68vDwMHBKEnJwcrF+7Git/Xo/anm//yLGzs0fMlSvYvfOXIntuivPq5UuYmsl+L5iamePVy5eFts/OzsaShfPh264D9PX1S329ikRRj+YXi8UKW5FrbW0NAEhISICNjY20PCEhAV5eXtI2iYmJMsfl5eUhKSlJery1tTUSEhJk2rzb/1ibd/UlodIJy5MnTzB16lSsX7++yDbZ2dkFxu9ysnOgpWJLrF8mxmPjigWY+ONyaGmVPDZtHV3MXbUVWZlvcOPqRWxatRCWNlXwiWd9PI69D4kkHyP7fSVzTF5uDgwMjaTXDQ7oKq2T5OcjLz8PfTo2lZZ17t4PnXuUboVX01btULuuN5KTXmL/zs1Y9MMEzFi0rlT3Ru8tmjsLsX/fx9I1G+U6z7Mnj5GVlYnRw2TnKeXl5qJajZrS/bbN3/+ykkgkyM3JkSn7vO0XGB1S+ETPotSp/ynWbtmF1JRkHIjYjWkhY7ByQzhM/pHs0Mf9tmcXGjVuWuxEVUGQwMTUDBOnzIC6ujpqutdCYmICNoWtx8AhQXjy+BGysjIxdGCAzHG5ubmo4fb+e6GJd13p1xJJPnJycmTK2n/REd9Pnl7qe8jNzcWEMSMhCEDIpGmlPp6Uz9nZGdbW1jh69Kg0QUlLS8P58+cxZMgQAICPjw9SUlJw+fJl1KtXDwBw7NgxSCQSeHt7S9tMnDgRubm50Pz/H65RUVGoUaMGTExMpG2OHj2KkSNHSq8fFRUFHx+fEsdb4oQlODi42PoXL16U+KIllZSUhI0bNxabsISGhmL6dNl/TINGTsDgUd8rPB55xN67g9SUJEwY0ktaJpHk488bVxH52w6EHzwLNXX1AsepqanBuoo9AMDJtQaePY5FxLYwfOJZH1mZb6Cmpo45KzZDTU32WG0dHQCAiZk55q7aKi0/f+YYLpw5hmETfpCW6f+/q9bQyBhqauoyE2wBIDU5CcYmsr90dPX0oaunDxs7B1Sv6YHvvmqJi2eOo/Fnbcvy8VRqi+bNQvSZk1iyOgyWViX/a6Mw7ybszlm4vMAQ3YcT4tZu2SX9+s9b17F62UIsWrlBWqanpyf92tTMHElJr2TOlZz0CqamspN+dXR0YWfvADt7B3zi4YmeXTrg4N496Nm3v1z3VNnEPX+GC+eiMW/h0mLbmZtbQENDE+of/NxwdnbBq5cvkJubgzdv3n4vLF6+qsBE3A/nP23buUf69Y0b17F04XysWf++F1hP733PiJm5OZJeyX4vJL16CTNz2e+F3NxcTBg7CnFxz7FqbVil710Bym/+RXp6Ou7fvy/dj42NRUxMDExNTeHg4ICRI0fihx9+QLVq1eDs7IzJkyfD1tZWupKoZs2aaNu2LQYMGIBVq1YhNzcXQUFB6NatG2xtbQEAPXr0wPTp0xEQEIDx48fj5s2bWLx4MRYuXCi97ogRI9C8eXMsWLAAHTp0wPbt23Hp0iWZpc8fU+KE5erVqx9t06xZs4+2+dDevXuLrf/7778/eo6QkJACydSdhJwiWpefWnU+xbw122XKVs6fgSr2jvjyW/9Ck5XCCIIEeblv78/JtQYkknykpiTLzGv5kLq6hjThAQAjY1NoaYllyt7R0NRE1epuuHH1Aj5t3ALA27++b169CN9O3xQTkwBBEJCbm1uie6C3BEHA4vmzcebEMSxauR42VezkPqeTsws0tbSQGB9f7PCPnb2D9OsXiQlQV9eQKfvQJx6euHLxvMzy6Uvno+Hu4VlsLIJEUuDlqPRxeyN+hYmp2UeXJnt61cWh3/dDIpFATe3tr8NHjx7C3MICmppaqOriAi0tLcTHxRU7/GPv4Cj9OiEhHuoa6jJlH6rt6YUL56PRo7e/tOz8ubOo7ekl3X+XrDx59Air123ksOD/ldfbmi9duoSWLVtK99/9vvT390dYWBjGjRuHjIwMDBw4ECkpKWjSpAkOHToEbW1t6THh4eEICgpCq1atoKamhi5dumDJkiXSeiMjIxw+fBiBgYGoV68ezM3NMWXKFJlntTRq1Ahbt27FpEmT8P3336NatWqIiIhArVrvH5nxMSVOWI4fP17ik5aUn58fRCIRinsUzMf+Jxc2nqeVonrPYdHR1YODs6tMmba2NvQNjaXly36cAlNzS/QICAIA7Nm2AS7Va8LK1g65Obm4euEPnD5yEAHD306wsrVzRJNW7bB87lT0HjgSzq41kJaajBtXL8KxajXU9W5S6jg7dOmJFXOnwaW6O1xqfIKDe7YiOysTLXzfPmMnIe4pzp6Igme9hjA0NsGrFwn4bXsYtLS0UadBY3k+okpn0dxZOBJ5ELPmL4aOrp50HoC+vj7E//9hMXvq9zC3tMTAwJEA3v4yeBj7AMDbYZ6XLxJx76870t4NXT09fNvTH8sWzoVEIoGHV11kpL/GzWtXoaunj7ZfdCp1nF269cKIQf3wS/hGNGzcFMcOH8LdP29h9PdTAbzt1dmy4Wc0atoCZuYWSE1JRsSu7XjxIhEtWrVRwCdVeUgkEuz9bQ+++NIPGhqyP56nfD8eFlaWGDbi7XObvv62O3ZsD8f8H2fh2+698PjxI2xYuxrderxNLPX09NHb/zssmBcKiUQCr7r1kP76Na7FXIGenn6x82OK0r1nbwz4rg82b1yPJs1a4PDvB3D71i1MnDIDwNvvz/GjR+DOn7exaNkq5Evy8fLl2953IyMjaGoWvrKNlKdFixYf/R07Y8YMzJgxo8g2pqam2Lp1a5H1AFC7dm2cPn262DZdu3ZF165di21TnHKdw2JjY4MVK1agU6fCf4jGxMRIx8wqg1eJ8VATve84zM7KxLolP+LVy0RoicWoYu+EoAkz0ajF+18CQ8ZMxa/h67B5zSIkvUyEoaExXGt6oF7DpoVd4qMatWiDtJRk7Ni4CinJr+DkUh0hs5dKh4Q0NcW4c+Mqfv91G9LT02BsYgY3jzqYuXhdgcm6VLzfdv8CABg5WHb+0PgpM9HuCz8AQEJCHERq75P2ly8SMaDX+3/wv2wJwy9bwuBZtz4Wr3o7pBMweBiMTUwRvnEt4mY/hb6BIarVqIleZRyaqVXbC5NnzsG6VcuwdsViVLF3xA/zFksnB6upqePxw1hEHtiL1JRkGBoZw839EyxdsxHOLq4fOTt96Py5s4iPe45Ofl8VqIuPfy7zvWBtbYNlq9Ziwdw56PZ1J1hYWqF7z97w/26AtM2QoBEwNjHFhnVr8Gz6UxgYGMCtpju+6z+oTPF5etXFrDnzsXLpIixfshAODk5YsHgZXKtVB/C2t+7kiWMAgO5d/WSOXb1uI+p/WrrVkRWJWvl0sFQocj/pVh5ffvklvLy8iszsrl27hjp16kAikZTqvP+FJ93Sv0sVn3RL5UcVn3RL5effeNJt8N47CjnPT1+6fbxRBVWuPSxjx45FRkZGkfWurq5KGYoiIiKi/5ZyTViaNi1+2EJPTw/NmyvvnRhERET/hvKadFuRqPRzWIiIiCoCzmGRX5mWhp8+fRq9evWCj48Pnj17BgDYvHkzzpw5o9DgiIiIiIAyJCy7d++Gr68vdHR0cPXqVelTZlNTU/m2ZiIiokIo6l1ClVmpE5YffvgBq1atws8//yx9BC8ANG7cGFeuXFFocERERBWBot7WXJmVeg7L3bt3C32irZGREVJSUhQRExERUYVSXo/mr0hK/RlaW1vLvJfgnTNnzqBq1aoKCYqIiIjoQ6VOWAYMGIARI0bg/PnzEIlEeP78OcLDwzFmzBjp2x2JiIjoPc5hkV+ph4QmTJgAiUSCVq1a4c2bN2jWrBnEYjHGjBmDYcOGKSNGIiKi/7TKPv9EEcr8aP6cnBzcv38f6enpcHd3V6nXh/PR/PRPfDQ/fYiP5qcP/RuP5p986J5CzjOzbTWFnOe/qMwPjtPS0oK7u7siYyEiIqqQ2MEiv1InLC1btiz2EcPHjh2TKyAiIqKKhk+6lV+pExYvLy+Z/dzcXMTExODmzZvw9/dXVFxEREREUqVOWBYuXFho+bRp05Ceni53QERERBUNJ93KT2HPsunVqxfWr1+vqNMRERFVGFzWLD+FJSzR0dHQ1tZW1OmIiIiIpEo9JPTVV1/J7AuCgLi4OFy6dAmTJ09WWGBEREQVBSfdyq/UCYuRkZHMvpqaGmrUqIEZM2agTZs2CguMiIioohCBGYu8SpWw5Ofno1+/fvDw8ICJiYmyYiIiIqpQ2MMiv1LNYVFXV0ebNm34VmYiIiL6V5V60m2tWrXw999/KyMWIiKiCklNpJitMit1wvLDDz9gzJgx2L9/P+Li4pCWliazERERkSyRSKSQrTIr8RyWGTNmYPTo0Wjfvj0A4Msvv5T58ARBgEgkQn5+vuKjJCIiokqtxAnL9OnTMXjwYBw/flyZ8RAREVU4lX04RxFKnLAIggAAaN68udKCISIiqogq+WiOQpRqDktlHz8jIiKi8lGq57BUr179o0lLUlKSXAERERFVNHz5ofxKlbBMnz69wJNuiYiIqHicwyK/UiUs3bp1g6WlpbJiISIiIipUiRMWzl8hIiIqG/4KlV+pVwkRERFR6ajx5YdyK3HCIpFIlBkHERFRhcUeFvmV+tH8RERERP+2Uk26JSIiotLjKiH5MWEhIiJSMj6HRX4cEiIiIqqAnJycCn3jc2BgIACgRYsWBeoGDx4sc47Hjx+jQ4cO0NXVhaWlJcaOHYu8vDyZNidOnEDdunUhFovh6uqKsLAwpdwPe1iIiIiUrDw6WC5evIj8/Hzp/s2bN/H555+ja9eu0rIBAwZgxowZ0n1dXV3p1/n5+ejQoQOsra1x9uxZxMXFoU+fPtDU1MTs2bMBALGxsejQoQMGDx6M8PBwHD16FP3794eNjQ18fX0Vej9MWIiIiJSsPIaELCwsZPbnzJkDFxcXmZcY6+rqwtrautDjDx8+jNu3b+PIkSOwsrKCl5cXZs6cifHjx2PatGnQ0tLCqlWr4OzsjAULFgAAatasiTNnzmDhwoUKT1g4JERERPQfkZ2djbS0NJktOzv7o8fl5ORgy5Yt+O6772QeBBseHg5zc3PUqlULISEhePPmjbQuOjoaHh4esLKykpb5+voiLS0Nt27dkrZp3bq1zLV8fX0RHR0t760WwISFiIhIyUQixWyhoaEwMjKS2UJDQz96/YiICKSkpKBv377Ssh49emDLli04fvw4QkJCsHnzZvTq1UtaHx8fL5OsAJDux8fHF9smLS0NmZmZZf24CsUhISIiIiVTVO9ASEgIgoODZcrEYvFHj1u3bh3atWsHW1tbadnAgQOlX3t4eMDGxgatWrXCgwcP4OLioqCIFYcJCxER0X+EWCwuUYLyoUePHuHIkSP49ddfi23n7e0NALh//z5cXFxgbW2NCxcuyLRJSEgAAOm8F2tra2nZh20MDQ2ho6NTqjg/hkNCRERESlbY8uKybGWxYcMGWFpaokOHDsW2i4mJAQDY2NgAAHx8fHDjxg0kJiZK20RFRcHQ0BDu7u7SNkePHpU5T1RUFHx8fMoUa3GYsBARESmZSEFbaUkkEmzYsAH+/v7Q0Hg/qPLgwQPMnDkTly9fxsOHD7F371706dMHzZo1Q+3atQEAbdq0gbu7O3r37o1r164hMjISkyZNQmBgoLSXZ/Dgwfj7778xbtw43LlzBytWrMCOHTswatSoMkRbPA4JERERKVl5Pen2yJEjePz4Mb777juZci0tLRw5cgSLFi1CRkYG7O3t0aVLF0yaNEnaRl1dHfv378eQIUPg4+MDPT09+Pv7yzy3xdnZGQcOHMCoUaOwePFi2NnZYe3atQpf0gwAIkEQBIWftZzFPH5d3iGQirEyKt2YL1VsBtqa5R0CqRB9sfKTiS2XnyrkPL3q2SnkPP9F7GEhIiJSMr5JSH5MWIiIiJSM7z6UHyfdEhERkcpjDwsREZGSlXVJMr3HhIWIiEjJOJwhP36GREREpPLYw0JERKRkHBKSHxMWIiIiJWO6Ij8OCREREZHKYw8LERGRknFISH4VMmG5nphS3iGQivGzrFLeIZAKeZL0prxDIBVS00ZP6dfgcIb8KmTCQkREpErYwyI/Jn1ERESk8tjDQkREpGTsX5EfExYiIiIl44iQ/DgkRERERCqPPSxERERKpsZBIbkxYSEiIlIyDgnJj0NCREREpPLYw0JERKRkIg4JyY0JCxERkZJxSEh+HBIiIiIilcceFiIiIiXjKiH5MWEhIiJSMg4JyY8JCxERkZIxYZEf57AQERGRymMPCxERkZJxWbP8mLAQEREpmRrzFblxSIiIiIhUHntYiIiIlIxDQvJjwkJERKRkXCUkPw4JERERkcpjDwsREZGScUhIfkxYiIiIlIyrhOTHISEiIiJSeexhISIiUjIOCcmPCQsREZGScZWQ/JiwEBERKRnzFflxDgsREVEFNG3aNIhEIpnNzc1NWp+VlYXAwECYmZlBX18fXbp0QUJCgsw5Hj9+jA4dOkBXVxeWlpYYO3Ys8vLyZNqcOHECdevWhVgshqurK8LCwpRyP0xYiIiIlExNJFLIVlqffPIJ4uLipNuZM2ekdaNGjcK+ffuwc+dOnDx5Es+fP8dXX30lrc/Pz0eHDh2Qk5ODs2fPYuPGjQgLC8OUKVOkbWJjY9GhQwe0bNkSMTExGDlyJPr374/IyEj5PrBCcEiIiIhIycprSEhDQwPW1tYFylNTU7Fu3Tps3boVn332GQBgw4YNqFmzJs6dO4eGDRvi8OHDuH37No4cOQIrKyt4eXlh5syZGD9+PKZNmwYtLS2sWrUKzs7OWLBgAQCgZs2aOHPmDBYuXAhfX1+F3gt7WIiIiP4jsrOzkZaWJrNlZ2cX2f7evXuwtbVF1apV0bNnTzx+/BgAcPnyZeTm5qJ169bStm5ubnBwcEB0dDQAIDo6Gh4eHrCyspK28fX1RVpaGm7duiVt8+E53rV5dw5FYsJCRESkbCLFbKGhoTAyMpLZQkNDC72kt7c3wsLCcOjQIaxcuRKxsbFo2rQpXr9+jfj4eGhpacHY2FjmGCsrK8THxwMA4uPjZZKVd/Xv6oprk5aWhszMzDJ8UEXjkBAREZGSKeo5LCEhIQgODpYpE4vFhbZt166d9OvatWvD29sbjo6O2LFjB3R0dBQSz7+JPSxERET/EWKxGIaGhjJbUQnLPxkbG6N69eq4f/8+rK2tkZOTg5SUFJk2CQkJ0jkv1tbWBVYNvdv/WBtDQ0OFJ0VMWIiIiJRMJFLMJo/09HQ8ePAANjY2qFevHjQ1NXH06FFp/d27d/H48WP4+PgAAHx8fHDjxg0kJiZK20RFRcHQ0BDu7u7SNh+e412bd+dQJCYsRERESqagKSylMmbMGJw8eRIPHz7E2bNn0blzZ6irq6N79+4wMjJCQEAAgoODcfz4cVy+fBn9+vWDj48PGjZsCABo06YN3N3d0bt3b1y7dg2RkZGYNGkSAgMDpb06gwcPxt9//41x48bhzp07WLFiBXbs2IFRo0bJ94EVgnNYiIiIKqCnT5+ie/fuePXqFSwsLNCkSROcO3cOFhYWAICFCxdCTU0NXbp0QXZ2Nnx9fbFixQrp8erq6ti/fz+GDBkCHx8f6Onpwd/fHzNmzJC2cXZ2xoEDBzBq1CgsXrwYdnZ2WLt2rcKXNAOASBAEQeFnLWebLj0p7xBIxfjVqlLeIZAKeZas2NUL9N9W00ZP6de4GJuqkPN86mykkPP8F7GHhYiISMn4tmb5MWEhIiJSMr6tWX6cdEtEREQqjz0sRERESsYOFvkxYSEiIlI2Zixy45AQERERqTz2sBARESkZVwnJjwkLERGRknGVkPw4JEREREQqjz0sRERESsYOFvkxYSEiIlI2Zixy45AQERERqTz2sBARESkZVwnJjwkLERGRknGVkPyYsBARESkZ8xX5cQ4LERERqTz2sPyLlo3oidSXCQXK67X+Em37DS9Qfufiafzx2zYkJzyDJD8fJlZV0LD91/Bo+rlS47x0+DecO7AD6alJsHJwQRv/IFRxcZPWH1y3ELE3ryA9+RW0tHVQpZo7Pus+AOa2DkqNqyLKyMjAquWLceLYESQnJaG6W02MHvc9PqnlUeQxO7aHY+f2rYh7/gxW1jb4bsAgdOjop9Q4L1+8gIXz5+DvB/f/f83B6Nips7R+w7o1OH40Co9i/4ZYrI3aXnUQNHI0nJyclRpXRfL7bztx6LedSIyPAwA4OFXFN/4DUc+7caHtj/6+F0t/nCZTpqmphZ1R55Qa58E9v2DP9k1ISXoFJ9fqGDB8HKrXrCWtX7HgB1y7fAHJL19AW0cHbrU80WfgcNg5VvLvBXaxyI0Jy7+o38zlECQS6f6Lp7HYGjoeNb2bFdpeR88AjTv1gLmtPdQ1NHHv6jnsWzMPukbGcKn9aZliuHYyEtdPR6L3pJ8Krb8dfRxHwleh3XcjYOtSExcO7cb2ORMweP4G6BmZAACsnauhVqNWMDS3RGb6a5z+dRO2zRmPwEVboKamXqa4Kqsfpk3Cg/v3MH3Wj7CwsMTvB/YhcNB32PHrflhaWRVov2vHNqxYshDfT5kB91oeuH3jOmbNmAIDAyM0a9GyTDE8f/YMndq3xsVrfxZa/+zpU4wMGoyvun6LmaHzcPH8OcyaPhnm5hbwadwEAHDl0kV0/bYH3D+phfz8fKxYuhDDBgdgx6/7oaOrW6a4KhszC0v0HjgctnYOEAQBxyP3IXTiKPz08zY4OLsUeoyunj6Wb/pVui+Sc6LE0d/34tihfZi1+OdC688ci8T6FT9hSPD3qF7TA3t3hWP62EAs37wHxiamAACX6jXRvHU7mFvaIP11KraHrca0sYFYvW0f1NUr788HTrqVHxOWf5GeobHM/tl922FiZQuHmp6Ftnd095LZb9D2K1w/fRhP7t6UJix5uTk4sWMDbkUfQ/abDFjYOeGzbv0LHFtS53/fDa+W7eHZvC0AoP13I3E/5jyunTyERl92BwDU/ewLaXtjC2s079oPa0MGIvVFAkysbMt03cooKysLx49GYf6iZahb7+3/z4FDgnD65HHs3rkNQ4JGFjjm4P696Pz1t2jTtj0AwM7OHrdu3cSmDWtlEpaIX3cifFMYnj97ChvbKvi2Ry90/bZHmeL8ded22FapglFjxgMAnKu6IObqZWzdslGasCxdKfsLbuqMULRp2Rh//nlLem9UvAaNmsvs9+ofhEO/7cLd2zeKTFgAwMTMvMi63JwcbFm7HKePHUJG+ms4OLuiz8Dh8KhTv0wx/rYzHG06dEardp0AAEOCJ+LyuTM4evA3dOnZDwDg27GLtL2VjS16BgzFyIBuSIx/Dpsq9mW6LhHAOSzlJj8vFzfPHIFn87Yl+qtIEATE3ryCpLincHCrLS2PDFuKZ/duo3PQJAwIXYOa3s2wbW4IkuKflimmuNi/4FyrrrRMpKYG51p18fTe7UKPycnKxPWTh2BsYQ1DM4tSX7Myy8/PR35+PrTEYplysVgbMVevFHpMbk4OtLS0ZMq0tcW4dfMG8nJzAQC/H9iH1SuWYkjQCOzYcwBDh43E6uVLsH9vRJnivHE9Bg0a+siUNWzUBDeuxxR5THr6awCAoaFRma5Z2eXn5+P00UhkZWXC7ZPaRbbLyszEgG/bI6BrO8yeOAqPYx/I1K9Z/CPu3r6O0VNCsWjdL2jcvDVmjAvC86ePSx1Tbm4uHtz9E7XreUvL1NTU4FnPG3dvXy8yvqO/74WVTRWYW1qX+poViUikmK0yYw9LObl76Q9kvUlH7WZtim2X9SYdS4K6IT8vFyI1NbTtOxxVPeoBAFJfJuDaqUgMW7IVBiZv/8pq2OEbPLh2EddORqLltwGliunN61QIEol06OcdPUMTvHr+RKbsUtRvOLbtZ+RmZ8HMxh49QuZCXUOzVNer7PT09ODh6YV1a1bC2dkFpmZmiPz9AG5cj4GdfeHzgRo2aoLf9uxCi89aw62mO/68fQsRv+5GXl4uUlKSYW5hiTUrl2Hk6PH4rPXb760qdnaI/fsBft31C7740q/Ucb56+RKm//gr3szMDBnp6cjKyoK2trZMnUQiwU9zQ+HpVReu1aqX+nqV2cO/72HC0L7IycmBto4OJsxcAHunqoW2reLgiGHjp8KxajW8yUhHxC+bMCGoH5Zs2AlzSyu8SIjD0d/3Yu2OgzA1f/vHhF+3Prhy4SyO/v4beg8YVqrYXqemQCLJh7GpqUy5kYkpnj5+KFN2MGIHNq1ajKysTFSxd8K0+SugqVm5fz5U8lxDIco9YcnMzMTly5dhamoKd3d3mbqsrCzs2LEDffr0KfL47OxsZGdny5Tl5mRDU0tcxBGq4dqJ3+Hi2UCaaBRFrK2L/rNXIycrEw9vXcWR8FUwsbSBo7sXEp/EQpBIsHJ0X5lj8vNyoaNvCOBtUrN63PvERSLJhyQvH3O/ez+s07hTDzTuVLrhglqNW6GqRz2kJyfh3MGd+HXJTPhPXQyNf/z1T8WbMetHzJg6Ee0/bw51dXXUcHNHm7YdcOfPW4W2Dxg4BK9evkS/3t0AQYCpqRm+6NgJm8LWQaSmhsw3b/D0yWPMnDYJs6ZPkR6Xn58HfX0D6f43nb9AfNzbyZ2CIAAAmjWsJ633qlsPS1asKdM9zZ09Aw8e3MPPYeFlOr4yq2LvhIVrtyEjIx3RJ49iSegUzFq8ttCkxe0TT7h98n442a1WbQT16YLIfbvRM2AoHv19HxJJPob28pM5Ljc3FwZGb3u+XiTEYZj/19K6t71+eejW9v1E3y69vkPXXqX746d563bwqt8Qya9eIOKXzZg3fTzmLN1QoDeRqDTKNWH566+/0KZNGzx+/BgikQhNmjTB9u3bYWNjAwBITU1Fv379ik1YQkNDMX36dJkyvwEj0XlgsFJjl0fqiwTE3ryKLiOnfrStSE0NptZVAADWTq54+fwxzu7dBkd3L+RmZUKkpoaAH1ZCpCY7uqelrQMAMDAxR//Zq6Xldy+ewZ2Lp9FpaIi0TOf/v8h0DYwgUlNDRmqyzLky0pIL9Lpo6+pDW1cfptZ2qFKtJhYM7Iy7l87gk0afleKTIDt7B6xZvxmZb94gIyMd5haWCBk7ClXs7Aptr62tjSkzZuH7ydPwKukVzM0tsGf3Dujp6cHExBTJyUkAgIlTZqCWh+xQwocTohcvX428vDwAQGJiAgYH+CN8x/vJm2Lx+14TM3NzJL16KXOuV69eQU9fv0DvytzZM3H61EmsWb8ZVlaVewigLDQ1NWFj97Z3zbWGO+7duYV9u7di6OhJHz1WQ0MTVau5If7Z297QrMw3UFNTx4I14VD7x88HbZ23E6FNzSywcO02aXn06WOIPnUMwRN/kJbp/39Yz8DIGGpq6khJSpI5V2pyEkxMzWTK9PQNoKdvAFs7B1R3r41eHZvj3JnjaNaqbUk/ioqHXSxyK9eEZfz48ahVqxYuXbqElJQUjBw5Eo0bN8aJEyfg4FCyJbIhISEIDpZNTnbeTFRGuApz7dQh6BoZo1qdhqU+VhAkyMt7O1fByqkaBIkEGWkpcHArfBmsmrq6NOEBAF1DY2hoacmUvaOuoQkb5+p4eOsKatR/+xeWIJHg4c2rqN+mUzExCRAEQTqHgkpPR1cXOrq6SEtLxbnoPzBs5Jhi22toakoTgsOHDqJJsxZQU1ODmZk5LCws8ezpE7Tr0LHI421s3///V1d/+2PA3sGx0LYetb3wx5lTMmUXzp2FR20v6b4gCJgX+gNOHDuCVes2FplwUekIggS5OSX7d5Wfn49Hf99HvYZv/+06V3ODRJKPlJQkfFK7bqHHqGtoSBMkADAyNoWWllim7B1NTU241KiJ61cuoGHTtxO8JRIJrl++gPadvy3uJiAIb+dfVWZcJSS/ck1Yzp49iyNHjsDc3Bzm5ubYt28fhg4diqZNm+L48ePQ09P76DnEYjHE/+hm1NRKVVbIchMkElw7GYnaTT+H2j+W+O1dOQcGJuZo2a0/AOCP37bCpmoNmFjZID83F/djLuDmmSNo228EAMDMxg61GrfC3lU/onXPQbB2dEXG61Q8vHkFlg5Vy5QQebfrgr2r58LGuQZsXWrgwqFfkZudhdr/XzWUnPgct6NPoGrt+tA1MMLrpJc4u287NLW04OrVQM5Pp/KJ/uMMBAhwdHTG0yePsHjhfDg5OePL/z/jZNnin/AiMQHTZ/0IAHj0MBa3bt5ALY/aeJ2WhvDNYfj7/j1MmzlHes6BQ4Mw/8fZ0Nc3gE/jJsjNzcXtWzfxOi0NPfv0LXWMX3Xthh3bt2LJwnn40q8LLl44hyOHD2Hh0lXSNj/OnoHI3w9g/qJl0NXTw8uXLwAA+voGBXphqHCb1yxFXe9GMLe0QWZmBk4fOYSbMZcxdd5yAMCi2ZNhZm6J3gPfzj35ZeMaVHf3gE0Ve2Skv0bE9k14kRCHzzu8/d6pYu+I5q3bYfHsKeg3dBSqVnNDakoyrl+5AKeq1VDfp2mpY+zUtScWh06Faw13VKv5Cfbt2oqsrEy0avclACD++VOcOX4YXvUbwsjYBK9eJGL31g0Qi8Wo17CJgj4pqqzKNWHJzMyEhsb7EEQiEVauXImgoCA0b94cW7duLcfolCP25hWkvUqEZ/N2BepSXyVCJHrfdZubnYVDG5bgddILaGiJYWZrj05DJsDd5/3y1S8GjsWZiHAcCV+N10kvoWtghCquNcuUrACAu09LZLxOxcldYchITYaVowu6jQ+F/v+HhDQ0tfDk7k1cPPQrMjPSoWdkAgc3D/hPXVJg2Ig+Lj39NZYvWYjEhHgYGhnhs1ZtMHTYSGj8f4Liy5cvEP//B4kBb/+iDd8UhkePYqGhoYH6n3pj7aZtsK3yvsfE76uu0NbWweaw9ViycB50dHThUq0auvcsemi1OFXs7LBo2Sr8NG8OtodvhqWVNSZOnSld0gwAu3dsBwAMDvCXOXbKjNkyD5ijoqWkJGHR7ClITnoJPT19OFathqnzlsOr/tt/yy8S4mV+PqS/TsOK+TORnPQK+vqGcKlRE3OWb5CZ7zJswjTs3LwWG1YsRNLLRBgYGaOGu0eZkhUAaPKZL1JTkrFtw0okJ72Cs2sNTJ27DMb/HxLS0hLj9vWr2LdrKzJep8HIxAyfeNbFnGUbpM9pqawq+wofRRAJ72bclYMGDRpg2LBh6N27d4G6oKAghIeHIy0tDfn5+aU676ZLTz7eiCoVv1oFh8Co8nqWnFneIZAKqWnz8d58ef0V/0Yh56luXXkfxFiuz2Hp3Lkztm3bVmjdsmXL0L17d5RjPkVERKQYIgVtlVi59rAoC3tY6J/Yw0IfYg8Lfehf6WFJUFAPi1Xl7WEp9+ewEBERVXRcJSQ/JixERERKxkm38uO7hIiIiEjlsYeFiIhIydjBIj8mLERERMrGjEVuHBIiIiIilcceFiIiIiXjKiH5MWEhIiJSMq4Skh+HhIiIiEjlMWEhIiJSsvJ4Mn9oaCg+/fRTGBgYwNLSEn5+frh7965MmxYtWkAkEslsgwcPlmnz+PFjdOjQAbq6urC0tMTYsWORl5cn0+bEiROoW7cuxGIxXF1dERYWVspoP44JCxERkbKVQ8Zy8uRJBAYG4ty5c4iKikJubi7atGmDjIwMmXYDBgxAXFycdJs7d660Lj8/Hx06dEBOTg7Onj2LjRs3IiwsDFOmTJG2iY2NRYcOHdCyZUvExMRg5MiR6N+/PyIjI0sX8EfwXUJUKfBdQvQhvkuIPvRvvEvo0atshZzH0Uxc5mNfvHgBS0tLnDx5Es2aNQPwtofFy8sLixYtKvSY33//HV988QWeP38OKysrAMCqVaswfvx4vHjxAlpaWhg/fjwOHDiAmzdvSo/r1q0bUlJScOjQoTLH+0/sYSEiIvqPyM7ORlpamsyWnV2yZCg1NRUAYGpqKlMeHh4Oc3Nz1KpVCyEhIXjz5v2LGqOjo+Hh4SFNVgDA19cXaWlpuHXrlrRN69atZc7p6+uL6OjoMt1jUZiwEBERKZlIpJgtNDQURkZGMltoaOhHry+RSDBy5Eg0btwYtWrVkpb36NEDW7ZswfHjxxESEoLNmzejV69e0vr4+HiZZAWAdD8+Pr7YNmlpacjMVFxvJpc1ExERKZmiVjWHhIQgODhYpkws/vgwUWBgIG7evIkzZ87IlA8cOFD6tYeHB2xsbNCqVSs8ePAALi4uiglaQdjDQkRE9B8hFothaGgos30sYQkKCsL+/ftx/Phx2NnZFdvW29sbAHD//n0AgLW1NRISEmTavNu3trYuto2hoSF0dHRKfnMfwYSFiIhIyRQ1JFQagiAgKCgIe/bswbFjx+Ds7PzRY2JiYgAANjY2AAAfHx/cuHEDiYmJ0jZRUVEwNDSEu7u7tM3Ro0dlzhMVFQUfH5/SBfwRTFiIiIiU7t9f1xwYGIgtW7Zg69atMDAwQHx8POLj46XzSh48eICZM2fi8uXLePjwIfbu3Ys+ffqgWbNmqF27NgCgTZs2cHd3R+/evXHt2jVERkZi0qRJCAwMlPbsDB48GH///TfGjRuHO3fuYMWKFdixYwdGjRol1yf2T1zWTJUClzXTh7ismT70byxrfpqco5Dz2JlolbitqIgumQ0bNqBv37548uQJevXqhZs3byIjIwP29vbo3LkzJk2aBENDQ2n7R48eYciQIThx4gT09PTg7++POXPmQEPj/TTYEydOYNSoUbh9+zbs7OwwefJk9O3bt8z3Wej9MGGhyoAJC32ICQt96N9IWJ6lKCZhqWJc8oSlouEqISIiIiXjuw/lxzksREREpPLYw0JERKRkpV3hQwUxYSEiIlIyEQeF5MaEhYiISNmYr8iNc1iIiIhI5bGHhYiISMnYwSI/JixERERKxkm38uOQEBEREak89rAQEREpGVcJyY8JCxERkbIxX5Ebh4SIiIhI5bGHhYiISMnYwSI/JixERERKxlVC8uOQEBEREak89rAQEREpGVcJyY8JCxERkZJxSEh+HBIiIiIilceEhYiIiFQeh4SIiIiUjENC8mPCQkREpGScdCs/DgkRERGRymMPCxERkZJxSEh+TFiIiIiUjPmK/DgkRERERCqPPSxERETKxi4WuTFhISIiUjKuEpIfh4SIiIhI5bGHhYiISMm4Skh+TFiIiIiUjPmK/JiwEBERKRszFrlxDgsRERGpPPawEBERKRlXCcmPCQsREZGScdKt/DgkRERERCpPJAiCUN5BkOJlZ2cjNDQUISEhEIvF5R0OqQB+T9CH+P1A/zVMWCqotLQ0GBkZITU1FYaGhuUdDqkAfk/Qh/j9QP81HBIiIiIilceEhYiIiFQeExYiIiJSeUxYKiixWIypU6dyMh1J8XuCPsTvB/qv4aRbIiIiUnnsYSEiIiKVx4SFiIiIVB4TFiIiIlJ5TFiIiIhI5TFhqaCWL18OJycnaGtrw9vbGxcuXCjvkKicnDp1Ch07doStrS1EIhEiIiLKOyQqR6Ghofj0009hYGAAS0tL+Pn54e7du+UdFtFHMWGpgH755RcEBwdj6tSpuHLlCjw9PeHr64vExMTyDo3KQUZGBjw9PbF8+fLyDoVUwMmTJxEYGIhz584hKioKubm5aNOmDTIyMso7NKJicVlzBeTt7Y1PP/0Uy5YtAwBIJBLY29tj2LBhmDBhQjlHR+VJJBJhz5498PPzK+9QSEW8ePEClpaWOHnyJJo1a1be4RAViT0sFUxOTg4uX76M1q1bS8vU1NTQunVrREdHl2NkRKSKUlNTAQCmpqblHAlR8ZiwVDAvX75Efn4+rKysZMqtrKwQHx9fTlERkSqSSCQYOXIkGjdujFq1apV3OETF0ijvAIiIqHwEBgbi5s2bOHPmTHmHQvRRTFgqGHNzc6irqyMhIUGmPCEhAdbW1uUUFRGpmqCgIOzfvx+nTp2CnZ1deYdD9FEcEqpgtLS0UK9ePRw9elRaJpFIcPToUfj4+JRjZESkCgRBQFBQEPbs2YNjx47B2dm5vEMiKhH2sFRAwcHB8Pf3R/369dGgQQMsWrQIGRkZ6NevX3mHRuUgPT0d9+/fl+7HxsYiJiYGpqamcHBwKMfIqDwEBgZi69at+O2332BgYCCd22ZkZAQdHZ1yjo6oaFzWXEEtW7YM8+bNQ3x8PLy8vLBkyRJ4e3uXd1hUDk6cOIGWLVsWKPf390dYWNi/HxCVK5FIVGj5hg0b0Ldv3383GKJSYMJCREREKo9zWIiIiEjlMWEhIiIilceEhYiIiFQeExYiIiJSeUxYiIiISOUxYSEiIiKVx4SFiIiIVB4TFiIV0LdvX/j5+Un3W7RogZEjR/7rcZw4cQIikQgpKSlKu8Y/77Us/o04iUi1MGEhKkLfvn0hEokgEomgpaUFV1dXzJgxA3l5eUq/9q+//oqZM2eWqO2//cvbyckJixYt+leuRUT0Dt8lRFSMtm3bYsOGDcjOzsbBgwcRGBgITU1NhISEFGibk5MDLS0thVzX1NRUIechIqoo2MNCVAyxWAxra2s4OjpiyJAhaN26Nfbu3Qvg/dDGrFmzYGtrixo1agAAnjx5gm+++QbGxsYwNTVFp06d8PDhQ+k58/PzERwcDGNjY5iZmWHcuHH45xsy/jkklJ2djfHjx8Pe3h5isRiurq5Yt24dHj58KH1PkImJCUQikfR9MBKJBKGhoXB2doaOjg48PT2xa9cumescPHgQ1atXh46ODlq2bCkTZ1nk5+cjICBAes0aNWpg8eLFhbadPn06LCwsYGhoiMGDByMnJ0daV5LYiahyYQ8LUSno6Ojg1atX0v2jR4/C0NAQUVFRAIDc3Fz4+vrCx8cHp0+fhoaGBn744Qe0bdsW169fh5aWFhYsWICwsDCsX78eNWvWxIIFC7Bnzx589tlnRV63T58+iI6OxpIlS+Dp6YnY2Fi8fPkS9vb22L17N7p06YK7d+/C0NBQ+sbd0NBQbNmyBatWrUK1atVw6tQp9OrVCxYWFmjevDmePHmCr776CoGBgRg4cCAuXbqE0aNHy/X5SCQS2NnZYefOnTAzM8PZs2cxcOBA2NjY4JtvvpH53LS1tXHixAk8fPgQ/fr1g5mZGWbNmlWi2ImoEhKIqFD+/v5Cp06dBEEQBIlEIkRFRQlisVgYM2aMtN7KykrIzs6WHrN582ahRo0agkQikZZlZ2cLOjo6QmRkpCAIgmBjYyPMnTtXWp+bmyvY2dlJryUIgtC8eXNhxIgRgiAIwt27dwUAQlRUVKFxHj9+XAAgJCcnS8uysrIEXV1d4ezZszJtAwIChO7duwuCIAghISGCu7u7TP348eMLnOufHB0dhYULFxZZ/0+BgYFCly5dpPv+/v6CqampkJGRIS1buXKloK+vL+Tn55co9sLumYgqNvawEBVj//790NfXR25uLiQSCXr06IFp06ZJ6z08PGTmrVy7dg3379+HgYGBzHmysrLw4MEDpKamIi4uDt7e3tI6DQ0N1K9fv8Cw0DsxMTFQV1cvVc/C/fv38ebNG3z++ecy5Tk5OahTpw4A4M8//5SJAwB8fHxKfI2iLF++HOvXr8fjx4+RmZmJnJwceHl5ybTx9PSErq6uzHXT09Px5MkTpKenfzR2Iqp8mLAQFaNly5ZYuXIltLS0YGtrCw0N2X8yenp6Mvvp6emoV68ewsPDC5zLwsKiTDG8G+IpjfT0dADAgQMHUKVKFZk6sVhcpjhKYvv27RgzZgwWLFgAHx8fGBgYYN68eTh//nyJz1FesRORamPCQlQMPT09uLq6lrh93bp18csvv8DS0hKGhoaFtrGxscH58+fRrFkzAEBeXh4uX76MunXrFtrew8MDEokEJ0+eROvWrQvUv+vhyc/Pl5a5u7tDLBbj8ePHRfbM1KxZUzqB+J1z5859/CaL8ccff6BRo0YYOnSotOzBgwcF2l27dg2ZmZnSZOzcuXPQ19eHvb09TE1NPxo7EVU+XCVEpEA9e/aEubk5OnXqhNOnTyM2NhYnTpzA8OHD8fTpUwDAiBEjMGfOHERERODOnTsYOnRosc9QcXJygr+/P7777jtERERIz7ljxw4AgKOjI0QiEfbv348XL14gPT0dBgYGGDNmDEaNGoWNGzfiwYMHuHLlCpYuXYqNGzcCAAYPHox79+5h7NixuHv3LrZu3YqwsLAS3eezZ88QExMjsyUnJ6NatWq4dOkSIiMj8ddff2Hy5Mm4ePFigeNzcnIQEBCA27dv4+DBg5g6dSqCgoKgpqZWotiJqBIq70k0RKrqw0m3pamPi4sT+vTpI5ibmwtisVioWrWqMGDAACE1NVUQhLeTbEeMGCEYGhoKxsbGQnBwsNCnT58iJ90KgiBkZmYKo0aNEmxsbAQtLS3B1dVVWL9+vbR+xowZgrW1tSASiQR/f39BEN5OFF60aJFQo0YNQVNTU7CwsBB8fX2FkydPSo/bt2+f4OrqKojFYqFp06bC+vXrSzTpFkCBbfPmzUJWVpbQt29fwcjISDA2NhaGDBkiTJgwQfD09CzwuU2ZMkUwMzMT9PX1hQEDBghZWVnSNh+LnZNuiSofkSAUMdOPiIiISEVwSIiIiIhUHhMWIiIiUnlMWIiIiEjlMWEhIiIilceEhYiIiFQeExYiIiJSeUxYiIiISOUxYSEiIiKVx4SFiIiIVB4TFiIiIlJ5TFiIiIhI5TFhISIiIpX3P4wUYGTZqjjNAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "logistic_regression = LogisticRegression(\n",
    "    solver='lbfgs',\n",
    "    penalty='l2'\n",
    ")\n",
    "logistic_regression.fit(X_train, y_train)\n",
    "\n",
    "evaluate_classification(logistic_regression, X_train, y_train, X_test, y_test, cv=10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7ab85e2c",
   "metadata": {
    "papermill": {
     "duration": 0.007569,
     "end_time": "2024-02-01T18:32:36.343723",
     "exception": false,
     "start_time": "2024-02-01T18:32:36.336154",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# (2) Decision Tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "d26652ab",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-01T18:32:36.360582Z",
     "iopub.status.busy": "2024-02-01T18:32:36.360300Z",
     "iopub.status.idle": "2024-02-01T18:32:51.918779Z",
     "shell.execute_reply": "2024-02-01T18:32:51.917858Z"
    },
    "papermill": {
     "duration": 15.569418,
     "end_time": "2024-02-01T18:32:51.920869",
     "exception": false,
     "start_time": "2024-02-01T18:32:36.351451",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.72      0.90      0.80     22158\n",
      "           1       0.74      0.35      0.48      7093\n",
      "           2       0.72      0.61      0.66     11782\n",
      "\n",
      "    accuracy                           0.72     41033\n",
      "   macro avg       0.73      0.62      0.65     41033\n",
      "weighted avg       0.72      0.72      0.70     41033\n",
      "\n",
      "cv_scores.mean: 0.7423\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "\n",
    "decision_tree = DecisionTreeClassifier(\n",
    "    criterion='gini',\n",
    "    max_depth=8,\n",
    "    max_features='log2'\n",
    ")\n",
    "decision_tree.fit(X_train, y_train)\n",
    "\n",
    "evaluate_classification(decision_tree, X_train, y_train, X_test, y_test, cv=10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "93876b3a",
   "metadata": {
    "papermill": {
     "duration": 0.008329,
     "end_time": "2024-02-01T18:32:51.938028",
     "exception": false,
     "start_time": "2024-02-01T18:32:51.929699",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# (3) Random Forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "d6146489",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-01T18:32:51.956510Z",
     "iopub.status.busy": "2024-02-01T18:32:51.956205Z",
     "iopub.status.idle": "2024-02-01T18:33:32.317180Z",
     "shell.execute_reply": "2024-02-01T18:33:32.316257Z"
    },
    "papermill": {
     "duration": 40.372872,
     "end_time": "2024-02-01T18:33:32.319332",
     "exception": false,
     "start_time": "2024-02-01T18:32:51.946460",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.76      0.95      0.85     22158\n",
      "           1       0.83      0.51      0.64      7093\n",
      "           2       0.87      0.66      0.75     11782\n",
      "\n",
      "    accuracy                           0.79     41033\n",
      "   macro avg       0.82      0.71      0.74     41033\n",
      "weighted avg       0.81      0.79      0.78     41033\n",
      "\n",
      "cv_scores.mean: 0.8152\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "random_forest = RandomForestClassifier(\n",
    "    criterion='gini',\n",
    "    max_depth=8,\n",
    "    max_features='log2',\n",
    "    n_estimators=4\n",
    ")\n",
    "random_forest.fit(X_train, y_train)\n",
    "\n",
    "evaluate_classification(random_forest, X_train, y_train, X_test, y_test, cv=10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "60370abc",
   "metadata": {
    "papermill": {
     "duration": 0.009133,
     "end_time": "2024-02-01T18:33:32.337768",
     "exception": false,
     "start_time": "2024-02-01T18:33:32.328635",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Unsupervised Learning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "755aac26",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-01T18:33:32.357246Z",
     "iopub.status.busy": "2024-02-01T18:33:32.356932Z",
     "iopub.status.idle": "2024-02-01T18:33:32.362405Z",
     "shell.execute_reply": "2024-02-01T18:33:32.361578Z"
    },
    "papermill": {
     "duration": 0.01736,
     "end_time": "2024-02-01T18:33:32.364224",
     "exception": false,
     "start_time": "2024-02-01T18:33:32.346864",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def preprocess_data_unsupervised(file_path):\n",
    "    df = pd.read_csv(file_path)\n",
    "\n",
    "    condition_encoding = {'no stress': 0, 'time pressure': 1, 'interruption': 2}\n",
    "    df['condition'] = df['condition'].map(condition_encoding)\n",
    "\n",
    "    scaler = StandardScaler()\n",
    "    standardized_features = scaler.fit_transform(df)\n",
    "\n",
    "    pca = PCA(n_components=0.95, random_state=42)\n",
    "    pca_features = pca.fit_transform(standardized_features)\n",
    "\n",
    "    return pca_features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "f20688e8",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-01T18:33:32.383320Z",
     "iopub.status.busy": "2024-02-01T18:33:32.383058Z",
     "iopub.status.idle": "2024-02-01T18:33:32.790280Z",
     "shell.execute_reply": "2024-02-01T18:33:32.788808Z"
    },
    "papermill": {
     "duration": 0.420822,
     "end_time": "2024-02-01T18:33:32.794116",
     "exception": false,
     "start_time": "2024-02-01T18:33:32.373294",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "X_test = preprocess_data_unsupervised('../input/swell-heart-rate-variability-hrv/hrv dataset/data/final/test.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "339a3dea",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-01T18:33:32.835930Z",
     "iopub.status.busy": "2024-02-01T18:33:32.835392Z",
     "iopub.status.idle": "2024-02-01T18:33:32.849451Z",
     "shell.execute_reply": "2024-02-01T18:33:32.848429Z"
    },
    "papermill": {
     "duration": 0.03872,
     "end_time": "2024-02-01T18:33:32.852952",
     "exception": false,
     "start_time": "2024-02-01T18:33:32.814232",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def evaluate_clustering(X_test, labels):\n",
    "    silhouette = silhouette_score(X_test, labels)\n",
    "    print('silhouette_score:', f'{silhouette:.4f}')\n",
    "\n",
    "    silhouette_vals = silhouette_samples(X_test, labels)\n",
    "    y_ticks = []\n",
    "    y_lower, y_upper = 0, 0\n",
    "\n",
    "    for i, cluster in enumerate(set(labels)):\n",
    "        cluster_silhouette_vals = silhouette_vals[labels == cluster]\n",
    "        cluster_silhouette_vals.sort()\n",
    "\n",
    "        y_upper += len(cluster_silhouette_vals)\n",
    "        color = plt.cm.viridis(float(i) / len(set(labels)))\n",
    "\n",
    "        plt.barh(range(y_lower, y_upper), cluster_silhouette_vals, height=1.0, edgecolor='none', color=color)\n",
    "        y_ticks.append((y_lower + y_upper) / 2)\n",
    "        y_lower += len(cluster_silhouette_vals)\n",
    "\n",
    "    plt.axvline(x=silhouette, color=\"red\", linestyle=\"--\")\n",
    "    plt.xlabel('Silhouette Coefficient')\n",
    "    plt.ylabel('Cluster')\n",
    "    plt.yticks(y_ticks, range(len(set(labels))))\n",
    "    plt.title('silhouette_samples')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "670d5d76",
   "metadata": {
    "papermill": {
     "duration": 0.019281,
     "end_time": "2024-02-01T18:33:32.892159",
     "exception": false,
     "start_time": "2024-02-01T18:33:32.872878",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# (4) K-means"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "698c0ccb",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-01T18:33:32.924845Z",
     "iopub.status.busy": "2024-02-01T18:33:32.924535Z",
     "iopub.status.idle": "2024-02-01T18:35:11.332488Z",
     "shell.execute_reply": "2024-02-01T18:35:11.331591Z"
    },
    "papermill": {
     "duration": 98.432534,
     "end_time": "2024-02-01T18:35:11.344518",
     "exception": false,
     "start_time": "2024-02-01T18:33:32.911984",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.10/site-packages/sklearn/cluster/_kmeans.py:870: FutureWarning: The default value of `n_init` will change from 10 to 'auto' in 1.4. Set the value of `n_init` explicitly to suppress the warning\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "silhouette_score: 0.2561\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.cluster import KMeans\n",
    "\n",
    "kmeans = KMeans(n_clusters=3)\n",
    "kmeans.fit(X_test)\n",
    "labels = kmeans.predict(X_test)\n",
    "\n",
    "evaluate_clustering(X_test, labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c38538dc",
   "metadata": {
    "papermill": {
     "duration": 0.009528,
     "end_time": "2024-02-01T18:35:11.363956",
     "exception": false,
     "start_time": "2024-02-01T18:35:11.354428",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# (5) Gaussian Mixture"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "1720949b",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-01T18:35:11.384585Z",
     "iopub.status.busy": "2024-02-01T18:35:11.384288Z",
     "iopub.status.idle": "2024-02-01T18:36:54.049574Z",
     "shell.execute_reply": "2024-02-01T18:36:54.048657Z"
    },
    "papermill": {
     "duration": 102.688452,
     "end_time": "2024-02-01T18:36:54.062019",
     "exception": false,
     "start_time": "2024-02-01T18:35:11.373567",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "silhouette_score: 0.0928\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.mixture import GaussianMixture\n",
    "\n",
    "gaussian_mixture = GaussianMixture(n_components=3)\n",
    "gaussian_mixture.fit(X_test)\n",
    "labels = gaussian_mixture.predict(X_test)\n",
    "\n",
    "evaluate_clustering(X_test, labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "151797b1",
   "metadata": {
    "papermill": {
     "duration": 0.010217,
     "end_time": "2024-02-01T18:36:54.082635",
     "exception": false,
     "start_time": "2024-02-01T18:36:54.072418",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# (6) BIRCH"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "5b4cb8fc",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-01T18:36:54.104334Z",
     "iopub.status.busy": "2024-02-01T18:36:54.104055Z",
     "iopub.status.idle": "2024-02-01T18:38:41.357336Z",
     "shell.execute_reply": "2024-02-01T18:38:41.356413Z"
    },
    "papermill": {
     "duration": 107.277488,
     "end_time": "2024-02-01T18:38:41.370062",
     "exception": false,
     "start_time": "2024-02-01T18:36:54.092574",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "silhouette_score: 0.2117\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.cluster import Birch\n",
    "\n",
    "birch = Birch(n_clusters=3)\n",
    "birch.fit(X_test)\n",
    "labels = birch.predict(X_test)\n",
    "\n",
    "evaluate_clustering(X_test, labels)"
   ]
  }
 ],
 "metadata": {
  "kaggle": {
   "accelerator": "nvidiaTeslaT4",
   "dataSources": [
    {
     "datasetId": 113673,
     "sourceId": 271117,
     "sourceType": "datasetVersion"
    }
   ],
   "dockerImageVersionId": 30626,
   "isGpuEnabled": true,
   "isInternetEnabled": false,
   "language": "python",
   "sourceType": "notebook"
  },
  "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": "3.10.12"
  },
  "papermill": {
   "default_parameters": {},
   "duration": 418.236412,
   "end_time": "2024-02-01T18:38:43.705814",
   "environment_variables": {},
   "exception": null,
   "input_path": "__notebook__.ipynb",
   "output_path": "__notebook__.ipynb",
   "parameters": {},
   "start_time": "2024-02-01T18:31:45.469402",
   "version": "2.4.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}