Text_Classification_model_1.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "d3e0c074",
   "metadata": {
    "id": "d3e0c074"
   },
   "outputs": [],
   "source": [
    "import glob\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import os\n",
    "import re\n",
    "from tqdm import tqdm\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "\n",
    "from keras.callbacks import ModelCheckpoint,EarlyStopping,TensorBoard\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "import tensorflow_addons as tf\n",
    "from tensorflow_addons.metrics import F1Score\n",
    "from keras.models import load_model\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras import *\n",
    "from tensorflow.keras.preprocessing.text import Tokenizer\n",
    "\n",
    "# import norm\n",
    "from keras.constraints import unit_norm\n",
    "from tensorflow.keras.preprocessing.sequence import pad_sequences\n",
    "from keras.layers import Concatenate, Embedding, Conv1D, Dense, Flatten, MaxPooling1D, Dropout, concatenate, MaxPool1D"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "XcHwj8bcogMr",
   "metadata": {
    "id": "XcHwj8bcogMr"
   },
   "outputs": [],
   "source": [
    "from google.colab import drive\n",
    "drive.mount('/content/drive')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "418ba68c",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "id": "418ba68c",
    "outputId": "d2380519-4591-4d06-d094-fe0640047b8b"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-bcfe3496-238a-4e8b-9511-b94419f91102\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>class</th>\n",
       "      <th>preprocessed_text</th>\n",
       "      <th>preprocessed_subject</th>\n",
       "      <th>preprocessed_emails</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>797</th>\n",
       "      <td>alt.atheism</td>\n",
       "      <td>article that was the original answer while do...</td>\n",
       "      <td>yet more rushdie islamic law</td>\n",
       "      <td>dbstu1 tu-bs edu buphy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>798</th>\n",
       "      <td>alt.atheism</td>\n",
       "      <td>article writes deletion had the old allegatio...</td>\n",
       "      <td>alleged deathbed conversions was asimov stamp</td>\n",
       "      <td>skyblu ccit arizona edu panix netcom CCIT ARIZ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>799</th>\n",
       "      <td>comp.graphics</td>\n",
       "      <td>call for presentations navy scientific visual...</td>\n",
       "      <td>call for presentations navy sciviz vr seminar</td>\n",
       "      <td>oasys navy mil ocean long legs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>800</th>\n",
       "      <td>comp.graphics</td>\n",
       "      <td>gnuplot etc make easy plot real valued functi...</td>\n",
       "      <td>graphical representation of vector valued func...</td>\n",
       "      <td>ucssun1 sdsu edu</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>801</th>\n",
       "      <td>comp.graphics</td>\n",
       "      <td>article can someone please tell where can ftp...</td>\n",
       "      <td>dmorph</td>\n",
       "      <td>coconut cis ufl edu usl ucs</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-bcfe3496-238a-4e8b-9511-b94419f91102')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-bcfe3496-238a-4e8b-9511-b94419f91102 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-bcfe3496-238a-4e8b-9511-b94419f91102');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ],
      "text/plain": [
       "             class                                  preprocessed_text  \\\n",
       "797    alt.atheism   article that was the original answer while do...   \n",
       "798    alt.atheism   article writes deletion had the old allegatio...   \n",
       "799  comp.graphics   call for presentations navy scientific visual...   \n",
       "800  comp.graphics   gnuplot etc make easy plot real valued functi...   \n",
       "801  comp.graphics   article can someone please tell where can ftp...   \n",
       "\n",
       "                                  preprocessed_subject  \\\n",
       "797                      yet more rushdie islamic law    \n",
       "798     alleged deathbed conversions was asimov stamp    \n",
       "799      call for presentations navy sciviz vr seminar   \n",
       "800  graphical representation of vector valued func...   \n",
       "801                                             dmorph   \n",
       "\n",
       "                                   preprocessed_emails  \n",
       "797                             dbstu1 tu-bs edu buphy  \n",
       "798  skyblu ccit arizona edu panix netcom CCIT ARIZ...  \n",
       "799                     oasys navy mil ocean long legs  \n",
       "800                                   ucssun1 sdsu edu  \n",
       "801                        coconut cis ufl edu usl ucs  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv('/content/drive/MyDrive/_ML_PROJECTS/CNN_Text_Classification/data.csv')\n",
    "df = df.drop(columns=['Unnamed: 0'])\n",
    "df[797:802]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "u98L0N4wAACy",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 282
    },
    "id": "u98L0N4wAACy",
    "outputId": "f6f28ca5-f530-434a-c448-ede41c27b548"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fa99f5f9c50>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df['class'].value_counts()[:20].plot(kind='barh')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "HKKaDtJwht_-",
   "metadata": {
    "id": "HKKaDtJwht_-"
   },
   "outputs": [],
   "source": [
    "cols = ['preprocessed_emails', 'preprocessed_subject', 'preprocessed_text']\n",
    "df['training'] = df[cols].apply(lambda row: ' '.join(row.values.astype(str)), axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "XZ2u2jTcqUg6",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 285
    },
    "id": "XZ2u2jTcqUg6",
    "outputId": "33df2fd5-5090-41ed-b929-b0042e5c1440"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-550d38da-c428-41b5-9444-6db6bb093914\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>class</th>\n",
       "      <th>preprocessed_text</th>\n",
       "      <th>preprocessed_subject</th>\n",
       "      <th>preprocessed_emails</th>\n",
       "      <th>training</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>797</th>\n",
       "      <td>alt.atheism</td>\n",
       "      <td>article that was the original answer while do...</td>\n",
       "      <td>yet more rushdie islamic law</td>\n",
       "      <td>dbstu1 tu-bs edu buphy</td>\n",
       "      <td>dbstu1 tu-bs edu buphy yet more rushdie islami...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>798</th>\n",
       "      <td>alt.atheism</td>\n",
       "      <td>article writes deletion had the old allegatio...</td>\n",
       "      <td>alleged deathbed conversions was asimov stamp</td>\n",
       "      <td>skyblu ccit arizona edu panix netcom CCIT ARIZ...</td>\n",
       "      <td>skyblu ccit arizona edu panix netcom CCIT ARIZ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>799</th>\n",
       "      <td>comp.graphics</td>\n",
       "      <td>call for presentations navy scientific visual...</td>\n",
       "      <td>call for presentations navy sciviz vr seminar</td>\n",
       "      <td>oasys navy mil ocean long legs</td>\n",
       "      <td>oasys navy mil ocean long legs call for presen...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>800</th>\n",
       "      <td>comp.graphics</td>\n",
       "      <td>gnuplot etc make easy plot real valued functi...</td>\n",
       "      <td>graphical representation of vector valued func...</td>\n",
       "      <td>ucssun1 sdsu edu</td>\n",
       "      <td>ucssun1 sdsu edu graphical representation of v...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>801</th>\n",
       "      <td>comp.graphics</td>\n",
       "      <td>article can someone please tell where can ftp...</td>\n",
       "      <td>dmorph</td>\n",
       "      <td>coconut cis ufl edu usl ucs</td>\n",
       "      <td>coconut cis ufl edu usl ucs dmorph  article ca...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-550d38da-c428-41b5-9444-6db6bb093914')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-550d38da-c428-41b5-9444-6db6bb093914 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-550d38da-c428-41b5-9444-6db6bb093914');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ],
      "text/plain": [
       "             class                                  preprocessed_text  \\\n",
       "797    alt.atheism   article that was the original answer while do...   \n",
       "798    alt.atheism   article writes deletion had the old allegatio...   \n",
       "799  comp.graphics   call for presentations navy scientific visual...   \n",
       "800  comp.graphics   gnuplot etc make easy plot real valued functi...   \n",
       "801  comp.graphics   article can someone please tell where can ftp...   \n",
       "\n",
       "                                  preprocessed_subject  \\\n",
       "797                      yet more rushdie islamic law    \n",
       "798     alleged deathbed conversions was asimov stamp    \n",
       "799      call for presentations navy sciviz vr seminar   \n",
       "800  graphical representation of vector valued func...   \n",
       "801                                             dmorph   \n",
       "\n",
       "                                   preprocessed_emails  \\\n",
       "797                             dbstu1 tu-bs edu buphy   \n",
       "798  skyblu ccit arizona edu panix netcom CCIT ARIZ...   \n",
       "799                     oasys navy mil ocean long legs   \n",
       "800                                   ucssun1 sdsu edu   \n",
       "801                        coconut cis ufl edu usl ucs   \n",
       "\n",
       "                                              training  \n",
       "797  dbstu1 tu-bs edu buphy yet more rushdie islami...  \n",
       "798  skyblu ccit arizona edu panix netcom CCIT ARIZ...  \n",
       "799  oasys navy mil ocean long legs call for presen...  \n",
       "800  ucssun1 sdsu edu graphical representation of v...  \n",
       "801  coconut cis ufl edu usl ucs dmorph  article ca...  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[797:802]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "93dc325a",
   "metadata": {
    "id": "93dc325a"
   },
   "outputs": [],
   "source": [
    "X,y = df['training'], df['class']\n",
    "\n",
    "#Encoding labels\n",
    "le = LabelEncoder()\n",
    "labels = le.fit_transform(y)\n",
    "y = np.asarray( tf.keras.utils.to_categorical(labels))\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y,\n",
    "                                                    stratify=y, \n",
    "                                                    test_size=0.25)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "244d3010",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "244d3010",
    "outputId": "14c03e15-0de0-4cf4-d6a7-51af717a3c75"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "training set: \n",
      " 14121 \n",
      "testing set: \n",
      " 4707\n"
     ]
    }
   ],
   "source": [
    "print(\"training set: \\n\",len(X_train),\"\\ntesting set: \\n\",len(X_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4WH9rEBhAfBq",
   "metadata": {
    "id": "4WH9rEBhAfBq"
   },
   "outputs": [],
   "source": [
    "length_sentence =[len(s.split()) for s in X_train]\n",
    "length_sentence.sort()\n",
    "length_sentence = np.array(length_sentence)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "xLpz44UpAjT6",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "xLpz44UpAjT6",
    "outputId": "be8cce49-4be8-45e4-b806-a6b0f43b2a85"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "839"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "percentile_98 = int(np.percentile(length_sentence,98))\n",
    "percentile_98"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0d10c114",
   "metadata": {
    "id": "0d10c114"
   },
   "outputs": [],
   "source": [
    "tokenizer = Tokenizer(filters='!\"#$%&()*+,-./:;<=>?@[\\\\]^`{|}~\\t\\n', oov_token=\"<OOV>\")\n",
    "tokenizer.fit_on_texts(X_train)\n",
    "word_index = tokenizer.word_index\n",
    "\n",
    "#Generating sequences\n",
    "#text_train = X_train.values\n",
    "#text_test = X_test.values\n",
    "train_sequences = tokenizer.texts_to_sequences(X_train.values)\n",
    "test_sequences = tokenizer.texts_to_sequences(X_test.values)\n",
    "\n",
    "#Padding\n",
    "max_length = 972\n",
    "x_train = pad_sequences(train_sequences, padding='post',maxlen=max_length)\n",
    "x_test = pad_sequences(test_sequences, padding='post',maxlen=max_length)\n",
    "\n",
    "train_ds = tf.data.Dataset.from_tensor_slices((x_train,y_train))\n",
    "test_ds = tf.data.Dataset.from_tensor_slices((x_test,y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "X3NGVgFEB7xx",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "X3NGVgFEB7xx",
    "outputId": "f6bc0935-c13a-4cf4-b305-47d8a73bc59e"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(14121, 972)\n",
      "(4707, 972)\n"
     ]
    }
   ],
   "source": [
    "print(x_train.shape)\n",
    "print(x_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b92550f7",
   "metadata": {
    "id": "b92550f7"
   },
   "outputs": [],
   "source": [
    "vocab_size = len(word_index) + 1\n",
    "max_length = 972"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "Kj9A8-ZNBhMR",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Kj9A8-ZNBhMR",
    "outputId": "c26703dc-a086-4973-d171-ab90d2495b1f"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "77875"
      ]
     },
     "execution_count": 135,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vocab_size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "10873448",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "10873448",
    "outputId": "9f9ebb5f-78d9-49d3-d5f1-b95b9ae63c7c"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "400000it [00:09, 40321.18it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded 400000 word vectors.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "embeddings_index = dict()\n",
    "f = open('/content/drive/MyDrive/_ML_PROJECTS/CNN_Text_Classification/glove.6B/glove.6B.100d.txt',encoding=\"utf8\")\n",
    "for line in tqdm(f):\n",
    "\tvalues = line.split()\n",
    "\tword = values[0]\n",
    "\tcoefs = np.asarray(values[1:], dtype='float32')\n",
    "\tembeddings_index[word] = coefs\n",
    "f.close()\n",
    "print('Loaded %s word vectors.' % len(embeddings_index))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8bb6fa41",
   "metadata": {
    "id": "8bb6fa41"
   },
   "outputs": [],
   "source": [
    "embedding_matrix = np.zeros((vocab_size, 100))\n",
    "for word, i in word_index.items():\n",
    "\tembedding_vector = embeddings_index.get(word)\n",
    "\tif embedding_vector is not None:\n",
    "\t\tembedding_matrix[i] = embedding_vector"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7e790541",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "7e790541",
    "outputId": "50930a12-5fd3-4153-f925-ee218ac57f94"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(77875, 100)"
      ]
     },
     "execution_count": 138,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "embedding_matrix.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9da03237",
   "metadata": {
    "id": "9da03237"
   },
   "source": [
    "## Model\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "483c7ced",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "483c7ced",
    "outputId": "42e7fdea-75a9-4016-a089-c8803c0cbd33",
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"TF_Functional_API\"\n",
      "__________________________________________________________________________________________________\n",
      " Layer (type)                   Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      " input_39 (InputLayer)          [(None, 972)]        0           []                               \n",
      "                                                                                                  \n",
      " embedding_38 (Embedding)       (None, 972, 100)     7787500     ['input_39[0][0]']               \n",
      "                                                                                                  \n",
      " conv1d_143 (Conv1D)            (None, 969, 16)      6416        ['embedding_38[0][0]']           \n",
      "                                                                                                  \n",
      " conv1d_144 (Conv1D)            (None, 969, 16)      6416        ['embedding_38[0][0]']           \n",
      "                                                                                                  \n",
      " conv1d_145 (Conv1D)            (None, 969, 16)      6416        ['embedding_38[0][0]']           \n",
      "                                                                                                  \n",
      " concatenate_2 (Concatenate)    (None, 969, 48)      0           ['conv1d_143[0][0]',             \n",
      "                                                                  'conv1d_144[0][0]',             \n",
      "                                                                  'conv1d_145[0][0]']             \n",
      "                                                                                                  \n",
      " max_pooling1d_70 (MaxPooling1D  (None, 323, 48)     0           ['concatenate_2[0][0]']          \n",
      " )                                                                                                \n",
      "                                                                                                  \n",
      " conv1d_146 (Conv1D)            (None, 321, 16)      2320        ['max_pooling1d_70[0][0]']       \n",
      "                                                                                                  \n",
      " conv1d_147 (Conv1D)            (None, 321, 16)      2320        ['max_pooling1d_70[0][0]']       \n",
      "                                                                                                  \n",
      " conv1d_148 (Conv1D)            (None, 321, 16)      2320        ['max_pooling1d_70[0][0]']       \n",
      "                                                                                                  \n",
      " concatenate_3 (Concatenate)    (None, 321, 48)      0           ['conv1d_146[0][0]',             \n",
      "                                                                  'conv1d_147[0][0]',             \n",
      "                                                                  'conv1d_148[0][0]']             \n",
      "                                                                                                  \n",
      " max_pooling1d_71 (MaxPooling1D  (None, 107, 48)     0           ['concatenate_3[0][0]']          \n",
      " )                                                                                                \n",
      "                                                                                                  \n",
      " conv1d_149 (Conv1D)            (None, 105, 16)      2320        ['max_pooling1d_71[0][0]']       \n",
      "                                                                                                  \n",
      " flatten_35 (Flatten)           (None, 1680)         0           ['conv1d_149[0][0]']             \n",
      "                                                                                                  \n",
      " dropout_35 (Dropout)           (None, 1680)         0           ['flatten_35[0][0]']             \n",
      "                                                                                                  \n",
      " dense_70 (Dense)               (None, 64)           107584      ['dropout_35[0][0]']             \n",
      "                                                                                                  \n",
      " dense_71 (Dense)               (None, 20)           1300        ['dense_70[0][0]']               \n",
      "                                                                                                  \n",
      "==================================================================================================\n",
      "Total params: 7,924,912\n",
      "Trainable params: 137,412\n",
      "Non-trainable params: 7,787,500\n",
      "__________________________________________________________________________________________________\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.7/dist-packages/keras/optimizer_v2/adam.py:105: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead.\n",
      "  super(Adam, self).__init__(name, **kwargs)\n"
     ]
    }
   ],
   "source": [
    "input_x = Input(shape=( max_length))\n",
    "x = layers.Embedding(vocab_size, 100, input_length=max_length,trainable=False,embeddings_initializer=tf.keras.initializers.Constant(embedding_matrix))(input_x)\n",
    "\n",
    "x_m = layers.Conv1D(filters=16,kernel_size=4, activation='relu', kernel_initializer =tf.keras.initializers.he_normal(),kernel_regularizer=tf.keras.regularizers.l2(0.001))(x)\n",
    "x_n = layers.Conv1D(filters=16,kernel_size=4, activation='relu', kernel_initializer =tf.keras.initializers.he_normal(),kernel_regularizer=tf.keras.regularizers.l2(0.001))(x)\n",
    "x_o = layers.Conv1D(filters=16,kernel_size=4, activation='relu', kernel_initializer =tf.keras.initializers.he_normal(),kernel_regularizer=tf.keras.regularizers.l2(0.001))(x)\n",
    "concatenated1 = layers.concatenate([x_m,x_n,x_o],axis = -1)\n",
    "max_pool1 = MaxPool1D(pool_size=3)(concatenated1)\n",
    "\n",
    "x_i = layers.Conv1D(filters=16, kernel_size=3, activation='relu', kernel_initializer =tf.keras.initializers.he_normal(),kernel_regularizer=tf.keras.regularizers.l2(0.001))(max_pool1)\n",
    "x_j = layers.Conv1D(filters=16,kernel_size=3, activation='relu', kernel_initializer =tf.keras.initializers.he_normal(),kernel_regularizer=tf.keras.regularizers.l2(0.001))(max_pool1)\n",
    "x_k = layers.Conv1D(filters=16,kernel_size=3, activation='relu', kernel_initializer =tf.keras.initializers.he_normal(),kernel_regularizer=tf.keras.regularizers.l2(0.001))(max_pool1)\n",
    "concatenated2 = layers.concatenate([x_i,x_j,x_k],axis = -1)\n",
    "max_pool2 = MaxPool1D(pool_size=3)(concatenated2)\n",
    "\n",
    "x1 = layers.Conv1D(filters=16, kernel_size=3, activation='relu', kernel_initializer =tf.keras.initializers.he_normal(),kernel_regularizer=tf.keras.regularizers.l2(0.001))(max_pool2)\n",
    "\n",
    "flatten = layers.Flatten()(x1)\n",
    "dropout1 = layers.Dropout(0.3)(flatten)\n",
    "dense1 = layers.Dense(64, activation='relu', kernel_initializer = tf.keras.initializers.he_normal())(dropout1)\n",
    "\n",
    "\n",
    "output= layers.Dense(20, activation='sigmoid', kernel_initializer= tf.keras.initializers.glorot_normal())(dense1)\n",
    "\n",
    "model = Model(inputs = input_x, outputs=output,name=\"TF_Functional_API\")\n",
    "\n",
    "checkpoint = ModelCheckpoint('best_model_L1.h5',  # model filename\n",
    "                             monitor='val_accuracy', # quantity to monitor\n",
    "                             verbose=0, # verbosity - 0 or 1\n",
    "                             save_best_only= True, # The latest best model will not be overwritten\n",
    "                             mode='auto') # The decision to overwrite model is made \n",
    "                                          # automatically depending on the quantity to monitor \n",
    "\n",
    "log_dir = \"logs\"\n",
    "tensorboard = TensorBoard(log_dir=log_dir,histogram_freq=1,write_graph=True)                                        \n",
    "\n",
    "early_stop = EarlyStopping(monitor=\"val_accuracy\",mode='max',patience=3)     \n",
    "reduce_ = tf.keras.callbacks.ReduceLROnPlateau(monitor='val_accuracy',patience=1,mode='auto',verbose=1,factor=0.8,min_lr=0.0001)\n",
    "\n",
    "callbacks = [reduce_,checkpoint,early_stop,tensorboard]     \n",
    "\n",
    "#LEARN_RATE = 2.0e-4\n",
    "model.compile(loss='categorical_crossentropy', # Better loss function for neural networks\n",
    "              optimizer=Adam(lr=0.001), # Adam optimizer with 1.0e-4 learning rate\n",
    "              metrics = ['accuracy', F1Score(average='micro',num_classes=20)]) # Metrics to be evaluated by the model\n",
    "\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "qDMVkTjqD6q-",
   "metadata": {
    "id": "qDMVkTjqD6q-"
   },
   "outputs": [],
   "source": [
    "tf.keras.utils.plot_model(model, \"multi_input_and_output_model.png\", show_shapes=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "VTmHqlsMEh06",
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "VTmHqlsMEh06"
   },
   "outputs": [],
   "source": [
    "#model = load_model('/content/best_model_L1.h5')\n",
    "epochs = 40\n",
    "\n",
    "history = model.fit(x_train,y_train,\n",
    "                    epochs= epochs ,\n",
    "                    validation_data=(x_test,y_test),\n",
    "                    batch_size =64,\n",
    "                    callbacks=[callbacks],\n",
    "                    verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "cSB2b3Y5Eh07",
   "metadata": {
    "id": "cSB2b3Y5Eh07"
   },
   "outputs": [],
   "source": [
    "history=np.load('./model_weights/my_history.npy',allow_pickle='TRUE').item()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "HROSD7TQitdj",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 295
    },
    "id": "HROSD7TQitdj",
    "outputId": "3801ed0f-7a85-4396-ed7d-7a29657b23ab"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "plt.plot(history['loss'], label=' training data')\n",
    "plt.plot(history['val_loss'], label='validation data)')\n",
    "plt.title('Loss for Text Classification')\n",
    "plt.ylabel('Loss value')\n",
    "plt.xlabel('No. epoch')\n",
    "plt.legend(loc=\"upper left\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "glEcqRZHitdl",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 295
    },
    "id": "glEcqRZHitdl",
    "outputId": "b1842fda-b75b-4abe-8578-8f0bb42cc3a4"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history['accuracy'], label=' training data')\n",
    "plt.plot(history['val_accuracy'], label='validation data)')\n",
    "plt.title('Accuracy for Text Classification')\n",
    "plt.ylabel('Loss value')\n",
    "plt.xlabel('No. epoch')\n",
    "plt.legend(loc=\"upper left\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "fd80b950",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train accuracy 0.8711139559745789\n",
      "Validation accuracy 0.7656681537628174\n"
     ]
    }
   ],
   "source": [
    "print(\"Train accuracy\",max(history['accuracy']))\n",
    "print(\"Validation accuracy\",max(history['val_accuracy']))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eGeMLs7Onria",
   "metadata": {
    "id": "eGeMLs7Onria"
   },
   "source": [
    "# Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "lD0DLh_knt9F",
   "metadata": {
    "id": "lD0DLh_knt9F"
   },
   "outputs": [],
   "source": [
    "model = load_model('/content/best_model_L1.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "LDp9aBz_oJcI",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "LDp9aBz_oJcI",
    "outputId": "ddad239f-605f-4910-b541-ee83ae202b14"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "prediction shape: (1, 20) [[1.09391166e-07 1.43946484e-01 2.06691548e-02 2.04064287e-02\n",
      "  7.82901645e-02 2.94220983e-04 9.99987602e-01 2.33618379e-01\n",
      "  1.56637151e-02 1.84288606e-01 6.66271970e-02 8.57936641e-07\n",
      "  4.94991094e-01 1.87705018e-06 3.49415362e-01 2.91094215e-09\n",
      "  1.65462232e-04 3.40243318e-08 1.67680415e-03 3.19320776e-07]]\n"
     ]
    }
   ],
   "source": [
    "prediction = model.predict(x_test[1:2])\n",
    "print(\"prediction shape:\", prediction.shape, prediction)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "uyU8ggD1oh14",
   "metadata": {
    "id": "uyU8ggD1oh14"
   },
   "outputs": [],
   "source": [
    "y_classes = prediction.argmax(axis=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "az9HZK9Xsv1i",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "az9HZK9Xsv1i",
    "outputId": "843305f6-74d9-4749-a8ba-343b878cb301"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['misc.forsale'], dtype=object)"
      ]
     },
     "execution_count": 115,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "le.inverse_transform(y_classes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9yOj-IvMtmaI",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "9yOj-IvMtmaI",
    "outputId": "d839ec6f-dde4-453e-e629-a7d8035a77f8"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "        0., 0., 0., 0.]], dtype=float32)"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test[1:2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "RsZZp32atyIj",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "RsZZp32atyIj",
    "outputId": "c345563e-e176-42ff-a73c-4df162ea7140"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['alt.atheism', 'comp.graphics', 'comp.os.ms-windows.misc',\n",
       "       'comp.sys.ibm.pc.hardware', 'comp.sys.mac.hardware',\n",
       "       'comp.windows.x', 'misc.forsale', 'rec.autos', 'rec.motorcycles',\n",
       "       'rec.sport.baseball', 'rec.sport.hockey', 'sci.crypt',\n",
       "       'sci.electronics', 'sci.med', 'sci.space',\n",
       "       'soc.religion.christian', 'talk.politics.guns',\n",
       "       'talk.politics.mideast', 'talk.politics.misc',\n",
       "       'talk.religion.misc'], dtype=object)"
      ]
     },
     "execution_count": 116,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "le.classes_"
   ]
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "collapsed_sections": [],
   "name": "Text_Classification_model_1.ipynb",
   "provenance": []
  },
  "gpuClass": "standard",
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}