{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "f3c92fc0",
   "metadata": {},
   "outputs": [],
   "source": [
    "from library.exercise import Exercise\n",
    "from library.dataset import DataSet, TrainTestData\n",
    "from library.GanExamples import StupidToyListGan\n",
    "from library.SimpleGan import SimpleGan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "d10f9c73",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "import numpy as np\n",
    "import random\n",
    "from imblearn.datasets import fetch_datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "bfdaaf87",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'folding': [(array([[ 7.4  ,  0.7  ,  0.   , ...,  3.51 ,  0.56 ,  9.4  ],\n",
      "       [ 7.8  ,  0.88 ,  0.   , ...,  3.2  ,  0.68 ,  9.8  ],\n",
      "       [ 7.8  ,  0.76 ,  0.04 , ...,  3.26 ,  0.65 ,  9.8  ],\n",
      "       ...,\n",
      "       [ 6.3  ,  0.51 ,  0.13 , ...,  3.42 ,  0.75 , 11.   ],\n",
      "       [ 5.9  ,  0.645,  0.12 , ...,  3.57 ,  0.71 , 10.2  ],\n",
      "       [ 6.   ,  0.31 ,  0.47 , ...,  3.39 ,  0.66 , 11.   ]]), array([0., 0., 0., ..., 0., 0., 0.]), array([[11.2 ,  0.28,  0.56, ...,  3.16,  0.58,  9.8 ],\n",
      "       [ 8.9 ,  0.62,  0.18, ...,  3.16,  0.88,  9.2 ],\n",
      "       [ 8.1 ,  0.56,  0.28, ...,  3.11,  1.28,  9.3 ],\n",
      "       ...,\n",
      "       [ 5.8 ,  0.61,  0.11, ...,  3.55,  0.66, 10.9 ],\n",
      "       [ 6.3 ,  0.55,  0.15, ...,  3.32,  0.82, 11.6 ],\n",
      "       [ 6.8 ,  0.62,  0.08, ...,  3.42,  0.82,  9.5 ]]), array([0., 0., 0., 0., 0., 1., 1., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 1., 0., 0., 0., 0.,\n",
      "       1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1.,\n",
      "       0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.])), (array([[ 7.8 ,  0.88,  0.  , ...,  3.2 ,  0.68,  9.8 ],\n",
      "       [11.2 ,  0.28,  0.56, ...,  3.16,  0.58,  9.8 ],\n",
      "       [ 7.4 ,  0.7 ,  0.  , ...,  3.51,  0.56,  9.4 ],\n",
      "       ...,\n",
      "       [ 6.2 ,  0.6 ,  0.08, ...,  3.45,  0.58, 10.5 ],\n",
      "       [ 5.9 ,  0.55,  0.1 , ...,  3.52,  0.76, 11.2 ],\n",
      "       [ 6.3 ,  0.51,  0.13, ...,  3.42,  0.75, 11.  ]]), array([0., 0., 0., ..., 0., 0., 0.]), array([[ 7.4  ,  0.7  ,  0.   , ...,  3.51 ,  0.56 ,  9.4  ],\n",
      "       [ 7.8  ,  0.76 ,  0.04 , ...,  3.26 ,  0.65 ,  9.8  ],\n",
      "       [ 7.9  ,  0.6  ,  0.06 , ...,  3.3  ,  0.46 ,  9.4  ],\n",
      "       ...,\n",
      "       [ 7.2  ,  0.39 ,  0.44 , ...,  3.3  ,  0.84 , 11.5  ],\n",
      "       [ 5.9  ,  0.645,  0.12 , ...,  3.57 ,  0.71 , 10.2  ],\n",
      "       [ 6.   ,  0.31 ,  0.47 , ...,  3.39 ,  0.66 , 11.   ]]), array([0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1.,\n",
      "       0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 1., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.])), (array([[ 7.4  ,  0.7  ,  0.   , ...,  3.51 ,  0.56 ,  9.4  ],\n",
      "       [ 7.8  ,  0.76 ,  0.04 , ...,  3.26 ,  0.65 ,  9.8  ],\n",
      "       [11.2  ,  0.28 ,  0.56 , ...,  3.16 ,  0.58 ,  9.8  ],\n",
      "       ...,\n",
      "       [ 5.9  ,  0.55 ,  0.1  , ...,  3.52 ,  0.76 , 11.2  ],\n",
      "       [ 5.9  ,  0.645,  0.12 , ...,  3.57 ,  0.71 , 10.2  ],\n",
      "       [ 6.   ,  0.31 ,  0.47 , ...,  3.39 ,  0.66 , 11.   ]]), array([0., 0., 0., ..., 0., 0., 0.]), array([[ 7.8  ,  0.88 ,  0.   , ...,  3.2  ,  0.68 ,  9.8  ],\n",
      "       [ 7.4  ,  0.7  ,  0.   , ...,  3.51 ,  0.56 ,  9.4  ],\n",
      "       [ 6.7  ,  0.58 ,  0.08 , ...,  3.28 ,  0.54 ,  9.2  ],\n",
      "       ...,\n",
      "       [ 7.5  ,  0.31 ,  0.41 , ...,  3.34 ,  0.85 , 11.4  ],\n",
      "       [ 6.6  ,  0.725,  0.2  , ...,  3.29 ,  0.54 ,  9.2  ],\n",
      "       [ 6.3  ,  0.51 ,  0.13 , ...,  3.42 ,  0.75 , 11.   ]]), array([0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0.,\n",
      "       0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0.,\n",
      "       0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 1., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.])), (array([[ 7.4  ,  0.7  ,  0.   , ...,  3.51 ,  0.56 ,  9.4  ],\n",
      "       [ 7.8  ,  0.88 ,  0.   , ...,  3.2  ,  0.68 ,  9.8  ],\n",
      "       [ 7.8  ,  0.76 ,  0.04 , ...,  3.26 ,  0.65 ,  9.8  ],\n",
      "       ...,\n",
      "       [ 6.3  ,  0.51 ,  0.13 , ...,  3.42 ,  0.75 , 11.   ],\n",
      "       [ 5.9  ,  0.645,  0.12 , ...,  3.57 ,  0.71 , 10.2  ],\n",
      "       [ 6.   ,  0.31 ,  0.47 , ...,  3.39 ,  0.66 , 11.   ]]), array([0., 0., 0., ..., 0., 0., 0.]), array([[ 7.4  ,  0.66 ,  0.   , ...,  3.51 ,  0.56 ,  9.4  ],\n",
      "       [ 5.6  ,  0.615,  0.   , ...,  3.58 ,  0.52 ,  9.9  ],\n",
      "       [ 6.3  ,  0.39 ,  0.16 , ...,  3.34 ,  0.56 ,  9.3  ],\n",
      "       ...,\n",
      "       [ 6.2  ,  0.7  ,  0.15 , ...,  3.54 ,  0.6  , 11.9  ],\n",
      "       [ 5.4  ,  0.74 ,  0.09 , ...,  3.67 ,  0.56 , 11.6  ],\n",
      "       [ 5.9  ,  0.55 ,  0.1  , ...,  3.52 ,  0.76 , 11.2  ]]), array([0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 1., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0.])), (array([[ 7.4  ,  0.7  ,  0.   , ...,  3.51 ,  0.56 ,  9.4  ],\n",
      "       [ 7.8  ,  0.88 ,  0.   , ...,  3.2  ,  0.68 ,  9.8  ],\n",
      "       [ 7.8  ,  0.76 ,  0.04 , ...,  3.26 ,  0.65 ,  9.8  ],\n",
      "       ...,\n",
      "       [ 6.3  ,  0.51 ,  0.13 , ...,  3.42 ,  0.75 , 11.   ],\n",
      "       [ 5.9  ,  0.645,  0.12 , ...,  3.57 ,  0.71 , 10.2  ],\n",
      "       [ 6.   ,  0.31 ,  0.47 , ...,  3.39 ,  0.66 , 11.   ]]), array([0., 0., 0., ..., 0., 0., 0.]), array([[ 7.8 ,  0.58,  0.02, ...,  3.36,  0.57,  9.5 ],\n",
      "       [ 7.5 ,  0.5 ,  0.36, ...,  3.35,  0.8 , 10.5 ],\n",
      "       [ 8.9 ,  0.22,  0.48, ...,  3.39,  0.53,  9.4 ],\n",
      "       ...,\n",
      "       [ 7.2 ,  0.66,  0.33, ...,  3.27,  0.78, 12.8 ],\n",
      "       [ 6.3 ,  0.51,  0.13, ...,  3.42,  0.75, 11.  ],\n",
      "       [ 6.2 ,  0.6 ,  0.08, ...,  3.45,  0.58, 10.5 ]]), array([0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0.,\n",
      "       1., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.])), (array([[ 7.4  ,  0.7  ,  0.   , ...,  3.51 ,  0.56 ,  9.4  ],\n",
      "       [ 7.8  ,  0.88 ,  0.   , ...,  3.2  ,  0.68 ,  9.8  ],\n",
      "       [ 7.8  ,  0.76 ,  0.04 , ...,  3.26 ,  0.65 ,  9.8  ],\n",
      "       ...,\n",
      "       [ 6.3  ,  0.51 ,  0.13 , ...,  3.42 ,  0.75 , 11.   ],\n",
      "       [ 5.9  ,  0.645,  0.12 , ...,  3.57 ,  0.71 , 10.2  ],\n",
      "       [ 6.   ,  0.31 ,  0.47 , ...,  3.39 ,  0.66 , 11.   ]]), array([0., 0., 0., ..., 0., 0., 0.]), array([[ 8.9 ,  0.62,  0.19, ...,  3.17,  0.93,  9.2 ],\n",
      "       [ 8.5 ,  0.28,  0.56, ...,  3.3 ,  0.75, 10.5 ],\n",
      "       [ 7.9 ,  0.32,  0.51, ...,  3.04,  1.08,  9.2 ],\n",
      "       ...,\n",
      "       [ 6.7 ,  0.32,  0.44, ...,  3.29,  0.8 , 11.6 ],\n",
      "       [ 6.3 ,  0.51,  0.13, ...,  3.42,  0.75, 11.  ],\n",
      "       [ 6.8 ,  0.62,  0.08, ...,  3.42,  0.82,  9.5 ]]), array([0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 1., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.])), (array([[ 7.4  ,  0.7  ,  0.   , ...,  3.51 ,  0.56 ,  9.4  ],\n",
      "       [ 7.8  ,  0.76 ,  0.04 , ...,  3.26 ,  0.65 ,  9.8  ],\n",
      "       [11.2  ,  0.28 ,  0.56 , ...,  3.16 ,  0.58 ,  9.8  ],\n",
      "       ...,\n",
      "       [ 6.3  ,  0.51 ,  0.13 , ...,  3.42 ,  0.75 , 11.   ],\n",
      "       [ 5.9  ,  0.645,  0.12 , ...,  3.57 ,  0.71 , 10.2  ],\n",
      "       [ 6.   ,  0.31 ,  0.47 , ...,  3.39 ,  0.66 , 11.   ]]), array([0., 0., 0., ..., 0., 0., 0.]), array([[ 7.8 ,  0.88,  0.  , ...,  3.2 ,  0.68,  9.8 ],\n",
      "       [ 7.4 ,  0.7 ,  0.  , ...,  3.51,  0.56,  9.4 ],\n",
      "       [ 7.6 ,  0.39,  0.31, ...,  3.52,  0.65,  9.7 ],\n",
      "       ...,\n",
      "       [ 6.2 ,  0.56,  0.09, ...,  3.54,  0.6 , 11.3 ],\n",
      "       [ 5.8 ,  0.61,  0.11, ...,  3.55,  0.66, 10.9 ],\n",
      "       [ 6.2 ,  0.6 ,  0.08, ...,  3.45,  0.58, 10.5 ]]), array([0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 1., 0., 0.,\n",
      "       0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 1., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.])), (array([[ 7.8  ,  0.88 ,  0.   , ...,  3.2  ,  0.68 ,  9.8  ],\n",
      "       [ 7.8  ,  0.76 ,  0.04 , ...,  3.26 ,  0.65 ,  9.8  ],\n",
      "       [11.2  ,  0.28 ,  0.56 , ...,  3.16 ,  0.58 ,  9.8  ],\n",
      "       ...,\n",
      "       [ 6.3  ,  0.51 ,  0.13 , ...,  3.42 ,  0.75 , 11.   ],\n",
      "       [ 5.9  ,  0.645,  0.12 , ...,  3.57 ,  0.71 , 10.2  ],\n",
      "       [ 6.   ,  0.31 ,  0.47 , ...,  3.39 ,  0.66 , 11.   ]]), array([0., 0., 0., ..., 0., 0., 0.]), array([[ 7.4 ,  0.7 ,  0.  , ...,  3.51,  0.56,  9.4 ],\n",
      "       [ 7.9 ,  0.6 ,  0.06, ...,  3.3 ,  0.46,  9.4 ],\n",
      "       [ 7.8 ,  0.58,  0.02, ...,  3.36,  0.57,  9.5 ],\n",
      "       ...,\n",
      "       [ 6.2 ,  0.46,  0.29, ...,  3.33,  0.62,  9.8 ],\n",
      "       [ 7.2 ,  0.39,  0.44, ...,  3.3 ,  0.84, 11.5 ],\n",
      "       [ 5.4 ,  0.74,  0.09, ...,  3.67,  0.56, 11.6 ]]), array([0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 1., 0., 0., 0., 0.,\n",
      "       1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       1., 1., 1., 0., 0., 0., 0., 0., 0., 0., 1., 0., 1., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.])), (array([[ 7.4  ,  0.7  ,  0.   , ...,  3.51 ,  0.56 ,  9.4  ],\n",
      "       [ 7.8  ,  0.88 ,  0.   , ...,  3.2  ,  0.68 ,  9.8  ],\n",
      "       [ 7.8  ,  0.76 ,  0.04 , ...,  3.26 ,  0.65 ,  9.8  ],\n",
      "       ...,\n",
      "       [ 5.9  ,  0.55 ,  0.1  , ...,  3.52 ,  0.76 , 11.2  ],\n",
      "       [ 5.9  ,  0.645,  0.12 , ...,  3.57 ,  0.71 , 10.2  ],\n",
      "       [ 6.   ,  0.31 ,  0.47 , ...,  3.39 ,  0.66 , 11.   ]]), array([0., 0., 0., ..., 0., 0., 0.]), array([[11.2  ,  0.28 ,  0.56 , ...,  3.16 ,  0.58 ,  9.8  ],\n",
      "       [ 7.5  ,  0.5  ,  0.36 , ...,  3.35 ,  0.8  , 10.5  ],\n",
      "       [ 8.1  ,  0.56 ,  0.28 , ...,  3.11 ,  1.28 ,  9.3  ],\n",
      "       ...,\n",
      "       [ 6.1  ,  0.715,  0.1  , ...,  3.57 ,  0.5  , 11.9  ],\n",
      "       [ 7.5  ,  0.31 ,  0.41 , ...,  3.34 ,  0.85 , 11.4  ],\n",
      "       [ 6.3  ,  0.51 ,  0.13 , ...,  3.42 ,  0.75 , 11.   ]]), array([0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.])), (array([[ 7.4 ,  0.7 ,  0.  , ...,  3.51,  0.56,  9.4 ],\n",
      "       [ 7.8 ,  0.88,  0.  , ...,  3.2 ,  0.68,  9.8 ],\n",
      "       [11.2 ,  0.28,  0.56, ...,  3.16,  0.58,  9.8 ],\n",
      "       ...,\n",
      "       [ 6.8 ,  0.62,  0.08, ...,  3.42,  0.82,  9.5 ],\n",
      "       [ 6.2 ,  0.6 ,  0.08, ...,  3.45,  0.58, 10.5 ],\n",
      "       [ 6.3 ,  0.51,  0.13, ...,  3.42,  0.75, 11.  ]]), array([0., 0., 0., ..., 0., 0., 0.]), array([[ 7.8  ,  0.76 ,  0.04 , ...,  3.26 ,  0.65 ,  9.8  ],\n",
      "       [ 7.4  ,  0.66 ,  0.   , ...,  3.51 ,  0.56 ,  9.4  ],\n",
      "       [ 7.3  ,  0.65 ,  0.   , ...,  3.39 ,  0.47 , 10.   ],\n",
      "       ...,\n",
      "       [ 5.9  ,  0.55 ,  0.1  , ...,  3.52 ,  0.76 , 11.2  ],\n",
      "       [ 5.9  ,  0.645,  0.12 , ...,  3.57 ,  0.71 , 10.2  ],\n",
      "       [ 6.   ,  0.31 ,  0.47 , ...,  3.39 ,  0.66 , 11.   ]]), array([0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       1., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0.,\n",
      "       0., 0., 1., 1., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.])), (array([[ 7.4  ,  0.7  ,  0.   , ...,  3.51 ,  0.56 ,  9.4  ],\n",
      "       [ 7.8  ,  0.88 ,  0.   , ...,  3.2  ,  0.68 ,  9.8  ],\n",
      "       [ 7.8  ,  0.76 ,  0.04 , ...,  3.26 ,  0.65 ,  9.8  ],\n",
      "       ...,\n",
      "       [ 5.9  ,  0.55 ,  0.1  , ...,  3.52 ,  0.76 , 11.2  ],\n",
      "       [ 6.3  ,  0.51 ,  0.13 , ...,  3.42 ,  0.75 , 11.   ],\n",
      "       [ 5.9  ,  0.645,  0.12 , ...,  3.57 ,  0.71 , 10.2  ]]), array([0., 0., 0., ..., 0., 0., 0.]), array([[11.2 ,  0.28,  0.56, ...,  3.16,  0.58,  9.8 ],\n",
      "       [ 7.5 ,  0.5 ,  0.36, ...,  3.35,  0.8 , 10.5 ],\n",
      "       [ 6.7 ,  0.58,  0.08, ...,  3.28,  0.54,  9.2 ],\n",
      "       ...,\n",
      "       [ 6.3 ,  0.55,  0.15, ...,  3.32,  0.82, 11.6 ],\n",
      "       [ 6.8 ,  0.62,  0.08, ...,  3.42,  0.82,  9.5 ],\n",
      "       [ 6.  ,  0.31,  0.47, ...,  3.39,  0.66, 11.  ]]), array([0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0.,\n",
      "       0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.])), (array([[ 7.4 ,  0.7 ,  0.  , ...,  3.51,  0.56,  9.4 ],\n",
      "       [11.2 ,  0.28,  0.56, ...,  3.16,  0.58,  9.8 ],\n",
      "       [ 7.4 ,  0.7 ,  0.  , ...,  3.51,  0.56,  9.4 ],\n",
      "       ...,\n",
      "       [ 5.9 ,  0.55,  0.1 , ...,  3.52,  0.76, 11.2 ],\n",
      "       [ 6.3 ,  0.51,  0.13, ...,  3.42,  0.75, 11.  ],\n",
      "       [ 6.  ,  0.31,  0.47, ...,  3.39,  0.66, 11.  ]]), array([0., 0., 0., ..., 0., 0., 0.]), array([[ 7.8  ,  0.88 ,  0.   , ...,  3.2  ,  0.68 ,  9.8  ],\n",
      "       [ 7.8  ,  0.76 ,  0.04 , ...,  3.26 ,  0.65 ,  9.8  ],\n",
      "       [ 7.3  ,  0.65 ,  0.   , ...,  3.39 ,  0.47 , 10.   ],\n",
      "       ...,\n",
      "       [ 6.6  ,  0.725,  0.2  , ...,  3.29 ,  0.54 ,  9.2  ],\n",
      "       [ 6.3  ,  0.51 ,  0.13 , ...,  3.42 ,  0.75 , 11.   ],\n",
      "       [ 5.9  ,  0.645,  0.12 , ...,  3.57 ,  0.71 , 10.2  ]]), array([0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 1., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 1., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.])), (array([[ 7.4  ,  0.7  ,  0.   , ...,  3.51 ,  0.56 ,  9.4  ],\n",
      "       [ 7.8  ,  0.88 ,  0.   , ...,  3.2  ,  0.68 ,  9.8  ],\n",
      "       [ 7.8  ,  0.76 ,  0.04 , ...,  3.26 ,  0.65 ,  9.8  ],\n",
      "       ...,\n",
      "       [ 6.3  ,  0.51 ,  0.13 , ...,  3.42 ,  0.75 , 11.   ],\n",
      "       [ 5.9  ,  0.645,  0.12 , ...,  3.57 ,  0.71 , 10.2  ],\n",
      "       [ 6.   ,  0.31 ,  0.47 , ...,  3.39 ,  0.66 , 11.   ]]), array([0., 0., 0., ..., 0., 0., 0.]), array([[ 7.4 ,  0.7 ,  0.  , ...,  3.51,  0.56,  9.4 ],\n",
      "       [ 7.9 ,  0.6 ,  0.06, ...,  3.3 ,  0.46,  9.4 ],\n",
      "       [ 7.5 ,  0.5 ,  0.36, ...,  3.35,  0.8 , 10.5 ],\n",
      "       ...,\n",
      "       [ 7.2 ,  0.39,  0.44, ...,  3.3 ,  0.84, 11.5 ],\n",
      "       [ 7.5 ,  0.31,  0.41, ...,  3.34,  0.85, 11.4 ],\n",
      "       [ 5.4 ,  0.74,  0.09, ...,  3.67,  0.56, 11.6 ]]), array([0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 1., 0., 0., 0., 0.,\n",
      "       0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.])), (array([[ 7.8  ,  0.88 ,  0.   , ...,  3.2  ,  0.68 ,  9.8  ],\n",
      "       [ 7.8  ,  0.76 ,  0.04 , ...,  3.26 ,  0.65 ,  9.8  ],\n",
      "       [11.2  ,  0.28 ,  0.56 , ...,  3.16 ,  0.58 ,  9.8  ],\n",
      "       ...,\n",
      "       [ 6.2  ,  0.6  ,  0.08 , ...,  3.45 ,  0.58 , 10.5  ],\n",
      "       [ 5.9  ,  0.645,  0.12 , ...,  3.57 ,  0.71 , 10.2  ],\n",
      "       [ 6.   ,  0.31 ,  0.47 , ...,  3.39 ,  0.66 , 11.   ]]), array([0., 0., 0., ..., 0., 0., 0.]), array([[ 7.4  ,  0.7  ,  0.   , ...,  3.51 ,  0.56 ,  9.4  ],\n",
      "       [ 7.4  ,  0.66 ,  0.   , ...,  3.51 ,  0.56 ,  9.4  ],\n",
      "       [ 5.6  ,  0.615,  0.   , ...,  3.58 ,  0.52 ,  9.9  ],\n",
      "       ...,\n",
      "       [ 7.2  ,  0.66 ,  0.33 , ...,  3.27 ,  0.78 , 12.8  ],\n",
      "       [ 5.9  ,  0.55 ,  0.1  , ...,  3.52 ,  0.76 , 11.2  ],\n",
      "       [ 6.3  ,  0.51 ,  0.13 , ...,  3.42 ,  0.75 , 11.   ]]), array([0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.])), (array([[ 7.4  ,  0.7  ,  0.   , ...,  3.51 ,  0.56 ,  9.4  ],\n",
      "       [ 7.8  ,  0.88 ,  0.   , ...,  3.2  ,  0.68 ,  9.8  ],\n",
      "       [ 7.8  ,  0.76 ,  0.04 , ...,  3.26 ,  0.65 ,  9.8  ],\n",
      "       ...,\n",
      "       [ 6.3  ,  0.51 ,  0.13 , ...,  3.42 ,  0.75 , 11.   ],\n",
      "       [ 5.9  ,  0.645,  0.12 , ...,  3.57 ,  0.71 , 10.2  ],\n",
      "       [ 6.   ,  0.31 ,  0.47 , ...,  3.39 ,  0.66 , 11.   ]]), array([0., 0., 0., ..., 0., 0., 0.]), array([[ 7.8  ,  0.61 ,  0.29 , ...,  3.26 ,  1.56 ,  9.1  ],\n",
      "       [ 7.6  ,  0.41 ,  0.24 , ...,  3.28 ,  0.59 ,  9.5  ],\n",
      "       [ 7.9  ,  0.43 ,  0.21 , ...,  3.17 ,  0.91 ,  9.5  ],\n",
      "       ...,\n",
      "       [ 6.1  ,  0.715,  0.1  , ...,  3.57 ,  0.5  , 11.9  ],\n",
      "       [ 6.7  ,  0.32 ,  0.44 , ...,  3.29 ,  0.8  , 11.6  ],\n",
      "       [ 6.2  ,  0.6  ,  0.08 , ...,  3.45 ,  0.58 , 10.5  ]]), array([0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.,\n",
      "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.]))], 'db_size': 1599, 'db_n_attr': 11, 'imbalanced_ratio': 29.169811320754718}\n"
     ]
    }
   ],
   "source": [
    "datasetName = \"folding_winequality-red-4\"\n",
    "pickle_in = open(f\"{datasetName}.pickle\",\"rb\")\n",
    "example_dict1 = pickle.load(pickle_in)\n",
    "\n",
    "print(example_dict1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "0f6b23e8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['folding', 'db_size', 'db_n_attr', 'imbalanced_ratio'])"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "example_dict1.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "ce509dbc",
   "metadata": {},
   "outputs": [],
   "source": [
    "myData = example_dict1[\"folding\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "0fabea38",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([[ 7.4  ,  0.7  ,  0.   , ...,  3.51 ,  0.56 ,  9.4  ],\n",
       "        [ 7.8  ,  0.88 ,  0.   , ...,  3.2  ,  0.68 ,  9.8  ],\n",
       "        [ 7.8  ,  0.76 ,  0.04 , ...,  3.26 ,  0.65 ,  9.8  ],\n",
       "        ...,\n",
       "        [ 6.3  ,  0.51 ,  0.13 , ...,  3.42 ,  0.75 , 11.   ],\n",
       "        [ 5.9  ,  0.645,  0.12 , ...,  3.57 ,  0.71 , 10.2  ],\n",
       "        [ 6.   ,  0.31 ,  0.47 , ...,  3.39 ,  0.66 , 11.   ]]),\n",
       " array([0., 0., 0., ..., 0., 0., 0.]),\n",
       " array([[11.2 ,  0.28,  0.56, ...,  3.16,  0.58,  9.8 ],\n",
       "        [ 8.9 ,  0.62,  0.18, ...,  3.16,  0.88,  9.2 ],\n",
       "        [ 8.1 ,  0.56,  0.28, ...,  3.11,  1.28,  9.3 ],\n",
       "        ...,\n",
       "        [ 5.8 ,  0.61,  0.11, ...,  3.55,  0.66, 10.9 ],\n",
       "        [ 6.3 ,  0.55,  0.15, ...,  3.32,  0.82, 11.6 ],\n",
       "        [ 6.8 ,  0.62,  0.08, ...,  3.42,  0.82,  9.5 ]]),\n",
       " array([0., 0., 0., 0., 0., 1., 1., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.,\n",
       "        0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 1., 0., 0., 0., 0.,\n",
       "        1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.,\n",
       "        0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "        0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "        0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "        0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "        0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1.,\n",
       "        0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "        0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.,\n",
       "        0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "        0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "        0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "        0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "        0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "        0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "        0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "        0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "        0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.]))"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "myData[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "4ff91174",
   "metadata": {},
   "outputs": [],
   "source": [
    "k = myData[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "9e31bb48",
   "metadata": {},
   "outputs": [],
   "source": [
    "data_con = np.concatenate((k[0],k[2]),axis = 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "727d84dc",
   "metadata": {},
   "outputs": [],
   "source": [
    "target = np.concatenate((k[1],k[3]),axis = 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "44270a01",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0., 0., 0., ..., 0., 0., 0.])"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "target"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "c95f51f4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 7.4 ,  0.7 ,  0.  , ...,  3.51,  0.56,  9.4 ],\n",
       "       [ 7.8 ,  0.88,  0.  , ...,  3.2 ,  0.68,  9.8 ],\n",
       "       [ 7.8 ,  0.76,  0.04, ...,  3.26,  0.65,  9.8 ],\n",
       "       ...,\n",
       "       [ 5.8 ,  0.61,  0.11, ...,  3.55,  0.66, 10.9 ],\n",
       "       [ 6.3 ,  0.55,  0.15, ...,  3.32,  0.82, 11.6 ],\n",
       "       [ 6.8 ,  0.62,  0.08, ...,  3.42,  0.82,  9.5 ]])"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_con"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "08c19220",
   "metadata": {},
   "outputs": [],
   "source": [
    "def loadDataset():\n",
    "    labels = target\n",
    "    features = data_con\n",
    "    label_1 = list(np.where(labels == 1)[0])\n",
    "    label_0 = list(np.where(labels == 0)[0])\n",
    "    features_1 = features[label_1]\n",
    "    features_0 = features[label_0]\n",
    "    \n",
    "    data0 = np.array(range(100)).reshape((25,4))\n",
    "    return DataSet(data0=features_0, data1=features_1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "5890f2ce",
   "metadata": {},
   "outputs": [],
   "source": [
    "data = loadDataset()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "f27cef3a",
   "metadata": {},
   "outputs": [],
   "source": [
    "def getRandGen(initValue, incValue=257, multValue=101, modulus=65537):\n",
    "    value = initValue\n",
    "    while True:\n",
    "        value = ((multValue * value) + incValue) % modulus\n",
    "        yield value\n",
    "            \n",
    "randGen = getRandGen(2021)\n",
    "random.seed(2021)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "0939e1eb",
   "metadata": {},
   "outputs": [],
   "source": [
    "def shuffler(data):\n",
    "    data = list(data)\n",
    "    size = len(data)\n",
    "    shuffled = []\n",
    "    while size > 0:\n",
    "        p = next(randGen) % size\n",
    "        size -= 1\n",
    "        shuffled.append(data[p])\n",
    "        data = data[0:p] + data[(p + 1):]\n",
    "    return np.array(shuffled)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "32cf8258",
   "metadata": {},
   "outputs": [],
   "source": [
    "ganName = \"SimpleGAN\"\n",
    "gan = SimpleGan(numOfFeatures=k[2].shape[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "40220314",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> Shuffling data\n",
      "### Start exercise for synthetic point generator\n",
      "\n",
      "====== Step 1/5 =======\n",
      "-> Shuffling data\n",
      "-> Spliting data to slices\n",
      "\n",
      "------ Step 1/5: Slice 1/5 -------\n",
      "-> Reset the GAN\n",
      "-> Train generator for synthetic samples\n",
      "Epoch 1/3\n",
      "Epoch 2/3\n",
      "Epoch 3/3\n",
      "-> create 1194 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWkAAAFpCAYAAABee9lOAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAABG4UlEQVR4nO3de1xUdf4/8NeZgeEqioqmZSiamhKmEpp5DbE10TZxzcXF/JZZabpRXmpts8taa+um609ra9M1KVNbcVPMVWKttDXvQeQtb2iKSoTAcJth5vz++DCHGW5ymeGcGV7Px8OHembmzOeAvufD+7w/748ky7IMIiLSJJ3aAyAiotoxSBMRaRiDNBGRhjFIExFpGIM0EZGGMUgTEWmYl9oDcKbcXCOsVm1WFAYH+yMvr1jtYbgEr8098dq0IySkVa2PcSbdTLy89GoPwWV4be6J1+YeGKSJiDSMQZqISMMYpImINIxBmohIwxikiYg0zKNK8GoiyzKMxnyUlBhhtVpUG8f16zpYrVannc/Ly4Dg4BDo9R7/LSRq0Tz+f3heXg4kSULbth2h13tBkiRVxuHlpUN5uXOCtCzLKCoqQF5eDtq37+SUcxKRNnl8usNkKkWbNu3g5eWtWoB2NkmSEBAQhPJyk9pDISIX8/ggDciQJM+7TE/5wCGiunle9CIi8iAM0hrz44+n8eST/4fRo4di2rRHcOLED2oPiYhUxCCtIWazGS+++Dyio2Pwn/98iWnTHkNi4jMoKjKqPTSiJjMvXwbT0CiYly9TeyhuhUFaQ44dO4Ly8nJMnhwPLy8vjB79ALp1C0NaWqraQyNqMnnLZqCoSPxO9cYgrSEXLpxDaGg3h2OhoV1x7twZlUZE5DxS3GQgIED8TvXm8XXS7qS4uBi+vj4Ox3x9fVFaWqrSiIicxztxHpA4T+1huB3OpOvJkpYK8+PTYHFh6sHPzw9lZWUOx0pLS+Hn5++y9yQibWOQrifrhiTIWVmwbkhy2Xt07RqGixezHI5lZV1A167dankFEXk6Bul60sUnQAoNhS4+wWXvMWBAJABg06aPUV5eji++2IWzZ89g+PBRLntPItI25qTrSR8dA310jEvfw9vbG8uW/Q1/+cub+OCD99CpUye8+eYyBAcHu/R9iUi7GKQ1JiysB959d43awyAijWC6g4hIwxikiYg0jEGaiEjDGKSJiDSMQZqISMMYpImINIxBmohIwxikiYg0jEGaiEjDGKQ16vjxTMTGunYZOhFpH5eFa4wsy0hJ+QyrV69QeyhEpAGcSWvMmjXvYevWf2HatMfVHgoRaQCDtMb8+tdxWLv2I/TufafaQyEiDWCQrqf9Z3PxYnIm9p/Nden7tG8f4tLzE5FruGr3JgbpetqWno3LN0qwLT1b7aEQkQa5avcmBul6mtCvE25t44cJ/TqpPRQi0iBX7d7E6o56urd7O9zbvZ3awyAijXLV7k2amUl/9913mDRpEgYOHIiYmBh8+umnag+JiEh1mphJW61WzJo1CwsXLsRDDz2EjIwMTJ06FXfddRd69+6t9vBUMWBAJP7zny/VHgYRqUwTM+n8/Hzk5uZClmXIsgxJkuDl5QVvb2+1h0ZEpCpNBOng4GD87ne/wwsvvIC+ffti0qRJSExMRPfu3dUeGhGRqjST7jAYDPjrX/+KMWPG4NixY5gzZw7CwsIwdOjQep+nXbvAaseuX9fBy0sTn0VOH4dOp0NISCunnrOxtDIOV+C1uSdPuTZNBOndu3fj2LFjWLhwIQAgKioKcXFx2LRpU4OCdG6uEVar7HDMarWivNzq1PE2hpeXzunjsFqtyMkpdOo5GyMkpJUmxuEKvDb35G7XVtcHiiammFevXoXJZHI45uXlBS8vTXyGEFEV5uXLYBoaBfPyZWoPxeNpIkjfd999+PHHH7Fp0ybIsozMzExs3rwZ48aNU3toRFQDectmoKhI/E4upYkgfccdd2DVqlXYtGkTIiMj8fzzz+P555/H6NGj1R4aEdVAipsMBASI38mlNJNPGDFiBEaMGKH2MIioHrwT5wGJ89QeRougiZk0ERHVTDMzaRIOHfoW7767Cj/9dAnBwcH47W8T8Otfx6k9LCJSCYO0hly7dhWLFi3EokWvYNiwETh16gSee24OOnXqjEGD7lV7eESkAqY7NOTq1WzExDyAESNGQafT4c47+6J//4H4/vt0tYdGRCrhTFpD+vXrj379+it/LyjIR0bGMTzwwIMqjoqI1MSZdD0dyT2Ipd+/iiO5B5vl/YxGIxYufA59+oRj2DBWvRC1VAzS9fTFlZ24VpKNL67sdPl7XbyYhSefnI62bdvi9deXQqfjt4mopeL//noa3XksOvp1wujOY136Pt99dxQzZ07HsGEj8ac/vQUfHx+Xvh8RaRtz0vU0sF0UBraLcul7XL78ExYsSMTMmU9j0qQpLn0vInIPDNIakpy8GcXFRXjvvdV4773VyvGHH/4NZs2aq+LIiEgtDNIaMmfOc5gz5zm1h0FEGsKcNBGRhjFIExFpGIM0EZGGMUgTEWlYiwjSsizf/EluxhOviYiq8/ggrdd7wWw23fyJbsZiKYdOp1d7GETkYh4fpAMD2+DGjRyYTGUeM/uUZSsKC/Pg5xeo9lCIyMU8vk7azy8AAJCf/zMslnLVxqHT6WC1Wp10NgkGgy8CA1s76XxEpFUeH6QBEahtwVotISGtkJNTqOoYiMj9eHy6g4jInTFIExFpGIM0EZGGMUgTEWkYgzQRkYYxSBMRaRiDNBGRhjFIExFpGIM0EZGGMUgTEWkYgzQRkYYxSBMRaRiDNBGRhjFIExFpGIM0EZGGMUgTEWkYgzQRkYZpJkhfv34ds2bNwsCBAzFkyBCsWLFC7SEREalOM0F61qxZCAkJwf/+9z9s2rQJ//73v7F9+3a1h0VEpCpN7HGYnp6OS5cu4ZNPPoG3tze6dOmCpKQk+Pj4qD00IiJVaWImnZmZiZ49e2LVqlUYNmwYRo8ejdTUVHTo0EHtoRERqUoTM+n8/HwcOXIEUVFRSEtLw7lz5zBjxgyEhIRg/Pjx9T5Pu3aBLhxl04WEtFJ7CC7Da3NPvDbt00SQNhgMCAwMxJw5cwAAvXv3xqRJk5CamtqgIJ2ba4TVKrtqmE0SEtIKOTmFag/DJXht7sWSlgrrhiQEPz0Txsihag/HJVzxfbN93XTxCdBHxzj13HV9oGgi3REWFoaSkhKYTCblmMViUXFERJ7LuiEJclYWjB+sVXsobsX2dbNuSGrW99VEkL7vvvvQtm1bLF26FCaTCadOncK//vUvjBs3Tu2hEXkcXXwCpNBQBM54TO2huBXb100Xn9C879us71YLHx8ffPTRR7h06RKGDRuGGTNmYMaMGXjggQfUHhqRx9FHx8B7zXr4jR2r9lBgXr4MpqFRMC9fpvZQamVJS4X58WkAAO81652e6rgZSZZlbSZxG4E5aXXw2tyTGtdWNa9rGhoFFBUBAQEw7DvotPdx5rWZH58GOSsLUmgovNesd8o5q9J8TpqIWoaqeV0pbjIQECB+1yi10hw2mqjuIKKWQRefoMykAcA7cR6QOE/lUdVNHx3T7CkOewzSRNRs1A547ojpDiIiDWOQJiLSMAZpIiINY5AmItIwBmkiIg1jkCYi0jAGaSIiDWOQJiKnsPW4sKSlqj0Uj8IgTUROoVYrT0/HIE1ETqF2jwtPxWXhROQUXPLtGpxJE2kA87lUGwZpIg1gPpdqwyBNpAHM51JtmJMm0gDmc6k2nEkTkVMwr+4aDNJE5BTMq7sGgzQROQXz6q7BnDQROQXz6q7BmTQRkYYxSBMRaRiDNBGRhjFIExFpGIM0ETWr/Wdz8WJyJvafzVV7KG6BQZqImtW29GxcvlGCbenZag/FLTBIE1GzmtCvE25t44cJ/TqpPRS3wDppIg9hXr4M8pbNkOImwztxntrDqdW93dvh3u7t1B6G2+BMmshDyFs2A0VF4nfyGAzSRB5CipsMBASI38ljMN1B5CG8E+cBGk5zUONwJk3UAlnSUpETN4ltRd0AgzRRC2TdkITyc+fZVtQNMEgTtUC6+AR4hYWxragb0FyQLigowMiRI5GcnKz2UIg8lj46BiFbPmVrUTeguSC9ePFiXLt2Te1hEBFpgqaC9NatW2E0GtGzZ0+1h0JEpAmaCdKXLl3CqlWr8MYbb6g9FCKPw01i3Zcm6qQtFgvmz5+PhQsXIiQkpNHnadcu0Imjcr6QkFZqD8FleG3adWPJEpj+/j7g6wv9p58gZMpE5bGGXlvJzp0wfrAGgTMeh9/Ysc4eqlO5+/fNRhNB+p133kG3bt0wZsyYJp0nN9cIq1V20qicKySkFXJyCtUehkvw2rTNtP4jwGoFSkth7tUHVyY8DF18Am6ZMrHB12Z+933IWVnIe/d9GCOHumjETedu37e6PlA0ke7YsWMHdu3ahcjISERGRuL06dN49dVX8corr6g9NCK3J8VNBgIDISVMBzIzIGdlNbo+mjuCNz9JlmXNTT0feughPProo5g4ceLNn2yHM2l18NrchyUtFdYNSY2eSbsLd/u+1TWTrjPdUVJSUu838fPzq/+IiEgV+ugY1ka7mTqD9L333ouysrJ6nejEiRNOGRAAfPbZZ047FxGRO6szSG/fvh1PPPEE/P398cILLzTXmIiokdyl8T/VX51BukuXLli7di3i4uJw5coV/PrXv26mYRFRYzg0/q8jSFvSUpHz6QZYfhPP9IfG3bS6o3Pnzli0aBG+/fbb5hgPETVBfRv/swue+9BkdUdjsbpDHbw292NJS4X+009g+c1vPXIm7W7fN5fWSefn5+Ojjz5q6mnIze0/m4sXkzOx/2yu2kOhemAXPPfR6BWH+/btw5YtW5CWloby8nL87ne/c+a4yM1sS8/G5Rsl2JaezZ2giZyoQUH6p59+QnJyMrZu3YqrV6/C398fkydPxtSpU101PnITE/p1wrb0bEzo10ntoRB5lJsGaZPJhP/85z/YsmULDh06BJ1Oh6ioKFy7dg0ff/wxevfu3RzjJI27t3s7zqCJXKDOIP3KK69gx44dMJvNGDJkCJYsWYL7778frVu3Rt++feHlpYn+TEREHqvOKLtx40Z07doVTz31FEaMGIHg4ODmGhcRaYh9zw/ebGxedVZ3rF+/Hvfccw/eeOMNDB06FAkJCUhKSsLVq1eba3xE5ARNbfpv3ZDUpO551Hh1BumoqCi8/vrr2LdvH95++220bt0ab731FkaNGgWr1Yrdu3fDaDQ211iJqJEsb70J+chhWN56s1GvZ4tS9TR4MUtBQQE+//xzbN++HUePHoWvry8efPBBLFmyxFVjrDcuZlEHr037TIP6A2VlgI8PDAeOAfCca6uJu12bUxezBAUFYcqUKfj444+RlpaGmTNn4rvvvmvK+IjIxaQpU0Xj/yksl3U39ZpJ37hxA0FBQdDpKmP6Dz/8gM6dO2vqZiJn0urgtbknXpt2NGkmvXbtWowaNQrp6ekOx5cuXYoRI0ZwSTiRG6h649C8fBmu9O4D8/JlKo+MbqbOIJ2SkoK3334bs2bNQq9evRwee+eddzBnzhwsXboUu3btcukgiahpqlZnyFs2QzYaRUtT0rQ6g/S6deswf/58pfG/vcDAQDzxxBOYNWsW1q5d69JBElHTVKvOCO0KyLL43YWaWvpHNwnS586dw6hRo+o8wZgxY3DmzBmnDoqInEsfHQPvNesrF6JkXQAkSfzuQqyvbro6g7Sfnx+Ki4vrPIHFYoGPj49TB0VEriXFTYYUGHjTzQGaivXVTVdnkI6MjLzpprBbt25Fnz59nDooIk9mSUuFaeJ4mCaOVy0N4J04D51PHnf5PojVZvDUYHUG6SeffBKffPIJ3nnnnWoz6qKiIqxatQofffQRnnzySZcOksgdmJcvg2lo1E0rJqwbkoCLWcBFpgHo5upssNSnTx/87W9/w4svvoh3330X3bp1Q6tWrVBQUIDz58+jbdu2WL58Oe65557mGi+RZtV3E1hdfAIseXnKn52NzZA8y017jY4YMQJffPEF9uzZgxMnTqCgoADBwcGIiIjA0KFDmY8mqiDFTYa8ZfNN87z66BiXBk/7m3UM0u6vziBdUlKCN954A7t27YLBYEB0dDQWLFiAwMDA5hofkdvwTpxX5wy6uejiE5SZNLm/OoP0ypUrsWfPHsyYMQM6nQ4ff/wx8vPzsWLFimYaHhEBjikMa0a6MmOv6cafq2fq1LzqDNK7du3CsmXLMHjwYACidenUqVNhNpvh7e3dLAMkIscUhnzqZL1y3+QZ6qzuuHbtGsLCwpS/33XXXZBlGbm5uS4fGJE7qm2FXVNX3tnXG0txk4GAAJfXOJM21DmTtlgs0Ov1yt8lSYK3tzfMZrPLB0bkjmq7adfUm3n2KQx9dAxn0C1Ig/tJE1Htalthx5V31Fg3LcFLTk52aK5ksVjw2WefVesjPXUqm4kT1XbTjjfzqLHqDNKdO3fGJ5984nCsffv2SE5OdjgmSRKDNGmKfTUEAI9e3MHFK56twXscahl3ZlGHFq/N/Pg0yFlZkEJDAUD5s/ea9Q06jyuurbFBtabXWdJSYVm8SHS0C+kAKTi4Xue1pKVC/+kGWH4T75GBXYv/Juvi1D0OidyBfQ7Y1fnghlZuNLZ9Z02vs25IAiSd6A0N1Pu81g1JKD93nr1D3ACDNHkk++5rTe3EdrMg3NCg29gPjZpep4tPgNSzJ/SvLoF+9tx6n1cXnwCvsDDeyHQDTHc0E3f78ashPP3arkx4uM50iTNzws2ZX/b075s7XZtbpDu++eYbTJw4EQMGDEBMTAw2btyo9pCIANx85tuUmXrVWbp1QxLk06dhWbyIW04RAI0E6ezsbMyZMwdPP/00Dh8+jL/+9a94++23sXfvXrWHRuTSxvVVUyW6+ARAtgKSxHwxAdBIkL58+TJiY2MRExMDnU6HiIgIREVF4ejRo2oPjcilqs7S9dEx0L+6BFLPXvXKF3OjV8+niSAdGRmJ1157Tfn7jRs3cPjwYW7LRS2SbeYOoM4AbElLheUPCyAfOwbL6pUuHxc/ENShiSBtr7CwEE8//TT69euH6OhotYdD5FJ1VYbcrGrEuiEJMJsBq6XGx2sKqk0JtNz5Wx03XRbenM6fP49Zs2ahR48eWLZsGXS6hn2GtGun7c0I6rqD6+54bY1T8vRMGD9Yi8AZj8HP7n1uLFkC0/FM6NoEo83TMx0eA4CSnTtRUJgPS8eO0LdqhaAF86o9J+fTDZAvXYT+008QMmUiSnbuRN4rLwE6HfSffgJMmdiga6ttrFrlKf8mNVOCd+jQIcyaNQtTpkzBc889B0mSGnwOluCpg9fmfKahUUBRERAQAMO+g9Uet19RWdsqSktaqpIG0c+eW1E5cgqQZehfXYJbpkzk900jNF+Cd/HiRTz55JOYO3cunn/++UYFaCJPYt8zuqYURV1lgZa0VJgmjq/MUxcUKLXXUs9eYuFLlUoV5pu1SxMz6TfffBPr1q1z6LYHAPHx8Zg/f369z8OZtDp4bc5R20KW+sya7V8v5+UBF7PEwdtDa+3nYX9t9X0Pd+Fu/ybrmklrIkg7C4O0OnhtzlFboKzvKkTTxPEiOLdtB1RsFq2fPbfW19hfm6d10nO3f5N1BWlN3TgkatHCI4AfMiFfuiS61NntxFJT4FRmzq2CgMMHAV3FLkqBgTAkb2/QW7PftXZpIidN1BLcNO+bmQGUlwO/5Na7k52clQV8tQcwGoGSYkj97oZ+9tyGvzdpFoM0eSQtBSXz8mUwDR4Ay8LnIZ8+5RCA7cepi08A2rUH9Hoxq65B1edLoaFAqyDxYHDbWpev2wK6ZfVKzXxdqH4YpMkjaWnhhbxlM1BaKmbJsgyERyiBsuoGtdJttwGtgiDv/armXcdXr4Sc/h0sq1dW9hRZ/BqkgZHQL/xD7R9O4RFAaQlgNGrm60L1wyBNHkntjV/3n83Fi8mZ2H82V5TT+foCHTpCipsMectmyKdPV5bFhYYqgRvhEQ67ydQaTHN/hmloFEzPPuNww6/qh5MtaMt7vwJ8/YDAQG6I62YYpMkj1bdzXW1Lp00Tx8M0cTxKdu686XvZB2SbbenZuHyjBNvSs+GdOA/6JUshdekigqUkAaUlkDO/hzUjXVRyZGaI/HJmhhh3LQ389bPnQup3N1BWJha7fP2lSGMsfQOmoVGQS0vFjLkiXaLkrQFIoaHQz57rso5+5BoM0tSi1bol1cUs4GIWjB+srfF19sF9W3o2Ll+7gc8271GCfa+OgSgrt6JXx0BlH0L59GkAgNSzl8g7m0wiFYKau+HVFExtxw9MmY0/jl+IA7/6nZh5F+SLoH3iuJgxZ2Y4nJfB2X2xBI9aNF18gsOu4rZjlrw8AEDgjMdgrOF19sH9wbtGICVHh3Gnv4b1jAX66BicumaEj5cOp64ZYd2RJGbPslWpWzYvXwZ5y2aRCkH9SuBsOWyERyCltCuyO92CzzsOwJDWA2FZ+oYI1IOHQCoscGx9Wo/A7Gl10p6EQZpatJqCmO2YJS0Vxg/WwJJfUu059sF90OJFGGQ0AjoddH9ZDgCY0K8TtqVnY0K/TtC1TnAIgJa0VCAzw2F5tn2QtGakKwHcO3Ge8p5K6uLUScTeFoGUnsMRe8ttInBbrZD6hsN7xapGfR2q3sAk7WCQJqqFdUMS5EsXIdcQuOyDu31Qtc2SB27ZjMi4yfCeOA/o7vhBUFNAtC+Rw/lzgCxDXv9PWCL6Vc68M9JFdcjQ4Rh0+CAGXc6EdLlnjT8NNJQzzkGuwZw0US3qu6O2d+I8GPYdhHfiPFjSUiEnrROlbhX5Zvv8tSUtVfTWCAqqvut3RVUHbJ0NZBG8lXOazUB5OaTCAuhfXQKEhIhzoTLIVi29Mz37DEwDwmF69pk6r8GVW4RR03AmTVQLfXQMQhrYztO6IQnw8QHKypR8s/3MWWl+dHuoQ0C0T7FYVq8UKwgDA5XgC4MPUFYKhHSotdyuxnTF118CVqv4ndwSZ9JETqSLT4DUpy/0f1mu5JMbUrOtj46BIXk7DLv3KD2gER4BdO4MdAuDfuEflCBsX2Ntm53LrYJgGhoF8/Jl4oTDRwI6nfid3BK74DUTd+vK1RC8tvprSBWF0tWuXXugyAhYLIBeX60ftH33PPnUyTo3C7DH75t2aL7pP5GzaKlnR03qyv1WG7vRKJaS5/4sSvjKyoDyclgWL3K8vool39/eNQJ/nPwqDvQcrKRayP0xSJNHqW/PjuYK5rb3MT37jGMaoqbn2vXlACB6Qnt5if7QIR2A9iHiuKRzvL7MDMDXDyklrZEd2B6fP5LoULpXm5KdOzX9gUYCgzR5lPrmf13RgKmmwK/UNn/9JVBUpFR81PRaXLksbvJVsC0B1y/8A6TgYKCkRJTghYRUqwyBTofYIynobPwZE/p1qtcYjR+sYbMlN8AgTR6lvqVkrmjAVFNzI9sNPdzZR6QsIqMcXqMEzdUrAW8D4O+v9IO2vxZdfAIgW8U5rl9zeC0AoLgIg84ewmubFyPqwtFaZ8hiM9rTsCxeBO8B/UUeu+rNRtIUBmlqkVxRF1w18Fs3JAEFBZCCgyH5+gJt2+GA1FZpxmRJS4Vl0ULIx44BubmAbFUWxNhTloNHRgEmE2CxiJpou/SI/ca1df2UYB/szUe/E82dDh+sc5ZP6mKQphavtvy0JS0VOXGT6p2zrRb4wyMAYyHkS5fEn3U6pAR2x+XLP2NberYIoqWlONAlAn8c+gQOdBuoNEayH5OSMjl8EAgIAPT6yp8ArFbg8k/QRfRTFtTU9VOCPjoG+leXQOrZC4EzHgPguDM5aQ+DNLV4tc08rRuSUH7ufI0z0roCu+24vPcr0ez/+jXImzYABfmIzdiNTjkXEStdV1YLpoSPRnbrjtjRJ9phFm4bky3oSnGTIfXspcyWpWEjAG9voMwEy6KFDf4w8Rs7FoDjiklX03r1jRYxSJPHaWggqG3mWXVZuH2facvqlQ6B3bb82vLayzWnGip2Zhl07QT+FHwdUSnrgIICoOMtiD2zD519JTw0eZTDQhUEBYlNaVevhC4+Ad6J86r1noaPLwAZKC1V3rM+N0Ub+lOCs2hpxxx3wSBNHqehgaCu3s0hWz51aIJk6zMNwDGw25Zf5+crfTmkYSMc38jPX3SqS5yn1DZLv3oQQ2ZOwWtfvoOB86crQVMfHSMqOn7JBS46XovDh0pxUeX5q+zuUtdN0bp+SnAltXfMcUcM0qRpjfnx2FmBoOpsUxefANweCtweWj0A25Zde3lDCg4WgT0zA9DpxfEuXSD1FB3rLGmp4iadJAGZGSJQXjgPnDtbWSMNiEBeXg6Ul4sdVyroo2OA8AhYFi8CDAZx0N8fctI6yD/8ULm7Sx03RevbPMrZ2Mip4RikSdMa8+OxrWStpq5wDX1v+9mm0lcjebuScrA9ZlixCtKjjwG+PsrWVSIAVrQpuH7d4byQdGJnlrw85aYiAODc2cpSuIqbiACAHzIrUy22IF9UBFgskAZGiiXjVitgKqs18Np/4FX9KYG0i0GaNK2xs2Jn5D7rmm3aj0vZ7HXnDqC0VNwwhOgzDZ1O3NxrE1w5nvAIUQbXJhgoKKjc97CCvGmDOF+rIMc3PXdWmW0rFRlTpoo8dYcO4jm33lpr4GU+2D2xVSlpWn23f6omPAI4dVKZ1Tb2vWtrVWo/LluDIxgdnydv2SxmuAEB0C94sXLnlQ1JYh/CgABIwcFiNm2pXGkISYKc/p3YB7FVKzFjlmXxC1BK7hDRr/KnhcuXKx67XOv1sLG/e+JMmjyGQ/66op+FQ8qgEeey7RZetXm+/XspLUMHDxG9NoxGMYbIKDFDDu3q0Pmu6uawuD1UzKx1OqBVUGUwbhMsNq0dPlL0qA5qrTRasrw4H5bnfw/5yGFYXltcr5akzAe7J7YqbSbu1jqxIbRybfYtO+1njQ3diBUALIsWAiYzvHp0h27zv2EaEC5yvjodDEczHd7Le836yvdP/w4otwA6STREslqBG3niBmCHjjDs3qO8n7J5rMUidl2RJKBbmAjahw+KIH/siKgYAYAOHZUl4VUZvjve4K+XVr5vruBu11ZXq1KmO8hjVA3MDZkxVts9paKaQvnIHz5SlNnd2Ucpc0NeHuS8POXmpJyXJ3o/X78mgnNBPqS+4ZDzb4hzFORX7rxy5bLyHgBEgPbyEq/N/Vm0Kd33tQjuNjnXxWzZagW6dBGpjYqGTKZnn4GhkZvQkrYx3UEeo+qP8w0p36t2g1KnB3x90XqBWIVnWLEKhqOZkHx9lYUkUrC48Wfro4GLWWLZ9ohRIuj6+Ijz3TdM/N3gA8urL4sbgPYBGhAzaB8f8byCAnGsvFw0XQIqzucrbkKGdYd0fwyk/gMqX//lf5v0tSPtYpAmj9WQagb7AK+fPRdS//7QL1mqLJ22qW3LKpw/J4Kq0ShSFZIEFBVVLoCRZZHaKKhIXUgS0LqN+LO3t3jPuMniebfdVvFmOsBSLm4edguD1LcvcOttoiJky2bxYaEMjP+VPRW/s+S2bjZTrk/5nu0c5uXLHGqIa7vBZnsMmRlKhzt8+7/Km335N0ROGgAsFhHAf/7ZblB6EaBDOoiZs5cX0OV26KNjRClecQnw00+An594vo+P6Hxnu55hI4DSEiAySnxYjLwfCAyElDDd4Zrsa6qdhX031MGcNLmFmvYGtJ8p1xZQb5aXVtIUmd8DrYJqPVc19iV+ly6J9IWXl9jVW7YCMIjtrg4fFL/b+PuJG4AFBUBQEKQuXRw/RKwVAd5qFYE3MwPyTz8BFy6Im5mdbwV8/SAVFig3LKtSZu8V19eQG6hA7fsw3uzrTa7BmTS5hZpSF3XNlBs862sT3LBFM3YlftLYcWKrq6HDIfXsKVqBPhJf2f4zuG3l62xN/4OCgNvFxrHWjHQAYicW+PqK2fatt1U2VCopEcG7tBQwGm86Tvvl6wAavICltjQR+26ogyV4zcTdSoIaojmurSG7bAOosUSuMeet7dpMzz4jbtYZDCId4eUN+PpU28nboZrD2yBm2b5+YkeUHzJF4PX1Bbr3AH7IFMHeaBQLWXrfCWRdAIqLK7fVCmoNw9f7axy3JS0V/0vZix3hMXhoZB/c271dnddX27U19GutRe72/62uEjwG6Wbibv9oGkKL19aQQNOYQGYaGiWCqb1WrSD17AXvNesrm/VfuiQ62UkSDnQOR8rAcYjNO4nBtwdBXv9PkcuWpMqcdhUHQvsjJXw0YjO/wKCsY4AkQZr2fxUNmnSQevZ0qNN+qcsYZAe2R6eC6/hT8PU6e0Rr8fvmLO52bXUFac2kO06ePIlHHnkEd999N8aPH4+MjMatFCP35cwbUw1ZXXezKpCadvyW4iYDXl440HUA/hi7AAcenAapZ69qTftRkC9mwWazaO7v0wYprXqIPh+2wFxTgK7o5SE2BLgFO6J/Jyo4fHwqO+jJViA8QrlJiPAIxOafRqf8a4hN38XtsDyEJoK0yWTCrFmzMHbsWBw6dAhPPfUUHn/8cRirzlSc7PFvpii/SH1qNACypKWK2a6xsMY+HyU7d1bsQ3gU+GqPshegd+I8SP3uRsrdv0J2u1vxeWSsw4wW4RE4cNcI/DF+CQ7cMUjs5p35BToViACKgnyg4y2VDZhatxE3Hm2/KlqQ2l7z0Lgo6P+yHFLfcGWHFv2rS0Ru3NbjOjMDw5a/hj+1zcGg3B+5HZaH0ER1x8GDB2E2mzF9+nQAwLhx4/DRRx/h888/x+TJ/IfWUjRnAyAlHZGXJ9IRQI19PowfrAFMZmVJOAyGyuAXHoHYb/dgx32TMKFfJ3He1SuBc+eAI4eRErsA2SUyUiIewOBAM4bEJ2BwRjrk3B+BwUMgFRZADggQlR46HVBkrFxhKEmAJGFQ1jEMunEehj9Nh+WCeEgX0Q96uzSGJS9PfHhUrH70TpwHNMNWWNQ8NDGTPnPmDLp37+5wLCwsDKdPn1ZpRNTcnHWzqj57DwJ26QhAqYSwtR21rzEOnPE40LWrqFcOCKjcWQUAMjMwKPcMXvvyHUS+kWj3niJ9Eft9qpg5H/wMcub3+N/3F/Fy6K9w5MOdwInjkI8cFotgbN3z7JeA6/ViNu3lBbQTNwAtixdBPn0alrfehGlQf5geuB/WjHRRqy3LwIXzDdrrkNyDJmbSxcXF8PX1dTjm5+eHkpKSBp2nXbvABj1/2693NOj5TVXXzQF319Rry/l0A+RLF6H/9BOETJnYpPNYz/wI6ysvoXVrP2XFYNXzlzw9EwVvLYMMGa0XzK98XtwkWC5dBADoP/0EmPEYDB3aAz27w5S2B3JGOrzfX4k2ixah5OmZMH6wFtZfcmGtOHfrPyzELzOfAqxWMQvOOqaMLSVHh6t++dh5XI+BuRULXGQZsFjg1aY1yk1lwI0bIkAD0LVrCymwFVovmIf8t/4iqjx8fMSCmbIy4NpVIPlTSH5+kAsLldx3Q76GtX3fSnbuhPGDNQic8Xi1VZfuwlP+v2kiSPv7+6PMvuAfQElJCfz9/Rt0HlZ3qMMZ12b5TTzkDUmw/Oa3TTqX5TfxkBcvAiQJee++D2Pk0JrPHzkUllbvQ87KUp5nSUuF5frPoq45MBDmXn2Ql/g8ZFTcpDObAQBF6z9CcZFJ3JiLjALKcwBJB9PxE/jlmbmi3WhBvgjAer1Y4FJagtjTXyNFp0PM57srS+oAwD8A5eUWoG070RFPloEOHWG9cgW4eg2/LF8J/HhGPLdTZ0jDRkDe+LHYNOBXD0LOzIAUHiFWLBYVwXQ9B1c3Jt/0J5K6vm/mdx2/Nu7G3f6/ab66o3v37jh//rzDsXPnzqFHjx4qjYiaW2N7HVdNY+ijYyp7YFTcCKwtlVJ1cYZ1Q5JY6t2lC/Sz54r+GKWlIlccGVW5VDu0K+RNG0QJ3pf/FTXQRUYRmEtLgfwbOHD73fjjuPk4cFsEYCoDZBmDrp/C69uXitm1LUjrdEBZqbLHIYzivfSz54rXWa2ifrqiVE8/ey68E+fBcOAYDLv+qyx48U6cB0Pydki33aY0fWoKVy1cMS9fplTIUP1oIkgPGjQIsixj3bp1MJvN2LFjB06dOoWYGPcspKfmU7UipOomrzU9x6bqB4MuPgEwmyAfPSIa6Us6ZWcVqbBAzIolCTh9yrGLXWmpeMzW5MjLCyl3xSC7dUekhI8W5wBEmqJdO/HnLl2AsO6ieVJpaWXQlmXgm72i1zQgKj/6hiv9OfTRMbCkpWLvzOew8PVPsPcPf3YIelWDa2PLGl21QYBtb0aWB9afJoK0wWDAP/7xD+zatQtRUVH4+9//jtWrV6Nt27Y3fzG1WJa0VKUTncNsWNIBsqwcq++sUB8dIyotZBnIvwGpZ08EPDVTqX+W4iaLnHBF2gOA+HvHW8QWVzqdCLy+vog9sw+dyvIRe2af45uUlcHw3XEYtu8SG9oWF1cfiNlc2ZO6vBwoKYH+1SWQsy7ANCAclldfRkqHCGT7tEaK1NEh6FUNrrYPKMvSNzQxg1X2ZmR5YL1xxWEzcbccWUOodW01Lf1ubJWIUpJ3/AfRK6N1axi+2o+QkFa4ujHZYX9C+dhREUD7hkP/2BOwrF6JAwhGyp2jEJt/CkMejhZ9owvyRR8Oby+x9LusDNIj8Q6rAE2REZVVHV5e4rx6LxzoEiHOZ1tp2KEj8HOOeFyScOCeB5DS5R7EXkvHoH3bAJ0OUsL0aisMlev6IVPcbAwIgGHfQQD8N6klXBauAe72j6Yh1Lo2Z5btWRYvEjPwkBBIwcHKrFv+659hvZIt+m74GIDgYNH1DhApiF69IZ8+hT8OfwrZQR3RqeAali75v8rttgDo//o3h/GZnn1G2eUFly5V9pi288cJLyA7qAM65WXj9R1/EUF+8BCxoKZVEPSLXxNjX7yocnl6YKASgKsyL18GectmSHGTlUDOf5Pawe2zyCM1eifxKkSKRFRw6GfPVY7JeXnAlWzxJLNJNOC3XwUbGQWEdgWOHUVs5heix0bWAZiGviuCeW6umG3bNUASi13Oitf/kCl+r1i4AqsMpcY6fZc434k9gK+vMgO3/fSg5NclSeStdbo6Uwhc4OK+OJNuJu72yd4Q7n5t1TahtZtV4/xZu40OJdj9BdLASMh5eThgCVIC6qBzh6s3RbJ1yqvhHAAqZ8nf/k+U69nWDBTkA97eStMm21gtq1eKDwtJAgICoJ89t1EfVu7+fauLu12b5kvwiNRkf7PN8tabIgCWlkA/ey4MD4yxe6ZdcPXygi4+AQcCb8eq4f+HrLa3IeXOUQAqmiIFVVR2AHYBuuIcYd1Fjtmmew+xOYC3twjQFekP6ZF4h6ZNtrECEDcWc65DCg5223aiVD8M0uTRqpagmZcvgykyAqZ7+imVDg7PuZEnXmg2w5qRDvM3+8UNvSp7CEpTp8G6IQkpYUMgyVbIkJRKDqWRUuYX1QfUujWk4GCxUYDND5liyy3bPoilpcD1a6JOu1UQLIsXOYwVFSsiIUmQ8/Ictv4iz8N0RzNxtx+/GkLL11a1AsShD3TFjTb75yA8AnLSOlFaJ8uiHE6nA7rcLvYqtL/J5+uHAxEjkBJyV2VqozY6feXWWVeuiIUqQGV9dGAg9K8ugTUjXSyUsVjEzi9FRjGOiqoM8+PTKqtLfHzE6sbSEmUjAfsmVU1Zceju3O3amO6gFqtqjbTDzbXQrtWe4504D/q/LK8IfqUiWJrNFasBq/ynLy3BoJ9/xOsZGx0C9IHQ/mK1YY97Kp9rtYia6KIiEVStVrtVh3plRxfvxHkwfHsU+j8vEznxwFaAj48ybl18AtC1GxDWHdKUqZBCQ0XrUrsAXZ/e2Jx1uw9Wd5BHq1oB4p04D6aKVW/IulDrayyLFlZ/wL7fBiAWseTkiL4eHW8RDY9gl5PuNRKDsr6rXPxitVY8p8rNQ6ul2pZbSpA1GCDdcYdSNnezihbL9s+AX3Ihh99V4+MOQdyuCZMnbJnlqTiTphZHWTnoH6BUS8jp34mqCZug1rW/fmAk9H/9m6hzLiwQM+SKrnYHQvujwCcAgWVFIidtvzpRITvmuCXJYdcX+zaqDe6fcfig+DD4+ssaZ8u1rb5UY8MFqh8GaWpxvBPnQeorFptYVq8ELv9UUaNcmQ6Qxo4Tu6VU1Tcc8pkfYZn3LLD3K3HMahWvlySkhI9GkV8rtCozVs9Re4vdVqDTAf7+kB59TFR0yLJo1GQ0Ql7/T7FTTFAQpGEjar0G2zir3jSU4iYreyY6fOhUqK0nhyt3AmdTpaZhkKYWyRaUAIg6Zn8/6GfPVWaU8t6vxCzZXlh3sWQ8P1/pAw0fHxHM27cHOnRE7Ln/odONq1UqO6SK32WR6/b3F8G0hp1gIAP4JVc08s/MqHV2q4xzy2aH53gnzgNCOojxNWD7OVc1VALYVKmpmJOmFsmW260pF6usNtR7AdbKGmf97LmOs9O+4ZB8fUVqIjAQuH4Ng659g0Envql8jk4nflllwM9fVHUEtRYfAgUF4pg5X3TRGzZC7FUIKDPa/6XsxY7wGDx0NheR2/6pLO223SREeASQmaHsKqOsnvTyEmPSAClusjJuajgGaWrRqt6Is2akQz51UizrNlcE6IrmRQDEDUdApCnOnoEc1Lqy+kOZMVfw9wdu6QTkXBevs5XvmUzizwYfMVu3LQu/mKX0DbGNaccFX1y5UYx/b9iNgTs2AGVlkLdshj5xXrVZr62UEEFBkG67rVn2iqwPLklvGgZpIjvyls0iTWBLFeh0wPCRkDd+DMv6daKSAxCB2WwWZXqSXTrDXkUbUqlnL8hHj1Qe9/YW+WljoUhLAGKfRaMR8sUsWPLylAA8LjMVKQFhiD2eJm5mFhfVOiNtSI00uQ/mpInsVA2AAU/NBPZ/I9p8ytbqL9DpKnttKCeRlMf0s+eKvhsjRonnBgaKcxUWALfeKv4+8n6Rgy4tFW1Lf/5ZuRk4JHYYXj++BYNwA9LYcZB69YYuop/Trpc39bSPQZo82v6zuXgxORP7z+YCqHkxh/0x78R5YtcUAEmREzGxqC+SIuNqPrmtjK7qhskGg3jM21s5pB//EKT+Axxv5l26JBaiFBZAPn1aBG6dHjCVKTcD9dExMCRvFxsE1HEjEWhYGZ0lLRU5cZPEXom8qadpDNLk0balZ+PyjRJsSxctR2sKZFWPSRXNj3bfORIleh/s7nFvzSe3WzWorDIM7S9mypIkSvzmJ8K8fJmoxT5yuNopbP05lNy0v59orFRDOdzNyuQaUkZn3ZCE8nPnxWa23ClF0xikyaNN6NcJt7bxw4R+nQA4BjLbDBrhEQ7BTRefAHToiDGnvoJfeSnGnK+5kb69ap3vbMvJrda6Z6nBwaJGGgCsVqUpvy2/bJo6GaYB4TA9+4xjt74afiJoSBmdLj4BXmFh0C94EYZ9B6vt6ELawQZLzcTdGr40hLteW03bb9kzTRwvOs4FtwWKixxTFTpd5TJxb28c6Bzu2EPajvToY9BF9INlfqJ4jS1NMnyk2KHFdh5JgjRgoPgAWbxIzKwLK7+u9ju83Gzs9eGu37f6cLdrY4MlIju1zaCrKSoCzGYcCOyCPz70okhl2Nj38TCbMSjrGF7f8Zfqqww73iKaNkXHiDK+il2/DUczYVixSgRqSRL10t3CKuufKzbTRd9wcR4vL4cUjStXCJK2sASPWhxbDvpAUFd8Pm4BJnTtBPuss7L7SUU/jpQ7RyFb9kFK+OjKIOzt7diXw35mXeFAWCRS+tyP2EdnY/Dd3cRNOnO5WMhSkV4wrFhV6xhtpXRVd44Bam60xCZJnokzaWoR7HO4tlnojvAYh5uKNtYNSWLln9UK+PkhNud7dCrMEUu99XrRc0Ovr3yBbZfvKlL63I/sVh2QclukCNBlZaJladYFsfHA8HthmjhejG35MpgGD4BpzChYM9LFODLSxYwfqFeumU2SPBODNLUI9gHMdoPtoZF9HG4qAiKYy3l5QNt2QNduCPi/RzHY8jNe3/lXMYuWJHEjUJJE6uLRx2oM0AAQe/y/6FR4HbE/HRYN/JU3sYh66IJ84GKWqPxIWle5I8vH62vsy2H7oPlm638dygodxh0U1KAUCPtLax+DNLUINeVw7+3eDm9ODMe93dspx6wbkoCCAkhdusCQvB3mo8fwbZEBf/zVczjQI0qkNYxGsbQ7Mgryh2trDtI6HQZl/4DXv3wHQ6b9GmjXTsy47QW1FisPz511PIcsi40BIqOU3WLMj08TwTwrC9tO/lLtJwBl3A3c85Czb+1jkKYWob7laVWDeeCMx5HScziyW3fEjoGxlZvKWiyid3NtunYTKRFJEgHw9lARiA0V7Uo73gLD1/uBstLK1wS1FisQAwLFrPpilqjcqFjEAoj+0hN6t632E0BjbyTyBqT28cYhEexuFkJ0u7MFc7+xYzHh0mfYfvIXjO/dFhh5vyibu7OP2JWluFjMfL28ROCumBHrZ8+FNSMd336TiZQuoxGbsRuDrFZAliENjKwMisNHijppnQ7Sw3Fi55iJ44ELRuDSRZgmjoc0bAQkQLkheB+A+yrGbH6j8kZhY24WNvZ11HwYpIlgd7Ow4s+W7Z8BX3+JnDGjcd+fV+A+2xMfvh9ARZ2y1Qqp/wClTtm8fJnYRNbgowT8lD5xyPZvi5Q+92PQucNAcFt4r1mv5IL18QmwFhaImXJFf2n97LmiTrq4GLhwAfL1a8oeiDaWtFSxxVdpKSw/ZIrXMdh6JKY7iFCxyvD2UOD2ih/9KxaZmHZ/4XBzzZKWir2JL+OlsFgcCOwCuVVQZd01IGbUxkIl4MdeOoROBdcRa70KaWAkpLHjHPLLtpI5+5SDPjoG+leXiJSJwRuQJFhWr3S4wWfdkASYKkoATWbmlD0YVxw2E3dbAdUQnnZt+8/m4t/JXyP2my0Y1rs9TKfPiqB7eyik4GC81GUMsv3bolP+Nbz+1buQevUWM+HSEjH7ta0qNBjEri0WS+VSdNtKwpAO1XpHV2WfggEgbgxWrDBUHjMagcBAhxRNfXna982eu10bVxwS1aC28rNt6dnIbncbPp/2AkLWrBGBsLwcyL4C+YdMxJ7bj05mI2J/SAMio5SZsBQ3ubJKw2oVN/9u5CkVGpbFi4AyE1BUBGnYiJveyHSo2Jg9t9ps25C8HYbde2BI3l4tFcKyOs/BIE0tkiUtFZbFiyCfPlUtVWBryhQrXUdO3KSK/tCSaElaWopBp/fj9Z3LMCj3R0iFBdBHx4jgmZlRudksIG4GTpmqVGhAkgBLORAQUPP+hlXYp0Ea0jyptrI6Bm/3xCBNLZJ9f4yq5We2+umolHWinWdAAA50tWtFGtS6onF/oWgzisrACL1O2QhASpgOee9XMI0YAjn9O8DPX/Tv8POH/EOmQ6N907PPKN3ubBq7OWxtZXWsiXZPDNLUIuniEyD17FmtaqLqc7zCwqCfPRcp909FdptbsCP6d9Av/ENl344v/yuqOvLyRF20wQcwGCA9Ei9my+fPAfk3RLqkuEjsqvJzDlBaCjlpXeWs1tYN7+svm3xttQV31kS7J944bCbudiOjIVrCte0/m4tt6dmY0K8T7u3eDqYB4ZX5Z19fschFliu3zvL3F7tkf7i28mQj7xe7sBz/oXI3F19f6JcsVUr+MHwkDCtWVWuqVFMNt7OuzRO527XVdeOQddJE9XBv93a4t3s7ZQEJWgWJGbK3t0h/XL8mArQsV6w01AGZGZAefUz05TAYRIBuFSRSJX5+IlBXlM8ZqvSErpaasKvhrqsahF3wPA+DNFEDKLnndu3ELwDSsBGi/eiVyyJoVymvMwOQN34M+cwZ0VRJliuXl7dqVS39UFOzJEteHgDUmqqw3Qi1LUNnkPYcmgnSH374IdavX48bN26gW7dueOGFFxAZGan2sKgFsk9tTKjyY6itKb8uPkGkIC5mQd77FQzJ22ufyWZmiCXj+TcqTqITs+6Kao+qAVUpvQsNVR6rT5tScSPUypyzh9HEjcPdu3fjgw8+wHvvvYdDhw7ht7/9LZ588kn88ssvag+NWqCqm9fas5XbWTck4UCr20V3vPZ3KI/Z70G4N/FlvLBmLw7GTherGTveAoR1FxUebdsBPj41bgDbmBt89bkRSu5JE0E6JycHTz31FHr06AGdToe4uDjo9XqcOnVK7aFRC1R189qqbCmPlB5Dkd3uVuwY9HC1GmTrhiSktO6JKwWlSJE7iIUnu/4L/ey5oh91STGkvuEOG8DaLy9vaOldY8v1SPuaLd1hMpmQn59f7bgkSZg6darDsUOHDqG4uBh33HFHcw2PSGG7SVgb20x6QnhHpMgdMKFfJ1jfSHTYVEAXn4DYlL3Y0SXUcVOB1SuB4hLA4K3sWF41fWLJy2OwJUWzBeljx45h2rRp1Y7r9XocP35c+fvp06eRmJiI3//+92jfvn1zDY+o3mztPW0tQwHAYpertj1nWHQMBj37DLBqD0xBQdC//Jp4sk4COt8qbipW7Ppd3wUm5uXLIG/ZDClussMsnDyXpuqk9+zZgwULFuCJJ57AzJkz1R4OUZ1Kdu6E8YM1CJzxOPzGjq3xOZe7hCr11F49e0KGDAkSghbMg9/YsRXnWAvvAXej9Is0h8dqcqV3H8hGI6TAQHQ+ebzG55Bn0VR1x4oVK/DnP/8ZDzzwQKPOwcUs6mip12Z+933IWVnIe/d9GCOH1nwCW1N/AOVXr4l66dBQGCOHwphTCEQOBSKHovjxaZB/ueH4WE0m/gbYshmY+Jsmf81b6vdNizTfBe/zzz/H8uXLsW7dukYHaKLmVp8qDMOKVWJLLAAoLAB0Osh5edWaHNW3osM7cR4M+w4y1dGCaGIm/Y9//AMmkwnTp093OP72229j1KhR6gyK6Cbqs/WUJS0V8A8AiooAHx+guAiwWqstOOE2VlQbTQTprVu3qj0EoiaraTGLdUOSyEl3C4MUHAyERwCZGVxwQvWmiSBN5Ans+23YgrT9CkXOlKkxGKSJnERXpQwPYBqDmk4TNw6JPEF9V/2Zly+DaWiUQ9N/G+6eQlUxSBM1k/1nc/Ficia+/fYEUFQEedOGagGZu6dQVQzSRM3E1rhpx32TxD6HQa2rBWTunkJVMSdN1Ewm9OskWqCOiIJh7sFqu68AzGFTdQzSRM2kauMmBmSqD6Y7iIg0jEGaiEjDGKSJiDSMQZqISMMYpIk0igtbCGCQJtIsLmwhgEGaSLO4sIUA1kkTaRbrqAngTJqISNMYpImINIxBmohIwxikiYg0jEGaiEjDGKSJiDSMQZqISMMYpImINIxBmohIwxikiYg0zKOWhet0ktpDqJPWx9cUvDb3xGvTPkmWZVntQRARUc2Y7iAi0jAGaSIiDWOQJiLSMAZpIiINY5AmItIwBmkiIg1jkCYi0jAGaSIiDWOQJiLSMAbpZvThhx8iOjoaAwcOxKRJk3D48GG1h9QkJ0+exCOPPIK7774b48ePR0ZGhtpDcppvvvkGEydOxIABAxATE4ONGzeqPSSnKygowMiRI5GcnKz2UJzm+vXrmDVrFgYOHIghQ4ZgxYoVag+p6WRqFrt27ZKHDh0q//jjj7LFYpH/9a9/yQMGDJBzc3PVHlqjlJWVyaNGjZL/+c9/yiaTSU5JSZEjIyPlwsJCtYfWZFeuXJH79+8v7969W7ZYLHJ6erp8zz33yF9//bXaQ3OqZ599Vu7du7e8ZcsWtYfiNHFxcfLLL78sl5aWyhcvXpRHjBghb9u2Te1hNQln0s0kJycHTz31FHr06AGdToe4uDjo9XqcOnVK7aE1ysGDB2E2mzF9+nR4e3tj3Lhx6NGjBz7//HO1h9Zkly9fRmxsLGJiYqDT6RAREYGoqCgcPXpU7aE5zdatW2E0GtGzZ0+1h+I06enpuHTpEl566SX4+PigS5cuSEpKwqBBg9QeWpN4VBc8tZlMJuTn51c7LkkSpk6d6nDs0KFDKC4uxh133NFcw3OqM2fOoHv37g7HwsLCcPr0aZVG5DyRkZGIjIxU/n7jxg0cPnwYDz30kIqjcp5Lly5h1apV2LhxI2bMmKH2cJwmMzMTPXv2xKpVq5CcnAwfHx/Ex8fjscceU3toTcIg7UTHjh3DtGnTqh3X6/U4fvy48vfTp08jMTERv//979G+ffvmHKLTFBcXw9fX1+GYn58fSkpKVBqRaxQWFuLpp59Gv379EB0drfZwmsxisWD+/PlYuHAhQkJC1B6OU+Xn5+PIkSOIiopCWloazp07hxkzZiAkJATjx49Xe3iNxiDtRIMGDbpp+mLPnj1YsGABnnjiCTzxxBPNNDLn8/f3R1lZmcOxkpIS+Pv7qzQi5zt//jxmzZqFHj16YNmyZdDp3D87+M4776Bbt24YM2aM2kNxOoPBgMDAQMyZMwcA0Lt3b0yaNAmpqaluHaTd/1+dG/nwww/x3HPP4U9/+hNmzpyp9nCapHv37jh//rzDsXPnzqFHjx4qjci5Dh06hMmTJ2P06NFYuXIlfHx81B6SU+zYsQO7du1SUjqnT5/Gq6++ildeeUXtoTVZWFgYSkpKYDKZlGMWi0XFETmJ2ncuW4odO3bI/fr1k7/77ju1h+IUZWVl8vDhwx2qO/r37++21Sr2srKy5P79+8vr169XeyguN2HCBI+p7igtLZWHDx8uv/baa3JZWZl88uRJefDgwfJ//vMftYfWJNyZpZk8/PDDOHXqVLUZ2dtvv41Ro0apNKqmOX36NBYvXoyTJ0/itttuwx/+8Afce++9ag+ryd58802sW7euWuomPj4e8+fPV2lUrvHQQw/h0UcfxcSJE9UeilNcunQJr7/+OtLT02EwGDB9+nQ8/vjjag+rSRikiYg0jDlpIiINY5AmItIwBmkiIg1jkCYi0jAGaSIiDWOQJiLSMAZp8mj3338/evXqpfzq06cPRo0ahaVLl6K4uFh53vHjxzF37lwMGTIE/fv3R1xcHLZv317jOQ8ePIhevXrh5ZdfrvHxs2fPIiEhAXfffTd+9atfYffu3S65NmoZGKTJ4z333HPYt28f9u3bhz179uCNN97Atm3bsGTJEgDAV199hd/+9rfo1KkTPvjgA2zduhXjx4/Hiy++iDVr1lQ737Zt29C1a1fs2LEDpaWlDo+VlZVhxowZ6NatG7Zs2YLJkyfjueeew4kTJ5rlWskDqbvgkci1Ro0aJSclJVU7/ve//10eOHCgbDQa5cGDB8v/7//9v2rPee+99+SIiAiHpe5lZWVyZGSknJycLN91113y1q1bHV6zdetWeciQIbLZbFaOPfPMM/KLL77ovIuiFoUzaWqR9Ho9DAYD9uzZg8LCQkyfPr3ac+Lj47F27VoEBQUpx/bs2QOj0YgRI0bgvvvuw5YtWxxec/ToUdx9993w8qpsMHnPPffgyJEjLrsW8mwM0tSiWK1WZGRk4KOPPsLo0aNx4sQJdOvWDYGBgdWeGxgYiIEDBzoE3G3btmHAgAFo27YtYmJicOjQIVy6dEl5/Pr16+jYsaPDeUJCQnDt2jXXXRR5NAZp8nh//vOf0b9/f/Tv3x933XUX4uPjERERgXnz5qGgoKDGAF2T/Px8fPXVV4iJiQEgbkrq9XqH2XRJSQkMBoPD6wwGg0P7TKKGYNN/8nhPPvkkJkyYAADw9vZG+/btlUAaHByMgoKCep1n586dMJvNSsP8Nm3aICoqClu3bsXcuXOh0+ng6+tbLSCbTKZqu9gQ1Rdn0uTxgoODERoaitDQUHTu3NlhphsREYHz58/DaDRWe11hYSESEhLw/fffAxCpDgAYPXo0+vTpgz59+mD//v24evUq9u3bBwDo2LEjcnJyHM6Tk5NTLQVCVF8M0tSiDR06FG3btsU///nPao9t3LgRx44dw6233orLly/j6NGjmDNnDv79738rv5KTkxEQEKCkPAYMGIBjx46hvLxcOc+hQ4fQv3//Zrsm8ixMd1CL5uvri1deeQXPPvssioqK8PDDD8PLywu7d+/G6tWrMX/+fLRt2xbvvvsufHx8MG3aNIdqD0Bs6LBp0ybk5eVhzJgxePvtt/HSSy9hxowZ2Lt3L7788kt8+umnKl0huTvOpKnFGz16NNasWYMff/wRjz76KCZNmoT//ve/eOutt/Doo48CALZv344HH3ywWoAGgKlTp6K8vBzbt2+Hv78/3n//fVy4cAEPP/wwNm/ejBUrVqB3797NfVnkIbgzCxGRhnEmTUSkYQzSREQaxiBNRKRhDNJERBrGIE1EpGEM0kREGsYgTUSkYQzSREQaxiBNRKRh/x9TOCnFWaUUsAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "LR tn, fp: 309, 1\n",
      "LR fn, tp: 11, 0\n",
      "LR f1 score: 0.000\n",
      "LR cohens kappa score: -0.006\n",
      "LR average precision score: 0.125\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 309, 1\n",
      "GB fn, tp: 11, 0\n",
      "GB f1 score: 0.000\n",
      "GB cohens kappa score: -0.006\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 310, 0\n",
      "KNN fn, tp: 11, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "------ Step 1/5: Slice 2/5 -------\n",
      "-> Reset the GAN\n",
      "-> Train generator for synthetic samples\n",
      "Epoch 1/3\n",
      "Epoch 2/3\n",
      "Epoch 3/3\n",
      "-> create 1194 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "LR tn, fp: 309, 1\n",
      "LR fn, tp: 11, 0\n",
      "LR f1 score: 0.000\n",
      "LR cohens kappa score: -0.006\n",
      "LR average precision score: 0.129\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 308, 2\n",
      "GB fn, tp: 11, 0\n",
      "GB f1 score: 0.000\n",
      "GB cohens kappa score: -0.011\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 310, 0\n",
      "KNN fn, tp: 11, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "------ Step 1/5: Slice 3/5 -------\n",
      "-> Reset the GAN\n",
      "-> Train generator for synthetic samples\n",
      "Epoch 1/3\n",
      "Epoch 2/3\n",
      "Epoch 3/3\n",
      "-> create 1194 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "LR tn, fp: 310, 0\n",
      "LR fn, tp: 11, 0\n",
      "LR f1 score: 0.000\n",
      "LR cohens kappa score: 0.000\n",
      "LR average precision score: 0.171\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 308, 2\n",
      "GB fn, tp: 11, 0\n",
      "GB f1 score: 0.000\n",
      "GB cohens kappa score: -0.011\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 310, 0\n",
      "KNN fn, tp: 11, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "------ Step 1/5: Slice 4/5 -------\n",
      "-> Reset the GAN\n",
      "-> Train generator for synthetic samples\n",
      "Epoch 1/3\n",
      "Epoch 2/3\n",
      "Epoch 3/3\n",
      "-> create 1194 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "LR tn, fp: 308, 2\n",
      "LR fn, tp: 10, 1\n",
      "LR f1 score: 0.143\n",
      "LR cohens kappa score: 0.130\n",
      "LR average precision score: 0.108\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 308, 2\n",
      "GB fn, tp: 10, 1\n",
      "GB f1 score: 0.143\n",
      "GB cohens kappa score: 0.130\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 310, 0\n",
      "KNN fn, tp: 11, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "------ Step 1/5: Slice 5/5 -------\n",
      "-> Reset the GAN\n",
      "-> Train generator for synthetic samples\n",
      "Epoch 1/3\n",
      "Epoch 2/3\n",
      "Epoch 3/3\n",
      "-> create 1196 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "LR tn, fp: 306, 0\n",
      "LR fn, tp: 9, 0\n",
      "LR f1 score: 0.000\n",
      "LR cohens kappa score: 0.000\n",
      "LR average precision score: 0.220\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 304, 2\n",
      "GB fn, tp: 8, 1\n",
      "GB f1 score: 0.167\n",
      "GB cohens kappa score: 0.155\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 306, 0\n",
      "KNN fn, tp: 9, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "====== Step 2/5 =======\n",
      "-> Shuffling data\n",
      "-> Spliting data to slices\n",
      "\n",
      "------ Step 2/5: Slice 1/5 -------\n",
      "-> Reset the GAN\n",
      "-> Train generator for synthetic samples\n",
      "Epoch 1/3\n",
      "Epoch 2/3\n",
      "Epoch 3/3\n",
      "-> create 1194 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "LR tn, fp: 309, 1\n",
      "LR fn, tp: 10, 1\n",
      "LR f1 score: 0.154\n",
      "LR cohens kappa score: 0.145\n",
      "LR average precision score: 0.132\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 308, 2\n",
      "GB fn, tp: 10, 1\n",
      "GB f1 score: 0.143\n",
      "GB cohens kappa score: 0.130\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 310, 0\n",
      "KNN fn, tp: 11, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "------ Step 2/5: Slice 2/5 -------\n",
      "-> Reset the GAN\n",
      "-> Train generator for synthetic samples\n",
      "Epoch 1/3\n",
      "Epoch 2/3\n",
      "Epoch 3/3\n",
      "-> create 1194 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "LR tn, fp: 310, 0\n",
      "LR fn, tp: 11, 0\n",
      "LR f1 score: 0.000\n",
      "LR cohens kappa score: 0.000\n",
      "LR average precision score: 0.107\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 308, 2\n",
      "GB fn, tp: 11, 0\n",
      "GB f1 score: 0.000\n",
      "GB cohens kappa score: -0.011\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 310, 0\n",
      "KNN fn, tp: 11, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "------ Step 2/5: Slice 3/5 -------\n",
      "-> Reset the GAN\n",
      "-> Train generator for synthetic samples\n",
      "Epoch 1/3\n",
      "Epoch 2/3\n",
      "Epoch 3/3\n",
      "-> create 1194 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "LR tn, fp: 308, 2\n",
      "LR fn, tp: 11, 0\n",
      "LR f1 score: 0.000\n",
      "LR cohens kappa score: -0.011\n",
      "LR average precision score: 0.162\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 305, 5\n",
      "GB fn, tp: 11, 0\n",
      "GB f1 score: 0.000\n",
      "GB cohens kappa score: -0.022\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 310, 0\n",
      "KNN fn, tp: 11, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "------ Step 2/5: Slice 4/5 -------\n",
      "-> Reset the GAN\n",
      "-> Train generator for synthetic samples\n",
      "Epoch 1/3\n",
      "Epoch 2/3\n",
      "Epoch 3/3\n",
      "-> create 1194 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "LR tn, fp: 310, 0\n",
      "LR fn, tp: 11, 0\n",
      "LR f1 score: 0.000\n",
      "LR cohens kappa score: 0.000\n",
      "LR average precision score: 0.209\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 310, 0\n",
      "GB fn, tp: 10, 1\n",
      "GB f1 score: 0.167\n",
      "GB cohens kappa score: 0.162\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 310, 0\n",
      "KNN fn, tp: 11, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "------ Step 2/5: Slice 5/5 -------\n",
      "-> Reset the GAN\n",
      "-> Train generator for synthetic samples\n",
      "Epoch 1/3\n",
      "Epoch 2/3\n",
      "Epoch 3/3\n",
      "-> create 1196 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "LR tn, fp: 306, 0\n",
      "LR fn, tp: 9, 0\n",
      "LR f1 score: 0.000\n",
      "LR cohens kappa score: 0.000\n",
      "LR average precision score: 0.085\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 303, 3\n",
      "GB fn, tp: 8, 1\n",
      "GB f1 score: 0.154\n",
      "GB cohens kappa score: 0.139\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 306, 0\n",
      "KNN fn, tp: 9, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "====== Step 3/5 =======\n",
      "-> Shuffling data\n",
      "-> Spliting data to slices\n",
      "\n",
      "------ Step 3/5: Slice 1/5 -------\n",
      "-> Reset the GAN\n",
      "-> Train generator for synthetic samples\n",
      "Epoch 1/3\n",
      "Epoch 2/3\n",
      "Epoch 3/3\n",
      "-> create 1194 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "LR tn, fp: 309, 1\n",
      "LR fn, tp: 11, 0\n",
      "LR f1 score: 0.000\n",
      "LR cohens kappa score: -0.006\n",
      "LR average precision score: 0.168\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 307, 3\n",
      "GB fn, tp: 10, 1\n",
      "GB f1 score: 0.133\n",
      "GB cohens kappa score: 0.117\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 310, 0\n",
      "KNN fn, tp: 11, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "------ Step 3/5: Slice 2/5 -------\n",
      "-> Reset the GAN\n",
      "-> Train generator for synthetic samples\n",
      "Epoch 1/3\n",
      "Epoch 2/3\n",
      "Epoch 3/3\n",
      "-> create 1194 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "LR tn, fp: 310, 0\n",
      "LR fn, tp: 11, 0\n",
      "LR f1 score: 0.000\n",
      "LR cohens kappa score: 0.000\n",
      "LR average precision score: 0.172\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 308, 2\n",
      "GB fn, tp: 11, 0\n",
      "GB f1 score: 0.000\n",
      "GB cohens kappa score: -0.011\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 310, 0\n",
      "KNN fn, tp: 11, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "------ Step 3/5: Slice 3/5 -------\n",
      "-> Reset the GAN\n",
      "-> Train generator for synthetic samples\n",
      "Epoch 1/3\n",
      "Epoch 2/3\n",
      "Epoch 3/3\n",
      "-> create 1194 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "LR tn, fp: 308, 2\n",
      "LR fn, tp: 11, 0\n",
      "LR f1 score: 0.000\n",
      "LR cohens kappa score: -0.011\n",
      "LR average precision score: 0.079\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 306, 4\n",
      "GB fn, tp: 11, 0\n",
      "GB f1 score: 0.000\n",
      "GB cohens kappa score: -0.019\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 310, 0\n",
      "KNN fn, tp: 11, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "------ Step 3/5: Slice 4/5 -------\n",
      "-> Reset the GAN\n",
      "-> Train generator for synthetic samples\n",
      "Epoch 1/3\n",
      "Epoch 2/3\n",
      "Epoch 3/3\n",
      "-> create 1194 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "LR tn, fp: 309, 1\n",
      "LR fn, tp: 11, 0\n",
      "LR f1 score: 0.000\n",
      "LR cohens kappa score: -0.006\n",
      "LR average precision score: 0.174\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 309, 1\n",
      "GB fn, tp: 11, 0\n",
      "GB f1 score: 0.000\n",
      "GB cohens kappa score: -0.006\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 310, 0\n",
      "KNN fn, tp: 11, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "------ Step 3/5: Slice 5/5 -------\n",
      "-> Reset the GAN\n",
      "-> Train generator for synthetic samples\n",
      "Epoch 1/3\n",
      "Epoch 2/3\n",
      "Epoch 3/3\n",
      "-> create 1196 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "LR tn, fp: 306, 0\n",
      "LR fn, tp: 9, 0\n",
      "LR f1 score: 0.000\n",
      "LR cohens kappa score: 0.000\n",
      "LR average precision score: 0.217\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 306, 0\n",
      "GB fn, tp: 8, 1\n",
      "GB f1 score: 0.200\n",
      "GB cohens kappa score: 0.195\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 306, 0\n",
      "KNN fn, tp: 9, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "====== Step 4/5 =======\n",
      "-> Shuffling data\n",
      "-> Spliting data to slices\n",
      "\n",
      "------ Step 4/5: Slice 1/5 -------\n",
      "-> Reset the GAN\n",
      "-> Train generator for synthetic samples\n",
      "Epoch 1/3\n",
      "Epoch 2/3\n",
      "Epoch 3/3\n",
      "-> create 1194 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "LR tn, fp: 310, 0\n",
      "LR fn, tp: 11, 0\n",
      "LR f1 score: 0.000\n",
      "LR cohens kappa score: 0.000\n",
      "LR average precision score: 0.345\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 309, 1\n",
      "GB fn, tp: 11, 0\n",
      "GB f1 score: 0.000\n",
      "GB cohens kappa score: -0.006\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 310, 0\n",
      "KNN fn, tp: 11, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "------ Step 4/5: Slice 2/5 -------\n",
      "-> Reset the GAN\n",
      "-> Train generator for synthetic samples\n",
      "Epoch 1/3\n",
      "Epoch 2/3\n",
      "Epoch 3/3\n",
      "-> create 1194 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWkAAAFpCAYAAABee9lOAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAABKrUlEQVR4nO3dfVxUZfo/8M+ZYYZnFBVJLfEpNSUQZUFLU0M0C7XEqPCH+U3Tzcqi1fyaW9aWma2luVpbm6ZhaJr4VVFLIrVszecgfIAURVIEQp4GBmY4c35/3Mwww+MMzMw5M3O9Xy9fyJmZc+6zbBe317nu6+YEQRBACCFEkmRiD4AQQkjLKEgTQoiEUZAmhBAJoyBNCCESRkGaEEIkjII0IYRImJvYA7CmkhIVdDrHqij09/dCaWm12MOwKbpH5+EK9ynGPQYE+Lb4Gs2kRebmJhd7CDZH9+g8XOE+pXaPFKQJIUTCKEgTQoiEUZAmhBAJoyBNCCESRkGaEEIkTDIleEVFRXjzzTdx4sQJuLu7Iy4uDi+//HKHzysIAlSqcqjVKuh0fMcHamVFRTLodDqLP+fmpoS/fwDkcsn8CAkhNiCZ/8IXLFiAoUOH4r///S+KioqQkJCA/v37Y8qUKR06b2lpMTiOQ5cugZDL3cBxnJVGbB1ubjLU1VkWpAVBQFVVBUpLi9GtWw8bjYwQIgWSCNIZGRnIz8/Htm3boFAocNdddyEpKQnu7u4dPrdGU4PAwDvBcc6T2eE4Dt7eflCpysQeCiHExiQRubKysjBw4ECsX78eY8aMwYQJE5CWlobu3btb4eyCUwVoPan9i4AQYhuSmEmXl5fjzJkziIiIQHp6OnJzczF37lwEBAR0ON1BCCGOTBJBWqlUwsfHBy+++CIAYPDgwZgxYwbS0tIsCtJdu/o0OVZUJIObm7Rn0sbj+/33HKxatQKXL/+OXr3uxLJlyzFkyNBmPyeTyVpd8y8ljjLOjnCFewRc4z6ldI+SCNL9+vWDWq2GRqOBUqkEAPC85ZUYzTVY0ul0Fj+YsyfjB4darRaLFyciLu4prF//Hxw5ko6FCxdg16598PZu+gtIp9OhuLjS3kO2WECAr0OMsyNc4R4B57lPPj0NuuQkyOITII+KNnlNjHuUfIOl+++/H126dMGqVaug0WiQnZ2Nb775Bo888ojYQ7Orc+fOoK6uDnFx8XBzc8OECZPQt28/pKeniT00QpyKLjkJQl4edMlJYg+lTZII0u7u7ti6dSvy8/MxZswYzJ07F3PnzsWkSZPEHppdXbuWi6CgvibHgoL6IDf3skgjIsQ5yeITwAUFQRafIPZQ2iSJdAcA3HXXXfjss8/EHoaoqqur4eFhWnbo4eGBmpoakUZEiHOSR0U3SXNIlSRm0o6AT0+Dds4s8DZMPXh6eqK2ttbkWE1NDTw9vWx2TUKItFGQNpM9clh9+vTD9et5Jsfy8q6hT5++LXyCECIVtprIUZA2kz1yWMOHhwMAvv76K9TV1eH777/DlSuX8cAD4212TUKIddhqIieZnLTU2SOHpVAosHr1R/jnP1fi888/RY8ePbBy5Wr4+/vb9LqEkI6TxScYyvqsiYK0xPTrNwCffLJR7GEQQixkq4kcpTsIIUTCKEgTQoiEUZAmhBAJoyBNCCESRkGaEEIkjII0IYRIGAVpQgiRMArShBAiYRSkCSFEwihIS9SFC1mIiXGMVoqEENuhZeESIwgCUlP3YMOGtWIPhRAiATSTlpiNGz/F7t3fYNasOWIPhRAiARSkJebRR2OxadNWDB58j9hDIYRIAAVpMx2/UoKlKVk4fqXEptfp1i3ApucnhDgWCtJm2ptRgBtlauzNKBB7KIQQF0JB2kxTQ3ugV2dPTA3tIfZQCCEuhKo7zDSqf1eM6t9V7GEQQlwMzaQJIUTCKEhL1PDh4fj22yNiD4MQIjIK0oQQImGSCtLffPMNhg4dirCwMMOf3bt3iz0sQggRjaQeHF64cAH/8z//g0WLFok9FEIIkQRJzaTPnz+Pe+6hlXaEEKInmSDN8zyys7OxZ88ejB49GtHR0fjss88gCILYQyOEENFIJt1x+/ZtBAcH49FHH8X69etx5coVLFiwAN7e3pg5c6ZZ5+ja1afJsaIiGdzcJPO7qFntHZ9MJkNAgK+VR2MbjjLOjnCFewRc4z6ldI+cIOGp6saNG/Hjjz9iy5YtZr2/pEQFnc70dm7dysMddwTZYnhW4eYmQ12drl2flfq96QUE+KK4uFLsYdiUK9wj4Br3KcY9tvZLQTJTzN9//x3r1q0zOabVauHu7i7SiAghRHySCdJ+fn744osvsGPHDuh0OmRlZSEpKQnTp08Xe2iEECIayeSkAwMD8fHHH+Of//wnVq5cCX9/fyxYsAAPPfSQ2EOzq1OnfsEnn6zHH3/kw9/fH089lYBHH40Ve1iEEJFIJkgDwKhRo5CSkiL2MERTWHgLy5YtwbJlb2LMmLHIzr6IV155ET169ERk5Cixh0cIEYFk0h0EuHWrANHRkzB27HjIZDLcc89QhIWNwG+/ZYg9NEKISCQ1k3Z1oaFhCA0NM3xfUVGOzMxzmDTpYRFHRQgRE82kzXSm5CRW/fYWzpSctMv1VCoVlix5BUOGBGPMmLF2uSYhRHooSJvp+5sHUaguwPc3D9r8Wtev52H+/Nno0qUL3n57FWQy+jER4qrov34zTeg5GYGePTCh52SbXufXX89i3rzZGDNmHN55532qEyfExVFO2kwjukZgRNcIm17jxo0/8OqriZg37znMmPGkTa9FCHEMFKQlJCVlB6qrq/Dppxvw6acbDMcfe+xxLFiwUMSREULEQkFaQl588RW8+OIrYg+DECIhlJMmhBAJoyBNCHE5fHoatHNmgU9PE3sobaIgTQhxObrkJAh5edAlJ4k9lDZRkCaEuBxZfAK4oCDI4hPEHkqbXOLBoSAI4DhO7GFYlYT3aiBE8uRR0ZBHRYs9DLM4/UxaLneDVqsRexhWx/N1kMnkYg+DEJtypNyxrTh9kPbx6YyysmJoNLVOM/sUBB0qK0vh6dl0T0dCxGTtoOpIuWNbcfp0h6enNwCgvPxP8HydyKNpSiaTQaezdI9DDkqlB3x8OtlkTIS0l3FQtUY6QRafAF1ykkPkjm3F6YM0wAK1PlhLjSts7Elchyw+AfyGdRBKS8Gnp3U4UDtS7thWnD7dQQixH3lUNDh/f6CiwqVTFNZEQZoQYlWOVN7mCFwi3UEIsR9KUVgXzaQJIUTCKEgTQoiEUZAmhBAJoyBNCCESRkGaEEIkTHJBuqKiAuPGjUNKSorYQyGEENFJLkgvX74chYWFYg+DENJB1BzJOiQVpHfv3g2VSoWBAweKPRRCSAdRcyTrkEyQzs/Px/r16/Huu++KPRRCnJ49Zrm08tA6JLHikOd5LF68GEuWLEFAQEC7z9O1q2O27gwI8BV7CDZH9ygtxTuTIeRfh3znNgQ8Od2iz5p9n09OZ38ckJR+lpII0h9//DH69u2LiRMndug8JSUq6HSO1TPaFbrg0T1KD/94PITkJPCPP2XRuB3tPttDjHts7ZcCJ0igE/5DDz2EoqIiyGQs+1JVVQWlUonHHnsMb775ptnnoSAtTXSPLL2g74vsyH0t6Gdpu2u2RBIz6W+//dbk+2nTpuHpp5/G9OmO+U8lQhqzdjN84jok8+CQEGdGD9FIe0liJt3Ynj17xB4CIVZF7TtJe9FMmhBiFj49DcWxM2hxip1RkCaEmGiphlqXnIS63Ku0OMXOKEgTQky0tFJQFp8At379KK9uZxSkCSEmWnrIKY+KRsCunZRbtzNJPjgkhIiHHnJKC82kCSEuQ7tmNTSjI6Bds1rsoZiNgjQhxGUIu3YAVVXsq4OgIE0IcRlcbBzg7c2+OgjKSRNCXIYicRGQuEjsYViEZtKEECJhFKQJIUTCKEgTQoiEUZAmhBAJoyBNCLE62inceihIE9IGCjiWo53CrYeCNCFtoIBjOdrkwHqoTpqQNsjiEwz7ExLzUP8P66GZNCEt0Kc5AECx8UsKOkQUFKQJaQGlOYgUUJAmDuX4lRIsTcnC8SslNr+W1PKqUnyAKcUxORsK0sSh7M0owI0yNfZmFNj8WvKoaEmlOaQ4s5fimJwNBWniUKaG9kCvzp6YGtpD7KHYndRm9oA0x+RsOEEQBLEHYS0lJSrodI51OwEBvigurhR7GDZF9+g8XOE+xbjHgADfFl+jmTQhxGzqgwcpB21nFKQJIWZTfb7R4hw0PVzsGEkF6cOHD2PKlCkICwvDhAkTsH37drGHRIjLaC2Y6l9TDA+zOAdNDxc7RjJBuqioCAsXLsSiRYtw7tw5fPTRR3j33Xdx/vx5sYdGiEtoLZjqX9Oe/dXiipeWHi7SDNs8kgnS3bt3x/HjxzF27FjodDqUlZVBLpfD29tb7KER4nSaC5CtVWroX/OZ+4zF12qplJFm2OaRTJAGAB8fH6jVatx777145plnMHPmTPTp00fsYRHidJoLkK3Vhetf85w82WozYEvK97RrVkMzOgLaNas7dE1HJLkSPJ1Oh7q6OmRnZ2PevHl45ZVX8Pjjj4s9LEKcivrgQag+3wSfuc/Ac/Jkiz5bHDsDdblX4davHwJ27bTRCE3dHDwEgkoFzscHPS9dsMs1pUJyQdrYmjVrkJ2djX//+99mvZ/qpKWJ7tF5BAT44tb2FENXQHutxtSuWQ1h1w5wsXFsx28bojrpFpw8eRLTp083OabRaODn5yfSiAghzRFjubwicRGUx07aPEBLkWSC9D333IPCwkJ88cUX4HkeZ8+exa5duzBjxgyxh0YIIaKRTJD29fXFZ599hkOHDiEiIgJvvPEG3nnnHURERIg9NEKcFpXBSZ+kdmYZOnQotm3bJvYwCHEZxlUeUun2R0xJZiZNCLE/c8rgaLYtLgrShLiwlh4CGgdmWnQiLgrShLgIS2bExoG5vT2jaQZuHRSkCXERlsyIjQNze0vuaAZuHZJ6cEgIsaHgECD7EvvaBnlUdIcfJMriEwwzcdJ+NJMmxFVkZQIenuxrI+1NTbTWU0Nqe0Q6KgrShLgI4xRG46Dc3tSEsGsHUFXFvhKboCBNiIswntk2DsrtfTjIxcYB3t7sK7EJSTdYshQ1WJImukfp0ZfWmdskSf9+/+fmQRU+us1zArB7EyZrkVqDJXpwSIgDOX6lBHszChDDFSEidXO7g6ClDwb1M2/V55uAFoJ049k5rWS0Dkp3EOJA9mYU4EaZGvsu3bZreZs5O7MYp0zMSZ+4ciN/S1CQJsSBTA3tgV6dPTFlcJd25ZDbQ7tmNfjly4DgkFY3CDDOeTeu7GguINNDR/NQTlpkjpbLbA+6R8emGR0BVFUB3t7olX2xXfdpfA7lsZMA7NvI3xJSy0nTTJoQJ2Pt5djGFRzqgwfbde7mqkBcuZG/JShIE+JkbLUcW8i7htLEv0HIyTGc29xfCBSQ24+CNCFOpr01zy3R547x4xEIHAcIOpMyO+rPYVsUpAlxMvqHdgCgmT4FmulTmp3pNp4FtzQr1qcq8MA4KIcMgfytFZBHRYNPT4NQWgr4+VF/DhuiB4cic+YHTnp0j+LQzpkFIeNXAAAXOswQuE1ez8sDFxQExcYvoXlgFFBRDvh1gvLH482e0/g+G3/eWTjUg0O1Wm32H0KItMjiE4DeQexPcEiTWXKTtEhlhelXM85vrzJAV9bqTHrYsGGora0160QXL1602qDai2bS0kT3KD5zZr2al18AfjzC0hpr1zd5nU9Pg3xnMvjH41tdRWjpknOpkdpMutVl4fv27cOzzz4LLy8v/O///q/VB0YIsY/Wejvrg6o8PgHyZoKzni45CUL+dQhtLPWmzW2tq9Ugfdddd2HTpk2IjY3FzZs38eijj9ppWIQQa2qtV0fjCg1+wzr2mecXmnxGFp8A+c5t4B9/qtVrUbN/62qzuqNnz55YtmwZfvnlF3uMhxBiZW3VMutzywgOYcu/r10FrrOgzaenGSpEACBg1842Z8fU7N+6zCrBi4mJwXvvvWfrsRDi8myxeWtbtcyGkr2sTICTAUol0DvIMCPGtatA7hXDDJvYV4frpMvLy7F169YOD+Tnn3/G9OnTMXz4cERHR2P79u0dPichjsYWi0PMrcKQxSeAGzgQ8hWrIH9+IRtDcAgL2jI5UFWF4tgZou3+7aq7j7c7SB87dgyJiYkYM2YM3n333Q4NoqCgAC+++CKee+45nD59Gh988AE+/PBD/PTTTx06LyGOxpyAammwMk4/tPbZ5nZuQVYm5CtWgQsLA7y9UZd71aJfINYMrK66utGiIP3HH39g3bp1GD9+PJ599ln8+OOPiIuLw/79+zs0iBs3biAmJgbR0dGQyWQICQlBREQEzp4926HzEuJozMnndiRYmftZ4zy1/iEgN2YsBLXarN3GrTHWlsbkag8k29yZRaPR4Ntvv8WuXbtw6tQpyGQyREREoLCwEF999RUGDx7c4UGEh4cjPDzc8H1ZWRlOnz6NadOmdfjchDgbWXwC+A3rIJSWstrlZgJ6S7XK5nwWaKgG0ddX64Ms5+kJoZndxlsbq7UqPSzdTcZZtDqTfvPNN3H//ffjjTfegLe3N1asWIGff/4ZmzZtAsdxcHOz/u5blZWVeO655xAaGoqoqCirn58QRyePigbn7w9UVLQ4Q21pBmvOZ4013m3FrV8/iwIuVXpYgdCKQYMGCZMmTRJ2794t3L592+S1IUOGCL///ntrH7dYbm6u8NBDDwkvvPCCUFNTY9VzE+JMqg8cEIqmzxCqDxww63X2faxQfeBAm59tSek77wh/9B8g3BjxF4s/S9qv1WXhJ0+exL59+/Ddd9+hqqoKw4cPx8SJExEdHY2oqCjs2bMHAwYMsMovi1OnTmHBggV48skn8corr4DjOIvPQcvCpYnuUXxtLQs3Zym3ZnQEoFIBALgR4U7VVMmY1JaFt5ruiIiIwNtvv41jx47hww8/RKdOnfD+++9j/Pjx0Ol0OHToEFT1P7SOuH79OubPn4+FCxfib3/7W7sCNCGkZS09dNMvVuGXLTFp5q9/zbgyg4uNAzw9ge6BkMUn4PiVEixNycLxKyV2vRdXY3Gr0oqKChw4cAD79u3D2bNn4eHhgYcffhgrVqxo9yBWrlyJzZs3w8vLy+R4fHw8Fi9ebPZ5aCYtTXSP0qOfOQulpcD1PECnA7y8DL2igeZn38b3uTQlCzfK1OjV2RMrpweLdi/WJrWZdIf6Sd+8eRN79uxBampqh8vwrIGCtDTRPUqPZvoUFpy7dAV8fAA07dXRXArE+D6PXynB3owCTA3tgVH9u9r/JmzEIYN0WVkZ/Pz8IJM1ZEfOnz+Pnj17wt/f3zqjtAIK0tJE9yg9hiDdOwjKlH1mf87R7rM9pBak21zMsmnTJowfPx4ZGRkmx1etWoWxY8daZUk4IcR2mlv1J39+IdsMoP51V11y7QhaLXROTU3Fhx9+iJdeegmDBg0yee3jjz/Gtm3bsGrVKgQEBGDSpEk2HSghpHX69ASCQyD8dBQADD04mu3vXFQIcDLDw0IhLw/8hnUO3bDfGbUapDdv3ozFixfj6aefbvKaj48Pnn32WdTV1WHTpk0UpAkRmaHfRvYloKbGcMx41Z/JA0Ojnb91mRnsc3I5hPqFLo2DNJ+ehmIzdmYh1tVquiM3Nxfjx49v9QQTJ07E5cuXrTooQojl9GV2XGycYW9D/Yy4SeMkANzAQQ3VHFmZgIcnIAhATfP9OXTJSdBevAR++TKz0iKUQrGOVoO0p6cnqqurWz0Bz/Nwd3e36qAIsTdnCCj6YKxIXMQeBvYOAr84ke1dWE8fyOXPLzRZrm1oqFRbC6hUEL490OT8svgEVqrHcWYtKXfVrnXW1mqQDg8Px549e1o9we7duzFkyBCrDooQe3PKgPLjERZUfzxiONRcLw3jXDbKy9jBstImp5NHRcN/zQfgBg4yq3+Hq3ats7ZWg/T8+fOxbds2fPzxx01m1FVVVVi/fj22bt2K+fPn23SQhNiaIwaUNmf/D4wz/FW7ZnWL59H/ghJ27WArCmUycE/ObPZ6qs83mv1QkZorWUebddJHjx7F0qVLUVlZib59+8LX1xcVFRW4evUqunTpgjfeeAMTJkyw13hbRXXS0kT32HHNLSxpqx8HUN9vo6oKUCrBBd/bbIA1mUlnZRq+Nn6vds4scPnXIdzV22n7dgDSq5Nus9fo2LFj8f333+Pw4cO4ePEiKioq4O/vj5CQEIwePZry0YTYQeMyOj49jVVo+Pm1OvvnYuPYDNnbp/kyPDTt02zcQ7o9u4W3xJwmTqSpVtMdarUar7/+OsaNG4eVK1eisrISr776KhITExEVFUUBmhA7aZyO0SUnARUV4Pz9Ww14isRFUB47CfmrSwE/Pwj5+ayhUisLWFpK/cijohGwaycAWPSQ1XCdDeucL+9vB63OpNetW4fDhw9j7ty5kMlk+Oqrr1BeXo61a9faaXiEEKDpbLfxjictzVKNjwNgC1iKCsEvTmQbzCoU0CUnQZeZAWHXDnCxcVAkLjJ7+y5zZsSGsj8/P4fL+0tBqzPp7777DqtXr8a8efMwd+5cfPTRR0hPT4dWq7XX+AhxCOqDB0Ut4WupOqXFqhWdji140Wohi09gKZGqKva1DZY+ZG2p7I+Yp9UgXVhYiH79+hm+v/feeyEIAkpKqH8sIcZUn2+06z/lGwfflgKnLD6BpTlKS8GNGQsE3gHoG6VxHODXiZ0jPAJwcwNqalqtBAEsr9qgKo+OaTVI8zwPuVxu+J7jOCgUCppJE9KIz9w5dv2nfOOg3Gr9MwBUVABZmSw37eXFVhd6ewM+7IEiV1kBuLsDPG/WbNr4Gvp/QTjDgiApsv5OsoS4IM/Jk6EKH93sa7aoatDnqI9fKcHelKxmezob54L1s2l+wzqAkwFugPytFYb36ft36PPS5mo8o7ckV03M02aQTklJMdkxhed57Nmzp0kf6Zkzmxa/E0Isf9Bmib0ZBbhRpsbejIImQVqoqQH+LAYCA8F5eEDIyQG0GqBnL7ZkfPky9qCwvuZZHhUNJC6y6PqNH2CaPKQkVtHqYpYHH3zQvJNwHNLT0602qPaixSzS5Aj32NHZbmv3aMv64JZ2R9GuWQ1hyyb2jUwG+T/XgF++DKjjATc5UFfH+nR4e0N57KTZY3WEn2VHOdRilh9++MHqgyFEKowDki1nu43L59qruQA6qn9XQ3A2tCH19QOOGP23q18e3j0QuPEHwLkBfp2AinLAy5t9rj7VAS9vQKejlIWEtLkzCyHOyjgw27p3hzUeqrXVBMqQgzZqqIRxD4KrrGC56IoKoNedrEXpktfADQ02BGR9CR4qyqmWWWIoSBOXZRyYbV0mZo0ue239IjG0G31gHODjA+7pZ8BVVjT0j25Uq2x8Pi42DvD2BvdEfIv/O/DpaSiOnUHVG3bWod3CpYZy0tJE92jdvLQ+Dx3DFSEidXOr5zRJ6WRmQPg6GfDrBG7yI4YmSgDMGhs1WLLtNVtCQVpkFMCcgy3vsXGAX5qShRtlavQszMM/0v/Vahc8Y5qRww3basHHB/DwZDNvwKSbXmtLzPUNlpw5Xy21IE110oRInD5Vot8kNiZmNlI7d0eMfxdwl9vOHxsWtbi7syDt5sbSG41m0ggOYR3wSkuBZvY5lEdFI+DJ6U7/C1dqKCdNiMQZcs0AhJxs/OWD/8U7nQpw/2MPGipTWssT8xvWQcj4FXD3ADciHNzMWc2mOpCVaZK/poeH0iC5mXRmZibmz5+P48ePiz0UQjrMGrlofQkfn57Gap05WUNf6Q3rgOt54PPz2Yy7pga4eAG4Zwg4D4+GQMvzQHERUJ+Lbm6VoPHCFGdOZzgaycykBUHAzp078cwzz1BvECIJ9iibs4Q8Khryt1aAGzjQdJar0wHFRWxF4fks9v35rIa67+cXsl3ABQHClk0skN8ugeDr12qFC/XikAbJBOl169Zh27ZteO6558QeCiEA7FM2Z6nGgVT+/EJAoWAvamqBocGsy93Q4Javqw/kRw+D37DOZOZsHJh1yUkQMjPA/+0lkx3HiX1JJt3x5JNP4qWXXsKJEyfEHgohAFiA5TesY42J0tPalQKwdLVhu9IjHMeaJvW6E8qvmnaw00yfwt7TXCHXddNVlo0X+PBnTrP3GS+QIXYlmSAdGBjY4XN07epjhZHYX2vlN87CIe/xyeko3pmMutyrkO/choAnp7f69vbco/rgQag+3wifuXPgOXkyincmQ8i/3uz1ylasQHXSV/BKmInOy5YBAIp3JoNXKsEpFPB/bQlw+hhUn2+EYngYtGfPwWfuHJS7ycDrA7RMBri7Qz5oIPjsbMg6+6Pzc/MMn/OKDIdW4Qafuc+w8aTth+bQ91BOnGC4P4f8WVpISvcomSBtDVQnLU2OfI/84/EQkpPAP/5Uq/fQ3nvUfvIZhLw8lH7yGVTho1u9nubLrUBVFaq+3ArtvIXs84OGAlnngdg4qMJHGzaR1Zw8Beh0uH3yFLiE2cD1/PoaaQ7o0RN81nmgrg46X53J57RaHoqNX0IFQFVcCby3Fsr32PWLiysd+mdpLqnVSUsmJ02IFNl6ubg5zfv1DEu36/s98+lprOdGfaN+Pj2toVxPp2Mf0ukg/HQU3BPxbOcVuQz480/WBQ9gTZaaGQeRDgrShLTC1hUOlvwSkIWEAt0DIfx0tGGBCsexlqMcZ8gtKzZ+yR4g6lVVQdj+FQvMPM8eMIIDOI4FbwvHQeyLgjRxCe0NtvqFIPyGdTYamfl0yUnA9TzDwz5ZfAK4gYNYQyVBAIJDDO/lPDwAXz+2/NvbG9DWz5yVSnBPxIMbMQLy1WuhsLDJP7E/yQXpyMhInD59WuxhECdjzXplWzDrl0hwCCCXA94+bOk2AMXGL9n+hB6eEH46Cs30KayaIzgE3MCBkL+1gpXp9ekD9OsP+YpVUCQuMsyatWtWQzM6os3NZ9s1XmIVkgvShNhCe3Ou8ucXggsdxgKdDZn1SyQrk5XalZcBN2822SkcAJB7Bci9AmHHNkMglkdFs/vw94cuM8MkuArbvwJUKvbV2uMlVuFU1R2EtKS9u6MY1w8bf28thtxycAg4oNVfIrL4BPCLXmbf1KiB4BDT3VjOnml4s1oNzcw4IO+aoZmSkJkJnDkNuCnAl5ay63p4sJy2TmdRLXjjvQ2J7VCQJqQNttxaS39uDjC0G+XT08CvepdVXoy8j+WhUb+6MKA7UFQIgGMPEHftYLPrKlXTxSrnswAAwtYt4P7f0yxAA0CdFrh5A0JxMRAQgBN3DEHqoLGISf0JY8y8P2ttCUbaRukOQtpgy/K05s7Nb1jHAnFNDVvpZ/SwkJv8CGs5ql/8VVfHAnSvXi1fhOdZs3+5nH3v14ktJRd0kD+/EPsnzkaBfw/sD6agK0U0kyakDbacNTZ7bpWq4e8PjDPMpPXL1MHzprNmd3fgxg3A15cdr60FGjcp02gALy/I31ph2HSWi42DPCoa04x2HNfTrllteA9VgIiLgjQhFrLmVljNBkMfH+B2CdA7CMq1602uaViIUnqb/V3Hs6XeSndAEMDFxrEFLsazZv05ewexVqde3oCHJ3sQCdMdx/XjQW0tUFfH/k5BWlSU7iDEQtaobNCXsAlfJ7PFJrsaGiM1V1HCv78SwpnTLLXh5gb4d2EBGmCrC93kQFAfdp6gPmxG3akz5Eteg/LQYShT9gGnT7a5I7hh13BBMFndSMRDQZoQC3U0R61v3i/k5LCZbqNgqN/J22THlbLS+g/zQO8gyJe8Bnh5NZy0poY9KKyqYk3/tXVA4S2TRTjm7AhueM/MWVAeO0mpDgmgdAchFupojtqwnFvQQb7kNcOuK9o5s4DgEAg/HQVu3gAUSkNFCffkTAhJmwGlOzh//4b+z++vBApvATyPE/eMRurdoxFTdxOR6V8DWtNqD0XiojZTF+a8h9gXBWlC7Ky5bar0KRRkX2KzYp0OcHMzzNYViYvAh4SabBgri09g6Q8A4Djsn5CAAs4DBzoPx31RIwzXMOSZ/f3ZA0ajrbWojE76KEgTYmfNzcT1gdswkwbLTTd+n5Cfb6h35s9nAf0HABfOAz6+mNJFi1TBH1NDe0DeP9gwQxe+/ILlmPVVI+ezIPj6QZecxCo9tn8FdPaH/NWlFLQliHLShEiAvgudLCSUpTOMArShT4a+flpPowXyrrFKjuoqRKRuxsrpwYZKDaC+5rqZHVlOdB2Av4c+hV9+ucgqOQpvtfkglE9PQ3HsDOrXYWc0kyZEQhpXjuiSk1gzpYoKwM8P6B7Iyu88vYBu3cCNGWuYecviE0zKA3WZGcDVXAAcoHBjn1FXA4KA1JCJKNDKkHrPeEReOQV07dbmg1BdchKE/OsQbLDykrSMZtLEYTlLJzbj+zCuHDHkqauqWK+O3kGs3vmu3pAv/wcr0cvKBDdmLDh/fwCmQV7YtYPNomUc5O+tZiV7dXWAmxtiKi+jR/VtxFz5GVzoMCi/+6HNwCuLT4Bbv37Ur8POOEFobndKx0TbZ0mTre5Rv+UTFxRk6Hshlo7cY0v3YWiepJ9JqyrZQ0WOYz2iuwey4zVqwMPTENwNZXe9g4DTJw0LZRovnOHT0wzvbS7/be37dBS0fRYhVmKrnhr2nqG3dB+G3VKeX8hakfp1YqsLBQGormYrDmvUQHgEez04hAXd63nAtWvA9TyTWufGO7sAYDnu4iJqOSphlJMmDstWPTVs2fWuuSXlbd2HPCqa5ZfPZwHdAthMuuRPtg2Wjy+4ygooNn7Jmv3nXmnyeUMJnrcP+xwaWq+CkwGCjlIYEkYzaUIa6cgMva1ZeEtLytv6nPB1Mkt1FBWyvh76/Qr9/EzHKZOzDnf9+oMbM5alUuob+6PwFtC1G9Cb3Rvbfqt+9xZ6EChZNJMmpJGOzNDbmoW31Cxfl5wE4fx58IteBh/Q3bAS0UDpzoI00NDhzqjTnXbOLHBjxgL+/mwDgNMnIRzczxbFdPYHiosAd3dwd95pkvem4Cx9FKQJsaK2dixp6ReALD4B/OJElm8uKmTd6gBDW1HwOvZGjjOte66qYu/lZIaNAzSjI1hFSF0duKHBhrE0Hpc1u/kR26EgTVySrQJUR7bp0iXMhpCcxGbINTUs+Gq1rBc0wB4a6nRsF/CAAHD+/mwFYnU1oFQaloCjtpYFcp4HgkNMct96+iZP4GT4b+pPOFDeA1NDe5gshCHSQDlp4pKkuJGqInERuJBQFoR1OvZQr7M/W1HIcfX9PBTsgSHq90P08WHBu2cvyKOi2azbKB1i3ALVmHGTp/3B0bhRpsbejAJ73SqxAAVp4pJsuSVWS8wp7dM/zMMD4wBBB+6hh9lmsbr6dIdbfcAuLmZ57/re0/qHhAiPYK8DAMe13A86OMSwScC0cUPQq7Onyc4sRDoo3UFckhgbqZpT2qcfl3bOLAj1u6dwsXGsusOvE5s5FxcZyub079dMn8Lqo7t0ZTPvujpALme10c3JyjTszjIqsSulOSSMZtKEmKkji1z49DS2crBxyRxYHbNmdATLJ9cznukrEhdB+ctZKA8dZjPngYOA8Ajwy5eZfAYA22FcP+vW6cBvWNfsmNv7L4kjF25haUoWjl8psehzpP0kE6QvXbqEJ554AsOGDcOUKVOQmZlp82vO+flJwx9CGmsclDuSx9YlJwHFxaZd7OoJXycDKhWEr5Mbmv8DLe6eIpSWAkcPAyoVfvn5NyxZ+Q1OPPAoS3s8Ec82pgUMX5sbs2E1o4X/mvj6l+uUv7YzSQRpjUaDBQsWYPLkyTh16hT++te/Ys6cOVAZ75pMiJ01DsodyWPL4hMAQQdwXNMgr3Q3fNVfs6UZsC45Cbh2lVVvcBxSgyeiwM0H+9Wd2DWyMhvqqWtrDUvKrZV7f2Jkb8pf25kkctInT56EVqvF7NmzAQCPPPIItm7digMHDiAujjbCJOJoXPPckTy28Q4sTQJmt25AdRXAAULWbywA/1kMQaEAX1rK+nGoVIYdv1mdNAf07YepoT2wN6sQj+SfAn/qd5av1j84fGBcO++8ZeOG3IGhAd5WPy9pmSSC9OXLl9G/f3+TY/369UNOTo5Nr7vx/u02PT9xbNZ6uGhck91ctz758wvrVxxmsZpo/WIVfSndtassz/znnyxd4uPL9kd8fiEiMs/iLyn1O690D6zf5duH1VFXVoBf9S5wuwR8aSktWHFQkgjS1dXV8PDwMDnm6ekJtVpt0Xm6dvWx5rDsprU2hc7Cle+xeGcydJd/h+7Nv6NTJ094Tp5s+oYnpwNPTkfZihWo+mILW4Si1UI5KRo+M2bg9osvsRSGuxLes/4f1P+3B7rbpfC4chHVKTsbgnpRIWthWqeBrKYayK+EUFkBcBzkbnKr/QwUn61DddJX8EqYic7LllnlnFIjpf+/SiJIe3l5oba21uSYWq2Gl/GW9WagftLS5Or3yD8eD2H5MoDjUPrJZ1CFj254zXjl47yF4E6cNvSWxntroQIgf+e9hk1lo6Jx/NBZpIb9P8QcOoqR0x8H6psvnQgKQ2rwBMRc/AGRKAPn7w8uOATIyoQuOAQ3pz7W4RWWAQG+qPpyK1BVhaovt0I7b2G7zyVV1E+6Gf3798fVq1dNjuXm5mLAgAEijYgQ65FHRbNFJYLAFpEYMefhpL4SA2CNlFKHPIgCv0CkBk+AInER5CtWAd0DkXpvNAo634HUkElAVRWE0lLIQkLZZ7MyrbbCkouNA7y9W14oQ6xKEkE6MjISgiBg8+bN0Gq12L9/P7KzsxEdTTk04iSMFo8YaxyUWyuN0wf0mGsn0KPqT8T8mQXNyOHgl/wNUKsRU5WLXv3vQkx5Dusbfb3typT21H4rEheZbCZAbEsy22fl5ORg+fLluHTpEu6880689tprGDVqlEXnoHSHNNE9tt3QqbnXDc36wyPAVVawWXhWZsN2WiV/NuSjfXzYzDYrEwgOYZvT6ktYfXxa3B7L0i3I6Gdpu2u2RDJB2hooSEuTK92jJd31tGtWs653gsCWc+t0JsFSM3J4Q80zx7HFKf0HABfOsyZMAFBRjhN9hiN12GTE/HoQkdfOsrc//QwL6Bm/su9DhzUbhC3tBuhKP0t7X7Mlkkh3EOIsjHPMraUS+PQ0CEmbWZldXR1QUd40HeHXqeHvggDU1LIttASBvf+xWHAjwpE6bDIKPP2ROjSq4e27drBz9Q5if4JDmh1Le1ceEvuhIE2IFRnnfltbRq5LTmpYaejmBu6J+CbBkpv8COBmXIBl/K9EDsjKhGLjl4i5eQY9KgoRk/U9e0mhABcbB3lUNJQp+6BM2cdm1efPg1+c2LTfB5E0SZTgEeIs9AtgWmuoBJiuZmzcjF+XnATh2lWgpLUmRgKEokJohgcjslcvRGYcBjr7Q/7BR63v/KLTsTw3PfRzGDSTJsQGdMlJQEUFOH//ZoNmS2kG/ey79QANVimSn89WIubns97Rd95pcj7j7nryqGhwCbMbHjASh0FBmhArMnSxCw5ps7FRczlrfboEXev7O7u7s51Xxj3I/nAc4OkJ9OwJDA1uOJlGA8HXz6TlqbBrB6uXrt+dhUrnHBMFaUI6iE9PQ3HsDLZv4IZ1EDJ+hfDT0TYfyDXOWWvXrAa/fBkEXz9wffpC/sFHQK87AXDAL/8FF9QHUCrZHoYAzvzjE7z+zAc4MWgUmyWfPslanm7ZBM2kB1nzf6WSZs4OjoI0IWZqqVpDl5yEutyrFq/ma7zARD/zxZEfIJw7B/61V1lTJR0P1NRA2P4Va8Ck0wFVVdibUYACn27YP3E2q4vWN/sHgMJbQHk50OtOmjk7OArShJippWoNWXwC3Pr1Yw8B6/cclD/ffE8LPj0NmulToBl7H/glf4Nw+TL4Ve+y7a/CI1iDpE6dWWCurWU7reh19mczaXBA4S08vOcT9BRq8EhWGts6q7ra5Fon+ozA62HxtIuKg6PqDkLM1Li/tJ48KhoBT043LIBoLcXBb1gH5F5pOFBexr7W79gif2sFe4/+uPF1Xl0KXWYGhC2bAACRGYcRWXABKL3NZtGdOrMaak0tMPI+pAaOQ0HnO7A3owCj+ne1eOEKkQaaSRNipsaNjtqz12ET+gb9AFDyJ+v/nHsFUChNX6sn/HS04XinTqzJv1bLVisOGADu7rsBH19wlRV4NG48egV2NuyiYun2Xx3Z05FYDwVpQizU3r0O+fQ0lmM21j2wIegapze0Rs3/AUAmA/+3l4DcKzjRexhen7IEJwaNAtw92Oe7B0IWn8Bm+X5+EEpLEXHtLFZODzbsBG7p9l8d2dORWA8FaUIs1FKwa27Xb2Osdrrc9GBZKTB2POBRH2z1vTr03N1ZHw4jqcETUODbHak9h7N9Ez09gfIy6DIzWD20vz9QXAR++TKTWbClS8A7sqcjsR4K0oRYqKVg17guuTFZfAKgUDQccHMDlO6sw13PXkBzOxHxPITTJ1k5Xb2YrO/RQ1WMmJtnWaqjuhqorYWw/StDjTbbqFbWoVkw9fWQBgrShHSAPm+rPniwxWb4htzuvj1sYUrgHZB/8BG4mbPYTLjRRgAm6upwQqXA6xNfxom+IwCOQ2RZLt7xL0LktXOmZXed/dlqxaxMyN9aAW7gQJoFOwFqVSoyav3o2PT9mJUD+gP/3tTqe3C7pH6jWG8oj51kx89nsRSHXM42kPX0ZDXORl5/ZDHyutwJgZPhhUv7EemlAY78YHqRfv0NG9rasnrDmX+WetSqlBAJM7ei4efdP2DJym/wy71jwQUFwWfuMy2+17DU+4FxLMVRUwPtmtVslqvR1F+YB1SVrE5aJjf5fMz5dAjgwOl4pAbc2zRAe3iAGzOW0hNOioI0IUbMrWjYe+k2Ctx8sF/dCYqNX5rsAK4P9JqXX4BmdARLcwBsWTfPA3V1EHbtYMH0gXENJ1W6sxWHOt7kWpHXzuKFH79AUOmNhnakegoFK9drtC0XcR60mIUQI40XrOi3sOJi40yWV08d3AX7Lt3GlMFdmpxDl5wEIScHqKxgB478AMHXFzh31vAAUJ+35iorIHQLAPz82HZYjVIdepF55xCZd870oI8PqwgRdE1yz8evlGBvRgGmhvYwlOARx0QzaeIy2rM4o3HFhv4cI/14vLd0Bu5/7MEmn5HFJ7AHgnr6lYD6jnYB3SEc3M+WggeHsABdVcV6bTRmXA3SWHgEuIGDwMXGGXaC0dtz5AJu5BVgz5ELZt8rkSYK0sRlmJPKaPwe44oNPj2NdanLyW5yDvXBg4ZfAPKoaMjfWgH068/+3Hkn2xRW6c662pX8yZaBX2eVGKyuubjhZPrFLW5ugJdXyzd0PQ+KjV9C+OkohIxfwW9Yh+NXSrA0JQt352agR3kh6+tBHBqlO4jLaKn3RmvvUSQuAh8SymaqPx0FOFmz6YXy9/8JIfcq+NJSw+4s+gd4muH1fZ8rK8A9Op1tPiuTAV26Gs7Dn88CNALg49uw4EWnM7QlbTpQebOH/2//SRTUArrOvfDOpd1UgucEKEgTl2EcOC15j2G3FD8/ICCg2c817bIBaF5+AfjxCODvD5SWsoeEWZns4aFOB1RWsC2tHhgH7ol4llLx8GC5bEFg7xl5X9NqDoA9XPT0bEhx9A6C/PmFiPlsO1IHjEbM5WNQfN10d3DieCjdQUgb9CV08ucXstRERUWTdIffq4ubtij98QgLtCUlkP9zDZRr17OZrVLJZsJqNXv9yA/1jZNkrKOd0dIFrrKC7cgia+Y/1Qvnm2zTNXLkPXj76CcYOfIeG/2vQeyNgjQhFjDuZ2H8ILL27FkI2Zegy8wAUN9Mybd+gYJCYQjq8qhoyFesAhcW1rD9lUwO3LwB1KjZMm89d3e2MGXKNHBhw023ywIADw8W9OVyCBm/QjPpQchCQk22yLJ1JzvqlGd7tOJQZLSCS/r0Kwa5oCBDq9LGx5GTDUGlYg2RhgazncIrKgA/P3D+/oZVgNo1qyF8ncwqPdw92GpDfQ5aLmepEJkM6NMX3JixbIZ94w+2DdbAQRDy8w29p9GvP5Qp+6AZHcEeTALgRoS3OEbj4+3V+Gdp7fNLAa04bMPmzZuxYMECsYdBiEFL3eCMj3slzGRpDK0WwpnTQP51wM8P8ucXQvD1A784EZqXX2B555oaVm5XVGjaFc8oQCtT9rH89fU8QFvHUiDBIaw2unugYRk4UF+B4ubGyvUa9QGxdSc76pRne5KZSVdVVWH9+vX44osv8OCDD+Ljjz+2+Bw0k5YmV7nHm1Mfg3DuHHuoJ5MDffqA8/eHcPZMQ57Zy4t1rfP0akiHcBwry+N5NmN+ciaEg/tZftrLC+jazdCXQ8jJBgQBXGwcm2UDkD+/kO3mcj0P6B3EArwN79MVfpY0k27G/PnzcePGDTzxxBNiD4UQi6kPHmQpjoAAwK8ToFQAf/4JIeNXVrGhV10NyGTghgyB8tBhyJe8xvp1CAL7U1vLAnRRIctPl5eDGzMWACD88QdQWckW13ydDFy7BuReYZsB5F0DHGyCQsxjtxI8jUaD8mZWVHEch27duuGDDz5AYGAg/vWvf6HYuLCfEAeg+nwjq7IICgIAVrKnqp+N+foBPXoC166yag6dDsIffxj2HMT1PDaL1isqZIG+PhUibNkE3YhwNtsGGlYvGtdQ8zxODBqF/RNnY9qVEloK7kTsFqTPnTuHWbNmNTkul8tx4cIFBAYGdvgaXbv6dPgcYmjtnzrOwtnvUT13DlSfb4Ji+DDUfp8OoYs/POJiUfP9D+Dzr7PKDXf3hsb+t0sg37kNnZ6bh4r3V6MuJ8fkfDJvL+iM8tX++vf98QebjZeXsxy4IAB1LGd9IGIKCupk2PvNYURH3WHS9MmanP1nCUjrHu0WpCMjI5GdnW3Ta1BOWppc4h4nT4YqfDSq58yCcLsMXFAQtPMWgv82DaipZX+GDAVystms2csLmqJi3N66Hbh5s8n5dLeMGi117YrycjV0vn6mTf61WshXrwW/fBlQVYWHT+7D/mGT8fCFdJTmVEEVPtr69+kKP0uJ5aRpxSEhVqA+eBDaTz6D4OvHZs3BIax2OPdKw5uyL7FZr6y+1K6iomE1oUzWkJfmuPrl5/UpkNJS9mCwooKlQfT7IHYPhDwqGrrMDPzyy0Wk3j8DU+9yh64oAK8Hx1Paw0lQkCakHfT5ZH39c0V97w64ubH+G1mZ0DXu8cxxLBjreFbXXF/bDIDlrfk6dkwQWIB2UwB1WpbWUKlY8A/qw3LVHAeUl0G7ZjUUiYtwICULBWVqpAqewJg43CxTY29GAQVpJyCZ6g5CHEnjbnkCwFIRGg1QXgbh8u9s4YkRLj4B6NO3+RO6u7MArCeTsYYgMhmbPet3Er9wnn1Vq4GaGkML1amhPdCrsyemhvYw+TtxfJKpk7YGyklLkzPeo6EyIzgEyMqEV2Q4qr7cyvpC6/+TkskBhRtrT9o7CPjlvyzdUVfX9IT9+uOEojtrjpT1PSLzMxryzx4e7JzaOraYRVVZ3+xfAJcw22QzAltzxp9lY1LLSdNMmpB20O8niKxMCHl50J79lfWQdndnb1AoABnX0Ivj9Ek2A24uQIMtSEm9fwYK/AKRGjzB9AFhTQ0rt1MqIF/+D8j/uYbNuvv0hSwk1LY3SkRHQZqQduLT09gCFj8/+Mx9xtA8Cd3ry0l1OuibmOq3ywJg0tHuRFAYXn/0NfxSIUfM1ePoUVvWdB9DvU6dwW9YB37Vu6whU3Fxm3sxEsdHDw4JsZA+1aFvosQFBcFz8mSUvrac5YjV6obFKT4+DY2S6lMU8PBgDwpL/kRq2GQUBPbBvku38I/fjiLydorpLNpY4/0P3dya7ZnR0r6MxDHRTJoQCxk2AQBMmgvp90M0BGiOYzuD79jGSvH0uerqanAPPQz0DkJM7nH0KLyGRy4ebkiN6LfP4jj20NCnmUVabm6Qv7Wi2U0MGu/LSBwbzaQJsZDxFlvGQZKLjWOBMTwCuHiBzXz1Nc2NCFs2AQAicQWRmYcbXtDPtgH2VV3Nvnp6AbU1bBau0bCdWFrYZUY/DpMUC3FYVN0hMnpa7hyM75FPT2PbYrWUttDjONZDWv8wsVNnVivN61hwlslwok8YUu95EDH5pzFSUWmoJmn8C8JeXO1nac9rtoRm0oRYmS45iS1Aqa012QrLhEwGLmE2AEDY/hXQ2R/yV5cCAPhlS1j5noxD6j0PoqBTIFK5vyDy6CfgAKdprk/MQzlpQtqpua2jDBUfPXuBm/U/DfnlxhQKCKdPsrRHbS2gVrMdyTesY+V7Xp5A126IufgDelQWI+bSUZbvbtTUvyXaNauhGR0B7ZrV1rhVIiKaSRPSTiarDp+cDgANzfe7dmv9wV1tLXA+q+H7ygoIOdk4EXgPUifGY1D/HshRy/FIVhre6VIM4dZ5wNuH7dZiBpOHh1Th4dBoJk1IO7W6dVRZKVDHN013uLk1zK4Viobjfn6AVovUIQ+iwKcbvlN54ibngQNj4tiCle6BQEBAq9tUGc/sudg4wNubHh46AQrShLSTftWh8QM8+fMLwYUOA0bdD2hqTT/g4QFu5iwgoHtDgPb1Yw8Py8uB2lrE3PoVPf08MMlHjZ6FeYjhihpm5/XXNGYcmI1n9orERSa7hhPHRekOQjpAHxzVz80DwkezVYdR0WwXbW8foLKi4c1KdygSF0GblQnhdgmr/hB0DXXVgoDIa+cQeS6d1Ux7eoK7PJTtQl5XZ9o1r55xYDYuDSTOg2bShHQAv2EdhIxfUf6+6QM6wdePBWilsuFg/cxaFp/AGi716cv6fdx1V8N7Sm/jRM9gvP7IYpzoPhhCZgarj3Zza3ZRi3HKpbmZPXF8FKQJsQIOwPErJViakoXjV0pYQyWALTwJvIMtQvHrBO2a1dAlJ0H+/MKGXb2NW5oKAlKDJzQ0WtJq2XZZHh6GDWmNWSswN1epQqSBgjQhHaDPQfu9ugh7Mwpwo77ZPhcbxzriBd4B+atLwQ0NZhvQ7toBIScb/PJl0Lz8AvhFL5ue0M0NMefT0UNVjJiLP7DcNc8D1dWs/0cj1gqujftjE+mgnDQhHaDPQXsG+GLqL9ewN6MAU0N7QNYpFLpGKwP1/aeFXTvY9lg/HjGt/ujaFSgpQWTeOURe/xUnZiXidfU0DLzyK3IC+yMm978YA2Dzz9dw8HwhJg8NxEyj4NqR2TTls6WLgjQhVjKqf1fDdlXad02Dp/4Pn54GXj8j7t2b7bQC4ETQcCQPnwpAQPyZPYi8fRmplV4o8PLFtXvGwb1Og9QBozEGwMHzhVBreBw8X4gEKwVX/fiI9FCQJsQGGs9MtWtWQ/jqS1al4eEBbmgwW5kolwMch9TgKBR0Yn2oU4MnAJc9UCH3gE9tFcbeykZO516I6cp6fIT08sPJa6UI6eUHeVSkSXA13nsRQLONoIhjoSBNiA0Yz0z59DQISZsbGi7V1rIAXVXFvvfvgpiqXFS6e0MQBAwsvIL1EU+B44A+tWWYPS0cAAu4fHpvqGp7oJOnAqpavsl1m+y9aIVUCBEXPTgkxMZYwyX3hgNubmxxiiCACx0G+ZLXMGbNP/CRTx4+SnkTOYH9wUGAwHGYFjce8qhok+A7NbQHego1ePinHSx9YvTwUBafAPj5sV8CwSEtr4gkDoNm0oTYkOblF4CzZ9iybwisZ4e+uT9MO9opEhdBc3A/YrK+R2rwBMQU/4ZRiVEATNMno/p3Rfi7/2FBu+IagIYZs2Ljlw2bEmRlUsc8J0AzaUJsSV/BUV4OjLwP8PYGOnVqcXGKfMlriJRX4O2LKbjvickNxxvVQxsvYmncQ6TVniLE4dBMmhBbemAccOQHQKEEV1kBxbGTTR7u6emPy59f2GYOuXE1RuO/Uw7aedBMmhAbUq5dD/kHH4ELCTEE5ZZWCVprQQmtHnQukgnSW7ZsQVRUFEaMGIEZM2bg9OnTYg+JEKswd+m2tdIUtHrQuUgiSB86dAiff/45Pv30U5w6dQpPPfUU5s+fj9u3b4s9NEIcDuWknYskgnRxcTH++te/YsCAAZDJZIiNjYVcLkd2drbYQyPEbvQz4P+m/tTQqKkdqBuec7Hbg0ONRoPy8vImxzmOw8yZM02OnTp1CtXV1bj77rvtNTxCRKcvs9sfHI2b9Y2a9MvMieuyW5A+d+4cZs2a1eS4XC7HhQsXDN/n5OQgMTERL730Erp162av4REiOn1VxrQrJYZGTeYwrhah2bPz4QShpT3n7e/w4cN49dVX8eyzz2LevHliD4cQh1AcOwN1uVfh1q8fAnbtBACoDx6E6vON8Jk7B56TJ7dxBiJlkgnSW7Zswdq1a/Hee+9h0qRJ7TpHSYkKOp0kbsdsAQG+KC6uFHsYNkX3aFvNzaS1c2ZByMsDFxRk1VWH9LO03TVbIonFLAcOHMCaNWuwZcsWhIaGij0cQiSvcWBunOag/tDOQxJB+j//+Q80Gg1mz55tcvzDDz/E+PHjxRkUIRKma6PZP606dB6SCNK7d+8WewiEOBSaKbsOSQRpQohlaKbsOiSxmIUQQkjzKEgTQoiEUZAmhBAJoyBNCCESRkGaEEIkjII0IYRIGAVpQgiRMArShBAiYRSkCSFEwihIE0KIhDnVsnCZjBN7CO3iqOO2BN2j83CF+5TSPUqmnzQhhJCmKN1BCCESRkGaEEIkjII0IYRIGAVpQgiRMArShBAiYRSkCSFEwihIE0KIhFGQJoQQCaMgTQghEkZBWgK2bNmCqKgojBgxAjNmzMDp06fFHpJVXLp0CU888QSGDRuGKVOmIDMzU+whWd3PP/+M6dOnY/jw4YiOjsb27dvFHpLNVFRUYNy4cUhJSRF7KDZRVFSEBQsWYMSIEbjvvvuwdu1asYfECERU3333nTB69Gjh999/F3ieF7755hth+PDhQklJidhD65Da2lph/PjxwhdffCFoNBohNTVVCA8PFyorK8UemtXcvHlTCAsLEw4dOiTwPC9kZGQIf/nLX4Qff/xR7KHZxMsvvywMHjxY2LVrl9hDsYnY2FjhjTfeEGpqaoTr168LY8eOFfbu3Sv2sASaSYusuLgYf/3rXzFgwADIZDLExsZCLpcjOztb7KF1yMmTJ6HVajF79mwoFAo88sgjGDBgAA4cOCD20Kzmxo0biImJQXR0NGQyGUJCQhAREYGzZ8+KPTSr2717N1QqFQYOHCj2UGwiIyMD+fn5+Pvf/w53d3fcddddSEpKQmRkpNhDc64ueFKl0WhQXl7e5DjHcZg5c6bJsVOnTqG6uhp33323vYZnE5cvX0b//v1NjvXr1w85OTkijcj6wsPDER4ebvi+rKwMp0+fxrRp00QclfXl5+dj/fr12L59O+bOnSv2cGwiKysLAwcOxPr165GSkgJ3d3fEx8fjmWeeEXtoFKTt4dy5c5g1a1aT43K5HBcuXDB8n5OTg8TERLz00kvo1q2bPYdoddXV1fDw8DA55unpCbVaLdKIbKuyshLPPfccQkNDERUVJfZwrIbneSxevBhLlixBQECA2MOxmfLycpw5cwYRERFIT09Hbm4u5s6di4CAAEyZMkXUsVGQtoPIyMg20xeHDx/Gq6++imeffRbPPvusnUZmO15eXqitrTU5plar4eXlJdKIbOfq1atYsGABBgwYgNWrV0Mmc54s4scff4y+ffti4sSJYg/FppRKJXx8fPDiiy8CAAYPHowZM2YgLS1N9CDtPP9vcmBbtmzBK6+8gnfeeQfz5s0TezhW0b9/f1y9etXkWG5uLgYMGCDSiGzj1KlTiIuLw4QJE7Bu3Tq4u7uLPSSr2r9/P7777jtDaicnJwdvvfUW3nzzTbGHZlX9+vWDWq2GRqMxHON5XsQRGRH7yaWr279/vxAaGir8+uuvYg/Fqmpra4UHHnjApLojLCzM4atWjOXl5QlhYWHCl19+KfZQ7Gbq1KlOWd1RU1MjPPDAA8I//vEPoba2Vrh06ZIwcuRI4dtvvxV7aALtzCKyxx57DNnZ2U1mYB9++CHGjx8v0qisIycnB8uXL8elS5dw55134rXXXsOoUaPEHpbVrFy5Eps3b26SwomPj8fixYtFGpVtTZs2DU8//TSmT58u9lCsLj8/H2+//TYyMjKgVCoxe/ZszJkzR+xh0fZZhBAiZZSTJoQQCaMgTQghEkZBmhBCJIyCNCGESBgFaUIIkTAK0oQQImEUpIlTe/DBBzFo0CDDnyFDhmD8+PFYtWoVqqurDe+7cOECFi5ciPvuuw9hYWGIjY3Fvn37mj3nyZMnMWjQILzxxhvNvn7lyhUkJCRg2LBheOihh3Do0CGb3BtxDRSkidN75ZVXcOzYMRw7dgyHDx/Gu+++i71792LFihUAgKNHj+Kpp55Cjx498Pnnn2P37t2YMmUKli5dio0bNzY53969e9GnTx/s378fNTU1Jq/V1tZi7ty56Nu3L3bt2oW4uDi88soruHjxol3ulTghcRc8EmJb48ePF5KSkpoc//e//y2MGDFCUKlUwsiRI4V//etfTd7z6aefCiEhISZL2Wtra4Xw8HAhJSVFuPfee4Xdu3ebfGb37t3CfffdJ2i1WsOxF154QVi6dKn1boq4FJpJE5ckl8uhVCpx+PBhVFZWYvbs2U3eEx8fj02bNsHPz89w7PDhw1CpVBg7dizuv/9+7Nq1y+QzZ8+exbBhw+Dm1tBg8i9/+QvOnDljs3shzo2CNHEpOp0OmZmZ2Lp1KyZMmICLFy+ib9++8PHxafJeHx8fjBgxwiTg7t27F8OHD0eXLl0QHR2NU6dOIT8/3/B6UVERAgMDTc4TEBCAwsJC290UcWoUpInTe++99xAWFoawsDDce++9iI+PR0hICBYtWoSKiopmA3RzysvLcfToUURHRwNgDyXlcrnJbFqtVkOpVJp8TqlUmrTAJMQS1PSfOL358+dj6tSpAACFQoFu3boZAqm/vz8qKirMOs/Bgweh1WoNDfA7d+6MiIgI7N69GwsXLoRMJoOHh0eTgKzRaJrsUkOIuWgmTZyev78/goKCEBQUhJ49e5rMdENCQnD16lWoVKomn6usrERCQgJ+++03ACzVAQATJkzAkCFDMGTIEBw/fhy3bt3CsWPHAACBgYEoLi42OU9xcXGTFAgh5qIgTVza6NGj0aVLF3zxxRdNXtu+fTvOnTuHXr164caNGzh79ixefPFF/N///Z/hT0pKCry9vQ0pj+HDh+PcuXOoq6sznOfUqVMICwuz2z0R50LpDuLSPDw88Oabb+Lll19GVVUVHnvsMbi5ueHQoUPYsGEDFi9ejC5duuCTTz6Bu7s7Zs2aZVLtAbCNG77++muUlpZi4sSJ+PDDD/H3v/8dc+fOxU8//YQjR45g586dIt0hcXQ0kyYub8KECdi4cSN+//13PP3005gxYwZ++OEHvP/++3j66acBAPv27cPDDz/cJEADwMyZM1FXV4d9+/bBy8sLn332Ga5du4bHHnsMO3bswNq1azF48GB73xZxErQzCyGESBjNpAkhRMIoSBNCiIRRkCaEEAmjIE0IIRJGQZoQQiSMgjQhhEgYBWlCCJEwCtKEECJhFKQJIUTC/j9zSPVZ51bcywAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "LR tn, fp: 309, 1\n",
      "LR fn, tp: 11, 0\n",
      "LR f1 score: 0.000\n",
      "LR cohens kappa score: -0.006\n",
      "LR average precision score: 0.183\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 309, 1\n",
      "GB fn, tp: 10, 1\n",
      "GB f1 score: 0.154\n",
      "GB cohens kappa score: 0.145\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 310, 0\n",
      "KNN fn, tp: 11, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "------ Step 4/5: Slice 3/5 -------\n",
      "-> Reset the GAN\n",
      "-> Train generator for synthetic samples\n",
      "Epoch 1/3\n",
      "Epoch 2/3\n",
      "Epoch 3/3\n",
      "-> create 1194 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "LR tn, fp: 310, 0\n",
      "LR fn, tp: 11, 0\n",
      "LR f1 score: 0.000\n",
      "LR cohens kappa score: 0.000\n",
      "LR average precision score: 0.062\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 309, 1\n",
      "GB fn, tp: 11, 0\n",
      "GB f1 score: 0.000\n",
      "GB cohens kappa score: -0.006\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 310, 0\n",
      "KNN fn, tp: 11, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "------ Step 4/5: Slice 4/5 -------\n",
      "-> Reset the GAN\n",
      "-> Train generator for synthetic samples\n",
      "Epoch 1/3\n",
      "Epoch 2/3\n",
      "Epoch 3/3\n",
      "-> create 1194 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "LR tn, fp: 307, 3\n",
      "LR fn, tp: 11, 0\n",
      "LR f1 score: 0.000\n",
      "LR cohens kappa score: -0.015\n",
      "LR average precision score: 0.147\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 306, 4\n",
      "GB fn, tp: 11, 0\n",
      "GB f1 score: 0.000\n",
      "GB cohens kappa score: -0.019\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 310, 0\n",
      "KNN fn, tp: 11, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "------ Step 4/5: Slice 5/5 -------\n",
      "-> Reset the GAN\n",
      "-> Train generator for synthetic samples\n",
      "Epoch 1/3\n",
      "Epoch 2/3\n",
      "Epoch 3/3\n",
      "-> create 1196 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "LR tn, fp: 306, 0\n",
      "LR fn, tp: 9, 0\n",
      "LR f1 score: 0.000\n",
      "LR cohens kappa score: 0.000\n",
      "LR average precision score: 0.049\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 302, 4\n",
      "GB fn, tp: 9, 0\n",
      "GB f1 score: 0.000\n",
      "GB cohens kappa score: -0.018\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 306, 0\n",
      "KNN fn, tp: 9, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "====== Step 5/5 =======\n",
      "-> Shuffling data\n",
      "-> Spliting data to slices\n",
      "\n",
      "------ Step 5/5: Slice 1/5 -------\n",
      "-> Reset the GAN\n",
      "-> Train generator for synthetic samples\n",
      "Epoch 1/3\n",
      "Epoch 2/3\n",
      "Epoch 3/3\n",
      "-> create 1194 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "LR tn, fp: 309, 1\n",
      "LR fn, tp: 11, 0\n",
      "LR f1 score: 0.000\n",
      "LR cohens kappa score: -0.006\n",
      "LR average precision score: 0.076\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 305, 5\n",
      "GB fn, tp: 11, 0\n",
      "GB f1 score: 0.000\n",
      "GB cohens kappa score: -0.022\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 310, 0\n",
      "KNN fn, tp: 11, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "------ Step 5/5: Slice 2/5 -------\n",
      "-> Reset the GAN\n",
      "-> Train generator for synthetic samples\n",
      "Epoch 1/3\n",
      "Epoch 2/3\n",
      "Epoch 3/3\n",
      "-> create 1194 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "LR tn, fp: 310, 0\n",
      "LR fn, tp: 11, 0\n",
      "LR f1 score: 0.000\n",
      "LR cohens kappa score: 0.000\n",
      "LR average precision score: 0.099\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 308, 2\n",
      "GB fn, tp: 10, 1\n",
      "GB f1 score: 0.143\n",
      "GB cohens kappa score: 0.130\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 310, 0\n",
      "KNN fn, tp: 11, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "------ Step 5/5: Slice 3/5 -------\n",
      "-> Reset the GAN\n",
      "-> Train generator for synthetic samples\n",
      "Epoch 1/3\n",
      "Epoch 2/3\n",
      "Epoch 3/3\n",
      "-> create 1194 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "LR tn, fp: 310, 0\n",
      "LR fn, tp: 11, 0\n",
      "LR f1 score: 0.000\n",
      "LR cohens kappa score: 0.000\n",
      "LR average precision score: 0.203\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 307, 3\n",
      "GB fn, tp: 10, 1\n",
      "GB f1 score: 0.133\n",
      "GB cohens kappa score: 0.117\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 310, 0\n",
      "KNN fn, tp: 11, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "------ Step 5/5: Slice 4/5 -------\n",
      "-> Reset the GAN\n",
      "-> Train generator for synthetic samples\n",
      "Epoch 1/3\n",
      "Epoch 2/3\n",
      "Epoch 3/3\n",
      "-> create 1194 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "LR tn, fp: 309, 1\n",
      "LR fn, tp: 10, 1\n",
      "LR f1 score: 0.154\n",
      "LR cohens kappa score: 0.145\n",
      "LR average precision score: 0.225\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 309, 1\n",
      "GB fn, tp: 10, 1\n",
      "GB f1 score: 0.154\n",
      "GB cohens kappa score: 0.145\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 310, 0\n",
      "KNN fn, tp: 11, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "------ Step 5/5: Slice 5/5 -------\n",
      "-> Reset the GAN\n",
      "-> Train generator for synthetic samples\n",
      "Epoch 1/3\n",
      "Epoch 2/3\n",
      "Epoch 3/3\n",
      "-> create 1196 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "LR tn, fp: 306, 0\n",
      "LR fn, tp: 9, 0\n",
      "LR f1 score: 0.000\n",
      "LR cohens kappa score: 0.000\n",
      "LR average precision score: 0.191\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 305, 1\n",
      "GB fn, tp: 9, 0\n",
      "GB f1 score: 0.000\n",
      "GB cohens kappa score: -0.006\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 306, 0\n",
      "KNN fn, tp: 9, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "### Exercise is done.\n",
      "\n",
      "-----[ LR ]-----\n",
      "maximum:\n",
      "LR tn, fp: 310, 3\n",
      "LR fn, tp: 11, 1\n",
      "LR f1 score: 0.154\n",
      "LR cohens kappa score: 0.145\n",
      "LR average precision score: 0.345\n",
      "\n",
      "\n",
      "average:\n",
      "LR tn, fp: 308.52, 0.68\n",
      "LR fn, tp: 10.48, 0.12\n",
      "LR f1 score: 0.018\n",
      "LR cohens kappa score: 0.014\n",
      "LR average precision score: 0.154\n",
      "\n",
      "\n",
      "minimum:\n",
      "LR tn, fp: 306, 0\n",
      "LR fn, tp: 9, 0\n",
      "LR f1 score: 0.000\n",
      "LR cohens kappa score: -0.015\n",
      "LR average precision score: 0.049\n",
      "\n",
      "\n",
      "-----[ GB ]-----\n",
      "maximum:\n",
      "GB tn, fp: 310, 5\n",
      "GB fn, tp: 11, 1\n",
      "GB f1 score: 0.200\n",
      "GB cohens kappa score: 0.195\n",
      "\n",
      "\n",
      "average:\n",
      "GB tn, fp: 307.04, 2.16\n",
      "GB fn, tp: 10.16, 0.44\n",
      "GB f1 score: 0.068\n",
      "GB cohens kappa score: 0.056\n",
      "\n",
      "\n",
      "minimum:\n",
      "GB tn, fp: 302, 0\n",
      "GB fn, tp: 8, 0\n",
      "GB f1 score: 0.000\n",
      "GB cohens kappa score: -0.022\n",
      "\n",
      "\n",
      "-----[ KNN ]-----\n",
      "maximum:\n",
      "KNN tn, fp: 310, 0\n",
      "KNN fn, tp: 11, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "average:\n",
      "KNN tn, fp: 309.2, 0.0\n",
      "KNN fn, tp: 10.6, 0.0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n",
      "\n",
      "minimum:\n",
      "KNN tn, fp: 306, 0\n",
      "KNN fn, tp: 9, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: 0.000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "exercise = Exercise(shuffleFunction=shuffler, numOfShuffles=5, numOfSlices=5)\n",
    "exercise.run(gan, data)\n",
    "exercise.saveResultsTo(f\"{datasetName}-{ganName}.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3b6bac36",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}