{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "1df0b7e7",
   "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": "230249b8",
   "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": "654990a6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'folding': [(array([[3, 3, 0, 0, 2, 1],\n",
      "       [3, 3, 0, 0, 2, 2],\n",
      "       [3, 3, 0, 0, 1, 2],\n",
      "       ...,\n",
      "       [1, 1, 3, 2, 1, 0],\n",
      "       [1, 1, 3, 2, 0, 1],\n",
      "       [1, 1, 3, 2, 0, 0]], dtype=int64), array([0, 0, 0, ..., 1, 0, 1], dtype=int64), array([[3, 3, 0, 0, 2, 0],\n",
      "       [3, 3, 0, 0, 1, 1],\n",
      "       [3, 3, 0, 1, 1, 0],\n",
      "       ...,\n",
      "       [1, 1, 3, 1, 1, 2],\n",
      "       [1, 1, 3, 2, 2, 2],\n",
      "       [1, 1, 3, 2, 0, 2]], dtype=int64), array([0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 1, 0, 0, 1, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 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], dtype=int64)), (array([[3, 3, 0, 0, 2, 2],\n",
      "       [3, 3, 0, 0, 2, 0],\n",
      "       [3, 3, 0, 0, 1, 1],\n",
      "       ...,\n",
      "       [1, 1, 3, 2, 0, 1],\n",
      "       [1, 1, 3, 2, 0, 2],\n",
      "       [1, 1, 3, 2, 0, 0]], dtype=int64), array([0, 0, 0, ..., 0, 0, 1], dtype=int64), array([[3, 3, 0, 0, 2, 1],\n",
      "       [3, 3, 0, 0, 0, 2],\n",
      "       [3, 3, 0, 1, 0, 1],\n",
      "       ...,\n",
      "       [1, 1, 3, 0, 0, 0],\n",
      "       [1, 1, 3, 1, 0, 1],\n",
      "       [1, 1, 3, 1, 0, 2]], dtype=int64), array([0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 1, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 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, 0, 0, 1, 1, 0,\n",
      "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0], dtype=int64)), (array([[3, 3, 0, 0, 2, 1],\n",
      "       [3, 3, 0, 0, 2, 0],\n",
      "       [3, 3, 0, 0, 1, 1],\n",
      "       ...,\n",
      "       [1, 1, 3, 2, 1, 1],\n",
      "       [1, 1, 3, 2, 1, 0],\n",
      "       [1, 1, 3, 2, 0, 2]], dtype=int64), array([0, 0, 0, ..., 0, 1, 0], dtype=int64), array([[3, 3, 0, 0, 2, 2],\n",
      "       [3, 3, 0, 0, 1, 2],\n",
      "       [3, 3, 0, 0, 0, 1],\n",
      "       ...,\n",
      "       [1, 1, 3, 2, 1, 2],\n",
      "       [1, 1, 3, 2, 0, 1],\n",
      "       [1, 1, 3, 2, 0, 0]], dtype=int64), array([0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 1, 0, 0, 0, 0, 1, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
      "       0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
      "       1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 1], dtype=int64)), (array([[3, 3, 0, 0, 2, 1],\n",
      "       [3, 3, 0, 0, 2, 2],\n",
      "       [3, 3, 0, 0, 2, 0],\n",
      "       ...,\n",
      "       [1, 1, 3, 2, 0, 1],\n",
      "       [1, 1, 3, 2, 0, 2],\n",
      "       [1, 1, 3, 2, 0, 0]], dtype=int64), array([0, 0, 0, ..., 0, 0, 1], dtype=int64), array([[3, 3, 0, 0, 1, 0],\n",
      "       [3, 3, 0, 1, 1, 1],\n",
      "       [3, 3, 0, 1, 1, 2],\n",
      "       ...,\n",
      "       [1, 1, 3, 0, 2, 2],\n",
      "       [1, 1, 3, 2, 1, 1],\n",
      "       [1, 1, 3, 2, 1, 0]], dtype=int64), array([0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 1,\n",
      "       0, 1, 0, 0, 0, 0, 0, 0, 1, 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, 0, 1, 0, 0, 0,\n",
      "       1, 0, 0, 0, 0, 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, 1, 0, 0, 1, 0, 0, 1], dtype=int64)), (array([[3, 3, 0, 0, 2, 1],\n",
      "       [3, 3, 0, 0, 2, 2],\n",
      "       [3, 3, 0, 0, 2, 0],\n",
      "       ...,\n",
      "       [1, 1, 3, 2, 0, 1],\n",
      "       [1, 1, 3, 2, 0, 2],\n",
      "       [1, 1, 3, 2, 0, 0]], dtype=int64), array([0, 0, 0, ..., 0, 0, 1], dtype=int64), array([[3, 3, 0, 0, 0, 0],\n",
      "       [3, 3, 0, 1, 2, 1],\n",
      "       [3, 3, 0, 2, 2, 2],\n",
      "       ...,\n",
      "       [1, 1, 3, 1, 1, 1],\n",
      "       [1, 1, 3, 2, 2, 1],\n",
      "       [1, 1, 3, 2, 2, 0]], dtype=int64), array([0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0,\n",
      "       0, 0, 1, 0, 1, 1, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 1, 0, 1, 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, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0], dtype=int64)), (array([[3, 3, 0, 0, 2, 1],\n",
      "       [3, 3, 0, 0, 2, 0],\n",
      "       [3, 3, 0, 0, 1, 1],\n",
      "       ...,\n",
      "       [1, 1, 3, 2, 1, 0],\n",
      "       [1, 1, 3, 2, 0, 1],\n",
      "       [1, 1, 3, 2, 0, 2]], dtype=int64), array([0, 0, 0, ..., 1, 0, 0], dtype=int64), array([[3, 3, 0, 0, 2, 2],\n",
      "       [3, 3, 0, 1, 0, 1],\n",
      "       [3, 3, 0, 1, 0, 2],\n",
      "       ...,\n",
      "       [1, 1, 3, 2, 2, 1],\n",
      "       [1, 1, 3, 2, 1, 1],\n",
      "       [1, 1, 3, 2, 0, 0]], dtype=int64), array([0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 1, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 1, 1, 0,\n",
      "       0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 0, 0,\n",
      "       0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1], dtype=int64)), (array([[3, 3, 0, 0, 2, 1],\n",
      "       [3, 3, 0, 0, 2, 2],\n",
      "       [3, 3, 0, 0, 2, 0],\n",
      "       ...,\n",
      "       [1, 1, 3, 2, 0, 1],\n",
      "       [1, 1, 3, 2, 0, 2],\n",
      "       [1, 1, 3, 2, 0, 0]], dtype=int64), array([0, 0, 0, ..., 0, 0, 1], dtype=int64), array([[3, 3, 0, 0, 1, 2],\n",
      "       [3, 3, 0, 0, 0, 2],\n",
      "       [3, 3, 0, 1, 2, 1],\n",
      "       ...,\n",
      "       [1, 1, 2, 2, 2, 2],\n",
      "       [1, 1, 3, 0, 2, 2],\n",
      "       [1, 1, 3, 0, 1, 2]], dtype=int64), array([0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 1, 1,\n",
      "       0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 1, 1, 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, 1, 0, 0, 0], dtype=int64)), (array([[3, 3, 0, 0, 2, 2],\n",
      "       [3, 3, 0, 0, 1, 2],\n",
      "       [3, 3, 0, 0, 1, 0],\n",
      "       ...,\n",
      "       [1, 1, 3, 2, 0, 1],\n",
      "       [1, 1, 3, 2, 0, 2],\n",
      "       [1, 1, 3, 2, 0, 0]], dtype=int64), array([0, 0, 0, ..., 0, 0, 1], dtype=int64), array([[3, 3, 0, 0, 2, 1],\n",
      "       [3, 3, 0, 0, 2, 0],\n",
      "       [3, 3, 0, 0, 1, 1],\n",
      "       ...,\n",
      "       [1, 1, 3, 0, 2, 0],\n",
      "       [1, 1, 3, 1, 1, 2],\n",
      "       [1, 1, 3, 1, 0, 1]], dtype=int64), array([0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 0, 1, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0,\n",
      "       0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0,\n",
      "       0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0], dtype=int64)), (array([[3, 3, 0, 0, 2, 1],\n",
      "       [3, 3, 0, 0, 2, 2],\n",
      "       [3, 3, 0, 0, 2, 0],\n",
      "       ...,\n",
      "       [1, 1, 3, 2, 0, 1],\n",
      "       [1, 1, 3, 2, 0, 2],\n",
      "       [1, 1, 3, 2, 0, 0]], dtype=int64), array([0, 0, 0, ..., 0, 0, 1], dtype=int64), array([[3, 3, 0, 1, 0, 0],\n",
      "       [3, 3, 1, 0, 2, 2],\n",
      "       [3, 3, 1, 0, 0, 2],\n",
      "       ...,\n",
      "       [1, 1, 3, 2, 2, 0],\n",
      "       [1, 1, 3, 2, 1, 2],\n",
      "       [1, 1, 3, 2, 1, 0]], dtype=int64), array([0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 1, 0, 0, 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, 1, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
      "       0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0,\n",
      "       1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1], dtype=int64)), (array([[3, 3, 0, 0, 2, 1],\n",
      "       [3, 3, 0, 0, 2, 2],\n",
      "       [3, 3, 0, 0, 2, 0],\n",
      "       ...,\n",
      "       [1, 1, 3, 2, 1, 2],\n",
      "       [1, 1, 3, 2, 1, 0],\n",
      "       [1, 1, 3, 2, 0, 0]], dtype=int64), array([0, 0, 0, ..., 0, 1, 1], dtype=int64), array([[3, 3, 0, 0, 1, 0],\n",
      "       [3, 3, 0, 2, 2, 2],\n",
      "       [3, 3, 0, 2, 0, 2],\n",
      "       ...,\n",
      "       [1, 1, 3, 2, 2, 2],\n",
      "       [1, 1, 3, 2, 0, 1],\n",
      "       [1, 1, 3, 2, 0, 2]], dtype=int64), array([0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0,\n",
      "       0, 0, 0, 0, 0, 0, 1, 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, 0, 1, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 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, 1, 0, 0, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0], dtype=int64)), (array([[3, 3, 0, 0, 2, 1],\n",
      "       [3, 3, 0, 0, 2, 2],\n",
      "       [3, 3, 0, 0, 2, 0],\n",
      "       ...,\n",
      "       [1, 1, 3, 2, 1, 0],\n",
      "       [1, 1, 3, 2, 0, 1],\n",
      "       [1, 1, 3, 2, 0, 0]], dtype=int64), array([0, 0, 0, ..., 1, 0, 1], dtype=int64), array([[3, 3, 0, 1, 2, 1],\n",
      "       [3, 3, 1, 0, 2, 1],\n",
      "       [3, 3, 1, 0, 1, 0],\n",
      "       ...,\n",
      "       [1, 1, 3, 0, 2, 2],\n",
      "       [1, 1, 3, 1, 2, 1],\n",
      "       [1, 1, 3, 2, 0, 2]], dtype=int64), array([0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0,\n",
      "       0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0,\n",
      "       0, 1, 0, 0, 0, 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, 0, 1, 0, 1, 0,\n",
      "       0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0], dtype=int64)), (array([[3, 3, 0, 0, 2, 1],\n",
      "       [3, 3, 0, 0, 2, 2],\n",
      "       [3, 3, 0, 0, 2, 0],\n",
      "       ...,\n",
      "       [1, 1, 3, 2, 0, 1],\n",
      "       [1, 1, 3, 2, 0, 2],\n",
      "       [1, 1, 3, 2, 0, 0]], dtype=int64), array([0, 0, 0, ..., 0, 0, 1], dtype=int64), array([[3, 3, 0, 0, 1, 1],\n",
      "       [3, 3, 0, 0, 0, 2],\n",
      "       [3, 3, 0, 1, 2, 2],\n",
      "       ...,\n",
      "       [1, 1, 3, 2, 2, 2],\n",
      "       [1, 1, 3, 2, 2, 0],\n",
      "       [1, 1, 3, 2, 1, 1]], dtype=int64), array([0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1,\n",
      "       0, 0, 0, 0, 0, 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, 0, 1, 0, 1, 0,\n",
      "       0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0,\n",
      "       1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1,\n",
      "       0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0], dtype=int64)), (array([[3, 3, 0, 0, 2, 1],\n",
      "       [3, 3, 0, 0, 1, 1],\n",
      "       [3, 3, 0, 0, 1, 0],\n",
      "       ...,\n",
      "       [1, 1, 3, 2, 1, 2],\n",
      "       [1, 1, 3, 2, 0, 1],\n",
      "       [1, 1, 3, 2, 0, 2]], dtype=int64), array([0, 0, 0, ..., 0, 0, 0], dtype=int64), array([[3, 3, 0, 0, 2, 2],\n",
      "       [3, 3, 0, 0, 2, 0],\n",
      "       [3, 3, 0, 0, 1, 2],\n",
      "       ...,\n",
      "       [1, 1, 3, 2, 2, 1],\n",
      "       [1, 1, 3, 2, 1, 0],\n",
      "       [1, 1, 3, 2, 0, 0]], dtype=int64), array([0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
      "       1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 1, 0, 0, 0, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 1], dtype=int64)), (array([[3, 3, 0, 0, 2, 2],\n",
      "       [3, 3, 0, 0, 2, 0],\n",
      "       [3, 3, 0, 0, 1, 1],\n",
      "       ...,\n",
      "       [1, 1, 3, 2, 0, 1],\n",
      "       [1, 1, 3, 2, 0, 2],\n",
      "       [1, 1, 3, 2, 0, 0]], dtype=int64), array([0, 0, 0, ..., 0, 0, 1], dtype=int64), array([[3, 3, 0, 0, 2, 1],\n",
      "       [3, 3, 0, 0, 1, 0],\n",
      "       [3, 3, 0, 0, 0, 1],\n",
      "       ...,\n",
      "       [1, 1, 3, 0, 0, 2],\n",
      "       [1, 1, 3, 1, 1, 1],\n",
      "       [1, 1, 3, 2, 1, 2]], dtype=int64), array([0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 1,\n",
      "       1, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1,\n",
      "       0, 0, 1, 0, 0, 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, 1, 0, 0, 0, 0], dtype=int64)), (array([[3, 3, 0, 0, 2, 1],\n",
      "       [3, 3, 0, 0, 2, 2],\n",
      "       [3, 3, 0, 0, 2, 0],\n",
      "       ...,\n",
      "       [1, 1, 3, 2, 1, 0],\n",
      "       [1, 1, 3, 2, 0, 2],\n",
      "       [1, 1, 3, 2, 0, 0]], dtype=int64), array([0, 0, 0, ..., 1, 0, 1], dtype=int64), array([[3, 3, 0, 1, 1, 1],\n",
      "       [3, 3, 0, 1, 1, 2],\n",
      "       [3, 3, 0, 1, 0, 0],\n",
      "       ...,\n",
      "       [1, 1, 3, 0, 0, 1],\n",
      "       [1, 1, 3, 1, 2, 2],\n",
      "       [1, 1, 3, 2, 0, 1]], dtype=int64), array([0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
      "       0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 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, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
      "       0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0,\n",
      "       1, 0, 0, 0, 1, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0], dtype=int64))], 'db_size': 1728, 'db_n_attr': 6, 'imbalanced_ratio': 25.584615384615386}\n"
     ]
    }
   ],
   "source": [
    "datasetName = \"folding_car-vgood\"\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": "20243fc0",
   "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": "fd784e03",
   "metadata": {},
   "outputs": [],
   "source": [
    "myData = example_dict1[\"folding\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "f7520973",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([[3, 3, 0, 0, 2, 1],\n",
       "        [3, 3, 0, 0, 2, 2],\n",
       "        [3, 3, 0, 0, 1, 2],\n",
       "        ...,\n",
       "        [1, 1, 3, 2, 1, 0],\n",
       "        [1, 1, 3, 2, 0, 1],\n",
       "        [1, 1, 3, 2, 0, 0]], dtype=int64),\n",
       " array([0, 0, 0, ..., 1, 0, 1], dtype=int64),\n",
       " array([[3, 3, 0, 0, 2, 0],\n",
       "        [3, 3, 0, 0, 1, 1],\n",
       "        [3, 3, 0, 1, 1, 0],\n",
       "        ...,\n",
       "        [1, 1, 3, 1, 1, 2],\n",
       "        [1, 1, 3, 2, 2, 2],\n",
       "        [1, 1, 3, 2, 0, 2]], dtype=int64),\n",
       " array([0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
       "        0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
       "        0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
       "        0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
       "        0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
       "        0, 0, 1, 0, 0, 1, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0,\n",
       "        0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0,\n",
       "        0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
       "        0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 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], dtype=int64))"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "myData[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "3946ca7f",
   "metadata": {},
   "outputs": [],
   "source": [
    "k = myData[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "23b5bdfd",
   "metadata": {},
   "outputs": [],
   "source": [
    "data_con = np.concatenate((k[0],k[2]),axis = 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "85fd78d1",
   "metadata": {},
   "outputs": [],
   "source": [
    "target = np.concatenate((k[1],k[3]),axis = 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "0a766eb8",
   "metadata": {},
   "outputs": [],
   "source": [
    "target = target.astype(float)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "0f3ddaba",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1728,)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "target.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "4d4643d3",
   "metadata": {},
   "outputs": [],
   "source": [
    "data_con = data_con.astype(float)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "2363c631",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1728, 6)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_con.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "9e5df4be",
   "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": 15,
   "id": "e8798a5a",
   "metadata": {},
   "outputs": [],
   "source": [
    "data = loadDataset()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "98e7807a",
   "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": 17,
   "id": "e4081536",
   "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": 18,
   "id": "42e6029a",
   "metadata": {},
   "outputs": [],
   "source": [
    "ganName = \"SimpleGAN\"\n",
    "gan = SimpleGan(numOfFeatures=k[2].shape[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "c985c643",
   "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 1278 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: 327, 6\n",
      "LR fn, tp: 12, 1\n",
      "LR f1 score: 0.100\n",
      "LR cohens kappa score: 0.076\n",
      "LR average precision score: 0.367\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 333, 0\n",
      "GB fn, tp: 0, 13\n",
      "GB f1 score: 1.000\n",
      "GB cohens kappa score: 1.000\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 333, 0\n",
      "KNN fn, tp: 11, 2\n",
      "KNN f1 score: 0.267\n",
      "KNN cohens kappa score: 0.259\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 1278 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: 326, 7\n",
      "LR fn, tp: 10, 3\n",
      "LR f1 score: 0.261\n",
      "LR cohens kappa score: 0.236\n",
      "LR average precision score: 0.305\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 332, 1\n",
      "GB fn, tp: 0, 13\n",
      "GB f1 score: 0.963\n",
      "GB cohens kappa score: 0.961\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 332, 1\n",
      "KNN fn, tp: 7, 6\n",
      "KNN f1 score: 0.600\n",
      "KNN cohens kappa score: 0.589\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 1278 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: 191, 142\n",
      "LR fn, tp: 5, 8\n",
      "LR f1 score: 0.098\n",
      "LR cohens kappa score: 0.031\n",
      "LR average precision score: 0.142\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 326, 7\n",
      "GB fn, tp: 4, 9\n",
      "GB f1 score: 0.621\n",
      "GB cohens kappa score: 0.604\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 328, 5\n",
      "KNN fn, tp: 9, 4\n",
      "KNN f1 score: 0.364\n",
      "KNN cohens kappa score: 0.343\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 1278 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: 324, 9\n",
      "LR fn, tp: 9, 4\n",
      "LR f1 score: 0.308\n",
      "LR cohens kappa score: 0.281\n",
      "LR average precision score: 0.291\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 333, 0\n",
      "GB fn, tp: 0, 13\n",
      "GB f1 score: 1.000\n",
      "GB cohens kappa score: 1.000\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 333, 0\n",
      "KNN fn, tp: 9, 4\n",
      "KNN f1 score: 0.471\n",
      "KNN cohens kappa score: 0.461\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 1280 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: 218, 113\n",
      "LR fn, tp: 8, 5\n",
      "LR f1 score: 0.076\n",
      "LR cohens kappa score: 0.009\n",
      "LR average precision score: 0.051\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 326, 5\n",
      "GB fn, tp: 2, 11\n",
      "GB f1 score: 0.759\n",
      "GB cohens kappa score: 0.748\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 330, 1\n",
      "KNN fn, tp: 8, 5\n",
      "KNN f1 score: 0.526\n",
      "KNN cohens kappa score: 0.515\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 1278 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: 326, 7\n",
      "LR fn, tp: 9, 4\n",
      "LR f1 score: 0.333\n",
      "LR cohens kappa score: 0.310\n",
      "LR average precision score: 0.225\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 333, 0\n",
      "GB fn, tp: 0, 13\n",
      "GB f1 score: 1.000\n",
      "GB cohens kappa score: 1.000\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 333, 0\n",
      "KNN fn, tp: 8, 5\n",
      "KNN f1 score: 0.556\n",
      "KNN cohens kappa score: 0.546\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 1278 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: 198, 135\n",
      "LR fn, tp: 6, 7\n",
      "LR f1 score: 0.090\n",
      "LR cohens kappa score: 0.023\n",
      "LR average precision score: 0.080\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 327, 6\n",
      "GB fn, tp: 0, 13\n",
      "GB f1 score: 0.813\n",
      "GB cohens kappa score: 0.804\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 329, 4\n",
      "KNN fn, tp: 7, 6\n",
      "KNN f1 score: 0.522\n",
      "KNN cohens kappa score: 0.506\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 1278 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: 325, 8\n",
      "LR fn, tp: 9, 4\n",
      "LR f1 score: 0.320\n",
      "LR cohens kappa score: 0.295\n",
      "LR average precision score: 0.252\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 333, 0\n",
      "GB fn, tp: 1, 12\n",
      "GB f1 score: 0.960\n",
      "GB cohens kappa score: 0.959\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 331, 2\n",
      "KNN fn, tp: 8, 5\n",
      "KNN f1 score: 0.500\n",
      "KNN cohens kappa score: 0.486\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 1278 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: 326, 7\n",
      "LR fn, tp: 12, 1\n",
      "LR f1 score: 0.095\n",
      "LR cohens kappa score: 0.069\n",
      "LR average precision score: 0.295\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 332, 1\n",
      "GB fn, tp: 1, 12\n",
      "GB f1 score: 0.923\n",
      "GB cohens kappa score: 0.920\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 333, 0\n",
      "KNN fn, tp: 10, 3\n",
      "KNN f1 score: 0.375\n",
      "KNN cohens kappa score: 0.366\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 1280 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: 329, 2\n",
      "LR fn, tp: 9, 4\n",
      "LR f1 score: 0.421\n",
      "LR cohens kappa score: 0.407\n",
      "LR average precision score: 0.566\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 331, 0\n",
      "GB fn, tp: 0, 13\n",
      "GB f1 score: 1.000\n",
      "GB cohens kappa score: 1.000\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 331, 0\n",
      "KNN fn, tp: 10, 3\n",
      "KNN f1 score: 0.375\n",
      "KNN cohens kappa score: 0.366\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 1278 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: 330, 3\n",
      "LR fn, tp: 11, 2\n",
      "LR f1 score: 0.222\n",
      "LR cohens kappa score: 0.206\n",
      "LR average precision score: 0.313\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 333, 0\n",
      "GB fn, tp: 2, 11\n",
      "GB f1 score: 0.917\n",
      "GB cohens kappa score: 0.914\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 333, 0\n",
      "KNN fn, tp: 8, 5\n",
      "KNN f1 score: 0.556\n",
      "KNN cohens kappa score: 0.546\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 1278 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: 210, 123\n",
      "LR fn, tp: 7, 6\n",
      "LR f1 score: 0.085\n",
      "LR cohens kappa score: 0.017\n",
      "LR average precision score: 0.054\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 330, 3\n",
      "GB fn, tp: 0, 13\n",
      "GB f1 score: 0.897\n",
      "GB cohens kappa score: 0.892\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 332, 1\n",
      "KNN fn, tp: 9, 4\n",
      "KNN f1 score: 0.444\n",
      "KNN cohens kappa score: 0.433\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 1278 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: 325, 8\n",
      "LR fn, tp: 7, 6\n",
      "LR f1 score: 0.444\n",
      "LR cohens kappa score: 0.422\n",
      "LR average precision score: 0.296\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 333, 0\n",
      "GB fn, tp: 0, 13\n",
      "GB f1 score: 1.000\n",
      "GB cohens kappa score: 1.000\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 333, 0\n",
      "KNN fn, tp: 9, 4\n",
      "KNN f1 score: 0.471\n",
      "KNN cohens kappa score: 0.461\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 1278 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: 328, 5\n",
      "LR fn, tp: 11, 2\n",
      "LR f1 score: 0.200\n",
      "LR cohens kappa score: 0.178\n",
      "LR average precision score: 0.390\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 333, 0\n",
      "GB fn, tp: 0, 13\n",
      "GB f1 score: 1.000\n",
      "GB cohens kappa score: 1.000\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 333, 0\n",
      "KNN fn, tp: 11, 2\n",
      "KNN f1 score: 0.267\n",
      "KNN cohens kappa score: 0.259\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 1280 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: 328, 3\n",
      "LR fn, tp: 10, 3\n",
      "LR f1 score: 0.316\n",
      "LR cohens kappa score: 0.299\n",
      "LR average precision score: 0.381\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 331, 0\n",
      "GB fn, tp: 0, 13\n",
      "GB f1 score: 1.000\n",
      "GB cohens kappa score: 1.000\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 331, 0\n",
      "KNN fn, tp: 10, 3\n",
      "KNN f1 score: 0.375\n",
      "KNN cohens kappa score: 0.366\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 1278 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: 200, 133\n",
      "LR fn, tp: 6, 7\n",
      "LR f1 score: 0.092\n",
      "LR cohens kappa score: 0.024\n",
      "LR average precision score: 0.061\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 330, 3\n",
      "GB fn, tp: 5, 8\n",
      "GB f1 score: 0.667\n",
      "GB cohens kappa score: 0.655\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 330, 3\n",
      "KNN fn, tp: 13, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: -0.014\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 1278 synthetic samples\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWkAAAFpCAYAAABee9lOAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAABRzElEQVR4nO2deXgUVbr/v9WdBEISkCU4MDqsimIIWwyiMIghejMGRglGDTfIlU1wZAZ+6FwvruMy6qDMoLggIIoCwg0oBFAYBhQcJCBcYkSDC7ssMQhJJ4FOuuv3Rydtd3Ul3Umf91RX5f08D09IJzl1qrvqe956z7soqqqqYBiGYSISm9ETYBiGYeqHRZphGCaCYZFmGIaJYFikGYZhIhgWaYZhmAiGRZphGCaCiTJ6AiIpLXXA7W56RGHbtq3w88+VAmckB7POGzDv3Hne8jHr3EOZd2JiQr0/Y0vah6gou9FTaBJmnTdg3rnzvOVj1rmHO28WaYZhmAgmokR669atGDlyJPr3748RI0ZgxYoVRk+JYRjGUCLGJ33mzBlMnz4dr7zyCoYNG4avvvoKd999N/r06YNrrrnG6OkxDMMYQsSIdMeOHbFz507Ex8fD7Xbj3LlzsNvtiIuLM3pqDMMwhhExIg0A8fHxqKqqQkpKCmpqajBp0iR07drV6GkxDMMYhhJpVfDcbjdqampQXFyMyZMnY+bMmbjjjjvCGu+nn37C2bM/w+VyCZypscTGtsTll1+O6Ohoo6fCMAwhESfSvsydOxfFxcV4/fXXQ/p9vTjps2fPQFEUJCRcArs9Coqi1Pv3UVE21NS4w5qzDFRVRUVFGS5cqESHDp2QmJiAkpJyo6fVJMw6d563fMw691DmbYo46YKCAowePdrvNafTidatW4c1rtN5AZdc0h5RUdENCrSZUBQFcXGtUVPjNHoqDMMQEzEiffXVV+P06dN466234HK5sHfvXuTl5WHMmDFhjqxCUSLmNIVhlQWHYZiGiRj1SkhIwIIFC7Bp0yakpqbisccew9NPP43U1FSjp8YwDGMYERXdcc0112D58uVGT8NQvv32IObM+Su+//5bdO78azz88GO4+mqOE2eY5krEWNIMUF1djYcf/n9IS0vHRx9tw7hx92LGjD+gosJh9NSYZoZry2ZUTxgH15bNRk+l2cMiHUHs2/cFampqkJ2dg6ioKIwYcQu6deuOLXyjMJJxL1sK9cgRuJctNXoqzR4W6Qji8OEf0KVLN7/XunTpih9++M6gGTGNwUrWpy0nF0qXLrDl5Bo9lWYPi3QEUVlZiZYtW/i91rJlS1y4cMGgGTGNQab1Sb0g2NPSEb3oHdjT0knGZ0KHRTpEZFhJsbGxuHjxot9rFy5cQGxsK7JjMuKw5eQCrVtD/flncmua3RHNBxbpEJFxU3Tt2h1Hjx7xe+3IkcPo2rVbPX/BNAYZ1qfSti1QVkYunuyOaD6wSIeIjJtiwIAUAMD777+Hmpoa/POfH+P777/Db387nOyYzQkZC60sa5rdEc0HFukQkXFTREdHY86cf2Dbtn8hI+MmvPPOYvz1r3PQtm1bsmM2J2QIqExr2koblUz9sEhHGN2798Rrry3C5s2f4p133sfAgdcaPSUpyBAcaQKalAxcqPJ8JYT90s0DFmkmIpAlOGpCa+BsqecrFUWFQMtYz1dC2C/dPGCRZiICaYKzpwBQVc9XItgvzYiERZqJCGQJjpKVDcTFeb4SIdMvbWbYpx4aLNJMxCDjpo2eMQsxOwoQPWMW2TEA87siZHwW7FMPDRZpJmKQddPK2qS05eTCvWypKS1FWeGKZl7IZMEizUQMsny5shYDiuNIcxFIiFBhn3posEgzIWGlEDlZiwGFpehethTqwWK4Hp9NK9SSIlSY4LBIMyEhzX+YlAw4yqEeO2b6hBMKS9GWk+uJTlFs7IpoJrBIMyEhrXhQUSFQUwOcLbVEwolo7GnpsD/5DJCYSJ45ya6IyIBFOkI5cKAImZmh3yBWKR5ky8kF2rUHoqJoBdTEj/OyPgsOkYsMWKQjDFVVsW7dB5g58w+oqakO+e+kuCMkbSYpl18OxCeQCijV47yVNvY4RC4yYJGOMBYtegNr1vwvxo2b0Ki/k+JDlJjuTOFa8RVQqsd51/x5UPfthWv2n02/scd+6ciARTrCuO22LCxe/C6uuurqRv2dDB+izHRnAMDRI3DNnydsXGmWodsNOKtNv7HHfunIgEU6RHZ+X4qHVxdh5/elpMfp0CGxyX9rFb80FVJKld4/Heh4KRATzTHGjBBYpENk7f6TOHGuCmv3nzR6KvUizS9dXgb1+HEyVwRQK3a/6eL9mQioLHTtMWT41JnQsMLmJ4t0iIzq2wm/viQWo/p2Mnoq9SMjrKyoEHC5gNKfhC4G2gXGzFa7zF6HZkWWeFph85NFOkQG92iPv45OwuAe7Zv092pZGdyHD0EtKxM8Mx8kbSZRhMjp+Vgp/K4UFnrAMbg7S1BkiacVNj9ZpCWhni0FLl70fCVCWnsogsd5PR8rhd+Vu7NEBlxzO3RYpCWhtGsPtGjh+RoCAwak4KOPtjXqGDITTswcY8zdWYLDm9CRA4u0JJTWrWHr2g1Ka0JhAExdvUyaVSipO4sM8TTzZ2EFV4QMWKQlYhW/NBWyblolKxto0QJoFWe62hdW8uWavea2LFikJSLLL23Gx19Anv8wesYsKNckAW636R61rebLNatPXSYs0hJprF+6KZjdgpMGgVvISjW3ATnnwy6P4LBI16KWlaHm++9JXRHS/NIEyLyZpGwgEriF9BYyknORUHMbqK1Dsv//yBJ/AGtEX1DDIl2LerYU6kUnqSvCzMi8mWRtWol2G+iNSXIusmpuMxEBi3QtSrv2UIhdEUxoyKqCJ9ptoDcmxRMIZc1tv/fo/ulQ+vbzJAAxhsEiXYvSujWienQ3pStCJmb2uwZYtRThipoxRT6B1L33AMjqg/i+R2Z3RZg1G1MLi7SFkHFRSu11KFhAA6xaQX5pv/edMATS972netrQvkdU15SlrlViWKQjjN27P8e99/4nbr55GO688zZ88EFeyH9rle4sAEjETmsZinJFaMWTbIPV572nquinfY+orimz7jsYAYt0BHH69CnMnv1n3HPPBHz00VY88cQzeOON+di1a2dIf2+17izU8d6iHud9xQAAnYvAgEQlqs9BVrKMFVLPI0akP/vsM4wePRoDBgxAeno6VqxYYfSUpHPq1Emkp9+CYcOGw2az4eqrr0H//gPx5Zf7Q/p7q3VnMUu8t1YMXFs2wzl6JJyjR5K6Iqgq+sloMybL322FOOyIEOmTJ0/igQcewNSpU7Fnzx68+OKLeOmll7B9+3ajpyaVvn3748EH/8f7fVnZeRQW7sMVV/QKeQwujNMwlJahd/GaPw84fBj44XvhrgjfNGppG6xEWCmLlZKIEOkTJ04gMzMT6enpsNlsSE5ORmpqKvbu3Wv01Lx8UVqA5798El+U0hXl8cXhcODPf56J3r2TMHTosJD/zsy+PjPftL6CCcDTPssm/vbSfr5kIX4SnpastBhQEhEinZKSgr/85S/e78+dO4c9e/agd+/eBs7Kn3/+uBGnq07inz9uJD/W0aNHMGXKeLRr1w5PPfU8bI252SVVwRNlwVVt3Oi9gUx/09a+98rQYbA/8zzQtZv3eKKPQRHiVwfX3I4sooyegJby8nJMnToVffv2RVpaWqP+tn37+IDXzpyxISoqdJGr73dvufx32HR8I26+LKNR4zWWffu+wEMPzcTtt4/B1Kl/gKIo9f6uzWZDYmICAHi/lhR/hZpWrRBVfMD7GgVVUyfDsXAx4ifei9gwjlOycBGUY0dhX7UcbQSNGTDXjRtx/oW/QQHQ+qEH4Vi1DGrtMRPvGt3kcbXvb0nxV3DabMDqVWgz9Do4OnZAzQ+Hwj6O9hjaz7dq40Y4Fi5C/MQJiM3IaPS8dY+T2A7OfXsRndKf7DpqyrXalLmIulbDIZz3UFFVVRU4l7A4dOgQpk2bhp49e2LOnDlo0aJFo/6+tNQBt9v/dE6dOoJf/apLSH8fFWVDTY27UccUyYkTx/Ff/zUWkydPxZgxdwX9/bpzS0xMQElJOQB4LVJbTq4p/HDRC+ah4p13oWRlI3rGLJJjVE8YB3X//wEAlL79vH7dcN4j3/e8DteWzXA9PhtQbFCuvBJISoaat1Loubm2bPb6ue33T4c9Ld1zfkeOQOnSBdGL3mn0vPVwDkkFKiqAuDjE7KBx8emdS0OEOvdII5R5NyTiEeHuAIDdu3cjOzsbI0aMwLx58xot0FZg9eqVqKyswBtvzEd6+lDvv1dfDX3ziSLDjdKXV713n19YGcUxtWnUlH5p+5PPQLnySs8xiWK9ZaSeU9bc9sucNPEmtCwiwt1x9OhRTJkyBTNmzEBurnlDZcLlgQdm4oEHZho9DS/aFGEK4idOwNlnn/duUlEc056W7hn3YBXUvJVwJfclOR+9pxhvdqBAfJ8EgNrFQdCiXDdu9IxZqC4qJPn8tck/FO+RlYgIS/q9995DRUUFXnrpJfTv39/7729/+5vRU2vWyIgxjc3I8LOmKNOdoboBRfFabaKt9jrxcc2f57UUKZ5qpMWQE23s+V5XVgiRoyYiRPrhhx9GcXEx9u3b5/fvwQcfNHpqzRrtDUQdFUGd7qxkZXt6G9aKjuhd/zrxASA8mkA7Vyq3EEVtEy3SriuLEBEizUQuvjcQWShTUSGgKB5XhMxC/4ItxbpkEwBA69aksctUmZMUtU2CYfYQOWpYpJkGkVE8yJaTCzidQGUVXPPnkaU7B7hSCKrguZctBcrKoLRtSxu7bOKu8FqskLpNSbMQ6QiKMhSGrHOSUTzInpYOtLkEcLsAh4MsmUI7LlUVPLJsPV9hllRsycxZoFbB8iJtt0ehutpp9DSE43LVQKmoRPWEcajaSJcFqVc8iOSmjY/3hMjFexKSyMRO4/+mqIJHFlbmI8zsimg+WF6k4+MvwblzJXA6L1rGolZVN8rLf0aLLVugHjkCx8LFtAf0ETZPKNtBuB6fLVQ8tS4OstRkCbHLMoo4AYQlUTXHZFeEsUREnDQlsbFxAIDz53+Cy1XT4O/abDa43cZlHIaOgpiYlkjo0Qtqly6In3gvHJSH89nYU7KygeJvvKFswmOZfeJyKWJobTm5ni7YtRa6MJFLSgaKvoR6/DgABM38awp67xE1vvHevt8z8rC8SAMeoa4T64YwXdppWjqQlo7YxAQ4CNPCbTm53nRnFBXC/uQzNAkIScmeBcCneJDoJA0yoSsqBFwuoPQnr6BRpOcbkfwhe2Fg/LG8u8PquLZsRknWGK/rgSo0CympQIUDakJroRs9vlXwqDbDZCRpaFPPqXy5RsQYm83lYbW4axZpk+NethTVX3/j9RGT3VB7CjyJIHvEFttxLFxEHxmhFWUqv7RvB++kZKC8DOrx4/Q1mQ8WC98j8MVs0RdW2+xkkTY5tpxcwO328xFT3FBKVjYQF+f5KpD4iRP8UoRlbBaSpp7Xjqtu/yTA/UGBJ91dBRSbZUQpXMxm+QeDRdrk2NPS0Sp3rF+6MwXRM2YhZkeB8HKisRkZ/osKlSvCR5RlxGED8HN/iERbx8Ov8p6AMamR0eLNTJZ/MFikLYC23CcVUm5kSeU9yTL2fLqz+Lk/BKJ9nBchSjJdBFZzR1DDIm0B4idOsExPOrJH1aRkwFEO9dgx0k1KGQknZu5rWHcsK7kjqGGRJkaG9akt90kF1c0lowwnigqBmhrgbCl9SVSChBPq90hmF3iruSOoYZEmRtajnQxLSFodYwL0urPIqA8iClM/xehgtTA5SlikiZH1GCnTEhKNrAUmwEcsoai9yDFlvEd1yTLU4sl+6dBp9iItY6dZlnia1den9x6R9TokKFWqxczuCJlPfma8Vo2g2Yu0lR4jZfj6ZHRnAegyJ2UUQqJCxnxlPvmxXzo0mr1Iy7jwzX5BSuvOokk4IYvyEFyqFJBXd5nKHcEdvCOXZi/SZhdQGUjrzmKzQf2qCNVz59BGeRC06qJavLTiT3UcGZ+xlZC58dnsRZoJjrTuLJUVwMWLUPNWCh3bF7I0aqJNSO7gHZnI3PhkkTY52ip4FEjb/ExJBRTF85UIUWnUARQVAopNuIVu1GYnhaVopbA7mU8bLNImx71sKWp+OES/oktofKqUlwHt2nu+EiLKUvQVHY+F/kuhK1HobXaaNbtUlvVptb6MLNImx5aTi6ju3elXdAmNT2VkNIpEWwzf/uQzQGJHkixG8kqBDRxT5JhmXWCMhEXa5NjT0pGYt0pKrzszZTQGRKQQ9GWUWV2PulJg0GMKGtOsC4yRsEgzIeG9WY8egWv+PGMnEwLaaAUZrggAZALq9zQg4alG97gCMNtiHwmwSDOGQ5Vd6OciyMqmqbktoesLYFyInGjXgSxr2lKblEZPgBGDlM2S+6dD6dsP9vunCx2XrC9jUjJcj89G9dw5UkqTAoHx3sIgSsIJBsmCIMFdYyW/NIs0MbJWdBkXJYXPuGrjRjLLUM1bCVRUQM1bKaU0aZ1fmiTeW7MYuLZshnP0SDhHjxR2PnpjkiwIJt6ENgIWaWK4YI0+de+LY+FisnRn376MUkuT1sV52+xkvlz3sqXA0SPAUXHXFsWYerBfunGwSBNjtuJK0nx5tY+80QP6AaBZzAL6Mmq7s4hC8/iulJcBdjtQWUHmy7Xl5AK/6QL8Rty1ZcvJBdp38MydOHKE64OEDos0MWZb0aX58mofeav3/h8ASYuZpjuL0HG1fmkKsdP4pWNWr4P9/ulhP4H4FVe67DIgoTV55IjZnvyMhEWa8UPWzVP3yOs6W0rbNkt7zHbtPQWWBFrTuvHSAsQu4KlGx5crYlENCFeUVKpUVoMBs8Mizfgh6+ape+RVz533Ckz13DlwDkkVGxUBjaV4+eWeUDyB1rTe47sIsdMKsN6YIhZVbQEtWa4IK0VgUMIizQQg7eZJSoZa9Ysv1zcaQyQBlmJcPFBTA/W778QtRBq/tIjkn1AyGkU8gQSMKyFEDmCXR6iwSDMByHrkRVEhlFifx3eCqAgAAb5cuGo8r58/F9ZCRJ0FqCfKakJr4Gyp56tA/ASTqOa2FrPt1xgFi7QJMaIvI1VWoO2SS7yLAUVUBIAAAVWysoEWLYBLfxWWFRcsC1BE8k/AgrmnwOOu2VPQ5DH18HVzISmZpua2DlbKDKSCRdqEGNGXkSor0NZOE1ZW54oQaSlqHt+jZ8xCzK59sD/0cHi+9yBZgBSuCCUrG4iJAeLiyar6oaiwtqJfIlesiwBYpE2IEX0ZqVwg8RMn+G1awVXjKfwfpqUYiisibIHQyQLUWoVCLEWfxSB6xizg15cBJWfgmv1nLonaDIg4kS4sLMTgwYONnkZEI8uX5yswVDdtbEZGoKXYogXQKi4sAQoprCzMDTJtnQ490RdiKer5iN1uwFlNWhKVMwMjg4gRaVVVsWrVKtx7772orq42ejoM9EPAqDt4R8+YBeWaJKCqMqz6zyGFlYW52aet06H3/ggLkfPxEdvvnw50vBSIiebMwGZAxIj0vHnzsHz5ckydOtXoqTB16ISVkaSea8UyKRmoqABcriaLQyhhZUIsRZ++jHXvj7twvzfeW5Rf2rcvoz0t3RPvHZ/AmYE+WHUTMmJE+q677sLq1auRlJRk9FRMg28lORIk1EYG6mm2GhMDXLwYnqXoK8w65yLCUlTKy4C4OGBPAVxbNsO1ZTPUpUsAh0NovLdudxaKOiTBjttEZAioZTch1Qjj888/VwcOHGj0NMKicsMG9czoLLVywwbS45wZnaX+2G+Aemb0GJLxPecxRvh56I3rey6VGzaoJ3pdrZ64+pqwzk075qkbb1JP3jjc77jhnqN2rmdGZ6nHu3ZXj3e+TD3eO4nsGjgzOks93qWberxLt7Deo1CuVRHXM/W1qqp016vRRBm9SIiktNQBt1tt8t8nJiagpKQ87HlUv7YA6pEj+Pm1BXCkDAl7vPpw3ZEDddlSxE+8V8i8A0gZAqQMgQOAw2f8ulKZdY/e4YwbC6CkpNx7Lq477oYjZQhsTzwN1/x5cJ4pwakVq5t0HNcdOXDXjnH+fBXcCa0DP5d6zjEY3mslZQgw+g6oeStR3as3bMl9gaKvPIWcHA6ya8B1Rw7ww2Gg7Dyqe/UO+fPXXuOhXKsirufqXtdALfqqUXMNNvcAmvhZUhOKriQmJtT7s4hxd1gJ2eVJYzMySI+jxTV/HtT9/xdWr0Otq0b7WC3CFRFqunPYj+I+rhRv1/AWLTzx3hcuNG3MIIjyS4filxdyPUvsy2g1WKQJkBlW5NqyGSVZY0y3WeJb9L9eRNSQ0PqlFQXqivf8uo+E68vU7c5y8aLnh18faPLUgy0eIjY+Q1kMRVzPZtqAjDRYpE2Oe9lS1PxwiDR9W4v9/umekp8/nmhyxbq6mzZ+4r3e10IpzdloigoBxebZxKtLd66u8es+Eq6A6Ardb28EbDbP1yYSbPEQFSJnRHKUTMwe9RFxIj1o0CDs2bPH6GmYB00lOfeypVAPFocVYxzsohbRx0/PVaN1o4iLMXZ7wuTq0p07dACiooSGFmrnGvP3VxCztwgxf38lvLlLcEVYXUDNHvURcSJtJaSs4JpKctrEh6YQykXt2z+wqQS4aioqPBtuFRUABNa+yMr2vCe1NTZExRhrMzJFC50sVwRgnLVpRB0as8EiTYj2AqSqJBfVvbv3AtQmPjR1zGAXdUD/wCbgXrYUzgNf/2L1x8V5LNy4uCaPqYteeysB6c5SLLSkZKC8DOrx4/XOVcR1JeNa1YNTz4PDIk2IrEpyiXmrhFZf0/t7iuJBtpxcKNp059908Y4tilCK5zeJEDY2hUSOuFxA6U/1zlXEdaXXjVyGi4BTz4PDIk2IXsEaWY9d4YqD9u8pigfZ09LRKncs4LwI9fhxAOJaNwUtDiUiciSEjU0hkSPt2vv50PV+R4RfOqAbuYzGDzCXO8IItxCLtERkNt8MVxy0G5BUxYOq9+7zsxTJXBFaURYQORLSXMNcDOrzoVP4w3VLlZaUhLUJHQpmckcYsQnJIi0Zmf0Dwy7D6euKqL2RAAgVh/iJE/wsRbKwMh2/tGjrUxeixYDKdRZQqrQ2MibSXREyfeiyrX4WacmIshSDXpQCynDqbUCKFofYjIxAS1GAKyKYq0mY9RZkrmSLgYRmsdrImKYiLczu4EFLWv0s0pIRZSkGE0uq+FkKS0K3Ch5BU1e9m0tkSrjemMJcXHruGgnNYkV8FrLC7Mxi9TcWFmkDEJWk0dAYVPGzMuKBZW5aCdnY8+nOojemsO4s2jBCCc1iRXwWssLsPE9+vUyxAdkYWKQNQFSSRp2FRlZPGvJ86Eb01wMgZGNPm32pFSVhmZPaMMKsbI/1KNjlIbptmvc6P3okrKJcoRzHLBuQjYFF2sSEVKQoTKisoGBV8Kh8riT1QXy6swCBTwZCu4aXnPnF7yqxKYNZQuSMgtLvziJtYuoE1HW2lPQxUmtJicxwq3eBESRAweK9RQiQUl4GtGvv+VoHwSKjdXFQLaAUm6xUiUqRAuUTJ4s0EVKq0dUKqHruPHnquejMSb0qeNqfU8RLUwiQnl/at/qeqIQirYuDyi0UasZpY8ekWOwjBcqnDRZpImT6cn1rd0iLnxVQx7ihhgW6j/dNQM+XS7HxqeuXDjPaQPezJIj3DkUshVxXmqcLWfeILIOJyh/OIk2EzO4svrU7ZNUGFmHBBWtYICKCQZoFp+OXRkoqUFEBNaF1k4YMJctThDiEIpYU3VlkRfFwqVJGF6N2mqUdV4DPVduwQIuIin4AAirJiai5DfiLvadreDyw87Nfur7sKfAsMnsKmjS+3mdJ4YrQE2CK0Eu9BUZGFI/ZNz5ZpC2IrDrW4fpcta4aPYQsOkWFQHU1cPoUXPPnCYsx9rXQvO6N6mpv1xclK9vT67BVnPAGDL6viyh0pX2PZbjNAHlPfrJq5lDAIm1BRNxgwYRehM9Vr8wq1cYnbLWXusMhLMZYG9vt6fqS6K1DEj1jFpRrkgC3O8xCV4HpztoFgjwLVBAykqP0MLPLg0Xagoi4wYJd1KIyvKo2boRz9Eivi4DKgkP7Dp5vzpz2RGAICPHTCoxuxTohha50FkOfcSmyS6lcEUaJpZldHizSkpEZmhfODRbKRS1CHBwLFwFHjwCHD8H1+GwgKZnmZvLp9qLmrSQrdCW6DonX6nc6/buzECSy6JZ3DdIVprEYJZZmzkZkkZaMEWnWTUHWRR0/cYLHynW7gRoXUFRIclz7/dOBjpcCLVtCycoWF6GiaZ5LUodErzsLVbKMdoEJ0hWmsZAVurIwLNKSkRmaF67YybhxYjMyoFx2GRAdDVyoanK4WjDsaemI2bQVMZ/v/aUvI4Gl6B3XxxUh4qkGv+kC/Kb++tgi0F1ggnSFaSqhdP5hPLBIS8ZMj10yrX64XJ5vasPVRCwQemP4vSbAUrTfPx1K334eS70OrYAKKOIUs3odYlavE5pQpIfvuACEdVbXwvVBQodFmqmX+oRAtIVtT0uHkjseaNnSG64mYoHQGyMgbC5MSzGksDKCOiSUKeEyysbqxUybOUyOEhZpE6OtJCea+lKzKSzsgHA1AT7XYBl7utEYTSBYWJkoK1GW9akmtAbOlkJNaC21Pgi7PPRhkTYx0kqVahI/qMTh8z7D8Oj1E/B5n2HCQuS01hlFHRI9cdFavcJadTnKoR47RlaHBEBgpqSEVl0Auzzqg0XaxASrJCcCb5JGYiJpkSIAyK9qg5MJHbG+qg1ZFTxAvNvAlpPrCZHb+wWcf/pDvccNm6JCoKYGOFvqHbd67hw4h6T+Un1PAEpWNhAX5/lad1wJrbrMtF8jExZpE1N3UQMgjcKQVaRo1FXt0OnCz7j10L8BgKZrOMS7Dexp6YCj3GN9frrNO6ZoX65elIeat9JTxKm2+p4IomfM8izMRYXezjIyWnUx+rBIWwDHwkXkvryCzPF4NO0BFGSOByAu9dy3Ct4Nt9+Epw9vxKAfvhC2aRXKxp4QC+7q3n5fKXy5elEe2up7ovD9fIUVumKaRIMiXVVVFfI/xjjiJ04g9+Xlqx1x8tIuyFc7el6gqoInOMYYkFMvQmnZEuiQ6PlaC5WPVbf63p6CiO/OwjSNqIZ+OHjwYFy8eDGkgb7++mshE2IaT2xGBhwpQ7zf14Ww1UUwiGBU3074cNsB/G77OrjanBSysWfLyYV91XK47rj7lxd1ag57reomorUKAfHvkS0nF/9+fyPyOw9A5uSZuP7ODI8FSiBqvkWXlKxsoPgbb20PUcfTmzvFdcUEp0GRXrduHSZNmoRWrVrhv//7v2XNiWkCvjeQniiFy+Ae7ZHy7JueccsOCxFPe1o6Eu8ajZKScu9rtpxcT6q1zyZluOegN6bo98ielo78gvM4aW+F/I7JuK52XAphs+XkeuqcKApQVAglK9vjkyaOvqC4rpjgNOjuuPzyy7F48WKcPHkSP/74I1JTU+v9xxgLdelKIDDGWNTjL3WShu6YBGFlo65qh06VZ5H5zTbSFlEBFQiJuoZr4RA5Y2jQkgaAzp07Y/bs2dixYwduu+02CVNimkRSsuext9aXS2HpkD6++1poPucijKRk7HJEI7/vGNz2fSlSCITthttvQmr+OKinD0LNOwBXcl+ac0HgZ6F9UqBA1OfPbpPGEVJ0R2ZmJp577jnquVgWaZ1SJFhTJGitWopzKSpE/tXDcdIVhbX7T9LWvvCt/yygg0191F1XgJhwRd8xuedg5BB2CN758+fx7rvvipiLZZFxUVKIjrTykYILEulhy8lF5vmD6Ny6JUb17USa7uzrihDRwaY+KFxcZr1WrUyTRXrHjh2YMWMGhg4dimeffVbknCyHrD5uokVHahU8gYXy9bCnpWPo3L/gma4XkPLsDM9xiNKdff31v7TqUoUfh2KPwKzXqpUJ6pP25fjx41i9ejXWrFmDU6dOoVWrVsjOzsbYsWOp5mcJqHy5AQj2f4qI4AgFbbQF5XH/nb8d+ZffjMz87biu7PAv6c7JfYV+Rr5+V5FV8Hx9uRTXlaxrVda1ZQWCirTT6cRHH32EvLw87N69GzabDampqTh9+jTee+89XHXVVTLmyYSCYAtU2uIC/zA5AN50d9HkdxuMkxeB9e0G4/prhtaGstmEh5VRLDpWCoGTeW2ZnQbdHU888QRuuOEGPPbYY4iLi8MzzzyDzz77DIsXL4aiKIiKapQh3iDffPMN7rzzTvTr1w8jR45EYaEJN8AMRlaHZwpkPQJnHtqJThWluPXQTtJ0ZypXhJV8udwyKzQaFOkVK1agffv2eOKJJ/DMM8/g9ttvR5s2bYRPwul0Ytq0acjIyMDu3btx3333YcKECXA4HMKPZWWohE6Wb1pbH4SC6zOH4qnC9zHoyD6hFf1kpJ7L9OVylEfk0KBIv/POO7j22mvx7LPPYsiQIcjNzcXSpUtx6tQpoZMoKChAdXU1xo8fj+joaNx6663o2bMnNmzYIPQ49UFdPF8mFBs/spIYAuqDECBzIaMQuoLM8Xh02H343N7B9ALKyTGh0aBIp6am4qmnnsKOHTvw0ksvoU2bNnjhhRcwfPhwuN1ubNq0SYi1+91336FHjx5+r3Xv3h0HDx4Me+xQkFE8XxZUFpyM4jqj+nbCry+Jxai+nUiPI6XTNmiELl/tiJPxHZDfeQCtBSqh0D8XbQqNkELwYmJicMstt+CVV17BZ599hscffxwDBgzAyy+/jKFDh2L27NlhTaKyshItfaqHAUBsbKy06noyiuebHRmPv4N7tMfTbU7+EiInkIAGtMSdtgEaS3FU307o3LolMs8fpLVAzZwcZTEUVVXVpv7xjz/+iA8//BD5+flYv359kyexZMkSfPLJJ3jrrbe8r82ePRutWrUKewFoDoz64Fbv/9fe1vTPoSFKssag+utvALcbbee+iNiMDLLj1PxwCFHduyMxbxXJuPET74Vj4WLET7xX6HlUbdxIOO4ixE+cQPa+6x9T/LkwjSckkT537hxat24Nm+0Xw/urr75C586d0bZt27AnsX37djz66KPYtm2b97W7774bt912G+68886QxyktdcDtbvKag8TEBL+KbGZhwmd3ef+/6IYVJMdwbdnsrbymXNlLWIic9j13bdkM1/x5AAD7/dOFPQqLrhch81qpnjAO6pEjULp0Cft9N+s1Dph37qHMOzExod6fBXV3LF68GMOHD8f+/fv9Xn/++ecxbNgwISnhgwYNgqqqWLJkCaqrq7F+/XoUFxcjPZ19VSGjAnCBtoUWUeZcwHEkdaemgMItZFToHZWLS1bonVVC/BoU6fz8fLz00kuYNm0aevXq5fezV199FQ888ACef/55fPzxx2FNIiYmBm+++SY+/vhjpKam4vXXX8f8+fPRrl27sMZtLqy9bT1efysGr/3lJO1mkslLYsoQHarSpDJ6TGqhivCQFXpnlRC/BkV6yZIlePDBB72F/32Jj4/HpEmTMG3aNCxeHH5UxJVXXonly5dj3759WLduHQYPHhz2mM0JGeFMsiw6qthlGaJDFSKn/XzNHCInK/TOKiF+DYr0Dz/8gOHDhzc4wM0334zvvvtO6KSYxmNPS/emH8vsGh7JBAgZVViZz7hUIXLahUtWISSKbvQyrtW641ghxK9BkY6NjUVlZWWDA7hcLrRo0ULopJimYeYyk1J8uVTuGp9xR/XthM72GmR+vVX4YqDtYCNLgCiuK1muCCv4pRsU6ZSUFHz44YcNDrBmzRr07t1b6KSYJiIpAUFkgfmSrDHSfLm2nFzs6jMMj14/CTu/LxV2HN/FIPXwXjy1fzkGlX4nfDEwSthIrHYJ1ypgDb90gyI9ZcoULF++HK+++mqARV1RUYFXXnkF7777LqZMmUI6SStgpe4sIgvM1/xwiLwvY52AAsCGodn4UWmJtftPCjuG3mJA4rvXETaK68o1fx7U/f/3SygkhdVusmvVSBoU6d69e+Mf//gH3n33XQwaNAijRo3C2LFjMXLkSAwaNAgrV67E3Llzce2118qar2kx80aPFmE3bVIy1Koqb19Gisd3rYCSpZ77CKj3mCVn4Hp8tjgB1RE2kuvK4QBqajxfiTDbJrSRBI2THjZsGP75z3/iueeew29/+1v06NEDN910E/7xj39g8+bNGDFihIx5mh6ZGz2iO3iTUVQIJVZC6rGPgA7u0R5/HZ2EwT3aiz2GXguwigrA5RImoHrCRnJdxccDNhtQdp43oSOABkW6qqoKjz76KG688Ub89a9/RXl5OR566CHMmDEDaWlpvGHYCMy2osuy/KO6d6d/FNUIqPNPf4BzQBKcf/qDsEMEiGVRIRATA1y4APXYMSFipydsFOGK9vunA61aAdExlnjyMzsNivS8efOwdetWTJw4EePHj8enn36KRx55RNbcGAORZfkn5q3yExiqKA+/c/l0G+B2A9v+RRZWZsvJBaKiAMUGnC0VJ3ZEG27ari+e7FI3V8GLABoU6Y8//hhz5szB5MmTMXHiRPzjH//Ali1bUF1dLWt+jEFobyAq90fVxo3kCScBYnB1bTSS3U4WVlbX9QWJiYDdLk7sigoBxebpy2jGcEUNVgiRo6ZBkT59+jS6d+/u/b5Pnz5QVRWlpeLCl5jIhTrdGQAcCxcFJpw4yoW5COrwPRelZUsgOhpwuaB+962w42gtdntaOpTLLgMSWocldr5zt+XkeixcRSEPV5ThirBCiBw1DYq0y+WC3W73fq8oCqKjo9mSbiZom6lS3LTxEydgV/eBeKRrBj5b8y+PmNXUiHURoPZcDh70VPJLSgZcLs8Pzp8Xdhy9x3cRUQy6rginE+rx42RuIVmuCPZLByekov9M80QbY0xx08ZmZCC/2/U42bIt1n1z1nPMdu09/lzRnVNqLVAUFULJHe85RnS00OPo9ToMN4pB1xXhcgGlP5G6hWQ1IGa/dMMEbfe9evVqv+JKLpcLH374YUAd6bFjx4qfHVMvvvWRcddokmPY09IDHkdF1mSuIzP2PPLPVuLWdtWwp42ptXqrPH7X5L5CjlU3hu/8q4sKoR45Erbf1fez0Fq9ALyvN9Va1H4OtpxcuGoXTkoLVO9cqBBd79tKNCjSnTt3xvLly/1e69ChA1avXu33mqIoLNKS8RNPIpEGPI1P135zFqOuaofUZUtIbtrrvvwEg2qL2gMTPCJU22CAUiBsObme7LpaC7Wpx/H7LJKSgeJv/Kxze1q6kMSfunGFjBcCot6fUJC5IJiNBkX6X//6l6x5MI0kXOssVDwdvBOQr8biOqJjas/FnpYOd+F+qHkrw3JFaK0zP780Ai3UpoqD30KWv4S8iFMd1NanqPcnFGRdz2aEfdImRc+XR+FDHNW3EzqrF/C77SsBQF7pSgEhYFpXjW5khIC4Y89C1gX5akeydGe9DTYpkRFE0TZa2DddPyzSFoLiph3coz2e+vebGPTlJ35+aerSlSLETi8kzv7kM1Cu7OWfGRjmYuBbD0Rq+y8ZAkoUbcOEDou0haAKZ1ITWgNnSz1fIa44jm+pUu2YIsROT9j0iueHey4B9UAkWZ8yBNSWkwv8pgvwG3OEyVkxOYZF2uT4Ch1Zx4s9BZ4GtHsKAIgrjuNbqlR3TIIUaIoQuQAkWZ+2nFwgLs6TlFO7gIrGnpaOmNXrYL9/etjXlZn7MhoJi7TJ8RW6uu9FX6RKVjYQF+f5WosIqz2gwJJWlAWlJgfLnBT9BOKN9VYU0sxJe1r6L0k5tQtouGPWh4jrykrlemXCIm12fGoyAzQXafSMWZ46FEWFfhZouBs99rR0fPXUy3jkfCdPpxSNKItyq2hjjGGzQf2qCNVz5wg7F+AXsQMA5fLLPU8fFJmTPkKnZGUDLVoAreKEZDTWh6hFmbqGtBU3IFmkzU5RIVRF8RbcobpI9W5kEY+v739+FCfOVWHt/pP6tS8EuCIKMsfj0bQHUJA53vO+VFYAFy96QvwEnkvAYhAXD9TUQP3uO7L6INEzZgGdfw2cLfV2UmnKmMHEU9SizDWkGw+LtMmx5eRCUVVAsXkvfCnlPiHm8fX2FqXofPoIMpUz9UcwCAyRAwCkpHqsXHuU0Op72toXcNV4fnD+XFjjUjeglSmeVnRHUMMiTYis2gdt574IJCZ6LSFRvr+g4iAgiqH/6kX4y5aXPUkgtcd0jh4J5+iRwkpm+sZ6u7ZshlJe5qnbUeHwC/kLVzy075GSlS2kPkiwz9N+/3Qofft5ivU3EdO1XmsCZo38YJEmRNZOc2xGBkkj1KDzFxDFED9xgp84uJctBY4eAY4eEXYu2lhvvSJOFOIRPWMWlL79gDaXhLf5GeRpQpQrwuriadbIDxZpQuoTGGp3hKjH12DWlYgY2tiMjMC45fYdvIXyRZ6Lb0U/5fLLgfgEoZEj9R03bAs1hKcJEdeU3hhWCpszq6ulWYs09QVYn8BQXJQBlpAAX26w0pV1MbQxq9eFZYFp3SraQvmiXBEynjb03qNwY9dtObn+NbdDmEdT8K1tQtkpR4uMqA/AvJEfzVqkjYrblLKiC4ox3vl9KR5eXYSd35fW+36Fu9gFSwkXdXPpPm2UnPETpXDG1DuX+l5rDPa0dL+a26HMoylQ1TYJBkd9NEyzFmkZYhlKajIFoqyTtftP1hsiV0e4IkSREg7oW7UBrhVNZExj0RuTYlHOjD2PTuVncGvsee9roqM+qGqbhIJZXREyaNYibdbHn1AQJXTa4kF671e4NxhVSniw2G5vK6owu2L7Pm1QLcrXffkJnto6H4P+9w2pTXvZFWE8zVqkzU6d4FRt3Kj7c5LiQQhuoTYJgpTwkGK7BRzH92mDCj2rX1YGnyxXhFlD5KhhkTYIkRlujoWLdX9O5TYg8eUTpITrLR5a4RZxHG0cNiBecPSsfmkCKsEvDYi5rqwo9CzSOpgl7KhOcOIn3hv0d8Lx9elt7In2H1L5pX1dEUCgcIs4zuAe7fGX9S9g0PYPvKnZJGn0ela/DAE1kV/arLHQDcEirYNZqnXVCU5sRkbQ3wnHFaFXU0NGarKI9ygUV4SI4+zqcAUe/Y+Z2NXhinrHFL3BCkCIgIYS620Wv7QVNyBZpHUwKuqjKfjWk6ZCVqsuvcUASclwPT7bW7GusfhufNZB4VNfP+h2nGzbCfmXX1tvtAXFBqsIAQ0WWglAmFtFRm6C1TInWaR1MNNOs7aeNBW6fumDxWHFGANA1caNDYaRqXkrgYoKv4p1jUFv45PCFfH7G3ujs7McmQc/bTCxheKpJlwBrU/otVX9RBguMp5SjfJLU50bi7TJseXkQrnkEvJHUd2EkzBjjAHAsXBRwxd2SiqgKJ6vTUTrl6ZwRQzu0R5PtTyMQd/u8hacklroKgy/dH1Cr03+EWK4SPChU5XVDQbVEziLNCGyquDZ2/nfYLJcEZ7EhyvDuii1BZa0KOVlQLv2nq9NROuXpnBFAAgoOEWZei66UqCeeAYrC9AkNHOVWlZXk/IuGqoncBZpQmTtNOtVkiOvDVLPa41FW2BJi6gQORl+aW11PapCVxSVAkMRelERSb5zlXWt6qa8mwQWaUJk7TTrVZKTtcNtVJGqxjC4R3s83eYkBk4dA+d1A1A9dw6ZKyKguh5BoStt9UFKv7T2dyK10FUoxw1IeTcJESfSS5YswbRp04yehhCM2oAUUXktVGQkIIiKn8WZ08CFC1DzVpK5IgIaIRA009WtPkjklxZdHwSor9BVCakrAjBXQIAvESPSFRUVeP755/Hcc88ZPRVLILNGL0UImGhx8MyzTe03dgBiwsoCFhBJfukABKXRa+cqrT6IJFeEGTMSI0akp0yZghMnTuDOO+80eiqWQJbLQ9ijtqaDt2hxsKelQ7niCo+/uLJCWFiZDFcEgKCWMoUrQtS4eugXulJNkXouG2ki7XQ6UVJSEvDvp59+AgC8+OKLmDdvHjp06CBrSpZG5qOdiCQNbQdvstTzug7e334LAMISihpyRQiLHNFYytrjCnFxaRYDKtcZRaGrUDBjRqKiqqoq40C7du3CuHHjAl632+04cOCA9/uXX34ZX3/9NV599VUZ02IihJIJE+D8eDPQpg3azXmhwVT3cPjxqt5Qy8sBADHXXYfEvFVhj1mSNQY1PxxCVPfu3vGqNm6EY+EixE+cIORcqjZuRNkLc6BCRZuHHkRsRkbAcfXmIeJcTg+/CTWHDiOqWzdcunVL2OdSdz6OhYsRP/FexGZk6J4f4yFK1oEGDRqE4uJi0mOUljrgdjd9zUlMTEBJSbnAGcmBet51oVJ1j+8iqZt7dclZT19DhwM/v7YAjpQhQo/jZfQdwHvvAIqC6l69m/y++b7nrjty4J4/D84zJTi1YjXsaemofm0B1CNHvOcS9nuYMgSuBM2Yd+RAXbYUrjvuRklJecD3weath9651JwrA6qrUXPuvLjrLGUIXOer8PNrC3D+fBXsaekB59fYuUcqocw7MTGh3p9FjE+aiVxkFZxCqzjA5YKa0FrYuNqNIr0O3uFuJoXiyxWSTKHjitBW9BPRnSUg2iI+3uPLj49v8rh6yKiuaAVYpJmgyCouD7fL882eAgBia25r052FJ1QE8eUKiWCQ6Lf1nav9/umezVCAPIvVjCFy1LBImxxZVfC0liJFKJOSlQ20aAG0ihMmnnrWWcD5iKgnoSOgvvMXkUxRn6VJ1mCguhrqsWMAxFXB0x5HRnVFsxNxIv3AAw/wpmEjkFUFT0Y6b/SMWVCuSQLcbrIQuTr8xiaKMRZtKdb392Sdcnzivc3UncVqRJxIM43DlpOLqO7dpcdDU/kPCzLH49G0B1CQOZ708dfXHYGkZJIYYwr0LE01oTVwtlSoL18b780hcsYhLbqDocGelo7Eu0bL2fVOSgaKv/EWD6IQz3y1I05emoB8NRY3gDaypM5qU+CJlw6bpGTgqyL/UqW1NbcBCJm/d8z/eQiuX1/m8RfvKfAkgtT68kWg9/m65s/za29GAdV1ZWbYkrYIUnx5EspMaivWiSoxqTdX4VZbUSHgrAbOnIZr/jxhNbd98Y5ZXeOtgqf15YuEojsL0zhYpC2CrDC5XX2G4dHrJ2Hn96UkleS0nVRE1XXQq8ks2kK35eQC9tpbqqLCJ93ZLcyXW7cBiQ4dvCVRtb58kVB0Z2EaB4u0RZBxA9nT0rFhaDZ+VFpi7f6TUooHCSsxqalOR1U4CJf/Bru6DsCj147DZ2v+ReLLpSqJqgdJdxamUbBIWwSSLho6ZCpn0Pn0EWQqZ8iK2gMEJTJ1qtNRLGr2+6cjP/lmnGxzKdZ9c5Ysxly3+h7Bxp6s64qpHxZpC+Hby4/K/ZGavwR/2TAH1774394kDYowOdHzt+XkAu07eFLPazc+KQoH2dPSkdmuGp3KS3Br7HmyqA/d6nsSiudziJx8WKQthG8vPypLMSAbTdAjcPXcOXAOSfWWKhUtOva0dCiXXYZdV16HRy50JV3IrvvyEzz170W47stPPC9oGwEQICsE0Gx+aStY/izSFmJU307orF7A77Z7yn1S+A+pav+qeSuBigpvqVIK0bHl5CK/dxpOXvIroT51veM01AiADAkJJyL90jIE1AqWP4u0JHxdEVQM7tEeT/37TQz68hPai5LA/6lkZQNxcZ6vdQgWHXtaOm7LHo7OMap3IZPmivBpUCsSP6EjaNVFiayIJDNZ/nqwSEvC1xVBCoE1pb1pKSzQ6Bmz8MXbG/FYl//4ZSEjWAwCFjIi6zNog1pBUITISbM+TWb5GwWLtCS0SRpkFBUCigI1b6UwAdXetFT+T+1CJiP1nCoqQq8MJ7VrRWSjWBmbkLJSzc0Oi7QktEkaVFBluWnFkkJAfX3qIrtTa/GknndBvtqRVDx9xxW1sGmfaijeI96EjCxYpCUiw9dHleUWvegdFHQd4PWrU4iDnk+dOvWcMkSOoqmrmbrAB4MqDNJqsEhLRJqvj+gxUoZf3c8VAZr3TC/1nCpcUXRRe1muCFnWtBWiL6hhkZaIrMc7quPI8Kv7uiIAmHpzycyuCABS3nt2eQSHS5XW4tqyGSWrlsF1R47pyzBSHWdwj/Z+PnWKIkWj+nbCh9sO4Hfb18HV5qTpN5co3qOCzPFYW3Qamcd243rCsqGeTWibZxM6uS9ZXW8zR17IgC3pWmR1ODE71KnnWr+02S0tivcoX+2Ik/EdkN95gPDr1XcPQFQFQiY8WKRrkdXhxOzISD1HUjJQXgb1+HEA4jInjUgRJouCsdcg8+utwl0RovsyNgeorysW6VrsaelIzFvFj15B0FbBI0kUKCoEXC6g9CehFpyeVUtxg/k+bVCkUace3oun9i/HoB+/EhoPD1irg7dVMidZpC2EjNTz1Pwl+MuWl5Gav4TsGFRp1HpWrYgbTCsGVFEwvl1qkJRM4oqQ1cHbSnU7qF1yLNIWQkaInKz4WYo0aj0Boohd9n3aEImvjxhFhdJcERRiZ6W6HdRPGyzSFkJGiJzMbDSKxUD7tEERu6ytuS0KrY9YliuCQuxkdRIyq6vGFxZpCyEr9VxW7LKM+iAi0M7VlpOLXZ2vwaM3/QH/zt8u7Dh1x6IWHlmp51YQUBmwSFsMI7qGUyGjPogofK1pAMi/YQxOJiRifbfBwo5RB/VnLMuXa4WC/DJgkbYYsnx9ZnFFAIHdyClqbmut6cxDO9GpohS3Htop7Bh1kEcTWKwVl9kXAxZpiyHL12cWVwQQKAba+iCi8H3vr88ciqcK38egI/uER0WoP/8MtG5N9hmbfd9Bi9nrg7BIWwwzbyZRbXzacnIBux1q0ZeonjvHk7F3ya+w7puzwjf26qq6ATRdX9zLlgJlZVDatqXt4M19GSMGFmmmSVAlaVBsfNrT0oEKB+B0Qs1b6fFLnzuFWw9sES4OflYbwQarnvVJYilK6svIUR7BYZFm6sWIjC2qhBzfHoqDe7THUy0PY9ChL8RHqPgKM8EGK1Wdai2y+jKaXUBlwCLN1IsRve7W7j+JE6fP4cOVW4UuDtEzZsH+5DNAUaHQpq0B+IxL5nPVWOhkJVEl9GVkgsMizdSLNF+ej7DJckXIaJsFiPNLU3QFD4asZghMw7BIM/WiZ6FRuCN8hS318F4prghZbbMoOnhTRkXIcEVoxzV7iBw1LNJMg8goHmRPS8euLv3xaPKdngw9Ca4IgNCi0ywGojp4+3YFp4qK0LoirFQIyaywSDMNor2BqMLk1iel42SbS7E+KZ30MVvbwZus1CrBhqHfXIlC5LTvvZUKIZkVFmmmQbQ3EFV9kN/f2BudW7fErUUewaF6zJbSH5CwtkmdZatu/4QkRE67GMiqesgRHvXDIm1i6m7Yqo0byY4h6wbSS9cmS9IgEFBZG3t1li0AoH0HwG4nD5GT1viW0YVF2sTU3bCOhYtJj6PdLCTzU2oEVNSjtgwBlbWxV/ceKUOHQbnsMiChNXmIHLsjjIVF2sTU3TzxE+/1vkYhoNrYZTI/JdHGnlZAKQSnIHM8Hh12Hz63dwBAkxIOQEoctl4Lrbp0d47AkE/EiPTbb7+NtLQ0DBw4EGPGjMGePXuMnlLEU+eKiM3I8L5GIaDa2GWzbexpIyMo3DcBHbyJ3Coyojz03iOOwDCOiBDpTZs2YeHChXjjjTewe/du3H333ZgyZQrOnj1r9NRMB4WAetOoD34uvIO3L7JEh+JpI6CDN5FbRW9jT4YrwmwuDyvFXkeESJeUlOC+++5Dz549YbPZkJWVBbvdjuLiYqOnZgqqNm6kr4VA1ME7ACIL1NevTmEVDu7R3tPBu/S7X1wRPpX3qJDliqCo7U2JlSx/aSLtdDpRUlIS8O+nn37C2LFjMXbsWO/v7t69G5WVlbjiiitkTc/UOBYuklPon6jgjh9EFqhvEo4MX6628p5ofMXOTIJklQ7eMomSdaB9+/Zh3LhxAa/b7XYcOHDA+/3BgwcxY8YM/PGPf0SHDh0adYz27ePDnmdiYkLYY8jm3ID+qP7qAFoNSsElVPO/azRKVi1DzQ+HEFV8QOj75DvWuUEpqDxYLPxcsuPPYeW3p5B92a/wK59zsa9ajsS7RjdpzID34K7Rnn+1nBv3n6hc+h5a5Y4V/rmUrFoG9dhR2FctR5upk+FYuBjxE+9FbAjHqe+zq9q4EY6FixA/cYLfPodIqqZORtkLc6CWn0f8nh2NPk7I153mszCacO4XaSI9aNCgoO6LrVu34qGHHsKkSZMwadKkRh+jtNQBt1tt6hSRmJiAkpLyJv+9YezdB7VFS1Tu2oNqwvm77siBe/48OM+U4NSK1UJcKtr3vHrXHpJzSXp/Aa45cgTKwS4ouXkIqntdA7XoK1T36t2kz7y+a6XOsrXl5MI+eTpsPa5G5bKluCDo/fIe544cqMuWwnXH3XCkDAFShsABwBHkXBq6xp3PPg8cPYKzZ35CTMoQYXP1I2UIXAkLoB45gp9fW+CZe4iY9f4MZd4NiXhE+KQBT3THzJkz8fTTT2Py5MlGT8dUxE+cIKUNkYzEBlmRIzLipfW+F4WZs/Ss5IqQQUSI9IYNGzB37lwsWbIEt9xyi9HTEQ71ZklsRoa0rDDqG4yy8pq2Oh11qVLXls3C3i8ZG272+6cDv+niPR7ZcUy8wBhBRIj0m2++CafTifHjx6N///7ef1u3bjV6akIwc5Eaqg7eDSEjo9Er2iVn4Hp8trBjaRcDUe+Xe9lSqAeLhc5Vi8wUcCuFyFETESK9Zs0aHDhwAPv27fP7N3z4cKOnJgbCgjt1UIknVQfvhpCZ0QhVBRSb0GORtbMimGsAEq5VwFohctREhEgbiZQVXVInDQqoSpM2hCxXhD0tHfYnn4Fy5ZVhCar2GqKIXbanpWP3/3sOj978R3xu70B3vZq864sVafYibWZXhAz0SpNSLWx14wI0tS/0HudFPIHoXUMU11VA6jkBsq5V9kuHTrMXaW4pHxytX5pqYZNRCInMFaGx/CmOE5B6ToDZr1UrIi1OOlKxp6XzBRkEX7/04B7tvY/yogVUTWgNnC2FmtRH6OfiF7tM8Hnb09Lx7/ztyG9zJTLzt2No7TFEH2dwj/ZI2b8caukRoKha6NhM49BeU5Q0e0uaCY7WL01mbe0p8GyO7SkQOqwMl5Zv+y9KZHRKMTtW68vIIs0EhapllhYlKxuIiQHi4knrI1Pg2/7L7AlFgLlD5Ky2z8QizYSEjJs2esYs4NeXAaU/wTV/nrBxZfhZ9dp/kSEhTM7MIXJW68vIIs2EhJlvWiBw85MCaa4ICWFyZo5IslpfRhZpJiSkhWbdPx1K336eFGWBfLjtAE4cOYkPtx0I/stNRJY4WM1SpMDMi4wWFmkmJOpuWgCkbg8qcbi1aDM6nT+NW4tofayyQjrNbClKqUNiob6MLNJMozCr2+P6zKHIPH8Q65PSSV0esixQisXAal1TzHqtamGRtgiybjCzPkba09KxYWg2flRaktchofgs9FLPRS8G0kRNUn0Qs16rWlikLYKsG8zMvkpZdUgoPgu9MUUvBrL6MqKoEFAUqHkruSRqCLBIWwSzWw0yoi9kxXtTWIp6m4WiFwPqvox1SKvoZxFYpC2C2a0GWSVRq+fOgXNIqlBLMcCiJQiR09ssJFmYU1IBRfF8JUJU9cHmAos0ExGM6tsJndUL+N122kdgNW8lUFEh1FLUWrRkIXIaC51iYVbKy4B27T1fCTG7USETFmkmKLJcEVIy9ggsRa1FSxYiJymJRVZtEDOnnsuERZoJirTuLBJ2/SksRT2rUFZJVNF4z+HoEaGp+XpYJUSOGhZpJijSurOYON1ZRoic2ZNYtJh9s1sWLNJMUGRFRZg53VlqjLGjHOqxY3TvEXcNjyhYpJmQkJXKa1ZLUWpxJWc1cOY0mTvCzJ+DFWGRZkLC7Nlo1IuM1OJKMdGAjfbWZVdE5MAizYSE2ctwSllkJGx82tPSYX/meSj9BwivFKg9DrsiIgMWaQtQtXGjZVwRZmpAG4CEjU9AnoByiFxkwCJtARwLF0lxRZi5s7oMYbNajDGHyEUGLNIWIH7iBDkF+U3+CCzDLw0AOHwYrv95yPQCyn7pyIBF2gLEZmSYWjxlIc0ydLuA6mruzsIIgUWaaT7I2Ni7fzrQ8VJP13PiDUQOk2sesEgzEYGUTaqiQkCxkdYxtqelQ7n8ciA+gXwDkd0RzQMWaSYikOVjheoGFMX0rgiA3RHNBRZpJiKQFTniqWPcixvFMqYhyugJMAxQK6ASLEJZx6nrVM2uCCZc2JJmIgZZyROy6pCwK4IRAYs0EzG4ly2FerAYrsdnmz7GmGFEwSLNRAyyGpRyVARjJtgnzUQMda4Bal+uLL80w4iARZqJKFhAGcYfdncwDMNEMCzSDMMwEUzEiPSCBQtw4403on///hgzZgz27Nlj9JQYhmEMJyJE+qOPPsJ7772Ht99+G3v37kVWVhamTZsGl8tl9NQYhmEMJSI2Dm+55RYMHToUcXFxuHDhAs6fP482bdrARtzHjWEYJtKRJtJOpxPnz58PeF1RFHTo0AFxcXH49NNPMWXKFNjtdvz973+HoiiypscwDBORKKqqqjIOtGvXLowbNy7gdbvdjgMHDgDwCLmiKNiwYQMeeeQRfPDBB+jRo4eM6TEMw0Qk0kS6sdx99924+eab8V//9V8h/01pqQNud9NPJzExASUl5U3+e6Mw67wB886d5y0fs849lHknJibU+7OIcPq++eabePzxx/1eczqdSEiof+IMwzDNgYgQ6YEDB+LDDz/Enj17UFNTg/fffx8nT57ETTfdZPTUGIZhDCUiojsGDBiAp59+GrNnz0ZpaSmuvvpqvPXWW2jXrl2jxrHZwt9oFDGGEZh13oB5587zlo9Z5x7OvCPWJ80wDMNEiLuDYRiG0YdFmmEYJoJhkWYYholgWKQZhmEiGBZphmGYCIZFmmEYJoJhkWYYholgWKQZhmEiGBZphmGYCIZFWsPbb7+NtLQ0DBw40JRtvJYsWYJp06YZPY0G+eabb3DnnXeiX79+GDlyJAoLC42eUqMpLCzE4MGDjZ5GyHz22WcYPXo0BgwYgPT0dKxYscLoKYXM1q1bMXLkSPTv3x8jRoww1dwBoKysDDfeeCNWr17dtAFUxsvHH3+sDhkyRP32229Vl8ul/u///q86YMAAtbS01OipBcXhcKjPPfec2qtXL3Xq1KlGT6deLl68qA4fPlx96623VKfTqebn56spKSlqeXm50VMLCbfbra5cuVIdOHCgOnDgQKOnExI//vij2r9/f3XTpk2qy+VS9+/fr1577bXqp59+avTUgnL69Gk1KSlJ3bZtm6qqqlpUVKT26dNHLSoqMnhmofOnP/1Jveqqq9S8vLwm/T1b0j6UlJTgvvvuQ8+ePWGz2ZCVlQW73Y7i4mKjpxaUKVOm4MSJE7jzzjuNnkqDFBQUoLq6GuPHj0d0dDRuvfVW9OzZExs2bDB6aiExb948LF++HFOnTjV6KiFz4sQJZGZmIj09HTabDcnJyUhNTcXevXuNnlpQOnbsiJ07d2LYsGFwu904d+4c7HY74uLijJ5aSKxZswYOhwNXXnllk8eIiCp4MmmojdfYsWP9Xtu9ezcqKytxxRVXyJpevQRrP/biiy/i0ksvxcsvv4ySkhIDZhga3333XUC3ne7du+PgwYMGzahx3HXXXfjjH/+IXbt2GT2VkElJSUFKSor3+3PnzmHPnj34/e9/b+CsQic+Ph5VVVVISUlBTU0NJk2ahK5duxo9raAcO3YMr7zyClasWIGJEyc2eZxmJ9L79u0L2sYLAA4ePIgZM2bgj3/8Izp06CBziroEm/ell15qwKwaT2VlJVq2bOn3WmxsLKqqqgyaUeMwy/tcH+Xl5Zg6dSr69u2LtLQ0o6cTMi1atMC+fftQXFyMyZMno0uXLrjjjjuMnla9uFwuPPjgg/jzn/+MxMTEsMZqdiI9aNCgoO6LrVu34qGHHsKkSZMwadIkSTNrmFDmbQZatWqFixcv+r1WVVWFVq1aGTSj5sOhQ4cwbdo09OzZE3PmzIHNZh5vp81mQ0xMDPr06YPs7Gxs2bIlokX61VdfRbdu3XDzzTeHPZZ5PiVJvP3225g5cyaefvppTJ482ejpWI4ePXrg0KFDfq/98MMP6Nmzp0Ezah7s3r0b2dnZGDFiBObNm4cWLVoYPaWQKCgowOjRo/1eczqdaN26tUEzCo3169fj448/9rqaDh48iCeffBJPPPFEo8dqdpZ0Q2zYsAFz587F22+/jb59+xo9HUsyaNAgqKqKJUuWYOzYsdi0aROKi4uRnp5u9NQsy9GjRzFlyhTMmDEDubm5Rk+nUVx99dU4ffo03nrrLYwbNw779+9HXl4eXnnlFaOn1iAfffSR3/e///3vcc899wQsOKHAlrQPb775JpxOJ8aPH4/+/ft7/23dutXoqVmGmJgYvPnmm/j444+RmpqK119/HfPnz290qzQmdN577z1UVFTgpZde8ruu//a3vxk9taAkJCRgwYIF2LRpE1JTU/HYY4/h6aefRmpqqtFTkwa3z2IYholg2JJmGIaJYFikGYZhIhgWaYZhmAiGRZphGCaCYZFmGIaJYFikGYZhIhgWacbS3HTTTejVq5f3X+/evTF8+HA8//zzqKys9P7egQMHMH36dFx//fXo378/srKysG7dOt0xCwoK0KtXLzz22GO6P//++++Rm5uLfv364T/+4z+wadMmknNjmgcs0ozlmTlzJnbs2IEdO3Zg69atePbZZ7F27Vo888wzAIBPPvkEd999Nzp16oSFCxdizZo1GDlyJB5++GEsWrQoYLy1a9eia9euWL9+PS5cuOD3s4sXL2LixIno1q0b8vLykJ2djZkzZ+Lrr7+Wcq6MBRFZ3JphIo3hw4erS5cuDXj99ddfVwcOHKg6HA71uuuuU19++eWA33njjTfU5ORkv6YPFy9eVFNSUtTVq1erffr0UdesWeP3N2vWrFGvv/56tbq62vvaH/7wB/Xhhx8Wd1JMs4ItaaZZYrfbERMTg61bt6K8vBzjx48P+J2cnBwsXrzYr5jP1q1b4XA4MGzYMNxwww3Iy8vz+5u9e/eiX79+iIr6pSzOtddeiy+++ILsXBhrwyLNNCvcbjcKCwvx7rvvYsSIEfj666/RrVs3xMfHB/xufHw8Bg4c6Ce4a9euxYABA9CuXTukp6dj9+7dOHbsmPfnZ86cCag5nZiYiNOnT9OdFGNpWKQZy/Pcc895iwr16dMHOTk5SE5OxqxZs1BWVqYr0HqcP38en3zyibdi30033QS73e5nTVdVVSEmJsbv72JiYuB0OsWdENOs4FKljOWZMmUKRo0aBQCIjo5Ghw4dvELatm1blJWVhTTOxo0bUV1d7S3kfskllyA1NRVr1qzB9OnTYbPZ0LJlywBBdjqdAd1oGCZU2JJmLE/btm3RpUsXdOnSBZ07d/azdJOTk3Ho0CE4HI6AvysvL0dubi6+/PJLAB5XBwCMGDECvXv3Ru/evbFz506cOnUKO3bsAOBpr6XtMVlSUmL6tluMcbBIM82aIUOGoF27dnjrrbcCfrZixQrs27cPv/71r3HixAns3bsXDzzwAD744APvv9WrVyMuLs7r8hgwYAD27duHmpoa7zi7d+9G//79pZ0TYy3Y3cE0a1q2bIknnngCf/rTn1BRUYHbb78dUVFR2LRpE+bPn48HH3wQ7dq1w2uvvYYWLVpg3LhxAa2bbr/9drz//vv4+eefcfPNN+Oll17CI488gokTJ2L79u3Ytm0bVq1aZdAZMmaHLWmm2TNixAgsWrQI3377Le655x6MGTMG//rXv/DCCy/gnnvuAQCsW7cOv/vd73R7640dOxY1NTVYt24dWrVqhQULFuDw4cO4/fbbsXLlSvz973/HVVddJfu0GIvAnVkYhmEiGLakGYZhIhgWaYZhmAiGRZphGCaCYZFmGIaJYFikGYZhIhgWaYZhmAiGRZphGCaCYZFmGIaJYFikGYZhIpj/D/FMFUs6Eyq9AAAAAElFTkSuQmCC\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: 329, 4\n",
      "LR fn, tp: 9, 4\n",
      "LR f1 score: 0.381\n",
      "LR cohens kappa score: 0.363\n",
      "LR average precision score: 0.516\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 332, 1\n",
      "GB fn, tp: 1, 12\n",
      "GB f1 score: 0.923\n",
      "GB cohens kappa score: 0.920\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 333, 0\n",
      "KNN fn, tp: 10, 3\n",
      "KNN f1 score: 0.375\n",
      "KNN cohens kappa score: 0.366\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 1278 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: 327, 6\n",
      "LR fn, tp: 12, 1\n",
      "LR f1 score: 0.100\n",
      "LR cohens kappa score: 0.076\n",
      "LR average precision score: 0.305\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 332, 1\n",
      "GB fn, tp: 0, 13\n",
      "GB f1 score: 0.963\n",
      "GB cohens kappa score: 0.961\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 333, 0\n",
      "KNN fn, tp: 11, 2\n",
      "KNN f1 score: 0.267\n",
      "KNN cohens kappa score: 0.259\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 1278 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: 330, 3\n",
      "LR fn, tp: 12, 1\n",
      "LR f1 score: 0.118\n",
      "LR cohens kappa score: 0.102\n",
      "LR average precision score: 0.273\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 332, 1\n",
      "GB fn, tp: 0, 13\n",
      "GB f1 score: 0.963\n",
      "GB cohens kappa score: 0.961\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 333, 0\n",
      "KNN fn, tp: 12, 1\n",
      "KNN f1 score: 0.143\n",
      "KNN cohens kappa score: 0.138\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 1280 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: 325, 6\n",
      "LR fn, tp: 10, 3\n",
      "LR f1 score: 0.273\n",
      "LR cohens kappa score: 0.250\n",
      "LR average precision score: 0.332\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 331, 0\n",
      "GB fn, tp: 0, 13\n",
      "GB f1 score: 1.000\n",
      "GB cohens kappa score: 1.000\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 331, 0\n",
      "KNN fn, tp: 7, 6\n",
      "KNN f1 score: 0.632\n",
      "KNN cohens kappa score: 0.623\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 1278 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: 318, 15\n",
      "LR fn, tp: 10, 3\n",
      "LR f1 score: 0.194\n",
      "LR cohens kappa score: 0.157\n",
      "LR average precision score: 0.203\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 333, 0\n",
      "GB fn, tp: 0, 13\n",
      "GB f1 score: 1.000\n",
      "GB cohens kappa score: 1.000\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 333, 0\n",
      "KNN fn, tp: 9, 4\n",
      "KNN f1 score: 0.471\n",
      "KNN cohens kappa score: 0.461\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 1278 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: 328, 5\n",
      "LR fn, tp: 11, 2\n",
      "LR f1 score: 0.200\n",
      "LR cohens kappa score: 0.178\n",
      "LR average precision score: 0.329\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 333, 0\n",
      "GB fn, tp: 0, 13\n",
      "GB f1 score: 1.000\n",
      "GB cohens kappa score: 1.000\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 333, 0\n",
      "KNN fn, tp: 9, 4\n",
      "KNN f1 score: 0.471\n",
      "KNN cohens kappa score: 0.461\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 1278 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: 325, 8\n",
      "LR fn, tp: 7, 6\n",
      "LR f1 score: 0.444\n",
      "LR cohens kappa score: 0.422\n",
      "LR average precision score: 0.268\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 333, 0\n",
      "GB fn, tp: 1, 12\n",
      "GB f1 score: 0.960\n",
      "GB cohens kappa score: 0.959\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 333, 0\n",
      "KNN fn, tp: 8, 5\n",
      "KNN f1 score: 0.556\n",
      "KNN cohens kappa score: 0.546\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 1278 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: 326, 7\n",
      "LR fn, tp: 11, 2\n",
      "LR f1 score: 0.182\n",
      "LR cohens kappa score: 0.156\n",
      "LR average precision score: 0.296\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 332, 1\n",
      "GB fn, tp: 0, 13\n",
      "GB f1 score: 0.963\n",
      "GB cohens kappa score: 0.961\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 333, 0\n",
      "KNN fn, tp: 6, 7\n",
      "KNN f1 score: 0.700\n",
      "KNN cohens kappa score: 0.692\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 1280 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: 319, 12\n",
      "LR fn, tp: 13, 0\n",
      "LR f1 score: 0.000\n",
      "LR cohens kappa score: -0.038\n",
      "LR average precision score: 0.139\n",
      "\n",
      "-> test with 'GB'\n",
      "GB tn, fp: 329, 2\n",
      "GB fn, tp: 0, 13\n",
      "GB f1 score: 0.929\n",
      "GB cohens kappa score: 0.926\n",
      "\n",
      "-> test with 'KNN'\n",
      "KNN tn, fp: 331, 0\n",
      "KNN fn, tp: 5, 8\n",
      "KNN f1 score: 0.762\n",
      "KNN cohens kappa score: 0.755\n",
      "\n",
      "### Exercise is done.\n",
      "\n",
      "-----[ LR ]-----\n",
      "maximum:\n",
      "LR tn, fp: 330, 142\n",
      "LR fn, tp: 13, 8\n",
      "LR f1 score: 0.444\n",
      "LR cohens kappa score: 0.422\n",
      "LR average precision score: 0.566\n",
      "\n",
      "\n",
      "average:\n",
      "LR tn, fp: 301.52, 31.08\n",
      "LR fn, tp: 9.44, 3.56\n",
      "LR f1 score: 0.214\n",
      "LR cohens kappa score: 0.182\n",
      "LR average precision score: 0.269\n",
      "\n",
      "\n",
      "minimum:\n",
      "LR tn, fp: 191, 2\n",
      "LR fn, tp: 5, 0\n",
      "LR f1 score: 0.000\n",
      "LR cohens kappa score: -0.038\n",
      "LR average precision score: 0.051\n",
      "\n",
      "\n",
      "-----[ GB ]-----\n",
      "maximum:\n",
      "GB tn, fp: 333, 7\n",
      "GB fn, tp: 5, 13\n",
      "GB f1 score: 1.000\n",
      "GB cohens kappa score: 1.000\n",
      "\n",
      "\n",
      "average:\n",
      "GB tn, fp: 331.32, 1.28\n",
      "GB fn, tp: 0.68, 12.32\n",
      "GB f1 score: 0.929\n",
      "GB cohens kappa score: 0.926\n",
      "\n",
      "\n",
      "minimum:\n",
      "GB tn, fp: 326, 0\n",
      "GB fn, tp: 0, 8\n",
      "GB f1 score: 0.621\n",
      "GB cohens kappa score: 0.604\n",
      "\n",
      "\n",
      "-----[ KNN ]-----\n",
      "maximum:\n",
      "KNN tn, fp: 333, 5\n",
      "KNN fn, tp: 13, 8\n",
      "KNN f1 score: 0.762\n",
      "KNN cohens kappa score: 0.755\n",
      "\n",
      "\n",
      "average:\n",
      "KNN tn, fp: 331.92, 0.68\n",
      "KNN fn, tp: 8.96, 4.04\n",
      "KNN f1 score: 0.442\n",
      "KNN cohens kappa score: 0.432\n",
      "\n",
      "\n",
      "minimum:\n",
      "KNN tn, fp: 328, 0\n",
      "KNN fn, tp: 5, 0\n",
      "KNN f1 score: 0.000\n",
      "KNN cohens kappa score: -0.014\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": "2ea0e554",
   "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
}