CNN_Toxic_Comment_Classification.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "6940f86e"
      },
      "outputs": [],
      "source": [
        "try:\n",
        "    %tensorflow_version 2.x\n",
        "except Exception:\n",
        "    pass\n",
        "\n",
        "import tensorflow as tf\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd\n",
        "import re"
      ],
      "id": "6940f86e"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1VAO1fllsZO_",
        "outputId": "d2082dbc-d70e-43c5-a2c0-09f47b9f919d"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ]
        }
      ],
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "id": "1VAO1fllsZO_"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ljA7St3msizP",
        "outputId": "f69d3603-d08b-4bc8-f5e6-d8226baaccd0"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Archive:  drive/MyDrive/jigsaw-toxic-comment-classification-challenge.zip\n",
            "replace data/sample_submission.csv.zip? [y]es, [n]o, [A]ll, [N]one, [r]ename: "
          ]
        }
      ],
      "source": [
        "!unzip drive/MyDrive/jigsaw-toxic-comment-classification-challenge.zip -d data"
      ],
      "id": "ljA7St3msizP"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "90d99ec8"
      },
      "outputs": [],
      "source": [
        "train_data=pd.read_csv('data/train.csv.zip')"
      ],
      "id": "90d99ec8"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "id": "6afdc10c",
        "outputId": "fbd31265-7389-4138-8bc5-8cfe69156b47"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-fddc2285-0b59-4fc0-9ed4-faeda4bc5136\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>id</th>\n",
              "      <th>comment_text</th>\n",
              "      <th>toxic</th>\n",
              "      <th>severe_toxic</th>\n",
              "      <th>obscene</th>\n",
              "      <th>threat</th>\n",
              "      <th>insult</th>\n",
              "      <th>identity_hate</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0000997932d777bf</td>\n",
              "      <td>Explanation\\nWhy the edits made under my usern...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>000103f0d9cfb60f</td>\n",
              "      <td>D'aww! He matches this background colour I'm s...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>000113f07ec002fd</td>\n",
              "      <td>Hey man, I'm really not trying to edit war. It...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0001b41b1c6bb37e</td>\n",
              "      <td>\"\\nMore\\nI can't make any real suggestions on ...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0001d958c54c6e35</td>\n",
              "      <td>You, sir, are my hero. Any chance you remember...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>159566</th>\n",
              "      <td>ffe987279560d7ff</td>\n",
              "      <td>\":::::And for the second time of asking, when ...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>159567</th>\n",
              "      <td>ffea4adeee384e90</td>\n",
              "      <td>You should be ashamed of yourself \\n\\nThat is ...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>159568</th>\n",
              "      <td>ffee36eab5c267c9</td>\n",
              "      <td>Spitzer \\n\\nUmm, theres no actual article for ...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>159569</th>\n",
              "      <td>fff125370e4aaaf3</td>\n",
              "      <td>And it looks like it was actually you who put ...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>159570</th>\n",
              "      <td>fff46fc426af1f9a</td>\n",
              "      <td>\"\\nAnd ... I really don't think you understand...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>159571 rows × 8 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-fddc2285-0b59-4fc0-9ed4-faeda4bc5136')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-fddc2285-0b59-4fc0-9ed4-faeda4bc5136 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-fddc2285-0b59-4fc0-9ed4-faeda4bc5136');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "                      id  ... identity_hate\n",
              "0       0000997932d777bf  ...             0\n",
              "1       000103f0d9cfb60f  ...             0\n",
              "2       000113f07ec002fd  ...             0\n",
              "3       0001b41b1c6bb37e  ...             0\n",
              "4       0001d958c54c6e35  ...             0\n",
              "...                  ...  ...           ...\n",
              "159566  ffe987279560d7ff  ...             0\n",
              "159567  ffea4adeee384e90  ...             0\n",
              "159568  ffee36eab5c267c9  ...             0\n",
              "159569  fff125370e4aaaf3  ...             0\n",
              "159570  fff46fc426af1f9a  ...             0\n",
              "\n",
              "[159571 rows x 8 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 88
        }
      ],
      "source": [
        "train_data"
      ],
      "id": "6afdc10c"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "d43a7346"
      },
      "outputs": [],
      "source": [
        "test_data=pd.read_csv('data/test.csv.zip')"
      ],
      "id": "d43a7346"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "id": "d06317cc",
        "outputId": "83767981-7294-4dc7-f0dc-cff1d858b363"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-e7bcfb31-2719-4dd2-b8e8-4bf25e91fb94\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>id</th>\n",
              "      <th>comment_text</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>00001cee341fdb12</td>\n",
              "      <td>Yo bitch Ja Rule is more succesful then you'll...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0000247867823ef7</td>\n",
              "      <td>== From RfC == \\n\\n The title is fine as it is...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>00013b17ad220c46</td>\n",
              "      <td>\" \\n\\n == Sources == \\n\\n * Zawe Ashton on Lap...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>00017563c3f7919a</td>\n",
              "      <td>:If you have a look back at the source, the in...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>00017695ad8997eb</td>\n",
              "      <td>I don't anonymously edit articles at all.</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>153159</th>\n",
              "      <td>fffcd0960ee309b5</td>\n",
              "      <td>. \\n i totally agree, this stuff is nothing bu...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>153160</th>\n",
              "      <td>fffd7a9a6eb32c16</td>\n",
              "      <td>== Throw from out field to home plate. == \\n\\n...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>153161</th>\n",
              "      <td>fffda9e8d6fafa9e</td>\n",
              "      <td>\" \\n\\n == Okinotorishima categories == \\n\\n I ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>153162</th>\n",
              "      <td>fffe8f1340a79fc2</td>\n",
              "      <td>\" \\n\\n == \"\"One of the founding nations of the...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>153163</th>\n",
              "      <td>ffffce3fb183ee80</td>\n",
              "      <td>\" \\n :::Stop already. Your bullshit is not wel...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>153164 rows × 2 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-e7bcfb31-2719-4dd2-b8e8-4bf25e91fb94')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-e7bcfb31-2719-4dd2-b8e8-4bf25e91fb94 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-e7bcfb31-2719-4dd2-b8e8-4bf25e91fb94');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "                      id                                       comment_text\n",
              "0       00001cee341fdb12  Yo bitch Ja Rule is more succesful then you'll...\n",
              "1       0000247867823ef7  == From RfC == \\n\\n The title is fine as it is...\n",
              "2       00013b17ad220c46  \" \\n\\n == Sources == \\n\\n * Zawe Ashton on Lap...\n",
              "3       00017563c3f7919a  :If you have a look back at the source, the in...\n",
              "4       00017695ad8997eb          I don't anonymously edit articles at all.\n",
              "...                  ...                                                ...\n",
              "153159  fffcd0960ee309b5  . \\n i totally agree, this stuff is nothing bu...\n",
              "153160  fffd7a9a6eb32c16  == Throw from out field to home plate. == \\n\\n...\n",
              "153161  fffda9e8d6fafa9e  \" \\n\\n == Okinotorishima categories == \\n\\n I ...\n",
              "153162  fffe8f1340a79fc2  \" \\n\\n == \"\"One of the founding nations of the...\n",
              "153163  ffffce3fb183ee80  \" \\n :::Stop already. Your bullshit is not wel...\n",
              "\n",
              "[153164 rows x 2 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 90
        }
      ],
      "source": [
        "test_data"
      ],
      "id": "d06317cc"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "zldoGNuvAGzM"
      },
      "outputs": [],
      "source": [
        "# from sklearn.feature_extraction.text import ENGLISH_STOP_WORDS\n",
        "\n",
        "# def remove_stop_word(text):\n",
        "#   words = [word for word in text.split() if word.lower() not in ENGLISH_STOP_WORDS]\n",
        "#   return \" \".join(words)"
      ],
      "id": "zldoGNuvAGzM"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "8SUZwI3HCp3x"
      },
      "outputs": [],
      "source": [
        "import string\n",
        "def remove_punc(text):\n",
        "  return text.translate(str.maketrans('', '', string.punctuation))"
      ],
      "id": "8SUZwI3HCp3x"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "NZLux8WqGOZh"
      },
      "outputs": [],
      "source": [
        "# import nltk\n",
        "# from nltk.stem.porter import PorterStemmer\n",
        "\n",
        "# def stem(text):\n",
        "#   porter_stemmer = PorterStemmer()\n",
        "#   words = [porter_stemmer.stem(word) for word in text.split()]\n",
        "#   return \" \".join(words)"
      ],
      "id": "NZLux8WqGOZh"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "4667a914"
      },
      "outputs": [],
      "source": [
        "def clean_text(text):\n",
        "    text = text.lower()\n",
        "    text = re.sub(r\"what's\", \"what is \", text)\n",
        "    text = re.sub(r\"\\'s\", \" \", text)\n",
        "    text = re.sub(r\"\\'ve\", \" have \", text)\n",
        "    text = re.sub(r\"can't\", \"cannot \", text)\n",
        "    text = re.sub(r\"n't\", \" not \", text)\n",
        "    text = re.sub(r\"i'm\", \"i am \", text)\n",
        "    text = re.sub(r\"\\'re\", \" are \", text)\n",
        "    text = re.sub(r\"\\'d\", \" would \", text)\n",
        "    text = re.sub(r\"\\'ll\", \" will \", text)\n",
        "    text = re.sub(r\"\\'scuse\", \" excuse \", text)\n",
        "    text = re.sub('\\W', ' ', text)\n",
        "    text = re.sub('\\s+', ' ', text)\n",
        "    text = text.strip(' ')\n",
        "    return text\n",
        "#https://www.kaggle.com/gtskyler/toxic-comments-bert"
      ],
      "id": "4667a914"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "rEiFxKy2IEcz"
      },
      "outputs": [],
      "source": [
        "# train_data['comment_texxxt'] = train_data['comment_text'].map(lambda text: clean_text(text)).map(lambda text: remove_punc(text)).map(lambda text: remove_stop_word(text))\n",
        "# test_data['comment_texxxt'] = test_data['comment_text'].map(lambda text: clean_text(text)).map(lambda text: remove_punc(text)).map(lambda text: remove_stop_word(text))\n",
        "train_data['comment_texxxt'] = train_data['comment_text'].map(lambda text: clean_text(text)).map(lambda text: remove_punc(text))\n",
        "test_data['comment_texxxt'] = test_data['comment_text'].map(lambda text: clean_text(text)).map(lambda text: remove_punc(text))"
      ],
      "id": "rEiFxKy2IEcz"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 641
        },
        "id": "b812599c",
        "outputId": "845f7a5e-010e-4f97-ef0d-e61bcf5ab958"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-6ae852b3-c39e-4c8e-8c63-68bb7d7a25ad\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>id</th>\n",
              "      <th>comment_text</th>\n",
              "      <th>toxic</th>\n",
              "      <th>severe_toxic</th>\n",
              "      <th>obscene</th>\n",
              "      <th>threat</th>\n",
              "      <th>insult</th>\n",
              "      <th>identity_hate</th>\n",
              "      <th>comment_texxxt</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0000997932d777bf</td>\n",
              "      <td>Explanation\\nWhy the edits made under my usern...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>explanation why the edits made under my userna...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>000103f0d9cfb60f</td>\n",
              "      <td>D'aww! He matches this background colour I'm s...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>d aww he matches this background colour i am s...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>000113f07ec002fd</td>\n",
              "      <td>Hey man, I'm really not trying to edit war. It...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>hey man i am really not trying to edit war it ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0001b41b1c6bb37e</td>\n",
              "      <td>\"\\nMore\\nI can't make any real suggestions on ...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>more i cannot make any real suggestions on imp...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0001d958c54c6e35</td>\n",
              "      <td>You, sir, are my hero. Any chance you remember...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>you sir are my hero any chance you remember wh...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>159566</th>\n",
              "      <td>ffe987279560d7ff</td>\n",
              "      <td>\":::::And for the second time of asking, when ...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>and for the second time of asking when your vi...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>159567</th>\n",
              "      <td>ffea4adeee384e90</td>\n",
              "      <td>You should be ashamed of yourself \\n\\nThat is ...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>you should be ashamed of yourself that is a ho...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>159568</th>\n",
              "      <td>ffee36eab5c267c9</td>\n",
              "      <td>Spitzer \\n\\nUmm, theres no actual article for ...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>spitzer umm theres no actual article for prost...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>159569</th>\n",
              "      <td>fff125370e4aaaf3</td>\n",
              "      <td>And it looks like it was actually you who put ...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>and it looks like it was actually you who put ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>159570</th>\n",
              "      <td>fff46fc426af1f9a</td>\n",
              "      <td>\"\\nAnd ... I really don't think you understand...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>and i really do not think you understand i cam...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>159571 rows × 9 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-6ae852b3-c39e-4c8e-8c63-68bb7d7a25ad')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-6ae852b3-c39e-4c8e-8c63-68bb7d7a25ad button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-6ae852b3-c39e-4c8e-8c63-68bb7d7a25ad');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "                      id  ...                                     comment_texxxt\n",
              "0       0000997932d777bf  ...  explanation why the edits made under my userna...\n",
              "1       000103f0d9cfb60f  ...  d aww he matches this background colour i am s...\n",
              "2       000113f07ec002fd  ...  hey man i am really not trying to edit war it ...\n",
              "3       0001b41b1c6bb37e  ...  more i cannot make any real suggestions on imp...\n",
              "4       0001d958c54c6e35  ...  you sir are my hero any chance you remember wh...\n",
              "...                  ...  ...                                                ...\n",
              "159566  ffe987279560d7ff  ...  and for the second time of asking when your vi...\n",
              "159567  ffea4adeee384e90  ...  you should be ashamed of yourself that is a ho...\n",
              "159568  ffee36eab5c267c9  ...  spitzer umm theres no actual article for prost...\n",
              "159569  fff125370e4aaaf3  ...  and it looks like it was actually you who put ...\n",
              "159570  fff46fc426af1f9a  ...  and i really do not think you understand i cam...\n",
              "\n",
              "[159571 rows x 9 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 96
        }
      ],
      "source": [
        "train_data"
      ],
      "id": "b812599c"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "id": "dHgdOG88XCSG",
        "outputId": "f65f42da-7f52-42a7-d24f-a20dc2b319b7"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-af985f20-418f-4b59-81a9-891ec8946512\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>id</th>\n",
              "      <th>comment_text</th>\n",
              "      <th>comment_texxxt</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>00001cee341fdb12</td>\n",
              "      <td>Yo bitch Ja Rule is more succesful then you'll...</td>\n",
              "      <td>yo bitch ja rule is more succesful then you wi...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0000247867823ef7</td>\n",
              "      <td>== From RfC == \\n\\n The title is fine as it is...</td>\n",
              "      <td>from rfc the title is fine as it is imo</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>00013b17ad220c46</td>\n",
              "      <td>\" \\n\\n == Sources == \\n\\n * Zawe Ashton on Lap...</td>\n",
              "      <td>sources zawe ashton on lapland</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>00017563c3f7919a</td>\n",
              "      <td>:If you have a look back at the source, the in...</td>\n",
              "      <td>if you have a look back at the source the info...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>00017695ad8997eb</td>\n",
              "      <td>I don't anonymously edit articles at all.</td>\n",
              "      <td>i do not anonymously edit articles at all</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>153159</th>\n",
              "      <td>fffcd0960ee309b5</td>\n",
              "      <td>. \\n i totally agree, this stuff is nothing bu...</td>\n",
              "      <td>i totally agree this stuff is nothing but too ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>153160</th>\n",
              "      <td>fffd7a9a6eb32c16</td>\n",
              "      <td>== Throw from out field to home plate. == \\n\\n...</td>\n",
              "      <td>throw from out field to home plate does it get...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>153161</th>\n",
              "      <td>fffda9e8d6fafa9e</td>\n",
              "      <td>\" \\n\\n == Okinotorishima categories == \\n\\n I ...</td>\n",
              "      <td>okinotorishima categories i see your changes a...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>153162</th>\n",
              "      <td>fffe8f1340a79fc2</td>\n",
              "      <td>\" \\n\\n == \"\"One of the founding nations of the...</td>\n",
              "      <td>one of the founding nations of the eu germany ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>153163</th>\n",
              "      <td>ffffce3fb183ee80</td>\n",
              "      <td>\" \\n :::Stop already. Your bullshit is not wel...</td>\n",
              "      <td>stop already your bullshit is not welcome here...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>153164 rows × 3 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-af985f20-418f-4b59-81a9-891ec8946512')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-af985f20-418f-4b59-81a9-891ec8946512 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-af985f20-418f-4b59-81a9-891ec8946512');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "                      id  ...                                     comment_texxxt\n",
              "0       00001cee341fdb12  ...  yo bitch ja rule is more succesful then you wi...\n",
              "1       0000247867823ef7  ...            from rfc the title is fine as it is imo\n",
              "2       00013b17ad220c46  ...                     sources zawe ashton on lapland\n",
              "3       00017563c3f7919a  ...  if you have a look back at the source the info...\n",
              "4       00017695ad8997eb  ...          i do not anonymously edit articles at all\n",
              "...                  ...  ...                                                ...\n",
              "153159  fffcd0960ee309b5  ...  i totally agree this stuff is nothing but too ...\n",
              "153160  fffd7a9a6eb32c16  ...  throw from out field to home plate does it get...\n",
              "153161  fffda9e8d6fafa9e  ...  okinotorishima categories i see your changes a...\n",
              "153162  fffe8f1340a79fc2  ...  one of the founding nations of the eu germany ...\n",
              "153163  ffffce3fb183ee80  ...  stop already your bullshit is not welcome here...\n",
              "\n",
              "[153164 rows x 3 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 97
        }
      ],
      "source": [
        "test_data"
      ],
      "id": "dHgdOG88XCSG"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "c7ebae5e"
      },
      "outputs": [],
      "source": [
        "from sklearn.feature_extraction.text import CountVectorizer\n",
        "from keras.preprocessing.text import Tokenizer\n",
        "from keras.preprocessing.sequence import pad_sequences\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Dense, Embedding, LSTM, SpatialDropout1D\n",
        "from sklearn.model_selection import train_test_split\n",
        "from keras.utils.np_utils import to_categorical"
      ],
      "id": "c7ebae5e"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "ad0242ca"
      },
      "outputs": [],
      "source": [
        "# train_values_x = train_data['comment_texxt'].values\n",
        "train_values_x = train_data['comment_texxxt'].values\n",
        "train_values_y = train_data[[\"toxic\",\"severe_toxic\",\"obscene\",\"threat\",\"insult\",\"identity_hate\"]].values"
      ],
      "id": "ad0242ca"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "ca9d0e91"
      },
      "outputs": [],
      "source": [
        "# Split training, validation set\n",
        "train_x, val_x, train_y, val_y = train_test_split(train_values_x, train_values_y, test_size=0.2, random_state=30)"
      ],
      "id": "ca9d0e91"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "42b018bf",
        "outputId": "a201f7eb-3779-4773-9c3c-9de2be4f9b09"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(127656,)\n",
            "(127656, 6)\n",
            "(31915,)\n",
            "(31915, 6)\n"
          ]
        }
      ],
      "source": [
        "print(train_x.shape)\n",
        "print(train_y.shape)\n",
        "print(val_x.shape)\n",
        "print(val_y.shape)"
      ],
      "id": "42b018bf"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "899411b7"
      },
      "outputs": [],
      "source": [
        "# Tokenize training data\n",
        "vocab_len = 100000\n",
        "tokenizer = Tokenizer(num_words = vocab_len, oov_token='<oov>')\n",
        "# tokenizer.fit_on_texts(train_data['comment_texxt'])\n",
        "tokenizer.fit_on_texts(train_data['comment_texxxt'])"
      ],
      "id": "899411b7"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "b15aa307"
      },
      "outputs": [],
      "source": [
        "# Encode training data sentences into sequences\n",
        "train_seq = tokenizer.texts_to_sequences(train_x)"
      ],
      "id": "b15aa307"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "c4b0cd08",
        "outputId": "36f8bf60-3765-436e-9bf5-64c2f6fa298f"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1403\n"
          ]
        }
      ],
      "source": [
        "# Get max training sequence length\n",
        "maxlen = max([len(x) for x in train_seq])\n",
        "print(maxlen)"
      ],
      "id": "c4b0cd08"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "5ec51db1"
      },
      "outputs": [],
      "source": [
        "# Pad the training sequences\n",
        "train_pad = pad_sequences(train_seq, maxlen = maxlen,\n",
        "                          padding = 'pre',\n",
        "                          truncating='pre')"
      ],
      "id": "5ec51db1"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "595f365b",
        "outputId": "363c4809-9335-4ecd-e192-a8160dfdd2c4"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "Padded training shape: (127656, 1403)\n",
            "Training sequences data type: <class 'list'>\n",
            "Padded Training sequences data type: <class 'numpy.ndarray'>\n"
          ]
        }
      ],
      "source": [
        "# print(\"\\nTraining sequences:\\n\", train_seq)\n",
        "# print(\"\\nPadded training sequences:\\n\", train_pad)\n",
        "print(\"\\nPadded training shape:\", train_pad.shape)\n",
        "print(\"Training sequences data type:\", type(train_seq))\n",
        "print(\"Padded Training sequences data type:\", type(train_pad))"
      ],
      "id": "595f365b"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "e4cf2eb8"
      },
      "outputs": [],
      "source": [
        "val_seq = tokenizer.texts_to_sequences(val_x)\n",
        "val_pad = pad_sequences(val_seq, maxlen = maxlen,\n",
        "                        padding = 'pre',\n",
        "                        truncating='pre')"
      ],
      "id": "e4cf2eb8"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "78e7230c",
        "outputId": "7a9d599d-7e5c-421f-bd8e-7558d3363f65"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "190214"
            ]
          },
          "metadata": {},
          "execution_count": 108
        }
      ],
      "source": [
        "vocab_size = len(tokenizer.word_index) + 1\n",
        "vocab_size"
      ],
      "id": "78e7230c"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "DgOInLsc8iRd"
      },
      "outputs": [],
      "source": [
        "embed_dim = 300"
      ],
      "id": "DgOInLsc8iRd"
    },
    {
      "cell_type": "code",
      "source": [
        "!unzip drive/MyDrive/crawl-300d-2M.vec.zip"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-AUik-66AH_3",
        "outputId": "1e44e6d7-10ae-432c-e9c9-f69399121161"
      },
      "id": "-AUik-66AH_3",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Archive:  drive/MyDrive/crawl-300d-2M.vec.zip\n",
            "  inflating: crawl-300d-2M.vec       \n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import io\n",
        "from tqdm.auto import tqdm\n",
        "\n",
        "print('loading word embeddings...')\n",
        "embeddings_index = {}\n",
        "f = io.open('crawl-300d-2M.vec', encoding='utf-8')\n",
        "for line in tqdm(f):\n",
        "    values = line.rstrip().rsplit(' ')\n",
        "    word = values[0]\n",
        "    coefs = np.asarray(values[1:], dtype='float32')\n",
        "    embeddings_index[word] = coefs\n",
        "f.close()\n",
        "print('found %s word vectors' % len(embeddings_index))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 84,
          "referenced_widgets": [
            "85c12b9a6bdd41829899f91e05069cdc",
            "3796170008e24d5891684ccbd23827a8",
            "448f4bd8756c497885b0cfd58b187f9b",
            "718b75d4942e43889b89f3d45ce6bc9e",
            "85295bb8e66443f2a32c0ea622af6320",
            "303495155f8d41c4b572eb24ae10351e",
            "37257c608a424d6d8096a7dd0229af52",
            "aee4d3229b984fa8be8069cc092783e9",
            "da1e147fd6dc447c85daabda97fc1e7e",
            "e8d40fd4c1b3441c92cfb7f56006caff",
            "bdf90828aaa14447a4d341d73533b8b1"
          ]
        },
        "id": "idvSV39IACBH",
        "outputId": "7fd159f0-0862-4692-b323-589157171874"
      },
      "id": "idvSV39IACBH",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "loading word embeddings...\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "85c12b9a6bdd41829899f91e05069cdc",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "0it [00:00, ?it/s]"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "found 1999996 word vectors\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Prepare embedding matrix\n",
        "print('preparing embedding matrix...')\n",
        "words_not_found = []\n",
        "nb_words = vocab_size\n",
        "fasttext_matrix = np.zeros((nb_words, 300))\n",
        "\n",
        "for word, i in tokenizer.word_index.items():\n",
        "    if i >= nb_words:\n",
        "        continue\n",
        "    embedding_vector = embeddings_index.get(word)\n",
        "    if (embedding_vector is not None) and len(embedding_vector) > 0:\n",
        "        # words not found in embedding index will be all-zeros.\n",
        "        fasttext_matrix[i] = embedding_vector\n",
        "    else:\n",
        "        words_not_found.append(word)\n",
        "        \n",
        "print('number of null word embeddings: %d' % np.sum(np.sum(fasttext_matrix, axis=1) == 0))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "R7EG69YQACrX",
        "outputId": "e19244dd-48f8-4e04-c695-d346ff6e7d06"
      },
      "id": "R7EG69YQACrX",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "preparing embedding matrix...\n",
            "number of null word embeddings: 87561\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "iiNrQ_qR8wjS"
      },
      "outputs": [],
      "source": [
        "# !wget --no-check-certificate \\http://nlp.stanford.edu/data/glove.6B.zip \\ -O /tmp/glove.6B.zip"
      ],
      "id": "iiNrQ_qR8wjS"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "o2YaWOM-Cw4e"
      },
      "outputs": [],
      "source": [
        "# !unzip drive/MyDrive/glove.6B.zip"
      ],
      "id": "o2YaWOM-Cw4e"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "r1aIRvAlDCSY",
        "outputId": "d4bee53a-e65d-423b-8c47-0aff4f5e5db9"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Found 400000 word vectors.\n"
          ]
        }
      ],
      "source": [
        "import numpy as np\n",
        "embeddings_index = {}\n",
        "f = open('glove.6B.300d.txt')\n",
        "for line in f:\n",
        "    values = line.split()\n",
        "    word = values[0]\n",
        "    coefs = np.asarray(values[1:], dtype='float32')\n",
        "    embeddings_index[word] = coefs\n",
        "f.close()\n",
        "\n",
        "print('Found %s word vectors.' % len(embeddings_index))"
      ],
      "id": "r1aIRvAlDCSY"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "MipN255pFyYZ"
      },
      "outputs": [],
      "source": [
        "word_index = tokenizer.word_index"
      ],
      "id": "MipN255pFyYZ"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "hnLF4aGyDpa2"
      },
      "outputs": [],
      "source": [
        "glove_matrix = np.zeros((len(word_index) + 1, embed_dim))\n",
        "for word, i in word_index.items():\n",
        "    embedding_vector = embeddings_index.get(word)\n",
        "    if embedding_vector is not None:\n",
        "        # words not found in embedding index will be all-zeros.\n",
        "        glove_matrix[i] = embedding_vector"
      ],
      "id": "hnLF4aGyDpa2"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "9LvuYHyPu2uJ"
      },
      "outputs": [],
      "source": [
        "def plot_learning_curve(history):\n",
        "  plt.plot(history.history['loss'], label='training')\n",
        "  plt.plot(history.history['val_loss'], label='validation')\n",
        "  plt.legend()\n",
        "  plt.show()"
      ],
      "id": "9LvuYHyPu2uJ"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "StGIHX7TXynD"
      },
      "outputs": [],
      "source": [
        "def CNN_model(embed_dim, kernel_num=32, kernel_sizes=[1], embed_weights=None, drop_out_rate=0.0):\n",
        "  # Input\n",
        "  input = tf.keras.Input(shape=(maxlen,))\n",
        " \n",
        "  # Embedding layers\n",
        "  if embed_weights is not None:\n",
        "    embedding = tf.keras.layers.Embedding(\n",
        "          input_dim=vocab_size,\n",
        "          output_dim=embed_dim, \n",
        "          input_length=maxlen,\n",
        "          # weights=[embed_weights], \n",
        "          embeddings_initializer=tf.keras.initializers.Constant(embed_weights),\n",
        "          trainable=False,\n",
        "          name='pretrained_embedding'\n",
        "    )\n",
        "  else:\n",
        "    embedding = tf.keras.layers.Embedding(\n",
        "          input_dim=vocab_size,\n",
        "          output_dim=embed_dim, \n",
        "          input_length=maxlen, \n",
        "          trainable=True,\n",
        "          name='trainable_embedding'\n",
        "    )\n",
        "\n",
        "  embed_out = tf.keras.layers.Reshape((maxlen, embed_dim, 1))(embedding(input))\n",
        "\n",
        "  # Convolutional layers (seperate convs for each word groups)\n",
        "  conv_list = [tf.keras.layers.Conv2D(\n",
        "      filters=kernel_num, \n",
        "      kernel_size=(k, embed_dim),\n",
        "      activation='relu',\n",
        "      padding='valid',\n",
        "      name=f'conv_group_{k}_word') for k in kernel_sizes]\n",
        "    \n",
        "  conv_outs = [conv(embed_out) for conv in conv_list]\n",
        " \n",
        "  # Max pooling layer (one for each convolutional layer)\n",
        "  max_pool_list = [tf.keras.layers.MaxPool1D(\n",
        "      pool_size = maxlen,\n",
        "      padding='same',\n",
        "      name=f'max_pool_group_{k}_word') for k in kernel_sizes]\n",
        "\n",
        "  max_pool_outs = []\n",
        "\n",
        "  for i in range(len(kernel_sizes)):\n",
        "    max_pool_outs.append(max_pool_list[i](tf.squeeze(conv_outs[i], axis=2)))\n",
        " \n",
        "  # Concatenate layer\n",
        "  concate = tf.keras.layers.Concatenate(axis=1, name='concate')(max_pool_outs)\n",
        "\n",
        "  # Flatten layer\n",
        "  flat_out = tf.keras.layers.Flatten(name='flatten')(concate)\n",
        "\n",
        "  # Dropout\n",
        "  drop_out = tf.keras.layers.Dropout(drop_out_rate, name='dropout')(flat_out)\n",
        " \n",
        "  # Output\n",
        "  output = tf.keras.layers.Dense(6, name='output')(drop_out)\n",
        " \n",
        "  # CNN model\n",
        "  return tf.keras.Model(inputs=[input], outputs=[output], name='cnn')"
      ],
      "id": "StGIHX7TXynD"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 680
        },
        "id": "w1ZZ79mwbgYz",
        "outputId": "62919c4a-1ddd-4f72-d606-fd97aa30381e"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "execution_count": 141
        }
      ],
      "source": [
        "cnn = CNN_model(embed_dim=embed_dim, embed_weights=glove_matrix, kernel_sizes=[3, 4, 5], drop_out_rate=0.2)\n",
        "tf.keras.utils.plot_model(cnn, show_shapes=True)"
      ],
      "id": "w1ZZ79mwbgYz"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YJ4m296sbjbb",
        "outputId": "65477fb5-f19a-40f6-ee33-5c5e2edeb4e6"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"cnn\"\n",
            "__________________________________________________________________________________________________\n",
            " Layer (type)                   Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            " input_7 (InputLayer)           [(None, 1403)]       0           []                               \n",
            "                                                                                                  \n",
            " pretrained_embedding (Embeddin  (None, 1403, 300)   57064200    ['input_7[0][0]']                \n",
            " g)                                                                                               \n",
            "                                                                                                  \n",
            " reshape_6 (Reshape)            (None, 1403, 300, 1  0           ['pretrained_embedding[0][0]']   \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv_group_3_word (Conv2D)     (None, 1401, 1, 32)  28832       ['reshape_6[0][0]']              \n",
            "                                                                                                  \n",
            " conv_group_4_word (Conv2D)     (None, 1400, 1, 32)  38432       ['reshape_6[0][0]']              \n",
            "                                                                                                  \n",
            " conv_group_5_word (Conv2D)     (None, 1399, 1, 32)  48032       ['reshape_6[0][0]']              \n",
            "                                                                                                  \n",
            " tf.compat.v1.squeeze_18 (TFOpL  (None, 1401, 32)    0           ['conv_group_3_word[0][0]']      \n",
            " ambda)                                                                                           \n",
            "                                                                                                  \n",
            " tf.compat.v1.squeeze_19 (TFOpL  (None, 1400, 32)    0           ['conv_group_4_word[0][0]']      \n",
            " ambda)                                                                                           \n",
            "                                                                                                  \n",
            " tf.compat.v1.squeeze_20 (TFOpL  (None, 1399, 32)    0           ['conv_group_5_word[0][0]']      \n",
            " ambda)                                                                                           \n",
            "                                                                                                  \n",
            " max_pool_group_3_word (MaxPool  (None, 1, 32)       0           ['tf.compat.v1.squeeze_18[0][0]']\n",
            " ing1D)                                                                                           \n",
            "                                                                                                  \n",
            " max_pool_group_4_word (MaxPool  (None, 1, 32)       0           ['tf.compat.v1.squeeze_19[0][0]']\n",
            " ing1D)                                                                                           \n",
            "                                                                                                  \n",
            " max_pool_group_5_word (MaxPool  (None, 1, 32)       0           ['tf.compat.v1.squeeze_20[0][0]']\n",
            " ing1D)                                                                                           \n",
            "                                                                                                  \n",
            " concate (Concatenate)          (None, 3, 32)        0           ['max_pool_group_3_word[0][0]',  \n",
            "                                                                  'max_pool_group_4_word[0][0]',  \n",
            "                                                                  'max_pool_group_5_word[0][0]']  \n",
            "                                                                                                  \n",
            " flatten (Flatten)              (None, 96)           0           ['concate[0][0]']                \n",
            "                                                                                                  \n",
            " dropout (Dropout)              (None, 96)           0           ['flatten[0][0]']                \n",
            "                                                                                                  \n",
            " output (Dense)                 (None, 6)            582         ['dropout[0][0]']                \n",
            "                                                                                                  \n",
            "==================================================================================================\n",
            "Total params: 57,180,078\n",
            "Trainable params: 115,878\n",
            "Non-trainable params: 57,064,200\n",
            "__________________________________________________________________________________________________\n"
          ]
        }
      ],
      "source": [
        "cnn.summary()"
      ],
      "id": "YJ4m296sbjbb"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Iq7RweY3Wqd6",
        "outputId": "04f649ca-1ec1-4f74-f1d6-ba893fcd872c"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/5\n",
            "3990/3990 [==============================] - 359s 90ms/step - loss: 0.0657 - val_loss: 0.0542\n",
            "Epoch 2/5\n",
            "3990/3990 [==============================] - 357s 89ms/step - loss: 0.0511 - val_loss: 0.0490\n",
            "Epoch 3/5\n",
            "3990/3990 [==============================] - 358s 90ms/step - loss: 0.0473 - val_loss: 0.0489\n",
            "Epoch 4/5\n",
            "3990/3990 [==============================] - 358s 90ms/step - loss: 0.0441 - val_loss: 0.0497\n",
            "Epoch 00004: early stopping\n"
          ]
        }
      ],
      "source": [
        "early_stopping = tf.keras.callbacks.EarlyStopping(\n",
        "    monitor='val_loss', \n",
        "    patience=1,\n",
        "    mode='min',\n",
        "    verbose=1\n",
        ")\n",
        "\n",
        "cnn.compile(\n",
        "    loss = tf.keras.losses.BinaryCrossentropy(from_logits=True), \n",
        "    optimizer = tf.keras.optimizers.Adam(learning_rate=0.0005)\n",
        ")\n",
        "\n",
        "history_cnn = cnn.fit(\n",
        "    train_pad,\n",
        "    train_y,\n",
        "    epochs = 5,\n",
        "    validation_data=(val_pad, val_y), \n",
        "    batch_size = 32,\n",
        "    callbacks=[early_stopping] \n",
        ")"
      ],
      "id": "Iq7RweY3Wqd6"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "9BxTrtVHt3H7",
        "outputId": "5ce744c7-dc72-4cb8-a635-9661f2f69114"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "plot_learning_curve(history_cnn)"
      ],
      "id": "9BxTrtVHt3H7"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "NGl-UfrDS-uE"
      },
      "outputs": [],
      "source": [
        "test_seq = tokenizer.texts_to_sequences(test_data.comment_texxxt)\n",
        "test_pad = pad_sequences(test_seq, maxlen = maxlen,\n",
        "                                padding = 'pre',\n",
        "                                truncating='pre')"
      ],
      "id": "NGl-UfrDS-uE"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "fLjcTVN3efDB"
      },
      "outputs": [],
      "source": [
        "def prepare_submit_file(model, test_pad, batch_size=200):\n",
        "  predicted = tf.math.sigmoid(model.predict(test_pad, batch_size = batch_size))\n",
        "  predict = np.hstack((test_data.id[:, np.newaxis], predicted))\n",
        "\n",
        "  subm = pd.DataFrame(predict, columns = ['id', 'toxic', 'severe_toxic', 'obscene', 'threat', 'insult', 'identity_hate'])\n",
        "  subm.to_csv('subm.csv', index = False)\n",
        "\n",
        "  print(subm)"
      ],
      "id": "fLjcTVN3efDB"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "M75XXyGf568F",
        "outputId": "5f23935d-b6b2-46d3-d50d-41182a4c760c"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:3: FutureWarning: Support for multi-dimensional indexing (e.g. `obj[:, None]`) is deprecated and will be removed in a future version.  Convert to a numpy array before indexing instead.\n",
            "  This is separate from the ipykernel package so we can avoid doing imports until\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "                      id        toxic  ...       insult identity_hate\n",
            "0       00001cee341fdb12     0.999838  ...     0.988103      0.752976\n",
            "1       0000247867823ef7   0.00424482  ...   0.00196032   0.000578403\n",
            "2       00013b17ad220c46    0.0123622  ...   0.00803353    0.00338043\n",
            "3       00017563c3f7919a   0.00273968  ...   0.00140995   0.000250498\n",
            "4       00017695ad8997eb    0.0348347  ...    0.0104682    0.00171533\n",
            "...                  ...          ...  ...          ...           ...\n",
            "153159  fffcd0960ee309b5     0.508009  ...    0.0204929    0.00339026\n",
            "153160  fffd7a9a6eb32c16    0.0413673  ...    0.0091051     0.0015539\n",
            "153161  fffda9e8d6fafa9e  0.000264021  ...  0.000198655   8.71405e-05\n",
            "153162  fffe8f1340a79fc2   0.00460601  ...   0.00157035    0.00848836\n",
            "153163  ffffce3fb183ee80      0.98798  ...     0.342285     0.0337524\n",
            "\n",
            "[153164 rows x 7 columns]\n"
          ]
        }
      ],
      "source": [
        "prepare_submit_file(cnn, test_pad, batch_size=200)"
      ],
      "id": "M75XXyGf568F"
    },
    {
      "cell_type": "code",
      "source": [
        "# Hyperparameter tuning\n",
        "pip install keras-tuner --upgrade\n",
        "import kerastuner as kt"
      ],
      "metadata": {
        "id": "jZDvyiCjaoQ-"
      },
      "id": "jZDvyiCjaoQ-",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def build_cnn_hyper_model(hp):\n",
        "  # Input\n",
        "  input = tf.keras.Input(shape=(maxlen,), name='input')\n",
        " \n",
        "  # Embedding layers\n",
        "  embedding = tf.keras.layers.Embedding(\n",
        "          input_dim=vocab_size,\n",
        "          output_dim=embed_dim, \n",
        "          input_length=maxlen,\n",
        "          weights=[fasttext_matrix], \n",
        "          trainable=False,\n",
        "          name='pretrained_embedding'\n",
        "  )\n",
        "\n",
        "  embed_out = tf.keras.layers.Reshape((maxlen, embed_dim, 1), name='reshape')(embedding(input))\n",
        "\n",
        "  # Convolution + Max pooling\n",
        "  conv_list = []\n",
        "  max_pool_list = []\n",
        "  max_pool_outs = []\n",
        "  # min_height = hp.Int('min_kernel_height', 2, 3)\n",
        "  kernel_num = hp.Int('kernel_num', 32, 128, step=32)\n",
        "  conv_num = hp.Int('conv_num', 1, 3)\n",
        "\n",
        "  for i in range(conv_num):  \n",
        "    kernel_height = 3 + i\n",
        "\n",
        "    # Convolutional layers (seperate convs for each word groups)\n",
        "    conv = tf.keras.layers.Conv2D(\n",
        "        filters=kernel_num,\n",
        "        kernel_size=(kernel_height, embed_dim),\n",
        "        activation='relu',\n",
        "        padding='valid',\n",
        "        name=f'conv_group_{kernel_height}'\n",
        "    )\n",
        "\n",
        "    conv_list.append(conv)\n",
        "\n",
        "    # Max pooling layers (one for each conv layer)\n",
        "    max_pool = tf.keras.layers.MaxPool1D(\n",
        "        pool_size = maxlen,\n",
        "        padding='same',\n",
        "        name=f'max_pooling_group_{kernel_height}')\n",
        "    \n",
        "    max_pool_list.append(max_pool)\n",
        "\n",
        "  conv_outs = [conv(embed_out) for conv in conv_list]\n",
        "\n",
        "  for i in range(len(conv_outs)):\n",
        "     max_pool_outs.append(max_pool_list[i](tf.squeeze(conv_outs[i], axis=2)))\n",
        " \n",
        "  # Concatenate layer\n",
        "  concate = tf.keras.layers.Concatenate(axis=1, name='concate')(max_pool_outs)\n",
        "\n",
        "  # Flatten layer\n",
        "  flat_out = tf.keras.layers.Flatten(name='flatten')(concate)\n",
        "\n",
        "  # Dropout\n",
        "  drop_out = tf.keras.layers.Dropout(hp.Float('dropout', 0.0, 0.5, step=0.1), name='dropout')(flat_out)\n",
        " \n",
        "  # Output\n",
        "  output = tf.keras.layers.Dense(6, name='output')(drop_out)\n",
        " \n",
        "  # CNN model\n",
        "  cnn_model = tf.keras.Model(inputs=[input], outputs=[output], name='cnn')\n",
        "\n",
        "  # Model compling\n",
        "  cnn_model.compile(\n",
        "      loss = tf.keras.losses.BinaryCrossentropy(from_logits=True), \n",
        "      optimizer = tf.keras.optimizers.Adam(learning_rate=0.0005)\n",
        "  )\n",
        "\n",
        "  return cnn_model"
      ],
      "metadata": {
        "id": "IZn3nqWua1Se"
      },
      "id": "IZn3nqWua1Se",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "tuner = kt.RandomSearch(\n",
        "    hypermodel=build_cnn_hyper_model,\n",
        "    objective='val_loss',\n",
        "    max_trials=50,\n",
        "    directory='tuning',\n",
        "    overwrite=False,\n",
        "    project_name='cnn_fasttext'\n",
        ")"
      ],
      "metadata": {
        "id": "7mR2FIYZbCe3"
      },
      "id": "7mR2FIYZbCe3",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "tuner.search_space_summary()"
      ],
      "metadata": {
        "id": "Eb2ZyxHDbErG"
      },
      "id": "Eb2ZyxHDbErG",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "early_stopping = tf.keras.callbacks.EarlyStopping(\n",
        "    monitor='val_loss', \n",
        "    patience=1,\n",
        "    mode='min',\n",
        "    verbose=1\n",
        ")\n",
        "\n",
        "tuner.search(\n",
        "    train_pad, \n",
        "    train_y, \n",
        "    epochs=5, \n",
        "    validation_data=(val_pad, val_y),\n",
        "    callbacks=[early_stopping])"
      ],
      "metadata": {
        "id": "KlEIBfE_bIAn"
      },
      "id": "KlEIBfE_bIAn",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "tuner.results_summary()"
      ],
      "metadata": {
        "id": "DQsXu2sobJCG"
      },
      "id": "DQsXu2sobJCG",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "best_model = tuner.get_best_models()[0]"
      ],
      "metadata": {
        "id": "bfGs9BzGbTWm"
      },
      "id": "bfGs9BzGbTWm",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "best_model.summary()"
      ],
      "metadata": {
        "id": "CdZBHzzNbVH9"
      },
      "id": "CdZBHzzNbVH9",
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "collapsed_sections": [],
      "name": "CNN_Toxic_Comment_Classification.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "display_name": "TF2",
      "language": "python",
      "name": "tf2"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.7.10"
    },
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "85c12b9a6bdd41829899f91e05069cdc": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_3796170008e24d5891684ccbd23827a8",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_448f4bd8756c497885b0cfd58b187f9b",
              "IPY_MODEL_718b75d4942e43889b89f3d45ce6bc9e",
              "IPY_MODEL_85295bb8e66443f2a32c0ea622af6320"
            ]
          }
        },
        "3796170008e24d5891684ccbd23827a8": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "448f4bd8756c497885b0cfd58b187f9b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_303495155f8d41c4b572eb24ae10351e",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": "",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_37257c608a424d6d8096a7dd0229af52"
          }
        },
        "718b75d4942e43889b89f3d45ce6bc9e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_aee4d3229b984fa8be8069cc092783e9",
            "_dom_classes": [],
            "description": "",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 1,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 1,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_da1e147fd6dc447c85daabda97fc1e7e"
          }
        },
        "85295bb8e66443f2a32c0ea622af6320": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_e8d40fd4c1b3441c92cfb7f56006caff",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 1999996/? [03:20&lt;00:00, 10523.52it/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_bdf90828aaa14447a4d341d73533b8b1"
          }
        },
        "303495155f8d41c4b572eb24ae10351e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "37257c608a424d6d8096a7dd0229af52": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "aee4d3229b984fa8be8069cc092783e9": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "da1e147fd6dc447c85daabda97fc1e7e": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": "20px",
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "e8d40fd4c1b3441c92cfb7f56006caff": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "bdf90828aaa14447a4d341d73533b8b1": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        }
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 5
}