multilabel_new2_version_cnn_lstm_attention_augOriginal_POS_NER.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Wpz5GhVd9HyQ"
      },
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import tensorflow as tf\n",
        "from tensorflow.keras.layers import Embedding, Conv1D, LSTM, Dense, Attention, GlobalMaxPooling1D, Bidirectional, Input\n",
        "from tensorflow.keras.models import Model\n",
        "from tensorflow.keras.preprocessing.text import Tokenizer\n",
        "from tensorflow.keras.preprocessing.sequence import pad_sequences\n",
        "from sklearn.preprocessing import MultiLabelBinarizer\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import classification_report"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LhOygIDlKjoL",
        "outputId": "b92e8e46-cba6-4b62-f83f-25fb65e089f1"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ],
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "7flw33tqI4sY"
      },
      "outputs": [],
      "source": [
        "\n",
        "# load the tweets dataset\n",
        "df_tweets = pd.read_csv('/content/drive/MyDrive/Colab Notebooks/propaganda/finalAugResults/augOriginalDS/augOriginal.csv')\n",
        "# load the test dataset\n",
        "df_tweets_test = pd.read_csv('/content/drive/MyDrive/Colab Notebooks/propaganda/testDS.csv')\n",
        "# load the validation dataset\n",
        "df_tweets_val = pd.read_csv('/content/drive/MyDrive/Colab Notebooks/propaganda/task2_dev.csv')\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 770
        },
        "id": "n0O5F1w65L8W",
        "outputId": "e178772c-f9c2-447d-e8b1-9d8f547a6f0d"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "      Unnamed: 0.1  Unnamed: 0  start  end                  technique  \\\n",
              "0                0           0      0    0               no technique   \n",
              "1                1           1      1   33  Exaggeration/Minimisation   \n",
              "2                2           2     37   86                     Smears   \n",
              "3                3           3      0    0               no technique   \n",
              "4                4           4     60   72            Loaded Language   \n",
              "...            ...         ...    ...  ...                        ...   \n",
              "5343          5375        6055     19   82   Appeal to fear/prejudice   \n",
              "5344          5376        3655      9   44   Appeal to fear/prejudice   \n",
              "5345          5377        2401      0   73   Appeal to fear/prejudice   \n",
              "5346          5378        6143     19   82   Appeal to fear/prejudice   \n",
              "5347          5379        4717     20   67   Appeal to fear/prejudice   \n",
              "\n",
              "                                                   text  \\\n",
              "0                                                   NaN   \n",
              "1                      ده مش معتقل ده أحسن من اللوكاندة   \n",
              "2     جدل وسخرية من زيارات تنظمها وزارة الداخلية للسجون   \n",
              "3                                                   NaN   \n",
              "4                                          مجزرة حقيقية   \n",
              "...                                                 ...   \n",
              "5343  انتشال ضحايا من عائلة كاملة قصف منزلها في شمال...   \n",
              "5344                حدث خطير في فلسطين يتطلب إرسال قوات   \n",
              "5345  أكثر من ألف وفاة في البرازيل وسط تحذيرات من مو...   \n",
              "5346  انتشال ضحايا من عائلة كاملة قصف منزلها في شمال...   \n",
              "5347    الركود يخنق التنمية والبطالة تغتال أحلام الشباب   \n",
              "\n",
              "                                              full_text                   id  \\\n",
              "0     بي بي سي ترندينغ النساء تثرثر كثيرا رئيس أولمب...  1358824915483435012   \n",
              "1     ده مش معتقل ده أحسن من اللوكاندة جدل وسخرية من...  1389927866356412416   \n",
              "2     ده مش معتقل ده أحسن من اللوكاندة جدل وسخرية من...  1389927866356412416   \n",
              "3                                الرجل الذي كان من فراغ  1364082975428677632   \n",
              "4     عاجل حركة حماس ما يجري في المسجد الأقصى مجزرة ...  1391667689656102914   \n",
              "...                                                 ...                  ...   \n",
              "5343  مراسلة فرانس انتشال ضحايا من عائلة كاملة قصف م...  1393098908218269697   \n",
              "5344        باكستان حدث خطير في فلسطين يتطلب إرسال قوات                 7331   \n",
              "5345  أكثر من ألف وفاة في البرازيل وسط تحذيرات من مو...                 7283   \n",
              "5346  مراسلة فرانس انتشال ضحايا من عائلة كاملة قصف م...  1393098908218269697   \n",
              "5347  الغزالة زاكورة الركود يخنق التنمية والبطالة تغ...  1393742634036785153   \n",
              "\n",
              "      start_new  end_new                                     augmented_text  \\\n",
              "0           0.0      0.0                                                NaN   \n",
              "1           0.0     32.0                   ده مش معتقل ده أحسن من اللوكاندة   \n",
              "2          33.0     82.0  جدل وسخرية من زيارات تنظمها وزارة الداخلية للسجون   \n",
              "3           0.0      0.0                                                NaN   \n",
              "4          40.0     52.0                                       مجزرة حقيقية   \n",
              "...         ...      ...                                                ...   \n",
              "5343       13.0     74.0  انتشال ضحايا من عائلة كاملة قصف منزلها في شمال...   \n",
              "5344        8.0     43.0               حدث خطير في باكستان يتطلب إرسال قوات   \n",
              "5345        0.0     67.0  أكثر من ألف وفاة في أذربيجان وسط تحذيرات من مو...   \n",
              "5346       13.0     74.0  انتشال ضحايا من عائلة كاملة قصف منزلها في شمال...   \n",
              "5347       15.0     62.0    الركود يخنق التنمية والبطالة تغتال ثامرة الشباب   \n",
              "\n",
              "                                    augmented_full_text  \n",
              "0     بي بي سي ترندينغ النساء تثرثر كثيرا رئيس أولمب...  \n",
              "1     ده مش معتقل ده أحسن من اللوكاندة جدل وسخرية من...  \n",
              "2     ده مش معتقل ده أحسن من اللوكاندة جدل وسخرية من...  \n",
              "3                                الرجل الذي كان من فراغ  \n",
              "4     عاجل حركة حماس ما يجري في المسجد الأقصى مجزرة ...  \n",
              "...                                                 ...  \n",
              "5343  مراسلة فرانس انتشال ضحايا من عائلة كاملة قصف م...  \n",
              "5344       باكستان حدث خطير في باكستان يتطلب إرسال قوات  \n",
              "5345  أكثر من ألف وفاة في أذربيجان وسط تحذيرات من مو...  \n",
              "5346  مراسلة فرانس انتشال ضحايا من عائلة كاملة قصف م...  \n",
              "5347  الغزالة زاكورة الركود يخنق التنمية والبطالة تغ...  \n",
              "\n",
              "[5348 rows x 12 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-7a99e4f0-fb82-4cca-acbd-f84a70654a9c\" 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>Unnamed: 0.1</th>\n",
              "      <th>Unnamed: 0</th>\n",
              "      <th>start</th>\n",
              "      <th>end</th>\n",
              "      <th>technique</th>\n",
              "      <th>text</th>\n",
              "      <th>full_text</th>\n",
              "      <th>id</th>\n",
              "      <th>start_new</th>\n",
              "      <th>end_new</th>\n",
              "      <th>augmented_text</th>\n",
              "      <th>augmented_full_text</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>no technique</td>\n",
              "      <td>NaN</td>\n",
              "      <td>بي بي سي ترندينغ النساء تثرثر كثيرا رئيس أولمب...</td>\n",
              "      <td>1358824915483435012</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>بي بي سي ترندينغ النساء تثرثر كثيرا رئيس أولمب...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>33</td>\n",
              "      <td>Exaggeration/Minimisation</td>\n",
              "      <td>ده مش معتقل ده أحسن من اللوكاندة</td>\n",
              "      <td>ده مش معتقل ده أحسن من اللوكاندة جدل وسخرية من...</td>\n",
              "      <td>1389927866356412416</td>\n",
              "      <td>0.0</td>\n",
              "      <td>32.0</td>\n",
              "      <td>ده مش معتقل ده أحسن من اللوكاندة</td>\n",
              "      <td>ده مش معتقل ده أحسن من اللوكاندة جدل وسخرية من...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>37</td>\n",
              "      <td>86</td>\n",
              "      <td>Smears</td>\n",
              "      <td>جدل وسخرية من زيارات تنظمها وزارة الداخلية للسجون</td>\n",
              "      <td>ده مش معتقل ده أحسن من اللوكاندة جدل وسخرية من...</td>\n",
              "      <td>1389927866356412416</td>\n",
              "      <td>33.0</td>\n",
              "      <td>82.0</td>\n",
              "      <td>جدل وسخرية من زيارات تنظمها وزارة الداخلية للسجون</td>\n",
              "      <td>ده مش معتقل ده أحسن من اللوكاندة جدل وسخرية من...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>no technique</td>\n",
              "      <td>NaN</td>\n",
              "      <td>الرجل الذي كان من فراغ</td>\n",
              "      <td>1364082975428677632</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>الرجل الذي كان من فراغ</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>4</td>\n",
              "      <td>60</td>\n",
              "      <td>72</td>\n",
              "      <td>Loaded Language</td>\n",
              "      <td>مجزرة حقيقية</td>\n",
              "      <td>عاجل حركة حماس ما يجري في المسجد الأقصى مجزرة ...</td>\n",
              "      <td>1391667689656102914</td>\n",
              "      <td>40.0</td>\n",
              "      <td>52.0</td>\n",
              "      <td>مجزرة حقيقية</td>\n",
              "      <td>عاجل حركة حماس ما يجري في المسجد الأقصى مجزرة ...</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",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5343</th>\n",
              "      <td>5375</td>\n",
              "      <td>6055</td>\n",
              "      <td>19</td>\n",
              "      <td>82</td>\n",
              "      <td>Appeal to fear/prejudice</td>\n",
              "      <td>انتشال ضحايا من عائلة كاملة قصف منزلها في شمال...</td>\n",
              "      <td>مراسلة فرانس انتشال ضحايا من عائلة كاملة قصف م...</td>\n",
              "      <td>1393098908218269697</td>\n",
              "      <td>13.0</td>\n",
              "      <td>74.0</td>\n",
              "      <td>انتشال ضحايا من عائلة كاملة قصف منزلها في شمال...</td>\n",
              "      <td>مراسلة فرانس انتشال ضحايا من عائلة كاملة قصف م...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5344</th>\n",
              "      <td>5376</td>\n",
              "      <td>3655</td>\n",
              "      <td>9</td>\n",
              "      <td>44</td>\n",
              "      <td>Appeal to fear/prejudice</td>\n",
              "      <td>حدث خطير في فلسطين يتطلب إرسال قوات</td>\n",
              "      <td>باكستان حدث خطير في فلسطين يتطلب إرسال قوات</td>\n",
              "      <td>7331</td>\n",
              "      <td>8.0</td>\n",
              "      <td>43.0</td>\n",
              "      <td>حدث خطير في باكستان يتطلب إرسال قوات</td>\n",
              "      <td>باكستان حدث خطير في باكستان يتطلب إرسال قوات</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5345</th>\n",
              "      <td>5377</td>\n",
              "      <td>2401</td>\n",
              "      <td>0</td>\n",
              "      <td>73</td>\n",
              "      <td>Appeal to fear/prejudice</td>\n",
              "      <td>أكثر من ألف وفاة في البرازيل وسط تحذيرات من مو...</td>\n",
              "      <td>أكثر من ألف وفاة في البرازيل وسط تحذيرات من مو...</td>\n",
              "      <td>7283</td>\n",
              "      <td>0.0</td>\n",
              "      <td>67.0</td>\n",
              "      <td>أكثر من ألف وفاة في أذربيجان وسط تحذيرات من مو...</td>\n",
              "      <td>أكثر من ألف وفاة في أذربيجان وسط تحذيرات من مو...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5346</th>\n",
              "      <td>5378</td>\n",
              "      <td>6143</td>\n",
              "      <td>19</td>\n",
              "      <td>82</td>\n",
              "      <td>Appeal to fear/prejudice</td>\n",
              "      <td>انتشال ضحايا من عائلة كاملة قصف منزلها في شمال...</td>\n",
              "      <td>مراسلة فرانس انتشال ضحايا من عائلة كاملة قصف م...</td>\n",
              "      <td>1393098908218269697</td>\n",
              "      <td>13.0</td>\n",
              "      <td>74.0</td>\n",
              "      <td>انتشال ضحايا من عائلة كاملة قصف منزلها في شمال...</td>\n",
              "      <td>مراسلة فرانس انتشال ضحايا من عائلة كاملة قصف م...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5347</th>\n",
              "      <td>5379</td>\n",
              "      <td>4717</td>\n",
              "      <td>20</td>\n",
              "      <td>67</td>\n",
              "      <td>Appeal to fear/prejudice</td>\n",
              "      <td>الركود يخنق التنمية والبطالة تغتال أحلام الشباب</td>\n",
              "      <td>الغزالة زاكورة الركود يخنق التنمية والبطالة تغ...</td>\n",
              "      <td>1393742634036785153</td>\n",
              "      <td>15.0</td>\n",
              "      <td>62.0</td>\n",
              "      <td>الركود يخنق التنمية والبطالة تغتال ثامرة الشباب</td>\n",
              "      <td>الغزالة زاكورة الركود يخنق التنمية والبطالة تغ...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5348 rows × 12 columns</p>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-7a99e4f0-fb82-4cca-acbd-f84a70654a9c')\"\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 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\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",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\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-7a99e4f0-fb82-4cca-acbd-f84a70654a9c 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-7a99e4f0-fb82-4cca-acbd-f84a70654a9c');\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",
              "\n",
              "\n",
              "<div id=\"df-cab59bc8-c268-41ad-b6bb-aa07541f8886\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-cab59bc8-c268-41ad-b6bb-aa07541f8886')\"\n",
              "            title=\"Suggest charts\"\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",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-cab59bc8-c268-41ad-b6bb-aa07541f8886 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "  <div id=\"id_7277815d-4648-4166-8c56-1c2dc69dc0e2\">\n",
              "    <style>\n",
              "      .colab-df-generate {\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-generate: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-generate {\n",
              "        background-color: #3B4455;\n",
              "        fill: #D2E3FC;\n",
              "      }\n",
              "\n",
              "      [theme=dark] .colab-df-generate: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",
              "    <button class=\"colab-df-generate\" onclick=\"generateWithVariable('df_tweets')\"\n",
              "            title=\"Generate code using this dataframe.\"\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=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "    <script>\n",
              "      (() => {\n",
              "      const buttonEl =\n",
              "        document.querySelector('#id_7277815d-4648-4166-8c56-1c2dc69dc0e2 button.colab-df-generate');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      buttonEl.onclick = () => {\n",
              "        google.colab.notebook.generateWithVariable('df_tweets');\n",
              "      }\n",
              "      })();\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "df_tweets",
              "summary": "{\n  \"name\": \"df_tweets\",\n  \"rows\": 5348,\n  \"fields\": [\n    {\n      \"column\": \"Unnamed: 0.1\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 1556,\n        \"min\": 0,\n        \"max\": 5379,\n        \"num_unique_values\": 5348,\n        \"samples\": [\n          2530,\n          3599,\n          486\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Unnamed: 0\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 6428,\n        \"min\": 0,\n        \"max\": 50088,\n        \"num_unique_values\": 2366,\n        \"samples\": [\n          1086,\n          4141,\n          688\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"start\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 49,\n        \"min\": 0,\n        \"max\": 274,\n        \"num_unique_values\": 145,\n        \"samples\": [\n          17,\n          115,\n          34\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"end\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 57,\n        \"min\": 0,\n        \"max\": 286,\n        \"num_unique_values\": 166,\n        \"samples\": [\n          166,\n          53,\n          187\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"technique\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 18,\n        \"samples\": [\n          \"no technique\",\n          \"Exaggeration/Minimisation\",\n          \"Doubt\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"text\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 849,\n        \"samples\": [\n          \"\\u0627\\u0644\\u0645 \\u0642\\u0627\\u0648\\u0645\\u0629\",\n          \"\\u0627\\u0644\\u0645\\u062c\\u0631\\u0645 \\u0631\\u0641\\u0639\\u062a \\u0627\\u0644\\u0627\\u0633\\u062f \\u064a\\u0634\\u0627\\u0631\\u0643 \\u0641\\u064a \\u0645\\u0633\\u0631\\u062d\\u064a\\u0629 \\u0627\\u0644\\u0627\\u0646\\u062a\\u062e\\u0627\\u0628\\u0627\\u062a \\u0627\\u0644\\u0631\\u0626\\u0627\\u0633\\u064a\\u0629\",\n          \"\\u0645\\u062c\\u0627\\u0632\\u0631 \\u0645\\u0631\\u0648\\u0639\\u0629\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"full_text\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 504,\n        \"samples\": [\n          \"\\u0627\\u0644\\u062e\\u0627\\u0631\\u062c\\u064a\\u0629 \\u0627\\u0644\\u0645\\u0635\\u0631\\u064a\\u0629 \\u0623\\u0643\\u062f\\u0646\\u0627 \\u0644\\u0648\\u0632\\u064a\\u0631 \\u0627\\u0644\\u062e\\u0627\\u0631\\u062c\\u064a\\u0629 \\u0627\\u0644\\u0625\\u0633\\u0631\\u0627\\u0626\\u064a\\u0644\\u064a \\u0623\\u0647\\u0645\\u064a\\u0629 \\u0627\\u062a\\u062e\\u0627\\u0630 \\u062a\\u062f\\u0627\\u0628\\u064a\\u0631 \\u062a\\u0639\\u0632\\u0632 \\u0627\\u0644\\u062a\\u0647\\u062f\\u0626\\u0629 \\u0648\\u062a\\u0645\\u0647\\u062f \\u0644\\u0625\\u062d\\u064a\\u0627\\u0621 \\u0627\\u0644\\u0645\\u0633\\u0627\\u0631 \\u0627\\u0644\\u0633\\u064a\\u0627\\u0633\\u064a\",\n          \"\\u0639\\u064a\\u0646 \\u0627\\u0644\\u064a\\u0645\\u0646 \\u0627\\u0644\\u0633\\u064a\\u0646\\u0627\\u0631\\u064a\\u0648 \\u0627\\u0644\\u0639\\u0631\\u0627\\u0642\\u064a \\u064a\\u0631\\u0639\\u0628 \\u062d\\u0632\\u0628 \\u0627\\u0644\\u0644\\u0647 \\u0641\\u064a \\u0644\\u0628\\u0646\\u0627\\u0646\",\n          \"\\u0634\\u0627\\u0647\\u062f \\u062a\\u0638\\u0627\\u0647\\u0631\\u0629 \\u0623\\u0645\\u0627\\u0645 \\u0645\\u0642\\u0631 \\u0633\\u0641\\u0627\\u0631\\u0629 \\u0627\\u0644\\u0627\\u062d\\u062a\\u0644\\u0627\\u0644 \\u0641\\u064a \\u0644\\u0646\\u062f\\u0646 \\u062a\\u0646\\u062f\\u064a\\u062f\\u0627 \\u0628\\u062c\\u0631\\u0627\\u0626\\u0645 \\u0627\\u0644\\u0627\\u062d\\u062a\\u0644\\u0627\\u0644 \\u0641\\u064a \\u0627\\u0644\\u0642\\u062f\\u0633 \\u0648\\u063a\\u0632\\u0629\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"id\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 688217731914662656,\n        \"min\": 7036,\n        \"max\": 1400444519536017411,\n        \"num_unique_values\": 504,\n        \"samples\": [\n          7322,\n          1192116108712726530,\n          1396079560697974790\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"start_new\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 45.560458681964555,\n        \"min\": 0.0,\n        \"max\": 236.0,\n        \"num_unique_values\": 134,\n        \"samples\": [\n          105.0,\n          41.0,\n          121.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"end_new\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 53.311723074913296,\n        \"min\": 0.0,\n        \"max\": 248.0,\n        \"num_unique_values\": 155,\n        \"samples\": [\n          106.0,\n          6.0,\n          54.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"augmented_text\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 4728,\n        \"samples\": [\n          \"\\u0647\\u0644 \\u064a\\u0646\\u062c\\u062d \\u0628\\u0647\\u0627\\u0621 \\u0627\\u0644\\u062f\\u064a\\u0646 \\u0628\\u0631\\u064a \\u0641\\u064a \\u062a\\u0637\\u0648\\u064a\\u0642 \\u0645\\u0641\\u0627\\u0639\\u064a\\u0644 \\u0627\\u0644\\u0645\\u0648\\u0627\\u062c\\u0647\\u0629 \\u0627\\u0644\\u0633\\u0646\\u064a\\u0629 \\u0627\\u0644\\u0645\\u0633\\u064a\\u062d\\u064a\\u0629 \\u0641\\u064a \\u0645\\u062c\\u0644\\u0633 \\u0627\\u0644\\u0646\\u0648\\u0627\\u0628\",\n          \"\\u0627\\u0646\\u0632\\u0644\\u0642\\u062a \\u0623\\u0648\\u0644 \\u0645\\u0631\\u0643\\u0628\\u0629 \\u0641\\u0636\\u0627\\u0626\\u064a\\u0629 \\u0623\\u0631\\u062f\\u0646\\u064a\\u0629 \\u0639\\u0644\\u0649 \\u0627\\u0644\\u0645\\u0631\\u064a\\u062e \\u0645\\u0646 \\u0645\\u0646\\u0635\\u0629 \\u0647\\u0628\\u0648\\u0637\\u0647\\u0627 \\u0648\\u0647\\u064a \\u062a\\u062a\\u062c\\u0648\\u0644 \\u0627\\u0644\\u0622\\u0646 \\u0639\\u0644\\u0649 \\u0633\\u0637\\u062d \\u0627\\u0644\\u0643\\u0648\\u0643\\u0628 \\u0627\\u0644\\u0623\\u062d\\u0645\\u0631\",\n          \"\\u0627\\u0644\\u0625\\u0645\\u0627\\u0631\\u0627\\u062a \\u062a\\u0647\\u0646\\u0649\\u0621 \\u0631\\u0648\\u0633\\u064a\\u0627 \\u0627\\u0644\\u0628\\u064a\\u0636\\u0627\\u0621 \\u0628\\u0631\\u0645\\u0636\\u0627\\u0646 \\u0648\\u0625\\u0633\\u0631\\u0627\\u0626\\u064a\\u0644 \\u0628\\u0627\\u0644\\u0646\\u0643\\u0628\\u0629\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"augmented_full_text\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 4695,\n        \"samples\": [\n          \"\\u0641\\u0648\\u0642 \\u0627\\u0644\\u0633\\u0644\\u0637\\u0629 \\u0627\\u0644\\u0625\\u0645\\u0627\\u0631\\u0627\\u062a \\u062a\\u0647\\u0646\\u0649\\u0621 \\u0627\\u0644\\u0646\\u0631\\u0648\\u064a\\u062c \\u0628\\u0631\\u0645\\u0636\\u0627\\u0646 \\u0648\\u0625\\u0633\\u0631\\u0627\\u0626\\u064a\\u0644 \\u0628\\u0627\\u0644\\u0646\\u0643\\u0628\\u0629\",\n          \"\\u062a\\u0642\\u0631\\u064a\\u0631 \\u0627\\u0633\\u062a\\u062e\\u0628\\u0627\\u0631\\u0627\\u062a\\u064a \\u0641\\u0644\\u0633\\u0637\\u064a\\u0646\\u064a \\u064a\\u062c\\u062f\\u062f \\u0627\\u0644\\u0646\\u0642\\u0627\\u0634 \\u062d\\u0648\\u0644 \\u0641\\u0631\\u0636\\u064a\\u0629 \\u062a\\u0633\\u0631\\u0628 \\u0641\\u064a\\u0631\\u0648\\u0633 \\u0643\\u0648\\u0631\\u0648\\u0646\\u0627 \\u0645\\u0646 \\u0645\\u0639\\u0645\\u0644 \\u0641\\u064a \\u0645\\u062f\\u064a\\u0646\\u0629 \\u0648\\u0648\\u0647\\u0627\\u0646 \\u0627\\u0644\\u0635\\u064a\\u0646\\u064a\\u0629\",\n          \"\\u0643\\u0644\\u0647\\u0627 \\u0641\\u0644\\u0633\\u0637\\u064a\\u0646 \\u062a\\u0638\\u0627\\u0647\\u0631\\u0629 \\u0636\\u062e\\u0645\\u0629 \\u0628\\u0627\\u0644\\u062f\\u0648\\u062d\\u0629 \\u0646\\u0635\\u0631\\u0629 \\u0644\\u0644\\u0642\\u062f\\u0633 \\u0648\\u063a\\u0632\\u0629 \\u0648\\u062a\\u0646\\u062f\\u064a\\u062f\\u0627 \\u0628\\u0639\\u062f\\u0648\\u0627\\u0646 \\u0627\\u0644\\u0627\\u062d\\u062a\\u0644\\u0627\\u0644 \\u062c\\u062f\\u0629 \\u062a\\u062d\\u062a \\u0627\\u0644\\u0642\\u0635\\u0641\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 4
        }
      ],
      "source": [
        "df_tweets"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 680
        },
        "id": "ardoCmKBIioH",
        "outputId": "dd92b53b-9626-4c72-b9ad-4136f3c8e5d0"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "technique\n",
              "Obfuscation, Intentional vagueness, Confusion    477\n",
              "Exaggeration/Minimisation                        462\n",
              "Loaded Language                                  446\n",
              "Name calling/Labeling                            443\n",
              "Flag-waving                                      421\n",
              "Whataboutism                                     405\n",
              "Appeal to authority                              386\n",
              "Black-and-white Fallacy/Dictatorship             348\n",
              "no technique                                     345\n",
              "Appeal to fear/prejudice                         324\n",
              "Glittering generalities (Virtue)                 284\n",
              "Slogans                                          254\n",
              "Causal Oversimplification                        220\n",
              "Smears                                           170\n",
              "Doubt                                            140\n",
              "Repetition                                        89\n",
              "Thought-terminating cliché                        72\n",
              "Presenting Irrelevant Data (Red Herring)          62\n",
              "Name: count, dtype: int64"
            ],
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>count</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>technique</th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Obfuscation, Intentional vagueness, Confusion</th>\n",
              "      <td>477</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Exaggeration/Minimisation</th>\n",
              "      <td>462</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Loaded Language</th>\n",
              "      <td>446</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Name calling/Labeling</th>\n",
              "      <td>443</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Flag-waving</th>\n",
              "      <td>421</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Whataboutism</th>\n",
              "      <td>405</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Appeal to authority</th>\n",
              "      <td>386</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Black-and-white Fallacy/Dictatorship</th>\n",
              "      <td>348</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>no technique</th>\n",
              "      <td>345</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Appeal to fear/prejudice</th>\n",
              "      <td>324</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Glittering generalities (Virtue)</th>\n",
              "      <td>284</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Slogans</th>\n",
              "      <td>254</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Causal Oversimplification</th>\n",
              "      <td>220</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Smears</th>\n",
              "      <td>170</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Doubt</th>\n",
              "      <td>140</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Repetition</th>\n",
              "      <td>89</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Thought-terminating cliché</th>\n",
              "      <td>72</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Presenting Irrelevant Data (Red Herring)</th>\n",
              "      <td>62</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div><br><label><b>dtype:</b> int64</label>"
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ],
      "source": [
        "df_tweets['technique'].value_counts()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "id": "ZRkp3bIc8uwT",
        "outputId": "53865c4f-0db6-42e7-f5e8-4f15fe9dc55c"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                    augmented_full_text  \\\n",
              "0     (بداية برنامج ترويج الارهابيين يا (شار يستبق ا...   \n",
              "1                   (بشار الأسد) دنس دائماً لينتخب نفسه   \n",
              "2     (ترامب) يتهم رئيس كوريا الجنوبية بنهب الولايات...   \n",
              "3     - الانتصار على القصف الإسرائيلي أمام قبة قاسم ...   \n",
              "4     -الحارس. يقتلون بالسكاكين والمناجل في شرق الكو...   \n",
              "...                                                 ...   \n",
              "4741                        يوم النصر إصرار عنبرة وبناء   \n",
              "4742  ٢٥ أيار عدد المتظاهرات في العراق ترفع شعار من ...   \n",
              "4743  ٢٥ أيار مظاهرات حاشدة في العراق ترفع شعار من ق...   \n",
              "4744  ٢٥ أيار مظاهرات حاشدة في العراق ترفع شعار من ق...   \n",
              "4745  ٢٥ أيار مظاهرات حاشدة في العراق ترفع شعار من ق...   \n",
              "\n",
              "                                              technique  \\\n",
              "0                                              [Smears]   \n",
              "1                                              [Smears]   \n",
              "2                                              [Smears]   \n",
              "3                                              [Smears]   \n",
              "4                               [Name calling/Labeling]   \n",
              "...                                                 ...   \n",
              "4741                                     [no technique]   \n",
              "4742                        [Exaggeration/Minimisation]   \n",
              "4743                                      [Flag-waving]   \n",
              "4744  [Loaded Language, Slogans, Flag-waving, Loaded...   \n",
              "4745                                          [Slogans]   \n",
              "\n",
              "                                         augmented_text  \n",
              "0              [(بداية برنامج ترويج الارهابيين يا (شار]  \n",
              "1                 [(بشار الأسد) دنس دائماً لينتخب نفسه]  \n",
              "2     [(ترامب) يتهم رئيس كوريا الجنوبية بنهب الولايا...  \n",
              "3     [- الانتصار على القصف الإسرائيلي أمام قبة قاسم...  \n",
              "4                                            [-الحارس.]  \n",
              "...                                                 ...  \n",
              "4741                                              [nan]  \n",
              "4742                                   [عدد المتظاهرات]  \n",
              "4743  [المطالبة بمحاسبة المسؤولين عن اغتيال النشطاء ...  \n",
              "4744  [من قتلني, من قتلني, مطالبة بمحاسبة المسؤولين ...  \n",
              "4745                                        [من قتلني؟]  \n",
              "\n",
              "[4746 rows x 3 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-a0551fe2-4ac8-4ecf-9062-152f385f59a0\" 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>augmented_full_text</th>\n",
              "      <th>technique</th>\n",
              "      <th>augmented_text</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>(بداية برنامج ترويج الارهابيين يا (شار يستبق ا...</td>\n",
              "      <td>[Smears]</td>\n",
              "      <td>[(بداية برنامج ترويج الارهابيين يا (شار]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>(بشار الأسد) دنس دائماً لينتخب نفسه</td>\n",
              "      <td>[Smears]</td>\n",
              "      <td>[(بشار الأسد) دنس دائماً لينتخب نفسه]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>(ترامب) يتهم رئيس كوريا الجنوبية بنهب الولايات...</td>\n",
              "      <td>[Smears]</td>\n",
              "      <td>[(ترامب) يتهم رئيس كوريا الجنوبية بنهب الولايا...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>- الانتصار على القصف الإسرائيلي أمام قبة قاسم ...</td>\n",
              "      <td>[Smears]</td>\n",
              "      <td>[- الانتصار على القصف الإسرائيلي أمام قبة قاسم...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>-الحارس. يقتلون بالسكاكين والمناجل في شرق الكو...</td>\n",
              "      <td>[Name calling/Labeling]</td>\n",
              "      <td>[-الحارس.]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4741</th>\n",
              "      <td>يوم النصر إصرار عنبرة وبناء</td>\n",
              "      <td>[no technique]</td>\n",
              "      <td>[nan]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4742</th>\n",
              "      <td>٢٥ أيار عدد المتظاهرات في العراق ترفع شعار من ...</td>\n",
              "      <td>[Exaggeration/Minimisation]</td>\n",
              "      <td>[عدد المتظاهرات]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4743</th>\n",
              "      <td>٢٥ أيار مظاهرات حاشدة في العراق ترفع شعار من ق...</td>\n",
              "      <td>[Flag-waving]</td>\n",
              "      <td>[المطالبة بمحاسبة المسؤولين عن اغتيال النشطاء ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4744</th>\n",
              "      <td>٢٥ أيار مظاهرات حاشدة في العراق ترفع شعار من ق...</td>\n",
              "      <td>[Loaded Language, Slogans, Flag-waving, Loaded...</td>\n",
              "      <td>[من قتلني, من قتلني, مطالبة بمحاسبة المسؤولين ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4745</th>\n",
              "      <td>٢٥ أيار مظاهرات حاشدة في العراق ترفع شعار من ق...</td>\n",
              "      <td>[Slogans]</td>\n",
              "      <td>[من قتلني؟]</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>4746 rows × 3 columns</p>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-a0551fe2-4ac8-4ecf-9062-152f385f59a0')\"\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 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\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",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\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-a0551fe2-4ac8-4ecf-9062-152f385f59a0 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-a0551fe2-4ac8-4ecf-9062-152f385f59a0');\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",
              "\n",
              "\n",
              "<div id=\"df-0917da38-ad9a-414a-8f1d-d5e7587ceae3\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-0917da38-ad9a-414a-8f1d-d5e7587ceae3')\"\n",
              "            title=\"Suggest charts\"\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",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-0917da38-ad9a-414a-8f1d-d5e7587ceae3 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "  <div id=\"id_a835ef75-6b9b-427f-9b3c-bf0130b8a4c4\">\n",
              "    <style>\n",
              "      .colab-df-generate {\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-generate: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-generate {\n",
              "        background-color: #3B4455;\n",
              "        fill: #D2E3FC;\n",
              "      }\n",
              "\n",
              "      [theme=dark] .colab-df-generate: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",
              "    <button class=\"colab-df-generate\" onclick=\"generateWithVariable('grouped_df')\"\n",
              "            title=\"Generate code using this dataframe.\"\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=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "    <script>\n",
              "      (() => {\n",
              "      const buttonEl =\n",
              "        document.querySelector('#id_a835ef75-6b9b-427f-9b3c-bf0130b8a4c4 button.colab-df-generate');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      buttonEl.onclick = () => {\n",
              "        google.colab.notebook.generateWithVariable('grouped_df');\n",
              "      }\n",
              "      })();\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "grouped_df",
              "summary": "{\n  \"name\": \"grouped_df\",\n  \"rows\": 4746,\n  \"fields\": [\n    {\n      \"column\": \"augmented_full_text\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 4746,\n        \"samples\": [\n          \"\\u0628\\u064a\\u0646 \\u0633\\u0627\\u0645 \\u0648\\u0639\\u0645\\u0627\\u0631 \\u062d\\u0631\\u0648\\u0628 \\u0648\\u0646\\u0632\\u0627\\u0639\\u0627\\u062a \\u062a\\u0646\\u0634\\u0623 \\u062d\\u062a\\u0649 \\u0627\\u0644\\u064a\\u0648\\u0645 \\u0628\\u0633\\u0628\\u0628 \\u0627\\u0646\\u0643\\u0627\\u0631 \\u0632\\u0627\\u0626\\u064a\\u0631 \\u0644\\u0644\\u0625\\u0628\\u0627\\u062f\\u0629 \\u0627\\u0644\\u0623\\u0631\\u0645\\u0646\\u064a\\u0629 \\u0634\\u0627\\u0647\\u062f \\u0627\\u0644\\u062d\\u0644\\u0642\\u0629 \\u0643\\u0627\\u0645\\u0644\\u0629 \\u0639\\u0628\\u0631 \\u0627\\u0644\\u0631\\u0627\\u0628\\u0637 \\u0634\\u0627\\u0647\\u062f \\u0627\\u0644\\u062d\\u0631\\u0629\",\n          \"\\u0639\\u0628\\u064a\\u062f \\u0627\\u0644\\u0628\\u0631\\u064a\\u0643\\u064a \\u0637\\u0627\\u0626\\u0641\\u0629 \\u0645\\u0627 \\u0642\\u062f\\u0645\\u062a \\u0643\\u0627\\u0646 \\u0627\\u0644\\u062f\\u0645\\u0627\\u0631 \\u0644\\u0644\\u0628\\u0644\\u0627\\u062f \\u0648 \\u0627\\u0644\\u0627\\u063a\\u062a\\u064a\\u0627\\u0644\\u0627\\u062a \\u0648 \\u0627\\u0644\\u062c\\u0647\\u0627\\u0632 \\u0627\\u0644\\u0633\\u0631\\u064a \\u0645\\u0627\\u0632\\u0644\\u062a \\u062a\\u0639\\u0645\\u0644 \\u0639\\u0644\\u064a\\u0647\\u0627 \\u0641\\u064a \\u062d\\u0632\\u0627\\u0645 \\u0633\\u064a\\u0627\\u0633\\u064a\",\n          \"\\u0623\\u0645\\u064a\\u0631 \\u0642\\u0637\\u0631 \\u0646\\u0633\\u0639\\u062f \\u0641\\u064a \\u0643\\u0644 \\u0639\\u0627\\u0645 \\u0628\\u062a\\u062e\\u0631\\u064a\\u062c \\u0646\\u062e\\u0628\\u0629 \\u0645\\u0646 \\u0623\\u0628\\u0646\\u0627\\u0626\\u0646\\u0627 \\u0648\\u0628\\u0646\\u0627\\u062a\\u0646\\u0627 \\u0637\\u0644\\u0627\\u0628 \\u062c\\u0627\\u0645\\u0639\\u0629 \\u0642\\u0637\\u0631 \\u0648\\u0646\\u062d\\u062b\\u0647\\u0645 \\u0639\\u0644\\u0649 \\u0645\\u0648\\u0627\\u0635\\u0644\\u0629 \\u0628\\u0630\\u0644 \\u0627\\u0644\\u062c\\u0647\\u062f \\u0648\\u0627\\u0644\\u0639\\u0637\\u0627\\u0621 \\u0644\\u0644\\u0625\\u0633\\u0647\\u0627\\u0645 \\u0641\\u064a \\u062a\\u062d\\u0642\\u064a\\u0642 \\u0631\\u0624\\u064a\\u0629 \\u0642\\u0637\\u0631 \\u0627\\u0644\\u0648\\u0637\\u0646\\u064a\\u0629 \\u0641\\u064a \\u0628\\u0646\\u0627\\u0621 \\u0645\\u0633\\u062a\\u0642\\u0628\\u0644 \\u0646\\u0635\\u0631 \\u0644\\u0623\\u062c\\u064a\\u0627\\u0644 \\u0627\\u0644\\u062d\\u0627\\u0636\\u0631 \\u0648\\u0627\\u0644\\u0645\\u0633\\u062a\\u0642\\u0628\\u0644\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"technique\",\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"augmented_text\",\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 6
        }
      ],
      "source": [
        "\n",
        "df_tweets.drop(columns=[\"Unnamed: 0\", \"start\", \"end\",\"text\", \"start_new\",\"end_new\"], inplace=True)\n",
        "\n",
        "#df_tweets_test['augmented_text'] = df_tweets_test['text']\n",
        "#df_tweets_test.drop(columns=[\"start\", \"end\",\"Unnamed: 0\", \"text\", \"start_new\",\"end_new\"], inplace=True)\n",
        "\n",
        "df_tweets_val['augmented_text'] = df_tweets_val['text']\n",
        "df_tweets_val['augmented_full_text'] = df_tweets_val['full_text']\n",
        "df_tweets_val.drop(columns=[\"start\", \"end\",\"Unnamed: 0\", \"text\", \"start_new\",\"end_new\"], inplace=True)\n",
        "\n",
        "\n",
        "df_all_tweets_list = [df_tweets,df_tweets_val]\n",
        "df_all_tweets = pd.concat(df_all_tweets_list)\n",
        "\n",
        "\n",
        "# First group tags Id wise\n",
        "df_tags = df_all_tweets[['technique','augmented_full_text', 'augmented_text']]\n",
        "classes = df_tags['technique'].unique()\n",
        "\n",
        "'''\n",
        "df_tags = df_tags.drop_duplicates(subset=['technique','augmented_full_text'], ignore_index = True)\n",
        "df_tags = df_tags.groupby('augmented_full_text').apply(lambda x:x['technique'].values).reset_index(name='techniques')\n",
        "'''\n",
        "\n",
        "# Group by 'augmented_full_text' and aggregate 'technique' and 'augmented_text' as lists\n",
        "grouped_df = df_tags.groupby('augmented_full_text').agg({\n",
        "    'technique': list,\n",
        "    'augmented_text': list\n",
        "}).reset_index()\n",
        "\n",
        "'''\n",
        "# Define a regular expression pattern to match non-Arabic characters\n",
        "non_arabic_pattern = r'[^\\u0600-\\u06FF\\s]'\n",
        "\n",
        "# Remove non-Arabic characters using the regular expression pattern\n",
        "df_tags['augmented_full_text'] = df_tags['augmented_full_text'].str.replace(non_arabic_pattern, '', regex=True)\n",
        "\n",
        "df_tags.shape\n",
        "'''\n",
        "df_new = grouped_df\n",
        "df_new"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7wKtA9ofVv-0",
        "outputId": "d0df3c41-a699-4d66-accd-a309464d4c6c"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array(['no technique', 'Exaggeration/Minimisation', 'Smears',\n",
              "       'Loaded Language', 'Appeal to fear/prejudice',\n",
              "       'Name calling/Labeling', 'Slogans', 'Repetition', 'Doubt',\n",
              "       'Obfuscation, Intentional vagueness, Confusion', 'Flag-waving',\n",
              "       'Glittering generalities (Virtue)',\n",
              "       'Presenting Irrelevant Data (Red Herring)', 'Appeal to authority',\n",
              "       'Whataboutism', 'Black-and-white Fallacy/Dictatorship',\n",
              "       'Thought-terminating cliché', 'Causal Oversimplification'],\n",
              "      dtype=object)"
            ]
          },
          "metadata": {},
          "execution_count": 7
        }
      ],
      "source": [
        "classes"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "id": "blgZIYnsgs5U",
        "outputId": "323b17c7-1fc8-4f34-f00e-74be8948e85c"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                    augmented_full_text  \\\n",
              "0     (بداية برنامج ترويج الارهابيين يا (شار يستبق ا...   \n",
              "1                   (بشار الأسد) دنس دائماً لينتخب نفسه   \n",
              "2     (ترامب) يتهم رئيس كوريا الجنوبية بنهب الولايات...   \n",
              "3     - الانتصار على القصف الإسرائيلي أمام قبة قاسم ...   \n",
              "4     -الحارس. يقتلون بالسكاكين والمناجل في شرق الكو...   \n",
              "...                                                 ...   \n",
              "4741                        يوم النصر إصرار عنبرة وبناء   \n",
              "4742  ٢٥ أيار عدد المتظاهرات في العراق ترفع شعار من ...   \n",
              "4743  ٢٥ أيار مظاهرات حاشدة في العراق ترفع شعار من ق...   \n",
              "4744  ٢٥ أيار مظاهرات حاشدة في العراق ترفع شعار من ق...   \n",
              "4745  ٢٥ أيار مظاهرات حاشدة في العراق ترفع شعار من ق...   \n",
              "\n",
              "                                              technique  \\\n",
              "0                                              [Smears]   \n",
              "1                                              [Smears]   \n",
              "2                                              [Smears]   \n",
              "3                                              [Smears]   \n",
              "4                               [Name calling/Labeling]   \n",
              "...                                                 ...   \n",
              "4741                                     [no technique]   \n",
              "4742                        [Exaggeration/Minimisation]   \n",
              "4743                                      [Flag-waving]   \n",
              "4744  [Loaded Language, Slogans, Flag-waving, Loaded...   \n",
              "4745                                          [Slogans]   \n",
              "\n",
              "                                         augmented_text  \n",
              "0              [(بداية برنامج ترويج الارهابيين يا (شار]  \n",
              "1                 [(بشار الأسد) دنس دائماً لينتخب نفسه]  \n",
              "2     [(ترامب) يتهم رئيس كوريا الجنوبية بنهب الولايا...  \n",
              "3     [- الانتصار على القصف الإسرائيلي أمام قبة قاسم...  \n",
              "4                                            [-الحارس.]  \n",
              "...                                                 ...  \n",
              "4741                                              [nan]  \n",
              "4742                                   [عدد المتظاهرات]  \n",
              "4743  [المطالبة بمحاسبة المسؤولين عن اغتيال النشطاء ...  \n",
              "4744  [من قتلني, من قتلني, مطالبة بمحاسبة المسؤولين ...  \n",
              "4745                                        [من قتلني؟]  \n",
              "\n",
              "[4746 rows x 3 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-2640b444-c9f0-412c-af31-494325d2bb25\" 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>augmented_full_text</th>\n",
              "      <th>technique</th>\n",
              "      <th>augmented_text</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>(بداية برنامج ترويج الارهابيين يا (شار يستبق ا...</td>\n",
              "      <td>[Smears]</td>\n",
              "      <td>[(بداية برنامج ترويج الارهابيين يا (شار]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>(بشار الأسد) دنس دائماً لينتخب نفسه</td>\n",
              "      <td>[Smears]</td>\n",
              "      <td>[(بشار الأسد) دنس دائماً لينتخب نفسه]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>(ترامب) يتهم رئيس كوريا الجنوبية بنهب الولايات...</td>\n",
              "      <td>[Smears]</td>\n",
              "      <td>[(ترامب) يتهم رئيس كوريا الجنوبية بنهب الولايا...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>- الانتصار على القصف الإسرائيلي أمام قبة قاسم ...</td>\n",
              "      <td>[Smears]</td>\n",
              "      <td>[- الانتصار على القصف الإسرائيلي أمام قبة قاسم...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>-الحارس. يقتلون بالسكاكين والمناجل في شرق الكو...</td>\n",
              "      <td>[Name calling/Labeling]</td>\n",
              "      <td>[-الحارس.]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4741</th>\n",
              "      <td>يوم النصر إصرار عنبرة وبناء</td>\n",
              "      <td>[no technique]</td>\n",
              "      <td>[nan]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4742</th>\n",
              "      <td>٢٥ أيار عدد المتظاهرات في العراق ترفع شعار من ...</td>\n",
              "      <td>[Exaggeration/Minimisation]</td>\n",
              "      <td>[عدد المتظاهرات]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4743</th>\n",
              "      <td>٢٥ أيار مظاهرات حاشدة في العراق ترفع شعار من ق...</td>\n",
              "      <td>[Flag-waving]</td>\n",
              "      <td>[المطالبة بمحاسبة المسؤولين عن اغتيال النشطاء ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4744</th>\n",
              "      <td>٢٥ أيار مظاهرات حاشدة في العراق ترفع شعار من ق...</td>\n",
              "      <td>[Loaded Language, Slogans, Flag-waving, Loaded...</td>\n",
              "      <td>[من قتلني, من قتلني, مطالبة بمحاسبة المسؤولين ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4745</th>\n",
              "      <td>٢٥ أيار مظاهرات حاشدة في العراق ترفع شعار من ق...</td>\n",
              "      <td>[Slogans]</td>\n",
              "      <td>[من قتلني؟]</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>4746 rows × 3 columns</p>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-2640b444-c9f0-412c-af31-494325d2bb25')\"\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 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\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",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\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-2640b444-c9f0-412c-af31-494325d2bb25 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-2640b444-c9f0-412c-af31-494325d2bb25');\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",
              "\n",
              "\n",
              "<div id=\"df-f348244e-3bd8-4a52-92ff-22bdcd3d6fc7\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-f348244e-3bd8-4a52-92ff-22bdcd3d6fc7')\"\n",
              "            title=\"Suggest charts\"\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",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-f348244e-3bd8-4a52-92ff-22bdcd3d6fc7 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "  <div id=\"id_07901f52-56d0-4e63-9636-3d21e6d5e777\">\n",
              "    <style>\n",
              "      .colab-df-generate {\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-generate: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-generate {\n",
              "        background-color: #3B4455;\n",
              "        fill: #D2E3FC;\n",
              "      }\n",
              "\n",
              "      [theme=dark] .colab-df-generate: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",
              "    <button class=\"colab-df-generate\" onclick=\"generateWithVariable('grouped_df')\"\n",
              "            title=\"Generate code using this dataframe.\"\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=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "    <script>\n",
              "      (() => {\n",
              "      const buttonEl =\n",
              "        document.querySelector('#id_07901f52-56d0-4e63-9636-3d21e6d5e777 button.colab-df-generate');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      buttonEl.onclick = () => {\n",
              "        google.colab.notebook.generateWithVariable('grouped_df');\n",
              "      }\n",
              "      })();\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "grouped_df",
              "summary": "{\n  \"name\": \"grouped_df\",\n  \"rows\": 4746,\n  \"fields\": [\n    {\n      \"column\": \"augmented_full_text\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 4746,\n        \"samples\": [\n          \"\\u0628\\u064a\\u0646 \\u0633\\u0627\\u0645 \\u0648\\u0639\\u0645\\u0627\\u0631 \\u062d\\u0631\\u0648\\u0628 \\u0648\\u0646\\u0632\\u0627\\u0639\\u0627\\u062a \\u062a\\u0646\\u0634\\u0623 \\u062d\\u062a\\u0649 \\u0627\\u0644\\u064a\\u0648\\u0645 \\u0628\\u0633\\u0628\\u0628 \\u0627\\u0646\\u0643\\u0627\\u0631 \\u0632\\u0627\\u0626\\u064a\\u0631 \\u0644\\u0644\\u0625\\u0628\\u0627\\u062f\\u0629 \\u0627\\u0644\\u0623\\u0631\\u0645\\u0646\\u064a\\u0629 \\u0634\\u0627\\u0647\\u062f \\u0627\\u0644\\u062d\\u0644\\u0642\\u0629 \\u0643\\u0627\\u0645\\u0644\\u0629 \\u0639\\u0628\\u0631 \\u0627\\u0644\\u0631\\u0627\\u0628\\u0637 \\u0634\\u0627\\u0647\\u062f \\u0627\\u0644\\u062d\\u0631\\u0629\",\n          \"\\u0639\\u0628\\u064a\\u062f \\u0627\\u0644\\u0628\\u0631\\u064a\\u0643\\u064a \\u0637\\u0627\\u0626\\u0641\\u0629 \\u0645\\u0627 \\u0642\\u062f\\u0645\\u062a \\u0643\\u0627\\u0646 \\u0627\\u0644\\u062f\\u0645\\u0627\\u0631 \\u0644\\u0644\\u0628\\u0644\\u0627\\u062f \\u0648 \\u0627\\u0644\\u0627\\u063a\\u062a\\u064a\\u0627\\u0644\\u0627\\u062a \\u0648 \\u0627\\u0644\\u062c\\u0647\\u0627\\u0632 \\u0627\\u0644\\u0633\\u0631\\u064a \\u0645\\u0627\\u0632\\u0644\\u062a \\u062a\\u0639\\u0645\\u0644 \\u0639\\u0644\\u064a\\u0647\\u0627 \\u0641\\u064a \\u062d\\u0632\\u0627\\u0645 \\u0633\\u064a\\u0627\\u0633\\u064a\",\n          \"\\u0623\\u0645\\u064a\\u0631 \\u0642\\u0637\\u0631 \\u0646\\u0633\\u0639\\u062f \\u0641\\u064a \\u0643\\u0644 \\u0639\\u0627\\u0645 \\u0628\\u062a\\u062e\\u0631\\u064a\\u062c \\u0646\\u062e\\u0628\\u0629 \\u0645\\u0646 \\u0623\\u0628\\u0646\\u0627\\u0626\\u0646\\u0627 \\u0648\\u0628\\u0646\\u0627\\u062a\\u0646\\u0627 \\u0637\\u0644\\u0627\\u0628 \\u062c\\u0627\\u0645\\u0639\\u0629 \\u0642\\u0637\\u0631 \\u0648\\u0646\\u062d\\u062b\\u0647\\u0645 \\u0639\\u0644\\u0649 \\u0645\\u0648\\u0627\\u0635\\u0644\\u0629 \\u0628\\u0630\\u0644 \\u0627\\u0644\\u062c\\u0647\\u062f \\u0648\\u0627\\u0644\\u0639\\u0637\\u0627\\u0621 \\u0644\\u0644\\u0625\\u0633\\u0647\\u0627\\u0645 \\u0641\\u064a \\u062a\\u062d\\u0642\\u064a\\u0642 \\u0631\\u0624\\u064a\\u0629 \\u0642\\u0637\\u0631 \\u0627\\u0644\\u0648\\u0637\\u0646\\u064a\\u0629 \\u0641\\u064a \\u0628\\u0646\\u0627\\u0621 \\u0645\\u0633\\u062a\\u0642\\u0628\\u0644 \\u0646\\u0635\\u0631 \\u0644\\u0623\\u062c\\u064a\\u0627\\u0644 \\u0627\\u0644\\u062d\\u0627\\u0636\\u0631 \\u0648\\u0627\\u0644\\u0645\\u0633\\u062a\\u0642\\u0628\\u0644\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"technique\",\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"augmented_text\",\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 8
        }
      ],
      "source": [
        "df_new"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "N7BCK5cwvaCa",
        "outputId": "c7f19fed-2e4a-4b48-cf6a-88090a4f91d5"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "18"
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ],
      "source": [
        "# Define necessary hyperparameters\n",
        "MAX_NUM_WORDS = 10000  # Maximum number of words to keep in the tokenizer (vocabulary size)\n",
        "max_seq_length = 50  # Maximum length of sequences after padding\n",
        "embedding_dim = 100  # Size of the embedding vectors\n",
        "num_classes = len(classes)  # Adjust according to your number of classes for multi-label classification\n",
        "num_classes"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "-S4xW5JoYvsb"
      },
      "outputs": [],
      "source": [
        "from tensorflow.keras.layers import Layer\n",
        "import tensorflow as tf\n",
        "\n",
        "# Custom Attention Layer\n",
        "class Attention(Layer):\n",
        "    def __init__(self, **kwargs):\n",
        "        super(Attention, self).__init__(**kwargs)\n",
        "\n",
        "    def build(self, input_shape):\n",
        "        self.W = self.add_weight(name=\"att_weight\", shape=(input_shape[-1], input_shape[-1]),\n",
        "                                 initializer=\"glorot_uniform\", trainable=True)\n",
        "        self.b = self.add_weight(name=\"att_bias\", shape=(input_shape[-1],),\n",
        "                                 initializer=\"zeros\", trainable=True)\n",
        "        super(Attention, self).build(input_shape)\n",
        "\n",
        "    def call(self, x):\n",
        "        e = tf.keras.activations.tanh(tf.tensordot(x, self.W, axes=(2, 0)) + self.b)\n",
        "        a = tf.keras.activations.softmax(e, axis=1)\n",
        "        output = tf.reduce_sum(x * a, axis=1)\n",
        "        return output\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "8VMLOogmI4sh"
      },
      "outputs": [],
      "source": [
        "from sklearn.preprocessing import MultiLabelBinarizer\n",
        "\n",
        "# Fit and transform labels using MultiLabelBinarizer\n",
        "mlb = MultiLabelBinarizer(classes = classes)\n",
        "\n",
        "labels = df_new['technique'].tolist()\n",
        "\n",
        "labels_encoded = mlb.fit_transform(labels)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "T-yYB4OwUvmi",
        "outputId": "1859e9e2-3c84-4723-b8b8-1d76fd9845f3"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0])"
            ]
          },
          "metadata": {},
          "execution_count": 12
        }
      ],
      "source": [
        "labels_encoded[0]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "2cryUZxjI4si"
      },
      "outputs": [],
      "source": [
        "x = df_new['augmented_full_text'].values\n",
        "y = labels_encoded"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Zh2G08flpR_X"
      },
      "source": [
        "# for test dataset"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "x3KMnaLlpbMO",
        "outputId": "8639d545-9259-45b4-e7b3-6f7182342f6f"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "<ipython-input-14-cea43e885653>:7: DeprecationWarning: DataFrameGroupBy.apply operated on the grouping columns. This behavior is deprecated, and in a future version of pandas the grouping columns will be excluded from the operation. Either pass `include_groups=False` to exclude the groupings or explicitly select the grouping columns after groupby to silence this warning.\n",
            "  test_tags = test_tags.groupby('augmented_full_text').apply(lambda x:x['technique'].values).reset_index(name='technique')\n"
          ]
        }
      ],
      "source": [
        "\n",
        "# First group tags Id wise\n",
        "df_tweets_test['augmented_full_text'] = df_tweets_test['full_text']\n",
        "#df_tweets_test.drop(columns=[\"start\", \"end\",\"Unnamed: 0\", \"text\", \"start_new\",\"end_new\"], inplace=True)\n",
        "test_tags = df_tweets_test[['technique','augmented_full_text']]\n",
        "test_classes = test_tags['technique'].unique()\n",
        "test_tags = test_tags.drop_duplicates(subset=['technique','augmented_full_text'], ignore_index = True)\n",
        "test_tags = test_tags.groupby('augmented_full_text').apply(lambda x:x['technique'].values).reset_index(name='technique')\n",
        "\n",
        "\n",
        "# Define a regular expression pattern to match non-Arabic characters\n",
        "non_arabic_pattern = r'[^\\u0600-\\u06FF\\s]'\n",
        "\n",
        "# Remove non-Arabic characters using the regular expression pattern\n",
        "test_tags['augmented_full_text'] = test_tags['augmented_full_text'].str.replace(non_arabic_pattern, '', regex=True)\n",
        "\n",
        "\n",
        "test_labels = test_tags['technique'].tolist()\n",
        "\n",
        "test_labels_encoded = mlb.fit_transform(test_labels)\n",
        "\n",
        "#df_tweets_test['augmented_text'] = df_tweets_test['text']\n",
        "#df_tweets_test.drop(columns=[\"start\", \"end\",\"Unnamed: 0\", \"full_text\" , \"text\", \"start_new\",\"end_new\"], inplace=True)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "2cDroh2HstIT"
      },
      "outputs": [],
      "source": [
        "\n",
        "x_ts = test_tags['augmented_full_text'].values\n",
        "y_ts = test_labels_encoded\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oRX_obwQ6Neo",
        "outputId": "cc190b91-5730-4bad-e7a0-330b6dd4229d"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array(['أربعة قتلى خلال تبادل لإطلاق النار بين الشرطة وإرهابيين وسط غرب تونس مصدر أمني',\n",
              "       'أول دولة أوروبية تصنف الاستيطان الإسرائيلي كاحتلال عين ليبيا',\n",
              "       'إفريقيا من منظمة الوحدة إلى الاتحاد',\n",
              "       'إيطاليا ارتفاع حصيلة ضحايا حادث تحطم قاطرة تلفريك إلى قتيلا',\n",
              "       'الأمم المتحدة السيد هلال يندد ب الارتباط الإيديولوجي و العمى السياسي لجنوب إفريقيا حول الصحراء المغربية',\n",
              "       'الاقتصاد الأخضر وتحدياته محور ندوة بالرباط',\n",
              "       'الباحث السياسي عز الدين عقيل مجموعة فاغنر التي تقاتل مع قوات الكرامة قد يخططون لانقلاب على حفتر داخل الرجمة خاصة أن الوضع محتقن بسبب هزيمة عقيلة صالح ليبيا',\n",
              "       'الحملة الإعلامية الإسبانية ضد المغرب الحديث عن ضعف التنمية في المملكة ينم عن تصورات قديمة لدى الجار الشمالي فالمغرب حقق نسب نمو مطردة رغم الأزمة',\n",
              "       'الحوار الاجتماعي الحكومة أوفت بكل التزاماتها رغم تداعيات الأزمة الصحية',\n",
              "       'السعيدية فتح بحث قضائي لتحديد المتورطين في محاولة تهريب أربعة أطنان و كيلوغرام من مخدر الشيرا عبر المسالك البحرية',\n",
              "       'الصحراء المغربية أي تصريح من قبل عضو في الحكومة لا يتوافق مع الموقف الرسمي لإسبانيا ما هو إلا مجرد رأي شخصي وزيرة الدفاع',\n",
              "       'الصحراء المغربية عمدة ليفيري الإيطالية يدعو الاتحاد الأوروبي إلى السير على خطى النموذج الأمريكي',\n",
              "       'الصحفي المتخصص في شؤون الجماعات المتطرفة ماهر فرغلي تركيا تدعم الفصائل المسلحة في سوريا بدون حدود وعلى العالم مواجهة هذه الممارسات العربية',\n",
              "       'الفريق شنقريحة يؤكد حرص الجيش الوطني الشعبي على مواصلة أداء مهامه الدستورية النبيلة',\n",
              "       'القيادة العسكرية الأمريكية في إفريقيا أفريكوم لدينا صور جوية تظهر فاغنر ومعدات لها في الخطوط الأمامية في سرت روسيا ومجموعة فاغنر تواصلان المشاركة في العمليات البرية والجوية في ليبيا تواصل روسيا لعب دور غير مفيد في ليبيا من خلال تسليم الإمدادات والمعدات إلى مجموعة فاغنر',\n",
              "       'الكتيبة مشاة التابعة لرئاسة الأركان العامة تلقي القبض على تشكيل عصابي داخل منطقة الماية غرب طرابلس يمتهن السطو المسلح والنصب والاحتيال على المواطنين',\n",
              "       'المجلس الوطني لحقوق الانسان بقاء الصحراء الغربية تحت وطأة الاستعمار نقطة سوداء في جبين افريقيا',\n",
              "       'المدير التنفيذي لشركة غازبروم الروسية سيرجي تومانوف قال إن الشركة استأنفت إنتاج النفط الخام في ليبيا بعد توقف دام أشهر بموجب اتفاق مشترك مع وينترشال الألمانية مضيفا أنه من المتوقع أن يرتفع الإنتاج الحالي في المشروع من ألف برميل في اليوم إلى ألفا مع نهاية عام',\n",
              "       'المغرب بلد موثوق به يأخذ مكافحة الهجرة غير الشرعية بجدية بالغة السفيرة الفرنسية',\n",
              "       'الناظور إحباط محاولة للتهريب الدولي للمخدرات وحجز خمسة أطنان و كلغ من مخدر الشيرا المديرية العامة للأمن الوطني',\n",
              "       'النموذج التنموي الجديد المحاور الاستراتيجية للتحول',\n",
              "       'بالتواطؤ مع شبكات إجرامية مغربية تفكيك شبكة إجرامية مكونة من ستة أشخاص بحوزتهم مبالغ مالية ضخمة بقيمة إجمالية تقدر ملايير و مليون سنتيم منها مبلغ مالي مزور بقيمة ملايير و مليون سنتيم حاولوا إدخالها وترويجها',\n",
              "       'بعد قرابة عقد من الزمان على أحداث بنغازي هل ستعيد إدارة بايدن فتح السفارة الأمريكية في طرابلس',\n",
              "       'بعيدا عن المغالطات التي تروجها وسائل إعلام إسبانية حول عدد القاصرين المغاربة الذين تمكنوا من العبور إلى مدينة سبتة المحتلة القاصرون المتبقون حاليا بالمدينة المحتلة يناهز حوالي قاصرا مغربيا السلطات المغربية تعمل على عودتهم واستقبالهم في مركز متخصص قبل الالتحاق بأسرهم',\n",
              "       'بوليساريو يروج زيفا لصور من ليبيا تدعي أنها لطائرة مغربية تم إسقاطها',\n",
              "       'تحذيرات من حرب جديدة في حال فشل الانتخابات القادمة',\n",
              "       'تونس مقتل ستة عناصر من الدرك على يد مجموعة إرهابية على الحدود مع الجزائر',\n",
              "       'حملت القبيلة قيادة قوات حفتر مسؤولية الانفلات الأمني والاستهتار بالأرواح عين ليبيا',\n",
              "       'ديوان المحاسبة الليبي يوقف رئيس المؤسسة الليبية للإعلام محمد بعيو عن العمل احتياطيا وذلك لعرقلته عمل اللجنة المكلفة بفحص حسابات المؤسسة',\n",
              "       'رئيس بعثة الأمم المتحدة للدعم في ليبيا يان كوبيش يشدد في الكلمة الافتتاحية لجلسة ملتقى الحوار السياسي على ضرورة إجراء الانتخابات في موعدها المقرر في ديسمبر القادم',\n",
              "       'شاهد المتحدث باسم مليشيات حفتر أحمد المسماري يقول إن الاحتفال عمل مدني وليس عسكريا وأنه لتغيير الأجواء والفرح وليس تهديدا لأحد',\n",
              "       'شاهد الناشط القبلي أسامة بوحلاق يهاجم تنظيم المداخلة للمرة الثانية خلال أسبوع ويصف منتسبيه بالمطبلين للحكام والمدعيين الإلتزام وأنهم لا يمثلون الدين في شيء',\n",
              "       'شاهد بالتزامن مع ذكرى نكبة الكرامة منذ سنوات التي دمرت وقتلت وهجرت ومزقت النسيج الاجتماعي بين الليبيين تفتتح المليشيات اليوم في بنغازي مطبعة عسكرية لطباعة صور حفتر وتعليقها في الشوارع',\n",
              "       'شاهد برتبة عقيد صدام ابن آمر المليشيات خليفة حفتر يكرم أفرادا من تنظيم المداخلة الإرهابي',\n",
              "       'شاهد بعد فشله عسكريا في السيطرة على العاصمة طرابلس و ليبيا المتحدث باسم مليشيات حفتر أحمد المسماري يدعو الله أن يخرج القوات التركية ويطالب المجتمع الدولي بالضغط على تركيا لخروج قواتها من ليبيا ويقر بالفوضى الأمنية في مدينة بنغازي',\n",
              "       'شاهد في بيان لشباب ثورة الفقراء بمدينة سبها يطالبون بتحسين أوضاعهم المعيشية ويهددون بإقفال النفط في الجنوب',\n",
              "       'شاهد فيديو جديد لأرتال مسلحة تابعة لمليشيات حفتر ببنغازي يظهر استحواذهم على منصات صواريخ سكود و لونا القادرة على حمل رؤوس حربية شديدة الانفجار أو نووية وكيميائية',\n",
              "       'شاهد مرتزق يوثق قدومه مع أرتال عسكرية قادمة إلى مدينة سبها من تشاد ويوجه رسالة إلى التشاديين أنهم سيأتونهم بخيرات ليبيا',\n",
              "       'شاهد وزير الصحة بحكومة الوحدة الوطنية علي الزناتي الوضع الوبائي مستقر في أرجاء ليبيا ولا توجد منطقة وبائية معينة أكثر من غيرها طرابلس',\n",
              "       'شمام يطالب شباب ونساء برقة أن يغيروا بأيديهم عين ليبيا',\n",
              "       'صور إدارة خدمات الطوارئ بشركة الخدمات العامة تواصل إزالة المباني العشوائية المخالفة بمناطق مختلفة من طرابلس ليبيا',\n",
              "       'صور مستعرضا قوته في بنغازي قبل أيام من مرور ذكرى هزيمته في طرابلس حفتر يدعو كلا من رئيس حكومة الوحدة الوطنية عبدالحميد الدبيبة ورئيس المجلس الرئاسي محمد المنفي ورئيس البرلمان عقيلة صالح لحضور عرض عسكري لمليشياته بمناسبة ذكرى نكبة الكرامة',\n",
              "       'عبد الرحمن الشاطر صدق أو لا تصدق كل المترشحين للرئاسي ورئيس الحكومة وعدوا بتحقيق الاستقرار ومحاربة الفساد وتوحيد المؤسسات وإجراء انتخابات معالجة الاقتصاد في عشرة أشهر فقط رجال المعجزات يتقدمون الصفوف منهم من ساهم في الفساد وقتل الليبيين وتدمير المدن على مين يا طبرق',\n",
              "       'غوتيريش المقاتلون الأجانب لا يزالون ينشطون في ليبيا عين ليبيا',\n",
              "       'فوق السلطة الإمارات تهنىء إيران برمضان وإسرائيل بالنكبة',\n",
              "       'فيديو القيادي في قوات حفتر محمود الورفلي يقتحم مقر شركة ماس وكيل شركة تويوتا في بنغازي ويقوم بتحطيم المقر بدعوى الفساد المالي',\n",
              "       'كارثة بيئية وصناعية بأستراليا ملايين الفئران تجتاح الحقول والمزارع وتغزو المنازل كارثة أستراليا فئران',\n",
              "       'مالي الإفراج عن رئيسي المرحلة الانتقالية والحكومة',\n",
              "       'مدينة شحات الأثرية إحدى المدن الليبية الواقعة شرق بنغازي والتي تتميز بامتزاج طبيعتها الريفية الخلابة مع الآثار التاريخية',\n",
              "       'مصادر إسبانية تكشف أن زعيم جبهة البوليساريو الانفصالية وافق على المثول أمام المحكمة العليا الإسبانية الأسبوع المقبل وذلك بعدما رفض سابقا تسلم استدعاء المحكمة وطلب مهلة للتشاور مع الجزائر المغرب إسبانيا الرباط مدريد الصحراء المغربية',\n",
              "       'مصادر إعلامية مجموعة من المسلحين الملثمين يقتحمون منزل تاجر قماش من عائلة قرقوم في بنغازي ليلة البارحة ويقومون بخطفه وسرقة مبلغ مالي ومجوهرات ثمنية وذهب من المنزل ويخلون سبيله فجر اليوم الأربعاء',\n",
              "       'ملتقى افريقيا للاستثمار دعوة الجزائر لقيادة الدول الافريقية نحو تجسيد سوقهم الموحدة',\n",
              "       'وزير الخارجية الأميركي العنف في منطقة الشرق الأوسط يبعدنا أكثر فأكثر عن حل الدولتين'],\n",
              "      dtype=object)"
            ]
          },
          "metadata": {},
          "execution_count": 16
        }
      ],
      "source": [
        "x_ts"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "2cDAdoOvY2sB"
      },
      "outputs": [],
      "source": [
        "\n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "# Define your random seed\n",
        "RANDOM_SEED = 42\n",
        "\n",
        "# First, split into training + validation and test sets\n",
        "x_train, x_val, y_train, y_val = train_test_split(x, y, test_size=0.2, random_state=RANDOM_SEED)\n",
        "\n",
        "# Then, split the training + validation set into training and validation sets\n",
        "#x_train_val, x_val, y_train_val, y_val = train_test_split(x_train, y_train, test_size=0.4, random_state=RANDOM_SEED)\n",
        "\n",
        "\n",
        "\n",
        "#x_val,x_ts,y_val,y_ts = train_test_split(X_test, y_test, test_size=0.8, random_state=RANDOM_SEED,shuffle=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uDMBKboJZf16",
        "outputId": "6408d97c-6ff8-468b-fc52-b78b08d8b804"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Collecting stanza\n",
            "  Downloading stanza-1.9.2-py3-none-any.whl.metadata (13 kB)\n",
            "Collecting emoji (from stanza)\n",
            "  Downloading emoji-2.14.0-py3-none-any.whl.metadata (5.7 kB)\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.10/dist-packages (from stanza) (1.26.4)\n",
            "Requirement already satisfied: protobuf>=3.15.0 in /usr/local/lib/python3.10/dist-packages (from stanza) (3.20.3)\n",
            "Requirement already satisfied: requests in /usr/local/lib/python3.10/dist-packages (from stanza) (2.32.3)\n",
            "Requirement already satisfied: networkx in /usr/local/lib/python3.10/dist-packages (from stanza) (3.4.2)\n",
            "Requirement already satisfied: torch>=1.3.0 in /usr/local/lib/python3.10/dist-packages (from stanza) (2.5.0+cu121)\n",
            "Requirement already satisfied: tqdm in /usr/local/lib/python3.10/dist-packages (from stanza) (4.66.6)\n",
            "Requirement already satisfied: tomli in /usr/local/lib/python3.10/dist-packages (from stanza) (2.0.2)\n",
            "Requirement already satisfied: filelock in /usr/local/lib/python3.10/dist-packages (from torch>=1.3.0->stanza) (3.16.1)\n",
            "Requirement already satisfied: typing-extensions>=4.8.0 in /usr/local/lib/python3.10/dist-packages (from torch>=1.3.0->stanza) (4.12.2)\n",
            "Requirement already satisfied: jinja2 in /usr/local/lib/python3.10/dist-packages (from torch>=1.3.0->stanza) (3.1.4)\n",
            "Requirement already satisfied: fsspec in /usr/local/lib/python3.10/dist-packages (from torch>=1.3.0->stanza) (2024.10.0)\n",
            "Requirement already satisfied: sympy==1.13.1 in /usr/local/lib/python3.10/dist-packages (from torch>=1.3.0->stanza) (1.13.1)\n",
            "Requirement already satisfied: mpmath<1.4,>=1.1.0 in /usr/local/lib/python3.10/dist-packages (from sympy==1.13.1->torch>=1.3.0->stanza) (1.3.0)\n",
            "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.10/dist-packages (from requests->stanza) (3.4.0)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests->stanza) (3.10)\n",
            "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests->stanza) (2.2.3)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests->stanza) (2024.8.30)\n",
            "Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.10/dist-packages (from jinja2->torch>=1.3.0->stanza) (3.0.2)\n",
            "Downloading stanza-1.9.2-py3-none-any.whl (1.1 MB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.1/1.1 MB\u001b[0m \u001b[31m13.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading emoji-2.14.0-py3-none-any.whl (586 kB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m586.9/586.9 kB\u001b[0m \u001b[31m33.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hInstalling collected packages: emoji, stanza\n",
            "Successfully installed emoji-2.14.0 stanza-1.9.2\n"
          ]
        }
      ],
      "source": [
        "!pip install stanza\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 154,
          "referenced_widgets": [
            "27dc81c687824079a11e08cffbc8bc15",
            "52550e94fecb430eb8d292ede83e9572",
            "af59a2f876fb4b9dbc249cca0eee99bf",
            "5d02a44463504b8284d289c6f11208b5",
            "0a066e9c12434a96b64c9ae56a92d1af",
            "94928004da8845ba976664334254bb57",
            "a3251a569a57420da82239bd27b29c27",
            "ed53626694b34ee5b0ce325a592a6906",
            "5589ebc3cd164624aeb171a87e9d1104",
            "d6a821c1aaf0450aaa329921df4bf88a",
            "86c0477fc27e498ca7d59fa61fe0276b",
            "8ad55e00fdef4c488320154259d5db86",
            "3612c3c9ce3f45cd9f6b6a94c65811fd",
            "c9927e6287e5483d986a5e669865283f",
            "0e7720c663ea40a4bf580132dfa472d4",
            "48a67a849a8c4e969c8456461bbbbb15",
            "3d0475b1fdd446a6b452ff6d596a2c59",
            "532d749ac10e4d83a5ad163bde92694f",
            "7fbcbf179a1249caa3e7a30a6dc6fd95",
            "f4e5fbdee8954b53b487e629fc67390c",
            "3f75dd0d46b344078bf007de152f41a1",
            "2be1ef8a2f0a4f9d9fe10368bb5b9464"
          ]
        },
        "id": "Tu6UwL-CaEO-",
        "outputId": "da470f10-17fa-411f-e9a4-1ce1788a5784"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading https://raw.githubusercontent.com/stanfordnlp/stanza-resources/main/resources_1.9.0.json:   0%|   …"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "27dc81c687824079a11e08cffbc8bc15"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "INFO:stanza:Downloaded file to /root/stanza_resources/resources.json\n",
            "INFO:stanza:Downloading default packages for language: ar (Arabic) ...\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading https://huggingface.co/stanfordnlp/stanza-ar/resolve/v1.9.0/models/default.zip:   0%|          | 0…"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "8ad55e00fdef4c488320154259d5db86"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "INFO:stanza:Downloaded file to /root/stanza_resources/ar/default.zip\n",
            "INFO:stanza:Finished downloading models and saved to /root/stanza_resources\n"
          ]
        }
      ],
      "source": [
        "import stanza\n",
        "stanza.download('ar')  # Download the Arabic model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 632,
          "referenced_widgets": [
            "22c0c080b0aa48d9a263db83bd4ba292",
            "eb369e73ada24787a5eeb43bf6ee0e58",
            "c43af64ec8ed4faea1bdd3d6172fa451",
            "8035da71d9ed45debb811fddbc652c5b",
            "8558097071f74d099dc9439dafeda48f",
            "b75db6785bb34fc391725093eb5d0d73",
            "756441376ecd4202b96527bc665d2cee",
            "be8b6e2eaafc4ca381735e927dfa949e",
            "0ae47ceac4b4408e87c105f17fb07403",
            "e401143ac189458091cb5bfa3d245e59",
            "e8011943f9a14a2da3485b035b2d1ce1"
          ]
        },
        "id": "wB4edQCtZovZ",
        "outputId": "fd7687fa-d03e-4a40-f773-5e8ea9979ab3"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "INFO:stanza:Checking for updates to resources.json in case models have been updated.  Note: this behavior can be turned off with download_method=None or download_method=DownloadMethod.REUSE_RESOURCES\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading https://raw.githubusercontent.com/stanfordnlp/stanza-resources/main/resources_1.9.0.json:   0%|   …"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "22c0c080b0aa48d9a263db83bd4ba292"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "INFO:stanza:Downloaded file to /root/stanza_resources/resources.json\n",
            "WARNING:stanza:Language ar package default expects mwt, which has been added\n",
            "INFO:stanza:Loading these models for language: ar (Arabic):\n",
            "============================\n",
            "| Processor | Package      |\n",
            "----------------------------\n",
            "| tokenize  | padt         |\n",
            "| mwt       | padt         |\n",
            "| pos       | padt_charlm  |\n",
            "| ner       | aqmar_charlm |\n",
            "============================\n",
            "\n",
            "INFO:stanza:Using device: cpu\n",
            "INFO:stanza:Loading: tokenize\n",
            "/usr/local/lib/python3.10/dist-packages/stanza/models/tokenization/trainer.py:82: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.\n",
            "  checkpoint = torch.load(filename, lambda storage, loc: storage)\n",
            "INFO:stanza:Loading: mwt\n",
            "/usr/local/lib/python3.10/dist-packages/stanza/models/mwt/trainer.py:201: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.\n",
            "  checkpoint = torch.load(filename, lambda storage, loc: storage)\n",
            "INFO:stanza:Loading: pos\n",
            "/usr/local/lib/python3.10/dist-packages/stanza/models/pos/trainer.py:139: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.\n",
            "  checkpoint = torch.load(filename, lambda storage, loc: storage)\n",
            "/usr/local/lib/python3.10/dist-packages/stanza/models/common/pretrain.py:56: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.\n",
            "  data = torch.load(self.filename, lambda storage, loc: storage)\n",
            "/usr/local/lib/python3.10/dist-packages/stanza/models/common/char_model.py:271: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.\n",
            "  state = torch.load(filename, lambda storage, loc: storage)\n",
            "INFO:stanza:Loading: ner\n",
            "/usr/local/lib/python3.10/dist-packages/stanza/models/ner/trainer.py:197: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.\n",
            "  checkpoint = torch.load(filename, lambda storage, loc: storage)\n",
            "INFO:stanza:Done loading processors!\n"
          ]
        }
      ],
      "source": [
        "from tensorflow.keras.preprocessing.text import Tokenizer\n",
        "from tensorflow.keras.preprocessing.sequence import pad_sequences\n",
        "\n",
        "# Initialize Keras Tokenizer\n",
        "tokenizer = Tokenizer(num_words=MAX_NUM_WORDS, oov_token='<OOV>')\n",
        "# Fit the tokenizer on the training data\n",
        "tokenizer.fit_on_texts(x_train)\n",
        "\n",
        "# Load Stanza pipeline for Arabic\n",
        "nlp = stanza.Pipeline('ar', processors='tokenize,pos,ner')\n",
        "import numpy as np\n",
        "\n",
        "\n",
        "def get_pos_ner_features_stanza(texts):\n",
        "    pos_features = []\n",
        "    ner_features = []\n",
        "    for text in texts:\n",
        "        doc = nlp(text)\n",
        "        pos_tags = [word.upos for sent in doc.sentences for word in sent.words]  # POS tags\n",
        "        entities = [ent.type for ent in doc.ents]  # NER tags\n",
        "        pos_features.append(pos_tags)\n",
        "        ner_features.append(entities)\n",
        "    return pos_features, ner_features\n",
        "\n",
        "# Get POS and NER features for your training, validation, and test sets\n",
        "pos_train, ner_train = get_pos_ner_features_stanza(x_train)\n",
        "pos_val, ner_val = get_pos_ner_features_stanza(x_val)\n",
        "pos_test, ner_test = get_pos_ner_features_stanza(x_ts)\n",
        "\n",
        "# Tokenize POS + NER features (Ensure tokenizer is defined as earlier)\n",
        "pos_train_sequences = tokenizer.texts_to_sequences(pos_train)\n",
        "ner_train_sequences = tokenizer.texts_to_sequences(ner_train)\n",
        "pos_val_sequences = tokenizer.texts_to_sequences(pos_val)\n",
        "ner_val_sequences = tokenizer.texts_to_sequences(ner_val)\n",
        "pos_test_sequences = tokenizer.texts_to_sequences(pos_test)\n",
        "ner_test_sequences = tokenizer.texts_to_sequences(ner_test)\n",
        "\n",
        "# Padding the sequences (same max length as text sequences)\n",
        "pos_train_padded = pad_sequences(pos_train_sequences, maxlen=max_seq_length, padding='post')\n",
        "ner_train_padded = pad_sequences(ner_train_sequences, maxlen=max_seq_length, padding='post')\n",
        "pos_val_padded = pad_sequences(pos_val_sequences, maxlen=max_seq_length, padding='post')\n",
        "ner_val_padded = pad_sequences(ner_val_sequences, maxlen=max_seq_length, padding='post')\n",
        "pos_test_padded = pad_sequences(pos_test_sequences, maxlen=max_seq_length, padding='post')\n",
        "ner_test_padded = pad_sequences(ner_test_sequences, maxlen=max_seq_length, padding='post')\n",
        "\n",
        "# Convert the text to sequences\n",
        "train_sequences = tokenizer.texts_to_sequences(x_train)\n",
        "test_sequences = tokenizer.texts_to_sequences(x_ts)\n",
        "val_sequences = tokenizer.texts_to_sequences(x_val)\n",
        "\n",
        "# Pad sequences to ensure equal length\n",
        "train_inputs = pad_sequences(train_sequences, maxlen=max_seq_length, padding='post')\n",
        "test_inputs = pad_sequences(test_sequences, maxlen=max_seq_length, padding='post')\n",
        "val_inputs = pad_sequences(val_sequences, maxlen=max_seq_length, padding='post')\n",
        "\n",
        "# Concatenate the padded POS and NER features with the text sequences\n",
        "train_inputs_combined = np.concatenate([train_inputs, pos_train_padded, ner_train_padded], axis=1)\n",
        "val_inputs_combined = np.concatenate([val_inputs, pos_val_padded, ner_val_padded], axis=1)\n",
        "test_inputs_combined = np.concatenate([test_inputs, pos_test_padded, ner_test_padded], axis=1)\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0LZzxabG4nmj",
        "outputId": "943623ec-2095-49ec-8869-df0774a9e361"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(3796,)"
            ]
          },
          "metadata": {},
          "execution_count": 21
        }
      ],
      "source": [
        " x_train.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tzu_lZIuDeRO",
        "outputId": "32187f6e-084e-4cae-bae1-7b1385c4e51b"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((53,), (950,), (3796,))"
            ]
          },
          "metadata": {},
          "execution_count": 22
        }
      ],
      "source": [
        "x_ts.shape , x_val.shape , x_train.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-CgjVaugI4sj",
        "outputId": "51337e0f-fb06-4896-abe0-9e34b999f0c8"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/keras/src/layers/core/embedding.py:90: UserWarning: Argument `input_length` is deprecated. Just remove it.\n",
            "  warnings.warn(\n"
          ]
        }
      ],
      "source": [
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import Embedding, Conv1D, MaxPooling1D, LSTM, Dense, Dropout\n",
        "\n",
        "def create_cnn_lstm_attention_model():\n",
        "    model = Sequential()\n",
        "    model.add(Embedding(input_dim=MAX_NUM_WORDS, output_dim=embedding_dim, input_length=max_seq_length))\n",
        "    model.add(Conv1D(filters=128, kernel_size=5, activation='relu'))\n",
        "    model.add(MaxPooling1D(pool_size=2))\n",
        "    model.add(LSTM(64, return_sequences=True))\n",
        "    model.add(Attention())  # Adding Attention Layer here\n",
        "    model.add(Dense(128, activation='relu'))\n",
        "    model.add(Dropout(0.5))\n",
        "    model.add(Dense(num_classes, activation='sigmoid'))\n",
        "    return model\n",
        "\n",
        "cnn_lstm_attention_model = create_cnn_lstm_attention_model()\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Ff57E0NsI4sl"
      },
      "outputs": [],
      "source": [
        "import pathlib\n",
        "import shutil\n",
        "import tempfile\n",
        "\n",
        "logdir = '/content/drive/MyDrive/Colab Notebooks/propaganda/tensorboard_logs_new'\n",
        "shutil.rmtree(logdir, ignore_errors=True)\n",
        "\n",
        "checkpoint_path = \"/content/drive/MyDrive/Colab Notebooks/propaganda/training_2/cp-{epoch:04d}.weights.h5\""
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "2VwJX_t9I4sl"
      },
      "outputs": [],
      "source": [
        "\n",
        "def get_callbacks(name, steps_per_epoch):\n",
        "    return [\n",
        "        tfdocs.modeling.EpochDots(),\n",
        "        tf.keras.callbacks.EarlyStopping(monitor='val_loss', patience=50),\n",
        "        tf.keras.callbacks.TensorBoard(logdir+'/'+name),\n",
        "        tf.keras.callbacks.ModelCheckpoint(\n",
        "            filepath=checkpoint_path,\n",
        "            verbose=1,\n",
        "            save_weights_only=True,\n",
        "            save_freq=steps_per_epoch * 100  # Save every 100 epochs\n",
        "        )\n",
        "    ]\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TjMasGPoEAS2",
        "outputId": "c523fdb1-bc02-4429-ab72-0b418f8c3f23"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "18"
            ]
          },
          "metadata": {},
          "execution_count": 26
        }
      ],
      "source": [
        "len(classes)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "gns9cqVBI4sm"
      },
      "outputs": [],
      "source": [
        "import tensorflow as tf\n",
        "# Optimizer, loss function, and metrics\n",
        "\n",
        "\n",
        "# Define the learning rate\n",
        "LEARNING_RATE = 0.001  # You can adjust this value as needed\n",
        "\n",
        "optimizer = tf.keras.optimizers.Adam(learning_rate=LEARNING_RATE)\n",
        "loss = 'binary_crossentropy'\n",
        "accuracy = tf.keras.metrics.CategoricalAccuracy()\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "iw8LxNwgyx6U"
      },
      "outputs": [],
      "source": [
        "# Define training hyperparameters\n",
        "EPOCHS = 100  # Number of epochs to train the model\n",
        "BATCH_SIZE = 128  # Number of samples per gradient update\n",
        "STEPS_PER_EPOCH = len(x_train) // BATCH_SIZE  # Number of batches per epoch"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "s5jhqZtOI4sn"
      },
      "outputs": [],
      "source": [
        "# Function to compile and fit the CNN-LSTM model\n",
        "def compile_and_fit(model_name, optimizer, loss_fn, accuracy, name, epochs_num, batch_size, steps_per_epoch):\n",
        "    # Compile the model\n",
        "    model_name.compile(optimizer=optimizer,\n",
        "                       loss=loss_fn,\n",
        "                       metrics=[\n",
        "                           tf.keras.metrics.BinaryCrossentropy(\n",
        "                               from_logits=False, name='binary_crossentropy'),\n",
        "                           'accuracy'\n",
        "                       ]\n",
        "                       )\n",
        "\n",
        "    model_name.summary()  # Display the model architecture\n",
        "\n",
        "\n",
        "\n",
        "    # Make sure you're only passing token IDs for CNN-LSTM\n",
        "    history = model_name.fit(train_inputs_combined, y_train,  # Ensure train_inputs contains only token IDs\n",
        "                             validation_data=(val_inputs_combined, y_val),  # Ensure val_inputs contains only token IDs\n",
        "                             epochs=epochs_num,\n",
        "                             batch_size=batch_size,\n",
        "                             steps_per_epoch=steps_per_epoch,\n",
        "                             callbacks=get_callbacks(name, steps_per_epoch),\n",
        "                             verbose=1)\n",
        "\n",
        "    return history\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "A3U-xF8HI4sn"
      },
      "outputs": [],
      "source": [
        "size_histories = {}"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gTNGwTPcI4sf",
        "outputId": "eb2ac79c-7dfb-4706-9646-ea2be0a09efb"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Collecting git+https://github.com/tensorflow/docs\n",
            "  Cloning https://github.com/tensorflow/docs to /tmp/pip-req-build-k_m99vxz\n",
            "  Running command git clone --filter=blob:none --quiet https://github.com/tensorflow/docs /tmp/pip-req-build-k_m99vxz\n",
            "  Resolved https://github.com/tensorflow/docs to commit 15e96c3ee154d7c10ec9bb807ef44b4e6f08e65b\n",
            "  Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "Collecting astor (from tensorflow-docs==2024.10.14.18741)\n",
            "  Downloading astor-0.8.1-py2.py3-none-any.whl.metadata (4.2 kB)\n",
            "Requirement already satisfied: absl-py in /usr/local/lib/python3.10/dist-packages (from tensorflow-docs==2024.10.14.18741) (1.4.0)\n",
            "Requirement already satisfied: jinja2 in /usr/local/lib/python3.10/dist-packages (from tensorflow-docs==2024.10.14.18741) (3.1.4)\n",
            "Requirement already satisfied: nbformat in /usr/local/lib/python3.10/dist-packages (from tensorflow-docs==2024.10.14.18741) (5.10.4)\n",
            "Requirement already satisfied: protobuf>=3.12 in /usr/local/lib/python3.10/dist-packages (from tensorflow-docs==2024.10.14.18741) (3.20.3)\n",
            "Requirement already satisfied: pyyaml in /usr/local/lib/python3.10/dist-packages (from tensorflow-docs==2024.10.14.18741) (6.0.2)\n",
            "Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.10/dist-packages (from jinja2->tensorflow-docs==2024.10.14.18741) (3.0.2)\n",
            "Requirement already satisfied: fastjsonschema>=2.15 in /usr/local/lib/python3.10/dist-packages (from nbformat->tensorflow-docs==2024.10.14.18741) (2.20.0)\n",
            "Requirement already satisfied: jsonschema>=2.6 in /usr/local/lib/python3.10/dist-packages (from nbformat->tensorflow-docs==2024.10.14.18741) (4.23.0)\n",
            "Requirement already satisfied: jupyter-core!=5.0.*,>=4.12 in /usr/local/lib/python3.10/dist-packages (from nbformat->tensorflow-docs==2024.10.14.18741) (5.7.2)\n",
            "Requirement already satisfied: traitlets>=5.1 in /usr/local/lib/python3.10/dist-packages (from nbformat->tensorflow-docs==2024.10.14.18741) (5.7.1)\n",
            "Requirement already satisfied: attrs>=22.2.0 in /usr/local/lib/python3.10/dist-packages (from jsonschema>=2.6->nbformat->tensorflow-docs==2024.10.14.18741) (24.2.0)\n",
            "Requirement already satisfied: jsonschema-specifications>=2023.03.6 in /usr/local/lib/python3.10/dist-packages (from jsonschema>=2.6->nbformat->tensorflow-docs==2024.10.14.18741) (2024.10.1)\n",
            "Requirement already satisfied: referencing>=0.28.4 in /usr/local/lib/python3.10/dist-packages (from jsonschema>=2.6->nbformat->tensorflow-docs==2024.10.14.18741) (0.35.1)\n",
            "Requirement already satisfied: rpds-py>=0.7.1 in /usr/local/lib/python3.10/dist-packages (from jsonschema>=2.6->nbformat->tensorflow-docs==2024.10.14.18741) (0.20.0)\n",
            "Requirement already satisfied: platformdirs>=2.5 in /usr/local/lib/python3.10/dist-packages (from jupyter-core!=5.0.*,>=4.12->nbformat->tensorflow-docs==2024.10.14.18741) (4.3.6)\n",
            "Downloading astor-0.8.1-py2.py3-none-any.whl (27 kB)\n",
            "Building wheels for collected packages: tensorflow-docs\n",
            "  Building wheel for tensorflow-docs (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for tensorflow-docs: filename=tensorflow_docs-2024.10.14.18741-py3-none-any.whl size=182607 sha256=8595ff22c5c445c2b998673aaa54708538c546e503bdf4e39875287a17e5d32b\n",
            "  Stored in directory: /tmp/pip-ephem-wheel-cache-im3erzxp/wheels/86/0f/1e/3b62293c8ffd0fd5a49508e6871cdb7554abe9c62afd35ec53\n",
            "Successfully built tensorflow-docs\n",
            "Installing collected packages: astor, tensorflow-docs\n",
            "Successfully installed astor-0.8.1 tensorflow-docs-2024.10.14.18741\n"
          ]
        }
      ],
      "source": [
        "\n",
        "!pip install git+https://github.com/tensorflow/docs\n",
        "import tensorflow_docs as tfdocs\n",
        "import tensorflow_docs.modeling\n",
        "import tensorflow_docs.plots"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "P-ZiLjK6I4sn",
        "outputId": "a2710563-963e-4300-91a9-184853e0bc2f"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1mModel: \"sequential\"\u001b[0m\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential\"</span>\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
              "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                        \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape               \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m        Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
              "│ embedding (\u001b[38;5;33mEmbedding\u001b[0m)                │ ?                           │     \u001b[38;5;34m0\u001b[0m (unbuilt) │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ conv1d (\u001b[38;5;33mConv1D\u001b[0m)                      │ ?                           │     \u001b[38;5;34m0\u001b[0m (unbuilt) │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ max_pooling1d (\u001b[38;5;33mMaxPooling1D\u001b[0m)         │ ?                           │     \u001b[38;5;34m0\u001b[0m (unbuilt) │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ lstm (\u001b[38;5;33mLSTM\u001b[0m)                          │ ?                           │     \u001b[38;5;34m0\u001b[0m (unbuilt) │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ attention (\u001b[38;5;33mAttention\u001b[0m)                │ ?                           │     \u001b[38;5;34m0\u001b[0m (unbuilt) │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense (\u001b[38;5;33mDense\u001b[0m)                        │ ?                           │     \u001b[38;5;34m0\u001b[0m (unbuilt) │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dropout (\u001b[38;5;33mDropout\u001b[0m)                    │ ?                           │     \u001b[38;5;34m0\u001b[0m (unbuilt) │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense_1 (\u001b[38;5;33mDense\u001b[0m)                      │ ?                           │     \u001b[38;5;34m0\u001b[0m (unbuilt) │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
              "┃<span style=\"font-weight: bold\"> Layer (type)                         </span>┃<span style=\"font-weight: bold\"> Output Shape                </span>┃<span style=\"font-weight: bold\">         Param # </span>┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
              "│ embedding (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Embedding</span>)                │ ?                           │     <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (unbuilt) │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ conv1d (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv1D</span>)                      │ ?                           │     <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (unbuilt) │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ max_pooling1d (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling1D</span>)         │ ?                           │     <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (unbuilt) │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ lstm (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">LSTM</span>)                          │ ?                           │     <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (unbuilt) │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ attention (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Attention</span>)                │ ?                           │     <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (unbuilt) │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                        │ ?                           │     <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (unbuilt) │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dropout (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)                    │ ?                           │     <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (unbuilt) │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                      │ ?                           │     <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (unbuilt) │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 366ms/step - accuracy: 0.0441 - binary_crossentropy: 0.5180 - loss: 0.5180\n",
            "Epoch: 0, accuracy:0.0544,  binary_crossentropy:0.3944,  loss:0.3944,  val_accuracy:0.0884,  val_binary_crossentropy:0.2303,  val_loss:0.2303,  \n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m20s\u001b[0m 429ms/step - accuracy: 0.0444 - binary_crossentropy: 0.5139 - loss: 0.5139 - val_accuracy: 0.0884 - val_binary_crossentropy: 0.2303 - val_loss: 0.2303\n",
            "Epoch 2/100\n",
            "\u001b[1m 1/29\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m6s\u001b[0m 215ms/step - accuracy: 0.0952 - binary_crossentropy: 0.2798 - loss: 0.2798"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/lib/python3.10/contextlib.py:153: UserWarning: Your input ran out of data; interrupting training. Make sure that your dataset or generator can generate at least `steps_per_epoch * epochs` batches. You may need to use the `.repeat()` function when building your dataset.\n",
            "  self.gen.throw(typ, value, traceback)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 47ms/step - accuracy: 0.0952 - binary_crossentropy: 0.2798 - loss: 0.2798 - val_accuracy: 0.0884 - val_binary_crossentropy: 0.2299 - val_loss: 0.2299\n",
            "Epoch 3/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m18s\u001b[0m 445ms/step - accuracy: 0.0735 - binary_crossentropy: 0.2636 - loss: 0.2636 - val_accuracy: 0.0884 - val_binary_crossentropy: 0.2242 - val_loss: 0.2242\n",
            "Epoch 4/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 47ms/step - accuracy: 0.1190 - binary_crossentropy: 0.2455 - loss: 0.2455 - val_accuracy: 0.0884 - val_binary_crossentropy: 0.2241 - val_loss: 0.2241\n",
            "Epoch 5/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m19s\u001b[0m 450ms/step - accuracy: 0.0666 - binary_crossentropy: 0.2470 - loss: 0.2470 - val_accuracy: 0.1579 - val_binary_crossentropy: 0.2220 - val_loss: 0.2220\n",
            "Epoch 6/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 27ms/step - accuracy: 0.0952 - binary_crossentropy: 0.2307 - loss: 0.2307 - val_accuracy: 0.1568 - val_binary_crossentropy: 0.2219 - val_loss: 0.2219\n",
            "Epoch 7/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 483ms/step - accuracy: 0.0998 - binary_crossentropy: 0.2375 - loss: 0.2375 - val_accuracy: 0.1968 - val_binary_crossentropy: 0.2070 - val_loss: 0.2070\n",
            "Epoch 8/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 47ms/step - accuracy: 0.1310 - binary_crossentropy: 0.2171 - loss: 0.2171 - val_accuracy: 0.2400 - val_binary_crossentropy: 0.2063 - val_loss: 0.2063\n",
            "Epoch 9/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m19s\u001b[0m 434ms/step - accuracy: 0.2202 - binary_crossentropy: 0.2133 - loss: 0.2133 - val_accuracy: 0.4011 - val_binary_crossentropy: 0.1754 - val_loss: 0.1754\n",
            "Epoch 10/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 47ms/step - accuracy: 0.2738 - binary_crossentropy: 0.2032 - loss: 0.2032 - val_accuracy: 0.4189 - val_binary_crossentropy: 0.1740 - val_loss: 0.1740\n",
            "Epoch 11/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m17s\u001b[0m 413ms/step - accuracy: 0.3907 - binary_crossentropy: 0.1773 - loss: 0.1773 - val_accuracy: 0.6716 - val_binary_crossentropy: 0.1378 - val_loss: 0.1378\n",
            "Epoch 12/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 49ms/step - accuracy: 0.6548 - binary_crossentropy: 0.1242 - loss: 0.1242 - val_accuracy: 0.6737 - val_binary_crossentropy: 0.1356 - val_loss: 0.1356\n",
            "Epoch 13/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m19s\u001b[0m 438ms/step - accuracy: 0.6410 - binary_crossentropy: 0.1294 - loss: 0.1294 - val_accuracy: 0.7274 - val_binary_crossentropy: 0.1097 - val_loss: 0.1097\n",
            "Epoch 14/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 27ms/step - accuracy: 0.7024 - binary_crossentropy: 0.1110 - loss: 0.1110 - val_accuracy: 0.7274 - val_binary_crossentropy: 0.1101 - val_loss: 0.1101\n",
            "Epoch 15/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m18s\u001b[0m 392ms/step - accuracy: 0.7386 - binary_crossentropy: 0.0987 - loss: 0.0987 - val_accuracy: 0.7695 - val_binary_crossentropy: 0.0952 - val_loss: 0.0952\n",
            "Epoch 16/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 29ms/step - accuracy: 0.7381 - binary_crossentropy: 0.1002 - loss: 0.1002 - val_accuracy: 0.7716 - val_binary_crossentropy: 0.0954 - val_loss: 0.0954\n",
            "Epoch 17/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 470ms/step - accuracy: 0.8215 - binary_crossentropy: 0.0728 - loss: 0.0728 - val_accuracy: 0.7737 - val_binary_crossentropy: 0.0869 - val_loss: 0.0869\n",
            "Epoch 18/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 47ms/step - accuracy: 0.7857 - binary_crossentropy: 0.0793 - loss: 0.0793 - val_accuracy: 0.7726 - val_binary_crossentropy: 0.0866 - val_loss: 0.0866\n",
            "Epoch 19/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m18s\u001b[0m 444ms/step - accuracy: 0.8543 - binary_crossentropy: 0.0601 - loss: 0.0601 - val_accuracy: 0.7821 - val_binary_crossentropy: 0.0889 - val_loss: 0.0889\n",
            "Epoch 20/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 28ms/step - accuracy: 0.8690 - binary_crossentropy: 0.0516 - loss: 0.0516 - val_accuracy: 0.7811 - val_binary_crossentropy: 0.0874 - val_loss: 0.0874\n",
            "Epoch 21/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m19s\u001b[0m 443ms/step - accuracy: 0.8754 - binary_crossentropy: 0.0494 - loss: 0.0494 - val_accuracy: 0.7842 - val_binary_crossentropy: 0.0824 - val_loss: 0.0824\n",
            "Epoch 22/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 47ms/step - accuracy: 0.8452 - binary_crossentropy: 0.0650 - loss: 0.0650 - val_accuracy: 0.7832 - val_binary_crossentropy: 0.0826 - val_loss: 0.0826\n",
            "Epoch 23/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m18s\u001b[0m 398ms/step - accuracy: 0.8864 - binary_crossentropy: 0.0465 - loss: 0.0465 - val_accuracy: 0.7842 - val_binary_crossentropy: 0.0810 - val_loss: 0.0810\n",
            "Epoch 24/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 47ms/step - accuracy: 0.8929 - binary_crossentropy: 0.0366 - loss: 0.0366 - val_accuracy: 0.7842 - val_binary_crossentropy: 0.0817 - val_loss: 0.0817\n",
            "Epoch 25/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m20s\u001b[0m 445ms/step - accuracy: 0.8936 - binary_crossentropy: 0.0439 - loss: 0.0439 - val_accuracy: 0.7968 - val_binary_crossentropy: 0.0818 - val_loss: 0.0818\n",
            "Epoch 26/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 27ms/step - accuracy: 0.9524 - binary_crossentropy: 0.0241 - loss: 0.0241 - val_accuracy: 0.7958 - val_binary_crossentropy: 0.0821 - val_loss: 0.0821\n",
            "Epoch 27/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m20s\u001b[0m 463ms/step - accuracy: 0.9198 - binary_crossentropy: 0.0403 - loss: 0.0403 - val_accuracy: 0.7842 - val_binary_crossentropy: 0.0841 - val_loss: 0.0841\n",
            "Epoch 28/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 47ms/step - accuracy: 0.9167 - binary_crossentropy: 0.0501 - loss: 0.0501 - val_accuracy: 0.7832 - val_binary_crossentropy: 0.0840 - val_loss: 0.0840\n",
            "Epoch 29/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m13s\u001b[0m 460ms/step - accuracy: 0.9215 - binary_crossentropy: 0.0343 - loss: 0.0343 - val_accuracy: 0.7905 - val_binary_crossentropy: 0.0821 - val_loss: 0.0821\n",
            "Epoch 30/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 27ms/step - accuracy: 0.9048 - binary_crossentropy: 0.0356 - loss: 0.0356 - val_accuracy: 0.7884 - val_binary_crossentropy: 0.0820 - val_loss: 0.0820\n",
            "Epoch 31/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m17s\u001b[0m 380ms/step - accuracy: 0.9240 - binary_crossentropy: 0.0336 - loss: 0.0336 - val_accuracy: 0.8000 - val_binary_crossentropy: 0.0829 - val_loss: 0.0829\n",
            "Epoch 32/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 48ms/step - accuracy: 0.9524 - binary_crossentropy: 0.0338 - loss: 0.0338 - val_accuracy: 0.8042 - val_binary_crossentropy: 0.0830 - val_loss: 0.0830\n",
            "Epoch 33/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 443ms/step - accuracy: 0.9248 - binary_crossentropy: 0.0340 - loss: 0.0340 - val_accuracy: 0.7916 - val_binary_crossentropy: 0.0845 - val_loss: 0.0845\n",
            "Epoch 34/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 47ms/step - accuracy: 0.9524 - binary_crossentropy: 0.0333 - loss: 0.0333 - val_accuracy: 0.7937 - val_binary_crossentropy: 0.0849 - val_loss: 0.0849\n",
            "Epoch 35/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m18s\u001b[0m 396ms/step - accuracy: 0.9316 - binary_crossentropy: 0.0291 - loss: 0.0291 - val_accuracy: 0.7800 - val_binary_crossentropy: 0.0839 - val_loss: 0.0839\n",
            "Epoch 36/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 27ms/step - accuracy: 0.9167 - binary_crossentropy: 0.0364 - loss: 0.0364 - val_accuracy: 0.7789 - val_binary_crossentropy: 0.0841 - val_loss: 0.0841\n",
            "Epoch 37/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 451ms/step - accuracy: 0.9368 - binary_crossentropy: 0.0250 - loss: 0.0250 - val_accuracy: 0.7895 - val_binary_crossentropy: 0.0858 - val_loss: 0.0858\n",
            "Epoch 38/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 47ms/step - accuracy: 0.8929 - binary_crossentropy: 0.0398 - loss: 0.0398 - val_accuracy: 0.7926 - val_binary_crossentropy: 0.0860 - val_loss: 0.0860\n",
            "Epoch 39/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m19s\u001b[0m 464ms/step - accuracy: 0.9432 - binary_crossentropy: 0.0254 - loss: 0.0254 - val_accuracy: 0.7832 - val_binary_crossentropy: 0.0868 - val_loss: 0.0868\n",
            "Epoch 40/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 47ms/step - accuracy: 0.9524 - binary_crossentropy: 0.0229 - loss: 0.0229 - val_accuracy: 0.7768 - val_binary_crossentropy: 0.0876 - val_loss: 0.0876\n",
            "Epoch 41/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m17s\u001b[0m 388ms/step - accuracy: 0.9442 - binary_crossentropy: 0.0242 - loss: 0.0242 - val_accuracy: 0.7895 - val_binary_crossentropy: 0.0883 - val_loss: 0.0883\n",
            "Epoch 42/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 47ms/step - accuracy: 0.9524 - binary_crossentropy: 0.0228 - loss: 0.0228 - val_accuracy: 0.7874 - val_binary_crossentropy: 0.0889 - val_loss: 0.0889\n",
            "Epoch 43/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m20s\u001b[0m 446ms/step - accuracy: 0.9494 - binary_crossentropy: 0.0211 - loss: 0.0211 - val_accuracy: 0.7916 - val_binary_crossentropy: 0.0897 - val_loss: 0.0897\n",
            "Epoch 44/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 27ms/step - accuracy: 0.9405 - binary_crossentropy: 0.0247 - loss: 0.0247 - val_accuracy: 0.7926 - val_binary_crossentropy: 0.0899 - val_loss: 0.0899\n",
            "Epoch 45/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m19s\u001b[0m 419ms/step - accuracy: 0.9420 - binary_crossentropy: 0.0224 - loss: 0.0224 - val_accuracy: 0.7916 - val_binary_crossentropy: 0.0907 - val_loss: 0.0907\n",
            "Epoch 46/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 28ms/step - accuracy: 0.9524 - binary_crossentropy: 0.0197 - loss: 0.0197 - val_accuracy: 0.7916 - val_binary_crossentropy: 0.0905 - val_loss: 0.0905\n",
            "Epoch 47/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 465ms/step - accuracy: 0.9445 - binary_crossentropy: 0.0223 - loss: 0.0223 - val_accuracy: 0.7789 - val_binary_crossentropy: 0.0881 - val_loss: 0.0881\n",
            "Epoch 48/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 29ms/step - accuracy: 0.9286 - binary_crossentropy: 0.0157 - loss: 0.0157 - val_accuracy: 0.7832 - val_binary_crossentropy: 0.0881 - val_loss: 0.0881\n",
            "Epoch 49/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m19s\u001b[0m 448ms/step - accuracy: 0.9486 - binary_crossentropy: 0.0208 - loss: 0.0208 - val_accuracy: 0.7853 - val_binary_crossentropy: 0.0925 - val_loss: 0.0925\n",
            "Epoch 50/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 47ms/step - accuracy: 0.9405 - binary_crossentropy: 0.0372 - loss: 0.0372 - val_accuracy: 0.7853 - val_binary_crossentropy: 0.0931 - val_loss: 0.0931\n",
            "Epoch 51/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m17s\u001b[0m 378ms/step - accuracy: 0.9410 - binary_crossentropy: 0.0182 - loss: 0.0182 - val_accuracy: 0.7863 - val_binary_crossentropy: 0.0989 - val_loss: 0.0989\n",
            "Epoch 52/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 46ms/step - accuracy: 0.9524 - binary_crossentropy: 0.0303 - loss: 0.0303 - val_accuracy: 0.7842 - val_binary_crossentropy: 0.0980 - val_loss: 0.0980\n",
            "Epoch 53/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 451ms/step - accuracy: 0.9451 - binary_crossentropy: 0.0198 - loss: 0.0198 - val_accuracy: 0.7916 - val_binary_crossentropy: 0.0988 - val_loss: 0.0988\n",
            "Epoch 54/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 47ms/step - accuracy: 0.9405 - binary_crossentropy: 0.0140 - loss: 0.0140 - val_accuracy: 0.7905 - val_binary_crossentropy: 0.0985 - val_loss: 0.0985\n",
            "Epoch 55/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m18s\u001b[0m 416ms/step - accuracy: 0.9455 - binary_crossentropy: 0.0180 - loss: 0.0180 - val_accuracy: 0.7653 - val_binary_crossentropy: 0.1013 - val_loss: 0.1013\n",
            "Epoch 56/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 47ms/step - accuracy: 0.9048 - binary_crossentropy: 0.0134 - loss: 0.0134 - val_accuracy: 0.7832 - val_binary_crossentropy: 0.0984 - val_loss: 0.0984\n",
            "Epoch 57/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m12s\u001b[0m 422ms/step - accuracy: 0.9580 - binary_crossentropy: 0.0168 - loss: 0.0168 - val_accuracy: 0.7642 - val_binary_crossentropy: 0.1007 - val_loss: 0.1007\n",
            "Epoch 58/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 28ms/step - accuracy: 0.9762 - binary_crossentropy: 0.0092 - loss: 0.0092 - val_accuracy: 0.7632 - val_binary_crossentropy: 0.1017 - val_loss: 0.1017\n",
            "Epoch 59/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m19s\u001b[0m 422ms/step - accuracy: 0.9472 - binary_crossentropy: 0.0176 - loss: 0.0176 - val_accuracy: 0.7747 - val_binary_crossentropy: 0.1028 - val_loss: 0.1028\n",
            "Epoch 60/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 48ms/step - accuracy: 0.9881 - binary_crossentropy: 0.0103 - loss: 0.0103 - val_accuracy: 0.7758 - val_binary_crossentropy: 0.1028 - val_loss: 0.1028\n",
            "Epoch 61/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m19s\u001b[0m 439ms/step - accuracy: 0.9486 - binary_crossentropy: 0.0168 - loss: 0.0168 - val_accuracy: 0.7779 - val_binary_crossentropy: 0.1047 - val_loss: 0.1047\n",
            "Epoch 62/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 47ms/step - accuracy: 0.9524 - binary_crossentropy: 0.0159 - loss: 0.0159 - val_accuracy: 0.7789 - val_binary_crossentropy: 0.1045 - val_loss: 0.1045\n",
            "Epoch 63/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m17s\u001b[0m 372ms/step - accuracy: 0.9440 - binary_crossentropy: 0.0153 - loss: 0.0153 - val_accuracy: 0.7779 - val_binary_crossentropy: 0.1046 - val_loss: 0.1046\n",
            "Epoch 64/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 47ms/step - accuracy: 0.9286 - binary_crossentropy: 0.0123 - loss: 0.0123 - val_accuracy: 0.7779 - val_binary_crossentropy: 0.1053 - val_loss: 0.1053\n",
            "Epoch 65/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 448ms/step - accuracy: 0.9473 - binary_crossentropy: 0.0163 - loss: 0.0163 - val_accuracy: 0.7800 - val_binary_crossentropy: 0.1028 - val_loss: 0.1028\n",
            "Epoch 66/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 47ms/step - accuracy: 0.9048 - binary_crossentropy: 0.0105 - loss: 0.0105 - val_accuracy: 0.7811 - val_binary_crossentropy: 0.1028 - val_loss: 0.1028\n",
            "Epoch 67/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m19s\u001b[0m 425ms/step - accuracy: 0.9574 - binary_crossentropy: 0.0126 - loss: 0.0126 - val_accuracy: 0.7821 - val_binary_crossentropy: 0.1013 - val_loss: 0.1013\n",
            "Epoch 68/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 29ms/step - accuracy: 0.9286 - binary_crossentropy: 0.0172 - loss: 0.0172 - val_accuracy: 0.7821 - val_binary_crossentropy: 0.1016 - val_loss: 0.1016\n",
            "Epoch 69/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m20s\u001b[0m 463ms/step - accuracy: 0.9478 - binary_crossentropy: 0.0126 - loss: 0.0126 - val_accuracy: 0.7726 - val_binary_crossentropy: 0.1104 - val_loss: 0.1104\n",
            "Epoch 70/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 29ms/step - accuracy: 0.9405 - binary_crossentropy: 0.0094 - loss: 0.0094 - val_accuracy: 0.7726 - val_binary_crossentropy: 0.1111 - val_loss: 0.1111\n",
            "Epoch 71/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 568ms/step - accuracy: 0.9499 - binary_crossentropy: 0.0154 - loss: 0.0154 - val_accuracy: 0.7821 - val_binary_crossentropy: 0.1120 - val_loss: 0.1120\n",
            "Epoch 72/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 29ms/step - accuracy: 0.9643 - binary_crossentropy: 0.0205 - loss: 0.0205 - val_accuracy: 0.7832 - val_binary_crossentropy: 0.1125 - val_loss: 0.1125\n",
            "Epoch 73/100\n",
            "\u001b[1m29/29\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m17s\u001b[0m 471ms/step - accuracy: 0.9529 - binary_crossentropy: 0.0123 - loss: 0.0123 - val_accuracy: 0.7842 - val_binary_crossentropy: 0.1113 - val_loss: 0.1113\n"
          ]
        }
      ],
      "source": [
        "# Compile and train the CNN-LSTM model\n",
        "size_histories['ORIGINAL'] = compile_and_fit(cnn_lstm_attention_model, optimizer, loss, accuracy, 'ORIGINAL', EPOCHS, BATCH_SIZE, STEPS_PER_EPOCH)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_I6NJVbXekEN"
      },
      "source": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 449
        },
        "id": "i5bclLODI4so",
        "outputId": "5db8ef0a-618e-4d5e-c4be-e99368bdc8c4"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "\n",
        "plotter = tfdocs.plots.HistoryPlotter(metric = 'binary_crossentropy', smoothing_std=10)\n",
        "plotter.plot(size_histories)\n",
        "\n",
        "#plt.ylim([-1, 2])\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Yd4Eqq27I4sp",
        "outputId": "a31c81aa-1c59-4134-b70d-7c9bfbea5b06"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 496ms/step\n"
          ]
        }
      ],
      "source": [
        "# Assuming you have defined your test inputs as test_inputs and your true labels as y_ts\n",
        "test_scores = cnn_lstm_attention_model.predict(test_inputs_combined)  # Use cnn_lstm_model instead of model\n",
        "test_pred_labels = (test_scores >= 0.5).astype(int)\n",
        "test_true_labels = y_ts.astype(int)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "huszOXpgI4sq",
        "outputId": "fcfc118f-005c-43a8-8e71-09440d37bef1"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Classification Report:\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.95      1.00      0.97       827\n",
            "           1       0.97      0.67      0.79       127\n",
            "\n",
            "    accuracy                           0.95       954\n",
            "   macro avg       0.96      0.83      0.88       954\n",
            "weighted avg       0.95      0.95      0.95       954\n",
            "\n",
            "F1 macro score: 0.566141903027149\n",
            "F1 micro score: 0.7906976744186046\n",
            "Confusion Matrix:\n",
            "[[824   3]\n",
            " [ 42  85]]\n",
            "Accuracy: 0.9528301886792453\n",
            "Recall: 0.8328318845271306\n"
          ]
        }
      ],
      "source": [
        "from sklearn.metrics import f1_score, confusion_matrix, accuracy_score, recall_score,classification_report\n",
        "\n",
        "# Calculate F1 score\n",
        "f1_macro = f1_score(test_true_labels, test_pred_labels, average='macro')\n",
        "f1_micro = f1_score(test_true_labels, test_pred_labels, average='micro')\n",
        "\n",
        "# Calculate confusion matrix, accuracy, and recall\n",
        "confusion_mat = confusion_matrix(test_true_labels.flatten(), test_pred_labels.flatten())\n",
        "accuracy = accuracy_score(test_true_labels.flatten(), test_pred_labels.flatten())\n",
        "recall = recall_score(test_true_labels.flatten(), test_pred_labels.flatten(), average='macro')\n",
        "classification_rep = classification_report(test_true_labels.flatten(), test_pred_labels.flatten())\n",
        "\n",
        "from sklearn import metrics\n",
        "\n",
        "print(\"Classification Report:\")\n",
        "print(classification_rep)\n",
        "print(\"F1 macro score:\", f1_macro)\n",
        "print(\"F1 micro score:\", f1_micro)\n",
        "print(\"Confusion Matrix:\")\n",
        "print(confusion_mat)\n",
        "print(\"Accuracy:\", accuracy)\n",
        "print(\"Recall:\", recall)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "AIePfeDqqTPM",
        "outputId": "c427cdcb-41c0-49e7-e294-78357c15a569"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                                 Text  \\\n",
              "0   أربعة قتلى خلال تبادل لإطلاق النار بين الشرطة ...   \n",
              "1   أول دولة أوروبية تصنف الاستيطان الإسرائيلي كاح...   \n",
              "2                 إفريقيا من منظمة الوحدة إلى الاتحاد   \n",
              "3   إيطاليا ارتفاع حصيلة ضحايا حادث تحطم قاطرة تلف...   \n",
              "4   الأمم المتحدة السيد هلال يندد ب الارتباط الإيد...   \n",
              "5          الاقتصاد الأخضر وتحدياته محور ندوة بالرباط   \n",
              "6   الباحث السياسي عز الدين عقيل مجموعة فاغنر التي...   \n",
              "7   الحملة الإعلامية الإسبانية ضد المغرب الحديث عن...   \n",
              "8   الحوار الاجتماعي الحكومة أوفت بكل التزاماتها ر...   \n",
              "9   السعيدية فتح بحث قضائي لتحديد المتورطين في محا...   \n",
              "10  الصحراء المغربية أي تصريح من قبل عضو في الحكوم...   \n",
              "11  الصحراء المغربية عمدة ليفيري الإيطالية يدعو ال...   \n",
              "12  الصحفي المتخصص في شؤون الجماعات المتطرفة ماهر ...   \n",
              "13  الفريق شنقريحة يؤكد حرص الجيش الوطني الشعبي عل...   \n",
              "14  القيادة العسكرية الأمريكية في إفريقيا أفريكوم ...   \n",
              "15  الكتيبة مشاة التابعة لرئاسة الأركان العامة تلق...   \n",
              "16  المجلس الوطني لحقوق الانسان بقاء الصحراء الغرب...   \n",
              "17  المدير التنفيذي لشركة غازبروم الروسية سيرجي تو...   \n",
              "18  المغرب بلد موثوق به يأخذ مكافحة الهجرة غير الش...   \n",
              "19  الناظور إحباط محاولة للتهريب الدولي للمخدرات و...   \n",
              "20  النموذج التنموي الجديد المحاور الاستراتيجية لل...   \n",
              "21  بالتواطؤ مع شبكات إجرامية مغربية تفكيك شبكة إج...   \n",
              "22  بعد قرابة عقد من الزمان على أحداث بنغازي هل ست...   \n",
              "23  بعيدا عن المغالطات التي تروجها وسائل إعلام إسب...   \n",
              "24  بوليساريو يروج زيفا لصور من ليبيا تدعي أنها لط...   \n",
              "25  تحذيرات من حرب جديدة في حال فشل الانتخابات الق...   \n",
              "26  تونس مقتل ستة عناصر من الدرك على يد مجموعة إره...   \n",
              "27  حملت القبيلة قيادة قوات حفتر مسؤولية الانفلات ...   \n",
              "28  ديوان المحاسبة الليبي يوقف رئيس المؤسسة الليبي...   \n",
              "29  رئيس بعثة الأمم المتحدة للدعم في ليبيا يان كوب...   \n",
              "30  شاهد المتحدث باسم مليشيات حفتر أحمد المسماري ي...   \n",
              "31  شاهد الناشط القبلي أسامة بوحلاق يهاجم تنظيم ال...   \n",
              "32  شاهد بالتزامن مع ذكرى نكبة الكرامة منذ سنوات ا...   \n",
              "33  شاهد برتبة عقيد صدام ابن آمر المليشيات خليفة ح...   \n",
              "34  شاهد بعد فشله عسكريا في السيطرة على العاصمة طر...   \n",
              "35  شاهد في بيان لشباب ثورة الفقراء بمدينة سبها يط...   \n",
              "36  شاهد فيديو جديد لأرتال مسلحة تابعة لمليشيات حف...   \n",
              "37  شاهد مرتزق يوثق قدومه مع أرتال عسكرية قادمة إل...   \n",
              "38  شاهد وزير الصحة بحكومة الوحدة الوطنية علي الزن...   \n",
              "39  شمام يطالب شباب ونساء برقة أن يغيروا بأيديهم ع...   \n",
              "40  صور إدارة خدمات الطوارئ بشركة الخدمات العامة ت...   \n",
              "41  صور مستعرضا قوته في بنغازي قبل أيام من مرور ذك...   \n",
              "42  عبد الرحمن الشاطر صدق أو لا تصدق كل المترشحين ...   \n",
              "43  غوتيريش المقاتلون الأجانب لا يزالون ينشطون في ...   \n",
              "44  فوق السلطة الإمارات تهنىء إيران برمضان وإسرائي...   \n",
              "45  فيديو القيادي في قوات حفتر محمود الورفلي يقتحم...   \n",
              "46  كارثة بيئية وصناعية بأستراليا ملايين الفئران ت...   \n",
              "47  مالي الإفراج عن رئيسي المرحلة الانتقالية والحكومة   \n",
              "48  مدينة شحات الأثرية إحدى المدن الليبية الواقعة ...   \n",
              "49  مصادر إسبانية تكشف أن زعيم جبهة البوليساريو ال...   \n",
              "50  مصادر إعلامية مجموعة من المسلحين الملثمين يقتح...   \n",
              "51  ملتقى افريقيا للاستثمار دعوة الجزائر لقيادة ال...   \n",
              "52  وزير الخارجية الأميركي العنف في منطقة الشرق ال...   \n",
              "\n",
              "                                        Actual Labels  \\\n",
              "0              Loaded Language, Name calling/Labeling   \n",
              "1   Name calling/Labeling, Glittering generalities...   \n",
              "2                                             Slogans   \n",
              "3                                        no technique   \n",
              "4   Smears, Name calling/Labeling, Obfuscation, In...   \n",
              "5                               Name calling/Labeling   \n",
              "6   Loaded Language, Appeal to fear/prejudice, App...   \n",
              "7   Name calling/Labeling, Glittering generalities...   \n",
              "8   Exaggeration/Minimisation, Obfuscation, Intent...   \n",
              "9              Loaded Language, Name calling/Labeling   \n",
              "10                                             Smears   \n",
              "11                                       no technique   \n",
              "12  Exaggeration/Minimisation, Loaded Language, Na...   \n",
              "13  Name calling/Labeling, Flag-waving, Glittering...   \n",
              "14                      Smears, Name calling/Labeling   \n",
              "15             Loaded Language, Name calling/Labeling   \n",
              "16  Exaggeration/Minimisation, Smears, Loaded Lang...   \n",
              "17                                       no technique   \n",
              "18  Exaggeration/Minimisation, Name calling/Labeli...   \n",
              "19                                    Loaded Language   \n",
              "20                              Name calling/Labeling   \n",
              "21  Exaggeration/Minimisation, Loaded Language, Na...   \n",
              "22                                       no technique   \n",
              "23  Smears, Loaded Language, Name calling/Labeling...   \n",
              "24                      Smears, Name calling/Labeling   \n",
              "25          Loaded Language, Appeal to fear/prejudice   \n",
              "26             Loaded Language, Name calling/Labeling   \n",
              "27  Exaggeration/Minimisation, Loaded Language, Na...   \n",
              "28                            Smears, Loaded Language   \n",
              "29                                       no technique   \n",
              "30  Loaded Language, Name calling/Labeling, Appeal...   \n",
              "31     Smears, Loaded Language, Name calling/Labeling   \n",
              "32  Exaggeration/Minimisation, Loaded Language, Na...   \n",
              "33     Smears, Loaded Language, Name calling/Labeling   \n",
              "34  Loaded Language, Appeal to fear/prejudice, Nam...   \n",
              "35  Loaded Language, Appeal to fear/prejudice, Nam...   \n",
              "36  Loaded Language, Appeal to fear/prejudice, Nam...   \n",
              "37  Loaded Language, Appeal to fear/prejudice, Nam...   \n",
              "38         Name calling/Labeling, Appeal to authority   \n",
              "39  Obfuscation, Intentional vagueness, Confusion,...   \n",
              "40                              Name calling/Labeling   \n",
              "41             Loaded Language, Name calling/Labeling   \n",
              "42  Exaggeration/Minimisation, Smears, Loaded Lang...   \n",
              "43         Name calling/Labeling, Appeal to authority   \n",
              "44              Smears, Loaded Language, Whataboutism   \n",
              "45     Smears, Loaded Language, Name calling/Labeling   \n",
              "46  Exaggeration/Minimisation, Loaded Language, Ap...   \n",
              "47                                       no technique   \n",
              "48  Exaggeration/Minimisation, Name calling/Labeli...   \n",
              "49             Loaded Language, Name calling/Labeling   \n",
              "50             Loaded Language, Name calling/Labeling   \n",
              "51                                       no technique   \n",
              "52           Presenting Irrelevant Data (Red Herring)   \n",
              "\n",
              "                                     Predicted Labels  \n",
              "0              Loaded Language, Name calling/Labeling  \n",
              "1   Name calling/Labeling, Glittering generalities...  \n",
              "2                                             Slogans  \n",
              "3                                        no technique  \n",
              "4       Obfuscation, Intentional vagueness, Confusion  \n",
              "5                               Name calling/Labeling  \n",
              "6                                 Appeal to authority  \n",
              "7   Name calling/Labeling, Glittering generalities...  \n",
              "8                    Glittering generalities (Virtue)  \n",
              "9                                                      \n",
              "10                                             Smears  \n",
              "11                                       no technique  \n",
              "12                                                     \n",
              "13         Name calling/Labeling, Appeal to authority  \n",
              "14                      Smears, Name calling/Labeling  \n",
              "15                                                     \n",
              "16  Exaggeration/Minimisation, Smears, Loaded Lang...  \n",
              "17                                       no technique  \n",
              "18                                                     \n",
              "19                                    Loaded Language  \n",
              "20                                                     \n",
              "21                                                     \n",
              "22                                       no technique  \n",
              "23  Loaded Language, Name calling/Labeling, Glitte...  \n",
              "24                      Smears, Name calling/Labeling  \n",
              "25          Loaded Language, Appeal to fear/prejudice  \n",
              "26             Loaded Language, Name calling/Labeling  \n",
              "27  Exaggeration/Minimisation, Loaded Language, Na...  \n",
              "28                   Glittering generalities (Virtue)  \n",
              "29                                       no technique  \n",
              "30  Loaded Language, Name calling/Labeling, Appeal...  \n",
              "31     Smears, Loaded Language, Name calling/Labeling  \n",
              "32  Exaggeration/Minimisation, Loaded Language, Na...  \n",
              "33     Smears, Loaded Language, Name calling/Labeling  \n",
              "34  Loaded Language, Appeal to fear/prejudice, Nam...  \n",
              "35  Loaded Language, Appeal to fear/prejudice, Nam...  \n",
              "36  Loaded Language, Appeal to fear/prejudice, Nam...  \n",
              "37                   Glittering generalities (Virtue)  \n",
              "38         Name calling/Labeling, Appeal to authority  \n",
              "39      Obfuscation, Intentional vagueness, Confusion  \n",
              "40                              Name calling/Labeling  \n",
              "41             Loaded Language, Name calling/Labeling  \n",
              "42  Exaggeration/Minimisation, Smears, Loaded Lang...  \n",
              "43         Name calling/Labeling, Appeal to authority  \n",
              "44                                       Whataboutism  \n",
              "45     Smears, Loaded Language, Name calling/Labeling  \n",
              "46  Exaggeration/Minimisation, Loaded Language, Ap...  \n",
              "47                                       no technique  \n",
              "48                                                     \n",
              "49             Loaded Language, Name calling/Labeling  \n",
              "50             Loaded Language, Name calling/Labeling  \n",
              "51                                       no technique  \n",
              "52                   Glittering generalities (Virtue)  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-c43ad87c-7add-4a05-b507-e52dbf88ddc2\" 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>Text</th>\n",
              "      <th>Actual Labels</th>\n",
              "      <th>Predicted Labels</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>أربعة قتلى خلال تبادل لإطلاق النار بين الشرطة ...</td>\n",
              "      <td>Loaded Language, Name calling/Labeling</td>\n",
              "      <td>Loaded Language, Name calling/Labeling</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>أول دولة أوروبية تصنف الاستيطان الإسرائيلي كاح...</td>\n",
              "      <td>Name calling/Labeling, Glittering generalities...</td>\n",
              "      <td>Name calling/Labeling, Glittering generalities...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>إفريقيا من منظمة الوحدة إلى الاتحاد</td>\n",
              "      <td>Slogans</td>\n",
              "      <td>Slogans</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>إيطاليا ارتفاع حصيلة ضحايا حادث تحطم قاطرة تلف...</td>\n",
              "      <td>no technique</td>\n",
              "      <td>no technique</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>الأمم المتحدة السيد هلال يندد ب الارتباط الإيد...</td>\n",
              "      <td>Smears, Name calling/Labeling, Obfuscation, In...</td>\n",
              "      <td>Obfuscation, Intentional vagueness, Confusion</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>الاقتصاد الأخضر وتحدياته محور ندوة بالرباط</td>\n",
              "      <td>Name calling/Labeling</td>\n",
              "      <td>Name calling/Labeling</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>الباحث السياسي عز الدين عقيل مجموعة فاغنر التي...</td>\n",
              "      <td>Loaded Language, Appeal to fear/prejudice, App...</td>\n",
              "      <td>Appeal to authority</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>الحملة الإعلامية الإسبانية ضد المغرب الحديث عن...</td>\n",
              "      <td>Name calling/Labeling, Glittering generalities...</td>\n",
              "      <td>Name calling/Labeling, Glittering generalities...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>الحوار الاجتماعي الحكومة أوفت بكل التزاماتها ر...</td>\n",
              "      <td>Exaggeration/Minimisation, Obfuscation, Intent...</td>\n",
              "      <td>Glittering generalities (Virtue)</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>السعيدية فتح بحث قضائي لتحديد المتورطين في محا...</td>\n",
              "      <td>Loaded Language, Name calling/Labeling</td>\n",
              "      <td></td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>الصحراء المغربية أي تصريح من قبل عضو في الحكوم...</td>\n",
              "      <td>Smears</td>\n",
              "      <td>Smears</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>الصحراء المغربية عمدة ليفيري الإيطالية يدعو ال...</td>\n",
              "      <td>no technique</td>\n",
              "      <td>no technique</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>الصحفي المتخصص في شؤون الجماعات المتطرفة ماهر ...</td>\n",
              "      <td>Exaggeration/Minimisation, Loaded Language, Na...</td>\n",
              "      <td></td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>الفريق شنقريحة يؤكد حرص الجيش الوطني الشعبي عل...</td>\n",
              "      <td>Name calling/Labeling, Flag-waving, Glittering...</td>\n",
              "      <td>Name calling/Labeling, Appeal to authority</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>القيادة العسكرية الأمريكية في إفريقيا أفريكوم ...</td>\n",
              "      <td>Smears, Name calling/Labeling</td>\n",
              "      <td>Smears, Name calling/Labeling</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15</th>\n",
              "      <td>الكتيبة مشاة التابعة لرئاسة الأركان العامة تلق...</td>\n",
              "      <td>Loaded Language, Name calling/Labeling</td>\n",
              "      <td></td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16</th>\n",
              "      <td>المجلس الوطني لحقوق الانسان بقاء الصحراء الغرب...</td>\n",
              "      <td>Exaggeration/Minimisation, Smears, Loaded Lang...</td>\n",
              "      <td>Exaggeration/Minimisation, Smears, Loaded Lang...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>17</th>\n",
              "      <td>المدير التنفيذي لشركة غازبروم الروسية سيرجي تو...</td>\n",
              "      <td>no technique</td>\n",
              "      <td>no technique</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18</th>\n",
              "      <td>المغرب بلد موثوق به يأخذ مكافحة الهجرة غير الش...</td>\n",
              "      <td>Exaggeration/Minimisation, Name calling/Labeli...</td>\n",
              "      <td></td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>19</th>\n",
              "      <td>الناظور إحباط محاولة للتهريب الدولي للمخدرات و...</td>\n",
              "      <td>Loaded Language</td>\n",
              "      <td>Loaded Language</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>20</th>\n",
              "      <td>النموذج التنموي الجديد المحاور الاستراتيجية لل...</td>\n",
              "      <td>Name calling/Labeling</td>\n",
              "      <td></td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>21</th>\n",
              "      <td>بالتواطؤ مع شبكات إجرامية مغربية تفكيك شبكة إج...</td>\n",
              "      <td>Exaggeration/Minimisation, Loaded Language, Na...</td>\n",
              "      <td></td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>22</th>\n",
              "      <td>بعد قرابة عقد من الزمان على أحداث بنغازي هل ست...</td>\n",
              "      <td>no technique</td>\n",
              "      <td>no technique</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23</th>\n",
              "      <td>بعيدا عن المغالطات التي تروجها وسائل إعلام إسب...</td>\n",
              "      <td>Smears, Loaded Language, Name calling/Labeling...</td>\n",
              "      <td>Loaded Language, Name calling/Labeling, Glitte...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24</th>\n",
              "      <td>بوليساريو يروج زيفا لصور من ليبيا تدعي أنها لط...</td>\n",
              "      <td>Smears, Name calling/Labeling</td>\n",
              "      <td>Smears, Name calling/Labeling</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25</th>\n",
              "      <td>تحذيرات من حرب جديدة في حال فشل الانتخابات الق...</td>\n",
              "      <td>Loaded Language, Appeal to fear/prejudice</td>\n",
              "      <td>Loaded Language, Appeal to fear/prejudice</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>26</th>\n",
              "      <td>تونس مقتل ستة عناصر من الدرك على يد مجموعة إره...</td>\n",
              "      <td>Loaded Language, Name calling/Labeling</td>\n",
              "      <td>Loaded Language, Name calling/Labeling</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>27</th>\n",
              "      <td>حملت القبيلة قيادة قوات حفتر مسؤولية الانفلات ...</td>\n",
              "      <td>Exaggeration/Minimisation, Loaded Language, Na...</td>\n",
              "      <td>Exaggeration/Minimisation, Loaded Language, Na...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>28</th>\n",
              "      <td>ديوان المحاسبة الليبي يوقف رئيس المؤسسة الليبي...</td>\n",
              "      <td>Smears, Loaded Language</td>\n",
              "      <td>Glittering generalities (Virtue)</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>29</th>\n",
              "      <td>رئيس بعثة الأمم المتحدة للدعم في ليبيا يان كوب...</td>\n",
              "      <td>no technique</td>\n",
              "      <td>no technique</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>30</th>\n",
              "      <td>شاهد المتحدث باسم مليشيات حفتر أحمد المسماري ي...</td>\n",
              "      <td>Loaded Language, Name calling/Labeling, Appeal...</td>\n",
              "      <td>Loaded Language, Name calling/Labeling, Appeal...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>31</th>\n",
              "      <td>شاهد الناشط القبلي أسامة بوحلاق يهاجم تنظيم ال...</td>\n",
              "      <td>Smears, Loaded Language, Name calling/Labeling</td>\n",
              "      <td>Smears, Loaded Language, Name calling/Labeling</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>32</th>\n",
              "      <td>شاهد بالتزامن مع ذكرى نكبة الكرامة منذ سنوات ا...</td>\n",
              "      <td>Exaggeration/Minimisation, Loaded Language, Na...</td>\n",
              "      <td>Exaggeration/Minimisation, Loaded Language, Na...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>33</th>\n",
              "      <td>شاهد برتبة عقيد صدام ابن آمر المليشيات خليفة ح...</td>\n",
              "      <td>Smears, Loaded Language, Name calling/Labeling</td>\n",
              "      <td>Smears, Loaded Language, Name calling/Labeling</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>34</th>\n",
              "      <td>شاهد بعد فشله عسكريا في السيطرة على العاصمة طر...</td>\n",
              "      <td>Loaded Language, Appeal to fear/prejudice, Nam...</td>\n",
              "      <td>Loaded Language, Appeal to fear/prejudice, Nam...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35</th>\n",
              "      <td>شاهد في بيان لشباب ثورة الفقراء بمدينة سبها يط...</td>\n",
              "      <td>Loaded Language, Appeal to fear/prejudice, Nam...</td>\n",
              "      <td>Loaded Language, Appeal to fear/prejudice, Nam...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>36</th>\n",
              "      <td>شاهد فيديو جديد لأرتال مسلحة تابعة لمليشيات حف...</td>\n",
              "      <td>Loaded Language, Appeal to fear/prejudice, Nam...</td>\n",
              "      <td>Loaded Language, Appeal to fear/prejudice, Nam...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37</th>\n",
              "      <td>شاهد مرتزق يوثق قدومه مع أرتال عسكرية قادمة إل...</td>\n",
              "      <td>Loaded Language, Appeal to fear/prejudice, Nam...</td>\n",
              "      <td>Glittering generalities (Virtue)</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>38</th>\n",
              "      <td>شاهد وزير الصحة بحكومة الوحدة الوطنية علي الزن...</td>\n",
              "      <td>Name calling/Labeling, Appeal to authority</td>\n",
              "      <td>Name calling/Labeling, Appeal to authority</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>39</th>\n",
              "      <td>شمام يطالب شباب ونساء برقة أن يغيروا بأيديهم ع...</td>\n",
              "      <td>Obfuscation, Intentional vagueness, Confusion,...</td>\n",
              "      <td>Obfuscation, Intentional vagueness, Confusion</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>40</th>\n",
              "      <td>صور إدارة خدمات الطوارئ بشركة الخدمات العامة ت...</td>\n",
              "      <td>Name calling/Labeling</td>\n",
              "      <td>Name calling/Labeling</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>41</th>\n",
              "      <td>صور مستعرضا قوته في بنغازي قبل أيام من مرور ذك...</td>\n",
              "      <td>Loaded Language, Name calling/Labeling</td>\n",
              "      <td>Loaded Language, Name calling/Labeling</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>42</th>\n",
              "      <td>عبد الرحمن الشاطر صدق أو لا تصدق كل المترشحين ...</td>\n",
              "      <td>Exaggeration/Minimisation, Smears, Loaded Lang...</td>\n",
              "      <td>Exaggeration/Minimisation, Smears, Loaded Lang...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>43</th>\n",
              "      <td>غوتيريش المقاتلون الأجانب لا يزالون ينشطون في ...</td>\n",
              "      <td>Name calling/Labeling, Appeal to authority</td>\n",
              "      <td>Name calling/Labeling, Appeal to authority</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>44</th>\n",
              "      <td>فوق السلطة الإمارات تهنىء إيران برمضان وإسرائي...</td>\n",
              "      <td>Smears, Loaded Language, Whataboutism</td>\n",
              "      <td>Whataboutism</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>45</th>\n",
              "      <td>فيديو القيادي في قوات حفتر محمود الورفلي يقتحم...</td>\n",
              "      <td>Smears, Loaded Language, Name calling/Labeling</td>\n",
              "      <td>Smears, Loaded Language, Name calling/Labeling</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>46</th>\n",
              "      <td>كارثة بيئية وصناعية بأستراليا ملايين الفئران ت...</td>\n",
              "      <td>Exaggeration/Minimisation, Loaded Language, Ap...</td>\n",
              "      <td>Exaggeration/Minimisation, Loaded Language, Ap...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>47</th>\n",
              "      <td>مالي الإفراج عن رئيسي المرحلة الانتقالية والحكومة</td>\n",
              "      <td>no technique</td>\n",
              "      <td>no technique</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>48</th>\n",
              "      <td>مدينة شحات الأثرية إحدى المدن الليبية الواقعة ...</td>\n",
              "      <td>Exaggeration/Minimisation, Name calling/Labeli...</td>\n",
              "      <td></td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>49</th>\n",
              "      <td>مصادر إسبانية تكشف أن زعيم جبهة البوليساريو ال...</td>\n",
              "      <td>Loaded Language, Name calling/Labeling</td>\n",
              "      <td>Loaded Language, Name calling/Labeling</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50</th>\n",
              "      <td>مصادر إعلامية مجموعة من المسلحين الملثمين يقتح...</td>\n",
              "      <td>Loaded Language, Name calling/Labeling</td>\n",
              "      <td>Loaded Language, Name calling/Labeling</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>51</th>\n",
              "      <td>ملتقى افريقيا للاستثمار دعوة الجزائر لقيادة ال...</td>\n",
              "      <td>no technique</td>\n",
              "      <td>no technique</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>52</th>\n",
              "      <td>وزير الخارجية الأميركي العنف في منطقة الشرق ال...</td>\n",
              "      <td>Presenting Irrelevant Data (Red Herring)</td>\n",
              "      <td>Glittering generalities (Virtue)</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-c43ad87c-7add-4a05-b507-e52dbf88ddc2')\"\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 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\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",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\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-c43ad87c-7add-4a05-b507-e52dbf88ddc2 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-c43ad87c-7add-4a05-b507-e52dbf88ddc2');\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",
              "\n",
              "\n",
              "<div id=\"df-d5367714-a5fd-45b3-9437-8a95d4cbe933\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-d5367714-a5fd-45b3-9437-8a95d4cbe933')\"\n",
              "            title=\"Suggest charts\"\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",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-d5367714-a5fd-45b3-9437-8a95d4cbe933 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "  <div id=\"id_8be0a83d-a68d-41bc-8e33-377fdefcc318\">\n",
              "    <style>\n",
              "      .colab-df-generate {\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-generate: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-generate {\n",
              "        background-color: #3B4455;\n",
              "        fill: #D2E3FC;\n",
              "      }\n",
              "\n",
              "      [theme=dark] .colab-df-generate: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",
              "    <button class=\"colab-df-generate\" onclick=\"generateWithVariable('results_df')\"\n",
              "            title=\"Generate code using this dataframe.\"\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=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "    <script>\n",
              "      (() => {\n",
              "      const buttonEl =\n",
              "        document.querySelector('#id_8be0a83d-a68d-41bc-8e33-377fdefcc318 button.colab-df-generate');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      buttonEl.onclick = () => {\n",
              "        google.colab.notebook.generateWithVariable('results_df');\n",
              "      }\n",
              "      })();\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "results_df",
              "summary": "{\n  \"name\": \"results_df\",\n  \"rows\": 53,\n  \"fields\": [\n    {\n      \"column\": \"Text\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 53,\n        \"samples\": [\n          \"\\u0627\\u0644\\u0646\\u0627\\u0638\\u0648\\u0631 \\u0625\\u062d\\u0628\\u0627\\u0637 \\u0645\\u062d\\u0627\\u0648\\u0644\\u0629 \\u0644\\u0644\\u062a\\u0647\\u0631\\u064a\\u0628 \\u0627\\u0644\\u062f\\u0648\\u0644\\u064a \\u0644\\u0644\\u0645\\u062e\\u062f\\u0631\\u0627\\u062a \\u0648\\u062d\\u062c\\u0632 \\u062e\\u0645\\u0633\\u0629 \\u0623\\u0637\\u0646\\u0627\\u0646 \\u0648 \\u0643\\u0644\\u063a \\u0645\\u0646 \\u0645\\u062e\\u062f\\u0631 \\u0627\\u0644\\u0634\\u064a\\u0631\\u0627 \\u0627\\u0644\\u0645\\u062f\\u064a\\u0631\\u064a\\u0629 \\u0627\\u0644\\u0639\\u0627\\u0645\\u0629 \\u0644\\u0644\\u0623\\u0645\\u0646 \\u0627\\u0644\\u0648\\u0637\\u0646\\u064a\",\n          \"\\u0635\\u0648\\u0631 \\u0645\\u0633\\u062a\\u0639\\u0631\\u0636\\u0627 \\u0642\\u0648\\u062a\\u0647 \\u0641\\u064a \\u0628\\u0646\\u063a\\u0627\\u0632\\u064a \\u0642\\u0628\\u0644 \\u0623\\u064a\\u0627\\u0645 \\u0645\\u0646 \\u0645\\u0631\\u0648\\u0631 \\u0630\\u0643\\u0631\\u0649 \\u0647\\u0632\\u064a\\u0645\\u062a\\u0647 \\u0641\\u064a \\u0637\\u0631\\u0627\\u0628\\u0644\\u0633 \\u062d\\u0641\\u062a\\u0631 \\u064a\\u062f\\u0639\\u0648 \\u0643\\u0644\\u0627 \\u0645\\u0646 \\u0631\\u0626\\u064a\\u0633 \\u062d\\u0643\\u0648\\u0645\\u0629 \\u0627\\u0644\\u0648\\u062d\\u062f\\u0629 \\u0627\\u0644\\u0648\\u0637\\u0646\\u064a\\u0629 \\u0639\\u0628\\u062f\\u0627\\u0644\\u062d\\u0645\\u064a\\u062f \\u0627\\u0644\\u062f\\u0628\\u064a\\u0628\\u0629 \\u0648\\u0631\\u0626\\u064a\\u0633 \\u0627\\u0644\\u0645\\u062c\\u0644\\u0633 \\u0627\\u0644\\u0631\\u0626\\u0627\\u0633\\u064a \\u0645\\u062d\\u0645\\u062f \\u0627\\u0644\\u0645\\u0646\\u0641\\u064a \\u0648\\u0631\\u0626\\u064a\\u0633 \\u0627\\u0644\\u0628\\u0631\\u0644\\u0645\\u0627\\u0646 \\u0639\\u0642\\u064a\\u0644\\u0629 \\u0635\\u0627\\u0644\\u062d \\u0644\\u062d\\u0636\\u0648\\u0631 \\u0639\\u0631\\u0636 \\u0639\\u0633\\u0643\\u0631\\u064a \\u0644\\u0645\\u0644\\u064a\\u0634\\u064a\\u0627\\u062a\\u0647 \\u0628\\u0645\\u0646\\u0627\\u0633\\u0628\\u0629 \\u0630\\u0643\\u0631\\u0649 \\u0646\\u0643\\u0628\\u0629 \\u0627\\u0644\\u0643\\u0631\\u0627\\u0645\\u0629\",\n          \"\\u0645\\u0627\\u0644\\u064a \\u0627\\u0644\\u0625\\u0641\\u0631\\u0627\\u062c \\u0639\\u0646 \\u0631\\u0626\\u064a\\u0633\\u064a \\u0627\\u0644\\u0645\\u0631\\u062d\\u0644\\u0629 \\u0627\\u0644\\u0627\\u0646\\u062a\\u0642\\u0627\\u0644\\u064a\\u0629 \\u0648\\u0627\\u0644\\u062d\\u0643\\u0648\\u0645\\u0629\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Actual Labels\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 31,\n        \"samples\": [\n          \"Smears, Loaded Language, Whataboutism\",\n          \"Exaggeration/Minimisation, Loaded Language, Name calling/Labeling, Repetition\",\n          \"Loaded Language, Appeal to fear/prejudice, Name calling/Labeling\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Predicted Labels\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 23,\n        \"samples\": [\n          \"Loaded Language, Appeal to fear/prejudice\",\n          \"Smears\",\n          \"Loaded Language, Name calling/Labeling\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 36
        }
      ],
      "source": [
        "import pandas as pd\n",
        "\n",
        "# Function to map binary labels to actual label names\n",
        "def binary_to_text_labels(binary_labels, class_names):\n",
        "    return [class_names[i] for i in range(len(binary_labels)) if binary_labels[i] == 1]\n",
        "\n",
        "# Create a list to store the results\n",
        "results = []\n",
        "\n",
        "# Iterate over the test samples and collect the text, actual labels, and predicted labels\n",
        "for i in range(len(x_ts)):  # test_texts is your list of test samples\n",
        "    actual_label_names = binary_to_text_labels(test_true_labels[i], classes)  # using 'classes' array\n",
        "    predicted_label_names = binary_to_text_labels(test_pred_labels[i], classes)  # using 'classes' array\n",
        "\n",
        "    # Append the results to the list\n",
        "    results.append({\n",
        "        \"Text\": x_ts[i],\n",
        "        \"Actual Labels\": ', '.join(actual_label_names),\n",
        "        \"Predicted Labels\": ', '.join(predicted_label_names)\n",
        "    })\n",
        "\n",
        "# Convert the results list to a pandas DataFrame\n",
        "results_df = pd.DataFrame(results)\n",
        "\n",
        "# Display the DataFrame\n",
        "results_df  # Use .head() to display the first few rows\n",
        "\n",
        "# If you want to display the entire table, you can use:\n",
        "# pd.set_option('display.max_rows', None)  # Uncomment this line to display all rows if needed\n",
        "# display(results_df)\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "m7YB94PUtB2P"
      },
      "source": [
        "# ** Saving the Model**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ipKjTZk6rP5M",
        "outputId": "4a4f8fe9-eb7d-40d0-cd49-7d8c91d2e115"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:absl:You are saving your model as an HDF5 file via `model.save()` or `keras.saving.save_model(model)`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')` or `keras.saving.save_model(model, 'my_model.keras')`. \n"
          ]
        }
      ],
      "source": [
        "# Save the model to a file\n",
        "cnn_lstm_attention_model.save(\"/content/drive/MyDrive/Colab Notebooks/propaganda/predict_resuls/cnn_lstm_attention_POS_NER/augOriginal/cnn_lstm_attention_model.h5\")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HfY8W6EXtI-f"
      },
      "source": [
        "# **Saving the Results**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "T66qCDnNtSZB"
      },
      "outputs": [],
      "source": [
        "# Save the results to a CSV file\n",
        "results_df.to_csv(\"/content/drive/MyDrive/Colab Notebooks/propaganda/predict_resuls/cnn_lstm_attention_POS_NER/augOriginal/results_df.csv\", index=False)\n",
        "\n",
        "# Alternatively, save to an Excel file\n",
        "results_df.to_excel(\"/content/drive/MyDrive/Colab Notebooks/propaganda/predict_resuls/cnn_lstm_attention_POS_NER/augOriginal/results_df.xlsx\", index=False)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Zi0kf2vltixW"
      },
      "source": [
        "# **Saving Evaluation Metrics**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "z_vsK6jttsh1"
      },
      "outputs": [],
      "source": [
        "# Save evaluation metrics to a text file\n",
        "with open(\"/content/drive/MyDrive/Colab Notebooks/propaganda/predict_resuls/cnn_lstm_attention_POS_NER/augOriginal/evaluation_metrics.txt\", \"w\") as file:\n",
        "    file.write(\"Classification Report:\\n\")\n",
        "    file.write(classification_rep)\n",
        "    file.write(f\"\\nF1 macro score: {f1_macro}\\n\")\n",
        "    file.write(f\"F1 micro score: {f1_micro}\\n\")\n",
        "    file.write(f\"\\nConfusion Matrix:\\n{confusion_mat}\\n\")\n",
        "    file.write(f\"Accuracy: {accuracy}\\n\")\n",
        "    file.write(f\"Recall: {recall}\\n\")\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dqWKvsEJujSo"
      },
      "source": [
        "# **Training and Validation Loss**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 564
        },
        "id": "GoIITQ4Yux41",
        "outputId": "8ca85449-2335-44e4-d3b7-e0c90879e5bc"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x600 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "def plot_loss(size_histories):\n",
        "    plt.figure(figsize=(10, 6))\n",
        "\n",
        "    # Assuming 'ORIGINAL' is the key you want to plot\n",
        "    key = 'ORIGINAL'\n",
        "\n",
        "    history = size_histories[key]\n",
        "    plt.plot(history.history['val_loss'], label=f'{key} Validation Loss')\n",
        "    plt.plot(history.history['loss'], label=f'{key} Training Loss')\n",
        "\n",
        "    plt.title(\"Training and Validation Loss over Epochs\")\n",
        "    plt.xlabel(\"Epochs\")\n",
        "    plt.ylabel(\"Loss\")\n",
        "    plt.legend()\n",
        "    plt.grid(True)\n",
        "    plt.show()\n",
        "\n",
        "plot_loss(size_histories)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "3VUtVsFMvHvY"
      },
      "outputs": [],
      "source": [
        "def plot_and_save_size_histories(size_histories, filename):\n",
        "    plt.figure(figsize=(10, 6))\n",
        "\n",
        "    for key, history in size_histories.items():\n",
        "        plt.plot(history.history['val_loss'], label=f'{key} Validation Loss')\n",
        "        plt.plot(history.history['loss'], '--', label=f'{key} Training Loss')\n",
        "\n",
        "    plt.title(\"Training and Validation Loss over Epochs\")\n",
        "    plt.xlabel(\"Epochs\")\n",
        "    plt.ylabel(\"Loss\")\n",
        "    plt.legend()\n",
        "    plt.grid(True)\n",
        "\n",
        "    # Save the plot to a file\n",
        "    plt.savefig(filename)\n",
        "    plt.close()  # Close the plot to prevent displaying it if running in a loop\n",
        "\n",
        "# Save the plot\n",
        "plot_and_save_size_histories(size_histories, filename=\"/content/drive/MyDrive/Colab Notebooks/propaganda/predict_resuls/cnn_lstm_attention_POS_NER/augOriginal/Training and Validation Loss over Epochs.png\")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Zq3ZBIfvzgq6"
      },
      "source": [
        "# **Training and Validation Accuracy**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "O6okXzUyxBj9",
        "outputId": "8b1c297a-059b-4afa-e9ec-4101fce87071"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Metrics available for ORIGINAL: dict_keys(['accuracy', 'binary_crossentropy', 'loss', 'val_accuracy', 'val_binary_crossentropy', 'val_loss'])\n"
          ]
        }
      ],
      "source": [
        "# Check available metrics in the size_histories\n",
        "for key, history in size_histories.items():\n",
        "    print(f\"Metrics available for {key}: {history.history.keys()}\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 564
        },
        "id": "jp2viLdaxEoh",
        "outputId": "d7732e0a-4fee-41fd-8709-ed01b9b60501"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x600 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "# Plot training and validation accuracy\n",
        "def plot_accuracy(size_histories):\n",
        "    plt.figure(figsize=(10, 6))\n",
        "\n",
        "    for key, history in size_histories.items():\n",
        "        plt.plot(history.history['val_accuracy'], label=f'{key} Validation Accuracy')\n",
        "        plt.plot(history.history['accuracy'], '--', label=f'{key} Training Accuracy')\n",
        "\n",
        "    plt.title(\"Training and Validation Accuracy over Epochs\")\n",
        "    plt.xlabel(\"Epochs\")\n",
        "    plt.ylabel(\"Accuracy\")\n",
        "    plt.legend()\n",
        "    plt.grid(True)\n",
        "    plt.show()\n",
        "\n",
        "plot_accuracy(size_histories)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "i3MNBoKkxRVe"
      },
      "outputs": [],
      "source": [
        "def plot_and_save_size_histories(size_histories, filename):\n",
        "    plt.figure(figsize=(10, 6))\n",
        "\n",
        "    for key, history in size_histories.items():\n",
        "        plt.plot(history.history['val_accuracy'], label=f'{key} Validation Accuracy')\n",
        "        plt.plot(history.history['accuracy'], '--', label=f'{key} Training Accuracy')\n",
        "\n",
        "    plt.title(\"Training and Validation Accuracy over Epochs\")\n",
        "    plt.xlabel(\"Epochs\")\n",
        "    plt.ylabel(\"Accuracy\")\n",
        "    plt.legend()\n",
        "    plt.grid(True)\n",
        "\n",
        "\n",
        "    # Save the plot to a file\n",
        "    plt.savefig(filename)\n",
        "    plt.close()  # Close the plot to prevent displaying it if running in a loop\n",
        "\n",
        "# Save the plot\n",
        "plot_and_save_size_histories(size_histories, filename=\"/content/drive/MyDrive/Colab Notebooks/propaganda/predict_resuls/cnn_lstm_attention_POS_NER/augOriginal/Training and Validation Accuracy over Epochs.png\")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "uA2_fAf2024v"
      },
      "source": [
        "# Binary **Crossentropy**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 564
        },
        "id": "MqEHaw7Lx-mr",
        "outputId": "66989233-b4f1-4b07-9d68-6a257c011161"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x600 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "# Plot training and validation binary crossentropy\n",
        "def plot_binary_crossentropy(size_histories):\n",
        "    plt.figure(figsize=(10, 6))\n",
        "\n",
        "    for key, history in size_histories.items():\n",
        "        plt.plot(history.history['val_binary_crossentropy'], label=f'{key} Validation Binary Crossentropy')\n",
        "        plt.plot(history.history['binary_crossentropy'], '--', label=f'{key} Training Binary Crossentropy')\n",
        "\n",
        "    plt.title(\"Training and Validation Binary Crossentropy over Epochs\")\n",
        "    plt.xlabel(\"Epochs\")\n",
        "    plt.ylabel(\"Binary Crossentropy\")\n",
        "    plt.legend()\n",
        "    plt.grid(True)\n",
        "    plt.show()\n",
        "\n",
        "plot_binary_crossentropy(size_histories)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "u_k3XX77x-4E"
      },
      "outputs": [],
      "source": [
        "def plot_and_save_size_histories(size_histories, filename):\n",
        "    plt.figure(figsize=(10, 6))\n",
        "\n",
        "    for key, history in size_histories.items():\n",
        "        plt.plot(history.history['val_binary_crossentropy'], label=f'{key} Validation Binary Crossentropy')\n",
        "        plt.plot(history.history['binary_crossentropy'], '--', label=f'{key} Training Binary Crossentropy')\n",
        "\n",
        "    plt.title(\"Training and Validation Binary Crossentropy over Epochs\")\n",
        "    plt.xlabel(\"Epochs\")\n",
        "    plt.ylabel(\"Binary Crossentropy\")\n",
        "    plt.legend()\n",
        "    plt.grid(True)\n",
        "\n",
        "    # Save the plot to a file\n",
        "    plt.savefig(filename)\n",
        "    plt.close()  # Close the plot to prevent displaying it if running in a loop\n",
        "\n",
        "# Save the plot\n",
        "plot_and_save_size_histories(size_histories, filename=\"/content/drive/MyDrive/Colab Notebooks/propaganda/predict_resuls/cnn_lstm_attention_POS_NER/augOriginal/Training and Validation Binary Crossentropy over Epochs.png\")"
      ]
    }
  ],
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.10.10"
    },
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "27dc81c687824079a11e08cffbc8bc15": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_52550e94fecb430eb8d292ede83e9572",
              "IPY_MODEL_af59a2f876fb4b9dbc249cca0eee99bf",
              "IPY_MODEL_5d02a44463504b8284d289c6f11208b5"
            ],
            "layout": "IPY_MODEL_0a066e9c12434a96b64c9ae56a92d1af"
          }
        },
        "52550e94fecb430eb8d292ede83e9572": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_94928004da8845ba976664334254bb57",
            "placeholder": "​",
            "style": "IPY_MODEL_a3251a569a57420da82239bd27b29c27",
            "value": "Downloading https://raw.githubusercontent.com/stanfordnlp/stanza-resources/main/resources_1.9.0.json: "
          }
        },
        "af59a2f876fb4b9dbc249cca0eee99bf": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_ed53626694b34ee5b0ce325a592a6906",
            "max": 48453,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_5589ebc3cd164624aeb171a87e9d1104",
            "value": 48453
          }
        },
        "5d02a44463504b8284d289c6f11208b5": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_d6a821c1aaf0450aaa329921df4bf88a",
            "placeholder": "​",
            "style": "IPY_MODEL_86c0477fc27e498ca7d59fa61fe0276b",
            "value": " 392k/? [00:00&lt;00:00, 6.05MB/s]"
          }
        },
        "0a066e9c12434a96b64c9ae56a92d1af": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "94928004da8845ba976664334254bb57": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "a3251a569a57420da82239bd27b29c27": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "ed53626694b34ee5b0ce325a592a6906": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "5589ebc3cd164624aeb171a87e9d1104": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "d6a821c1aaf0450aaa329921df4bf88a": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "86c0477fc27e498ca7d59fa61fe0276b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "8ad55e00fdef4c488320154259d5db86": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_3612c3c9ce3f45cd9f6b6a94c65811fd",
              "IPY_MODEL_c9927e6287e5483d986a5e669865283f",
              "IPY_MODEL_0e7720c663ea40a4bf580132dfa472d4"
            ],
            "layout": "IPY_MODEL_48a67a849a8c4e969c8456461bbbbb15"
          }
        },
        "3612c3c9ce3f45cd9f6b6a94c65811fd": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_3d0475b1fdd446a6b452ff6d596a2c59",
            "placeholder": "​",
            "style": "IPY_MODEL_532d749ac10e4d83a5ad163bde92694f",
            "value": "Downloading https://huggingface.co/stanfordnlp/stanza-ar/resolve/v1.9.0/models/default.zip: 100%"
          }
        },
        "c9927e6287e5483d986a5e669865283f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_7fbcbf179a1249caa3e7a30a6dc6fd95",
            "max": 459506904,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_f4e5fbdee8954b53b487e629fc67390c",
            "value": 459506904
          }
        },
        "0e7720c663ea40a4bf580132dfa472d4": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_3f75dd0d46b344078bf007de152f41a1",
            "placeholder": "​",
            "style": "IPY_MODEL_2be1ef8a2f0a4f9d9fe10368bb5b9464",
            "value": " 460M/460M [00:04&lt;00:00, 171MB/s]"
          }
        },
        "48a67a849a8c4e969c8456461bbbbb15": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "3d0475b1fdd446a6b452ff6d596a2c59": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "532d749ac10e4d83a5ad163bde92694f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "7fbcbf179a1249caa3e7a30a6dc6fd95": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "f4e5fbdee8954b53b487e629fc67390c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "3f75dd0d46b344078bf007de152f41a1": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "2be1ef8a2f0a4f9d9fe10368bb5b9464": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "22c0c080b0aa48d9a263db83bd4ba292": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_eb369e73ada24787a5eeb43bf6ee0e58",
              "IPY_MODEL_c43af64ec8ed4faea1bdd3d6172fa451",
              "IPY_MODEL_8035da71d9ed45debb811fddbc652c5b"
            ],
            "layout": "IPY_MODEL_8558097071f74d099dc9439dafeda48f"
          }
        },
        "eb369e73ada24787a5eeb43bf6ee0e58": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_b75db6785bb34fc391725093eb5d0d73",
            "placeholder": "​",
            "style": "IPY_MODEL_756441376ecd4202b96527bc665d2cee",
            "value": "Downloading https://raw.githubusercontent.com/stanfordnlp/stanza-resources/main/resources_1.9.0.json: "
          }
        },
        "c43af64ec8ed4faea1bdd3d6172fa451": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_be8b6e2eaafc4ca381735e927dfa949e",
            "max": 48453,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_0ae47ceac4b4408e87c105f17fb07403",
            "value": 48453
          }
        },
        "8035da71d9ed45debb811fddbc652c5b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_e401143ac189458091cb5bfa3d245e59",
            "placeholder": "​",
            "style": "IPY_MODEL_e8011943f9a14a2da3485b035b2d1ce1",
            "value": " 392k/? [00:00&lt;00:00, 9.07MB/s]"
          }
        },
        "8558097071f74d099dc9439dafeda48f": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "b75db6785bb34fc391725093eb5d0d73": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "756441376ecd4202b96527bc665d2cee": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "be8b6e2eaafc4ca381735e927dfa949e": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "0ae47ceac4b4408e87c105f17fb07403": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "e401143ac189458091cb5bfa3d245e59": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "e8011943f9a14a2da3485b035b2d1ce1": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        }
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}