{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "maritime-diagram",
   "metadata": {},
   "outputs": [],
   "source": [
    "from library.exercise import Exercise, allTesters\n",
    "from library.dataset import DataSet, TrainTestData\n",
    "from library.interfaces import GanBaseClass, TesterNetworkBaseClass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "changing-enhancement",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "acquired-dover",
   "metadata": {},
   "outputs": [],
   "source": [
    "def loadFailingDataset():\n",
    "    data0 = np.array(range(100)).reshape((25,4))\n",
    "    return DataSet(data0=data0[1:], data1=data0 * 2)\n",
    "\n",
    "def loadEqualDataset():\n",
    "    data0 = np.array(range(100)).reshape((25,4))\n",
    "    return DataSet(data0=data0, data1=data0 * 2)\n",
    "\n",
    "def loadDataset():\n",
    "    data0 = np.array(range(100)).reshape((25,4))\n",
    "    return DataSet(data0=data0, data1=(data0 * 2)[0:12])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "searching-sweet",
   "metadata": {},
   "outputs": [],
   "source": [
    "exercise = Exercise(allTesters)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "concrete-range",
   "metadata": {},
   "outputs": [],
   "source": [
    "gan = GanBaseClass()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "binary-biodiversity",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "### Start exercise for synthetic point generator\n",
      "\n",
      "====== Step {shuffleStep + 1}/{self.numOfShuffles} =======\n",
      "-> Spliting data to slices\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=20\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [5 0 5 0]\n",
      "f1 score_KNN: 0.0\n",
      "balanced accuracy_KNN: 0.5\n",
      "confusion matrix_KNN\n",
      "[[5 0]\n",
      " [5 0]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=20\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [5 0 5 0]\n",
      "f1 score_KNN: 0.0\n",
      "balanced accuracy_KNN: 0.5\n",
      "confusion matrix_KNN\n",
      "[[5 0]\n",
      " [5 0]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=20\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [5 0 5 0]\n",
      "f1 score_KNN: 0.0\n",
      "balanced accuracy_KNN: 0.5\n",
      "confusion matrix_KNN\n",
      "[[5 0]\n",
      " [5 0]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=20\n",
      "{1: 20, 0: 20}\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWkAAAFpCAYAAABee9lOAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAAmwklEQVR4nO3df1RUdf4/8CcjEiKkzARmwLYu7UhWM+Umv77QJmKhu6motRqQa7WRpO4Rd1M/o0CouLa2pqct6IdYUosGxLGDn/LH7n60PfzwUx1mN0WpCAmTpZkRBxpiYOb7h+t8xBlowJm574Hn45zOLve+5973fTc9efOa99zrY7VarSAiIiHJpO4AERENjCFNRCQwhjQRkcAY0kREAmNIExEJjCFNRCQwX6k74G10uk5YLJ5btRgcHACD4TuPnc8bcEzscUzsedOYhIQEDbiPM2nB+fqOkboLwuGY2OOY2BspY8KQJiISGEOaiEhgDGkiIoExpImIBMaQJiISGJfguUhfXy8Mhnb09va49Lj//rcMFovFpcccCl9fPwQHh2DMGL5ViKTA//JcxGBoh79/AMaPvxk+Pj4uO66vrwy9vdKEtNVqRVfXJRgM7bjppsmS9IFotGO5w0V6e3swfvyNLg1oqfn4+GD8+Btd/tcBETmPIe1CIymgrxiJ10TkTRjSREQCY0gTEQmMIe0Bnd29+NpgQmd3r9Rd6WffvmLMn5+CWbP+HzZsWAuDQS91l4joGgxpD7hoMsPcZ8FFk1nqrthUVR3EW28VIzt7HQoL98BoNCIvTyN1t4joGgxpD5g4bizGjpFh4rixUnfFprx8P5YsScPPfz4TP/3pVGzYkIOPPz6JL7/8XOquEdFVGNIeEOjvi/DgcQj0F2NZek9PDz7/vBHTp99r2xYWFo7Jk2/BqVP/krBnRHQthvQodOlSBywWC4KD5f22T5w4EQaDQaJeEZEjDGmB9B07AvMTj6Hv2BG3nsdq9dyTZYjo+jCkBWIueQu9TV/BXPKWW88zYcJEyGQyu9UcFy9eRHBwsFvPTURDI3RIFxUVISEhAWq1GllZWdDpdAO2bWpqQkZGBlQqFZKSklBRUdFv/6ZNm5CUlIS77roLiYmJyMvLQ1dXl7svYUgupS5Bb/iPcCl1iVvP4+fnh9tu+yk++eR/bdvOn2/FN9+cx7Rpd7r13EQ0NMKGdHl5OQoLC5Gbm4vS0lIYjUZkZ2c7bGs2m5GZmQmFQoGysjKsWLECOTk5qKurs7W566678Mc//hEffPABXnjhBdTV1WH79u2euhynBDzwIDp2vYaABx50+7kWLnwE+/e/gxMn/o7GxrP4wx82Y/r0e/GTn9zm9nMTkfPEWG7gQElJCZYvX47Zs2cDAAoKCpCcnIyzZ89CqVT2a3v8+HG0tbWhsrISAQEBUCqVOHnyJEpKShAdHQ0AeOSRR2ztw8LCsHTpUuzfv99zF+SEQH9fj60A+eUv58Ng0GPHjm3o7OzEjBkxePZZrpMmEo2QId3T04OGhgZs2LDBti0iIgJhYWGor6+3C2mtVguVSoWAgADbtri4OLz44osOj//tt9/i8OHDmD59ulv67y0yMpYjI2O51N0gokEIGdIGgwEWiwUKhaLfdrlcDr3e/qvLer3eYdtra9iFhYUoLCyEyWTCzJkzsXHjxiH3TaEIdLj93/+WwdfXPdUjdx3XWTKZDCEhQZL24Vqi9UcEHBN7I2FMhAzpoXJ2SdmSJUvw4IMPorm5GTt27MDOnTuxbt26IZ1Lp+uExWJ/PovF4pab80t50/8rLBYL2tuNkvbhaiEhQUL1RwQcE3veNCaD/TIRMqSDg4Mhk8mg0+kQGRlp267X6yGXy+3aKxQKNDc399vmaHY9ceJETJw4EVOmTEFQUBDS09OxcuVKjB8/3j0X4kad3b24aDJj4rixwnyTkYhcT8jVHX5+foiKikJtba1tW0tLC1pbW6FWq+3aq1QqaLVamEwm27aamhqHba+4MvuWyYQcgh8k4k2biMj1hE2otLQ0FBcX4+jRo2hoaIBGo0FMTAyUSiW0Wi1SUlLQ1tYGAEhMTERoaCg0Gg0aGxtRVlaGqqoqpKenAwAuXLiAV199FadOnUJrayv+8Y9/ID8/H0lJSRg3bpyUlzlsIt60iYhcT9i/kxcvXgydToe8vDwYjUbEx8dj8+bNAACTyYSmpiaYzZdnkX5+figqKkJubi4WLlyIkJAQ5Ofn25bfjR07FidPnsSePXvQ1dWFSZMmYfbs2cjKypLs+q6XJ5frEZF0fKy8kcOQDPTB4YULzbj55ltdfj4RPjh017UNlzd9IOQpHBN73jQmg31wKGy5g4iIGNIjnqiP7iIi5zCkRziuAiHybvzkaYSbOG6sbT31Ff/zP39FRcW7aGg4ha6uLnz00f8OcgQikhJn0iOco0d3dXd3Y/r0e5Ge/mvpOkZETuFMehR68MG5ANDvftJEJCbOpImIBMaQJiISGENaINVf6LCh4l+o/mLgx4S5E5frEYmHIS2Qg/XfoPWiCQfrv5Hk/FyuRyQehrRA5qknI2ziOMxTT5bk/LxpE5F4uLpDIHGRCsRFKn644XW6dKkDbW0X0Nr6NQCgsfEMAODHP/4JwoO9866ARCMVQ3oU+uij4ygoeM728/LlaQCAd989iMmTb5GqW0TkAEN6FJo79yHMnfuQ1N0gIiewJk1DJvUqFKLRhCFNQyb1KhSi0YQhTUMm9SoUotGENWkaMk+tQiEizqRdaiQ+iWwkXhORN2FIu4hMNgZ9fSPv69R9fb2QycZI3Q2iUYsh7SLjxgXCaLwIq1Xah8a6ktVqgdFowLhxgVJ3hWjUYk3aRQIDJ8BgaEdb29cAXFcikMlksFikCn4f+Pn5IzBwwrCPUP2FDgfrv8E89WTWsYmGgSHtIj4+PpDLQ11+XG96LL0jVy/XY0gTDR3LHeRWXK5HdH04kya34nI9ouvDmTQRkcAY0kREAmNIkxB40yYixxjSJATetInIMYY0CYGrQIgc4+oOEgJXgRA5xpk0EZHAGNJERAJjSBMRCYwhTV7n76cucLkejRoMafI6+2vOcbkejRpCh3RRURESEhKgVquRlZUFnW7gmVNTUxMyMjKgUqmQlJSEiooK2z6z2Yzt27djzpw5UKvVSEpKwu7du9HbO/Ju0j8a/Cr2R1yuR6OGsEvwysvLUVhYiOeffx7h4eEoKChAdnY23nzzTbu2ZrMZmZmZmDZtGsrKylBfX4+cnByEh4cjOjoa3d3daGxsRHZ2NpRKJb766iv813/9F2QyGVauXCnB1dH1uH/azbgjZLzU3SDyCGFDuqSkBMuXL8fs2bMBAAUFBUhOTsbZs2ehVCr7tT1+/Dja2tpQWVmJgIAAKJVKnDx5EiUlJYiOjkZQUBBef/11W/tbb70Vy5cvR1VVFUOaiIQmZLmjp6cHDQ0NiI2NtW2LiIhAWFgY6uvr7dprtVqoVCoEBATYtsXFxTlse4XBYEBQUJBrO05E5GJCzqQNBgMsFgsUiv7fQJPL5dDr9Xbt9Xq9w7YD1bDPnz+PAwcOYOPGjUPum0Lh+ef9hYTwl8m1nBmTv5+6gP015/Cr2B/h/mk3e6BX0uL7xN5IGBMhQ3qorFbnnynY0dGBzMxMzJo1C/Pnzx/yuXS6TlgsrnuG4Q/x9sdnuYOzY1JyogmtF00oOdE04mvYfJ/Y86YxGeyXiZDljuDgYMhkMruZsF6vh1wut2uvUCgctr12dt3V1YWnnnoK4eHh2LJli+s7TkLhTZtoJBAypP38/BAVFYXa2lrbtpaWFrS2tkKtVtu1V6lU0Gq1MJlMtm01NTX92nZ3d+Ppp5+Gv78/du3aBV/fEfFHBA0iLlKBbQvv5I2byKsJGdIAkJaWhuLiYhw9ehQNDQ3QaDSIiYmBUqmEVqtFSkoK2traAACJiYkIDQ2FRqNBY2MjysrKUFVVhfT0dACXl+itWrUKer0e+fn56OjoQHt7O9rb26W8RCKiHyTsdHLx4sXQ6XTIy8uD0WhEfHw8Nm/eDAAwmUxoamqC2WwGcHnmXVRUhNzcXCxcuBAhISHIz89HdHQ0AKCtrQ3Hjx8HADzwwAP9znPmzBkPXhUR0dD4WIfyqRvxg0MBcEzscUzsedOYeN0Hh0SexmcskqgY0kTgMxZJXAxpInC5HolL2A8OiTyJz1gkUXEmTUQkMIY0EZHAGNJEQ8BVIORpDGmiIeAqEPI0hjTREHAVCHkaV3cQDQFXgZCncSZNRCQwhjQRkcAY0kREAmNIE7kJl+uRKzCkidyEy/XIFRjSRG7C5XrkClyCR+QmXK5HrsCZNBGRwBjSREQCY0gTSYyrQGgwDGkiiXEVCA2GIU0kMa4CocFwdQeRxLgKhAbDmTQRkcAY0kREAmNIE3kJrgIZnRjSRF6Cq0BGJ4Y0kZfgKpDRias7iLwEV4GMTpxJExEJjCFNRCQwhjQRkcAY0kREAmNIExEJjCFNRCQwhjQRkcCEDumioiIkJCRArVYjKysLOt3AX4dtampCRkYGVCoVkpKSUFFR0W9/aWkpli5dCrVajaSkJHd3nYjIJYQN6fLychQWFiI3NxelpaUwGo3Izs522NZsNiMzMxMKhQJlZWVYsWIFcnJyUFdXZ2vz/fffY9asWVi6dKmnLoGI6LoJ+43DkpISLF++HLNnzwYAFBQUIDk5GWfPnoVSqezX9vjx42hra0NlZSUCAgKgVCpx8uRJlJSUIDo6GgCwbNkyAEBFRQUOHz7s2YshIhomIWfSPT09aGhoQGxsrG1bREQEwsLCUF9fb9deq9VCpVIhICDAti0uLs5hWyIibyJkSBsMBlgsFigU/e9TIJfLodfr7drr9XqHbQerYRMReQNhyx1DYbVaPXYuhSLQY+e6IiQkyOPnFB3HxB7HxN5IGBMhQzo4OBgymQw6nQ6RkZG27Xq9HnK53K69QqFAc3Nzv22OZteuoNN1wmLx3C+FkJAgtLcbPXY+b8AxsccxsedNYzLYLxMhyx1+fn6IiopCbW2tbVtLSwtaW1uhVqvt2qtUKmi1WphMJtu2mpoah22JiLyJkCENAGlpaSguLsbRo0fR0NAAjUaDmJgYKJVKaLVapKSkoK2tDQCQmJiI0NBQaDQaNDY2oqysDFVVVUhPT7cdr729HadPn8b58+fR09OD06dP4/Tp01JdHhGRU4QsdwDA4sWLodPpkJeXB6PRiPj4eGzevBkAYDKZ0NTUBLPZDODyzLuoqAi5ublYuHAhQkJCkJ+fb1t+B1z+MstLL71k+3nBggUAgDNnznjuooiIhsjH6slP3UYA1qSlxzGxxzGx501j4nU1aSIiuowhTUQkMIY0EZHAGNJERAJjSBMRCYwhTUQkMIY0EZHAGNJERAJjSBMRCYwhTUQkMIY0EZHAGNJERAJjSBMRCcwlIf3dd9/h5MmTrjgUERFdxSUhfe7cOTz22GOuOBQREV2F5Q4iIoE59WSWO++80939ICIiB5wKaV9fXyxduhRKpdLh/tbWVvz5z392aceIiK6o/kKHg/XfYJ56MuIiFVJ3x6OcCmmlUolbbrkFqampDvc3NDQwpInIbQ7Wf4PWiyYcrP9m1IW0UzXpu+++G83NzQPuDwgIwL333uuyThERXW2eejLCJo7DPPVkqbvicXwQ7RDxQbTS45jY45jY86Yx8ciDaM+dO+eqQxER0X9cV0j39vbi0KFDWLZsGVJSUlzVJyIi+g+nPji81ldffYUDBw6gsrISvb29mDlzJl555RVX942IaNRzOqTNZjMOHz6M0tJSfPrpp4iPj8fFixdRWVk54NI8IiJPu7JcLz1xCu4IGS91d66bU+WO7du347777sMrr7yCn//85/jb3/6GV199FT4+PpDJ+KVFIhLHleV6+2tGxudkTs2k33zzTTz99NNYuXIlQ5mIhDZPPRkH67/Br2J/JHVXXMKpxF27di0OHTqE++67D9u3b8fZs2fd3S8iomGJi1Rg28I7cf+0m6Xuiks4NZN+4okn8MQTT6CmpgYHDhzAokWLcNttt8FqteK7775zdx+JiEatIa3uiI2NRWxsLPR6PcrLy/Hdd9/h0UcfxYwZMzB37lw8/PDD7uonEdGoNKwCs1wux29+8xt8+OGHeO2113DjjTciPz/f1X0jInKb6i902FDxL1R/oZO6K4Ma1jrpq8XFxSEuLg56vd4V/SEi8ghvuWmTUzPpU6dOIT09HUaj/ffgL126hIyMDLS3t7u8c0RE7uItN21yKqSLi4sxY8YMBAXZ3wTkxhtvRHR0NIqLi13eOSIid7myCkTkWTTgZEjX19cjOTl5wP2zZs3Cxx9/7LJOERHRZU6F9IULFzBx4sQB90+YMAFtbW2u6hMREf2HUyEdFBQ06K1Iz50757AUQkRE18epkJ4+fTrKysoG3P/uu+/innvucVmnrigqKkJCQgLUajWysrKg0w28VKapqQkZGRlQqVRISkpCRUVFv/29vb3Ytm0bYmJiMH36dGzYsIFfxCEip0i5XM+pkH788cfxwQcf4LnnnutX1mhra0NeXh4+/PBDPP744y7tWHl5OQoLC5Gbm4vS0lIYjUZkZ2c7bGs2m5GZmQmFQoGysjKsWLECOTk5qKurs7V5+eWXUVVVhRdffBF79+6FVqvF5s2bXdpnIhqZrl6u52lOPz6rvLwczz33HMxmMwIDAwEAnZ2dGDt2LDZt2uTybxumpqZi5syZWL16NQCgpaUFycnJeP/99+1ujXrs2DFkZ2ejuroaAQEBAIBnn30W3d3d2L17NywWC+Lj4/H73/8eixYtAgBUV1fjySefRHV1NW688Uan+8XHZ0mPY2KPY2LPlWPi7qeVD/b4LKe/zLJo0SIkJCSgvLwc7e3tsFqtmDJlClJSUjBp0iSXdPSKnp4eNDQ0YMOGDbZtERERCAsLQ319vV1Ia7VaqFQqW0ADl79k8+KLLwK4HPAGgwGxsbG2/dHR0bBarfjss88QFxfn0v4T0cgSF6mQbKmeUyF98eJFrF+/HidOnIDFYoFarcaOHTsQHh7ulk4ZDAZYLBYoFP0HRS6XO/xmo16vd9j2Sg37yv9e3WbMmDGYMGHCoHVuRxSKwCG1d4XBfsuOVhwTexwTeyNhTJwK6Z07d6K+vh6rVq3CDTfcgHfeeQe5ubl444033N0/p/xQxcaVD0RnuUN6HBN7HBN73jQm113uOHHiBLZu3YqkpCQAQGJiIubNmwez2YyxY8e6ppdXCQ4Ohkwmg06nQ2RkpG27Xq+HXC63a69QKNDc3Nxv29Wz65tuugnA5Rl1WFgYAKCvrw8dHR12M3AiouFyR+3aqdUdbW1tuOOOO2w/33bbbRg7dqzb7tfh5+eHqKgo1NbW2ra1tLSgtbUVarXarr1KpYJWq4XJZLJtq6mpsbWNiIhAcHBwv+OdPHkSPj4+mDZtmluugYhGH3esAnEqpPv6+uxmzDKZDH19fS7ryLXS0tJQXFyMo0ePoqGhARqNBjExMVAqldBqtUhJSbEtB0xMTERoaCg0Gg0aGxtRVlaGqqoqpKen2/q6dOlS7Ny5EzU1NdBqtdi6dSvmzZuHCRMmuO0aiGh0ccdNm5xe3fHb3/62X1D39PRg3bp18Pf3t23bs2ePyzq2ePFi6HQ65OXlwWg0Ij4+3rau2WQyoampCWazGcDlmXdRURFyc3OxcOFChISEID8/H9HR0bbjPfPMM+jq6sLq1athNpvx4IMPYtOmTS7rLxGRO1aBOLVO+uqlcIPZtm3bdXdIdPzgUHocE3scE3veNCbX/cHhaAhfIiIRDevxWURE5BkMaSIigTGkiYgExpAmIhIYQ5qISGAMaSIigTGkiYgExpAmIhIYQ5qISGAMaSIigTGkiYgExpAmIhIYQ5qISGAMaSIigTGkiYgExpAmIhIYQ5qISGAMaSIigTGkiYgExpAmIhIYQ5qISGAMaSIigTGkiYgExpAmIhIYQ5qISGAMaSIigTGkiYgExpAmIhIYQ5qISGAMaSIigTGkiYgExpAmIhIYQ5qISGAMaSIigTGkiYgEJmxIFxUVISEhAWq1GllZWdDpdIO2b2pqQkZGBlQqFZKSklBRUdFvf2lpKZYuXQq1Wo2kpCR3dp2IyGWEDOny8nIUFhYiNzcXpaWlMBqNyM7OHrC92WxGZmYmFAoFysrKsGLFCuTk5KCurs7W5vvvv8esWbOwdOlST1wCEZFL+ErdAUdKSkqwfPlyzJ49GwBQUFCA5ORknD17Fkql0q798ePH0dbWhsrKSgQEBECpVOLkyZMoKSlBdHQ0AGDZsmUAgIqKChw+fNhzF0NEdB2Em0n39PSgoaEBsbGxtm0REREICwtDfX29w9dotVqoVCoEBATYtsXFxQ3YnojIWwg3kzYYDLBYLFAoFP22y+Vy6PV6h6/R6/UO2/9QHXs4FIpAp9ua/vu/0fn6Gwh88gmMmzNn2OcMCQka9mtHKo6JPY6JvZEwJh4N6fXr1+O9994bcH9qairWrFkz5ONardbr6daQ6HSdsFicO5/5lVdhbW6G4ZVX0XlvwrDOFxIShPZ247BeO1JxTOxxTOx505gM9svEoyGt0Wiwdu3aAff7+/vjhhtugEwmg06nQ2RkpG2fXq+HXC53+DqFQoHm5uZ+2xzNrj1N9mgGLO/sg+zRDEn7QUTey6MhHRQUhKCgH/7zIyoqCrW1tbYP/VpaWtDa2gq1Wu2wvUqlwt69e2EymTBu3DgAQE1NzYDtPWXMrNkYM2t2v219x47YgvvafURE1xLug0MASEtLQ3FxMY4ePYqGhgZoNBrExMTYVnZotVqkpKSgra0NAJCYmIjQ0FBoNBo0NjairKwMVVVVSE9Ptx2zvb0dp0+fxvnz59HT04PTp0/j9OnTHr82yzv7YG1uhuWdfR4/NxF5H+E+OASAxYsXQ6fTIS8vD0ajEfHx8di8ebNtv8lkQlNTE8xmMwDAz88PRUVFyM3NxcKFCxESEoL8/HzbTBy4/GWWl156yfbzggULAABnzpzxzEX9B0sgRDQUPlZPfuo2Agzlg0NnDVYC8aYPPzyFY2KPY2LPm8ZksA8OhSx3jDYsgRDRQBjSApA9mgGfW29lCYSI7AhZkx5tBlsFYlrxFDDMNdZE5P04kxbUlRJI5+t7pO4KEUmIIS2oKyWQwCcfl7orRCQhljsEdaUEMi4kCJ1e8gk1EbkeZ9Jepu/YEZifeAx9x45I3RUi8gCGtJfhcj2i0YUh7WW4XI9odGFN2svwpk1Eowtn0iMASyBEIxdDegRgCYRo5GK5YwRgCYRo5OJMeoRiCYRoZGBIj1AsgRCNDCx3jFAsgRCNDJxJjyIsgRB5H4b0KMISCJH3YbljFHFUAiEisXEmTbxpE5HAGNLEWjWRwBjSxFo1kcBYkyYu1yMSGGfS5BBLIERiYEiTQyyBEImB5Q5yiCUQIjFwJk1OYwmEyPMY0uQ0lkCIPI/lDnIaSyBEnseZNF0XlkCI3IshTdeFJRAi92K5g64Lb9pE5F6cSZNb8KZNRK7BkCa3YK2ayDUY0uQWrFUTuQZr0uQWXK5H5BrCzqSLioqQkJAAtVqNrKws6HS6Qds3NTUhIyMDKpUKSUlJqKiosO0zm83Yvn075syZA7VajaSkJOzevRu9vb3uvgy6CksgREMnZEiXl5ejsLAQubm5KC0thdFoRHZ29oDtzWYzMjMzoVAoUFZWhhUrViAnJwd1dXUAgO7ubjQ2NiI7OxsHDx5Ebm4u9u/fj8LCQk9dEoElEKLh8LFarVapO3Gt1NRUzJw5E6tXrwYAtLS0IDk5Ge+//z6USqVd+2PHjiE7OxvV1dUICAgAADz77LPo7u7G7t27HZ7j9ddfR1VVFd57770h9U2n64TF4rkhCwkJQnu70WPn87ThlEBG+pgMB8fEnjeNSUhI0ID7hJtJ9/T0oKGhAbGxsbZtERERCAsLQ319vcPXaLVaqFQqW0ADQFxc3IDtAcBgMCAoaOCBIc9gCYRocMKFtMFggMVigUKh6LddLpdDr9c7fI1er3fYfqA69vnz53HgwAEsWrTINZ2mYWMJhGhwHl3dsX79+kHLC6mpqVizZs2QjzuUik1HRwcyMzMxa9YszJ8/f8jnUigCh/ya6zXYn0Jeb8nCy/8M0Ygek2HimNgbCWPi0ZDWaDRYu3btgPv9/f1xww03QCaTQafTITIy0rZPr9dDLpc7fJ1CoUBzc3O/bY5m111dXXjqqacQHh6OLVu2DOsaWJOWHsfEHsfEnjeNyWC/TDwa0kFBQU7VgaOiolBbW4vo6GgAlz84bG1thVqtdthepVJh7969MJlMGDduHACgpqamX/vu7m48/fTT8Pf3x65du+DryyXiRCQ+4WrSAJCWlobi4mIcPXoUDQ0N0Gg0iImJsa3s0Gq1SElJQVtbGwAgMTERoaGh0Gg0aGxsRFlZGaqqqpCeng7g8hK9VatWQa/XIz8/Hx0dHWhvb0d7e7tk10hE5Awhp5OLFy+GTqdDXl4ejEYj4uPjsXnzZtt+k8mEpqYmmM1mAICfnx+KioqQm5uLhQsXIiQkBPn5+baZeFtbG44fPw4AeOCBB/qd68yZMx66KiKioRNynbTIWJOWHsfEHsfEnjeNiVetkyYiov/DkCYiEhhDmohIYAxpIiKBMaSJiATGkCYiEhhDmohIYAxpIiKBMaSJiATGkCYiEhhDmka1vmNHYH7iMfQdOyJ1V4gcYkjTqMbHd5HoGNI0qvHxXSQ6IW9VSuQpY2bNdvop5URS4Eya6BqsU5NIGNJE12CdmkTCkCa6BuvUJBLWpImu4ahO3XfsCCzv7IPs0QzWsMmjOJMmcgJLICQVhjSRE1gCIamw3EHkBJZASCqcSRMNE0sg5AkMaaJhYgmEPIHlDqJhGujbiiyDkCtxJk3kYiyDkCsxpIlcjGUQciWWO4hcjDdtIlfiTJqISGAMaSIigTGkiYgExpAmIhIYQ5qISGAMaSIigTGkiYgExpAmIhIYQ5qISGAMaSIigfFr4UMkk/mMinOKjmNij2NibySMiY/VarVK3QkiInKM5Q4iIoExpImIBMaQJiISGEOaiEhgDGkiIoExpImIBMaQJiISGEOaiEhgDGkiIoExpImIBMaQ9hKvvPIK5s+fj7vvvhv33XcftmzZgq6uLqm75XFFRUVISEiAWq1GVlYWdDqd1F2SDN8TP+yZZ57B1KlTUVtbK3VXho0h7SU+/fRTPPnkk6ioqMALL7yAjz76CFu2bJG6Wx5VXl6OwsJC5ObmorS0FEajEdnZ2VJ3SzJ8TwyusrISJpNJ6m5cPyt5pUOHDllnzJghdTc8asGCBdZdu3bZfj537pxVqVRaz5w5I2GvxDEa3xMDuXDhgvX++++3tra2WpVKpbWmpkbqLg0bZ9JeymAwICgoSOpueExPTw8aGhoQGxtr2xYREYGwsDDU19dL2DNxjLb3xGA0Gg0yMzNxyy23SN2V68aQ9kJGoxF79uzBokWLpO6KxxgMBlgsFigUin7b5XI59Hq9RL0Sx2h8TwyktLQUvb29WLJkidRdcQne9F9i69evx3vvvTfg/tTUVPzhD3+w/dzT04NVq1YhIiICTz31lCe6SILje+L/nD9/Hi+99BJKS0ul7orLMKQlptFosHbt2gH3+/v72/5/b28v1qxZg66uLuzduxe+vqPnX19wcDBkMhl0Oh0iIyNt2/V6PeRyuYQ9k9Zofk84curUKXz77bd44IEH+m3/9a9/jdTUVBQUFEjUs+Eb3f9GBRAUFORUHdFisWDdunU4d+4c9u3bh/Hjx3ugd+Lw8/NDVFQUamtrER0dDQBoaWlBa2sr1Gq1xL2Txmh/TzgSGxuLgwcP9tv20EMPYcuWLUhISJCoV9eHIe0lNm3ahNraWrz22mswm81ob28HcLkmO2bMGIl75xlpaWkoKCjA7bffjvDwcBQUFCAmJgZKpVLqrkmC7wl7gYGBDt8P4eHhmDRpkgQ9un58xqGXmDp1qsPtx44dQ3h4uId7I52ioiLs27cPRqMR8fHx2Lx5M2666SapuyUJviecM3XqVLz11luIiYmRuivDwpAmIhIYl+AREQmMIU1EJDCGNBGRwBjSREQCY0gTEQmMIU1EJDCGNBGRwBjSRNdYv349pk6diqlTp2LatGmYOXMmcnJyYDAYbG0+/fRTrFy5EvHx8bjrrruQnJyM3/3ud/jss8/sjvfGG2/g9ttvx7Zt2xyer7OzExs3bkRMTAzuvvtuPPnkkzh37pzbro+8C0OayIF7770XH330Ef76179Co9Hg8OHDWLduHYDLT4hJS0uDr68vduzYgUOHDmHnzp0ICwvD1q1b7Y514MABZGZmorKyEj09PXb7n332WVRXV2PXrl145513YLVasXz5cnR3d7v9OskLSPrIASIBrVu3zrps2bJ+215++WVrVFSU9cKFC9Y777zTumnTJoevvXjxYr+fq6urrXFxcVaz2WydO3eu9eDBg/32f/nll1alUmk9ceJEv2Pccccd1vLyctdcEHk1zqSJnODv7w+LxYKysjL09PQgKyvLYbsJEyb0+3n//v146KGH4OvriwULFuDAgQP99n/yyScYO3Ys4uLi+h1DpVLh448/dv2FkNdhSBP9gM8//xxvv/021Go12tvbERgYiJtvvvkHX6fX63HkyBGkpqYCAObPn49PPvkEX375pa1Ne3s7Jk6caHfXuptuusl2Vzsa3RjSRA7U1dXhnnvugUqlwi9/+UtERETghRdegHUI9yMrLy9HZGQkoqKiAAChoaGIj4/Hu+++69TrfXx8htV3Gll4P2kiB1QqFbZv344xY8YgNDQUfn5+AIApU6ags7MTFy5cGHQ2bbVaUVZWhubmZkybNs223WKx4J///CfWrFkDPz8/hISE4OLFi+jr6+s3m9bpdPjxj3/stusj78GZNJED/v7+uPXWWxEeHm4LaACYM2cO/Pz88PLLLzt8XUdHBwCgpqYGLS0t+Mtf/oLKysp+//T29uLIkSMAgOnTp8NsNqOmpsZ2jEuXLqG+vh4/+9nP3HiF5C04kyYagkmTJiEnJwc5OTkwGo145JFHEBERgY6ODhw7dgy1tbV4++23UVpaihkzZuCee+6xO0ZSUhL279+PX/ziF5gyZQpmzZqFvLw8bN26FUFBQfjTn/6ESZMmYe7cuRJcIYmGM2miIXr44Yexb98+fP/998jOzkZKSgpWr16Nr7/+Ghs3boROp8OxY8cwZ84ch6+fO3cu6urq8NVXXwEAnn/+eURHR2PlypVYsmQJLBYL9uzZ0+8hxDR68cksREQC40yaiEhgDGkiIoExpImIBMaQJiISGEOaiEhgDGkiIoExpImIBMaQJiISGEOaiEhg/x+/1Nyf7Q5DrgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_KNN: 1.0\n",
      "balanced accuracy_KNN: 1.0\n",
      "confusion matrix_KNN\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=20\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [1 4 0 5]\n",
      "f1 score_KNN: 0.7142857142857143\n",
      "balanced accuracy_KNN: 0.6\n",
      "confusion matrix_KNN\n",
      "[[1 4]\n",
      " [0 5]]\n",
      "-> check results\n",
      "\n",
      "====== Step {shuffleStep + 1}/{self.numOfShuffles} =======\n",
      "-> Spliting data to slices\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=20\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [5 0 5 0]\n",
      "f1 score_KNN: 0.0\n",
      "balanced accuracy_KNN: 0.5\n",
      "confusion matrix_KNN\n",
      "[[5 0]\n",
      " [5 0]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=20\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [5 0 5 0]\n",
      "f1 score_KNN: 0.0\n",
      "balanced accuracy_KNN: 0.5\n",
      "confusion matrix_KNN\n",
      "[[5 0]\n",
      " [5 0]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=20\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [5 0 5 0]\n",
      "f1 score_KNN: 0.0\n",
      "balanced accuracy_KNN: 0.5\n",
      "confusion matrix_KNN\n",
      "[[5 0]\n",
      " [5 0]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=20\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_KNN: 1.0\n",
      "balanced accuracy_KNN: 1.0\n",
      "confusion matrix_KNN\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=20\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [1 4 0 5]\n",
      "f1 score_KNN: 0.7142857142857143\n",
      "balanced accuracy_KNN: 0.6\n",
      "confusion matrix_KNN\n",
      "[[1 4]\n",
      " [0 5]]\n",
      "-> check results\n",
      "\n",
      "====== Step {shuffleStep + 1}/{self.numOfShuffles} =======\n",
      "-> Spliting data to slices\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=20\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [5 0 5 0]\n",
      "f1 score_KNN: 0.0\n",
      "balanced accuracy_KNN: 0.5\n",
      "confusion matrix_KNN\n",
      "[[5 0]\n",
      " [5 0]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=20\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [5 0 5 0]\n",
      "f1 score_KNN: 0.0\n",
      "balanced accuracy_KNN: 0.5\n",
      "confusion matrix_KNN\n",
      "[[5 0]\n",
      " [5 0]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=20\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [5 0 5 0]\n",
      "f1 score_KNN: 0.0\n",
      "balanced accuracy_KNN: 0.5\n",
      "confusion matrix_KNN\n",
      "[[5 0]\n",
      " [5 0]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=20\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_KNN: 1.0\n",
      "balanced accuracy_KNN: 1.0\n",
      "confusion matrix_KNN\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=20\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [1 4 0 5]\n",
      "f1 score_KNN: 0.7142857142857143\n",
      "balanced accuracy_KNN: 0.6\n",
      "confusion matrix_KNN\n",
      "[[1 4]\n",
      " [0 5]]\n",
      "-> check results\n",
      "\n",
      "====== Step {shuffleStep + 1}/{self.numOfShuffles} =======\n",
      "-> Spliting data to slices\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=20\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [5 0 5 0]\n",
      "f1 score_KNN: 0.0\n",
      "balanced accuracy_KNN: 0.5\n",
      "confusion matrix_KNN\n",
      "[[5 0]\n",
      " [5 0]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=20\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [5 0 5 0]\n",
      "f1 score_KNN: 0.0\n",
      "balanced accuracy_KNN: 0.5\n",
      "confusion matrix_KNN\n",
      "[[5 0]\n",
      " [5 0]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=20\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [5 0 5 0]\n",
      "f1 score_KNN: 0.0\n",
      "balanced accuracy_KNN: 0.5\n",
      "confusion matrix_KNN\n",
      "[[5 0]\n",
      " [5 0]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=20\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_KNN: 1.0\n",
      "balanced accuracy_KNN: 1.0\n",
      "confusion matrix_KNN\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=20\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [1 4 0 5]\n",
      "f1 score_KNN: 0.7142857142857143\n",
      "balanced accuracy_KNN: 0.6\n",
      "confusion matrix_KNN\n",
      "[[1 4]\n",
      " [0 5]]\n",
      "-> check results\n",
      "\n",
      "====== Step {shuffleStep + 1}/{self.numOfShuffles} =======\n",
      "-> Spliting data to slices\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=20\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [5 0 5 0]\n",
      "f1 score_KNN: 0.0\n",
      "balanced accuracy_KNN: 0.5\n",
      "confusion matrix_KNN\n",
      "[[5 0]\n",
      " [5 0]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=20\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [5 0 5 0]\n",
      "f1 score_KNN: 0.0\n",
      "balanced accuracy_KNN: 0.5\n",
      "confusion matrix_KNN\n",
      "[[5 0]\n",
      " [5 0]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=20\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [5 0 5 0]\n",
      "f1 score_KNN: 0.0\n",
      "balanced accuracy_KNN: 0.5\n",
      "confusion matrix_KNN\n",
      "[[5 0]\n",
      " [5 0]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=20\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_KNN: 1.0\n",
      "balanced accuracy_KNN: 1.0\n",
      "confusion matrix_KNN\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=20\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 5]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [1 4 0 5]\n",
      "f1 score_KNN: 0.7142857142857143\n",
      "balanced accuracy_KNN: 0.6\n",
      "confusion matrix_KNN\n",
      "[[1 4]\n",
      " [0 5]]\n",
      "-> check results\n",
      "### Exercise is done.\n"
     ]
    }
   ],
   "source": [
    "exercise.run(gan, loadEqualDataset())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "forward-throat",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "### Start exercise for synthetic point generator\n",
      "\n",
      "====== Step {shuffleStep + 1}/{self.numOfShuffles} =======\n",
      "-> Spliting data to slices\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=9\n",
      "-> create 11 synthetic samples\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 3]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 3]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 3]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 3]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [0 5 0 3]\n",
      "f1 score_KNN: 0.5454545454545454\n",
      "balanced accuracy_KNN: 0.5\n",
      "confusion matrix_KNN\n",
      "[[0 5]\n",
      " [0 3]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=9\n",
      "-> create 11 synthetic samples\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 3]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 3]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 3]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 3]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [2 3 2 1]\n",
      "f1 score_KNN: 0.28571428571428575\n",
      "balanced accuracy_KNN: 0.3666666666666667\n",
      "confusion matrix_KNN\n",
      "[[2 3]\n",
      " [2 1]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=9\n",
      "-> create 11 synthetic samples\n",
      "{1: 20, 0: 20}\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWkAAAFpCAYAAABee9lOAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAAkxElEQVR4nO3dfXRU9YH/8U+GECEkYjIE1CRLFDuA2Ikg5kmwpwEUaCvy4PoAqU1hjWTRrmHlYaMJBomlx4eiXUysFVpSC5gA5WxwW6XnLLrHACvdZH9qBGsEDBDTmYADBjPJzO8P6pSQBCYwk/mGvF/ncJB7vzPzuTmTT75+7+TeMK/X6xUAwEiWUAcAAHSNkgYAg1HSAGAwShoADEZJA4DBKGkAMFh4qAP0Ng7HSXk8nX9qMSYmUk1NX/Vwou4h46UzPZ9ExkDpqYxxcdFd7mMmHUDh4f1CHeGCyHjpTM8nkTFQTMhISQOAwShpADAYJQ0ABqOkAcBglDQAGIyP4AVIW1urPv30U506ZfZHir74wiKPx+P3+PDwCMXExKlfP94qQCjwnRcgTU2NGjRokIYNG6KwsLBQx+lSeLhFra3+lbTX69WpU1+qqalRQ4ZcE+RkADrDckeAtLa2KCpqsNEF3V1hYWEaNOhKtba2hDoK0GdR0gF0ORX0Ny7HYwJ6E0oaAAxGSQOAwSjpPmzDhnWaMWOqJk26TcuXL1ZTkzPUkYCQWf/fn+neV3Zr/X9/Fuoo7VDSfVRl5Xb95jfrlJe3VCUlr8nlcmnFivxQxwJC5s0PGtTc0qY3P2gIdZR2KOk+qqJik+67b66+853v6lvfGqnlywv0/vt79emnn4Q6GhAS08YM08CIfpo2Zlioo7RDSfdBLS0t+uSTAxo3brxvW3x8gq655lp9+OH/C2EywD/BWJr40W1J2vRQqn50W1LAnjMQjC7p0tJSTZgwQcnJycrNzZXD4ehybF1dnbKysmS325WZmaktW7Z0Oq61tVWzZ8/WyJEj9fnnnwcrutG+/PKEPB6PYmJi222/6qqr1NTUFKJUgP9MXZoIBmNLuqKiQiUlJSosLNTGjRvlcrmUl5fX6Vi3262cnBxZrVaVl5dr4cKFKigo0J49ezqMLS0t1eDBg4Md/6K07XxL7vk/VNvOt4L6Ol5v53eWAXoLU5cmgsHYXwsvKytTdna2pkyZIkkqLi7W5MmTtX//ftlstnZjd+3apYaGBm3btk2RkZGy2Wzau3evysrKlJKS4hv30UcfaevWrfrFL36hGTNm9Ojx+MPz+gZ5Dx6U5/UN6jdpStBeZ/Dgq2SxWNTU5FRS0nW+7cePH1dMTEzQXhcIlB/dlmTcskSwGDmTbmlpUW1trdLS0nzbEhMTFR8fr+rq6g7ja2pqZLfbFRkZ6duWnp7ebmxLS4uWLl2qwsJCRUVFBfcALpLlgSyFDR8uywNZQX2diIgI3XDDt7Rv3//4th05Uq+jR4/oxhtvCupro+8x9aNtvYWRM+mmpiZ5PB5ZrdZ222NjY+V0dvwsr9Pp7HTs2WvYL730km666SZNnDjxktairdbOC/6LL878vAsPv/ife+F33indeedFP97v1wm36J577tULLzyn0aNH65prrtXPf/6sbrllfIf/S5Eki8Vy3htlBkNPv153mZ5PMifjf354Zv34Pz9s0ON3f7vdPlMynk+oMxpZ0t11oTXWmpoaVVZW6ve///0lv1ZXdwv/5vKf/l5hLlS+uQretGl36a9/dWj16lU6efKkbr01VUuW5Hea3+PxqLHR1WMZ4+Kie/T1usv0fJJZGafeOExvftCgqTcOa5fJpIxd6amM5/tBYGRJx8TEyGKxyOFwaMSIEb7tTqdTsbGxHcZbrVYdPHiw3bazZ9f79u3TkSNHlJqa2m7MHXfcodzcXC1atCgIR2G+rKxsZWVlhzoGLnN9af04GIws6YiICI0aNUq7d+/2nfg7fPiw6uvrlZyc3GG83W7X+vXr1dzcrIEDB0qSqqqqfGNnzJihjIwM3/gvvvhC8+fP1yuvvKLRo0f3wBEBwMUx8sShJM2dO1fr1q3T22+/rdraWuXn5ys1NVU2m001NTWaOnWqGhrOfEZy4sSJGjp0qPLz83XgwAGVl5ersrJS8+bNk3RmZm6z2Xx/kpKSJElJSUkd1rIBwCRGzqQlac6cOXI4HFqxYoVcLpcyMjK0cuVKSVJzc7Pq6urkdrslnZl5l5aWqrCwULNmzVJcXJyKioraffwOAHqjMC+/2dAtXZ04PHbsoBISrus1Jw6749ixg7r66uFBStSR6SeUTM8nkTFQTDhxaOxyBwCAkgYAo1HSAGAwShoADEZJA4DBKGkAMJixn5NG8PzXf/1JW7a8odraD3Xq1Cm9++7/XPhBfcSqyo+057MmpSTFKP97/DYqQo+ZdB90+vRpjRs3XvPm/SjUUYyz57Mmebxn/gZMwEy6D7rzzumS1O560jgjJSnGN5MGTEBJo9cKxtIESxwwDcsd6LVYmkBfQEkb5L2/OLR8y//Te3/p+q7o+LuUpBhZwsTSBC5rLHcYZHv1UdUfb9b26qNKH3F5XUJ1/X9/pjc/aNC0McMCdgF4libQFzCTNshdydco/qqBuiv5mlBHCbg3Pzhzn7s3P2gIdRSgV2EmbZD0EdYemUF/+eUJNTQcU339mRvyHjjwsSQpKel69e/fPyivOW3MMN9MGoD/KOk+6N13d6m4+Cnfv7Oz50qS3nhju6655tqgvCb3uQMuDiXdB02f/gNNn/6DLvf/9eTXOtHs1uCB/TUk6ooeTAbgXKxJo4MTzW55vGf+BhBalDQ6GDywvyxhZ/4GEFosd/RywViaGBJ1BcscgCGYSfdyLE0AlzdKOoBCceP1YC9NcDN5ILRY7ggQi6Wf2tpaJfXr0dcN9tJEW1urLJaePSYAf8dMOkAGDozSl182yev1dDnmrye/1l8aT+qvJ7/uwWQXz+v1yOVq0sCBUaGOAvRZzKQDJCpqsJqbm9TQ8LmkzpcIGl1fy+uVvjoutUaH5sScxWKRx9P1D5L2whQRMUBRUYODmglA1yjpAAkLC9Pw4cPV2Ojqcsx//uWsiwx9a3gPpvu7uLjo82YEYBZKugfxq9EAuos1aQAwGCUNAAajpAHAYJQ0ABiMkgYAg1HSAGAwShoADEZJA4DBKGkAMBglDQAGo6QBwGCUNAAYjJIGAINR0gBgMEoaAAxGSQOAwShpADAYJQ0ABqOkAcBglDQAGIySBgCDUdIAYDBKGgAMRkkDgMEoaQAwGCUNAAajpAHAYJQ0ABjM6JIuLS3VhAkTlJycrNzcXDkcji7H1tXVKSsrS3a7XZmZmdqyZYtvn9vt1urVqzVt2jQlJycrMzNTL774olpbW3viMADgohlb0hUVFSopKVFhYaE2btwol8ulvLy8Tse63W7l5OTIarWqvLxcCxcuVEFBgfbs2SNJOn36tA4cOKC8vDxt375dhYWF2rRpk0pKSnrykACg28JDHaArZWVlys7O1pQpUyRJxcXFmjx5svbv3y+bzdZu7K5du9TQ0KBt27YpMjJSNptNe/fuVVlZmVJSUhQdHa1XX33VN3748OHKzs5WZWWlFi1a1KPHBQDdYeRMuqWlRbW1tUpLS/NtS0xMVHx8vKqrqzuMr6mpkd1uV2RkpG9benp6p2O/0dTUpOjo6MAGB4AAM3Im3dTUJI/HI6vV2m57bGysnE5nh/FOp7PTsV2tYR85ckSbN2/WE0880e1sVmvUeffHxZlf/GS8dKbnk8gYKKHOaGRJd5fX6/V77IkTJ5STk6NJkyZpxowZ3X4th+OkPJ7OXy8uLlqNja5uP2dPIuOlMz2fRMZA6amM5/tBYORyR0xMjCwWS4eZsNPpVGxsbIfxVqu107Hnzq5PnTqlhx56SAkJCXr66acDHxwAAszIko6IiNCoUaO0e/du37bDhw+rvr5eycnJHcbb7XbV1NSoubnZt62qqqrd2NOnT+vhhx/WgAEDtGbNGoWHXxb/EwHgMmdkSUvS3LlztW7dOr399tuqra1Vfn6+UlNTZbPZVFNTo6lTp6qhoUGSNHHiRA0dOlT5+fk6cOCAysvLVVlZqXnz5kk68xG9Rx55RE6nU0VFRTpx4oQaGxvV2NgYykMEgAsydjo5Z84cORwOrVixQi6XSxkZGVq5cqUkqbm5WXV1dXK73ZLOzLxLS0tVWFioWbNmKS4uTkVFRUpJSZEkNTQ0aNeuXZKkO+64o93rfPzxxz14VADQPWHe7px1AycOe4DpGU3PJ5ExUDhxCAA4L0oaAAxGSQOAwShpADAYJQ0ABqOkAcBglDQAGIySBgCDUdIAYDBKGgAMRkkDgMEoaQAwGCUNAAajpAHAYJQ0ABiMkgYAg1HSAGAwShoADEZJA4DBKGkAMBglDQAGo6QBwGCUNAAYjJIGAINR0gBgMEoaAAxGSQOAwShpADAYJQ0ABqOkAcBglDQAGIySBgCDUdIAYDBKGgAMRkkDgMEoaQAwGCUNAAajpAHAYJQ0ABiMkgYAg1HSAGAwShoADEZJA4DBKGkAMFhASvqrr77S3r17A/FUAICzBKSkDx06pB/+8IeBeCoAwFlY7gAAg4X7M+imm24Kdg4AQCf8Kunw8HDdf//9stlsne6vr6/Xv//7vwc0GADAz5K22Wy69tprNXPmzE7319bWUtIAEAR+rUnffPPNOnjwYJf7IyMjNX78+ICFAgCc4ddM+t/+7d/Ou/8f/uEftGHDhoAEAgD8XcA+3XHo0KFAPRUA4G8uqaRbW1u1Y8cOPfjgg5o6dWqgMgEA/uaiSvqzzz7Tz372M91+++1asWKFrr76ar388suBzqbS0lJNmDBBycnJys3NlcPh6HJsXV2dsrKyZLfblZmZqS1btrTb39raqmeeeUapqakaN26cli9frq+++irgmQEgkPwuabfbrcrKSmVlZen73/++PvnkEx0/flxlZWVavXq1vvOd7wQ0WEVFhUpKSlRYWKiNGzfK5XIpLy+vy2w5OTmyWq0qLy/XwoULVVBQoD179vjGrF27VpWVlfr5z3+u9evXq6amRitXrgxoZgAINL9OHK5evVrbtm2T1WrV3Xffreeff15xcXEaM2aMLJbg/NJiWVmZsrOzNWXKFElScXGxJk+erP3793f4vPauXbvU0NCgbdu2KTIyUjabTXv37lVZWZlSUlLk8Xj0+uuv6/HHH1d6erok6YknntCCBQu0fPlyXXnllUE5BgC4VH417K9//Wvdf//92r59uxYsWKC4uLighmppaVFtba3S0tJ82xITExUfH6/q6uoO42tqamS32xUZGenblp6e7ht7+PBhNTU1tXu+lJQUeb1effDBB0E8EgC4NH6V9OLFi7Vjxw7dfvvtWr16tfbv3x/UUE1NTfJ4PLJare22x8bGyul0dhjvdDo7HfvNGvY3f589pl+/fho8ePB517kBINT8Wu6YP3++5s+fr6qqKm3evFmzZ8/WDTfcIK/Xa8TJN6/Xe0n7u8NqjTrv/ri46IC9VrCQ8dKZnk8iY6CEOqNfJf2NtLQ0paWlyel0qqKiQl999ZUeeOAB3XrrrZo+fbruueeegISKiYmRxWKRw+HQiBEjfNudTqdiY2M7jLdarR1+I/Ls2fWQIUMknZlRx8fHS5La2tp04sSJDjPwC3E4Tsrj6bz04+Ki1djo6tbz9TQyXjrT80lkDJSeyni+HwQXddYvNjZW//RP/6Q//OEP+uUvf6krr7xSRUVFFx3wXBERERo1apR2797t23b48GHV19crOTm5w3i73a6amho1Nzf7tlVVVfnGJiYmKiYmpt3z7d27V2FhYbrxxhsDlhsAAq1bM+nOpKenKz09vdO14ksxd+5cFRcXa/To0UpISFBxcbFSU1Nls9lUU1OjJUuW6Ne//rWGDRumiRMnaujQocrPz9fChQtVXV2tyspKvfbaa5Iki8Wi+++/Xy+88IKuvfZaRUZGatWqVbrrrrs0ePDggOYGgEDyayb94Ycfat68eXK5Ok77v/zyS2VlZamxsTGgwebMmaOcnBytWLFC9957rwYNGqTnn39ektTc3Ky6ujq53W5JZ2bepaWlamxs1KxZs7R27VoVFRUpJSXF93z//M//rGnTpunRRx/Vgw8+qDFjxujJJ58MaGYACLQwrx9n1R5//HElJCToJz/5Saf7X3rpJdXX1+unP/1pwAOahjXp4DM9o+n5JDIGSq9Zk66urtbkyZO73D9p0iS9//773U8GADgvv0r62LFjuuqqq7rcP3jwYDU0NAQqEwDgb/wq6ejo6PNeivTQoUOKjjb/844A0Nv4VdLjxo1TeXl5l/vfeOMNjR07NmChAABn+PURvB//+MeaN2+errzySj388MMaNmyYJKmhoUEvv/yy/vCHP3BnFgAIAr9KeuzYsSoqKtJTTz2ljRs3KirqzK9Gnzx5Uv3799eKFSs0bty4oAYFgL7I719mmT17tiZMmKCKigo1NjbK6/Xquuuu09SpU30zawBAYPlV0sePH9eyZcv0zjvvyOPxKDk5Wc8++6wSEhKCnQ8A+jS/Thy+8MILqq6u1iOPPKIlS5bI4XCosLAw2NkAoM/zayb9zjvvaNWqVcrMzJQkTZw4UXfddZfcbrf69+8f1IAA0Jf5NZNuaGjQmDFjfP++4YYb1L9//4BfrwMA0J5fJd3W1tZhxmyxWNTW1haUUACAM/z+dMdPfvKTdkXd0tKipUuXasCAAb5t31waFAAQGH6V9MyZMztsu+uuuwIeBgDQnl8l/cwzzwQ7BwCgExd1+ywAQM+gpAHAYJQ0ABiMkgYAg1HSAGAwShoADEZJA4DBKGkAMBglDQAGo6QBwGCUNAAYjJIGAINR0gBgMEoaAAxGSQOAwShpADAYJQ0ABqOkAcBglDQAGIySBgCDUdIAYDBKGgAMRkkDgMEoaQAwGCUNAAajpAHAYJQ0ABiMkgYAg1HSAGAwShoADEZJA4DBKGkAMBglDQAGo6QBwGCUNAAYjJIGAINR0gBgMEoaAAxGSQOAwShpADCYsSVdWlqqCRMmKDk5Wbm5uXI4HOcdX1dXp6ysLNntdmVmZmrLli2+fW63W6tXr9a0adOUnJyszMxMvfjii2ptbQ32YQDAJTGypCsqKlRSUqLCwkJt3LhRLpdLeXl5XY53u93KycmR1WpVeXm5Fi5cqIKCAu3Zs0eSdPr0aR04cEB5eXnavn27CgsLtWnTJpWUlPTUIQHARQkPdYDOlJWVKTs7W1OmTJEkFRcXa/Lkydq/f79sNluH8bt27VJDQ4O2bdumyMhI2Ww27d27V2VlZUpJSVF0dLReffVV3/jhw4crOztblZWVWrRoUY8dFwB0l3Ez6ZaWFtXW1iotLc23LTExUfHx8aquru70MTU1NbLb7YqMjPRtS09P73K8JDU1NSk6OjpwwQEgCIwr6aamJnk8Hlmt1nbbY2Nj5XQ6O32M0+nsdHxX69hHjhzR5s2bNXv27MCEBoAg6dHljmXLlmnr1q1d7p85c6Yee+yxbj+v1+v1e+yJEyeUk5OjSZMmacaMGd1+Las16rz74+LMn52T8dKZnk8iY6CEOmOPlnR+fr4WL17c5f4BAwboiiuukMVikcPh0IgRI3z7nE6nYmNjO32c1WrVwYMH223rbHZ96tQpPfTQQ0pISNDTTz99UcfgcJyUx9P5D4W4uGg1Nrou6nl7Chkvnen5JDIGSk9lPN8Pgh4t6ejoaL/WgUeNGqXdu3crJSVFknT48GHV19crOTm50/F2u13r169Xc3OzBg4cKEmqqqpqN/706dN6+OGHNWDAAK1Zs0bh4UaeMwWAdoxbk5akuXPnat26dXr77bdVW1ur/Px8paam+j7ZUVNTo6lTp6qhoUGSNHHiRA0dOlT5+fk6cOCAysvLVVlZqXnz5kk68xG9Rx55RE6nU0VFRTpx4oQaGxvV2NgYsmMEAH8YOZ2cM2eOHA6HVqxYIZfLpYyMDK1cudK3v7m5WXV1dXK73ZKkiIgIlZaWqrCwULNmzVJcXJyKiop8M/GGhgbt2rVLknTHHXe0e62PP/64h44KALovzNuds25gTboHmJ7R9HwSGQPFhDVpI5c7AABnUNIAYDBKGgAMRkkDgMEoaQAwGCUNAAajpAHAYJQ0ABiMkgYAg1HSAGAwShoADEZJA4DBKGkAMBglDQAGo6QBwGCUNAAYjJIGAINR0gBgMEoaAAxGSQOAwShpADAYJQ0ABqOkAcBglDQAGIySBoAAadv5ltzzf6i2nW8F7DkpaQAIEM/rG+Q9eFCe1zcE7DkpaQC4AH9nyJYHshQ2fLgsD2QF7LUpaQB9UneWJvydIfebNEX9f/Ub9Zs0JVAxKWkAfVN3liaCMUP2FyUN4LISjKWJYMyQ/RXe468IAEF09gz5fKXab9KUkJRudzGTBmC87qwfh3JpIhiYSQMwnr+zY6n3zJD9xUwaQMiE8qNtvQUlDSCgestH23oLShpAQPWWj7b1FpQ0gAsK1om7vjxD9hcnDgFcUF8+cRdqzKSBPowTd+ajpIHLTNvOt9Q4ew4n7i4TlDRwmfG8vkGtn9Zx4u4ywZo0cJmxPJClfm/8Tm333H/Bsawfm4+SBi4z/SZNUdx9s9TY6Ap1FAQAyx0AYDBKGgAMRkkDgMEoaQAwGCUNAAajpAHAYJQ0ABiMksZlqTtXbQNMRknjstSdaxoDJqOk0Wv05ZuRou+ipBFy/pZvd2bHXLUNlwtKGiHnb/kyO0ZfZGxJl5aWasKECUpOTlZubq4cDsd5x9fV1SkrK0t2u12ZmZnasmVLp+NaW1s1e/ZsjRw5Up9//nkwokPBWZpgdoy+yMiSrqioUElJiQoLC7Vx40a5XC7l5eV1Od7tdisnJ0dWq1Xl5eVauHChCgoKtGfPng5jS0tLNXjw4GDGh1iaAALFyJIuKytTdna2pkyZotGjR6u4uFhVVVXav39/p+N37dqlhoYGFRcXy2az6Z577tH06dNVVlbWbtxHH32krVu3asmSJT1xGJclbrcE9CzjSrqlpUW1tbVKS0vzbUtMTFR8fLyqq6s7fUxNTY3sdrsiIyN929LT09uNb2lp0dKlS1VYWKioqKjgHUAv1J2lCW63BPQs40q6qalJHo9HVqu13fbY2Fg5nc5OH+N0Ojsdf/Y69ksvvaSbbrpJEydODHzoXq47SxPMkIGe1aN3Zlm2bJm2bt3a5f6ZM2fqscce6/bzer3e8+6vqalRZWWlfv/733f7uc9ltZ5/Fh4XF33JrxEIzW++qZOv/kpRC+Zr4LRp7fadm7F54UM6+eprilrwYw28UP77Zp35E2SmfB27Yno+iYyBEuqMPVrS+fn5Wrx4cZf7BwwYoCuuuEIWi0UOh0MjRozw7XM6nYqNje30cVarVQcPHmy37ezZ9b59+3TkyBGlpqa2G3PHHXcoNzdXixYt8vsYHI6T8ng6/6EQFxdtzC2L3C+/Iu/Bg2p6+RWdHD/Bt73TjOMnSOMn6KSkkwbkN+nr2BnT80lkDJSeyni+HwQ9WtLR0dGKjr7wT6VRo0Zp9+7dSklJkSQdPnxY9fX1Sk5O7nS83W7X+vXr1dzcrIEDB0qSqqqqfONnzJihjIwM3/gvvvhC8+fP1yuvvKLRo0df6mH1qLadb8nz+oYzNxs9z3qv5YEs3zgAvZdxa9KSNHfuXK1bt05vv/22amtrlZ+fr9TUVNlsNklnli+mTp2qhoYGSdLEiRM1dOhQ5efn68CBAyovL1dlZaXmzZsnSYqJiZHNZvP9SUpKkiQlJSV1WMs2HSfugL7FyJKeM2eOcnJytGLFCt17770aNGiQnn/+ed/+5uZm1dXVye12S5IiIiJUWlqqxsZGzZo1S2vXrlVRUZFvJm46rkkBoCth3guddUM73VmT9ndpwj3/h/IePKiw4cPV/1e/CXjm82U0kekZTc8nkTFQTFiTNnImfbngmhQALlWPnjjsa/w9eddv0hTWjgF0ipIOIsoXwKViuQMADEZJA4DBKGkAMBglDQAGo6QBwGCUNAAYjJIGAINR0gBgMEoaAAxGSQOAwfi18G6yWMIuab8JyHjpTM8nkTFQQp2RS5UCgMFY7gAAg1HSAGAwShoADEZJA4DBKGkAMBglDQAGo6QBwGCUNAAYjJIGAINR0gBgMEo6CF588UXdeeedstvtysjI0OLFi9XY2BjqWO28/PLLmjFjhm6++Wbdfvvtevrpp3Xq1KlQx/L54x//qAcffFC33HKLRo4cGeo4PqWlpZowYYKSk5OVm5srh8MR6kg+pn7Nzmb6++4bJn0PU9JBcP311+upp57Sjh07VFJSoqNHj2rZsmWhjtXOn//8Zy1YsEBbtmzRc889p3fffVdPP/10qGP5NDc3Ky0tTQ899FCoo/hUVFSopKREhYWF2rhxo1wul/Ly8kIdy8fEr9m5TH/ffcOo72Evgm7nzp3em2++OdQxzmvHjh3eW2+9NdQxOqiqqvLabLZQx/B6vV7v3Xff7V2zZo3v34cOHfLabDbvxx9/HMJUHZn0NbsQU9935wrl9zAz6SBzuVz6j//4D40bNy7UUc6rqalJ0dHRoY5hrJaWFtXW1iotLc23LTExUfHx8aqurg5hst6tN7zvQv09TEkHyfbt2zV27FiNHz9en3/+uZ577rlQR+qSy+XSa6+9ptmzZ4c6irGamprk8XhktVrbbY+NjZXT6QxRqt7N9PedKd/DXPS/G5YtW6atW7d2uX/mzJn66U9/KknKzMyU3W7X0aNH9dJLL6mgoEAvvviiURmlMzPERx55RImJiT2yltndfLg89fT77mKE6nv4XJR0N+Tn52vx4sVd7h8wYIDvv6OiohQVFaWkpCRdf/31uv322/XJJ5/ohhtuMCZja2urHnvsMZ06dUrr169XeHjw3w7dyWeSmJgYWSwWORwOjRgxwrfd6XQqNjY2hMl6n1C87y5GqL6Hz2XmV8dQ0dHRF7V+5v3bzW8sluCvLvmb0ePxaOnSpTp06JA2bNigQYMGBT2bdPFfw1CLiIjQqFGjtHv3bqWkpEiSDh8+rPr6eiUnJ4c4Xe8RqvfdperJ7+FzUdIB5na79Ytf/EKTJk2S1WrV0aNHtWbNGo0ZM0ZJSUmhjufz5JNPavfu3frlL38pt9vt+wxobGys+vXrF+J00vHjx3X06FEdOnRIkvTRRx9JkkaMGKGIiIiQZJo7d66Ki4s1evRoJSQkqLi4WKmpqbLZbCHJcy4Tv2bnMv19J5n3Pcw9DgOstbVV//Iv/6L//d//1fHjxxUXF6fbbrtNjz76qIYOHRrqeD5d/bLDzp07lZCQ0MNpOtqyZYuWL1/eYXuo85WWlmrDhg1yuVzKyMjQypUrNWTIkJDlOZupX7Ozmf6+k8z7HqakAcBgfAQPAAxGSQOAwShpADAYJQ0ABqOkAcBglDQAGIySBgCDUdLAOZYtW6aRI0dq5MiRuvHGG/Xd735XBQUFampq8o3585//rEWLFikjI0Pf/va3NXnyZP3rv/6rPvjggw7P96tf/UqjR4/WM8880+nrnTx5Uk888YRSU1N18803a8GCBb7fGgQoaaAT48eP17vvvqs//elPys/P1x//+EctXbpU0pk7tMydO1fh4eF69tlntWPHDr3wwguKj4/XqlWrOjzX5s2blZOTo23btqmlpaXD/iVLlui9997TmjVr9Prrr8vr9So7O1unT58O+nGiFwjJrQYAgy1dutT74IMPttu2du1a76hRo7zHjh3z3nTTTd4nn3yy08ceP3683b/fe+89b3p6utftdnunT5/u3b59e7v9n376qddms3nfeeedds8xZswYb0VFRWAOCL0aM2nADwMGDJDH41F5eblaWlqUm5vb6bjBgwe3+/emTZv0gx/8QOHh4br77ru1efPmdvv37dun/v37Kz09vd1z2O12vf/++4E/EPQ6lDRwAZ988ol++9vfKjk5WY2NjYqKitLVV199wcc5nU699dZbmjlzpiRpxowZ2rdvnz799FPfmMbGRl111VUdrgA3ZMgQ4+4wj9CgpIFO7NmzR2PHjpXdbtf3v/99JSYm6rnnnvNdV9gfFRUVGjFihEaNGiVJGjp0qDIyMvTGG2/49fiwsLCLyo7LC9eTBjpht9u1evVq9evXT0OHDvVdj/m6667TyZMndezYsfPOpr1er8rLy3Xw4EHdeOONvu0ej0f/93//p8cee0wRERGKi4vT8ePH1dbW1m427XA4jLr+OEKHmTTQiQEDBmj48OFKSEhod8H8adOmKSIiQmvXru30cSdOnJAkVVVV6fDhw/rd736nbdu2tfvT2tqqt956S5I0btw4ud1uVVVV+Z7jyy+/VHV1tW655ZYgHiF6C2bSQDcMGzZMBQUFKigokMvl0j/+4z8qMTFRJ06c0M6dO7V792799re/1caNG3Xrrbdq7NixHZ4jMzNTmzZt0ve+9z1dd911mjRpklasWKFVq1YpOjpazz//vIYNG6bp06eH4AhhGmbSQDfdc8892rBhg77++mvl5eVp6tSpevTRR/X555/riSeekMPh0M6dOzVt2rROHz99+nTt2bNHn332mSTpZz/7mVJSUrRo0SLdd9998ng8eu2114y9KS96FndmAQCDMZMGAINR0gBgMEoaAAxGSQOAwShpADAYJQ0ABqOkAcBglDQAGIySBgCD/X+0zv5IP6rVUQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "tn, fp, fn, tp: [5 0 0 3]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 3]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 3]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 3]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [1 4 2 1]\n",
      "f1 score_KNN: 0.25\n",
      "balanced accuracy_KNN: 0.26666666666666666\n",
      "confusion matrix_KNN\n",
      "[[1 4]\n",
      " [2 1]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=9\n",
      "-> create 11 synthetic samples\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 3]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 3]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 3]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 3]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [2 3 3 0]\n",
      "f1 score_KNN: 0.0\n",
      "balanced accuracy_KNN: 0.2\n",
      "confusion matrix_KNN\n",
      "[[2 3]\n",
      " [3 0]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=12\n",
      "-> create 8 synthetic samples\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5]\n",
      "average_pr_score: nan\n",
      "f1 score_LR: 0.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5]\n",
      "f1 score_SVM: 0.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [0 5 0 0]\n",
      "f1 score_KNN: 0.0\n",
      "balanced accuracy_KNN: 0.0\n",
      "confusion matrix_KNN\n",
      "[[0 5]\n",
      " [0 0]]\n",
      "-> check results\n",
      "\n",
      "====== Step {shuffleStep + 1}/{self.numOfShuffles} =======\n",
      "-> Spliting data to slices\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=9\n",
      "-> create 11 synthetic samples\n",
      "{1: 20, 0: 20}\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/kristian/anaconda3/lib/python3.8/site-packages/sklearn/metrics/_ranking.py:681: RuntimeWarning: invalid value encountered in true_divide\n",
      "  recall = tps / tps[-1]\n",
      "/home/kristian/anaconda3/lib/python3.8/site-packages/sklearn/metrics/_classification.py:1464: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 due to no true nor predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(\n",
      "/home/kristian/anaconda3/lib/python3.8/site-packages/sklearn/metrics/_classification.py:1464: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 due to no true nor predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(\n",
      "/home/kristian/anaconda3/lib/python3.8/site-packages/sklearn/metrics/_classification.py:1814: UserWarning: y_pred contains classes not in y_true\n",
      "  warnings.warn('y_pred contains classes not in y_true')\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",
      "tn, fp, fn, tp: [5 0 0 3]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 3]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 3]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 3]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [5 0 3 0]\n",
      "f1 score_KNN: 0.0\n",
      "balanced accuracy_KNN: 0.5\n",
      "confusion matrix_KNN\n",
      "[[5 0]\n",
      " [3 0]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=9\n",
      "-> create 11 synthetic samples\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 3]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 3]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 3]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 3]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [4 1 3 0]\n",
      "f1 score_KNN: 0.0\n",
      "balanced accuracy_KNN: 0.4\n",
      "confusion matrix_KNN\n",
      "[[4 1]\n",
      " [3 0]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=9\n",
      "-> create 11 synthetic samples\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5 0 0 3]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 3]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 3]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 3]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [3 2 3 0]\n",
      "f1 score_KNN: 0.0\n",
      "balanced accuracy_KNN: 0.3\n",
      "confusion matrix_KNN\n",
      "[[3 2]\n",
      " [3 0]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=9\n",
      "-> create 11 synthetic samples\n",
      "{1: 20, 0: 20}\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWkAAAFpCAYAAABee9lOAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAAl0UlEQVR4nO3dfVTT590/8DcRKTKjkhhsBU71tgv40ERdBaHQeyJ1aFcpPuzXFmnr7ESpuiNuPpxYwlBp3emT1lbSdsVO26EFpPbg1io7u203Ae/aQ3bbonRFpVhTlkQNNoxg8vvDNZPyRUHzcIW8X+f0WL/fK998vlc5b65+ciUJcblcLhARkZBk/i6AiIh6x5AmIhIYQ5qISGAMaSIigTGkiYgExpAmIhIYQ5qISGBCh7TBYEBKSgq0Wi3y8vJgNpt7Hdvc3IycnBxoNBqkpaWhsrKyx5i//e1vWLhwIe6++25Mnz4dxcXF3iyfiOiWCRvSFRUVKCkpgV6vR1lZGWw2G/Lz8yXHOhwO5ObmQqlUory8HMuXL0dBQQHq6+vdY2pra7Fy5UrMnTsXBw4cwFtvvYXk5GRf3Q4R0U0JEfUdh1lZWZgxYwZWrVoFAGhpaUF6ejref/99qNXqbmNramqQn5+Po0ePIiIiAgCwdu1adHR0YPv27QCAefPmIT09HXl5eb69ESKiWxDq7wKkdHZ2orGxERs2bHAfi42NRXR0NBoaGnqEtNFohEajcQc0ACQlJeGll14CALS1teHEiRPIzMzE/PnzYTKZMHnyZOh0Otxxxx39qs1sbofTKeTvNQBAZGQErNZv/V2GcDgv0jgv0nw9LyqVvNdzQoa01WqF0+mEUqnsdlyhUMBisfQYb7FYJMd+18NubW0FcLXHrdPpEBsbi1deeQVLly5FVVUVBg0a1OfalMqh/b0dn7vef/BgxnmRxnmRJsq8CBnS/XWjjo3T6QQAPProo3jggQcAAM8++yzuvfdeGI1GTJkypc/PJfpKWqWSo63N5u8yhMN5kcZ5kebrebneLwQhXziMjIyETCbrsZvDYrFAoVD0GK9UKiXHfre6HjlyJABg7Nix3Z4jMjISX3/9tafLJyLyGCFDOiwsDPHx8airq3Mfa2lpQWtrK7RabY/xGo0GRqMRdrvdfay2ttY9NiYmBkqlEmfOnHGfv3TpEqxWK0aPHu3FOyEiujVChjQAZGdno7S0FIcPH0ZjYyN0Oh0SExOhVqthNBqRkZEBk8kEAEhNTUVUVBR0Oh2amppQXl6O6upqLFq0CAAgk8nw2GOP4a233sJf/vIX/OMf/8DGjRuhVqtx9913+/M2iYiuS9ie9IIFC2A2m1FYWAibzYbk5GRs2rQJAGC329Hc3AyHwwHg6srbYDBAr9dj3rx5UKlUKCoqQkJCgvt6S5cuhd1ux4YNG9DV1YVp06ahpKSkXy8aXo/L5YLV2obOzg4A/utZf/ONzN2Dv3UhCAsLR2SkCiEhIR66JhH1h7D7pEXV2wuHNtsFdHU5MGKEEiEh/vsflNBQGbq6PBPSLpcTFy78E6GhYZDLR3jkmv7CF8ikcV6k8YXDAchub4dcPsKvAe1pISEyyOWRsNvb/V0KUdAaOIniZ07nFQwaJGz36KYNGhQKp/OKv8sgCloMaQ8aiH3bgXhPRIGEIU1EJDCGtBe5Ll2C83QzXJcu+bsUIgpQDGkvclnMwL/+dfVPAe3eXYrMzAzMnHkvNmxYA6u15+eiEJF/MaS9KEShBG677eqfgqmuPoDf/74U+fnrUFLyJmw2GwoLdf4ui4i+hyHtRSHDhkE2ZixChg3zdyk9VFTsxcMPZ+O//3sGfvjDOGzYUIBPPjmGL7/8wt+lEdE1GNIC8HXvurOzE1980YSpU+9xH4uOjsEdd4zGZ5/9n09qIKK+YUgLwNe960uXLsLpdCIysvsnCo4YMQJWq9UnNRBR3zCkBfBd79r597/DseQxXKk55NXn4ycBEAUOhrQAvutdu6rK4TpzBs53dvc61hOtkeHDR0Amk/XYzXHhwgVERkbe9HWJyPMY0gKRPZqDkDvvhOzRnF7HeKI1EhYWhrvu+iGOH/9f97Fz51rx9dfnMGHCpJu+LhF53sD7sIkANmjm/Rg08/7rjglRKOGymG95W9+8eT/D9u0vQK2Ow+23j8bLL7+AqVPvwX/91123dF0i8iyGdIAJGTbMI1v6fvrTTFitFjz33DNob2/HtGmJWLuW+6SJRMOQHqBcly65V9y9hXpOzmLk5Cz2cWVE1B/sSQ9Qor8lnYj6hiE9QIn8lnQi6ju2OwYoT/Wuici/uJImfqQqkcAY0sT+NZHAGNLE/jWRwBjS1K+PVL1Sc8gnny9CRFcxpKlfnO/svuHnixCR5zCkqV/68vkiROQ53IIXhP7nf/6Mysp30dj4GS5fvoyPP/7fGz/o3/ry+SJE5DlcSQehjo4OTJ16DxYtesJrz8HeNZFncCUdhH7ykzkA0O2jSj3t2t41V95EN48rafIK9q6JPIMrafKK/vSur9QcgvOd3ZA9msNVN9H3cCVNfsdtfUS9Y0gL5Og/zNhQ+X84+o/gens2WyNEvWO7QyAHGr5G6wU7DjR8jaRxwfMWbW7rI+odV9ICmau9A9EjhmCu9g6vPs+lSxfR1HQSra1fAQCamk6iqekkHA6HV5/3VnFbHwUjrqQFkjRO6ZMV9McfH0Fx8W/cf1+8OBsA8O67B3DHHaO9/vw3i9v6KBgxpIPQnDkPYs6cB/1dRr/JHs1x7wIhChYMaQoY3NZHwYg9aRqQuK2PBgqGNA1I3NZHAwXbHTQgcVsfDRRCr6QNBgNSUlKg1WqRl5cHs7n3N3k0NzcjJycHGo0GaWlpqKyslBzX1dWF+fPnIy4uDl999ZW3SqcAwW19JDphQ7qiogIlJSXQ6/UoKyuDzWZDfn6+5FiHw4Hc3FwolUqUl5dj+fLlKCgoQH19fY+xBoMBw4cP90rNLpfLK9f1p4F4T9di75pEJ2xI79mzB4sXL8b999+P8ePHo7i4GLW1tTh16lSPsUeOHIHJZEJxcTHUajUWLlyIOXPmYM+ePd3Gff7559i/fz/Wrl3r8XpDQ8Nw+fKlARVqLpcLly9fQmhomL9L8Rr2rkl0QvakOzs70djYiA0bNriPxcbGIjo6Gg0NDVCr1d3GG41GaDQaREREuI8lJSXhpZde6nbNdevWQa/XY+jQoR6vOTJSBau1De3tFzx+7f6QyWRwOp0eu15oaBgiI1Ueu55o2Lsm0QkZ0larFU6nE0pl93ffKRQKWCyWHuMtFovk2Gt72C+//DImTZqE1NTUW+pFK5W9B/ztt0fe9HXJ+1Qq+S093v7HP6L9jd9h6JNLMGT2bA9V5X+3Oi8DlSjzImRI99eNWgxGoxHV1dV47733bvm5zOZ2OJ3itjRUKjna2mz+LkM4npgXx87X4DpzBtadr6H9nhQPVeZf/HmR5ut5ud4vBCF70pGRkZDJZD12c1gsFigUih7jlUql5NjvVtfHjx/HuXPnkJiYiAkTJmDWrFkAgFmzZmHHjh1eugsaaNi/Jn8QciUdFhaG+Ph41NXVISEhAQDQ0tKC1tZWaLXaHuM1Gg127doFu92OIUOGAABqa2vdYzMzM5GcnOwe/80332DJkiV47bXXMH78eB/cEQ0EfFs6+YOQK2kAyM7ORmlpKQ4fPozGxkbodDokJiZCrVbDaDQiIyMDJpMJAJCamoqoqCjodDo0NTWhvLwc1dXVWLRoEYCrK3O1Wu3+Z8yYMQCAMWPG9OhlE3kCt/aRpwi5kgaABQsWwGw2o7CwEDabDcnJydi0aRMAwG63o7m52f35x2FhYTAYDNDr9Zg3bx5UKhWKiorcq3AiX+Mn9pGnhLgG0sZeH+ALh4GJ8yKN8yKNLxwSEQC+LZ1ujCFN5EfsXdONMKSJ/Ijb+uhGhH3hkCgYcFsf3QhX0kQBgq2R4MSQJgoQbI0EJ7Y7iAIEP7EvOHElTUQkMIY0EZHAGNJERAJjSBMRCYwhTUQkMIY0EZHAGNJERAJjSBMRCYwhTUQkMIY0EZHAGNJERAJjSBMRCYwhTUQkMIY0EZHAGNJERAJjSBMRCYwhTUQkMIY0EZHAGNJEJLwrNYfgWPIYrtQc8ncpPseQJiLhBfM3pTOkiUh4wfxN6fy2cCISXn++Kf1KzSE439kN2aM5A+Lb1bmSJqIBZaC1RhjSRDSgDLTWCNsdRDSg9Kc1Egi4kiaioBQo2/oY0kQUlAKld82QJqKgFCi9a/akiSgoBcq2Pq6kiYhuwJ+tEYY0EdEN+LM1wnYHEdEN+HNbH1fSREQe4o1tfUKHtMFgQEpKCrRaLfLy8mA2m3sd29zcjJycHGg0GqSlpaGystJ9zuFwYOvWrZg9eza0Wi3S0tKwfft2dHV1+eI2iChIeKN3LWxIV1RUoKSkBHq9HmVlZbDZbMjPz5cc63A4kJubC6VSifLycixfvhwFBQWor68HAHR0dKCpqQn5+fk4cOAA9Ho99u7di5KSEl/eEhENcN7oXYe4XC6Xx67mQVlZWZgxYwZWrVoFAGhpaUF6ejref/99qNXqbmNramqQn5+Po0ePIiIiAgCwdu1adHR0YPv27ZLXf+ONN1BdXY39+/f3qy6zuR1Op5BTBgBQqeRoa7P5uwzhcF6kcV6k+XpeVCp5r+eEXEl3dnaisbER06dPdx+LjY1FdHQ0Ghoaeow3Go3QaDTugAaApKQkybHfsVqtkMt7nxgiIhEIubvDarXC6XRCqVR2O65QKGCxWHqMt1gskmN762GfO3cO+/btw8aNG/tdm1I5tN+P8bXr/VYOZpwXaZwXaaLMi5Ah3V/96dhcvHgRubm5mDlzJjIzM/v9XGx3BCbOizTOizS2O24gMjISMpmsx0rYYrFAoVD0GK9UKiXHfn91ffnyZSxduhQxMTHYvHmz5wsnIvIwIUM6LCwM8fHxqKurcx9raWlBa2srtFptj/EajQZGoxF2u919rLa2ttvYjo4OLFu2DOHh4di2bRtCQwfE/0QQ0QAnZEgDQHZ2NkpLS3H48GE0NjZCp9MhMTERarUaRqMRGRkZMJlMAIDU1FRERUVBp9OhqakJ5eXlqK6uxqJFiwBc3aK3cuVKWCwWFBUV4eLFi2hra0NbW5s/b5GI6IaEXU4uWLAAZrMZhYWFsNlsSE5OxqZNmwAAdrsdzc3NcDgcAK6uvA0GA/R6PebNmweVSoWioiIkJCQAAEwmE44cOQIAmDVrVrfnOXnypA/vioiof4TdJy0qvnAYmDgv0jgv0vjCIRER9QlDmohIYAxpIiKBMaSJiATGkCYiEhhDmohIYAxpIiKBMaSJiATGkCYiEhhDmohIYAxpIiKBMaSJiATGkCYiEhhDmohIYAxpIiKBMaSJiATGkCYiEhhDmohIYAxpIiKBMaSJiATGkCYiEhhDmohIYAxpIiKBMaSJiATGkCYiEhhDmohIYB4J6W+//RbHjh3zxKWIiOgaHgnps2fP4rHHHvPEpYiI6BpsdxARCSy0L4MmTZrk7TqIiEhCn0I6NDQUjzzyCNRqteT51tZWvPLKKx4tjIiI+hjSarUao0ePRlZWluT5xsZGhjQRkRf0qSc9efJknDlzptfzERERuOeeezxWFBERXRXicrlc/i4ikJjN7XA6xZ0ylUqOtjabv8sQDudFGudFmq/nRaWS93rOY7s7zp4966lLERHRv91SSHd1deHgwYN4/PHHkZGR4amaiIjo3/r0wuH3nT59Gvv27UNVVRW6urowY8YM7Ny509O1EREFvT6HtMPhwIcffoiysjJ8+umnSE5OxoULF1BVVdXr1jwiIro1fQrprVu3oqqqCkqlEg899BBeeOEFqFQqTJw4ETIZ37RIROQtfQrpt956C8uWLcOKFSsYykREPtSnxF2zZg0OHjyI++67D1u3bsWpU6e8XRcAwGAwICUlBVqtFnl5eTCbzb2ObW5uRk5ODjQaDdLS0lBZWdntfFdXF5555hkkJiZi6tSp2LBhA7799ltv3wIR0S3pU0gvWbIEf/rTn/Dcc8/BZDJh/vz5yMrKgsvl8lrQVVRUoKSkBHq9HmVlZbDZbMjPz5cc63A4kJubC6VSifLycixfvhwFBQWor693j3n11VdRXV2Nl156Cbt27YLRaMSmTZu8UjsRkafc1JtZLBYLKioqUF5ejtbWVkybNg1z5szBwoULPVZYVlYWZsyYgVWrVgEAWlpakJ6ejvfff7/HC5U1NTXIz8/H0aNHERERAQBYu3YtOjo6sH37djidTiQnJ+PXv/415s+fDwA4evQonnzySRw9ehTDhg3rc118M0tg4rxI47xIC/g3sygUCvziF7/ABx98gNdffx3Dhg1DUVHRTRf4fZ2dnWhsbMT06dPdx2JjYxEdHY2GhoYe441GIzQajTugASApKck9tqWlBVartdv1EhIS4HK5cOLECY/VTUTkaTe1T/paSUlJSEpKgsVi8UQ9AACr1Qqn0wmlUtntuEKhkHwei8UiOfa7HvZ3f147ZtCgQRg+fPh1+9xSlMqh/RrvD9f7rRzMOC/SOC/SRJmXPoX0Z599huLiYuzcuRNyeffCL126hKeeegobN26EQqHwSpE3cqOOjSc/noTtjsDEeZHGeZEWcO2O0tJSTJs2rUdAA8CwYcOQkJCA0tLSm6/weyIjIyGTyXqsci0Wi+QvAqVSKTn2u5XzyJEjAaDbmCtXruDixYs9VuBERCLpU0g3NDQgPT291/MzZ87EJ5984rGiwsLCEB8fj7q6OvexlpYWtLa2QqvV9hiv0WhgNBpht9vdx2pra91jY2NjERkZ2e16x44dQ0hICCZMmOCxuomIPK1PIX3+/HmMGDGi1/PDhw+HyWTyVE0AgOzsbJSWluLw4cNobGyETqdDYmIi1Go1jEYjMjIy3M+ZmpqKqKgo6HQ6NDU1oby8HNXV1Vi0aBEAQCaT4ZFHHsGLL76I2tpaGI1GbNmyBXPnzsXw4cM9WjcRkSf1qSctl8tx9uxZREdHS54/e/asZCvkVixYsABmsxmFhYWw2WxITk5272u22+1obm6Gw+EAcHXlbTAYoNfrMW/ePKhUKhQVFSEhIcF9vaeeegqXL1/GqlWr4HA48JOf/ARPP/20R2smIvK0Pu2TXrlyJcLCwvD8889Lns/Pz0dnZyd27Njh8QJFwxcOAxPnRRrnRZpILxz2aSX985//HIsWLcKwYcOwbNkyjBo1CgBgMpmwc+dOfPDBB9i9e7dnqiUiIrc+hfSUKVNQVFSE3/zmNygrK8PQoVf3Cre3t2Pw4MEoLCzE1KlTvVooEVEw6vObWebPn4+UlBRUVFSgra0NLpcLY8eORUZGhntlTUREntWnkL5w4QLWr1+Pjz76CE6nE1qtFs899xxiYmK8XR8RUVDr0xa8F198EQ0NDVi5ciXWrl0Ls9kMvV7v7dqIiIJen1bSH330EbZs2YK0tDQAV/clz507Fw6HA4MHD/ZqgUREwaxPK2mTyYSJEye6/37XXXdh8ODBaGtr81phRETUx5C+cuVKjxWzTCbDlStXvFIUERFd1efdHb/85S+7BXVnZyfWrVuH8PBw97E333zTs9UREQW5PoV0VlZWj2Nz5871eDFERNRdn0L6mWee8XYdREQk4aa+PouIiHyDIU1EJDCGNBGRwBjSREQCY0gTEQmMIU1EJDCGNBGRwBjSREQCY0gTEQmMIU1EJDCGNBGRwBjSREQCY0gTEQmMIU1EJDCGNBGRwBjSREQCY0gTEQmMIU1EJDCGNBGRwBjSREQCY0gTEQmMIU1EJDCGNBGRwBjSREQCY0gTEQmMIU1EJDCGNBGRwBjSREQCY0gTEQlM2JA2GAxISUmBVqtFXl4ezGbzdcc3NzcjJycHGo0GaWlpqKysdJ9zOBzYunUrZs+eDa1Wi7S0NGzfvh1dXV3evg0iolsiZEhXVFSgpKQEer0eZWVlsNlsyM/P73W8w+FAbm4ulEolysvLsXz5chQUFKC+vh4A0NHRgaamJuTn5+PAgQPQ6/XYu3cvSkpKfHVLREQ3JdTfBUjZs2cPFi9ejPvvvx8AUFxcjPT0dJw6dQpqtbrH+CNHjsBkMqGqqgoRERFQq9U4duwY9uzZg4SEBMjlcrzxxhvu8XfeeScWL16M6upqrFixwmf3RUTUX8KtpDs7O9HY2Ijp06e7j8XGxiI6OhoNDQ2SjzEajdBoNIiIiHAfS0pK6nU8AFitVsjlcs8VTkTkBcKtpK1WK5xOJ5RKZbfjCoUCFotF8jEWi0VyfG997HPnzmHfvn3YuHFjv+tTKof2+zG+plLxl48Uzos0zos0UebFpyG9fv167N+/v9fzWVlZWL16db+v63K5+jz24sWLyM3NxcyZM5GZmdnv5zKb2+F09v35fE2lkqOtzebvMoTDeZHGeZHm63m53i8En4a0TqfDmjVrej0fHh6O2267DTKZDGazGePGjXOfs1gsUCgUko9TKpU4c+ZMt2NSq+vLly9j6dKliImJwebNm2/hToiIfMOnIS2Xy/vUB46Pj0ddXR0SEhIAAC0tLWhtbYVWq5Ucr9FosGvXLtjtdgwZMgQAUFtb2218R0cHli1bhvDwcGzbtg2hocJ1eoiIehDuhUMAyM7ORmlpKQ4fPozGxkbodDokJia6d3YYjUZkZGTAZDIBAFJTUxEVFQWdToempiaUl5ejuroaixYtAnB1i97KlSthsVhQVFSEixcvoq2tDW1tbX67RyKivhByOblgwQKYzWYUFhbCZrMhOTkZmzZtcp+32+1obm6Gw+EAAISFhcFgMECv12PevHlQqVQoKipyr8RNJhOOHDkCAJg1a1a35zp58qSP7oqIqP9CXP151Y34wmGA4rxI47xIE+mFQyHbHUREdBVDmohIYAxpIiKBMaSJiATGkCYiEhhDmohIYAxpIiKBMaSJiATGkCYiEhhDmohIYAxpIiKBMaSJiATGkCYiEhhDmohIYAxpIiKBMaSJiATGkCYiEhhDmohIYAxpH9r119P4f6/VYddfT/u7FCIKEAxpH/rjCRPsnVfwxxMmf5dCRAGCIe1DsyeOwpCwQZg9cZS/SyGiABHq7wKCyRP3jsET947xdxlEFEC4kiYiEhhDmohIYAxpIiKBsScd4LZUf47601YkjImE7oHx/i6HiDyMK+kAV3/aCqfr6p9ENPAwpANcwphIyEKu/klEAw/bHQGOLQ6igY0raSIigTGkiYgExpAmIhIYe9LUA7f1EYmDK2nqgdv6iMTBkKYeuK2PSBxsd1APbHEQiYMraSIigTGkiYgExpAmIhIYQ5p8gl/CS3RzhA1pg8GAlJQUaLVa5OXlwWw2X3d8c3MzcnJyoNFokJaWhsrKSslxXV1dmD9/PuLi4vDVV195o3SSwC/hJbo5QoZ0RUUFSkpKoNfrUVZWBpvNhvz8/F7HOxwO5ObmQqlUory8HMuXL0dBQQHq6+t7jDUYDBg+fLg3yycJ/BJeopsj5Ba8PXv2YPHixbj//vsBAMXFxUhPT8epU6egVqt7jD9y5AhMJhOqqqoQEREBtVqNY8eOYc+ePUhISHCP+/zzz7F//37s2LEDmZmZPrsf4pfwEt0s4VbSnZ2daGxsxPTp093HYmNjER0djYaGBsnHGI1GaDQaREREuI8lJSV1G9/Z2Yl169ZBr9dj6NCh3rsBIiIPEi6krVYrnE4nlEplt+MKhQIWi0XyMRaLRXL8tX3sl19+GZMmTUJqaqrniyYi8hKftjvWr1+P/fv393o+KysLq1ev7vd1XS7Xdc8bjUZUV1fjvffe6/e1v0+pFH8VrlLJ/V2CkDgv0jgv0kSZF5+GtE6nw5o1a3o9Hx4ejttuuw0ymQxmsxnjxo1zn7NYLFAoFJKPUyqVOHPmTLdj166ujx8/jnPnziExMbHbmFmzZiEvLw8rVqzo8z2Yze1wOq//S8GfVCo52tps/i7DJ3b99TT+eMKE2RNH3bDfHUzz0h+cF2m+npfr/ULwaUjL5XLI5Tf+7RQfH4+6ujr3i34tLS1obW2FVquVHK/RaLBr1y7Y7XYMGTIEAFBbW+sen5mZieTkZPf4b775BkuWLMFrr72G8eP5ORWB6tptfXxRkgYq4XrSAJCdnY3S0lIcPnwYjY2N0Ol0SExMdO/sMBqNyMjIgMl0dc9tamoqoqKioNPp0NTUhPLyclRXV2PRokUAgMjISKjVavc/Y8aMAQCMGTOmRy+bAge39VEwEHIL3oIFC2A2m1FYWAibzYbk5GRs2rTJfd5ut6O5uRkOhwMAEBYWBoPBAL1ej3nz5kGlUqGoqKjb9jsaeLitj4JBiOtGr7pRN+xJBybOizTOizSRetJCtjuIiOgqhjQRkcAY0kTX4Kf1kWgY0kTX4Kf1kWgY0kTX4LY+Eo2QW/CI/IXb+kg0XEkTEQmMIU1EJDCGNBGRwBjSRD7ArX10sxjSRD7ArX10sxjSRD7ArX10s7gFj8gHuLWPbhZX0kREAmNIExEJjCFNRCQwhjRRgOK2vuDAkCYKUNzWFxwY0kQBitv6ggO34BEFKG7rCw5cSRMRCYwhTUQkMIY0EZHAGNJERAJjSBORG/dei4chTURu3HstHoY0Eblx77V4uE+aiNy491o8XEkTEQmMIU1EJDCGNBGRwBjSREQCY0gTEQmMIU1EJDCGNBGRwBjSREQCY0gTEQmM7zjsJ5ksxN8l3FAg1OgPnBdpnBdposxLiMvlcvm7CCIiksZ2BxGRwBjSREQCY0gTEQmMIU1EJDCGNBGRwBjSREQCY0gTEQmMIU1EJDCGNBGRwBjSREQCY0gPUDt37kRmZiYmT56M++67D5s3b8bly5f9XZZfGAwGpKSkQKvVIi8vD2az2d8l+RV/Nm7sqaeeQlxcHOrq6vxdCkN6oPr000/x5JNPorKyEs8//zw+/vhjbN682d9l+VxFRQVKSkqg1+tRVlYGm82G/Px8f5flV/zZuL6qqirY7XZ/l/EfLgoKBw8edE2bNs3fZfjcQw895Nq2bZv772fPnnWp1WrXyZMn/ViVWIL1Z0PK+fPnXT/+8Y9dra2tLrVa7aqtrfV3SS6upIOE1WqFXC73dxk+1dnZicbGRkyfPt19LDY2FtHR0WhoaPBjZWIJxp+N3uh0OuTm5mL06NH+LsWNIR0EbDYb3nzzTcyfP9/fpfiU1WqF0+mEUqnsdlyhUMBisfipKrEE68+GlLKyMnR1deHhhx/2dynd8EP/A8z69euxf//+Xs9nZWXh2Wefdf+9s7MTK1euRGxsLJYuXeqLEilA8GfjP86dO4cdO3agrKzM36X0wJAOMDqdDmvWrOn1fHh4uPvfu7q6sHr1aly+fBm7du1CaGhw/eeOjIyETCaD2WzGuHHj3MctFgsUCoUfK/O/YP/Z+L7PPvsM//znPzFr1qxux5944glkZWWhuLjYT5UxpAOOXC7vU//Q6XRi3bp1OHv2LHbv3o0f/OAHPqhOLGFhYYiPj0ddXR0SEhIAAC0tLWhtbYVWq/Vzdf7Dn42epk+fjgMHDnQ79uCDD2Lz5s1ISUnxU1VXMaQHqKeffhp1dXV4/fXX4XA40NbWBuBqP3bQoEF+rs53srOzUVxcjPHjxyMmJgbFxcVITEyEWq32d2l+w5+NnoYOHSr5MxETE4NRo0b5oaL/4HccDlBxcXGSx2tqahATE+PjavzLYDBg9+7dsNlsSE5OxqZNmzBy5Eh/l+U3/Nnom7i4OPz+979HYmKiX+tgSBMRCYxb8IiIBMaQJiISGEOaiEhgDGkiIoExpImIBMaQJiISGEOaiEhgDGmi71m/fj3i4uIQFxeHCRMmYMaMGSgoKIDVanWP+fTTT7FixQokJyfj7rvvRnp6On71q1/hxIkTPa73u9/9DuPHj8czzzwj+Xzt7e3YuHEjEhMTMXnyZDz55JM4e/as1+6PAgtDmkjCPffcg48//hh//vOfodPp8OGHH2LdunUArn7bS3Z2NkJDQ/Hcc8/h4MGDePHFFxEdHY0tW7b0uNa+ffuQm5uLqqoqdHZ29ji/du1aHD16FNu2bcM777wDl8uFxYsXo6Ojw+v3SQHAr185QCSgdevWuR5//PFux1599VVXfHy86/z5865Jkya5nn76acnHXrhwodvfjx496kpKSnI5HA7XnDlzXAcOHOh2/ssvv3Sp1WrXRx991O0aEydOdFVUVHjmhiigcSVN1Afh4eFwOp0oLy9HZ2cn8vLyJMcNHz6829/37t2LBx98EKGhoXjooYewb9++buePHz+OwYMHIykpqds1NBoNPvnkE8/fCAUchjTRDXzxxRd4++23odVq0dbWhqFDh+L222+/4eMsFgsOHTqErKwsAEBmZiaOHz+OL7/80j2mra0NI0aM6PHpcyNHjnR/Oh0FN4Y0kYT6+npMmTIFGo0GP/3pTxEbG4vnn38ern58HllFRQXGjRuH+Ph4AEBUVBSSk5Px7rvv9unxISEhN1U7DSz8PGkiCRqNBlu3bsWgQYMQFRWFsLAwAMDYsWPR3t6O8+fPX3c17XK5UF5ejjNnzmDChAnu406nE3//+9+xevVqhIWFQaVS4cKFC7hy5Uq31bTZbMaYMWO8dn8UOLiSJpIQHh6OO++8EzExMe6ABoDZs2cjLCwMr776quTjLl68CACora1FS0sL/vCHP6CqqqrbP11dXTh06BAAYOrUqXA4HKitrXVf49KlS2hoaMCPfvQjL94hBQqupIn6YdSoUSgoKEBBQQFsNht+9rOfITY2FhcvXkRNTQ3q6urw9ttvo6ysDNOmTcOUKVN6XCMtLQ179+7FAw88gLFjx2LmzJkoLCzEli1bIJfL8cILL2DUqFGYM2eOH+6QRMOVNFE/LVy4ELt378a//vUv5OfnIyMjA6tWrcJXX32FjRs3wmw2o6amBrNnz5Z8/Jw5c1BfX4/Tp08DAH77298iISEBK1aswMMPPwyn04k333yz25cKU/DiN7MQEQmMK2kiIoExpImIBMaQJiISGEOaiEhgDGkiIoExpImIBMaQJiISGEOaiEhgDGkiIoH9f+clMyzkF+c7AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-> test with 'LR'\n",
      "tn, fp, fn, tp: [5 0 0 3]\n",
      "average_pr_score: 1.0\n",
      "f1 score_LR: 1.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5 0]\n",
      " [0 3]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5 0 0 3]\n",
      "f1 score_SVM: 1.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5 0]\n",
      " [0 3]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [2 3 3 0]\n",
      "f1 score_KNN: 0.0\n",
      "balanced accuracy_KNN: 0.2\n",
      "confusion matrix_KNN\n",
      "[[2 3]\n",
      " [3 0]]\n",
      "-> check results\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=12\n",
      "-> create 8 synthetic samples\n",
      "{1: 20, 0: 20}\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",
      "tn, fp, fn, tp: [5]\n",
      "average_pr_score: nan\n",
      "f1 score_LR: 0.0\n",
      "balanced accuracy_LR: 1.0\n",
      "confusion matrix_LR\n",
      "[[5]]\n",
      "-> test with 'SVM'\n",
      "tn, fp, fn, tp: [5]\n",
      "f1 score_SVM: 0.0\n",
      "balanced accuracy_SVM: 1.0\n",
      "confusion matrix_SVM\n",
      "[[5]]\n",
      "-> test with 'KNN'\n",
      "tn, fp, fn, tp: [0 5 0 0]\n",
      "f1 score_KNN: 0.0\n",
      "balanced accuracy_KNN: 0.0\n",
      "confusion matrix_KNN\n",
      "[[0 5]\n",
      " [0 0]]\n",
      "-> check results\n",
      "\n",
      "====== Step {shuffleStep + 1}/{self.numOfShuffles} =======\n",
      "-> Spliting data to slices\n",
      "\n",
      "------ Step {shuffleStep + 1}/{self.numOfShuffles}: Slice {sliceNr + 1}/{self.numOfSlices} -------\n",
      "-> Train generator for synthetic samples\n",
      "Train GAN with |class 0|=20, |class 1|=9\n",
      "-> create 11 synthetic samples\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/kristian/anaconda3/lib/python3.8/site-packages/sklearn/metrics/_ranking.py:681: RuntimeWarning: invalid value encountered in true_divide\n",
      "  recall = tps / tps[-1]\n",
      "/home/kristian/anaconda3/lib/python3.8/site-packages/sklearn/metrics/_classification.py:1464: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 due to no true nor predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(\n",
      "/home/kristian/anaconda3/lib/python3.8/site-packages/sklearn/metrics/_classification.py:1464: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 due to no true nor predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(\n",
      "/home/kristian/anaconda3/lib/python3.8/site-packages/sklearn/metrics/_classification.py:1814: UserWarning: y_pred contains classes not in y_true\n",
      "  warnings.warn('y_pred contains classes not in y_true')\n"
     ]
    },
    {
     "ename": "IndexError",
     "evalue": "index 11 is out of bounds for axis 0 with size 9",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mIndexError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-7-e4cc87710ca8>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mexercise\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrun\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mgan\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mloadDataset\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m/smallTank/home/kristian/Dokumente/Programming/LoGAN/library/exercise.py\u001b[0m in \u001b[0;36mrun\u001b[0;34m(self, gan, dataset)\u001b[0m\n\u001b[1;32m     52\u001b[0m                 \u001b[0msliceTitle\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m\"Slice {sliceNr + 1}/{self.numOfSlices}\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     53\u001b[0m                 \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdebug\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf\"\\n------ {stepTitle}: {sliceTitle} -------\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 54\u001b[0;31m                 \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_exerciseWithDataSlice\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mgan\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msliceData\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     55\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdebug\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"### Exercise is done.\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     56\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/smallTank/home/kristian/Dokumente/Programming/LoGAN/library/exercise.py\u001b[0m in \u001b[0;36m_exerciseWithDataSlice\u001b[0;34m(self, gan, dataSlice)\u001b[0m\n\u001b[1;32m     63\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mnumOfNeededSamples\u001b[0m \u001b[0;34m>\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     64\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdebug\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf\"-> create {numOfNeededSamples} synthetic samples\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 65\u001b[0;31m             \u001b[0mnewSamples\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0masarray\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mgan\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mgenerateData\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0m_\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnumOfNeededSamples\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     66\u001b[0m             train = DataSet(\n\u001b[1;32m     67\u001b[0m                 \u001b[0mdata0\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdataSlice\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdata0\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/smallTank/home/kristian/Dokumente/Programming/LoGAN/library/exercise.py\u001b[0m in \u001b[0;36m<listcomp>\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m     63\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mnumOfNeededSamples\u001b[0m \u001b[0;34m>\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     64\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdebug\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf\"-> create {numOfNeededSamples} synthetic samples\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 65\u001b[0;31m             \u001b[0mnewSamples\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0masarray\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mgan\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mgenerateData\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0m_\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnumOfNeededSamples\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     66\u001b[0m             train = DataSet(\n\u001b[1;32m     67\u001b[0m                 \u001b[0mdata0\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdataSlice\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdata0\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/smallTank/home/kristian/Dokumente/Programming/LoGAN/library/interfaces.py\u001b[0m in \u001b[0;36mgenerateData\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m     28\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnextIndex\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     29\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 30\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mexampleItems\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     31\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     32\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mIndexError\u001b[0m: index 11 is out of bounds for axis 0 with size 9"
     ]
    }
   ],
   "source": [
    "exercise.run(gan, loadDataset())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "extraordinary-hughes",
   "metadata": {},
   "outputs": [],
   "source": [
    "try:\n",
    "    exercise.run(gan, loadFailingDataset())\n",
    "except AttributeError as e:\n",
    "    print(e)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "focused-respect",
   "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}