BERT_final.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "BERT_final.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "6kdLUTPGmo5M"
      },
      "source": [
        "!wget --quiet https://raw.githubusercontent.com/tensorflow/models/master/official/nlp/bert/tokenization.py"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "feauyo34m8e0",
        "outputId": "fa7a78a2-1491-4963-d8df-c624ed548554"
      },
      "source": [
        "!pip install sentencepiece"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Collecting sentencepiece\n",
            "  Downloading sentencepiece-0.1.96-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.2 MB)\n",
            "\u001b[?25l\r\u001b[K     |▎                               | 10 kB 25.6 MB/s eta 0:00:01\r\u001b[K     |▌                               | 20 kB 28.1 MB/s eta 0:00:01\r\u001b[K     |▉                               | 30 kB 12.4 MB/s eta 0:00:01\r\u001b[K     |█                               | 40 kB 9.2 MB/s eta 0:00:01\r\u001b[K     |█▍                              | 51 kB 5.5 MB/s eta 0:00:01\r\u001b[K     |█▋                              | 61 kB 5.9 MB/s eta 0:00:01\r\u001b[K     |██                              | 71 kB 5.7 MB/s eta 0:00:01\r\u001b[K     |██▏                             | 81 kB 6.3 MB/s eta 0:00:01\r\u001b[K     |██▍                             | 92 kB 6.4 MB/s eta 0:00:01\r\u001b[K     |██▊                             | 102 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |███                             | 112 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |███▎                            | 122 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |███▌                            | 133 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |███▉                            | 143 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |████                            | 153 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |████▎                           | 163 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |████▋                           | 174 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |████▉                           | 184 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |█████▏                          | 194 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |█████▍                          | 204 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |█████▊                          | 215 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████                          | 225 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████▏                         | 235 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████▌                         | 245 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████▊                         | 256 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████                         | 266 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████▎                        | 276 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████▋                        | 286 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████▉                        | 296 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |████████                        | 307 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |████████▍                       | 317 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |████████▋                       | 327 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |█████████                       | 337 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |█████████▏                      | 348 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |█████████▌                      | 358 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |█████████▊                      | 368 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████████                      | 378 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████████▎                     | 389 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████████▌                     | 399 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████████▉                     | 409 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████████                     | 419 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████████▍                    | 430 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████████▋                    | 440 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████████▉                    | 450 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |████████████▏                   | 460 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |████████████▍                   | 471 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |████████████▊                   | 481 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |█████████████                   | 491 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |█████████████▎                  | 501 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |█████████████▌                  | 512 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |█████████████▊                  | 522 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████████████                  | 532 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████████████▎                 | 542 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████████████▋                 | 552 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████████████▉                 | 563 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████████████▏                | 573 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████████████▍                | 583 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████████████▋                | 593 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |████████████████                | 604 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |████████████████▏               | 614 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |████████████████▌               | 624 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |████████████████▊               | 634 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |█████████████████               | 645 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |█████████████████▎              | 655 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |█████████████████▌              | 665 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |█████████████████▉              | 675 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████████████████              | 686 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████████████████▍             | 696 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████████████████▋             | 706 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████████████████             | 716 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████████████████▏            | 727 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████████████████▍            | 737 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████████████████▊            | 747 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |████████████████████            | 757 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |████████████████████▎           | 768 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |████████████████████▌           | 778 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |████████████████████▉           | 788 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |█████████████████████           | 798 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |█████████████████████▎          | 808 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |█████████████████████▋          | 819 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |█████████████████████▉          | 829 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████████████████████▏         | 839 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████████████████████▍         | 849 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████████████████████▊         | 860 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████████████████████         | 870 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████████████████████▏        | 880 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████████████████████▌        | 890 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████████████████████▊        | 901 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |████████████████████████        | 911 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |████████████████████████▎       | 921 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |████████████████████████▋       | 931 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |████████████████████████▉       | 942 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |█████████████████████████       | 952 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |█████████████████████████▍      | 962 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |█████████████████████████▋      | 972 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████████████████████████      | 983 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████████████████████████▏     | 993 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████████████████████████▌     | 1.0 MB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████████████████████████▊     | 1.0 MB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████████████████████████     | 1.0 MB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████████████████████████▎    | 1.0 MB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████████████████████████▌    | 1.0 MB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████████████████████████▉    | 1.1 MB 5.3 MB/s eta 0:00:01\r\u001b[K     |████████████████████████████    | 1.1 MB 5.3 MB/s eta 0:00:01\r\u001b[K     |████████████████████████████▍   | 1.1 MB 5.3 MB/s eta 0:00:01\r\u001b[K     |████████████████████████████▋   | 1.1 MB 5.3 MB/s eta 0:00:01\r\u001b[K     |████████████████████████████▉   | 1.1 MB 5.3 MB/s eta 0:00:01\r\u001b[K     |█████████████████████████████▏  | 1.1 MB 5.3 MB/s eta 0:00:01\r\u001b[K     |█████████████████████████████▍  | 1.1 MB 5.3 MB/s eta 0:00:01\r\u001b[K     |█████████████████████████████▊  | 1.1 MB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████████████████████████████  | 1.1 MB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████████████████████████████▎ | 1.1 MB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████████████████████████████▌ | 1.2 MB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████████████████████████████▊ | 1.2 MB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████████████████████████████ | 1.2 MB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████████████████████████████▎| 1.2 MB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████████████████████████████▋| 1.2 MB 5.3 MB/s eta 0:00:01\r\u001b[K     |███████████████████████████████▉| 1.2 MB 5.3 MB/s eta 0:00:01\r\u001b[K     |████████████████████████████████| 1.2 MB 5.3 MB/s \n",
            "\u001b[?25hInstalling collected packages: sentencepiece\n",
            "Successfully installed sentencepiece-0.1.96\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fxRj2SiPm1vq"
      },
      "source": [
        "import tensorflow_hub as hub\n",
        "import tokenization1\n",
        "module_url = 'https://tfhub.dev/tensorflow/bert_en_uncased_L-12_H-768_A-12/2'\n",
        "bert_layer = hub.KerasLayer(module_url, trainable=True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cnakOzRRnXfB",
        "outputId": "420313c0-656b-4538-b09d-fcece0b6edd4"
      },
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "df=pd.read_csv(\"/content/FINAL_DATA2.csv\")\n",
        "df.cleanData=df.cleanData.astype(str)\n",
        "df.dtypes"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "twitter_id             int64\n",
              "content               object\n",
              "location              object\n",
              "search_word            int64\n",
              "word_length            int64\n",
              "cleanData             object\n",
              "location_extracted    object\n",
              "dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FHmvziSznQcc"
      },
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "X_train,X_test, y_train,y_test = train_test_split(df.cleanData,df.search_word, test_size = 0.2, stratify = df.search_word,random_state = 42)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EKkBAAHIniN8"
      },
      "source": [
        "vocab_file = bert_layer.resolved_object.vocab_file.asset_path.numpy()\n",
        "do_lower_case = bert_layer.resolved_object.do_lower_case.numpy()\n",
        "tokenizer = tokenization1.FullTokenizer(vocab_file, do_lower_case)\n",
        "\n",
        "def bert_encode(texts, tokenizer, max_len=171):\n",
        "    all_tokens = []\n",
        "    all_masks = []\n",
        "    all_segments = []\n",
        "    \n",
        "    for text in texts:\n",
        "        text = tokenizer.tokenize(text)\n",
        "            \n",
        "        text = text[:max_len-2]\n",
        "        input_sequence = [\"[CLS]\"] + text + [\"[SEP]\"]\n",
        "        pad_len = max_len - len(input_sequence)\n",
        "        \n",
        "        tokens = tokenizer.convert_tokens_to_ids(input_sequence) + [0] * pad_len\n",
        "        pad_masks = [1] * len(input_sequence) + [0] * pad_len\n",
        "        segment_ids = [0] * max_len\n",
        "        \n",
        "        all_tokens.append(tokens)\n",
        "        all_masks.append(pad_masks)\n",
        "        all_segments.append(segment_ids)\n",
        "    \n",
        "    return np.array(all_tokens), np.array(all_masks), np.array(all_segments)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-sJ_kBJ8oSkH"
      },
      "source": [
        "def build_model(bert_layer, max_len=171):\n",
        "    input_word_ids = tf.keras.Input(shape=(max_len,), dtype=tf.int32, name=\"input_word_ids\")\n",
        "    input_mask = tf.keras.Input(shape=(max_len,), dtype=tf.int32, name=\"input_mask\")\n",
        "    segment_ids = tf.keras.Input(shape=(max_len,), dtype=tf.int32, name=\"segment_ids\")\n",
        "\n",
        "    pooled_output, sequence_output = bert_layer([input_word_ids, input_mask, segment_ids])\n",
        "    clf_output = sequence_output[:, 0, :]\n",
        "    net = tf.keras.layers.Dense(64, activation='relu')(clf_output)\n",
        "    net = tf.keras.layers.Dropout(0.2)(net)\n",
        "    out = tf.keras.layers.Dense(3, activation='softmax')(net)\n",
        "    \n",
        "    model = tf.keras.models.Model(inputs=[input_word_ids, input_mask, segment_ids], outputs=out)\n",
        "    model.compile(tf.keras.optimizers.Adam(lr=1e-5), loss='sparse_categorical_crossentropy', metrics=['accuracy'])\n",
        "    \n",
        "    return model"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vOX4YLP9oZim"
      },
      "source": [
        "import keras\n",
        "max_len = 171\n",
        "train_input = bert_encode(X_train, tokenizer, max_len=max_len)\n",
        "test_input = bert_encode(X_test, tokenizer, max_len=max_len)\n",
        "train_labels = y_train"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QYKSio9Uoa0w",
        "outputId": "93dabebc-d424-47ef-e40d-bef9ee0cba24"
      },
      "source": [
        "import tensorflow as tf\n",
        "model = build_model(bert_layer, max_len=max_len)\n",
        "model.summary()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model_2\"\n",
            "__________________________________________________________________________________________________\n",
            " Layer (type)                   Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            " input_word_ids (InputLayer)    [(None, 171)]        0           []                               \n",
            "                                                                                                  \n",
            " input_mask (InputLayer)        [(None, 171)]        0           []                               \n",
            "                                                                                                  \n",
            " segment_ids (InputLayer)       [(None, 171)]        0           []                               \n",
            "                                                                                                  \n",
            " keras_layer_1 (KerasLayer)     [(None, 768),        109482241   ['input_word_ids[0][0]',         \n",
            "                                 (None, 171, 768)]                'input_mask[0][0]',             \n",
            "                                                                  'segment_ids[0][0]']            \n",
            "                                                                                                  \n",
            " tf.__operators__.getitem_2 (Sl  (None, 768)         0           ['keras_layer_1[2][1]']          \n",
            " icingOpLambda)                                                                                   \n",
            "                                                                                                  \n",
            " dense_6 (Dense)                (None, 64)           49216       ['tf.__operators__.getitem_2[0][0\n",
            "                                                                 ]']                              \n",
            "                                                                                                  \n",
            " dropout_4 (Dropout)            (None, 64)           0           ['dense_6[0][0]']                \n",
            "                                                                                                  \n",
            " dense_7 (Dense)                (None, 3)            195         ['dropout_4[0][0]']              \n",
            "                                                                                                  \n",
            "==================================================================================================\n",
            "Total params: 109,531,652\n",
            "Trainable params: 109,531,651\n",
            "Non-trainable params: 1\n",
            "__________________________________________________________________________________________________\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "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"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "aE3aiYVipN_d",
        "outputId": "16689265-9c5c-4168-eb82-fc67ed8fc19a"
      },
      "source": [
        "checkpoint = tf.keras.callbacks.ModelCheckpoint('model.h5', monitor='val_accuracy', mode='max',save_best_only=True, verbose=1)\n",
        "earlystopping = tf.keras.callbacks.EarlyStopping(monitor='val_loss',mode='min', patience=8, verbose=1)\n",
        "\n",
        "train_history = model.fit(\n",
        "    train_input, train_labels, \n",
        "    validation_data=(test_input,y_test),\n",
        "    epochs=60,\n",
        "    callbacks=[checkpoint, earlystopping],\n",
        "    batch_size=30,verbose=1)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/60\n",
            "66/66 [==============================] - ETA: 0s - loss: 0.5090 - accuracy: 0.8146\n",
            "Epoch 00001: val_accuracy improved from -inf to 0.90467, saving model to model.h5\n",
            "66/66 [==============================] - 138s 2s/step - loss: 0.5090 - accuracy: 0.8146 - val_loss: 0.2687 - val_accuracy: 0.9047\n",
            "Epoch 2/60\n",
            "66/66 [==============================] - ETA: 0s - loss: 0.2054 - accuracy: 0.9345\n",
            "Epoch 00002: val_accuracy improved from 0.90467 to 0.92495, saving model to model.h5\n",
            "66/66 [==============================] - 137s 2s/step - loss: 0.2054 - accuracy: 0.9345 - val_loss: 0.1974 - val_accuracy: 0.9249\n",
            "Epoch 3/60\n",
            "66/66 [==============================] - ETA: 0s - loss: 0.1306 - accuracy: 0.9543\n",
            "Epoch 00003: val_accuracy improved from 0.92495 to 0.94929, saving model to model.h5\n",
            "66/66 [==============================] - 137s 2s/step - loss: 0.1306 - accuracy: 0.9543 - val_loss: 0.1615 - val_accuracy: 0.9493\n",
            "Epoch 4/60\n",
            "66/66 [==============================] - ETA: 0s - loss: 0.0822 - accuracy: 0.9741\n",
            "Epoch 00004: val_accuracy did not improve from 0.94929\n",
            "66/66 [==============================] - 132s 2s/step - loss: 0.0822 - accuracy: 0.9741 - val_loss: 0.1760 - val_accuracy: 0.9493\n",
            "Epoch 5/60\n",
            "66/66 [==============================] - ETA: 0s - loss: 0.0410 - accuracy: 0.9883\n",
            "Epoch 00005: val_accuracy improved from 0.94929 to 0.95335, saving model to model.h5\n",
            "66/66 [==============================] - 137s 2s/step - loss: 0.0410 - accuracy: 0.9883 - val_loss: 0.1914 - val_accuracy: 0.9533\n",
            "Epoch 6/60\n",
            "66/66 [==============================] - ETA: 0s - loss: 0.0217 - accuracy: 0.9929\n",
            "Epoch 00006: val_accuracy did not improve from 0.95335\n",
            "66/66 [==============================] - 132s 2s/step - loss: 0.0217 - accuracy: 0.9929 - val_loss: 0.2335 - val_accuracy: 0.9432\n",
            "Epoch 7/60\n",
            "66/66 [==============================] - ETA: 0s - loss: 0.0111 - accuracy: 0.9975\n",
            "Epoch 00007: val_accuracy did not improve from 0.95335\n",
            "66/66 [==============================] - 132s 2s/step - loss: 0.0111 - accuracy: 0.9975 - val_loss: 0.2899 - val_accuracy: 0.9473\n",
            "Epoch 8/60\n",
            "66/66 [==============================] - ETA: 0s - loss: 0.0107 - accuracy: 0.9964\n",
            "Epoch 00008: val_accuracy did not improve from 0.95335\n",
            "66/66 [==============================] - 132s 2s/step - loss: 0.0107 - accuracy: 0.9964 - val_loss: 0.2497 - val_accuracy: 0.9432\n",
            "Epoch 9/60\n",
            "66/66 [==============================] - ETA: 0s - loss: 0.0108 - accuracy: 0.9964\n",
            "Epoch 00009: val_accuracy did not improve from 0.95335\n",
            "66/66 [==============================] - 132s 2s/step - loss: 0.0108 - accuracy: 0.9964 - val_loss: 0.2712 - val_accuracy: 0.9533\n",
            "Epoch 10/60\n",
            "66/66 [==============================] - ETA: 0s - loss: 0.0043 - accuracy: 0.9995\n",
            "Epoch 00010: val_accuracy did not improve from 0.95335\n",
            "66/66 [==============================] - 132s 2s/step - loss: 0.0043 - accuracy: 0.9995 - val_loss: 0.3586 - val_accuracy: 0.9513\n",
            "Epoch 11/60\n",
            "66/66 [==============================] - ETA: 0s - loss: 0.0025 - accuracy: 1.0000\n",
            "Epoch 00011: val_accuracy did not improve from 0.95335\n",
            "66/66 [==============================] - 132s 2s/step - loss: 0.0025 - accuracy: 1.0000 - val_loss: 0.3308 - val_accuracy: 0.9473\n",
            "Epoch 00011: early stopping\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ev-t25whsO2B",
        "outputId": "220edb91-ae11-480a-cdb9-0ec95762970b"
      },
      "source": [
        "model.evaluate(test_input,y_test) "
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "16/16 [==============================] - 11s 669ms/step - loss: 0.3308 - accuracy: 0.9473\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.3307691216468811, 0.9472616910934448]"
            ]
          },
          "metadata": {},
          "execution_count": 38
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Dq7k1zZtxvu_"
      },
      "source": [
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "nZg9PbU6yLR4",
        "outputId": "1403a596-4df0-4eab-cdc1-ca10bfd0104a"
      },
      "source": [
        "plt.plot(train_history.history['accuracy'],c='b',label='train accuracy')\n",
        "plt.plot(train_history.history['val_accuracy'],c='r',label='validation accuracy')\n",
        "plt.legend(loc='lower right')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "d8tZm95eyT-V",
        "outputId": "7d3474f1-ca2d-4da4-9f24-3c6884360ddc"
      },
      "source": [
        "plt.plot(train_history.history['loss'],c='b',label='train loss')\n",
        "plt.plot(train_history.history['val_loss'],c='r',label='validation loss')\n",
        "plt.legend(loc='lower right')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CgH2L_39zpty"
      },
      "source": [
        "X_new=[\"There is an accident in the area , 3 got dead on site and others are injured heavily\"]\n",
        "new_input = bert_encode(X_new, tokenizer, max_len=max_len)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "w0ieD5UxzsPs",
        "outputId": "6d0febc1-e2ed-4cc0-d347-4c94096807b6"
      },
      "source": [
        "pred=model.predict(new_input)\n",
        "pred"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0.9613284 , 0.03411213, 0.00455958]], dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 61
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "S_CBaQi31Zqq",
        "outputId": "246719ab-014a-4800-f84b-bb729851a176"
      },
      "source": [
        "import numpy as np\n",
        "labels = ['Accident','Traffic','Potholes']\n",
        "print(labels[np.argmax(pred)])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accident\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "43D8y9wo2Ykw",
        "outputId": "ff0bcff2-d409-4eb2-a497-8d0e4ab8dc52"
      },
      "source": [
        "from sklearn.metrics import classification_report\n",
        "\n",
        "y_pred=model.predict(test_input)\n",
        "predicted = np.argmax(y_pred, axis=1)\n",
        "report = classification_report(y_test, predicted)\n",
        "print(report)\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.87      0.79      0.83        66\n",
            "           1       0.96      0.98      0.97       418\n",
            "           2       1.00      0.44      0.62         9\n",
            "\n",
            "    accuracy                           0.95       493\n",
            "   macro avg       0.94      0.74      0.80       493\n",
            "weighted avg       0.95      0.95      0.94       493\n",
            "\n"
          ]
        }
      ]
    }
  ]
}