HW_4_ignatiev.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Домашняя работа, Игнатьев Д. Д."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from sklearn.feature_extraction.text import CountVectorizer, TfidfVectorizer\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import classification_report, accuracy_score\n",
    "from sklearn.metrics.pairwise import cosine_distances, cosine_similarity\n",
    "\n",
    "from IPython.display import Image\n",
    "from IPython.core.display import HTML "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## TF-IDF"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 6,
     "metadata": {
      "image/png": {
       "height": 500,
       "width": 1000
      }
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(data=\"tfIdf.png\",\n",
    "     width=1000, height=500)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Близости текстов"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = pd.read_csv('labeled.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>comment</th>\n",
       "      <th>toxic</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Верблюдов-то за что? Дебилы, бл...\\n</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Хохлы, это отдушина затюканого россиянина, мол...</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Собаке - собачья смерть\\n</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Страницу обнови, дебил. Это тоже не оскорблени...</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>тебя не убедил 6-страничный пдф в том, что Скр...</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                             comment  toxic\n",
       "0               Верблюдов-то за что? Дебилы, бл...\\n    1.0\n",
       "1  Хохлы, это отдушина затюканого россиянина, мол...    1.0\n",
       "2                          Собаке - собачья смерть\\n    1.0\n",
       "3  Страницу обнови, дебил. Это тоже не оскорблени...    1.0\n",
       "4  тебя не убедил 6-страничный пдф в том, что Скр...    1.0"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "vectorizer = TfidfVectorizer(min_df=6, max_df=0.5, max_features=1500)\n",
    "X = vectorizer.fit_transform(data.comment)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Комментарий: в двух постах нет совпадающих слов, поэтому косинусная близость в любом случае будет нулевой"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Собаке - собачья смерть\n",
      "\n",
      "ну, как же: от Владивостока до ЮЖХ 1700 км по автодорогам, а между ними Татарский пролив)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(data[\"comment\"][2])\n",
    "print(data[\"comment\"][12665])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.]])"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cosine_similarity(X[2], X[12665])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Близкие тексты"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([  42, 1983, 2168, 2044], dtype=int64)"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dist = cosine_distances(X[42], X)\n",
    "dist.argsort()[0,:4]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Косинусная близость: [0.54327201]\n",
      "Текст: 1983    регулярно игнорировались После чего орда прихо...\n",
      "Name: comment, dtype: object\n",
      "\n",
      "\n",
      "Косинусная близость: [0.49476779]\n",
      "Текст: 2168    что генетически русские ебать даун\\n\n",
      "Name: comment, dtype: object\n",
      "\n",
      "\n",
      "Косинусная близость: [0.48523005]\n",
      "Текст: 2044    РРЯЯЯЯ, КУБАНОИДЫ НЕ РУССКИЕ, НАХУЙ ВЫ НУЖНЫ К...\n",
      "Name: comment, dtype: object\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "for index in [1983, 2168, 2044]:\n",
    "    print(f\"Косинусная близость: {1-dist[:,index]}\")\n",
    "    text = data.loc[[index], \"comment\"]\n",
    "    print(f\"Текст: {text}\")\n",
    "    print(\"\\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Классификация"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(file=\"../2ch_corpus.txt\", mode=\"r\", encoding=\"UTF-8\") as dvach:\n",
    "    texts = dvach.read().split(\"\\n\")\n",
    "    ye_olde_texts = pd.DataFrame(data=texts, columns=[\"text\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Анимублядский WebM-треддля приличных анимубля...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Посоны, поделитесь ссылкой на ютуб-версию...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Побольше тебе стандопавера, бро.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Перекатился. Благословил.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>У меня нофап- мне туда даже заходить нельзя</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                text\n",
       "0   Анимублядский WebM-треддля приличных анимубля...\n",
       "1       Посоны, поделитесь ссылкой на ютуб-версию...\n",
       "2                   Побольше тебе стандопавера, бро.\n",
       "3                          Перекатился. Благословил.\n",
       "4        У меня нофап- мне туда даже заходить нельзя"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ye_olde_texts.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Мерджим таблицы"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(14412,)\n",
      "(85904,)\n"
     ]
    }
   ],
   "source": [
    "print(data[\"comment\"].shape)\n",
    "print(ye_olde_texts[\"text\"].shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [],
   "source": [
    "merged = pd.concat([data[\"comment\"], ye_olde_texts[\"text\"]], axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(100316,)"
      ]
     },
     "execution_count": 98,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "merged.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Обучение векторайзеров"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "from nltk.tokenize import RegexpTokenizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "tokenizer = RegexpTokenizer(r\"\\w+\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "C_vectorizer = CountVectorizer(min_df=6, max_df=0.5, lowercase=True, tokenizer=tokenizer.tokenize, max_features=3000, ngram_range=(1,2))\n",
    "T_vectorizer = TfidfVectorizer(min_df=6, max_df=0.5, lowercase=True, tokenizer=tokenizer.tokenize, max_features=3000, ngram_range=(1,2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "counts = C_vectorizer.fit_transform(merged)\n",
    "tfs = T_vectorizer.fit_transform(merged)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_C = counts[:14412,:]\n",
    "X_predict_C = counts[14412:,:]\n",
    "X_T = tfs[:14412,:]\n",
    "X_predict_T = tfs[14412:,:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = data.toxic"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Деление на трейн-тест"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Для того, чтобы понять качество обученной модели нам нужно разделить данные на обучающую и тестовую выборку."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_C_train, X_C_test, y_C_train, y_C_test = train_test_split(X_C, y, test_size=0.1, shuffle=True)\n",
    "X_T_train, X_T_test, y_T_train, y_T_test = train_test_split(X_T, y, test_size=0.1, shuffle=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf_log = LogisticRegression(C=1, n_jobs=3)\n",
    "clf_NB = MultinomialNB(alpha=1., fit_prior=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ОБучение"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.83      0.89      0.86       937\n",
      "         1.0       0.76      0.66      0.71       505\n",
      "\n",
      "    accuracy                           0.81      1442\n",
      "   macro avg       0.80      0.77      0.78      1442\n",
      "weighted avg       0.81      0.81      0.81      1442\n",
      "\n"
     ]
    }
   ],
   "source": [
    "clf_log.fit(X_C_train, y_C_train)\n",
    "predict_log = clf_log.predict(X_C_test)\n",
    "print(classification_report(y_C_test, predict_log))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.85      0.88      0.87       949\n",
      "         1.0       0.75      0.71      0.73       493\n",
      "\n",
      "    accuracy                           0.82      1442\n",
      "   macro avg       0.80      0.79      0.80      1442\n",
      "weighted avg       0.82      0.82      0.82      1442\n",
      "\n"
     ]
    }
   ],
   "source": [
    "clf_NB.fit(X_T_train, y_T_train)\n",
    "predict_NB = clf_NB.predict(X_T_test)\n",
    "print(classification_report(y_T_test, predict_NB))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {},
   "outputs": [],
   "source": [
    "toxic_log = clf_log.predict_proba(X_predict_C)\n",
    "toxic_NB = clf_NB.predict_proba(X_predict_T)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Результаты"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [],
   "source": [
    "ye_olde_texts[\"toxic_log\"] = [n[1] for n in toxic_log]\n",
    "ye_olde_texts[\"toxic_NB\"] = [n[1] for n in toxic_NB]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>text</th>\n",
       "      <th>toxic_log</th>\n",
       "      <th>toxic_NB</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>51892</th>\n",
       "      <td>Че ебать, еблет закрой. Есть блять тебе по еб...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.999269</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>78161</th>\n",
       "      <td>Ебаный школьник как ты заебал блядь рот твой ...</td>\n",
       "      <td>0.999990</td>\n",
       "      <td>0.999228</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30710</th>\n",
       "      <td>Я ЕБАЛ ТЕБЯ В РОТ ЙОХАН И МАТЬ ТВОЮ ШЛЮХУ СП...</td>\n",
       "      <td>0.999950</td>\n",
       "      <td>0.998991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51907</th>\n",
       "      <td>Обосрался? Ахуеть ты блять чмошник. Вы тольк...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.998647</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33918</th>\n",
       "      <td>Сказочный долбоёб просто. Нахуй ты убежал, де...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.998646</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>71359</th>\n",
       "      <td>Иди нахуй, всем похуй на твой тред и говно бр...</td>\n",
       "      <td>0.999882</td>\n",
       "      <td>0.998507</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41658</th>\n",
       "      <td>Маня Быдло За щёку Бугурт Бомбануло Пригорел ...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.998443</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28369</th>\n",
       "      <td>Что ты здесь извиваешься петух? За шлюху свою...</td>\n",
       "      <td>0.999989</td>\n",
       "      <td>0.998099</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>83364</th>\n",
       "      <td>Я ебал твою мамашу Сходи нюхни трусы своего п...</td>\n",
       "      <td>0.999974</td>\n",
       "      <td>0.998025</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59820</th>\n",
       "      <td>Ты дебил тупой блять читай пуе хуесосина преж...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.997979</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    text  toxic_log  toxic_NB\n",
       "51892   Че ебать, еблет закрой. Есть блять тебе по еб...   1.000000  0.999269\n",
       "78161   Ебаный школьник как ты заебал блядь рот твой ...   0.999990  0.999228\n",
       "30710    Я ЕБАЛ ТЕБЯ В РОТ ЙОХАН И МАТЬ ТВОЮ ШЛЮХУ СП...   0.999950  0.998991\n",
       "51907    Обосрался? Ахуеть ты блять чмошник. Вы тольк...   1.000000  0.998647\n",
       "33918   Сказочный долбоёб просто. Нахуй ты убежал, де...   1.000000  0.998646\n",
       "71359   Иди нахуй, всем похуй на твой тред и говно бр...   0.999882  0.998507\n",
       "41658   Маня Быдло За щёку Бугурт Бомбануло Пригорел ...   1.000000  0.998443\n",
       "28369   Что ты здесь извиваешься петух? За шлюху свою...   0.999989  0.998099\n",
       "83364   Я ебал твою мамашу Сходи нюхни трусы своего п...   0.999974  0.998025\n",
       "59820   Ты дебил тупой блять читай пуе хуесосина преж...   1.000000  0.997979"
      ]
     },
     "execution_count": 135,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ye_olde_texts.sort_values(\"toxic_NB\", ascending=False).head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>text</th>\n",
       "      <th>toxic_log</th>\n",
       "      <th>toxic_NB</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>65189</th>\n",
       "      <td>Бомжиха Света Было это, когда я учился в шест...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.946105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75128</th>\n",
       "      <td>ИЩЕМ ШКУР, КОТОРЫЕ СНИМАЛИСЬ В ПОРНО ИЛИ РАБ...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.891571</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>67154</th>\n",
       "      <td>Сап, двощ! Пишу тебе с большой просьбой к адм...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.906598</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51892</th>\n",
       "      <td>Че ебать, еблет закрой. Есть блять тебе по еб...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.999269</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>67089</th>\n",
       "      <td>ХВАТИТ УЖЕ ДУМАТЬ ОБ ЭТОМ. ТЫ ПРОСТО ИСПОЛЬЗУ...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.981874</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>64360</th>\n",
       "      <td>Когда я вижу такие треды, я не чувствую нен...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.984079</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37188</th>\n",
       "      <td>\"Понимание\" — оно в процессе отношений и даже...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.980289</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>67551</th>\n",
       "      <td>&gt;Наверняка тебе не раз приходила в голову так...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.978702</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>65197</th>\n",
       "      <td>История про цирк Мы с корешем срали под дверь...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.966062</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37240</th>\n",
       "      <td>Вот так вот почитаешь пасты про всех этих пёз...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.975902</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    text  toxic_log  toxic_NB\n",
       "65189   Бомжиха Света Было это, когда я учился в шест...        1.0  0.946105\n",
       "75128    ИЩЕМ ШКУР, КОТОРЫЕ СНИМАЛИСЬ В ПОРНО ИЛИ РАБ...        1.0  0.891571\n",
       "67154   Сап, двощ! Пишу тебе с большой просьбой к адм...        1.0  0.906598\n",
       "51892   Че ебать, еблет закрой. Есть блять тебе по еб...        1.0  0.999269\n",
       "67089   ХВАТИТ УЖЕ ДУМАТЬ ОБ ЭТОМ. ТЫ ПРОСТО ИСПОЛЬЗУ...        1.0  0.981874\n",
       "64360     Когда я вижу такие треды, я не чувствую нен...        1.0  0.984079\n",
       "37188   \"Понимание\" — оно в процессе отношений и даже...        1.0  0.980289\n",
       "67551   >Наверняка тебе не раз приходила в голову так...        1.0  0.978702\n",
       "65197   История про цирк Мы с корешем срали под дверь...        1.0  0.966062\n",
       "37240   Вот так вот почитаешь пасты про всех этих пёз...        1.0  0.975902"
      ]
     },
     "execution_count": 139,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ye_olde_texts.sort_values(\"toxic_log\", ascending=False).head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Совпадает текст 51892. В большинстве текстов присутствует обсценная лексика, что служит для алгоритма показателем токсичности. Прогнозы классификаторов для \"топовых\" текстов почти не расходятся, оба предсказывают высокую вероятность."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\deprecation.py:143: FutureWarning: The sklearn.metrics.scorer module is  deprecated in version 0.22 and will be removed in version 0.24. The corresponding classes / functions should instead be imported from sklearn.metrics. Anything that cannot be imported from sklearn.metrics is now part of the private API.\n",
      "  warnings.warn(message, FutureWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\deprecation.py:143: FutureWarning: The sklearn.feature_selection.base module is  deprecated in version 0.22 and will be removed in version 0.24. The corresponding classes / functions should instead be imported from sklearn.feature_selection. Anything that cannot be imported from sklearn.feature_selection is now part of the private API.\n",
      "  warnings.warn(message, FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "import eli5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "    <style>\n",
       "    table.eli5-weights tr:hover {\n",
       "        filter: brightness(85%);\n",
       "    }\n",
       "</style>\n",
       "\n",
       "\n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "        \n",
       "\n",
       "    \n",
       "\n",
       "        \n",
       "            \n",
       "                \n",
       "                \n",
       "    \n",
       "        <p style=\"margin-bottom: 0.5em; margin-top: 0em\">\n",
       "            <b>\n",
       "    \n",
       "        y=1.0\n",
       "    \n",
       "</b>\n",
       "\n",
       "top features\n",
       "        </p>\n",
       "    \n",
       "    <table class=\"eli5-weights\"\n",
       "           style=\"border-collapse: collapse; border: none; margin-top: 0em; table-layout: auto; margin-bottom: 2em;\">\n",
       "        <thead>\n",
       "        <tr style=\"border: none;\">\n",
       "            \n",
       "                <th style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\" title=\"Feature weights. Note that weights do not account for feature value scales, so if feature values have different scales, features with highest weights might not be the most important.\">\n",
       "                    Weight<sup>?</sup>\n",
       "                </th>\n",
       "            \n",
       "            <th style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">Feature</th>\n",
       "            \n",
       "        </tr>\n",
       "        </thead>\n",
       "        <tbody>\n",
       "        \n",
       "            <tr style=\"background-color: hsl(120, 100.00%, 80.00%); border: none;\">\n",
       "    <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "        +3.605\n",
       "    </td>\n",
       "    <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "        хохлов\n",
       "    </td>\n",
       "    \n",
       "</tr>\n",
       "        \n",
       "            <tr style=\"background-color: hsl(120, 100.00%, 80.50%); border: none;\">\n",
       "    <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "        +3.478\n",
       "    </td>\n",
       "    <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "        хохлы\n",
       "    </td>\n",
       "    \n",
       "</tr>\n",
       "        \n",
       "            <tr style=\"background-color: hsl(120, 100.00%, 83.08%); border: none;\">\n",
       "    <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "        +2.839\n",
       "    </td>\n",
       "    <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "        дебил\n",
       "    </td>\n",
       "    \n",
       "</tr>\n",
       "        \n",
       "            <tr style=\"background-color: hsl(120, 100.00%, 83.26%); border: none;\">\n",
       "    <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "        +2.797\n",
       "    </td>\n",
       "    <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "        тупые\n",
       "    </td>\n",
       "    \n",
       "</tr>\n",
       "        \n",
       "            <tr style=\"background-color: hsl(120, 100.00%, 84.55%); border: none;\">\n",
       "    <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "        +2.493\n",
       "    </td>\n",
       "    <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "        сука\n",
       "    </td>\n",
       "    \n",
       "</tr>\n",
       "        \n",
       "            <tr style=\"background-color: hsl(120, 100.00%, 84.72%); border: none;\">\n",
       "    <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "        +2.455\n",
       "    </td>\n",
       "    <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "        хуйню\n",
       "    </td>\n",
       "    \n",
       "</tr>\n",
       "        \n",
       "            <tr style=\"background-color: hsl(120, 100.00%, 84.85%); border: none;\">\n",
       "    <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "        +2.426\n",
       "    </td>\n",
       "    <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "        русских\n",
       "    </td>\n",
       "    \n",
       "</tr>\n",
       "        \n",
       "            <tr style=\"background-color: hsl(120, 100.00%, 84.92%); border: none;\">\n",
       "    <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "        +2.409\n",
       "    </td>\n",
       "    <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "        моча\n",
       "    </td>\n",
       "    \n",
       "</tr>\n",
       "        \n",
       "            <tr style=\"background-color: hsl(120, 100.00%, 85.28%); border: none;\">\n",
       "    <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "        +2.327\n",
       "    </td>\n",
       "    <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "        бабы\n",
       "    </td>\n",
       "    \n",
       "</tr>\n",
       "        \n",
       "            <tr style=\"background-color: hsl(120, 100.00%, 85.33%); border: none;\">\n",
       "    <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "        +2.316\n",
       "    </td>\n",
       "    <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "        блять\n",
       "    </td>\n",
       "    \n",
       "</tr>\n",
       "        \n",
       "            <tr style=\"background-color: hsl(120, 100.00%, 85.51%); border: none;\">\n",
       "    <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "        +2.275\n",
       "    </td>\n",
       "    <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "        хохол\n",
       "    </td>\n",
       "    \n",
       "</tr>\n",
       "        \n",
       "            <tr style=\"background-color: hsl(120, 100.00%, 85.58%); border: none;\">\n",
       "    <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "        +2.260\n",
       "    </td>\n",
       "    <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "        быдло\n",
       "    </td>\n",
       "    \n",
       "</tr>\n",
       "        \n",
       "            <tr style=\"background-color: hsl(120, 100.00%, 85.61%); border: none;\">\n",
       "    <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "        +2.252\n",
       "    </td>\n",
       "    <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "        00\n",
       "    </td>\n",
       "    \n",
       "</tr>\n",
       "        \n",
       "            <tr style=\"background-color: hsl(120, 100.00%, 86.40%); border: none;\">\n",
       "    <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "        +2.078\n",
       "    </td>\n",
       "    <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "        шлюхи\n",
       "    </td>\n",
       "    \n",
       "</tr>\n",
       "        \n",
       "            <tr style=\"background-color: hsl(120, 100.00%, 86.74%); border: none;\">\n",
       "    <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "        +2.004\n",
       "    </td>\n",
       "    <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "        шлюх\n",
       "    </td>\n",
       "    \n",
       "</tr>\n",
       "        \n",
       "            <tr style=\"background-color: hsl(120, 100.00%, 86.86%); border: none;\">\n",
       "    <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "        +1.978\n",
       "    </td>\n",
       "    <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "        со своим\n",
       "    </td>\n",
       "    \n",
       "</tr>\n",
       "        \n",
       "            <tr style=\"background-color: hsl(120, 100.00%, 86.91%); border: none;\">\n",
       "    <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "        +1.968\n",
       "    </td>\n",
       "    <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "        народа\n",
       "    </td>\n",
       "    \n",
       "</tr>\n",
       "        \n",
       "            <tr style=\"background-color: hsl(120, 100.00%, 87.07%); border: none;\">\n",
       "    <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "        +1.934\n",
       "    </td>\n",
       "    <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "        очередной\n",
       "    </td>\n",
       "    \n",
       "</tr>\n",
       "        \n",
       "            <tr style=\"background-color: hsl(120, 100.00%, 87.21%); border: none;\">\n",
       "    <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "        +1.904\n",
       "    </td>\n",
       "    <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "        да еще\n",
       "    </td>\n",
       "    \n",
       "</tr>\n",
       "        \n",
       "        \n",
       "            <tr style=\"background-color: hsl(120, 100.00%, 87.21%); border: none;\">\n",
       "                <td colspan=\"2\" style=\"padding: 0 0.5em 0 0.5em; text-align: center; border: none; white-space: nowrap;\">\n",
       "                    <i>&hellip; 1354 more positive &hellip;</i>\n",
       "                </td>\n",
       "            </tr>\n",
       "        \n",
       "\n",
       "        \n",
       "            <tr style=\"background-color: hsl(0, 100.00%, 86.50%); border: none;\">\n",
       "                <td colspan=\"2\" style=\"padding: 0 0.5em 0 0.5em; text-align: center; border: none; white-space: nowrap;\">\n",
       "                    <i>&hellip; 1602 more negative &hellip;</i>\n",
       "                </td>\n",
       "            </tr>\n",
       "        \n",
       "        \n",
       "            <tr style=\"background-color: hsl(0, 100.00%, 86.50%); border: none;\">\n",
       "    <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "        -2.056\n",
       "    </td>\n",
       "    <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "        и только\n",
       "    </td>\n",
       "    \n",
       "</tr>\n",
       "        \n",
       "\n",
       "        </tbody>\n",
       "    </table>\n",
       "\n",
       "            \n",
       "        \n",
       "\n",
       "        \n",
       "\n",
       "\n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "\n",
       "\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": 141,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "eli5.show_weights(clf_log, top=20, feature_names=T_vectorizer.get_feature_names())"
   ]
  }
 ],
 "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": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}