HW6_NB.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Наивный Байес и спам\n",
    "__Суммарное количество баллов: 10__\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "import numpy as np\n",
    "import pandas\n",
    "import random\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib\n",
    "import copy\n",
    "#import spacy\n",
    "from nltk.stem.snowball import SnowballStemmer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "import re\n",
    "def read_dataset(filename):\n",
    "    file = open(filename, encoding=\"utf-8\")\n",
    "    x = []\n",
    "    y = []\n",
    "    for line in file:\n",
    "        cl, sms = re.split(\"^(ham|spam)[\\t\\s]+(.*)$\", line)[1:3]\n",
    "        x.append(sms)\n",
    "        y.append(cl)\n",
    "    return np.array(x, dtype=np.str), np.array(y, dtype=np.str)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_precision_recall_accuracy(y_pred, y_true):\n",
    "    classes = np.unique(list(y_pred) + list(y_true))\n",
    "    true_positive = dict((c, 0) for c in classes)\n",
    "    true_negative = dict((c, 0) for c in classes)\n",
    "    false_positive = dict((c, 0) for c in classes)\n",
    "    false_negative = dict((c, 0) for c in classes)\n",
    "    for c_pred, c_true in zip(y_pred, y_true):\n",
    "        for c in classes:\n",
    "            if c_true == c:\n",
    "                if c_pred == c_true:\n",
    "                    true_positive[c] = true_positive.get(c, 0) + 1\n",
    "                else:\n",
    "                    false_negative[c] = false_negative.get(c, 0) + 1\n",
    "            else:\n",
    "                if c_pred == c:\n",
    "                    false_positive[c] = false_positive.get(c, 0) + 1\n",
    "                else:\n",
    "                    true_negative[c] = true_negative.get(c, 0) + 1\n",
    "    precision = dict((c, true_positive[c] / (true_positive[c] + false_positive[c])) for c in classes)\n",
    "    recall = dict((c, true_positive[c] / (true_positive[c] + false_negative[c])) for c in classes)\n",
    "    accuracy = sum([true_positive[c] for c in classes]) / len(y_pred)\n",
    "    return precision, recall, accuracy\n",
    "\n",
    "def plot_precision_recall(X_train, y_train, X_test, y_test, bow_method, voc_sizes=range(4, 200, 5)):\n",
    "    classes = np.unique(list(y_train) + list(y_test))\n",
    "    precisions = dict([(c, []) for c in classes])\n",
    "    recalls = dict([(c, []) for c in classes])\n",
    "    accuracies = []\n",
    "    for v in voc_sizes:\n",
    "        bow = bow_method(X_train, voc_limit=v)\n",
    "        X_train_transformed = bow.transform(X_train)\n",
    "        X_test_transformed = bow.transform(X_test)\n",
    "        classifier = NaiveBayes(0.001)\n",
    "        classifier.fit(X_train_transformed, y_train)\n",
    "        y_pred = classifier.predict(X_test_transformed)\n",
    "        precision, recall, acc = get_precision_recall_accuracy(y_pred, y_test)\n",
    "        for c in classes:\n",
    "            precisions[c].append(precision[c])\n",
    "            recalls[c].append(recall[c])\n",
    "        accuracies.append(acc)\n",
    "    def plot(x, ys, ylabel, legend=True):        \n",
    "        plt.figure(figsize = (12, 3))\n",
    "        plt.xlabel(\"Vocabulary size\")\n",
    "        plt.ylabel(ylabel)\n",
    "        plt.xlim(x[0], x[-1])\n",
    "        plt.ylim(np.min(list(ys.values()))-0.01, np.max(list(ys.values()))+0.01)\n",
    "        for c in ys.keys():\n",
    "            plt.plot(x, ys[c], label=\"Class \" + str(c))\n",
    "        if legend:\n",
    "            plt.legend()\n",
    "        plt.tight_layout()\n",
    "        plt.show()\n",
    "    \n",
    "    plot(voc_sizes, recalls, \"Recall\")\n",
    "    plot(voc_sizes, precisions, \"Precision\")\n",
    "    plot(voc_sizes, {\"\":accuracies}, \"Accuracy\", legend=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "X, y = read_dataset(\"hw6_data/spam\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['Talk With Yourself Atleast Once In A Day...!!! Otherwise You Will Miss Your Best FRIEND In This WORLD...!!! -Shakespeare- SHESIL  &lt;#&gt;'\n",
      " 'You will be in the place of that man'\n",
      " 'Thanks 4 your continued support Your question this week will enter u in2 our draw 4 £100 cash. Name the NEW US President? txt ans to 80082'\n",
      " ...\n",
      " 'Natalja (25/F) is inviting you to be her friend. Reply YES-440 or NO-440 See her: www.SMS.ac/u/nat27081980 STOP? Send STOP FRND to 62468'\n",
      " 'We tried to contact you re our offer of New Video Phone 750 anytime any network mins HALF PRICE Rental camcorder call 08000930705 or reply for delivery Wed'\n",
      " 'Ü give me some time to walk there.']\n"
     ]
    }
   ],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, train_size=0.9)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Задание 1 (3 балла)\n",
    "Реализуйте классификацию методом Байеса в предположении независимости компонент `x`. При реализации следует считать, что значения каждой компоненты дискретны, а `p(x|y)` имеет мультиномиальное распределение.\n",
    "\n",
    "#### Методы\n",
    "`fit(X, y)` - оценивает параметры распределения `p(x|y)` для каждого `y`. \n",
    "\n",
    "`log_proba(X)` - для каждого элемента набора `X` считает логарифм вероятности отнести его к каждому классу. По теореме Байеса: `p(y|x) = p(y)p(x|y)/p(x)`. Соответственно, после логарифмирования: `log p(y|x) = log p(y) + log p(x|y) - log p(x)`. Поскольку `log p(x)` является независящей от класса `y` константой, это слагаемое никак не влияет на решение задачи максимизации. Поэтому его можно просто убрать для простоты."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "from collections import Counter\n",
    "\n",
    "class NaiveBayes:\n",
    "    def __init__(self, alpha):\n",
    "        self.alpha = alpha # Параметр аддитивной регуляризации\n",
    "    \n",
    "    def fit(self, X, y):\n",
    "        self.py = Counter(y)\n",
    "        self.classes = dict(zip(range(len(self.py.keys())), self.py.keys()))\n",
    "        self.pxy = dict() # feat -> {x -> {y -> cnt}}\n",
    "        for row, cl in zip(X, y):\n",
    "            for i, f in enumerate(row):\n",
    "                if i not in self.pxy:\n",
    "                    self.pxy[i] = dict()\n",
    "                if f not in self.pxy[i]:\n",
    "                    self.pxy[i][f] = dict()\n",
    "                if cl not in self.pxy[i][f]:\n",
    "                    self.pxy[i][f][cl] = 0\n",
    "                self.pxy[i][f][cl] += 1\n",
    "        \n",
    "        total = sum(self.py.values())\n",
    "        self.py = {k : v/total for k,v in self.py.items()}\n",
    "        for feat, row in self.pxy.items():\n",
    "            for x, vals in row.items():\n",
    "                self.pxy[feat][x] = {cl : v/self.py[cl] for cl,v in vals.items()}\n",
    "        \n",
    "    def predict(self, X):\n",
    "        return [self.classes[i] for i in np.argmax(self.log_proba(X), axis=1)]\n",
    "    \n",
    "    def log_proba(self, X):\n",
    "        ans = np.zeros((X.shape[0], len(self.classes)))\n",
    "        for i, row in enumerate(X):\n",
    "            for cl in range(len(self.classes.keys())):\n",
    "                ans[i][cl] = np.log(self.py[self.classes[cl]]) + sum([np.log(self.pxy[i].get(x, dict()).get(self.classes[cl], self.alpha)) for i, x in enumerate(row)])\n",
    "        return ans"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Задание 2 (3 балла)\n",
    "Теперь реализуем метод bag of words. Задача состоит в том, чтобы посчитать количество вхождений каждого слова в предложение.\n",
    "\n",
    "#### Методы\n",
    "`__init__(X, voc_limit)` - инициализирует BoW, составляя словарь, который будет использоваться для векторизации предложений. Размер словаря должен быть не больше `voc_limit`, в него должны входить самые часто встречающиеся в выборке слова.\n",
    "\n",
    "`transform(X)` - векторизует сообщения из набора."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "class BoW:\n",
    "    def __init__(self, X, voc_limit=1000):\n",
    "        c = Counter()\n",
    "        for row in X:\n",
    "            c.update(row.split())\n",
    "        self.bow = dict() # word -> pos\n",
    "        for i, (k, v) in enumerate(sorted(c.items(), key=lambda x: x[1], reverse=True)[:voc_limit]):\n",
    "            self.bow[k] = i\n",
    "        \n",
    "    def transform(self, X):\n",
    "        ans = np.zeros((X.shape[0], len(self.bow)))\n",
    "        for i, row in enumerate(X):\n",
    "            for w in row.split():\n",
    "                t = self.bow.get(w, -1)\n",
    "                if t == -1:\n",
    "                    pass\n",
    "                else:\n",
    "                    ans[i][t] += 1\n",
    "        return ans"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "bow = BoW(X_train, voc_limit=500)\n",
    "X_train_bow = bow.transform(X_train)\n",
    "X_test_bow = bow.transform(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "({'ham': 0.9755600814663951, 'spam': 0.9701492537313433},\n",
       " {'ham': 0.9958419958419958, 'spam': 0.8441558441558441},\n",
       " 0.974910394265233)"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictor = NaiveBayes(0.001)\n",
    "predictor.fit(X_train_bow, y_train)\n",
    "get_precision_recall_accuracy(predictor.predict(X_test_bow), y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1gAAADQCAYAAAAalMCAAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/d3fzzAAAACXBIWXMAAAsTAAALEwEAmpwYAAA8MElEQVR4nO3deXhb5Zn38e8tWd7ixNns7CELkB0COAEChADD3hK2lqXNsDMMyxRmuk3bGabAUFraQjtQeKGllA6UlrZQ2tJhaROHnTipA9lIQuIQO5uzOfFuSc/7xzm2ZUVOnMSybOv3uS5dR+c5j45u6Uj2ufUsx5xziIiIiIiIyOELpDoAERERERGR3kIJloiIiIiISCdRgiUiIiIiItJJlGCJiIiIiIh0EiVYIiIiIiIinSQj1QEcrMGDB7sxY8akOgwREREREUljixcv3u6cK4gv73EJ1pgxYygpKUl1GCIiIiIiksbMbEOi8h6XYImIiEjqOOeIRB1NEUdjOEpjJEpTzK0x7FrvR6I0hqM0RWLKmh/jl7fWab45GsLx+4zSGHH+Y5qf0xF/Lc/YVYdLUNb6GuLL9scSldm+pYnqxWvv+RJdl7S9upnBAKFggFDQCAUDZGbErTdvz2jd3voYr17rYwIYxBwz/1iF2643xr33bY9F6zFurpeMq6zu73VnZgS9ZezrzLD9vu5gAKxDR026WtS51s9fzN+D9j6f7f2tiK0TjXbsO9bud9E5BrpdjI9uYHy0jPHRsnbjV4Il0l2EG2DLR1BeAtVbOnnnBgOOgGHHQuFkyMjq5P2LSDPnHPVNUWoaw9Q1RqhpDFPTEKG2MUxtY8RLHmISk30SkEiUprgk5UAntJ3NO7lx+yQ+zc+d4PzjsGUE2p78ZsadDIcyAmQFA2SFAuRlZ5ARCBCIOTduzndiT5hbymLr0VIxpqx9CV9qgkKXoNC5ts+9Twz7bjhgkQPCcce/piHcut58vMKuJZFt9MsP9riZEZOwJUjW/AQuFAyQEwrSLzujNWnr5Lwl0euubgi3vNZ9vhuH8bqle2v5PGYE9vl8Jvpb0ZxcBwOJp55I9FHNdA0MaSxjRP06hjd8wvAGb5kXqWqpsztjcLsxKsESSQXnYNd6KF8MFSVQvshLriKN3vZAKPF/5UMVjYCL+PvOgIKJXrI19Bh/ORWy+nbe80lSOeeoqmuiYncdFbvq2LS7jk1V9VTs9u5vraoHaPnn03pSFPMPKBggKyPmn5J/0pSZ0fYfVXNZwIhLAtr+itgUjf3F0LVt1QjHrfsn7hlBIzczgz6ZQXKz/GVmBn2ygi3lOZlB+mRlkJsZpE9mBrlZ3rK1jlcWCu77j9NrZWmbjBzwfjhKONr6C2hjxDt5rW2M+LfWZKmmMULdPuteQnWoJ3Rm3slDpn9M2muRiD2hzfBbIDpTwCzmM9H2c5MZ85lJmAgleExzy0FmzOcxFPt5CwQIBNSSkGzN34nYFsSmSJSoc3HJbfMJqSVsqetp2nvd4QQtGtI9GLT+nYj525fR2Z9J56BqI2xdDluX+cvlsGMtOP/Hq4wcGDIZhsyFIVNhyBQonEz/3IHwH4ljsUTNYN1ZUVGR0xgs6XHqdkHF4taEqmIx1O7wtoVyYdh0GFkEI4twI04g2ndEpz69i0aI7CwjuqkUNn9IYOtHZGz9kGDddm87RkP+WGoHTGbvwClU5U9id/5EajMG+P+EWk+Sm39BjDrnnfRmZXg3/6Q3LyuD3KwM8vz1jAQnvp2lMRxtObGt9U+CaxrD1DZEWloPvBNFiztxaP31q3k9o03Xkrbbgl184tcYjrJ1T2vCtGl3nZdM7a5vWa9tjLR5TGZGgBH9cxjRP4ch/bL9hChxF6zG+JaJcHNZ5KBOPOJ/RQwFrG1SF/eLYux6RtCIRN2+SUtMsnIwLTOZwQDZoQBRR8vntLPPnbIyAuTGJYHx616S2DZhjF3PyQySHQp2m8+aiEjaaNgL21a2TaS2LoeGPa11BoxpTaKGTPHuDxgDgWDCXZrZYudc0T7lSrBEOqa51aBybwOV1Q1U7m1ge3Uju2oaafBPTBsjUSJNjRTUrmFkzQpG161gbP1KhoXLAYhibAyMYmXwKJbb0XzojmRVdBT1UWvzK3oXvSIK2c3UwHqmWBlTAhuYGljPSNveUqPCDWJFdAzLomNY7rzlFgbSsZEGnsyMAHlZ3gmol4R5J5x5LYlZa5JmRktyFJsk1fhdq5pbEmoawtQ1RbrsvQrEdJPJaNMK5P+iFpNMZMS0OCSqm+hxe+vDMclUPVv31u/TAjI4L5Ph/XMYnp/DiAE5DO+fw4j+2V5Z/xwG9cls/VWvqgJyB0Io55BfczTqaIq2dl+LRF1yf0VMoCkSbZOAtX4WEidkdY0RggFrkyy3TZwD+9lm/vHwXmNGoLX1LjcrSG4ouT8WSApVV8KKl2D5i94v2QUTva7UQ6ZA4SQYfHTqulXXbPdOALet8G5bV3i9H5wDC7TeAkH/vvnLYNvtFoBA3HpsHfB+rW9zi/hL5y2jkQR14m6xdZIh4Wvez+sOBGO2JXjdZhzM/zPpSof4uXMu5rMbt71+d+vus/rFJFF+IlU46aB78yjBEknAOcfehrCXLPmJU+uy0Vu2JFMNCU7oHaMCO5iR8QnHBdZyDGuZyHqy8Lr67bL+rAlNYG3mRNZnTWJjzkTCob5tus+0dptpLUvGr9gZfjeclhaHQOuJZHOXnoxAgJzwHvKrVpC3awV9dq4gZ/syQrs/wfzxBdGcQUSHTCM6ciZ7jrmB6kBfqmMSH+9+mOoGb90bf+KdCDevVzd4LU7N9WsaI0T85oaA0aYrWG6bloGMllaz2OW+rQje43JCQRzsMzC+tdWm/e5i+xsXE462DuSPv98U9hKTlnEB/v1wzLiBcEvXNe81N7c+De+fzfB8P3kakOOX5TAsP5vsUOJfz1pEo7DmVXj3USh70zuJKJgIw6d73UCHTfe6gmb26fTPlkiPU7cLVv4Jlv0O1hd7J1/NiVXlx7B9NUSbvLoWhEFHel2ECv2ka8hk6D/GS1o6Q0M1VK5qTaKaE6qaytY6uYO8+AYd6XX1bi8R6shJafx22DchO6zELRmJi4t5jfEn1Ilec6QD70nPOgdOO4FESXEHfjBo/swG4sr6FHr/B4dMgfxRnTIUQwmWpIWmSJTdtU3srm1kl7/cXdvErtpGdtd565V7G9skTQ0JuiEFA8agPpkU9M2ioG8Wg/NalyMyqxm3ZzFDdrxL303vEKj61HtQRrZ3IjuiCEae4C37j+7csVSp0lDt/Yq65UPYvNS7bV0GOQPgH74N079wWCcazrmW45CVEegVff47wvkTCYSCh9Ea1FgLS5+Dd38COz/x/mmccC2E62FTKWwubT1Js4D3a/yw6d5ndfh0GDpN4+8kPTRUw8d/8ZKqtW94CdSAsTD1Mu82ZHJr3UiTNwZj63KvS1FzwrOrrLVOKDemtWuytyycDHmF7f/dP5j9xu5zyBToU9A7/p+I9CJKsKTHqWuMsHVPPbvr/ASptpFdNU0tidI+CVRtE9UN4Xb3lxEw+udmMjjPT5zyshjcssykIC/bT6IyGZCb2TrguqEaNrzj/cq5boGXWABk5cPY02DsbBg102teDoaS/8Z0F1s+gj9/GTa+ByNnwoXf907apWvs3QIfPAklP/N+jR9+PMy6HSbNhWDM/EXOwd7NXlLcnHBtXuqVAWDeL+KxLV3DjoHs/C5/SSKdrqkO1rwGy34Pq1+FcB30GwFTLvGSquHHHVzScjAtTYWToXCi9/3cttKrH98yNvgor0UsWS1jIpJUSrCk24lGHZXVDXy6s5ZPd9Ty6c5aNu70lp/urGXb3oaEjzODftkhBuSG6J+bSf/cEAPilv1zM73tOX55n0z6ZAY71koQafImoVi3ANYVQ/kHEA1DMBNGnwRjT4dxZ3gno8E0n4gzGoUPn4fX/gPqdsKMG+GMb0JO/1RH1nttWeZ1A/zoBe9zOekzcPLtMOrEgztR3LvVb40s9ROvpbCnvHX7wHFtW7oGHemVN89IGY3E3Q+3dsOJhvdTJ9LaVccC3vcqmAkZmRDMirsf8sa+xN/XyafsT7gR1s33WqpW/Rkaq73Wn+akauTMzv8MVVe2JlstyddKaKrxtueP8hOuSd1jbJeIdAolWJIStY1hNu6sa0maYhOojTtr23TPM4Ph+TmMGpjD6IG5jB6Yy9D8HAb2CZGf4yVMA3Iz6ZcT6twxSs55/xDX+S1UG972/iFj3snluDnebfRJhzVpQK9Wtxvm/zcs+qn36+3Z98CxV6k7S2eJRuGTv8K7j3if0VAfOO6LcNItXiLUWaorYUtcS9fuTztv/53Fgn6yldm6DGZ6388BY7yT14KJ3nLQUV7CJr1bJAwb3vKSqhUve4PZs/vD5Iu8pOqIU7v+B7FoFPZUQHY/tQiL9FJKsISmSJTFG3Yx/+NtvLl6O9UN4ZarnzfPgBZ7bZLYa+LsUx57DRR/+mWAit31bNxZy4YdNXy6s47t1W1bofKyMhg1MJfRA3M4YlAf/753G94/m6yMAwzk7yy7P21NqNYvhJptXvnA8X5CdTqMOc2biU06bvNS+PO/edf1Gn0yXPB9b0CpHJqmOvjwN16L1faPoe8wOPGfvDFWOQO6Joband5x3VXWOug9kOElOQF/YHEgwx9M7Je1bA/GbA+0ljVvdxHv2m/hRm8ZaWjnvn8LN8Tdb2pbr7HGG9+yc13rdd+aJygonAgFk1qXg8anV5fe3igahY3v+0nVS143vcy+MPFCL6kaN0fJtYgklRKsNFW5t4Hi1ZXMX7WNhWsq2VvvJVVFRwxkaH52m+vjtCxjrpETu2yMWbb3sQkYDMtvbYEaPSi3TRI1IDeUugkMwg1eK8vKP3onYODNKDPudO8f8djTof+o1MTWm0SjUPosvHG317I182Y449/1C+7BqK70WgMX/RRqt3sXhJ51B0y+WCeMHRFugO1rWsfKbFsFlSth53rwZ8MkEPLGvzS3dDUvB4xV19/uKBqF3WUx165Z5l1XcO8m7yKgR5/rJVVHna2eBiLSZZRgpYlo1PFRRRV/W7WNBR9vY2l5FQCFfbM4Y0IhZ0ws4JQjB9M3+/B+uW2+2GxL4hWJEo06hvTLJjOjG46PqNsFz3/B6/531Dkw/kwvoSqcpG5syVK7E/52H5Q85Y1/OOc+OObzer/3Z9sqeO9RWPprr2Xm6PPh5NtgzKl63zpDU503yUBzwrXNv+3e0FonmOWNjSmc6CVdBRO9bpgDxkBmbspCTyt1u/1xTMtjLgi6onU8E+a1QA6dBhMuhAnnaSZMEUkJJVi9WFVdE2+uqWT+qkqKV29je3UjZnDcqP5+UlXIlOH90mbq633s2gDPfs67OOPFj8G0y1MdUXqpWAKvfNmbOOSIU7xug7HTIac757wB+e8+6k0dnZED06+Gk/7Za2GR5Gus8a59VLnKS7ial1Ub29brO9xLtgaO9ZfjWtd1gn/wImHv0gItSZR/i33fs/t7iVTsBUELJinZFZFuQQlWL+KcY822av62ahvzV22jZMMuIlFHfk6I048u4IyJBZx+dCED+6grEZv+Ds9d4V0T6MrnvJYA6XrRKPz9GXjjv6B+j5c8nP41b/B3Oqnb3Xoi33wSv20lVG+BvCEw8yY44XroMyjVkQpAw16vq+HOdV73wp3rWm/N4zab9SncN+lqvp8us2pGo95MkdGwNxV588yR0bA3Rm5XWdsufttWeS214I3LG3x0TCLlXwy07zC13opIt5WSBMvMzgN+BASBnzrnHojbPgB4ChgP1APXO+eW7W+f6Zpg1TVGeOeT7cz/eBvzV1VSsbsOgEnD+nHGhALOnFjI9FH9yQh2w+55qbL6NXjhWm+iii/81uvyI6lVswP++m1Y8oyXUJz73964id52AlW3OyaJWtV6v+XaU3gXEx18tNdNdexs733QlM09R8PefZOu5vW9m9rWzRnYNvnKyvOTjnDihCTS5JdF/PLw/utH27/+3yFz0dYY2sQTjokpNoFqomV824H0KfSSp6FTWxMpTVkuIj1QlydYZhYEVgNnA+XAIuAq59yKmDoPAtXOuW+b2UTgUefcWfvbb29MsJoiUSr3NrB1Tz3b9jawzV9ubVk2sK6ymoZwlNzMIKccOZgzJxYyZ0IBw/I1mDehkp97s9kNnQpX/wb6Dk11RBKrvMQ7PptLvdkaL/h+z0yA63Z5LVLbVvotUyu9hKp6S2udUC4UTPC6NRVM8CdUmAD5o3U9p96qsdZrrWmTfPkJWNVG9klELOjNaNg8G2Mgw5uEo3m9ZVvcLRhqnamx03+ksNb9x8cT//yJYmupE3O/3wgvmcor7ORYRURSo70EK5lTJc0E1jrn1vkBPA/MBVbE1JkMfAfAObfKzMaY2RDn3NYkxtVlGsNRKqv9RGlPA9v2esutMQlU5d4GdtQ07vPYgMHgvCwK+2UxPD+bWeMHMWdCATPHDuy6qcx7Iufgb/fCmz/wJrO4/Ofer8XSvYwsgpv+Boufhr/eA4+fAifd6k0/HmmCplpvQoKm2rb3G+PLm9frvHE0LdtilmbeyWHzSWow5F0zqeUEMeSfBMbVSbie4c1QV/mxd2uTSPXxEqfxZ7adICF/lBKpdJOZ640zTDTWsHmq+ZakJRnJkYiIpFIyE6wRQOwI4XLgxLg6S4FLgbfMbCZwBDASaJNgmdnNwM0Ao0ePTla8h2xPfRMrNu1h+aY9LK+oYuWWvWzdU8/OBIlTMGAMzstkSL9sRg7I4fgjBlDYN4vCvtkM6de6HJSX1bkX000H4Qb4w23w0Qtw/DVw4Q813XJ3FgjCjBtg8lxvSvd3fuzdOvTYkNcyFMrxTmab74dyvOtDhXL8smwv6Y42+d2rmlq7O0WaWtfDjRCtSbAtwWMCGd7kE0ee1doyVTgR+o1UIiUHlpGlrnAiIr1cMs8+E2UH8f0RHwB+ZGalwEfA34F9OpM7554AngCvi2DnhnlwdlQ3sHzTHpZtqmJ5xR6Wb6qibEdty/Yh/bKYNKwfx4/u35o0+YlTYb8sBvVR4pQUdbvh11+EsjfhzP+A0/5Nvwr3FH0Gw9xHYcaNsGWZnzD1iUmS4pOoXF0gVkRERLqtZCZY5UDsVVtHAm1G/jrn9gDXAZg3h/h6/5Zyzjm27KlnmZ9ENS83V9W31Bk9MJcpw/vxuaJRTBnejynD8ynoq18mu9zujfDs5bDjE7jkCTj2ilRHJIdi+HHeTURERKQHS2aCtQg4yszGAhXAlcDVsRXMrD9Q65xrBG4EFvpJV5dyzvHpzlqWVXgtU8sqqlixaU/L2CgzGF+Qx4ljBzJ1RD6Th/djyrB88nP1K3rKbV4Kz37eG2sz7/febGwiIiIiIimStATLORc2s9uBV/GmaX/KObfczG7xtz8OTAKeMbMI3uQXNyQrnmbVDWE+3rKXj7fsZdWWPazaspeVm/awt8HrmRgKGkcP6ctZkwqZOiKfKcPzmTSsL7mZGsvT7ax53ZuGPbs/3PCqNzubiIiIiEgK9doLDYcjUcp21LZJpFZt2cPGnXUtdfKyMpgwtC+Th/Vj6givi99RQ/I0S19PsPgX8Ke7vFm6rn4B+g1LdUQiIiIikkZSMU17l6nc27BPIrVmq3fdKPBm7hs7uA/HjuzPFUWjmDi0HxOG9mXkgBwsXSdCeOcRyOoLx17Zs2a0cg7m/zcsfBDGnwWf/4X3OkREREREuoEel2DVNUb4TcnGloTq4y172V7dOh16Qd8sJg7tyz+efERLInVkYR7ZIbVKtdi4CF77pnd//v1w8m1QdF33T1TCjfDyHfDh83DcPPjMQ5pNTkRERES6lR7XRTBr2FFu2DUPkx0KMGFIXyYM7cvEof2YONS7PyivB7XGpMpzV8DG92HuT+D9x2D9QsjOh5k3w4m3eNNmdzf1Vd407OsXwhnfgtlf1jTsIiIiIpIy7XUR7HEJ1lFTjnWvFb/D6IG5up7UodjyETx+Ksz5Bsz5mldWXgJvPQSr/gQZOXD8P8Ks26F/N7moc1U5PPs52L7au17SsVemOiIRERERSXO9ZgxWfk6IsYP7pDqMnuvNH0BmXzjx5taykUVw5bNQ+TG8/SMo+Zl3m/Y5OOVLqZ2db/OH8NznobEGvvg7GDcndbGIiIiIiBxAINUBSBfavgaWvwQzb4ScAftuL5gAF/8EvrQUZtwEK/4APzkJfnW1N26rq0SaYNPf4d1H4efngwXg+v9TciUiIiIi3V6Pa8GSw/DWQ5CRDSfdtv96+SPh/Adg9lfggyfg/cfh4z/DEafCaXd5s/d15vin6koo/wA2fgDli6BiCYT96fSHH++1rvUb3nnPJyIiIiKSJEqw0sWuDbD0eZh5E+QVdOwxfQbBGf8Os+6AJb/wpnb/38tg6DQ49S6YfDEEDnJ2xkgYti1vTaY2fgC71nvbAhkw7Fg44VoYNQNGzvSSPU1mISIiIiI9hBKsdPH2j7yudrP+5eAfm5XnTeU+40b48Dfw9sPw2+thwL3eGK3pV7d/La2aHX4i9X5r61RTjbctbwiMnOFNET9yJgyfDqGcQ32FIiIiIiIp1+NmESwqKnIlJSWpDqNn2bsFHj7Gm33voh8f/v6iEW/GwTd/CJtLIW8onHwrHH8NVG1s2zq18xPvMRb0Wr5GzfSSqVEzoP8Rap0SERERkR6p18wiKIfgnf+BaBOcemfn7C8QhMlzYdJFsG6BN7br9f/0bs1yB3vJ1HFfhFEnwvDjIDO3c55fRERERKSbUoLV29XsgJKfw9TLYeC4zt23GYw/w7tVLIZVf4bBR3uJ1YCxap0SERERkbSjBKu3e/8xb8zTaf+W3OcZcYJ3ExERERFJY7oOVm9WXwXvPwGTPguFE1MdjYiIiIhIr6cEqzdb9FNoqEp+65WIiIiIiABKsHqvxlp49ydw5D94E0yIiIiIiEjSJTXBMrPzzOxjM1trZl9PsD3fzP5oZkvNbLmZXZfMeNLKkl9A7XY47cupjkREREREJG0kLcEysyDwKHA+MBm4yswmx1W7DVjhnDsWmAP8wMwykxVT2gg3eBcWPuJUOOLkVEcjIiIiIpI2ktmCNRNY65xb55xrBJ4H5sbVcUBfMzMgD9gJhJMYU3oofQ72bobZGnslIiIiItKVkplgjQA2xqyX+2WxHgEmAZuAj4AvOeei8Tsys5vNrMTMSiorK5MVb+8QCcPbD8Pw42HcGamORkREREQkrSQzwUp0lVkXt34uUAoMB6YDj5hZv30e5NwTzrki51xRQUFBZ8fZuyz7Hewqg9lf1oV+RURERES6WDITrHJgVMz6SLyWqljXAb93nrXAekAXbDpU0Si8+QMonAJHn5/qaERERERE0k4yE6xFwFFmNtafuOJK4OW4Op8CZwGY2RBgArAuiTH1bqv+CNs/htP+FQKagV9EREREpKtlJGvHzrmwmd0OvAoEgaecc8vN7BZ/++PAvcDTZvYRXpfCrznnticrpl7NOVj4fRg4HqZckupoRERERETSUtISLADn3CvAK3Flj8fc3wSck8wY0sbaN2DLh3DRIxAIpjoaEREREZG0pH5kvUFz61X+KDjmilRHIyIiIiKStpRg9QYb3oaN78EpX4IMXadZRERERCRVlGD1BgsfhD6FcNwXUx2JiIiIiEhaU4LV05UvhnULYNbtEMpJdTQiIiIiImlNCVZP9+b3Ibs/FF2f6khERERERNKeEqyebOty+PgVOOlWyOqb6mhERERERNKeEqye7M0fQGZfOPHmVEciIiIiIiIoweq5tq+FZb+HGTdAzoBURyMiIiIiIijB6rneeggysuDk21IdiYiIiIiI+JRg9US7P4UPn4cTroW8wlRHIyIiIiIiPiVYPdHbPwYMZv1LqiMREREREZEYSrB6mr1bYckzMP0qyB+R6mhERERERCRGRkcqmdkpwH8BR/iPMcA558YlLzRJ6N3/gWgTnHJnqiMREREREZE4HUqwgJ8BdwGLgUjywpH9qt0Ji56CqZfDoPGpjkZEREREROJ0NMGqcs79JamRyIG9/zg01cBp/5rqSEREREREJIGOJljzzexB4PdAQ3Ohc25JUqKSfdXv8RKsiZ+BwkmpjkZERERERBLoaIJ1or8siilzwJn7e5CZnQf8CAgCP3XOPRC3/SvAF2JimQQUOOd2djCu9FHyM6ivgtlfTnUkIiIiIiLSjg4lWM65Mw52x2YWBB4FzgbKgUVm9rJzbkXMfh8EHvTrfxa4S8lVAo218M4jMP4sGH5cqqMREREREZF2dHQWwXzgbmC2X1QM3OOcq9rPw2YCa51z6/x9PA/MBVa0U/8q4FcdiadHiIQhXAdN9f4y5tZc3lQL4fqY8vrEdfZuhtrtMPsrqX5VIiIiIiKyHx3tIvgUsAz4vL8+D/g5cOl+HjMC2BizXk5rV8M2zCwXOA+4vZ3tNwM3A4wePbqDIadA/R546yH44Elo3Hto+wiEIJTj3TKyW+/PvBmOOLlz4xURERERkU7V0QRrvHPuspj1b5tZ6QEeYwnKXDt1Pwu83V73QOfcE8ATAEVFRe3tI3WiEfj7L+Fv90FNJUy5BAont02QMrIhlAuhbMjIab882NFDIiIiIiIi3U1Hz+brzOxU59xb0HLh4boDPKYcGBWzPhLY1E7dK+mp3QPXLYBXvwlbl8GoE+GqX8PIE1IdlYiIiIiIpEBHE6x/Bn7hj8UyYCdw7QEeswg4yszGAhV4SdTV8ZX8fZ4OfLGDsXQP29fCa9+C1X+B/NFw+c+9litL1HAnIiIiIiLpoKOzCJYCx5pZP399TwceEzaz24FX8aZpf8o5t9zMbvG3P+5XvQR4zTlXcwjxd73anVD8PVj0pNel76y74aRbvS5+IiIiIiKS1sy59oc0mdkXnXP/a2b/mmi7c+6HSYusHUVFRa6kpKSrnxYiTbDoZ1D8gHc9quPmwZnfgrzCro9FRERERERSyswWO+eK4ssP1ILVx1/27fyQegjnYPWrXnfAHWtg7Olw7v0wdGqqIxMRERERkW5mvwmWc+7/+ctvd0043czW5fDqN7yJLAYdCVc9D0efp3FWIiIiIiKSUKAjlczse2bWz8xCZvZXM9tuZj1rUoqDUV0Jf/wSPH4qbCqF874Lt74HE85XciUiIiIiIu3q6CyC5zjnvmpml+BNv/45YD7wv0mLLBWa6uH9x2DhDyBcBzP/CU7/KuQOTHVkIiIiIiLSA3Q0wQr5ywuAXznndlpvaslxDla8BK/fDbs3wNHnwzn3wuCjUh2ZiIiIiIj0IB1NsP5oZqvwLi58q5kVAPXJC6sLbf4Q/vJV+PRdKJwC816C8WekOioREREREemBOnodrK+b2XeBPc65iJnVAHOTG1oXiEbgfy8DHHz2R97U64FgqqMSEREREZEear8Jlpmd6Zz7m5ldGlMWW+X3yQqsS2z5EGq2waVPwjGfT3U0IiIiIiLSwx2oBet04G/AZxNsc/T0BGtdsbccOzu1cYiIiIiISK9woOtg3e0vr+uacLrY+mIomAh9h6Y6EhERERER6QU6eh2s+82sf8z6ADO7L2lRdYVwA2x4F8aenupIRERERESkl+hQggWc75zb3bzinNuFN2V7z1W+yLvW1TglWCIiIiIi0jk6mmAFzSyrecXMcoCs/dTv/tYVgwXgiFNSHYmIiIiIiPQSHb0O1v8CfzWzn+NNbnE98IukRdUV1hfD8OMgp3+qIxERERERkV6io9fB+p6ZfQj8A2DAvc65V5MaWTI17IWKxTDrX1IdiYiIiIiI9CIdbcECWAmEnXNvmFmumfV1zu1NVmBJteEdiIY1/kpERERERDpVR2cRvAn4LfD//KIRwEsdeNx5Zvaxma01s6+3U2eOmZWa2XIzK+5g3IdnXTEEs2DUiV3ydCIiIiIikh462oJ1GzATeB/AObfGzAr39wAzCwKPAmcD5cAiM3vZObcipk5/4CfAec65Tw+0z06zvhhGnwihnC55OhERERERSQ8dnUWwwTnX2LxiZhl4k13sz0xgrXNunf/Y54G5cXWuBn7vnPsUwDm3rYPxHLrqSti6DMbNSfpTiYiIiIhIeuloglVsZt8AcszsbOAF4I8HeMwIYGPMerlfFutoYICZLTCzxWb2j4l2ZGY3m1mJmZVUVlZ2MOR2lC30lmPnHN5+RERERERE4nQ0wfoaUAl8BPwT8ArwrQM8xhKUxbd6ZQAnABcC5wL/YWZH7/Mg555wzhU554oKCgo6GHI71hVDVj4Mn354+xEREREREYlzwDFYZhYAPnTOTQWePIh9lwOjYtZHApsS1NnunKsBasxsIXAssPognufgrC+GMadCIJi0pxARERERkfR0wBYs51wUWGpmow9y34uAo8xsrJllAlcCL8fV+QNwmpllmFkucCLedPDJsavMu2l6dhERERERSYKOziI4DFhuZh8ANc2FzrmL2nuAcy5sZrcDrwJB4Cnn3HIzu8Xf/rhzbqWZ/R/wIRAFfuqcW3aIr+XA1vmzwI9VgiUiIiIiIp2vownWtw9l5865V/DGa8WWPR63/iDw4KHs/6CtL4a8oVAwoUueTkRERERE0st+EywzywZuAY7Em+DiZ865cFcE1umcg/ULYdwZYInm3xARERERETk8BxqD9QugCC+5Oh/4QdIjSpZtK6CmUuOvREREREQkaQ7URXCyc24agJn9DPgg+SElicZfiYiIiIhIkh2oBaup+U6P7RrYbH0xDBwH/UcduK6IiIiIiMghOFAL1rFmtse/b0COv26Ac871S2p0nSUShrK3YdrlqY5ERERERER6sf0mWM653nE13k1LoHGvxl+JiIiIiEhSHfBCw71C8/irMbNTG4eIiIiIiPRq6ZFgrS+GodOgz6BURyIiIiIiIr1Y70+wGmth4/uaPVBERERERJKu9ydYG9+DSCOMm5PqSEREREREpJfr/QnWumIIZMDok1MdiYiIiIiI9HK9P8FaXwwjZ0JWXqojERERERGRXq53J1h1u2BTqaZnFxERERGRLtG7E6yytwCnCS5ERERERKRL9O4Ea10xhPrAiBNSHYmIiIiIiKSBXp5gLYAjZkFGZqojERERERGRNJDUBMvMzjOzj81srZl9PcH2OWZWZWal/u0/O+3J92yCHWs0/kpERERERLpMRrJ2bGZB4FHgbKAcWGRmLzvnVsRVfdM595lOD2BdsbfU+CsREREREekiyWzBmgmsdc6tc841As8Dc5P4fG2tL4bcQTBkapc9pYiIiIiIpLdkJlgjgI0x6+V+WbyTzWypmf3FzKYk2pGZ3WxmJWZWUllZeeBnds5rwRpzGgR69zAzERERERHpPpKZfViCMhe3vgQ4wjl3LPA/wEuJduSce8I5V+ScKyooKDjwM+9YC3s3afyViIiIiIh0qWQmWOXAqJj1kcCm2ArOuT3OuWr//itAyMwGH/Yzr1vgLTX+SkREREREulAyE6xFwFFmNtbMMoErgZdjK5jZUDMz//5MP54dh/3M64shfxQMHHfYuxIREREREemopM0i6JwLm9ntwKtAEHjKObfczG7xtz8OXA78s5mFgTrgSudcfDfCgxONwPo3YeJnwBL1UhQREREREUmOpCVY0NLt75W4ssdj7j8CPNKpT7rlQ6jfrfFXIiIiIiLS5ZKaYKVEy/WvZqc2DhERERGRJGhqaqK8vJz6+vpUh5IWsrOzGTlyJKFQqEP1e1+Ctb4YCiZC36GpjkREREREpNOVl5fTt29fxowZg2lITFI559ixYwfl5eWMHTu2Q4/pXReJCjfAhnc1e6CIiIiI9Fr19fUMGjRIyVUXMDMGDRp0UK2FvSvBKl8E4ToYNyfVkYiIiIiIJI2Sq65zsO9170qw1hWDBWDMKamORERERERE0lDvSrDWF8Pw4yE7P9WRiIiIiIj0Wlu2bOHKK69k/PjxTJ48mQsuuIDVq1dTVlbG1KlTk/Kc1157Lb/97W+Tsu/O1HsSrIa9ULFY07OLiIiIiCSRc45LLrmEOXPm8Mknn7BixQruv/9+tm7dmurQuoXeM4vghncgGtYEFyIiIiKSNr79x+Ws2LSnU/c5eXg/7v7slHa3z58/n1AoxC233NJSNn36dADKyspaysrKypg3bx41NTUAPPLII8yaNYvNmzdzxRVXsGfPHsLhMI899hizZs3ihhtuoKSkBDPj+uuv56677trnuRcuXMgPf/hDtmzZwve+9z0uv/xyqqurmTt3Lrt27aKpqYn77ruPuXPnUlZWxnnnncepp57Ke++9x7HHHst1113H3XffzbZt23j22WeZOXNm57xpMXpPgrVuAWRkw6gTUx2JiIiIiEivtWzZMk444YQD1issLOT1118nOzubNWvWcNVVV1FSUsJzzz3Hueeeyze/+U0ikQi1tbWUlpZSUVHBsmXLANi9e3fCfW7evJm33nqLVatWcdFFF3H55ZeTnZ3Niy++SL9+/di+fTsnnXQSF110EQBr167lhRde4IknnmDGjBk899xzvPXWW7z88svcf//9vPTSS531trToRQlWsZdchbJTHYmIiIiISJfYX0tTqjU1NXH77bdTWlpKMBhk9erVAMyYMYPrr7+epqYmLr74YqZPn864ceNYt24dd9xxBxdeeCHnnHNOwn1efPHFBAIBJk+e3NIl0TnHN77xDRYuXEggEKCioqJl29ixY5k2bRoAU6ZM4ayzzsLMmDZtWpvWts7UO8ZgVVfCtuUafyUiIiIikmRTpkxh8eLFB6z30EMPMWTIEJYuXUpJSQmNjY0AzJ49m4ULFzJixAjmzZvHM888w4ABA1i6dClz5szh0Ucf5cYbb0y4z6ysrJb7zjkAnn32WSorK1m8eDGlpaUMGTKk5bpVsfUDgUDLeiAQIBwOH9obcAC9I8FaX+wtx85JZRQiIiIiIr3emWeeSUNDA08++WRL2aJFiyguLm5Tr6qqimHDhhEIBPjlL39JJBIBYMOGDRQWFnLTTTdxww03sGTJErZv3040GuWyyy7j3nvvZcmSJR2Op6qqisLCQkKhEPPnz2fDhg2d80IPUe/oIri+GLLyYfj0VEciIiIiItKrmRkvvvgid955Jw888ADZ2dmMGTOGhx9+uE29W2+9lcsuu4wXXniBM844gz59+gCwYMECHnzwQUKhEHl5eTzzzDNUVFRw3XXXEY1GAfjOd77T4Xi+8IUv8NnPfpaioiKmT5/OxIkTO+21HgprblrrKYqKilxJSUnbwoePgSFT4arnUhOUiIiIiEgXWblyJZMmTUp1GGkl0XtuZoudc0XxdXt+F8FdZbB7g8ZfiYiIiIhIyvX8BGtd8/grJVgiIiIiIpJaSU2wzOw8M/vYzNaa2df3U2+GmUXM7PKDfpL1xZA3FAomHFasIiIiIiIihytpCZaZBYFHgfOBycBVZja5nXrfBV496CdxDtYvhLGzwewwIxYRERERETk8yWzBmgmsdc6tc841As8DcxPUuwP4HbDtoJ9h2wqoqdT4KxERERER6RaSmWCNADbGrJf7ZS3MbARwCfD4/nZkZjebWYmZlVRWVrZu0PgrERERERHpRpKZYCXqsxc/J/zDwNecc5H97cg594Rzrsg5V1RQUNC6YX0xDBwH/UcdbqwiIiIiItJBW7Zs4corr2T8+PFMnjyZCy64gNWrV1NWVsbUqVNTHV5KJfNCw+VAbOYzEtgUV6cIeN688VODgQvMLOyce+mAe4+EoextOOZznROtiIiIiIgckHOOSy65hGuuuYbnn38egNLSUrZu3cqoUWr4SGaCtQg4yszGAhXAlcDVsRWcc2Ob75vZ08CfOpRcAWxaAo171T1QRERERNLXX74OWz7q3H0OnQbnP9Du5vnz5xMKhbjllltayqZPnw5AWVlZS1lZWRnz5s2jpqYGgEceeYRZs2axefNmrrjiCvbs2UM4HOaxxx5j1qxZ3HDDDZSUlGBmXH/99dx1111tnveFF17g29/+NsFgkPz8fBYuXMjTTz/Niy++SENDA+vXr+fqq6/m7rvvBuDiiy9m48aN1NfX86UvfYmbb74ZgLy8PG677TbeeOMNBgwYwP33389Xv/pVPv30Ux5++GEuuuiiw3r7kpZgOefCZnY73uyAQeAp59xyM7vF377fcVcHtK4YMG8GQRERERER6RLLli3jhBNOOGC9wsJCXn/9dbKzs1mzZg1XXXUVJSUlPPfcc5x77rl885vfJBKJUFtbS2lpKRUVFSxbtgyA3bt377O/e+65h1dffZURI0a02f7BBx+wbNkycnNzmTFjBhdeeCFFRUU89dRTDBw4kLq6OmbMmMFll13GoEGDqKmpYc6cOXz3u9/lkksu4Vvf+havv/46K1as4Jprrum+CRaAc+4V4JW4soSJlXPu2oPa+fpiL7vOHXjI8YmIiIiI9Gj7aWlKtaamJm6//XZKS0sJBoOsXr0agBkzZnD99dfT1NTExRdfzPTp0xk3bhzr1q3jjjvu4MILL+Scc87ZZ3+nnHIK1157LZ///Oe59NJLW8rPPvtsBg0aBMCll17KW2+9RVFRET/+8Y958cUXAdi4cSNr1qxh0KBBZGZmct555wEwbdo0srKyCIVCTJs2rU0L3KFK6oWGk6axFja+r+nZRURERES62JQpU1i8ePEB6z300EMMGTKEpUuXUlJSQmNjIwCzZ89m4cKFjBgxgnnz5vHMM88wYMAAli5dypw5c3j00Ue58cYb99nf448/zn333cfGjRuZPn06O3bsAMDirodrZixYsIA33niDd999l6VLl3LcccdRX18PQCgUanlMIBAgKyur5X44HD70N8bXMxOsT9+FSCOMnZPqSERERERE0sqZZ55JQ0MDTz75ZEvZokWLKC4ublOvqqqKYcOGEQgE+OUvf0kk4k0cvmHDBgoLC7npppu44YYbWLJkCdu3bycajXLZZZdx7733smTJkn2e95NPPuHEE0/knnvuYfDgwWzc6F0R6vXXX2fnzp3U1dXx0ksvccopp1BVVcWAAQPIzc1l1apVvPfee0l8R9pKahfBpFlfDIEQHHFyqiMREREREUkrZsaLL77InXfeyQMPPEB2djZjxozh4YcfblPv1ltv5bLLLuOFF17gjDPOoE+fPgAsWLCABx98kFAoRF5eHs888wwVFRVcd911RKNRAL7zne/s87xf+cpXWLNmDc45zjrrLI499lhKS0s59dRTmTdvHmvXruXqq6+mqKiIadOm8fjjj3PMMccwYcIETjrppKS/L83MufhLU3VvRUVFruSmPhDKhev/kupwRERERES61MqVK5k0aVKqw+gWnn76aUpKSnjkkUeS+jyJ3nMzW+ycK4qv2/O6CEbDsHmpxl+JiIiIiEi30/O6CDZWA07XvxIRERERSXPXXnst1157barDaKPntWA1VEOoD4w48Nz7IiIiIiK9UU8b5tOTHex73QMTrL1wxCzIyEx1JCIiIiIiXS47O5sdO3YoyeoCzjl27NhBdnZ2hx/T87oIhus1/kpERERE0tbIkSMpLy+nsrIy1aGkhezsbEaOHNnh+j0vwQKNvxIRERGRtBUKhRg7dmyqw5B29LwugoEMGDI11VGIiIiIiIjso+clWAUTIdDzwhYRERERkd6v52UqwVCqIxAREREREUnIetrsI2ZWCWxIdRy91GBge6qDkIR0bLovHZvuTcen+9Kx6b50bLovHZvu5QjnXEF8YY9LsCR5zKzEOVeU6jhkXzo23ZeOTfem49N96dh0Xzo23ZeOTc/Q87oIioiIiIiIdFNKsERERERERDqJEiyJ9USqA5B26dh0Xzo23ZuOT/elY9N96dh0Xzo2PYDGYImIiIiIiHQStWCJiIiIiIh0EiVYIiIiIiIinUQJVhoys1FmNt/MVprZcjP7kl/+X2ZWYWal/u2CVMearsyszMw+8o9DiV820MxeN7M1/nJAquNMN2Y2Ieb7UWpme8zsTn13UsPMnjKzbWa2LKas3e+Jmf27ma01s4/N7NzURJ0+2jk+D5rZKjP70MxeNLP+fvkYM6uL+Q49nrLA00A7x6bdv2P67nSddo7Nr2OOS5mZlfrl+t50UxqDlYbMbBgwzDm3xMz6AouBi4HPA9XOue+nMj7xEiygyDm3Pabse8BO59wDZvZ1YIBz7mupijHdmVkQqABOBK5D350uZ2azgWrgGefcVL8s4ffEzCYDvwJmAsOBN4CjnXORFIXf67VzfM4B/uacC5vZdwH84zMG+FNzPUmudo7Nf5Hg75i+O10r0bGJ2/4DoMo5d4++N92XWrDSkHNus3NuiX9/L7ASGJHaqKQD5gK/8O//Ai8pltQ5C/jEObch1YGkK+fcQmBnXHF735O5wPPOuQbn3HpgLd4JoyRJouPjnHvNORf2V98DRnZ5YNLed6c9+u50of0dGzMzvB/Df9WlQclBU4KV5vxfP44D3veLbve7bjylLmgp5YDXzGyxmd3slw1xzm0GL0kGClMWnQBcSdt/cvrudA/tfU9GABtj6pWjH5ZS7XrgLzHrY83s72ZWbGanpSqoNJfo75i+O93HacBW59yamDJ9b7ohJVhpzMzygN8Bdzrn9gCPAeOB6cBm4Aepiy7tneKcOx44H7jN7zIg3YSZZQIXAS/4RfrudH+WoEx95FPEzL4JhIFn/aLNwGjn3HHAvwLPmVm/VMWXptr7O6bvTvdxFW1/2NP3pptSgpWmzCyEl1w965z7PYBzbqtzLuKciwJPoi4AKeOc2+QvtwEv4h2Lrf74ueZxdNtSF2HaOx9Y4pzbCvrudDPtfU/KgVEx9UYCm7o4NgHM7BrgM8AXnD8Q3O9+tsO/vxj4BDg6dVGmn/38HdN3pxswswzgUuDXzWX63nRfSrDSkN+H92fASufcD2PKh8VUuwRYFv9YST4z6+NPPoKZ9QHOwTsWLwPX+NWuAf6QmgiFuF8R9d3pVtr7nrwMXGlmWWY2FjgK+CAF8aU1MzsP+BpwkXOuNqa8wJ84BjMbh3d81qUmyvS0n79j+u50D/8ArHLOlTcX6HvTfWWkOgBJiVOAecBHzVN9At8ArjKz6XhN/2XAP6UiOGEI8KKXB5MBPOec+z8zWwT8xsxuAD4FPpfCGNOWmeUCZ9P2+/E9fXe6npn9CpgDDDazcuBu4AESfE+cc8vN7DfACryuabdpFrTkauf4/DuQBbzu/417zzl3CzAbuMfMwkAEuMU519FJGOQgtXNs5iT6O6bvTtdKdGyccz9j33G/oO9Nt6Vp2kVERERERDqJugiKiIiIiIh0EiVYIiIiIiIinUQJloiIiIiISCdRgiUiIiIiItJJlGCJiIiIiIh0EiVYIiKSFGa2wMzOjSu708x+0kn7v9bMHjnIxzxtZpd3xvN34LmKzOzHXfFcIiLSfSjBEhGRZPkV3rVbYiW6lku3ZWaHfL1I51yJc+5fOjMeERHp/pRgiYhIsvwW+IyZZQGY2RhgOPCWmV1lZh+Z2TIz+27zA8zsPDNbYmZLzeyvftlMM3vHzP7uLyfEPMcoM/s/M/vYzO5ufh4zWxazzy+b2X/FB2dm/2lmi/wYnjD/yrd+y9v9ZlYMfNPM1ptZyN/Wz8zKmtdj9vU5fz9LzWyhXzbHzP7k33/FzEr9W5WZXWNmQTN70I/hQzPTBapFRHqBQ/5lTkREZH+cczvM7APgPOAPeK1XvwaGAd8FTgB2Aa+Z2cXA28CTwGzn3HozG+jvapVfFjazfwDuBy7zt80EpgK1wCIz+zOwvYMhPuKcuwfAzH4JfAb4o7+tv3PudH/bGOBC4CX/NfzOOdcUt6//BM51zlWYWf8E78UF/r5OAH7u7+sGoMo5N8NPQt82s9ecc+s7GL+IiHRDasESEZFkiu0m2Nw9cAawwDlX6ZwLA88Cs4GTgIXNCYZzbqf/uHzgBb9V6iFgSsz+X3fO7XDO1QG/B049iNjOMLP3zewj4My4/f465v5Pgev8+9fhJUjx3gaeNrObgGCiJzOzwcAvgaudc1XAOcA/mlkp8D4wCDjqIOIXEZFuSC1YIiKSTC8BPzSz44Ec59wSMxvdTl0DXILye4H5zrlL/NakBTHb4us7IEzbHxCz93kis2zgJ0CRc26j34Uwtl5Nyw6de9vvdng6EHTOLSOOc+4WMzsRr6Wr1Mymxz1fEHgeuCfm8Qbc4Zx7NcFrFhGRHkotWCIikjTOuWq8hOgpWie3eB843cwG+4nHVUAx8K5fPhYgpotgPlDh37827inONrOBZpYDXIzXkrQVKDSzQX7Xu88kCK05mdpuZnnAgWYWfMaPP1HrFWY23jn3vnPuP/G6KI6Kq/IA8KFz7vmYsleBf44Z33W0mfU5QBwiItLNqQVLRESS7Vd43feuBHDObTazfwfm47XivOKc+wOAmd0M/N7MAsA24Gzge8AvzOxfgb/F7fstvG53RwLPOedK/P3cg5fIrccbw9WGc263mT0JfASUAYsO8BqeBe6j/RkQHzSzo/zX81dgKXB6zPYvA8v97oDgjdn6KTAGWOJPsFGJlySKiEgPZs4l6o0hIiIizfxrZ811zs1LdSwiItK9qQVLRERkP8zsf4DzgQtSHYuIiHR/asESERERERHpJJrkQkREREREpJMowRIREREREekkSrBEREREREQ6iRIsERERERGRTqIES0REREREpJP8fyBk9qzZYW/MAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 864x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_precision_recall(X_train, y_train, X_test, y_test, BoW)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Задание 3 (2 балла)\n",
    "Теперь добавим в BoW стемминг. Для этого нам понадобится класс SnowballStemmer из пакета `nltk`\n",
    "\n",
    "#### Методы\n",
    "`__init__(X, voc_limit)` - инициализирует BoW, составляя словарь, который будет использоваться для векторизации предложений. Размер словаря должен быть не больше `voc_limit`, в него должны входить самые часто встречающиеся в выборке слова.\n",
    "\n",
    "`transform(X)` - векторизует сообщения из набора."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [],
   "source": [
    "class BowStem:\n",
    "    def __init__(self, X, voc_limit=1000):\n",
    "        self.stemmer = SnowballStemmer(\"english\")\n",
    "        c = Counter()\n",
    "        for row in X:\n",
    "            c.update([self.stemmer.stem(i) for i in row.split()])\n",
    "        self.bow = dict() # word -> pos\n",
    "        for i, (k, v) in enumerate(sorted(c.items(), key=lambda x: x[1], reverse=True)[:voc_limit]):\n",
    "            self.bow[k] = i\n",
    "        \n",
    "    def transform(self, X):\n",
    "        ans = np.zeros((X.shape[0], len(self.bow)))\n",
    "        for i, row in enumerate(X):\n",
    "            for w in [self.stemmer.stem(i) for i in row.split()]:\n",
    "                t = self.bow.get(w, -1)\n",
    "                if t == -1:\n",
    "                    pass\n",
    "                else:\n",
    "                    ans[i][t] += 1\n",
    "        return ans"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "bows = BowStem(X_train, voc_limit=500)\n",
    "X_train_bows = bows.transform(X_train)\n",
    "X_test_bows = bows.transform(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "({'ham': 0.9696969696969697, 'spam': 0.9841269841269841},\n",
       " {'ham': 0.997920997920998, 'spam': 0.8051948051948052},\n",
       " 0.9713261648745519)"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictor = NaiveBayes(0.001)\n",
    "predictor.fit(X_train_bows, y_train)\n",
    "get_precision_recall_accuracy(predictor.predict(X_test_bows), y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1gAAADQCAYAAAAalMCAAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/d3fzzAAAACXBIWXMAAAsTAAALEwEAmpwYAAAsq0lEQVR4nO3deXxddbnv8c83SZukQ4bSUtIpLTO1hdKGgqKIckQcDgUcLjghg5VzxQk9F4/negb1KupBLyoeTlFQFAERvKLHIyLiqaBCk9IW2lLoPCS06ZB0Stomee4fe6Xspmmb0r2zd5Lv+/XKK3v91rCf3dW1s5/9+/2epYjAzMzMzMzMjl1BrgMwMzMzMzPrL5xgmZmZmZmZZYgTLDMzMzMzswxxgmVmZmZmZpYhTrDMzMzMzMwypCjXAWTSyJEjY+LEibkOw8zMzMzM+rm6urrNETGqa3u/SrAmTpxIbW1trsMwMzMzM7N+TtKa7to9RNDMzMzMzCxDnGCZmZmZmZllSFYTLEmXSFomabmkz3WzvlLSLyQtkvSMpClp6z4tabGk5yXdJ6kkm7GamZmZmZkdq6wlWJIKgduBtwGTgaskTe6y2eeBBRFxJvAh4LZk37HAJ4CaiJgCFAJXZitWMzMzMzOzTMhmD9ZMYHlErIyIvcD9wKwu20wGHgeIiBeAiZJGJ+uKgFJJRcAQoD6LsZqZmZmZmR2zbCZYY4F1acvrk7Z0C4ErACTNBKqBcRGxAfg3YC3QADRHxO+6exJJsyXVSqptbGzM8EswMzMzMzPruWwmWOqmLbos3wJUSloAfBx4FmiTVEmqt2sSMAYYKukD3T1JRMyJiJqIqBk16qAy9GZmZmZmZr0mm/fBWg+MT1seR5dhfhGxHbgGQJKAVcnPW4FVEdGYrHsYeB3wkyzGa2ZmZmZmdkyy2YM1DzhF0iRJg0kVqXgkfQNJFck6gOuBuUnStRY4T9KQJPG6CFiaxVjNzMzMzMyOWdZ6sCKiTdKNwKOkqgDeFRGLJd2QrL8DOAO4R1I7sAS4Lln3tKSfA/OBNlJDB+dkK1YzMzMzM7NMUETXaVF9V01NTdTW1uY6DDMzs16ze28b9U2tNDS3sHXXXjL9Z736uCG8Zkw5g4uyeutMM7O8t6+9g5ebW2loTr3nXnb2uLqIqOm6XTbnYJmZmdkx2NPWzsbmPdQ3t9DQ3LI/kapvaqW+qYWG5laaW/ZlPY7iogLOHFfOjOoRzKiuZPqECo4bVpz15zUz6y0dHUHjzj3731vTf9c3t9LQ1ELjzj09+hLLPVhmZmY50N4RbNrRmpY0vZJApf6ot7J5556D9qscMoiq8lLGVJRQVV5KVUUJY8pLqSovYeTwYgrUXRHfVx/jSxt3ULdmG3Vrt/H8hmb2tac+N0waOZTpEyqpmVjJjOpKTh41jIKCzD23WVdt7R205/nnViH39uahiGDb7n37k6YDv7BKPd64vZW2jgP/f5UOKjzgPbaqopQxab9PPaGs2x4sJ1hmZmYZFhFs2bW3m6Qp+ePe1MLGHXto7/LHfFhx0YF/xJMEamxF8se9vJTSwYU5elXQuq+d5zc0U7tmG3VrtjF/zTa27NoLwPCSIqZPSCVbNdWVnDW+gqHFHihjPXOo3oP0Htue9h7k2vCSotQH8uRLkDFdPphXlZdQMih313F/tKN130HvsfVdEqnWfR0H7DOoUJxQ/so5GlNResB775iKEspLB6HDfGklyQmWmZnZsYoItre0HTRsr6GpNWlLjc/f23bgH/PBRQUHJE2dH8DGVJTuf1xWMihHr+rViQjWbNl9QML14qYdRECB4IyqMmZUV+7/GVtRetgPK9Y/vdregyGDC6nq/OBbXsIJ5aUU53nvUESweWfy5UryvtD5JUS644YOPigBS3+to8tKGFSY36+1t7Tuaz8waTogeUr9G+/Y03bAPgWC0WUlB31hld7zP3Jo8TH3ujvBMjMz64H0ohH7k6a05Km+qYXde9sP2KewQJyQ/sc8bUhJ5wemEUMHD4jkorllHwvWNVG3eit1a7fx7Nqm/f9eo8uKOf2EMgYV9v9/B4Pde9v3J1Rdew8GFxYkvQevXCOdH4Q7v3QoKy3qF9dM6752Xm5+5b2kofmVRKHzPWV768EJwvHDS6iqKGHEkMH0g3+Go5IaQr2HhuZWtnaToI4cNjiVKHXz/6eqopTRw4sp6oUE1QmWmZkNeIcqGtHQ1MqGQxSNkGDksOJXhpB0M/9p1PBiCj3/qFtt7R0s27iD+Wu2UbtmGysad/aJYV527IqLCrr0Hrxy7Rw3dLDn7KXZuaft4B6a5D1p2+6DE4z+ToJRw4q7HS49uix/hlg6wTIzs7ySPmY+0+XFA9i2ay/1nUNIDlM0oqKzaER5yYFD9pKEanRZiSetm5nZQQ6VYHn2qZmZZdzhxsx3Drfb0WVITDYMHVy4f27DGSeUHTTnqaq8hCGD/afQzMwyx39VzMz6udZ97RnuHQq27tp7yIpNh5rUPWLoYKrKS5hw3BDOO3HEAZO6Rw4rpjDDkwzKhwyirKR/zOEwM7O+wwmWmVk/FBH8ecUW5sxdyX+/2Jj15xteXLS/d2jq2AqXJTYzswHLCZaZWT+yr72D3zzXwJy5K1lcv52Rw4q54Y0nUTEks+W/y0sHHVC9aXgfKy9uZmaWLU6wzMz6gZ172rj/mbXc/dRqNjS1cNKooXztXVOZNW2se47MzMx6kRMsM7M+7OXmVu7+8yp++vRadrS2ce6kEXxx1mt402nHuwSymZlZDjjBMjPrg5a9vIM5c1fyyMINtHcEb5taxUfecCLTxlfkOjQzM7MBzQmWmVkf0bVwRemgQt5/bjXXnj+JCccNyXV4ZmZmhhMsM8tjLXvbac/wzdBLBxVS2MeGznVXuOKzF5/K+8+tpnLo4FyHZ2ZmZmmcYJlZXmjvCF7cuIO6NduYv2YbdWu3sWbL7ow/T1GBGF2WusHs/jLi+x+nbj573NDBeXHvpK6FK04cNZRbrpjKZWe7cIWZmVm+ymqCJekS4DagEPh+RNzSZX0lcBdwEtAKXBsRzyfrKoDvA1OASNb9JZvxmlnv2dG6jwXrmqhbs426NdtYsLaJHXvaABg5rJia6kreM2McxUWZSySCoLllHw1NrdQ3t7BofROPPt/K3vaOA7YrLipIJV1JwjWmy++q8tKs3sB24/ZW7n5qNfc+vYYdrW3MnDSCf730Nbz5dBeuMDMzy3dZS7AkFQK3A28B1gPzJD0SEUvSNvs8sCAiLpd0erL9Rcm624DfRsS7JQ0GPMHArI+KCNZtbaFu7VZqV6cSqmUbdxABBYLTTihj1tljmFFdyYwJIxg/orTXepA6OoItu/bS0NxCfVMrDc0tNDS3Ut+U+v3XFVvYuGMP7R0HDlUcOriQMRWlHF9WTEEGY21rD2rXbE0VrphSxfVvmMTZEyozdnwzMzPLrmz2YM0ElkfESgBJ9wOzgPQEazLwVYCIeEHSREmjgRbgAuDDybq9wN4sxmpmGbSnrZ3nN2ynbs3WpIeqic079wAwvLiIaRMquGTKCdRUj+Cs8eU5vUltQYEYNbyYUcOLOXNc99u0tXewaceeA5Kwzt+NO/aQ2Vli8L6ZE7j29ZOoPm5oho9sZmZm2ZbNBGsssC5teT1wbpdtFgJXAE9KmglUA+OAdqARuFvSWUAd8MmI2NX1SSTNBmYDTJgwIdOvwcx6oHHHntTcqbWp3qnn1jfvH3ZXfdwQLjhlJDMmVjKjupJTjh/e54pMFBUWMKailDEVpcyoznU0ZmZmls+ymWB19wmq6xe9twC3SVoAPAc8C7QBg4DpwMcj4mlJtwGfA75w0AEj5gBzAGpqajL9RbKZdXG4YhSDiwo4c2w515w/kenVlUyfUMmo4cU5jtjMzMys92QzwVoPjE9bHgfUp28QEduBawCUmnCxKvkZAqyPiKeTTX9OKsEys17Wk2IUHzi3munVlUwZW5bRohRmZmZmfU02E6x5wCmSJgEbgCuB96VvkFQK3J3MsboemJskXdslrZN0WkQsI1X4YglmllXpxSg6504te3k7Hd0Uo6ipHsG4yt4rRmFmZmbWF2QtwYqINkk3Ao+SKtN+V0QslnRDsv4O4AzgHkntpBKo69IO8XHg3qSC4EqSni4zy5zOYhTzk96purXbaNyRKkYxrLiIsydU8NaLTmFGdSXTxlfktBiFmZmZWV+giP4zbammpiZqa2tzHYZZ3mrcsYf5a1Nzp2q7KUYxY0Il06tTxShOHd33ilGYmZmZ9RZJdRFR07U9qzcaNrPcae8IXtqUKkZRt7pLMYrCAqaOK+fq11Uzo3oEM6pdjMLMzMwsE5xgmfUTRypGMaO6gvefO4EZ1SNcjMLMzMwsS5xgmfVBhytGIcFpo4dz6bQx1EysZMaEEYwf4WIUZmZmZr3BCZZZH9Cyt50lDS5GYWZmZpbvnGCZ5di+9g5ebm6lobmVhuYW6ptaqW9q2f+4obmFbbv37d9+woghvOHkkS5GYWZmZpaHnGCZZVFHR9C4c0+SMLXu/52eSDXu3EPXYp5lJUWMqSilqryEaRMqGFNewsnHD2N6dSXHDy/JzYsxMzMzsyNygmWWYX9ZsYXv/OEl1m7dzcbtrexrPzB7Kh1USFVFCWPKS3njqaOoqihlTHnJAb+HFfvSNDMzM+uL/CnOLENa97Xztd++wN1PrWZsRSnnTKx8JWkqL2VMRSljKkooLx3kghNmZmZm/ZQTLLMMeHbtNj7z4EJWNu7i6tdWc/PbTmfIYF9eZmZmZgONPwGaHYO9bR18+/GX+N4fl3NCWQk/ue5cXn/KyFyHZWZmZmY54gTL7FVa2rCdm362kKUN23n3jHH8099Opszl0c3MzMwGNCdYZkeprb2DOX9aybcee5Hy0sHc+aEa3jJ5dK7DMjMzM7M84ATL7CisbNzJZx5cyLNrm3j71BP48mVTGTF0cK7DMjMzM7M84QTLrAc6OoJ7/rKaW377AsVFhdx25TQuPWuMqwGamZmZ2QGcYJkdwfptu/lfP1/En1ds4cLTRvG1d53J6DLf7NfMzMzMDuYEy+wQIoIH69bzxV8tISL46hVTufKc8e61MjMzM7NDcoJl1o1NO1r5/MPP8fulm5g5aQS3vucsxo8YkuuwzMzMzCzPFWTz4JIukbRM0nJJn+tmfaWkX0haJOkZSVO6rC+U9KykX2czTrN0/7mogbd+ay5zX9rM/37HGdz/kfOcXJmZmZlZj2StB0tSIXA78BZgPTBP0iMRsSRts88DCyLickmnJ9tflLb+k8BSoCxbcZp1atq9ly/8cjG/WljPWePKufW9Z3Hy8cNzHZaZmZmZ9SHZ7MGaCSyPiJURsRe4H5jVZZvJwOMAEfECMFHSaABJ44B3AN/PYoxmADzxwiYu/tZc/uu5Bm56y6k89Hevc3JlZmZmZkctm3OwxgLr0pbXA+d22WYhcAXwpKSZQDUwDtgI/F/gfwGH/ZQraTYwG2DChAmZiNv6qV172mhobqG+qfWA32u27ObpVVs5bfRw7vrwOUwZW57rUM3MzMysj8pmgtVdqbXosnwLcJukBcBzwLNAm6R3Apsiok7ShYd7koiYA8wBqKmp6Xp8GyD2tLXzcnPr/qSpobmVDU0tNDSlHtc3tbC9te2AfSQYNayYqopSPvHmk/nYm0+muKgwR6/AzMzMzPqDbCZY64HxacvjgPr0DSJiO3ANgFK1r1clP1cCl0p6O1AClEn6SUR8IIvxWp5r3dfOovXNLFrfxPptLdQnyVNDcwubd+49aPvKIYOoKi9lXOUQZk4aQVV5KWMqSqgqL6WqvITRZSUMLspqnRczMzMzG2COmGAlvUm/iYiOozz2POAUSZOADaSSpvd1OXYFsDuZo3U9MDdJuv4h+SHpwfqsk6uBZ+P2VurWbKN29Tbq1m5j8YZm2jpSnZTDiov2J0tTxpbtT5rGVKR+V5WXUjrYvVFmZmZm1rt60oN1JalhfA8Bd0fE0p4cOCLaJN0IPAoUAndFxGJJNyTr7wDOAO6R1A4sAa57NS/C+r629g5eeHkHdWu27f/Z0NQCQHFRAWeNr2D2BScyo7qSaeMrOG5YcY4jNjMzMzM7mCKOPG1JUhlwFanhfAHcDdwXETuyG97Rqampidra2lyHYT3Q3LKP+Wu3MT9Jphasa2L33nYATigrYcbESmZMqGRGdSVnVJV5KJ+ZmZmZ5RVJdRFR07W9R3OwImJ70oNVCnwKuBz4e0nfjojvZDRS63ciglWbd1G3Zhvz16YSqhc37gSgsEBMrirjvTXjmV6dSqjGlJeQmpJnZmZmZta39GQO1t8C1wInAT8GZkbEJklDSN0E2AmWdWtJ/XZ+8OQqnli2ia27UkUoykqKmFFdyaVnjWF6dSVnjatgaHE2a62YmZmZmfWennyyfQ/wrYiYm94YEbslXZudsKyvigieXL6ZOXNX8qeXNjNkcCGXTDmBmRNHMKO6kpNGDaOgwL1TZmZmZtY/9STB+megoXNBUikwOiJWR8TjWYvM+pR97R38elE9c+auYmnDdkYNL+bv33oaHzi3mvIhg3IdnpmZmZlZr+hJgvUg8Lq05fak7ZysRGR9yo7Wfdz/zDruemoVDc2tnHL8ML7+7jOZNW2Mb9prZmZmZgNOTxKsouQ+VQBExF5Jg7MYk/UBDc0t/PCp1fz06bXs2NPGeSeO4CuXT+WNp47yEEAzMzMzG7B6kmA1Sro0Ih4BkDQL2JzdsCxfLW3Yzp1zV/LIwno6InjHmWP4yBsmcea4ilyHZmZmZmaWcz1JsG4A7pX0XUDAOuBDWY3K8kpE8NTyLfzH3BX7C1d88LXVXHv+JMaPGJLr8MzMzMzM8sYRE6yIWAGcJ2kYqRsT59XNhS17DlW44v3nTqBiiEeJmpmZmZl11aMbEEl6B/AaoKTzBrAR8cUsxmU51LVwxcnHD+Pr7zqTWWe7cIWZmZmZ2eH05EbDdwBDgDcB3wfeDTyT5bgsB3btaePbf3iJn/71lcIV/+fyKVx46vEuXGFmZmZm1gM96cF6XUScKWlRRPyrpFuBh7MdmPWu1Zt38dEf1/Hiph28Y2oVsy840YUrzMzMzMyOUk8SrNbk925JY4AtwKTshWS97Yllm/jkfc9SUCDuuXYmbzhlVK5DMjMzMzPrk3qSYP1KUgXwDWA+EMCd2QzKekdHR/C9Py7n1sde5PQTypjzwRmuCmhmZmZmdgwOm2BJKgAej4gm4CFJvwZKIqK5N4Kz7Nm5p43P/GwBjy7eyGXTxvDVK86kdLALWJiZmZmZHYvDJlgR0ZHMuXptsrwH2NMbgVn2rGjcyUd/XMeqzbv4wjsnc+35E+msDmlmZmZmZq9eT4YI/k7Su4CHIyKyHZBl12NLNnLTAwsYVFTAT647l9eedFyuQzIzMzMz6zcKerDNTcCDwB5J2yXtkLS9JweXdImkZZKWS/pcN+srJf1C0iJJz0iakrSPl/SEpKWSFkv65FG9KjtIR0fwzcde5CP31DJx5FB+9fHXO7kyMzMzM8uwI/ZgRcTwV3NgSYXA7cBbgPXAPEmPRMSStM0+DyyIiMslnZ5sfxHQBnwmIuZLGg7USXqsy77WQ80t+7jpgQU8/sIm3j1jHF++bAolgzzfyszMzMws03pyo+ELumuPiLlH2HUmsDwiVibHuR+YBaQnSZOBrybHe0HSREmjI6IBaEjad0haCoztsq/1wEsbdzD7x3Ws27qbL816DR84r9rzrczMzMzMsqQnc7D+Pu1xCanEqQ548xH2GwusS1teD5zbZZuFwBXAk5JmAtXAOGBj5waSJgJnA0939ySSZgOzASZMmHCEkAaW/3qugc8+uJDSwUXcN/s8zpk4ItchmZmZmZn1az0ZIvi36cuSxgNf78Gxu+sm6Vok4xbgNkkLgOeAZ0kND+x8rmHAQ8CnIqLbeV8RMQeYA1BTU+MiHEB7R3Dr75bxvT+uYNr4Cu74wAxOKC/JdVhmZmZmZv1eT3qwuloPTOnhduPTlscB9ekbJEnTNQBKjVtblfwgaRCp5OreiHj4VcQ5IDXt3ssn7l/A3BcbuWrmBP7l0skUF3m+lZmZmZlZb+jJHKzv8ErPUwEwjdTQviOZB5wiaRKwAbgSeF+XY1cAuyNiL3A9MDcitifJ1g+ApRHxzZ69FFvasJ2P/riOl5tb+eoVU7lqpodMmpmZmZn1pp70YNWmPW4D7ouIp460U0S0SboReBQoBO6KiMWSbkjW3wGcAdwjqZ1UAYvrkt3PBz4IPJcMHwT4fET8pgfxDkiPLKzn5p8voqy0iPs/eh7TJ1TmOiQzMzMzswFHR7p3sKShQGtEtCfLhUBxROzuhfiOSk1NTdTW1h55w36krb2Drz+6jDlzV3LOxEpuf/90jh/u+VZmZmZmZtkkqS4iarq29+RGw48DpWnLpcDvMxWYvXpbd+3l6rufYc7clXzotdXce/15Tq7MzMzMzHKoJ0MESyJiZ+dCROyUNCSLMdkR1De1cPdTq7jvmXXsbe/gG+8+k/fUjD/yjmZmZmZmllU9SbB2SZoeEfMBJM0AWrIblnVnSf127vzTSn61sJ4A3nlmFf/zwpM57YThuQ7NzMzMzMzoWYL1KeBBSZ0l1quA/5G1iOwAEcGfXtrMnX9ayZ9e2szQwYVc/bqJXHP+RMZVuiPRzMzMzCyf9ORGw/MknQ6cRurmwS9ExL6sRzbA7W3r4NeL6pkzdyUvvLyD44cXc/Mlp/O+cydQXjoo1+GZmZmZmVk3enIfrI+Rutnv88lypaSrIuJ7WY9uANreuo/7n1nLXU+u5uXtrZw6ehjfePeZzJo2lsFFPalJYmZmZmZmudKTIYIfiYjbOxciYpukjwBOsDKovqmFH/55NT99ei0797TxupOO45Z3TeWNp44idd9lMzMzMzPLdz1JsAokKZIbZiX3wRqc3bAGjq6FK94xtYrZF5zIlLHluQ7NzMzMzMyOUk8SrEeBn0m6AwjgBuC/shpVPxcRPLl8M3PmunCFmZmZmVl/0pME62ZgNvB3pIpcPEuqkqAdpUMWrpg5gfIhLlxhZmZmZtbX9aSKYIekvwInkirPPgJ4KNuB9TcrGnfyoR88w4amlv2FKy6dNobiosJch2ZmZmZmZhlyyARL0qnAlcBVwBbgAYCIeFPvhNZ/7Gjdx+x7amnd187dHz6HC09z4QozMzMzs/7ocD1YLwB/Av42IpYDSPp0r0TVj3R0BJ9+YCFrtuzmJ9efy3knHpfrkMzMzMzMLEsOd2OldwEvA09IulPSRaTmYNlR+PYfXuL3Szfyv99xhpMrMzMzM7N+7pAJVkT8IiL+B3A68Efg08BoSf8u6eJeiq9Pe2zJRv7v71/iXdPHcfXrJuY6HDMzMzMzy7LD9WABEBG7IuLeiHgnMA5YAHwu24H1dSsad3LTAwuYOrac/3P5FM+5MjMzMzMbAI6YYKWLiK0R8R8R8eZsBdQfdBa1GFxUwH98cAYlg1wp0MzMzMxsIDiqBOtoSbpE0jJJyyUd1OslqVLSLyQtkvSMpCk93TdfdRa1WL1lN7e/fzpjKkpzHZKZmZmZmfWSrCVYkgqB24G3AZOBqyRN7rLZ54EFEXEm8CHgtqPYNy995w/LXdTCzMzMzGyAymYP1kxgeUSsjIi9wP3ArC7bTAYeB4iIF4CJkkb3cN+88/slG/nW71/kiulj+bCLWpiZmZmZDTjZTLDGAuvSltcnbekWAlcASJoJVJMqpNGTfUn2my2pVlJtY2NjhkI/eisad/LppKjFVy6f6qIWZmZmZmYDUDYTrO4yjOiyfAtQKWkB8HHgWaCth/umGiPmRERNRNSMGjXqGMJ99VzUwszMzMzMAIqyeOz1wPi05XFAffoGEbEduAZAqS6fVcnPkCPtmy86OoKbfpYqavGT6851UQszMzMzswEsmz1Y84BTJE2SNBi4EngkfQNJFck6gOuBuUnSdcR988V3/rCcx5akilq89iQXtTAzMzMzG8iy1oMVEW2SbgQeBQqBuyJisaQbkvV3AGcA90hqB5YA1x1u32zF+mq5qIWZmZmZmaVTRLdTm/qkmpqaqK2t7ZXnWtG4k8u++xQTRw7lwRte63lXZmZmZmYDiKS6iKjp2p7VGw33V+lFLe5wUQszMzMzM0tks8hFv9S1qMVYF7UwMzMzM7OEe7COkotamJmZmZnZoTjBOgouamFmZmZmZofjBKuHVjTu5NMPLGDq2HK+cvlUUrftMjMzMzMze4UTrB5wUQszMzMzM+sJF7k4Ahe1MDMzMzOznnIP1hF0FrX4x7e7qIWZmZmZmR2eE6zDeHzpK0Utrjl/Yq7DMTMzMzOzPOcE6xBWNO7kU/e7qIWZmZmZmfWcE6xubN21l+t/VMsgF7UwMzMzM7Oj4ASri9Z97Vz/o3nUN7Vw54dmuKiFmZmZmZn1mKsIpmnvCD55/7M8u66Jf3//dGZUj8h1SGZmZmZm1oe4BysREXzp10t4dPFGvvCOyVwypSrXIZmZmZmZWR/jBCvxgydX8cM/r+a610/i2tdPynU4ZmZmZmbWBznBAn69qJ4v/+dS3j71BP7x7WfkOhwzMzMzM+ujBnyC9cyqrdz0wEJqqiv55nunUVDgcuxmZmZmZvbqZDXBknSJpGWSlkv6XDfryyX9StJCSYslXZO27tNJ2/OS7pNUkun4lm/awUfuqWXciFLu/FCNy7GbmZmZmdkxyVqCJakQuB14GzAZuErS5C6bfQxYEhFnARcCt0oaLGks8AmgJiKmAIXAlZmMb9OOVq6+ax6DCgv40TUzqRw6OJOHNzMzMzOzASibPVgzgeURsTIi9gL3A7O6bBPAcEkChgFbgbZkXRFQKqkIGALUZyqwXXvauPaH89i6ay93fbiG8SOGZOrQZmZmZmY2gGUzwRoLrEtbXp+0pfsucAap5Ok54JMR0RERG4B/A9YCDUBzRPyuuyeRNFtSraTaxsbGIwbV1t7BjT+dz5L67dz+/rM5c1zF0b4uMzMzMzOzbmUzwequWkR0WX4rsAAYA0wDviupTFIlqd6uScm6oZI+0N2TRMSciKiJiJpRo0YdNqCI4Au/fJ4nljXy5cum8ubTRx/VCzIzMzMzMzucbCZY64HxacvjOHiY3zXAw5GyHFgFnA78DbAqIhojYh/wMPC6Yw3ou39Yzn3PrOPGN53M+86dcKyHMzMzMzMzO0A2E6x5wCmSJkkaTKpIxSNdtlkLXAQgaTRwGrAyaT9P0pBkftZFwNJjCeahuvXc+tiLXHH2WD5z8anHcigzMzMzM7NuFWXrwBHRJulG4FFSVQDviojFkm5I1t8BfAn4oaTnSA0pvDkiNgObJf0cmE+q6MWzwJxXG8uTL23m5ocWcf7Jx3HLu84klbOZmZmZmZllliK6Tovqu2pqaqK2tvaAtqUN23nPHX9hXGUpP7vhtZSVDMpRdGZmZmZm1l9IqouImq7tWb3RcK41NLdwzd3zGFZcxN3XnOPkyszMzMzMsqrfJljbW/fx4bvmsWtPG3dfcw5V5aW5DsnMzMzMzPq5rM3ByqW9bR189J46VjTu5EfXzuSMqrJch2RmZmZmZgNAv0uwIoKbH1rEX1Zu4ZvvPYvzTx6Z65DMzMzMzGyA6HdDBP/td8v4xbMb+OzFp3LF9HG5DsfMzMzMzAaQfpVgbd21l9ufWMFVM8fzsTednOtwzMzMzMxsgOlXCdaGphbedNoovjRriu91ZWZmZmZmva5fJVilgwr57vumU1TYr16WmZmZmZn1Ef0qE5k4cihDi/td3Q4zMzMzM+sj+lWCVVTgYYFmZmZmZpY7/SrBMjMzMzMzyyUnWGZmZmZmZhniBMvMzMzMzCxDnGCZmZmZmZlliBMsMzMzMzOzDHGCZWZmZmZmliFOsMzMzMzMzDIkqwmWpEskLZO0XNLnullfLulXkhZKWizpmrR1FZJ+LukFSUslvTabsZqZmZmZmR2rrCVYkgqB24G3AZOBqyRN7rLZx4AlEXEWcCFwq6TBybrbgN9GxOnAWcDSbMVqZmZmZmaWCdnswZoJLI+IlRGxF7gfmNVlmwCGSxIwDNgKtEkqAy4AfgAQEXsjoimLsZqZmZmZmR2zbCZYY4F1acvrk7Z03wXOAOqB54BPRkQHcCLQCNwt6VlJ35c0tLsnkTRbUq2k2sbGxoy/CDMzMzMzs54qyuKx1U1bdFl+K7AAeDNwEvCYpD8lcU0HPh4RT0u6Dfgc8IWDDhgxB5gDIKlR0pqMvQLrNBLYnOsg7CA+L/nL5yY/+bzkJ5+X/OVzk598XvJHdXeN2Uyw1gPj05bHkeqpSncNcEtEBLBc0irgdGAtsD4ink62+zmpBOuwImLUMUdtB5FUGxE1uY7DDuTzkr98bvKTz0t+8nnJXz43+cnnJf9lc4jgPOAUSZOSwhVXAo902WYtcBGApNHAacDKiHgZWCfptGS7i4AlWYzVzMzMzMzsmGWtBysi2iTdCDwKFAJ3RcRiSTck6+8AvgT8UNJzpIYU3hwRnV2eHwfuTZKzlaR6u8zMzMzMzPJWNocIEhG/AX7Tpe2OtMf1wMWH2HcB4O7P/DAn1wFYt3xe8pfPTX7yeclPPi/5y+cmP/m85Dmlpj+ZmZmZmZnZscrmHCwzMzMzM7MBxQmWmZmZmZlZhjjBsv0kjZf0hKSlkhZL+mTS/i+SNkhakPy8PdexDkSSVkt6LjkHtUnbCEmPSXop+V2Z6zgHEkmnpV0XCyRtl/QpXzO5IekuSZskPZ/WdshrRNI/SFouaZmkt+Ym6v7vEOflG5JekLRI0i8kVSTtEyW1pF07dxzywHZMDnFeDvne5eul9xzi3DyQdl5WS1qQtPuayUOeg2X7SaoCqiJivqThQB1wGfBeYGdE/Fsu4xvoJK0GatIqbSLp68DWiLhF0ueAyoi4OVcxDmSSCoENwLmkqp76mullki4AdgL3RMSUpK3ba0TSZOA+YCYwBvg9cGpEtOco/H7rEOflYuAPScXhrwEk52Ui8OvO7Sx7DnFe/oVu3rt8vfSu7s5Nl/W3As0R8UVfM/nJPVi2X0Q0RMT85PEOYCkwNrdR2RHMAn6UPP4RqYTYcuMiYEVErMl1IANVRMwFtnZpPtQ1Mgu4PyL2RMQqYDmpD4+WYd2dl4j4XUS0JYt/Bcb1emAD3CGul0Px9dKLDnduJInUF9/39WpQdlScYFm3km9EzgaeTppuTIZy3OVhaDkTwO8k1UmanbSNjogGSCXIwPE5i86u5MA/eL5m8sOhrpGxwLq07dbjL5Ry5Vrgv9KWJ0l6VtJ/S3pDroIawLp77/L1kj/eAGyMiJfS2nzN5BknWHYQScOAh4BPRcR24N+Bk4BpQANwa+6iG9DOj4jpwNuAjyVDCCwPJDdEvxR4MGnyNZP/1E2bx8z3Mkn/CLQB9yZNDcCEiDgbuAn4qaSyXMU3AB3qvcvXS/64igO/zPM1k4ecYNkBJA0ilVzdGxEPA0TExohoj4gO4E48LCAnkhtzExGbgF+QOg8bk7lznXPoNuUuwgHtbcD8iNgIvmbyzKGukfXA+LTtxgH1vRzbgCbpauCdwPsjmRCeDEHbkjyuA1YAp+YuyoHlMO9dvl7ygKQi4Arggc42XzP5yQmW7ZeM6/0BsDQivpnWXpW22eXA8133teySNDQpPIKkocDFpM7DI8DVyWZXA7/MTYQD3gHfKPqaySuHukYeAa6UVCxpEnAK8EwO4huQJF0C3AxcGhG709pHJQVjkHQiqfOyMjdRDjyHee/y9ZIf/gZ4ISLWdzb4mslPRbkOwPLK+cAHgec6y38CnweukjSN1HCA1cBHcxHcADca+EUqB6YI+GlE/FbSPOBnkq4D1gLvyWGMA5KkIcBbOPC6+Lqvmd4n6T7gQmCkpPXAPwO30M01EhGLJf0MWEJqiNrHXBEtOw5xXv4BKAYeS97X/hoRNwAXAF+U1Aa0AzdERE8LMdhROMR5ubC79y5fL72ru3MTET/g4Lm+4GsmL7lMu5mZmZmZWYZ4iKCZmZmZmVmGOMEyMzMzMzPLECdYZmZmZmZmGeIEy8zMzMzMLEOcYJmZmZmZmWWIEywzM8s6SX+U9NYubZ+S9L0MHf/Dkr57lPv8UNK7M/H8PXiuGknf7o3nMjOz3HKCZWZmveE+UvdwSdfdPV3ylqRXfe/IiKiNiE9kMh4zM8tPTrDMzKw3/Bx4p6RiAEkTgTHAk5KukvScpOclfa1zB0mXSJovaaGkx5O2mZL+LOnZ5Pdpac8xXtJvJS2T9M+dzyPp+bRjflbSv3QNTtI/SZqXxDBHyd1vk563r0j6b+AfJa2SNChZVyZpdedy2rHekxxnoaS5SduFkn6dPP6NpAXJT7OkqyUVSvpGEsMiSb45tZlZH/Wqv40zMzPrqYjYIukZ4BLgl6R6rx4AqoCvATOAbcDvJF0GPAXcCVwQEaskjUgO9ULS1ibpb4CvAO9K1s0EpgC7gXmS/hPY3MMQvxsRXwSQ9GPgncCvknUVEfHGZN1E4B3A/0tew0MRsa/Lsf4JeGtEbJBU0c2/xduTY80A7k6OdR3QHBHnJEnoU5J+FxGrehi/mZnlCfdgmZlZb0kfJtg5PPAc4I8R0RgRbcC9wAXAecDczgQjIrYm+5UDDya9Ut8CXpN2/MciYktEtAAPA68/itjeJOlpSc8Bb+5y3AfSHn8fuCZ5fA2pBKmrp4AfSvoIUNjdk0kaCfwYeF9ENAMXAx+StAB4GjgOOOUo4jczszzhHiwzM+st/w/4pqTpQGlEzJc04RDbCohu2r8EPBERlye9SX9MW9d1+wDaOPDLxJKDnkgqAb4H1ETEumQIYfp2u/YfMOKpZNjhG4HCiHieLiLiBknnkurpWiBpWpfnKwTuB76Ytr+Aj0fEo928ZjMz60Pcg2VmZr0iInaSSoju4pXiFk8Db5Q0Mkk8rgL+G/hL0j4JIG2IYDmwIXn84S5P8RZJIySVApeR6knaCBwv6bhk6N07uwmtM5naLGkYcKTKgvck8XfXe4WkkyLi6Yj4J1JDFMd32eQWYFFE3J/W9ijwd2nzu06VNPQIcZiZWR5yD5aZmfWm+0gN37sSICIaJP0D8ASpXpzfRMQvASTNBh6WVABsAt4CfB34kaSbgD90OfaTpIbdnQz8NCJqk+N8kVQit4rUHK4DRESTpDuB54DVwLwjvIZ7gS9z6AqI35B0SvJ6HgcWAm9MW/9ZYHEyHBBSc7a+D0wE5icFNhpJJYlmZtbHKKK7ERhmZmbWneTeWbMi4oO5jsXMzPKPe7DMzMx6SNJ3gLcBb891LGZmlp/cg2VmZmZmZpYhLnJhZmZmZmaWIU6wzMzMzMzMMsQJlpmZmZmZWYY4wTIzMzMzM8sQJ1hmZmZmZmYZ8v8Bq1ruVe9CpqQAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 864x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_precision_recall(X_train, y_train, X_test, y_test, BowStem)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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": 2
}