eval_svm.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import os\n",
    "import json\n",
    "import time\n",
    "\n",
    "from data_utils import prepare_data, flatten_data, shuffle_data\n",
    "from sklearn.metrics import confusion_matrix, precision_score, recall_score, f1_score, accuracy_score\n",
    "\n",
    "from sklearn.svm import SVC\n",
    "\n",
    "def f1_metric(val_1, val2):\n",
    "    if val_1 + val2 == 0:\n",
    "        return 0  # To avoid division by zero\n",
    "    return 2 * (val_1 * val2) / (val_1 + val2)\n",
    "\n",
    "PATH = \"results/svm\"\n",
    "NPY_DATA_DIR = \"npy_datasets\"\n",
    "\n",
    "spectogram_map = {\n",
    "    256: [8, 64, 128, 250],\n",
    "    512: [0, 256, 511],\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "256 8 X_train.shape (40, 35, 16)\n",
      "256 8 X_test.shape (40, 35, 16)\n",
      "y_train_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "y_test_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "Linear SVM Training Time (W:256-O:8): 0.0017409324645996094 seconds\n",
      "Polynomial SVM Training Time (W:256-O:8): 0.0015769004821777344 seconds\n",
      "RBF SVM Training Time (W:256-O:8): 0.0014469623565673828 seconds\n",
      "Linear SVM Metrics:\n",
      "Precision: 69.16666666666667, Recall: 75.0, F1 Score: 70.66666666666667, Accuracy: 75.0\n",
      "\n",
      "Polynomial SVM Metrics:\n",
      "Precision: 80.0, Recall: 80.0, F1 Score: 78.5, Accuracy: 80.0\n",
      "\n",
      "RBF SVM Metrics:\n",
      "Precision: 65.0, Recall: 75.0, F1 Score: 68.83333333333333, Accuracy: 75.0\n",
      "--------------------------------------------------\n",
      "Linear Kernel Accuracy: 75.0\n",
      "Polynomial Kernel Accuracy: 80.0\n",
      "RBF Kernel Accuracy: 75.0\n",
      "--------------------------------------------------\n",
      "256 8 X_train.shape (40, 35, 16)\n",
      "256 8 X_test.shape (40, 35, 16)\n",
      "y_train_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "y_test_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "Linear SVM Training Time (W:256-O:8): 0.0015888214111328125 seconds\n",
      "Polynomial SVM Training Time (W:256-O:8): 0.001580953598022461 seconds\n",
      "RBF SVM Training Time (W:256-O:8): 0.0014960765838623047 seconds\n",
      "Linear SVM Metrics:\n",
      "Precision: 67.5952380952381, Recall: 65.0, F1 Score: 61.412698412698404, Accuracy: 65.0\n",
      "\n",
      "Polynomial SVM Metrics:\n",
      "Precision: 62.26190476190476, Recall: 62.5, F1 Score: 57.72222222222222, Accuracy: 62.5\n",
      "\n",
      "RBF SVM Metrics:\n",
      "Precision: 64.76190476190476, Recall: 62.5, F1 Score: 58.72222222222223, Accuracy: 62.5\n",
      "--------------------------------------------------\n",
      "Linear Kernel Accuracy: 65.0\n",
      "Polynomial Kernel Accuracy: 62.5\n",
      "RBF Kernel Accuracy: 62.5\n",
      "--------------------------------------------------\n",
      "256 64 X_train.shape (40, 35, 21)\n",
      "256 64 X_test.shape (40, 35, 21)\n",
      "y_train_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "y_test_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "Linear SVM Training Time (W:256-O:64): 0.001676321029663086 seconds\n",
      "Polynomial SVM Training Time (W:256-O:64): 0.001743316650390625 seconds\n",
      "RBF SVM Training Time (W:256-O:64): 0.0016353130340576172 seconds\n",
      "Linear SVM Metrics:\n",
      "Precision: 71.66666666666666, Recall: 77.5, F1 Score: 72.49999999999999, Accuracy: 77.5\n",
      "\n",
      "Polynomial SVM Metrics:\n",
      "Precision: 65.0, Recall: 70.0, F1 Score: 65.33333333333331, Accuracy: 70.0\n",
      "\n",
      "RBF SVM Metrics:\n",
      "Precision: 58.33333333333333, Recall: 67.5, F1 Score: 59.999999999999986, Accuracy: 67.5\n",
      "--------------------------------------------------\n",
      "Linear Kernel Accuracy: 77.5\n",
      "Polynomial Kernel Accuracy: 70.0\n",
      "RBF Kernel Accuracy: 67.5\n",
      "--------------------------------------------------\n",
      "256 64 X_train.shape (40, 35, 21)\n",
      "256 64 X_test.shape (40, 35, 21)\n",
      "y_train_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "y_test_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "Linear SVM Training Time (W:256-O:64): 0.0009865760803222656 seconds\n",
      "Polynomial SVM Training Time (W:256-O:64): 0.0009951591491699219 seconds\n",
      "RBF SVM Training Time (W:256-O:64): 0.0010077953338623047 seconds\n",
      "Linear SVM Metrics:\n",
      "Precision: 68.33333333333333, Recall: 77.5, F1 Score: 70.66666666666667, Accuracy: 77.5\n",
      "\n",
      "Polynomial SVM Metrics:\n",
      "Precision: 60.83333333333333, Recall: 72.5, F1 Score: 65.33333333333333, Accuracy: 72.5\n",
      "\n",
      "RBF SVM Metrics:\n",
      "Precision: 61.16666666666667, Recall: 70.0, F1 Score: 64.02380952380952, Accuracy: 70.0\n",
      "--------------------------------------------------\n",
      "Linear Kernel Accuracy: 77.5\n",
      "Polynomial Kernel Accuracy: 72.5\n",
      "RBF Kernel Accuracy: 70.0\n",
      "--------------------------------------------------\n",
      "256 128 X_train.shape (40, 35, 31)\n",
      "256 128 X_test.shape (40, 35, 31)\n",
      "y_train_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "y_test_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "Linear SVM Training Time (W:256-O:128): 0.0012493133544921875 seconds\n",
      "Polynomial SVM Training Time (W:256-O:128): 0.0012519359588623047 seconds\n",
      "RBF SVM Training Time (W:256-O:128): 0.0013170242309570312 seconds\n",
      "Linear SVM Metrics:\n",
      "Precision: 70.0, Recall: 72.5, F1 Score: 68.66666666666667, Accuracy: 72.5\n",
      "\n",
      "Polynomial SVM Metrics:\n",
      "Precision: 62.083333333333336, Recall: 67.5, F1 Score: 61.83333333333333, Accuracy: 67.5\n",
      "\n",
      "RBF SVM Metrics:\n",
      "Precision: 67.5, Recall: 72.5, F1 Score: 67.66666666666666, Accuracy: 72.5\n",
      "--------------------------------------------------\n",
      "Linear Kernel Accuracy: 72.5\n",
      "Polynomial Kernel Accuracy: 67.5\n",
      "RBF Kernel Accuracy: 72.5\n",
      "--------------------------------------------------\n",
      "256 128 X_train.shape (40, 35, 31)\n",
      "256 128 X_test.shape (40, 35, 31)\n",
      "y_train_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "y_test_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "Linear SVM Training Time (W:256-O:128): 0.0012018680572509766 seconds\n",
      "Polynomial SVM Training Time (W:256-O:128): 0.0011620521545410156 seconds\n",
      "RBF SVM Training Time (W:256-O:128): 0.0012137889862060547 seconds\n",
      "Linear SVM Metrics:\n",
      "Precision: 67.5, Recall: 72.5, F1 Score: 67.0, Accuracy: 72.5\n",
      "\n",
      "Polynomial SVM Metrics:\n",
      "Precision: 66.66666666666666, Recall: 72.5, F1 Score: 66.33333333333333, Accuracy: 72.5\n",
      "\n",
      "RBF SVM Metrics:\n",
      "Precision: 70.0, Recall: 72.5, F1 Score: 68.16666666666666, Accuracy: 72.5\n",
      "--------------------------------------------------\n",
      "Linear Kernel Accuracy: 72.5\n",
      "Polynomial Kernel Accuracy: 72.5\n",
      "RBF Kernel Accuracy: 72.5\n",
      "--------------------------------------------------\n",
      "256 250 X_train.shape (40, 35, 641)\n",
      "256 250 X_test.shape (40, 35, 641)\n",
      "y_train_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "y_test_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "Linear SVM Training Time (W:256-O:250): 0.058225154876708984 seconds\n",
      "Polynomial SVM Training Time (W:256-O:250): 0.045758724212646484 seconds\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RBF SVM Training Time (W:256-O:250): 0.059877872467041016 seconds\n",
      "Linear SVM Metrics:\n",
      "Precision: 76.66666666666666, Recall: 82.5, F1 Score: 78.83333333333333, Accuracy: 82.5\n",
      "\n",
      "Polynomial SVM Metrics:\n",
      "Precision: 74.16666666666666, Recall: 77.5, F1 Score: 74.0, Accuracy: 77.5\n",
      "\n",
      "RBF SVM Metrics:\n",
      "Precision: 76.66666666666666, Recall: 82.5, F1 Score: 78.83333333333333, Accuracy: 82.5\n",
      "--------------------------------------------------\n",
      "Linear Kernel Accuracy: 82.5\n",
      "Polynomial Kernel Accuracy: 77.5\n",
      "RBF Kernel Accuracy: 82.5\n",
      "--------------------------------------------------\n",
      "256 250 X_train.shape (40, 35, 641)\n",
      "256 250 X_test.shape (40, 35, 641)\n",
      "y_train_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "y_test_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "Linear SVM Training Time (W:256-O:250): 0.0450894832611084 seconds\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Polynomial SVM Training Time (W:256-O:250): 0.15875458717346191 seconds\n",
      "RBF SVM Training Time (W:256-O:250): 0.07480764389038086 seconds\n",
      "Linear SVM Metrics:\n",
      "Precision: 76.66666666666666, Recall: 85.0, F1 Score: 79.66666666666666, Accuracy: 85.0\n",
      "\n",
      "Polynomial SVM Metrics:\n",
      "Precision: 75.0, Recall: 82.5, F1 Score: 77.0, Accuracy: 82.5\n",
      "\n",
      "RBF SVM Metrics:\n",
      "Precision: 74.16666666666666, Recall: 80.0, F1 Score: 74.66666666666666, Accuracy: 80.0\n",
      "--------------------------------------------------\n",
      "Linear Kernel Accuracy: 85.0\n",
      "Polynomial Kernel Accuracy: 82.5\n",
      "RBF Kernel Accuracy: 80.0\n",
      "--------------------------------------------------\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "512 0 X_train.shape (40, 70, 8)\n",
      "512 0 X_test.shape (40, 70, 8)\n",
      "y_train_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "y_test_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "Linear SVM Training Time (W:512-O:0): 0.0009639263153076172 seconds\n",
      "Polynomial SVM Training Time (W:512-O:0): 0.0008749961853027344 seconds\n",
      "RBF SVM Training Time (W:512-O:0): 0.0008256435394287109 seconds\n",
      "Linear SVM Metrics:\n",
      "Precision: 76.16666666666666, Recall: 75.0, F1 Score: 72.52380952380953, Accuracy: 75.0\n",
      "\n",
      "Polynomial SVM Metrics:\n",
      "Precision: 74.16666666666666, Recall: 72.5, F1 Score: 70.66666666666667, Accuracy: 72.5\n",
      "\n",
      "RBF SVM Metrics:\n",
      "Precision: 73.66666666666666, Recall: 72.5, F1 Score: 70.02380952380953, Accuracy: 72.5\n",
      "--------------------------------------------------\n",
      "Linear Kernel Accuracy: 75.0\n",
      "Polynomial Kernel Accuracy: 72.5\n",
      "RBF Kernel Accuracy: 72.5\n",
      "--------------------------------------------------\n",
      "512 0 X_train.shape (40, 70, 8)\n",
      "512 0 X_test.shape (40, 70, 8)\n",
      "y_train_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "y_test_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "Linear SVM Training Time (W:512-O:0): 0.0008432865142822266 seconds\n",
      "Polynomial SVM Training Time (W:512-O:0): 0.0008461475372314453 seconds\n",
      "RBF SVM Training Time (W:512-O:0): 0.0008034706115722656 seconds\n",
      "Linear SVM Metrics:\n",
      "Precision: 77.5, Recall: 72.5, F1 Score: 71.0, Accuracy: 72.5\n",
      "\n",
      "Polynomial SVM Metrics:\n",
      "Precision: 74.16666666666666, Recall: 70.0, F1 Score: 68.33333333333333, Accuracy: 70.0\n",
      "\n",
      "RBF SVM Metrics:\n",
      "Precision: 71.16666666666667, Recall: 67.5, F1 Score: 64.85714285714286, Accuracy: 67.5\n",
      "--------------------------------------------------\n",
      "Linear Kernel Accuracy: 72.5\n",
      "Polynomial Kernel Accuracy: 70.0\n",
      "RBF Kernel Accuracy: 67.5\n",
      "--------------------------------------------------\n",
      "512 256 X_train.shape (40, 70, 15)\n",
      "512 256 X_test.shape (40, 70, 15)\n",
      "y_train_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "y_test_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "Linear SVM Training Time (W:512-O:256): 0.00115966796875 seconds\n",
      "Polynomial SVM Training Time (W:512-O:256): 0.001171112060546875 seconds\n",
      "RBF SVM Training Time (W:512-O:256): 0.0012090206146240234 seconds\n",
      "Linear SVM Metrics:\n",
      "Precision: 84.16666666666666, Recall: 85.0, F1 Score: 83.0, Accuracy: 85.0\n",
      "\n",
      "Polynomial SVM Metrics:\n",
      "Precision: 83.33333333333333, Recall: 82.5, F1 Score: 80.66666666666666, Accuracy: 82.5\n",
      "\n",
      "RBF SVM Metrics:\n",
      "Precision: 80.83333333333333, Recall: 82.5, F1 Score: 80.16666666666667, Accuracy: 82.5\n",
      "--------------------------------------------------\n",
      "Linear Kernel Accuracy: 85.0\n",
      "Polynomial Kernel Accuracy: 82.5\n",
      "RBF Kernel Accuracy: 82.5\n",
      "--------------------------------------------------\n",
      "512 256 X_train.shape (40, 70, 15)\n",
      "512 256 X_test.shape (40, 70, 15)\n",
      "y_train_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "y_test_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "Linear SVM Training Time (W:512-O:256): 0.0011463165283203125 seconds\n",
      "Polynomial SVM Training Time (W:512-O:256): 0.0011491775512695312 seconds\n",
      "RBF SVM Training Time (W:512-O:256): 0.0011975765228271484 seconds\n",
      "Linear SVM Metrics:\n",
      "Precision: 78.92857142857143, Recall: 75.0, F1 Score: 73.38888888888889, Accuracy: 75.0\n",
      "\n",
      "Polynomial SVM Metrics:\n",
      "Precision: 76.66666666666666, Recall: 75.0, F1 Score: 72.16666666666666, Accuracy: 75.0\n",
      "\n",
      "RBF SVM Metrics:\n",
      "Precision: 75.33333333333333, Recall: 75.0, F1 Score: 72.19047619047618, Accuracy: 75.0\n",
      "--------------------------------------------------\n",
      "Linear Kernel Accuracy: 75.0\n",
      "Polynomial Kernel Accuracy: 75.0\n",
      "RBF Kernel Accuracy: 75.0\n",
      "--------------------------------------------------\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "512 511 X_train.shape (40, 70, 3585)\n",
      "512 511 X_test.shape (40, 70, 3585)\n",
      "y_train_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "y_test_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "Linear SVM Training Time (W:512-O:511): 0.21903276443481445 seconds\n",
      "Polynomial SVM Training Time (W:512-O:511): 0.22972679138183594 seconds\n",
      "RBF SVM Training Time (W:512-O:511): 0.25325775146484375 seconds\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Linear SVM Metrics:\n",
      "Precision: 86.66666666666666, Recall: 87.5, F1 Score: 85.50000000000001, Accuracy: 87.5\n",
      "\n",
      "Polynomial SVM Metrics:\n",
      "Precision: 85.83333333333334, Recall: 85.0, F1 Score: 82.66666666666666, Accuracy: 85.0\n",
      "\n",
      "RBF SVM Metrics:\n",
      "Precision: 81.66666666666667, Recall: 87.5, F1 Score: 83.33333333333334, Accuracy: 87.5\n",
      "--------------------------------------------------\n",
      "Linear Kernel Accuracy: 87.5\n",
      "Polynomial Kernel Accuracy: 85.0\n",
      "RBF Kernel Accuracy: 87.5\n",
      "--------------------------------------------------\n",
      "512 511 X_train.shape (40, 70, 3585)\n",
      "512 511 X_test.shape (40, 70, 3585)\n",
      "y_train_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "y_test_idx [ 0  0  1  1  2  2  3  3  4  4  5  5  6  6  7  7  8  8  9  9 10 10 11 11\n",
      " 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19]\n",
      "Linear SVM Training Time (W:512-O:511): 0.32654285430908203 seconds\n",
      "Polynomial SVM Training Time (W:512-O:511): 0.23546195030212402 seconds\n",
      "RBF SVM Training Time (W:512-O:511): 0.25396060943603516 seconds\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/user/yahya/other/listener_cls_ffr/venv/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Linear SVM Metrics:\n",
      "Precision: 80.0, Recall: 85.0, F1 Score: 81.33333333333333, Accuracy: 85.0\n",
      "\n",
      "Polynomial SVM Metrics:\n",
      "Precision: 75.83333333333333, Recall: 77.5, F1 Score: 73.5, Accuracy: 77.5\n",
      "\n",
      "RBF SVM Metrics:\n",
      "Precision: 90.83333333333331, Recall: 87.5, F1 Score: 86.83333333333334, Accuracy: 87.5\n",
      "--------------------------------------------------\n",
      "Linear Kernel Accuracy: 85.0\n",
      "Polynomial Kernel Accuracy: 77.5\n",
      "RBF Kernel Accuracy: 87.5\n",
      "--------------------------------------------------\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1IAAANXCAYAAAA/6260AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAA9hAAAPYQGoP6dpAABvtUlEQVR4nO3deXxN1/7/8fdJIhMSQyYxJGmoWWsoNafmoVQp11A1laqZVnFLFTVWW60qt701tIS2iotb81xFDaVXtaiaSVAkpkgk+/dHf863pwmyNHFO5PV8PM7jkbP22ut8ztateVt7r22zLMsSAAAAACDd3JxdAAAAAABkNQQpAAAAADBEkAIAAAAAQwQpAAAAADBEkAIAAAAAQwQpAAAAADBEkAIAAAAAQwQpAAAAADBEkAIAAAAAQwQpAAAesI0bN8pms2njxo0uPeaDEBUVpaioKGeXAQDGCFIAXM7s2bNls9nsLw8PDxUsWFCdO3fW6dOnU/WPiopy6O/p6amIiAj16NFDJ0+evOvYf34NHTr0nrUtW7ZMtWvXVlBQkHx9ffXII4+oTZs2WrlypSTp3Xfflc1m09q1a+84xieffCKbzaalS5c61F+sWLE0+69Zs8Ze48KFC+9Z420///yzbDabvL29dfny5XTvB2R3x44du+PfEwsWLHDo+/3336tXr16qWLGicuTIIZvNluaYJ0+e1KhRo1S5cmXlzZtXAQEBioqKuuvfFQBcm4ezCwCAOxk9erQiIiKUkJCg7du3a/bs2fr222+1f/9+eXt7O/QtVKiQxo8fL0lKTEzUgQMHNGPGDK1atUo///yzfH190xz7z8qUKXPXeiZPnqzBgwerdu3aGjZsmHx9ffXrr79q7dq1WrBggRo1aqS2bdtq8ODBio6OVr169dIcJzo6Wvnz51fjxo3tbd7e3vr111/1/fffq3Llyg79582bJ29vbyUkJNz9gP3F3LlzFRISokuXLmnhwoV68cUXjfZH5qlVq5Zu3LghT09PZ5fidKtXr3Z2CXfUrl07NWnSxKGtatWqDu+/+eYb/fvf/1a5cuX0yCOP6NChQ2mO9Z///EcTJ05UixYt1KlTJ926dUufffaZ6tevr5kzZ6pLly6Z9j0AZBILAFzMrFmzLEnWzp07HdqHDBliSbK++OILh/batWtbpUuXTjXOhx9+aEmyVq9efc+x7yUpKcny8/Oz6tevn+b22NhY+89169a1/P39rYSEhFT9Tp06Zbm5uVk9e/ZMVX/x4sWtAQMGOPS/ceOG5efnZ7Vq1cqSZH311VfpqjclJcUKDw+3Bg0aZD377LNWVFRUuvZzhqtXrzq7hIfChg0bLEnWhg0bnF1Klnf06FFLkvX222/fs29MTIx1/fp1y7Isq3fv3tadfrXav3+/df78eYe2hIQEq0SJElahQoX+ftEAHjgu7QOQZdSsWVOSdOTIkXT1DwkJkSR5ePz9yfcLFy4oPj5e1atXT3N7UFCQ/efnn39ecXFx+u9//5uq34IFC5SSkqIOHTqk2tauXTt98cUXSklJsbctW7ZM169fV5s2bYzq3bp1q44dO6a2bduqbdu22rx5s06dOpWqX0pKit5//32VLVtW3t7eCgwMVKNGjbRr1y6HfnPnzlXlypXl6+urvHnzqlatWg4zCTabTW+++Waq8cPDw9W5c2f7+9uXVm7atEm9evVSUFCQChUqJEk6fvy4evXqpeLFi8vHx0f58+dX69atdezYsVTjXr58WQMHDlR4eLi8vLxUqFAhvfDCC7pw4YKuXr2qnDlzqn///qn2O3XqlNzd3e2zl39Xy5YtVaFCBYe2Zs2aOVy6KUk7duyQzWbTihUrJKV9P1NUVJTKlCmjAwcO6KmnnpKvr68KFiyoSZMmpfk9WrRooZw5cyooKEgDBw7UzZs306zxq6++UsWKFeXj46OAgAA9//zzDpfILl26VDabTT/++KO97euvv5bNZlPLli0dxipZsqT+8Y9/pPv4xMTEqEuXLipUqJC8vLxUoEABPfPMMw5/pn+9Ryo8PPyOl9X9+XidPn1aXbt2VXBwsLy8vFS6dGnNnDkz3bWl17Vr15SYmHjH7cHBwfLx8bnnOKVLl1ZAQIBDm5eXl5o0aaJTp07pypUrf7tWAA8WQQpAlnH7l6+8efOm2pacnKwLFy7owoULOnv2rNavX6+RI0eqaNGiaYafuLg4e//br7sJCgqSj4+Pli1bposXL961b8uWLeXt7a3o6OhU26KjoxUWFpZmTe3bt9fZs2cdflmMjo5W3bp1HYJaesybN0+RkZF64okn1KxZM/n6+mr+/Pmp+nXr1k0DBgxQ4cKFNXHiRA0dOlTe3t7avn27vc+oUaPUsWNH5ciRQ6NHj9aoUaNUuHBhrV+/3qimP+vVq5cOHDigN954w35v2s6dO/Xdd9+pbdu2+uCDD9SzZ0+tW7dOUVFRun79un3fq1evqmbNmpo6daoaNGig999/Xz179tQvv/yiU6dOKVeuXHr22Wf1xRdfKDk52eFz58+fL8uy0gyy96NmzZrat2+f4uPjJUmWZWnr1q1yc3PTli1b7P22bNkiNze3Owbx2y5duqRGjRrpscce0zvvvKMSJUpoyJAh9gAmSTdu3FDdunW1atUq9enTR6+//rq2bNmi1157LdV4s2fPVps2bezhsXv37lq0aJFq1Khhv2+uRo0astls2rx5c6p6v/32W3vb+fPn9csvv6hWrVrpPj6tWrXS4sWL1aVLF3300Ufq16+frly5ohMnTtxxnylTpujzzz93eFWoUEFubm7Knz+/JCk2NlZPPvmk1q5dqz59+uj9999X0aJF1a1bN02ZMsVhvL+e53d6pRVER40apVy5csnb21tPPPFEplyGGBMTI19f31SXHwPIApw9JQYAf3X78ru1a9da58+ft06ePGktXLjQCgwMtLy8vKyTJ0869K9du7YlKdWrZMmS1m+//Zbm2Gm97uWNN96wJFk5c+a0GjdubI0dO9bavXt3mn1bt25teXt7W3Fxcfa2X375xZJkDRs2LFX9ty9NrFSpktWtWzfLsizr0qVLlqenpzVnzhz7ZVvpubQvMTHRyp8/v/X666/b29q3b2899thjDv3Wr19vSbL69euXaoyUlBTLsizr8OHDlpubm/Xss89aycnJafaxLMuSZI0cOTLVOGFhYVanTp3s728f/xo1ali3bt1y6Hv78qg/27ZtmyXJ+uyzz+xtt/8cFi1adMe6V61aZUmyVqxY4bC9XLlyVu3atVPtd7927txpSbK++eYby7Is68cff7QkWa1bt7aqVKli79e8eXOrfPny9vdpXYZ3+7/jP3/XmzdvWiEhIVarVq3sbVOmTLEkWV9++aW97dq1a1bRokUdxkxMTLSCgoKsMmXKWDdu3LD3Xb58uSXJeuONN+xtpUuXttq0aWN/X6FCBat169aWJOvnn3+2LMuyFi1aZEmy9u3bl65jc+nSpXRdHle7du27/pl8+eWXliRr9OjR9rZu3bpZBQoUsC5cuODQt23btpa/v7/Df0t3Ot//+po1a5Z9n+PHj1sNGjSwpk+fbi1dutSaMmWKVaRIEcvNzc1avnz5HWu926V9aTl8+LDl7e1tdezYMd37AHAdzEgBcFn16tVTYGCgChcurOeee045c+bU0qVL7ZeC/Vl4eLjWrFmjNWvWaMWKFZoyZYri4uLUuHFjnT9/PlX/adOm2fvfft3LqFGjFB0drfLly2vVqlV6/fXXVbFiRVWoUEE///yzQ9/nn39eCQkJWrRokb3t9gzV3WZD2rdvr0WLFikxMVELFy6Uu7u7nn322XvW9mcrVqzQ77//rnbt2tnb2rVrp3379umnn36yt92+fGvkyJGpxri98tiSJUuUkpKiN954Q25ubmn2uR/du3eXu7u7Q9ufL49KSkrS77//rqJFiypPnjzas2ePQ92PPfZYmsfldk316tVTaGio5s2bZ9+2f/9+/fjjj3r++efvu+6/Kl++vHLlymWfzdmyZYv9MsM9e/bo+vXrsixL3377rf3S1LvJlSuXQ32enp6qXLmyfvvtN3vbN998owIFCui5556zt/n6+qpHjx4OY+3atUvnzp1Tr169HBZnadq0qUqUKOFw6WnNmjXtM2hXrlzRvn371KNHDwUEBNjbt2zZojx58txzUZbbfHx85OnpqY0bN+rSpUvp2uevDhw4oK5du+qZZ57R8OHDJf0x6/f111+rWbNmsizLYVapYcOGiouLc/jv5a/n+Z1eDRs2tO9TpEgRrVq1Sj179lSzZs3Uv39//fDDDwoMDNQrr7xyX9/lr65fv67WrVvLx8dHEyZMyJAxATxYrNoHwGVNmzZNjz76qOLi4jRz5kxt3rxZXl5eafbNmTOnwyp5jRo1Uo0aNVSpUiVNmDBB77zzjkP/ypUrq1KlSsY1tWvXTu3atVN8fLx27Nih2bNnKzo6Ws2aNXNYTbBx48bKly+foqOj7fcIzZ8/X4899phKly59x/Hbtm2rV199VStWrNC8efP09NNPK3fu3EY1zp07VxEREfLy8tKvv/4qSYqMjJSvr6/mzZuncePGSfrjXrPQ0FDly5fvjmMdOXJEbm5uKlWqlFEN9/LXFROlPy5ZGz9+vGbNmqXTp0/Lsiz7tri4OIeaWrVqddfx3dzc1KFDB02fPl3Xr1+3f3dvb2+1bt36rvueP3/e4ZLAXLlyKVeuXGn2dXd3V9WqVR3CRs2aNVWjRg0lJydr+/btCg4O1sWLF9MVpAoVKpQqoObNm9fh/qXjx4+raNGiqfoVL17c4f3x48fTbJekEiVKOFy2V7NmTc2YMUO//vqrjhw5IpvNpqpVq9oDVvfu3bVlyxZVr149VaC+Ey8vL02cOFGvvPKKgoOD9eSTT+rpp5/WCy+8YL9/8W7i4+PVsmVLFSxYUJ999pn9+54/f16XL1/Wxx9/rI8//jjNfc+dO2f/+U6rZ5rKly+funTpogkTJujUqVNp/oNOeiUnJ6tt27Y6cOCAVqxYodDQ0AypEcCDxYwUAJdVuXJl1atXT61atdLSpUtVpkwZtW/fXlevXk3X/hUrVpS/v7/DvR8Zxc/PT/Xr19e8efPUqVMnHTlyRDt27LBvz5Ejh9q0aaP169crNjZWO3fu1OHDh+95b06BAgUUFRWld955R5s3b1b79u2N6oqPj9eyZct09OhRFStWzP4qVaqUrl+/rujoaIeAktn+eo/SbWndnN+3b1+NHTtWbdq00ZdffqnVq1drzZo1yp8/v8MCHOn1wgsv6OrVq1qyZIksy1J0dLSefvpp+fv733W/J554QgUKFLC/Jk+efNf+NWrU0M6dO5WQkGAPUrdnbrZs2WIPWekJUn+dpbsts//MatSoIUnavHmztmzZogoVKihnzpz2IHX16lX98MMP6foOfzZgwAAdOnRI48ePl7e3t0aMGKGSJUvqhx9+uOe+nTt31pkzZ7RkyRL5+fnZ22//t/D888/fcXbpz/eixcTEpOt148aNe9ZUuHBhSbrnfZL30r17dy1fvlyzZ89WnTp1/tZYAJyHGSkAWcLtm+Wfeuopffjhh+l6eK70xy/y6Q1e96tSpUqaM2eOzp4969DeoUMHzZgxQ1988YWOHj0qm83mcLndnbRv314vvvii8uTJk+oZNveyaNEiJSQkaPr06alWCDt48KCGDx+urVu3qkaNGoqMjNSqVat08eLFO85KRUZGKiUlRQcOHNDjjz9+x8/Nmzdvqof+JiYmpjomd7Nw4UJ16tTJYfYwISEh1biRkZHav3//PccrU6aMypcvr3nz5qlQoUI6ceKEpk6des/95s2b5/BL9SOPPHLX/jVr1lRiYqLmz5+v06dP28NGrVq1tGXLFgUHB+vRRx9VcHDwPT87PcLCwrR//35ZluUwK3Xw4MFU/W63//WX9YMHD9q3S39cylakSBFt2bJFv/32m8N3GDRokL766islJycbLTRxW2RkpF555RW98sorOnz4sB5//HG98847mjt37h33mTBhgpYsWaJFixapRIkSDtsCAwOVO3duJScnp2u2qUCBAumqc9asWQ4rTKbl9iWWgYGB6RozLYMHD9asWbM0ZcqUdP19AMB1EaQAZBlRUVGqXLmypkyZogEDBqR6KO9fbdiwQVevXtVjjz32tz/7+vXr2rdvX6qHcUqyr6j210uoqlevrvDwcM2dO1enTp1S7dq103U50HPPPaeTJ0+qePHixg9snTt3rh555BH17Nkz1babN29qwoQJmjdvnmrUqKFWrVpp2rRpGjVqlN5//32Hvrd/SW/RooWGDBmi0aNHa+HChQ6Xdf35F/nIyMhUM38ff/zxHWek0uLu7p5q5mXq1KmpxmjVqpVGjx6txYsXp7pP6q/homPHjnrttdfk5eWV6iHId3KvlfX+qkqVKsqRI4cmTpyofPny2S/drFmzpmbNmqU8efKoUaNGRmPeTZMmTbR69WotXLjQfpni9evXU13mVqlSJQUFBWnGjBnq2rWr/bLYFStW6Oeff9Ybb7zh0L9mzZpav369zp07p0GDBkmSHn/8ceXOnVsTJkyQj4+PKlasmO46r1+/Ljc3N4fzNDIyUrlz577jUu2StHbtWg0fPlyvv/66WrRokWq7u7u7WrVqpejoaO3fvz/VPVvnz593CDrpuf9RksMlt38dQ/pjufWZM2eqXLly6Q5nf/X2229r8uTJ+uc//5nm8vwAshaCFIAsZfDgwWrdurVmz57tEBbi4uLs/8J969YtHTx4UNOnT5ePj0+6Z6/u5vr166pWrZqefPJJNWrUSIULF9bly5e1ZMkSbdmyRS1atFD58uUd9rHZbGrfvr39nqTRo0en67P8/f3TfCbTvZw5c0YbNmxQv3790tzu5eWlhg0b6quvvtIHH3ygp556Sh07dtQHH3ygw4cPq1GjRkpJSdGWLVv01FNPqU+fPipatKhef/11jRkzRjVr1lTLli3l5eWlnTt3KjQ01P48phdffFE9e/ZUq1atVL9+fe3bt0+rVq1KNSt2N08//bQ+//xz+fv7q1SpUtq2bZvWrl1rX/L6tsGDB9tDRNeuXVWxYkVdvHhRS5cu1YwZMxyCc/v27fXaa69p8eLFevnll5UjRw7j43ovvr6+qlixorZv325/hpT0x2zOtWvXdO3aNeNL4u6me/fu+vDDD/XCCy9o9+7dKlCggD7//PNUy2ffDnddunRR7dq11a5dO8XGxur9999XeHi4Bg4c6NC/Zs2amjdvnmw2m/1SP3d3d1WrVk2rVq1SVFSUUbA/dOiQ6tatqzZt2qhUqVLy8PDQ4sWLFRsbq7Zt295xv3bt2ikwMFDFihVLNWtVv359BQcHa8KECdqwYYOqVKmi7t27q1SpUrp48aL27NmjtWvXOlx6dz/3SL322ms6cuSI6tatq9DQUB07dkz/+te/dO3atVT/6HD8+HF9/vnnkmR//tpbb70l6Y9ZwY4dO0qSFi9erNdee03FihVTyZIl7/jdAGQhzlksEADu7PYS2Tt37ky1LTk52YqMjLQiIyPty2f/dflzm81m5cuXz2revHmq5cnvNvbdJCUlWZ988onVokULKywszPLy8rJ8fX2t8uXLW2+//bZ18+bNNPf76aefLEmWl5eXdenSpTT7/Hn58ztJz/Ln77zzjiXJWrdu3R37zJ4925Jk/ec//7Esy7Ju3bplvf3221aJEiUsT09PKzAw0GrcuHGq4zZz5kyrfPnylpeXl5U3b16rdu3a1po1a+zbk5OTrSFDhlgBAQGWr6+v1bBhQ+vXX3+94/LnaR3/S5cuWV26dLECAgKsXLlyWQ0bNrR++eWXVGNYlmX9/vvvVp8+fayCBQtanp6eVqFChaxOnTqlWg7bsiyrSZMmliTru+++u+Nx+bsGDx5sSbImTpzo0H57SfIjR444tN9p+fO0/jvo1KmTFRYW5tB2/Phxq3nz5pavr68VEBBg9e/f31q5cmWqMS3Lsr744gv7n12+fPmsDh06WKdOnUr1Obf/Wy1ZsqRD+1tvvWVJskaMGJGOI/F/Lly4YPXu3dsqUaKElTNnTsvf39+qUqWKw7Ltt7/3n5c/112WKP/zd4uNjbV69+5tFS5c2MqRI4cVEhJi1a1b1/r444+N6kxLdHS0VatWLSswMNDy8PCwAgICrGeffTbNxx3c/rNM6/Xn7zVy5Mh0fzcAWYPNsh7gXccAADxgzz77rP73v//ZVzAEACAjsGofAOChdfbsWf33v/+1X14FAEBG4R4pAMBD5+jRo9q6dav+/e9/K0eOHHrppZecXdJDJS4u7p7LhafnWVEAkJURpAAAD51NmzapS5cuKlKkiObMmcMv9Rmsf//+mjNnzl37cOcAgIedU++R2rx5s95++23t3r1bZ8+e1eLFix2WOrUsSyNHjtQnn3yiy5cvq3r16po+fbqKFStm73Px4kX17dtXy5Ytk5ubm1q1aqX333//jk+hBwAAf8+BAwd05syZu/a5n9XyACArceqM1LVr1/TYY4+pa9euatmyZartkyZN0gcffKA5c+YoIiJCI0aMUMOGDXXgwAH7cyk6dOigs2fPas2aNUpKSlKXLl3Uo0cPRUdHP+ivAwBAtlCqVCmVKlXK2WUAgFO5zKp9NpvNYUbKsiyFhobqlVde0auvvirpj2uyg4ODNXv2bLVt21Y///yzSpUqpZ07d6pSpUqSpJUrV6pJkyY6deqUQkNDnfV1AAAAADzEXPYeqaNHjyomJsbh0gB/f39VqVJF27ZtU9u2bbVt2zblyZPHHqKkPy4lcHNz044dO1I98f62mzdvOjxVPSUlRRcvXlT+/PntD1IEAAAAkP1YlqUrV64oNDRUbm53XuTcZYNUTEyMJKV6yndwcLB9W0xMjIKCghy2e3h4KF++fPY+aRk/frxGjRqVwRUDAAAAeFicPHlShQoVuuN2lw1SmWnYsGEaNGiQ/X1cXJyKFCmikydPys/Pz4mVAQAAAHCm+Ph4FS5cWLlz575rP5cNUreXqo2NjVWBAgXs7bGxsXr88cftfc6dO+ew361bt3Tx4sW7LnXr5eUlLy+vVO1+fn4EKQAAAAD3vOXnzhf9OVlERIRCQkK0bt06e1t8fLx27NihqlWrSpKqVq2qy5cva/fu3fY+69evV0pKiqpUqfLAawYAAACQPTh1Rurq1av69ddf7e+PHj2qvXv3Kl++fCpSpIgGDBigt956S8WKFbMvfx4aGmpf2a9kyZJq1KiRunfvrhkzZigpKUl9+vRR27ZtWbEPAAAAQKZxapDatWuXnnrqKfv72/ctderUSbNnz9Zrr72ma9euqUePHrp8+bJq1KihlStX2p8hJUnz5s1Tnz59VLduXfsDeT/44IMH/l0AAAAAZB8u8xwpZ4qPj5e/v7/i4uK4RwoAAADIxtKbDVz2HikAAAAAcFUEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAAAAAEMEKQAAAAAwRJACAADIBpKTkzVixAhFRETIx8dHkZGRGjNmjCzLsve5evWq+vTpo0KFCsnHx0elSpXSjBkznFh1xsnu31/iGGQ0D2cXAAAAgMw3ceJETZ8+XXPmzFHp0qW1a9cudenSRf7+/urXr58kadCgQVq/fr3mzp2r8PBwrV69Wr169VJoaKiaN2/u5G/w92T37y9xDDIaM1IAAADZwHfffadnnnlGTZs2VXh4uJ577jk1aNBA33//vUOfTp06KSoqSuHh4erRo4cee+wxhz5ZVXb//hLHIKMRpAAAALKBatWqad26dTp06JAkad++ffr222/VuHFjhz5Lly7V6dOnZVmWNmzYoEOHDqlBgwbOKjvDZPfvL3EMMhqX9gEAAGQDQ4cOVXx8vEqUKCF3d3clJydr7Nix6tChg73P1KlT1aNHDxUqVEgeHh5yc3PTJ598olq1ajmx8oyR3b+/xDHIaAQpAACAbODLL7/UvHnzFB0drdKlS2vv3r0aMGCAQkND1alTJ0l//BK9fft2LV26VGFhYdq8ebN69+6t0NBQ1atXz8nf4O/J7t9f4hhkNJv152U6sqn4+Hj5+/srLi5Ofn5+zi4HAAAgwxUuXFhDhw5V79697W1vvfWW5s6dq19++UU3btyQv7+/Fi9erKZNm9r7vPjiizp16pRWrlzpjLIzTHb//hLHIL3Smw24RwoAACAbuH79utzcHH/1c3d3V0pKiiQpKSlJSUlJd+2TlWX37y9xDDIal/YBAABkA82aNdPYsWNVpEgRlS5dWj/88IPeffddde3aVZLk5+en2rVra/DgwfLx8VFYWJg2bdqkzz77TO+++66Tq//7svv3lzgGGY1L+8SlfQAA4OF35coVjRgxQosXL9a5c+cUGhqqdu3a6Y033pCnp6ckKSYmRsOGDdPq1at18eJFhYWFqUePHho4cKBsNpuTv8Hfk92/v8QxSK/0ZgOClAhSAAAAAP7APVIAAAAAkEkIUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiAfyAgAAuIBs8oieO7LmZe8DYDvs7AqcyxqZ9Z7IxIwUAAAAABgiSAEAAACAIYIUAAAAABgiSAEAAACAIYIUAAAAABgiSAEAAACAIYIUAAAAABgiSAEAAACAIYIUAAAAABgiSAEAAACAIYIUAAAAABgiSAEAAACAIYIUAAAAABgiSAEAAACAIYIUAAAAABgiSAEAAACAIYIUAAAAABgiSAEAAACAIYIUAAAAABgiSAEAAACAIYIUAAAAABgiSAFANpKcnKwRI0YoIiJCPj4+ioyM1JgxY2RZliQpKSlJQ4YMUdmyZZUzZ06FhobqhRde0JkzZ5xcOQAArsXD2QUAAB6ciRMnavr06ZozZ45Kly6tXbt2qUuXLvL391e/fv10/fp17dmzRyNGjNBjjz2mS5cuqX///mrevLl27drl7PIBAHAZBCkAyEa+++47PfPMM2ratKkkKTw8XPPnz9f3338vSfL399eaNWsc9vnwww9VuXJlnThxQkWKFHngNQMA4Iq4tA8AspFq1app3bp1OnTokCRp3759+vbbb9W4ceM77hMXFyebzaY8efI8oCoBAHB9zEgBQDYydOhQxcfHq0SJEnJ3d1dycrLGjh2rDh06pNk/ISFBQ4YMUbt27eTn5/eAqwUAwHURpAAgG/nyyy81b948RUdHq3Tp0tq7d68GDBig0NBQderUyaFvUlKS2rRpI8uyNH36dCdVDACAayJIAUA2MnjwYA0dOlRt27aVJJUtW1bHjx/X+PHjHYLU7RB1/PhxrV+/ntkoAAD+giAFANnI9evX5ebmeHusu7u7UlJS7O9vh6jDhw9rw4YNyp8//4MuEwAAl0eQAoBspFmzZho7dqyKFCmi0qVL64cfftC7776rrl27SvojRD333HPas2ePli9fruTkZMXExEiS8uXLJ09PT2eWDwCAy7BZt5/CmI3Fx8fL399fcXFxXL4C4KF25coVjRgxQosXL9a5c+cUGhqqdu3a6Y033pCnp6eOHTumiIiINPfdsGGDoqKiHmzBQDZiszm7Auey5mXvA2A77OwKnMsa6TqRJL3ZgCAlghQAAHA+glT2PgAEKdeJJOnNBjxHCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAM8RwpAHAh2X7VLtdZtAkAgLtiRgoAAAAADBGkAAAAAMAQQQoAAAAADBGkAAAAAMAQQQoAAAAADBGkAAAAAMAQQQoAAAAADBGkAAAAAMAQQQoAAAAADBGkAAAAAMAQQQoAAAAADBGkAAAAAMAQQQoAAAAADBGkAADIRsLDw2Wz2VK9evfuLUk6cuSInn32WQUGBsrPz09t2rRRbGysk6sGANdDkAIAIBvZuXOnzp49a3+tWbNGktS6dWtdu3ZNDRo0kM1m0/r167V161YlJiaqWbNmSklJcXLlAOBaXDpIJScna8SIEYqIiJCPj48iIyM1ZswYWZZl72NZlt544w0VKFBAPj4+qlevng4fPuzEqgEAcF2BgYEKCQmxv5YvX67IyEjVrl1bW7du1bFjxzR79myVLVtWZcuW1Zw5c7Rr1y6tX7/e2aUDgEtx6SA1ceJETZ8+XR9++KF+/vlnTZw4UZMmTdLUqVPtfSZNmqQPPvhAM2bM0I4dO5QzZ041bNhQCQkJTqwcAADXl5iYqLlz56pr166y2Wy6efOmbDabvLy87H28vb3l5uamb7/91omVAoDrcekg9d133+mZZ55R06ZNFR4erueee04NGjTQ999/L+mP2agpU6Zo+PDheuaZZ1SuXDl99tlnOnPmjJYsWeLc4gEAcHFLlizR5cuX1blzZ0nSk08+qZw5c2rIkCG6fv26rl27pldffVXJyck6e/asc4sFABfj0kGqWrVqWrdunQ4dOiRJ2rdvn7799ls1btxYknT06FHFxMSoXr169n38/f1VpUoVbdu27Y7j3rx5U/Hx8Q4vAACym08//VSNGzdWaGiopD8u+/vqq6+0bNky5cqVS/7+/rp8+bIqVKggNzeX/pUBAB44D2cXcDdDhw5VfHy8SpQoIXd3dyUnJ2vs2LHq0KGDJCkmJkaSFBwc7LBfcHCwfVtaxo8fr1GjRmVe4QAAuLjjx49r7dq1WrRokUN7gwYNdOTIEV24cEEeHh7KkyePQkJC9MgjjzipUgBwTS79z0tffvml5s2bp+joaO3Zs0dz5szR5MmTNWfOnL817rBhwxQXF2d/nTx5MoMqBgAga5g1a5aCgoLUtGnTNLcHBAQoT548Wr9+vc6dO6fmzZs/4AoBwLW59IzU4MGDNXToULVt21aSVLZsWR0/flzjx49Xp06dFBISIkmKjY1VgQIF7PvFxsbq8ccfv+O4Xl5eDjfSAgCQnaSkpGjWrFnq1KmTPDwcfxWYNWuWSpYsqcDAQG3btk39+/fXwIEDVbx4cSdVCwCuyaVnpK5fv57qmmx3d3f7sywiIiIUEhKidevW2bfHx8drx44dqlq16gOtFQCArGLt2rU6ceKEunbtmmrbwYMH1aJFC5UsWVKjR4/W66+/rsmTJzuhSgBwbS49I9WsWTONHTtWRYoUUenSpfXDDz/o3Xfftf/Fb7PZNGDAAL311lsqVqyYIiIiNGLECIWGhqpFixbOLR4AABfVoEEDh2cy/tmECRM0YcKEB1wRAGQ9Lh2kpk6dqhEjRqhXr146d+6cQkND9dJLL+mNN96w93nttdd07do19ejRQ5cvX1aNGjW0cuVKeXt7O7FyAAAAAA8zm3Wnf5LKRuLj4+Xv76+4uDj5+fk5uxwA2ZjN5uwKnIv/IyE7y/bn/7zsfQBsh51dgXNZI13nfwDpzQYufY8UAAAAALgighQAAAAAGCJIAQAAAIAhghQAAAAAGHLpVfsAAMhObKOy9832rnSzOQDcCzNSAAAAAGCIIAUAAAAAhghSAAAAAGCIIAUAAAAAhghSAAAAAGCIIAUAAAAAhghSAAAAAGCIIAUAAAAAhghSAAAAAGCIIAUAAAAAhghSAIBsJTw8XDabLdWrd+/ekqSoqKhU23r27OnkqgEArsbD2QUAAPAg7dy5U8nJyfb3+/fvV/369dW6dWt7W/fu3TV69Gj7e19f3wdaIwDA9RGkAADZSmBgoMP7CRMmKDIyUrVr17a3+fr6KiQk5EGXBgDIQri0DwCQbSUmJmru3Lnq2rWrbDabvX3evHkKCAhQmTJlNGzYMF2/ft2JVQIAXBEzUgCAbGvJkiW6fPmyOnfubG9r3769wsLCFBoaqh9//FFDhgzRwYMHtWjRIucVCgBwOQQpAEC29emnn6px48YKDQ21t/Xo0cP+c9myZVWgQAHVrVtXR44cUWRkpDPKBAC4IC7tAwBkS8ePH9fatWv14osv3rVflSpVJEm//vrrgygLAJBFEKQAANnSrFmzFBQUpKZNm9613969eyVJBQoUeABVAQCyCi7tAwBkOykpKZo1a5Y6deokD4//+1/hkSNHFB0drSZNmih//vz68ccfNXDgQNWqVUvlypVzYsUAAFdDkAIAZDtr167ViRMn1LVrV4d2T09PrV27VlOmTNG1a9dUuHBhtWrVSsOHD3dSpQAAV0WQAgBkOw0aNJBlWanaCxcurE2bNjmhIgBAVsM9UgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiFX7AACuI9rm7AoAAEgXZqQAAAAAwBBBCgAAAAAMEaSyofDwcNlstlSv3r176+LFi+rbt6+KFy8uHx8fFSlSRP369VNcXJyzywYAAABcBvdIZUM7d+5UcnKy/f3+/ftVv359tW7dWmfOnNGZM2c0efJklSpVSsePH1fPnj115swZLVy40IlVAwAAAK6DIJUNBQYGOryfMGGCIiMjVbt2bdlsNn399df2bZGRkRo7dqyef/553bp1Sx4e/CcDAAAAcGlfNpeYmKi5c+eqa9eustnSXi0rLi5Ofn5+hCgAAADg/yNIZXNLlizR5cuX1blz5zS3X7hwQWPGjFGPHj0ebGEAAACACyNIZXOffvqpGjdurNDQ0FTb4uPj1bRpU5UqVUpvvvnmgy8OAAAAcFFcq5WNHT9+XGvXrtWiRYtSbbty5YoaNWqk3Llza/HixcqRI4cTKgQAAABcEzNS2disWbMUFBSkpk2bOrTHx8erQYMG8vT01NKlS+Xt7e2kCgEAAADXxIxUNpWSkqJZs2apU6dODotI3A5R169f19y5cxUfH6/4+HhJf6z25+7u7qySAQAAAJdBkMqm1q5dqxMnTqhr164O7Xv27NGOHTskSUWLFnXYdvToUYWHhz+oEgEAAACXRZDKpho0aCDLslK1R0VFpdkOAAAA4P9wjxQAAAAAGCJIAQAAAIAhghQAAAAAGCJIAQAAAIAhghQAAAAAGGLVPhdkszm7Audi0UAAAAC4OmakAAAAAMAQQQoAAAAADBGkAAAAAMAQQQrIZsLDw2Wz2VK9evfuLUlKSEhQ7969lT9/fuXKlUutWrVSbGysk6sGAABwLQQpIJvZuXOnzp49a3+tWbNGktS6dWtJ0sCBA7Vs2TJ99dVX2rRpk86cOaOWLVs6s2QAAACXw6p9QDYTGBjo8H7ChAmKjIxU7dq1FRcXp08//VTR0dGqU6eOJGnWrFkqWbKktm/frieffNIZJQMAALgcZqSAbCwxMVFz585V165dZbPZtHv3biUlJalevXr2PiVKlFCRIkW0bds2J1YKAADgWghSQDa2ZMkSXb58WZ07d5YkxcTEyNPTU3ny5HHoFxwcrJiYmAdfIAAAgIsiSAHZ2KeffqrGjRsrNDTU2aUAAABkKdwjBWRTx48f19q1a7Vo0SJ7W0hIiBITE3X58mWHWanY2FiFhIQ4oUoAAADXxIwUkE3NmjVLQUFBatq0qb2tYsWKypEjh9atW2dvO3jwoE6cOKGqVas6o0wAAACXxIwUkA2lpKRo1qxZ6tSpkzw8/u+vAX9/f3Xr1k2DBg1Svnz55Ofnp759+6pq1aqs2AcAAPAnBCkgG1q7dq1OnDihrl27ptr23nvvyc3NTa1atdLNmzfVsGFDffTRR06oEgAAwHURpIBsqEGDBrIsK81t3t7emjZtmqZNm/aAqwIAAMg6uEcKAAAAAAwRpAAAAADAEEEKAAAAAAwRpAAAAADAEEEKAAAAAAyxah/gYmyjbM4uwamskWmvJggAAOBKmJECAAAAAEMEKWRLp0+f1vPPP6/8+fPLx8dHZcuW1a5du+zbr169qj59+qhQoULy8fFRqVKlNGPGDCdWDAAAAFfCpX3Idi5duqTq1avrqaee0ooVKxQYGKjDhw8rb9689j6DBg3S+vXrNXfuXIWHh2v16tXq1auXQkND1bx5cydWDwAAAFdAkEK2M3HiRBUuXFizZs2yt0VERDj0+e6779SpUydFRUVJknr06KF//etf+v777wlSAAAA4NI+ZD9Lly5VpUqV1Lp1awUFBal8+fL65JNPHPpUq1ZNS5cu1enTp2VZljZs2KBDhw6pQYMGTqoaAAAAroQghWznt99+0/Tp01WsWDGtWrVKL7/8svr166c5c+bY+0ydOlWlSpVSoUKF5OnpqUaNGmnatGmqVauWEysHAACAq+DSPmQ7KSkpqlSpksaNGydJKl++vPbv368ZM2aoU6dOkv4IUtu3b9fSpUsVFhamzZs3q3fv3goNDVW9evWcWT4AAABcAEEK2U6BAgVUqlQph7aSJUvq66+/liTduHFD//znP7V48WI1bdpUklSuXDnt3btXkydPJkgBAACAS/uQ/VSvXl0HDx50aDt06JDCwsIkSUlJSUpKSpKbm+Pp4e7urpSUlAdWJwAAAFwXM1LIdgYOHKhq1app3LhxatOmjb7//nt9/PHH+vjjjyVJfn5+ql27tgYPHiwfHx+FhYVp06ZN+uyzz/Tuu+86uXoAAAC4AoIUsp0nnnhCixcv1rBhwzR69GhFRERoypQp6tChg73PggULNGzYMHXo0EEXL15UWFiYxo4dq549ezqxcgAAALgKghSypaefflpPP/30HbeHhIQ4PGcKAAAA+DPukQIAAAAAQwQpAAAAADBEkAIAAAAAQwQpAAAAADBEkAIAAAAAQwQpAAAAADDE8udwPdE2Z1cAAAAA3BUzUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgiCAFAAAAAIYIUgAAAABgyOWD1OnTp/X8888rf/788vHxUdmyZbVr1y77dsuy9MYbb6hAgQLy8fFRvXr1dPjwYSdWDAAAAOBh59JB6tKlS6pevbpy5MihFStW6MCBA3rnnXeUN29ee59Jkybpgw8+0IwZM7Rjxw7lzJlTDRs2VEJCghMrBwAAAPAw83B2AXczceJEFS5cWLNmzbK3RURE2H+2LEtTpkzR8OHD9cwzz0iSPvvsMwUHB2vJkiVq27btA68ZAAAAwMPPpWekli5dqkqVKql169YKCgpS+fLl9cknn9i3Hz16VDExMapXr569zd/fX1WqVNG2bdvuOO7NmzcVHx/v8AIAAACA9HLpIPXbb79p+vTpKlasmFatWqWXX35Z/fr105w5cyRJMTExkqTg4GCH/YKDg+3b0jJ+/Hj5+/vbX4ULF868LwEAAADgoePSQSolJUUVKlTQuHHjVL58efXo0UPdu3fXjBkz/ta4w4YNU1xcnP118uTJDKoYAAAAQHbg0kGqQIECKlWqlENbyZIldeLECUlSSEiIJCk2NtahT2xsrH1bWry8vOTn5+fwAgAAAID0cukgVb16dR08eNCh7dChQwoLC5P0x8ITISEhWrdunX17fHy8duzYoapVqz7QWgEAAABkHy69at/AgQNVrVo1jRs3Tm3atNH333+vjz/+WB9//LEkyWazacCAAXrrrbdUrFgxRUREaMSIEQoNDVWLFi2cWzwAAACAh5ZLB6knnnhCixcv1rBhwzR69GhFRERoypQp6tChg73Pa6+9pmvXrqlHjx66fPmyatSooZUrV8rb29uJlQMAAAB4mNksy7KcXYSzxcfHy9/fX3FxcS5xv5TN5uwKnMual70PgO2wsytwLmtk9v4rifM/ex8Azn/O/+yM89/ZFTiXK53/6c0GLn2PFAAAAAC4IoIUAAAAABgiSAEAAACAIYIUAAAAABgiSAEAAACAIYIUAAAAABgiSAEAAACAIYIUAAAAABgiSAEAAACAIYIUAAAAABgiSAEAAACAIYIUAAAAABgiSAEAAACAIYIUAAAAABgiSAEAAACAIYIUAAAAABgiSAEAAACAIYIUAAAAABgiSAEAAACAIYIUAAAAABgiSAEAAACAIYIUAAAAABgiSAEAAACAIYIUAAAAABgiSAEAAACAIYIUAAAAABj6W0Hq5s2bGVUHAAAAAGQZRkFqxYoV6tSpkx555BHlyJFDvr6+8vPzU+3atTV27FidOXMms+oEAAAAAJeRriC1ePFiPfroo+ratas8PDw0ZMgQLVq0SKtWrdK///1v1a5dW2vXrtUjjzyinj176vz585ldNwAAAAA4jUd6Ok2aNEnvvfeeGjduLDe31NmrTZs2kqTTp09r6tSpmjt3rgYOHJixlQIAAACAi0hXkNq2bVu6BitYsKAmTJjwtwoCAAAAAFf3t1ftu3btmuLj4zOiFgAAAADIEu47SB04cECVKlVS7ty5lTdvXpUtW1a7du3KyNoAAAAAwCXdd5B66aWX1KdPH129elW///67WrZsqU6dOmVkbQAAAADgktIdpJ555hmdPn3a/v78+fNq3ry5fH19lSdPHjVp0kSxsbGZUiQAAAAAuJJ0LTYhSc8//7zq1Kmj3r17q2/fvurTp49Kly6t2rVrKykpSevXr9crr7ySmbUCAAAAgEtI94xU69at9f333+vAgQN68sknVb16da1evVrVq1dXzZo1tXr1ag0fPjwzawUAAAAAl5DuGSlJ8vf314wZM/Ttt9+qU6dOql+/vsaMGSNfX9/Mqg8AAAAAXI7RYhMXL17U7t27VbZsWe3evVt+fn4qX768vvnmm8yqDwAAAABcTrqDVHR0tAoVKqSmTZsqLCxMK1as0MiRI/Wf//xHkyZNUps2bVhsAgAAAEC2kO4gNWzYMM2cOVMxMTFat26dRowYIUkqUaKENm7cqPr166tq1aqZVigAAAAAuIp0B6mrV6+qePHikqTIyEhdv37dYXv37t21ffv2jK0OAAAAAFxQuheb6NSpk5o2baqoqCjt2rVLHTt2TNUnKCgoQ4sDAAAAAFeU7iD17rvv6qmnntIvv/yizp07q0GDBplZFwAAAAC4LKPlz5s1a6ZmzZplVi0AAAAAkCWk6x6pBQsWpHvAkydPauvWrfddEAAAAAC4unQFqenTp6tkyZKaNGmSfv7551Tb4+Li9M0336h9+/aqUKGCfv/99wwvFAAAAABcRbou7du0aZOWLl2qqVOnatiwYcqZM6eCg4Pl7e2tS5cuKSYmRgEBAercubP279+v4ODgzK4bAAAAAJwm3fdINW/eXM2bN9eFCxf07bff6vjx47px44YCAgJUvnx5lS9fXm5u6V5NHQAAAACyLKPFJiQpICBALVq0yIRSAAAAACBrYAoJAAAAAAwRpAAAAADAEEEKAAAAAAwRpAAAAADAkHGQ2rBhQ2bUAQAAAABZhnGQatSokSIjI/XWW2/p5MmTmVETAAAAALg04yB1+vRp9enTRwsXLtQjjzyihg0b6ssvv1RiYmJm1AcAAAAALsc4SAUEBGjgwIHau3evduzYoUcffVS9evVSaGio+vXrp3379mVGnQAAAADgMv7WYhMVKlTQsGHD1KdPH129elUzZ85UxYoVVbNmTf30008ZVSMAAAAAuJT7ClJJSUlauHChmjRporCwMK1atUoffvihYmNj9euvvyosLEytW7fO6FoBAAAAwCV4mO7Qt29fzZ8/X5ZlqWPHjpo0aZLKlClj354zZ05NnjxZoaGhGVooAAAAALgK4yB14MABTZ06VS1btpSXl1eafQICAlgmHQAAAMBDyzhIrVu37t6Deniodu3a91UQAAAAALg643ukxo8fr5kzZ6ZqnzlzpiZOnJghRQEAAACAKzMOUv/6179UokSJVO2lS5fWjBkzMqQoAAAAAHBlxkEqJiZGBQoUSNUeGBios2fPZkhRAAAAAODKjINU4cKFtXXr1lTtW7duZaU+AAAAANmC8WIT3bt314ABA5SUlKQ6depI+mMBitdee02vvPJKhhcIAAAAAK7GOEgNHjxYv//+u3r16qXExERJkre3t4YMGaJhw4ZleIEAAAAA4GqMg5TNZtPEiRM1YsQI/fzzz/Lx8VGxYsXu+EwpAAAAAHjYGAep23LlyqUnnngiI2sBAAAAgCzhvoLUrl279OWXX+rEiRP2y/tuW7RoUYYUBgAAAACuynjVvgULFqhatWr6+eeftXjxYiUlJemnn37S+vXr5e/vnxk1AgAAAIBLMQ5S48aN03vvvadly5bJ09NT77//vn755Re1adNGRYoUyYwaAQAAAMClGAepI0eOqGnTppIkT09PXbt2TTabTQMHDtTHH3+c4QUCAAAAgKsxDlJ58+bVlStXJEkFCxbU/v37JUmXL1/W9evXM7Y6AAAAAHBBxotN1KpVS2vWrFHZsmXVunVr9e/fX+vXr9eaNWtUt27dzKgRAAAAAFyKcZD68MMPlZCQIEl6/fXXlSNHDn333Xdq1aqVhg8fnuEFAgAAAICrMQpSt27d0vLly9WwYUNJkpubm4YOHZophQEAAACAqzK6R8rDw0M9e/a0z0gBAAAAQHZkvNhE5cqVtXfv3kwoBQAAAACyBuN7pHr16qVBgwbp5MmTqlixonLmzOmwvVy5chlWHAAAAAC4IuMg1bZtW0lSv3797G02m02WZclmsyk5OTnjqgMAAAAAF2QcpI4ePZoZdQAAAABAlmEcpMLCwjKjDgAAAADIMoyD1GeffXbX7S+88MJ9FwMAAAAAWYFxkOrfv7/D+6SkJF2/fl2enp7y9fUlSAEAAAB46Bkvf37p0iWH19WrV3Xw4EHVqFFD8+fPz4waAQAAAMClGAeptBQrVkwTJkxINVsFAAAAAA+jDAlSkuTh4aEzZ85k1HAAAAAA4LKM75FaunSpw3vLsnT27Fl9+OGHql69eoYVBgAAAACuyjhItWjRwuG9zWZTYGCg6tSpo3feeSej6gIAAAAAl2UcpFJSUjKjDgAAAADIMjLsHikAAAAAyC6Mg1SrVq00ceLEVO2TJk1S69atM6QoAAAAAHBlxkFq8+bNatKkSar2xo0ba/PmzRlSFAAAAAC4MuMgdfXqVXl6eqZqz5Ejh+Lj4zOkKAAAAABwZcZBqmzZsvriiy9StS9YsEClSpXKkKIAAAAAwJUZr9o3YsQItWzZUkeOHFGdOnUkSevWrdP8+fP11VdfZXiBAAAAAOBqjINUs2bNtGTJEo0bN04LFy6Uj4+PypUrp7Vr16p27dqZUSMAAAAAuBTjICVJTZs2VdOmTTO6FgAAAADIEozvkdq5c6d27NiRqn3Hjh3atWtXhhQFAAAAAK7MOEj17t1bJ0+eTNV++vRp9e7dO0OKAgAAAABXZhykDhw4oAoVKqRqL1++vA4cOJAhRQEAAACAKzMOUl5eXoqNjU3VfvbsWXl43NctVwAAAACQpRgHqQYNGmjYsGGKi4uzt12+fFn//Oc/Vb9+/QwtDgAAAABckfEU0uTJk1WrVi2FhYWpfPnykqS9e/cqODhYn3/+eYYXCAAAAACuxjhIFSxYUD/++KPmzZunffv2ycfHR126dFG7du2UI0eOzKgRAAAAAFzKfd3UlDNnTvXo0SOjawEAAACALOG+V4c4cOCATpw4ocTERIf25s2b/+2iAAAAAMCVGQep3377Tc8++6z+97//yWazybIsSZLNZpMkJScnZ2yFAAAAAOBijFft69+/vyIiInTu3Dn5+vrqp59+0ubNm1WpUiVt3LgxE0oEAAAAANdiPCO1bds2rV+/XgEBAXJzc5Obm5tq1Kih8ePHq1+/fvrhhx8yo04AAAAAcBnGM1LJycnKnTu3JCkgIEBnzpyRJIWFhengwYMZWx0AAAAAuCDjGakyZcpo3759ioiIUJUqVTRp0iR5enrq448/1iOPPJIZNQIAAACASzEOUsOHD9e1a9ckSaNHj9bTTz+tmjVrKn/+/Priiy8yvEAAAAAAcDXGQaphw4b2n4sWLapffvlFFy9eVN68ee0r9wEAAADAw+y+nyP1Z/ny5cuIYQAAAAAgSzBebAIAAAAAsjuCFAAAAAAYIkgBAAAAgCGCFAAAAAAYIkgBAAAAgCGCFAAAAAAYIkgBAAAAgCGCFAAAAAAYIkgBAAAAgCGCFAAAAAAYIkgBAAAAgCGCFAAAAAAYIkgBAAAAgCGCFAAAAAAYIkgBAAAAgCGCFAAAAAAYIkgBAAAAgCGCFAAAAAAYIkgBAAAAgCGCFAAAAAAYIkgBAAAAgCGCFAAAAAAYIkgBAAAAgCGCFAAAAAAYIkgBAAAAgKEsFaQmTJggm82mAQMG2NsSEhLUu3dv5c+fX7ly5VKrVq0UGxvrvCIBAAAAPPSyTJDauXOn/vWvf6lcuXIO7QMHDtSyZcv01VdfadOmTTpz5oxatmzppCoBAAAAZAdZIkhdvXpVHTp00CeffKK8efPa2+Pi4vTpp5/q3XffVZ06dVSxYkXNmjVL3333nbZv3+7EigEAAAA8zLJEkOrdu7eaNm2qevXqObTv3r1bSUlJDu0lSpRQkSJFtG3btjuOd/PmTcXHxzu8AAAAACC9PJxdwL0sWLBAe/bs0c6dO1Nti4mJkaenp/LkyePQHhwcrJiYmDuOOX78eI0aNSqjSwUAAACQTbj0jNTJkyfVv39/zZs3T97e3hk27rBhwxQXF2d/nTx5MsPGBgAAAPDwc+kgtXv3bp07d04VKlSQh4eHPDw8tGnTJn3wwQfy8PBQcHCwEhMTdfnyZYf9YmNjFRIScsdxvby85Ofn5/ACAAAAgPRy6Uv76tatq//9738ObV26dFGJEiU0ZMgQFS5cWDly5NC6devUqlUrSdLBgwd14sQJVa1a1RklAwAAAMgGXDpI5c6dW2XKlHFoy5kzp/Lnz29v79atmwYNGqR8+fLJz89Pffv2VdWqVfXkk086o2QAAAAA2YBLB6n0eO+99+Tm5qZWrVrp5s2batiwoT766CNnlwUAAADgIZblgtTGjRsd3nt7e2vatGmaNm2acwoCAAAAkO249GITAAAAAOCKCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGCFIAAAAAYIggBQAAAACGXDpIjR8/Xk888YRy586toKAgtWjRQgcPHnTok5CQoN69eyt//vzKlSuXWrVqpdjYWCdVDAAAACA7cOkgtWnTJvXu3Vvbt2/XmjVrlJSUpAYNGujatWv2PgMHDtSyZcv01VdfadOmTTpz5oxatmzpxKoBAAAAPOw8nF3A3axcudLh/ezZsxUUFKTdu3erVq1aiouL06effqro6GjVqVNHkjRr1iyVLFlS27dv15NPPumMsgEAAAA85Fx6Ruqv4uLiJEn58uWTJO3evVtJSUmqV6+evU+JEiVUpEgRbdu27Y7j3Lx5U/Hx8Q4vAAAAAEivLBOkUlJSNGDAAFWvXl1lypSRJMXExMjT01N58uRx6BscHKyYmJg7jjV+/Hj5+/vbX4ULF87M0gEAAAA8ZLJMkOrdu7f279+vBQsW/O2xhg0bpri4OPvr5MmTGVAhAAAAgOzCpe+Ruq1Pnz5avny5Nm/erEKFCtnbQ0JClJiYqMuXLzvMSsXGxiokJOSO43l5ecnLyyszSwYAAADwEHPpGSnLstSnTx8tXrxY69evV0REhMP2ihUrKkeOHFq3bp297eDBgzpx4oSqVq36oMsFAAAAkE249IxU7969FR0drf/85z/KnTu3/b4nf39/+fj4yN/fX926ddOgQYOUL18++fn5qW/fvqpatSor9gEAAADINC4dpKZPny5JioqKcmifNWuWOnfuLEl677335ObmplatWunmzZtq2LChPvroowdcKQAAAIDsxKWDlGVZ9+zj7e2tadOmadq0aQ+gIgAAAABw8XukAAAAAMAVEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwJCHswsAAAAAXF2yzUdJHoGSzZYp44flzJRhs4yEhIQH/pk5cuSQu7v7fe9PkAIAAADu4qp3GZ0q8LosD/9MC1IzCmbKsFnG0aNHH/hn2mw2FSpUSLly5bqv/QlSAAAAwB0k23x0qsDr8s0XoUA/d2VOjJKuJWXSwFlERFDEA/08y7J0/vx5nTp1SsWKFbuvmSmCFAAAAHAHSR6Bsjz8FejnLh/PTPwgKxPHzgK8vb0f+GcGBgbq2LFjSkpKuq8gxWITAAAAwJ3YbJLNlmkzUXAe29+8TJMgBQAAAACGCFIAAAAAYIggBQAAAACGWGwCAAAAMGALqJRpY+88vStd/WoVq2X/+cb1G/L28bbf8/Plxi8VUjAk3Z/55oA3VTi8sLoN6ObQvnHlRq3+z2r1+WcfPfPkM/Lx9ZEkBQQF6IXeL6hF+xb3HHvZF8u0YtEKffTFR+muJy2dO3dW0aJFNXz48L81TkYiSAEAAABZzObDm+0/V3+kur7Y8IVCC4dm6Gds27BNVaOqSpLc3d3tn7l1/Va90vkVPV75cYUXDc/Qz8xKuLQPAAAAeEjEXYrT8N7D1aBcAz1T9Rkt/3K5fdvSBUvVrHIz1X60tlrVbKXd3+3W8i+Xa+Xilfr0/U9Vq1gtjR8y3t5/++bt9iD1Z9XrVJdfHj8dPnBYkpScnKx/Tf6XmlVupoaPNdR7o97TrVu3dOr4KU0YNkF7tu1RrWK11OapNvescebMmQoLC1Pu3LlVvHhxbdy4UXPmzNG8efM0ZswY5cqVSz179sysw2eEGSkAAADgITGy30iFFwvX8p3LdebkGb3c5mUVL1NchcIL6d2R7+qzFZ+pyCNFdPbUWaWkpKhitYra9d2uVJf2Hfv1mHLmyqmA4ACdOXnG3m5Zljav2azLFy+rUHghSVL0x9Hau2OvPlvxmTw8PDS422At+nyR2nRpo6Hjh6a6tC+tGlvWaanIyEgNGDBAu3fvVrFixXT8+HGlpKQoKipKGzZscLlL+5iRAgAAAB4CF85d0O5tu9VnWB95enkqvGi4GrZoqA0rNvzRwSb9dug3Jd5MVIFCBVSwSME7jvXdhu9U9an/m41KTk7WUyWfUo3IGnq1y6vq888+KlmupCTpP/P/o56v9VTe/HmV2z+3OrzUQev+u86oxkWLFv1Ros2mn376STdv3lRYWJgiIiIy6OhkPGakAAAAgIdA7OlYJd5MVP1y9e1tKckpavRsI/n4+mjstLGaO2OuRg0cpSo1q+iV0a8oMCQwzbG2bdymzr0729+7u7trw88bdCvplqa/PV07t+7UC71ekCTFnI5R/4797YtdWJaloJAgoxo7Pt9ROXPm1Pz58zV58mR17txZ9evX1/vvv6/Q0Iy99yujEKQAAACAh0BgSKB8cvpo/YH19lDzZ9XrVlf1utV1/dp1jR86XtMmTNObU95M1TfhRoIO7T+kx554LNUYHjk81GtIL7WJaqONKzcqqlGUgkKCNG7GOJUoWyJV/7+OfacaK4X+sRJikyZN1KRJE129elU9e/bUP//5T82ePTvN7+NsXNoHAAAAPASCCgSpbMWymj5xuhJuJOjWrVv65X+/6LdDv+n3879ry5otSriRIE9PT3n7eMvd3V2SlC8gn86eOmsfZ8/2PSpbsaw8cqQ95+Lu7q723dtr1tRZkqRmbZtp+qTpuhB7QZZl6czJM9q9bbckKW9AXp07e063bt26a40HDhxQbGysli9frhs3bsjLy0u+vr72GoOCgnTs2LHMOnT3hRkpAAAAwIB1IX3PejKx62bGjDNm6hhNGTVFz1R9RreSbimyeKQGvjlQVoqlz6d/rhF9RsjN3U3lKpbT8Ml/LNzQ7B/NNPSloXqq5FNq2KKhcuTIkeZqfX/WtHVTzXh7hr7f8r06vtxRybeS1a1FN8VdjFNIoRD7ZX9PVH9CBQoXUINyDRQcGqz5a+enWeMn0z5RSkqKJk2apPbt28vd3V3VqlXTv//9b0lS165d9dxzzylPnjxq3769Pvro7z2XKiPYLMuynF2Es8XHx8vf319xcXHy8/NzdjlywZnLB8qal70PgO2wsytwLmtk9v4rifM/ex8Azn/O/+zMVc//BM8wHQ2foYiCAfL2zLzPyagglRFa126t9+e+n+HPpbqb25f2PUgJCQk6evSoIiIi5O3tbW9Pbzbg0j4AAAAAkqTEm4lq3q75Aw1RWRVBCgAAAIAkydPLUx17dnR2GVkCQQoAAAAADBGkAAAAAMAQQQoAAAAADBGkAAAAAMAQQQoAAAAADBGkAAAAAMCQh7MLAAAAALKUVU9k+JC3H0e7K2pnho+NzEGQAgAAALKg5lWa6+KFi3Jzc1PO3DlVt2ldDXxzoNzd3e+4z5mTZ9SyekttP7H9b332si+WacWiFfroi48c2uMuxal9vfaKj4uXzWaTJN24fkPePt72919u/FIhBUPS/VmdO3dW0aJFNXz48L9Vc0YjSAEAAABZ1IfzP9TjlR/XqWOn1KNVDz1S/BG1fL6l0+rZsXmHnqj5hN6c8qa9rfoj1fXFhi8UWjjUaXVlBu6RAgAAALK4QuGF9Filx3Top0OSpHXL16nNU21Ut3RdDeo8SBcvXJQk9W3XV8nJyapVrJZqFaulmNMxSk5O1r8m/0vNKjdTw8ca6r1R7+nWrVuSpP/t/p86NOigqOJRalKxiaI/jtap46c0YdgE7dm2R7WK1VKbp9rY6/huw3eq9lS1O9YZdylOw3sPV4NyDfRM1We0/Mvl9m0zZ85UWFiYcufOreLFi2vjxo2aM2eO5s2bpzFjxihXrlzq2bNnZhy++0KQAgAAALK440eO64fvf1ChsEL66Yef9O7IdzXuo3FatXeVwouGa+KwiZKkqfOnyt3dXZsPb9bmw5sVUjBE0R9Ha++OvfpsxWdauHmhDv7voBZ9vkiS9O7Id/V8z+e18eBGLVi3QBWrVVShsEIaOn6oKlStoM2HN+vLDV/a69i5Zaeq1KpyxzpH9hupgOAALd+5XO9//r6mTZimwwcO69q1axowYIDWrl2rK1euaPXq1QoLC1OnTp3UoUMHjRgxQlevXtWMGTMy90AaIEgBAAAAWVT/5/urVrFaeq7WcypXqZxad26tpQuW6rlOz6loyaLyyOGhbgO6adPqTfZZpr/6z/z/qOdrPZU3f17l9s+tDi910Lr/rpMkeeTw0KljpxR3KU5+efxUvEzxO9ZycP9BBRUIkn9e/zS3Xzh3Qbu37VafYX3k6eWp8KLhatiioTas2CBJstls+umnn3Tz5k2FhYUpIiLibx6dzMU9UgAAAEAW9f7c9/XYE49p8+rNmjxism5cv6GYMzH65utv9Nn0z+z9PDw89Pu539McI+Z0jPp37G9fDMKyLAWFBEmSXn/7df3r7X+pZY2WiigWoX7D+6lcpXJpjrNt4zZVfarqHWuNPR2rxJuJql+uvr0tJTlFjZ5tpJw5c2r+/PmaPHmyOnfurPr16+v9999XaKjr3ldFkAIAAACyMJvNptoNa2vz6s36dMqnCgwJVM/BPdXhpQ6p+p49dTZVW1BIkMbNGKcSZUuk2hZeNFzj/zVet27d0qLPF2l47+FaumOpPXT92baN29RnWJ871hkYEiifnD5af2B9mvs3adJETZo00dWrV9WzZ0/985//1OzZs9Ps6wq4tA8AAAB4CDzf83n9Z/5/1KRVE305+0v7whNxl+K0adUmSVKefHmUkpKi2DOx9v2atW2m6ZOm60LsBVmWpTMnz2j3tt2SpBWLVujyxcvy8PCQb05fubn/ER/yBuTVubPn7JcLXrt6TcePHFfp8qXvWF9QgSCVrVhW0ydOV8KNBN26dUu//O8X/XboN8XGxmr58uW6ceOGvLy85Ovra1/GPSgoSMeOHcvw4/V3MSMFAAAAmGiY8Q/N3XXz748RUSxCFZ6soB2bd6j/iP56s/+bOnPyjPzy+Kles3qq3bC2fHx91Kl3J3Vo0EHJycmav3a+Or7cUcm3ktWtRTfFXYxTSKEQvdDrBUnS1vVb9c4b7ygpMUlFHimiUe+PkiQ9Uf0JFShcQA3KNVBwaLBeevUlVXiygtzc7j5PM2bqGE0ZNUXPVH1Gt5JuKbJ4pAa+OVApKSmaNGmS2rdvL3d3d1WrVk3//ve/JUldu3bVc889pzx58qh9+/b66KOP7voZD4rNsizL2UU4W3x8vPz9/RUXFyc/Pz9nlyMXnb18YKx52fsA2A47uwLnskZm77+SOP+z9wHg/Of8z85c9fxP8AzT0fAZiigYIG/PzPucjAhSzjR+yHiVqVBGzf7R7L72rxRaKYMrureEhAQdPXpUERER8vb2trenNxtwaR8AAACAv6V4meKqUa+Gs8t4oLi0DwAAAMDf0rJjS2eX8MAxIwUAAAAAhghSAAAAAGCIIAUAAAAAhghSAAAAAGCIIAUAAAAAhghSAAAAANK07Itl6vWPXnfc3uapNvpx148Z/rnh4eH69ttvM3zcjMTy5wAAAIAB24dPZNrYO7vvNOq/eN5ifTHzC50+flr+ef1VqVol9Xi1h0ILh2ZShY6+3PDlA/mcP5s9e7ZefPFFh4fojhs3Tv369dPChQs1efJk7d27V23bttXs2bMzrQ6CFAAAAJAFfTrlU3015yuNmDxClapXUkpKilYsWqFdW3epedvmzi4vU0VFRWnt2rWp2vPly6dXX31V3333nS5evJipNXBpHwAAAJDFXIm7ollTZ2no+KGqXre6vLy95OPro5bPt1Tzts0VczpG/Tv2V51SddQmqo02rdpk3/el517SjLdnqGOjjqpVrJbGvDJGF2Iv6OXWLyuqeJRe6/6aEm8m2vtbKZbeGvyWoopHqX399jq4/6B9W/MqzbX3+72SpDcHvKnJIybr5TYvq/ajtdWnXR/FXYqz9929bbc6Nuqop0o+pZeee0mnjp2yb1uxYoWKFi2qfPnyadSoUfd9XOrUqaPnnntOQUFB9z1GehGkAAAAgCzmx90/6lbSLdWoVyPN7cN7D9cjjz6ilT+s1JBxQ/RG3zd08uhJ+/b136zX25++rUVbF+nbdd9qUJdBevWtV/Xf3f/Vid9OaOWSlfa+P+z4QSXKlNDa/Wv1TLtn9Fr313Tr1q00P3fNsjUa9OYgrf5xtZKTk7Xg0wWSpJjTMRraY6gGjRqktfvXqk6TOvpnr39Kki79fklt2rTRlClTFBMTo2vXrunUqVNpju9KCFIAAABAFhN3KU7++fzl4ZH6Tp2Y0zH6+cef9dKrL8nTy1MVq1VUjXo1tHb5/10K90zbZxRSMEQBQQGq8GQFlatYTpHFI5UzV05Vr1Ndh386bO8bHBqs5zo9J48cHmrTpY1uJd3STz/8lGZd9Z6up2KlisnL20t1mtTR4QN/jLNy8UpFNY5S+Srl5e7urn90/YfOnjyrMyfPaOu6rapYsaKefvppeXp66s0335Sb291jyqZNm5QnTx7769y5c/dzGP8W7pECAAAAshj/vP6KuxinW7dupQpTF2IvKE++PPL2+b/FGEIKhehC7AX7+7wBee0/e3l7KW9+x/d/viQvqMD/XSZns9kUXCBYv5/7Pc268gXks//s7eOt69euS/oj3K34eoXWLvu/MJeUmKTzMed14dwFFS5c2N7u6+ur/PnzS5JOnDihUqVKSZLCwsL0009/BLjatWuneY/Ug0SQAgAAALKYchXLySOHh7au26raDWs7bAsIDtDli5d1M+GmvLy9JEmxp2MVWSLyvj7r3FnH2Z7Ys7HKH5TfaIygkCC1aN9Cr455NdW2k0dPav93++3vb9y4od9//yOoFSlSRFevXr2PqjMfl/YBAAAAWUxu/9zq0reLJv5zor7b8J0SbyYq4UaCli5Yqu+3fK/iZYrr43c+VlJikvZs36Mta7aobtO69/VZsWditWjuIt1KuqUvZ38pDw8PlS5f2miMhs821Nrla/XDjh+UkpKia1evad3ydZKk6nWra/fu3frmm2+UmJioUaNGKSUl5b5qTU5OVkJCgm7duuXwc2ZgRgoAAAAwYPUxe9ZTeuy6ab5PtwHdlDcgrz546wOdOnZK/nn99UT1J/TS4JdUuWZljX1trBo81kABQQEaOWWkijxS5L5qK1+lvA7sO6D3R7+vgkUKasLHE9K8N+tuChYpqLEfjdUHYz7QsSPH5OPro0rVKqnu03WVN39ezZ8/X3379tXFixc1YMAAFSpU6L5q/fzzz9WlSxf7+7lz52rkyJF6880372u8u7FZlmVl+KhZTHx8vPz9/RUXFyc/Pz9nlyObzdkVOJc1L3sfANvhe/d5mFkjs/dfSZz/2fsAcP5z/mdnrnr+J3iG6Wj4DEUUDJC3Z+Z9zv0EqYdJpdBKD/wzExISdPToUUVERDg83De92YBL+wAAAADAEEEKAAAAAAwRpAAAAADAEEEKAAAAAAwRpAAAAIA7sSzJspS9l0J5OP3dNfdY/hwAAAC4gxy3zst2K07n4/Mp0M9dmba2YOY86ijLSEhIeKCfZ1mWzp8/L5vNphw5ctzXGAQpAAAA4A7crRsqdHasTul1HYv3z7R16i8kZcqwWcbRa0cf+GfabDYVKlRI7u7u97U/QQoAAAC4i1wJ+1XsWFcleQRmWpBqfCxThs0yfunzywP/zBw5ctx3iJIeoiA1bdo0vf3224qJidFjjz2mqVOnqnLlys4uCwAAAA8Bd+uG3JNOZNr4x69l2tBZwp8fiJtVPBSLTXzxxRcaNGiQRo4cqT179uixxx5Tw4YNde7cOWeXBgAAAOAh9FAEqXfffVfdu3dXly5dVKpUKc2YMUO+vr6aOXOms0sDAAAA8BDK8pf2JSYmavfu3Ro2bJi9zc3NTfXq1dO2bdvS3OfmzZu6efOm/X1cXJwkKT4+PnOLRbrEX3d2BU72YBetcTmch9kb57+zC3Auzv/sjfPf2QU4lyud/7drudfy6Fk+SF24cEHJyckKDg52aA8ODtYvv6R909r48eM1atSoVO2FCxfOlBphxr+7syuAM/lP8Hd2CXAizv/sjfM/e+P8z95c8fy/cuWK/P3vXFeWD1L3Y9iwYRo0aJD9fUpKii5evKj8+fPLlkkrsSBriI+PV+HChXXy5En5+fk5uxwADxDnP5B9cf7jzyzL0pUrVxQaGnrXflk+SAUEBMjd3V2xsbEO7bGxsQoJCUlzHy8vL3l5eTm05cmTJ7NKRBbk5+fHX6RANsX5D2RfnP+47W4zUbdl+cUmPD09VbFiRa1bt87elpKSonXr1qlq1apOrAwAAADAwyrLz0hJ0qBBg9SpUydVqlRJlStX1pQpU3Tt2jV16dLF2aUBAAAAeAg9FEHqH//4h86fP6833nhDMTExevzxx7Vy5cpUC1AA9+Ll5aWRI0emuvQTwMOP8x/Ivjj/cT9s1r3W9QMAAAAAOMjy90gBAAAAwINGkAIAAAAAQwQpAAAAADBEkAIAAAAAQwQpIA0JCQnq3bu38ufPr1y5cqlVq1apHvoMIGu5ePGi+vbtq+LFi8vHx0dFihRRv379FBcX59DPZrOlei1YsMChz82bN/X6668rLCxMXl5eCg8P18yZMx/k1wFgKCoqKtW53bNnT4c+/fr1U8WKFeXl5aXHH3881RgJCQnq3LmzypYtKw8PD7Vo0eLBFA+X9FAsfw78HYmJifL09HRoGzhwoP773//qq6++kr+/v/r06aOWLVtq69atTqoSgKm/nttnzpzRmTNnNHnyZJUqVUrHjx9Xz549debMGS1cuNBh31mzZqlRo0b293ny5HHY3qZNG8XGxurTTz9V0aJFdfbsWaWkpGTq9wGQfmn9v12SunfvrtGjR9vf+/r6purTtWtX7dixQz/++GOqbcnJyfLx8VG/fv309ddfZ2zRyHIIUsh2oqKiVKZMGXl4eGju3LkqW7asNmzYYN8eFxenTz/9VNHR0apTp46kP36pKlmypLZv364nn3zSWaUDuIt7ndtlypRx+MUnMjJSY8eO1fPPP69bt27Jw+P//peYJ08ehYSEpPk5K1eu1KZNm/Tbb78pX758kqTw8PDM+VIA0uVe5/9tvr6+dzy3JemDDz6QJJ0/fz7NIJUzZ05Nnz5dkrR161Zdvnw5Y74AsiQu7UO2NGfOHHl6emrr1q2aMWOGw7bdu3crKSlJ9erVs7eVKFFCRYoU0bZt2x50qQAM3O3cTktcXJz8/PwcQpQk9e7dWwEBAapcubJmzpypPz9ycenSpapUqZImTZqkggUL6tFHH9Wrr76qGzduZPj3AZB+6Tn/582bp4CAAJUpU0bDhg3T9evXH3CVeJgwI4VsqVixYpo0aVKa22JiYuTp6ZnqUp7g4GDFxMQ8gOoA3K+7ndt/deHCBY0ZM0Y9evRwaB89erTq1KkjX19frV69Wr169dLVq1fVr18/SdJvv/2mb7/9Vt7e3lq8eLEuXLigXr166ffff9esWbMy/DsBSJ97nf/t27dXWFiYQkND9eOPP2rIkCE6ePCgFi1a9ACrxMOEIIVsqWLFipKkcePGady4cfb2AwcOOKskABngbud2kSJF7O/j4+PVtGlTlSpVSm+++abDGCNGjLD/XL58eV27dk1vv/22PUilpKTIZrNp3rx58vf3lyS9++67eu655/TRRx/Jx8cns74egLu41/n/5380KVu2rAoUKKC6devqyJEjioyMfOD1Iuvj0j5kSzlz5pQk9ezZU3v37rW/QkNDFRISosTExFTXPcfGxt71umoAzne3c/u2K1euqFGjRsqdO7cWL16sHDly3HXMKlWq6NSpU7p586YkqUCBAipYsKA9RElSyZIlZVmWTp06lQnfCkB6pOf8/7MqVapIkn799dcHViMeLsxIIVvLly+f/Wbx2ypWrKgcOXJo3bp1atWqlSTp4MGDOnHihKpWreqMMgEYSuvclv6YiWrYsKG8vLy0dOlSeXt733OsvXv3Km/evPLy8pIkVa9eXV999ZWuXr2qXLlySZIOHTokNzc3FSpUKGO/CABjdzr//2rv3r2S/vjHEeB+EKSAv/D391e3bt00aNAg5cuXT35+furbt6+qVq3Kin1AFhYfH68GDRro+vXrmjt3ruLj4xUfHy9JCgwMlLu7u5YtW6bY2Fg9+eST8vb21po1azRu3Di9+uqr9nHat2+vMWPGqEuXLho1apQuXLigwYMHq2vXrlzWB7ioI0eOKDo6Wk2aNFH+/Pn1448/auDAgapVq5bKlStn7/frr7/q6tWriomJ0Y0bN+xhq1SpUvbl1A8cOKDExERdvHhRV65csfdJ67lTeLgRpIA0vPfee3Jzc1OrVq108+ZNNWzYUB999JGzywLwN+zZs0c7duyQJBUtWtRh29GjRxUeHq4cOXJo2rRpGjhwoCzLUtGiRfXuu++qe/fu9r65cuXSmjVr1LdvX1WqVEn58+dXmzZt9NZbbz3Q7wMg/Tw9PbV27VpNmTJF165dU+HChdWqVSsNHz7cod+LL76oTZs22d+XL19e0v/9HSFJTZo00fHjx1P1+fPqnsgebBZ/6gAAAABghMUmAAAAAMAQQQoAAAAADBGkAAAAAMAQQQoAAAAADBGkAAAAAMAQQQoAAAAADBGkAAAAAMAQQQoAAAAADBGkAAC4h6ioKA0YMMDZZQAAXAhBCgAAAAAMEaQAALiDxMREZ5cAAHBRBCkAQJZx8+ZN9evXT0FBQfL29laNGjW0c+dOpaSkqFChQpo+fbpD/x9++EFubm46fvy4JOny5ct68cUXFRgYKD8/P9WpU0f79u2z93/zzTf1+OOP69///rciIiLk7e2dZh2ff/65KlWqpNy5cyskJETt27fXuXPn7Ns3btwom82m//73vypXrpy8vb315JNPav/+/ZlwVAAAzkCQAgBkGa+99pq+/vprzZkzR3v27FHRokXVsGFDXb58We3atVN0dLRD/3nz5ql69eoKCwuTJLVu3Vrnzp3TihUrtHv3blWoUEF169bVxYsX7fv8+uuv+vrrr7Vo0SLt3bs3zTqSkpI0ZswY7du3T0uWLNGxY8fUuXPnVP0GDx6sd955Rzt37lRgYKCaNWumpKSkDDseAADnsVmWZTm7CAAA7uXatWvKmzevZs+erfbt20v6I9CEh4drwIABql+/vipUqKBjx46pSJEiSklJUZEiRTR8+HD17NlT3377rZo2bapz587Jy8vLPm7RokX12muvqUePHnrzzTc1btw4nT59WoGBgfY+UVFRevzxxzVlypQ0a9u1a5eeeOIJXblyRbly5dLGjRv11FNPacGCBfrHP/4hSbp48aIKFSqk2bNnq02bNpl3oAAADwQzUgCALOHIkSNKSkpS9erV7W05cuRQ5cqV9fPPP+vxxx9XyZIl7bNSmzZt0rlz59S6dWtJ0r59+3T16lXlz59fuXLlsr+OHj2qI0eO2McMCwtzCFFp2b17t5o1a6YiRYood+7cql27tiTpxIkTDv2qVq1q/zlfvnwqXry4fv755793IAAALsHD2QUAAJBROnTooOjoaA0dOlTR0dFq1KiR8ufPL0m6evWqChQooI0bN6baL0+ePPafc+bMedfPuHbtmho2bKiGDRtq3rx5CgwM1IkTJ9SwYUMWpwCAbIQZKQBAlhAZGSlPT09t3brV3paUlKSdO3eqVKlSkqT27dtr//792r17txYuXKgOHTrY+1aoUEExMTHy8PBQ0aJFHV4BAQHpruOXX37R77//rgkTJqhmzZoqUaKEw0ITf7Z9+3b7z5cuXdKhQ4dUsmRJ068OAHBBBCkAQJaQM2dOvfzyyxo8eLBWrlypAwcOqHv37rp+/bq6desmSQoPD1e1atXUrVs3JScnq3nz5vb969Wrp6pVq6pFixZavXq1jh07pu+++06vv/66du3ale46ihQpIk9PT02dOlW//fabli5dqjFjxqTZd/To0Vq3bp3279+vzp07KyAgQC1atPhbxwEA4BoIUgCALGPChAlq1aqVOnbsqAoVKujXX3/VqlWrlDdvXnufDh06aN++fXr22Wfl4+Njb7fZbPrmm29Uq1YtdenSRY8++qjatm2r48ePKzg4ON01BAYGavbs2frqq69UqlQpTZgwQZMnT75jvf3791fFihUVExOjZcuWydPT8/4PAADAZbBqHwAAGez2qn2XLl1yuP8KAPDwYEYKAAAAAAwRpAAAAADAEJf2AQAAAIAhZqQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAMEaQAAAAAwBBBCgAAAAAM/T85YfoM6cIKzAAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 1000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def train_svm(train_set, test_set, window_size, overlap, exp_name):\n",
    "    training_times = {}\n",
    "    \n",
    "    X_train, y_train = prepare_data(train_set)\n",
    "    X_test, y_test = prepare_data(test_set)\n",
    "    \n",
    "    print(window_size, overlap, 'X_train.shape', X_train.shape)\n",
    "    print(window_size, overlap, 'X_test.shape', X_test.shape)\n",
    "    \n",
    "    X_train_flat = flatten_data(X_train)\n",
    "    X_test_flat = flatten_data(X_test)\n",
    "\n",
    "    # Convert one-hot encoded labels to class indices\n",
    "    y_train_idx = np.argmax(y_train, axis=1)\n",
    "    y_test_idx = np.argmax(y_test, axis=1)\n",
    "    \n",
    "    print(\"y_train_idx\", y_train_idx)\n",
    "    print(\"y_test_idx\", y_test_idx)\n",
    "\n",
    "    X_train_flat, y_train_idx = shuffle_data(X_train_flat, y_train_idx)\n",
    "\n",
    "    # Linear Kernel\n",
    "    svm_linear = SVC(kernel=\"linear\", gamma=\"scale\")\n",
    "\n",
    "    # Polynomial Kernel\n",
    "    svm_poly = SVC(\n",
    "        kernel=\"poly\", degree=3, gamma=\"scale\"\n",
    "    )  # degree is a hyperparameter for polynomial kernel\n",
    "\n",
    "    # RBF Kernel\n",
    "    svm_rbf = SVC(kernel=\"rbf\", gamma=\"scale\")\n",
    "\n",
    "    start_time = time.time()\n",
    "    svm_linear.fit(X_train_flat, y_train_idx)\n",
    "    linear_training_time = time.time() - start_time\n",
    "    print(f\"Linear SVM Training Time (W:{window_size}-O:{overlap}): {linear_training_time} seconds\")\n",
    "    \n",
    "    start_time = time.time()\n",
    "    svm_poly.fit(X_train_flat, y_train_idx)\n",
    "    poly_training_time = time.time() - start_time\n",
    "    print(f\"Polynomial SVM Training Time (W:{window_size}-O:{overlap}): {poly_training_time} seconds\")\n",
    "    \n",
    "    start_time = time.time()\n",
    "    svm_rbf.fit(X_train_flat, y_train_idx)\n",
    "    rbf_training_time = time.time() - start_time\n",
    "    print(f\"RBF SVM Training Time (W:{window_size}-O:{overlap}): {rbf_training_time} seconds\")\n",
    "    \n",
    "    training_times = {\n",
    "        'linear': linear_training_time,\n",
    "        'poly': poly_training_time,\n",
    "        'rbf': rbf_training_time\n",
    "    }\n",
    "\n",
    "    # Making predictions and evaluating accuracy for each model\n",
    "    y_pred_linear = svm_linear.predict(X_test_flat)\n",
    "    y_pred_poly = svm_poly.predict(X_test_flat)\n",
    "    y_pred_rbf = svm_rbf.predict(X_test_flat)\n",
    "\n",
    "    accuracy_linear = accuracy_score(y_test_idx, y_pred_linear) * 100\n",
    "    accuracy_poly = accuracy_score(y_test_idx, y_pred_poly) * 100\n",
    "    accuracy_rbf = accuracy_score(y_test_idx, y_pred_rbf) * 100\n",
    "    \n",
    "    # For the linear SVM model\n",
    "    precision_linear = precision_score(y_test_idx, y_pred_linear, average='weighted') * 100\n",
    "    recall_linear = recall_score(y_test_idx, y_pred_linear, average='weighted') * 100\n",
    "    f1_linear = f1_score(y_test_idx, y_pred_linear, average='weighted') * 100\n",
    "\n",
    "    # For the polynomial SVM model\n",
    "    precision_poly = precision_score(y_test_idx, y_pred_poly, average='weighted') * 100\n",
    "    recall_poly = recall_score(y_test_idx, y_pred_poly, average='weighted') * 100\n",
    "    f1_poly = f1_score(y_test_idx, y_pred_poly, average='weighted') * 100\n",
    "\n",
    "    # For the RBF SVM model\n",
    "    precision_rbf = precision_score(y_test_idx, y_pred_rbf, average='weighted') * 100\n",
    "    recall_rbf = recall_score(y_test_idx, y_pred_rbf, average='weighted') * 100\n",
    "    f1_rbf = f1_score(y_test_idx, y_pred_rbf, average='weighted') * 100\n",
    "    \n",
    "    print(\"Linear SVM Metrics:\")\n",
    "    print(f\"Precision: {precision_linear}, Recall: {recall_linear}, F1 Score: {f1_linear}, Accuracy: {accuracy_linear}\")\n",
    "\n",
    "    print(\"\\nPolynomial SVM Metrics:\")\n",
    "    print(f\"Precision: {precision_poly}, Recall: {recall_poly}, F1 Score: {f1_poly}, Accuracy: {accuracy_poly}\")\n",
    "\n",
    "    print(\"\\nRBF SVM Metrics:\")\n",
    "    print(f\"Precision: {precision_rbf}, Recall: {recall_rbf}, F1 Score: {f1_rbf}, Accuracy: {accuracy_rbf}\")\n",
    "    \n",
    "    cm_linear = confusion_matrix(y_test_idx, y_pred_linear)\n",
    "    cm_poly = confusion_matrix(y_test_idx, y_pred_poly)\n",
    "    cm_rbf = confusion_matrix(y_test_idx, y_pred_rbf)\n",
    "    \n",
    "    # for cm, acc, type in zip(\n",
    "    #     [cm_linear, cm_poly, cm_rbf], \n",
    "    #     [accuracy_linear, accuracy_poly, accuracy_rbf],\n",
    "    #     ['linear', 'poly', 'rbf']\n",
    "    # ):\n",
    "    #     sns.heatmap(cm, annot=True, fmt='d')\n",
    "    #     plt.title(f'{type} - {window_size}_{overlap} - {exp_name} - {acc:.2f}%')\n",
    "    #     plt.xlabel('Predicted')\n",
    "    #     plt.ylabel('True')\n",
    "    #     plt.savefig(f'{PATH}/plots/svm_{type}_{window_size}_{overlap}_{exp_name}.png')\n",
    "    #     plt.show()\n",
    "\n",
    "    print(\"-\" * 50)\n",
    "    print(f\"Linear Kernel Accuracy: {accuracy_linear}\")\n",
    "    print(f\"Polynomial Kernel Accuracy: {accuracy_poly}\")\n",
    "    print(f\"RBF Kernel Accuracy: {accuracy_rbf}\")\n",
    "    print(\"-\" * 50)\n",
    "\n",
    "    # save results in dict\n",
    "    metrics = {\n",
    "        'accuracy': {\n",
    "            'linear': accuracy_linear,\n",
    "            'poly': accuracy_poly,\n",
    "            'rbf': accuracy_rbf           \n",
    "        },\n",
    "        'training_time': training_times,\n",
    "        'precision': {\n",
    "            'linear': precision_linear,\n",
    "            'poly': precision_poly,\n",
    "            'rbf': precision_rbf\n",
    "        },\n",
    "        'recall': {\n",
    "            'linear': recall_linear,\n",
    "            'poly': recall_poly,\n",
    "            'rbf': recall_rbf\n",
    "        },\n",
    "        'f1': {\n",
    "            'linear': f1_linear,\n",
    "            'poly': f1_poly,\n",
    "            'rbf': f1_rbf\n",
    "        },\n",
    "    }\n",
    "\n",
    "    return metrics\n",
    "\n",
    "results = {}\n",
    "\n",
    "for window_size, overlap_list in spectogram_map.items():\n",
    "    linear_tr_metrics = []\n",
    "    linear_rt_metrics = []\n",
    "    linear_names = []\n",
    "    \n",
    "    poly_tr_metrics = []\n",
    "    poly_rt_metrics = []\n",
    "    poly_names = []\n",
    "    \n",
    "    rbf_tr_metrics = []\n",
    "    rbf_rt_metrics = []\n",
    "    rbf_names = []\n",
    "    \n",
    "    metric = 'accuracy'\n",
    "\n",
    "    for overlap in overlap_list:\n",
    "        # Load the dataset\n",
    "        test_dataset = np.load(\n",
    "            f\"{NPY_DATA_DIR}/test_{window_size}_{overlap}.npy\", allow_pickle=True\n",
    "        )\n",
    "        retest_dataset = np.load(\n",
    "            f\"{NPY_DATA_DIR}/retest_{window_size}_{overlap}.npy\", allow_pickle=True\n",
    "        )\n",
    "\n",
    "        test_retest_metrics = train_svm(test_dataset, retest_dataset, window_size, overlap, 'test_retest')\n",
    "        retest_test_metrics = train_svm(retest_dataset, test_dataset, window_size, overlap, 'retest_test')\n",
    "        \n",
    "        results[f\"{window_size}_{overlap}\"] = {\n",
    "            \"test_retest\": test_retest_metrics,\n",
    "            \"retest_test\": retest_test_metrics,\n",
    "        }\n",
    "\n",
    "        linear_tr_metrics.append(test_retest_metrics[metric]['linear'])\n",
    "        linear_rt_metrics.append(retest_test_metrics[metric]['linear'])\n",
    "        linear_names.append(f'l-{overlap}')\n",
    "\n",
    "        poly_tr_metrics.append(test_retest_metrics[metric]['poly'])\n",
    "        poly_rt_metrics.append(retest_test_metrics[metric]['poly'])\n",
    "        poly_names.append(f'p-{overlap}')\n",
    "        \n",
    "        rbf_tr_metrics.append(test_retest_metrics[metric]['rbf'])\n",
    "        rbf_rt_metrics.append(retest_test_metrics[metric]['rbf'])\n",
    "        rbf_names.append(f'r-{overlap}')\n",
    "        \n",
    "    combined_linear_f1 = []\n",
    "    for tr_f1, rt_f1 in zip(linear_tr_metrics, linear_rt_metrics):\n",
    "        combined_linear_f1.append(f1_metric(tr_f1, rt_f1))\n",
    "        \n",
    "    combined_poly_f1 = []\n",
    "    for tr_f1, rt_f1 in zip(poly_tr_metrics, poly_rt_metrics):\n",
    "        combined_poly_f1.append(f1_metric(tr_f1, rt_f1))\n",
    "        \n",
    "    combined_rbf_f1 = []\n",
    "    for tr_f1, rt_f1 in zip(rbf_tr_metrics, rbf_rt_metrics):\n",
    "        combined_rbf_f1.append(f1_metric(tr_f1, rt_f1))\n",
    "\n",
    "    for tr_metrics, rt_metrics, combined, names, svm_type in zip(\n",
    "        [linear_tr_metrics, poly_tr_metrics, rbf_tr_metrics],\n",
    "        [linear_rt_metrics, poly_rt_metrics, rbf_rt_metrics],\n",
    "        [combined_linear_f1, combined_poly_f1, combined_rbf_f1],\n",
    "        [linear_names, poly_names, rbf_names],\n",
    "        ['Linear', 'Poly', 'RBF']\n",
    "    ):\n",
    "        n = len(tr_metrics)  # Number of data pairs\n",
    "        ind = np.arange(n)  # The x locations for the groups\n",
    "        width = 0.25        # Width of the bars\n",
    "\n",
    "        fig, ax = plt.subplots(figsize=(10, 10))\n",
    "        x_bars = ax.bar(ind - width, tr_metrics, width, label='Test/Retest', color='blue')\n",
    "        y_bars = ax.bar(ind, rt_metrics, width, label='Retest/Test', color='orange')\n",
    "        z_bars = ax.bar(ind + width, combined, width, label='Combined-F1', color='green')\n",
    "            \n",
    "        for bar in x_bars + y_bars + z_bars:\n",
    "            height = bar.get_height()\n",
    "            ax.annotate(f'{height:.0f}',  # Adjust format as needed\n",
    "                        xy=(bar.get_x() + bar.get_width() / 2, height),\n",
    "                        xytext=(0, 3),  # Offset\n",
    "                        textcoords='offset points',\n",
    "                        ha='center', va='bottom')\n",
    "\n",
    "        ax.set_xlabel('overlap')\n",
    "        ax.set_ylabel(metric + ' (%)')\n",
    "        ax.set_title(f'{svm_type} SVM Accuracy - window_size={window_size}')\n",
    "        ax.set_xticks(ind)\n",
    "        ax.set_xticklabels(names)\n",
    "        ax.set_ylim([0, 100])\n",
    "        ax.legend(fontsize='small', loc='lower right')\n",
    "\n",
    "        plt.savefig(f'{PATH}/plots/svm_{svm_type}_{window_size}_{metric}.png')\n",
    "        plt.show()\n",
    "        \n",
    "# save results as pretty printed json\n",
    "with open(f\"{PATH}/svm_results.json\", \"w\") as f:\n",
    "    json.dump(results, f, indent=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "84.91423529411765\n"
     ]
    }
   ],
   "source": [
    "print(f1_metric(87.7,82.3))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "venv",
   "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": "3.7.16"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}