bert.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zgJ92o8-t3Ph",
        "outputId": "4114deb2-2478-4ef3-c409-898c3a29dafb"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "env: PATH=/home/jupyter/.local/bin:/opt/bin:/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/tools/node/bin:/tools/google-cloud-sdk/bin\n"
          ]
        }
      ],
      "source": [
        "import os\n",
        "import warnings\n",
        "\n",
        "warnings.filterwarnings(\"ignore\")\n",
        "os.environ[\"TF_CPP_MIN_LOG_LEVEL\"] = \"2\"\n",
        "\n",
        "# Set `PATH` to include the directory containing saved_model_cli\n",
        "PATH = %env PATH\n",
        "%env PATH=/home/jupyter/.local/bin:{PATH}"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install -U \"tensorflow-text==2.15.*\"\n",
        "!pip install -U \"tf-models-official==2.15.*\""
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cTcgcQyRFdr8",
        "outputId": "9572a87d-d418-4a88-9e30-9f6846ffc851"
      },
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: tensorflow-text==2.15.* in /usr/local/lib/python3.10/dist-packages (2.15.0)\n",
            "Requirement already satisfied: tensorflow-hub>=0.13.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow-text==2.15.*) (0.16.1)\n",
            "Requirement already satisfied: tensorflow<2.16,>=2.15.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow-text==2.15.*) (2.15.1)\n",
            "Requirement already satisfied: absl-py>=1.0.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (1.4.0)\n",
            "Requirement already satisfied: astunparse>=1.6.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (1.6.3)\n",
            "Requirement already satisfied: flatbuffers>=23.5.26 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (24.3.25)\n",
            "Requirement already satisfied: gast!=0.5.0,!=0.5.1,!=0.5.2,>=0.2.1 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (0.6.0)\n",
            "Requirement already satisfied: google-pasta>=0.1.1 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (0.2.0)\n",
            "Requirement already satisfied: h5py>=2.9.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (3.11.0)\n",
            "Requirement already satisfied: libclang>=13.0.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (18.1.1)\n",
            "Requirement already satisfied: ml-dtypes~=0.3.1 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (0.3.2)\n",
            "Requirement already satisfied: numpy<2.0.0,>=1.23.5 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (1.26.4)\n",
            "Requirement already satisfied: opt-einsum>=2.3.2 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (3.3.0)\n",
            "Requirement already satisfied: packaging in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (24.1)\n",
            "Requirement already satisfied: protobuf!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.20.3 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (3.20.3)\n",
            "Requirement already satisfied: setuptools in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (71.0.4)\n",
            "Requirement already satisfied: six>=1.12.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (1.16.0)\n",
            "Requirement already satisfied: termcolor>=1.1.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (2.4.0)\n",
            "Requirement already satisfied: typing-extensions>=3.6.6 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (4.12.2)\n",
            "Requirement already satisfied: wrapt<1.15,>=1.11.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (1.14.1)\n",
            "Requirement already satisfied: tensorflow-io-gcs-filesystem>=0.23.1 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (0.37.1)\n",
            "Requirement already satisfied: grpcio<2.0,>=1.24.3 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (1.64.1)\n",
            "Requirement already satisfied: tensorboard<2.16,>=2.15 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (2.15.2)\n",
            "Requirement already satisfied: tensorflow-estimator<2.16,>=2.15.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (2.15.0)\n",
            "Requirement already satisfied: keras<2.16,>=2.15.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (2.15.0)\n",
            "Requirement already satisfied: tf-keras>=2.14.1 in /usr/local/lib/python3.10/dist-packages (from tensorflow-hub>=0.13.0->tensorflow-text==2.15.*) (2.15.1)\n",
            "Requirement already satisfied: wheel<1.0,>=0.23.0 in /usr/local/lib/python3.10/dist-packages (from astunparse>=1.6.0->tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (0.44.0)\n",
            "Requirement already satisfied: google-auth<3,>=1.6.3 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.16,>=2.15->tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (2.27.0)\n",
            "Requirement already satisfied: google-auth-oauthlib<2,>=0.5 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.16,>=2.15->tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (1.2.1)\n",
            "Requirement already satisfied: markdown>=2.6.8 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.16,>=2.15->tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (3.7)\n",
            "Requirement already satisfied: requests<3,>=2.21.0 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.16,>=2.15->tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (2.32.3)\n",
            "Requirement already satisfied: tensorboard-data-server<0.8.0,>=0.7.0 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.16,>=2.15->tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (0.7.2)\n",
            "Requirement already satisfied: werkzeug>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.16,>=2.15->tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (3.0.4)\n",
            "Requirement already satisfied: cachetools<6.0,>=2.0.0 in /usr/local/lib/python3.10/dist-packages (from google-auth<3,>=1.6.3->tensorboard<2.16,>=2.15->tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (5.5.0)\n",
            "Requirement already satisfied: pyasn1-modules>=0.2.1 in /usr/local/lib/python3.10/dist-packages (from google-auth<3,>=1.6.3->tensorboard<2.16,>=2.15->tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (0.4.0)\n",
            "Requirement already satisfied: rsa<5,>=3.1.4 in /usr/local/lib/python3.10/dist-packages (from google-auth<3,>=1.6.3->tensorboard<2.16,>=2.15->tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (4.9)\n",
            "Requirement already satisfied: requests-oauthlib>=0.7.0 in /usr/local/lib/python3.10/dist-packages (from google-auth-oauthlib<2,>=0.5->tensorboard<2.16,>=2.15->tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (1.3.1)\n",
            "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.10/dist-packages (from requests<3,>=2.21.0->tensorboard<2.16,>=2.15->tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (3.3.2)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests<3,>=2.21.0->tensorboard<2.16,>=2.15->tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (3.8)\n",
            "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests<3,>=2.21.0->tensorboard<2.16,>=2.15->tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (2.0.7)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests<3,>=2.21.0->tensorboard<2.16,>=2.15->tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (2024.8.30)\n",
            "Requirement already satisfied: MarkupSafe>=2.1.1 in /usr/local/lib/python3.10/dist-packages (from werkzeug>=1.0.1->tensorboard<2.16,>=2.15->tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (2.1.5)\n",
            "Requirement already satisfied: pyasn1<0.7.0,>=0.4.6 in /usr/local/lib/python3.10/dist-packages (from pyasn1-modules>=0.2.1->google-auth<3,>=1.6.3->tensorboard<2.16,>=2.15->tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (0.6.0)\n",
            "Requirement already satisfied: oauthlib>=3.0.0 in /usr/local/lib/python3.10/dist-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib<2,>=0.5->tensorboard<2.16,>=2.15->tensorflow<2.16,>=2.15.0->tensorflow-text==2.15.*) (3.2.2)\n",
            "Requirement already satisfied: tf-models-official==2.15.* in /usr/local/lib/python3.10/dist-packages (2.15.0)\n",
            "Requirement already satisfied: Cython in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (3.0.11)\n",
            "Requirement already satisfied: Pillow in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (9.4.0)\n",
            "Requirement already satisfied: gin-config in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (0.5.0)\n",
            "Requirement already satisfied: google-api-python-client>=1.6.7 in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (2.137.0)\n",
            "Requirement already satisfied: immutabledict in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (4.2.0)\n",
            "Requirement already satisfied: kaggle>=1.3.9 in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (1.6.17)\n",
            "Requirement already satisfied: matplotlib in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (3.7.1)\n",
            "Requirement already satisfied: numpy>=1.20 in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (1.26.4)\n",
            "Requirement already satisfied: oauth2client in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (4.1.3)\n",
            "Requirement already satisfied: opencv-python-headless in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (4.10.0.84)\n",
            "Requirement already satisfied: pandas>=0.22.0 in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (2.1.4)\n",
            "Requirement already satisfied: psutil>=5.4.3 in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (5.9.5)\n",
            "Requirement already satisfied: py-cpuinfo>=3.3.0 in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (9.0.0)\n",
            "Requirement already satisfied: pycocotools in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (2.0.8)\n",
            "Requirement already satisfied: pyyaml>=6.0.0 in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (6.0.2)\n",
            "Requirement already satisfied: sacrebleu in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (2.4.3)\n",
            "Requirement already satisfied: scipy>=0.19.1 in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (1.13.1)\n",
            "Requirement already satisfied: sentencepiece in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (0.1.99)\n",
            "Requirement already satisfied: seqeval in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (1.2.2)\n",
            "Requirement already satisfied: six in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (1.16.0)\n",
            "Requirement already satisfied: tensorflow-datasets in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (4.9.6)\n",
            "Requirement already satisfied: tensorflow-hub>=0.6.0 in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (0.16.1)\n",
            "Requirement already satisfied: tensorflow-model-optimization>=0.4.1 in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (0.8.0)\n",
            "Requirement already satisfied: tensorflow-text~=2.15.0 in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (2.15.0)\n",
            "Requirement already satisfied: tensorflow~=2.15.0 in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (2.15.1)\n",
            "Requirement already satisfied: tf-slim>=1.1.0 in /usr/local/lib/python3.10/dist-packages (from tf-models-official==2.15.*) (1.1.0)\n",
            "Requirement already satisfied: httplib2<1.dev0,>=0.19.0 in /usr/local/lib/python3.10/dist-packages (from google-api-python-client>=1.6.7->tf-models-official==2.15.*) (0.22.0)\n",
            "Requirement already satisfied: google-auth!=2.24.0,!=2.25.0,<3.0.0.dev0,>=1.32.0 in /usr/local/lib/python3.10/dist-packages (from google-api-python-client>=1.6.7->tf-models-official==2.15.*) (2.27.0)\n",
            "Requirement already satisfied: google-auth-httplib2<1.0.0,>=0.2.0 in /usr/local/lib/python3.10/dist-packages (from google-api-python-client>=1.6.7->tf-models-official==2.15.*) (0.2.0)\n",
            "Requirement already satisfied: google-api-core!=2.0.*,!=2.1.*,!=2.2.*,!=2.3.0,<3.0.0.dev0,>=1.31.5 in /usr/local/lib/python3.10/dist-packages (from google-api-python-client>=1.6.7->tf-models-official==2.15.*) (2.19.2)\n",
            "Requirement already satisfied: uritemplate<5,>=3.0.1 in /usr/local/lib/python3.10/dist-packages (from google-api-python-client>=1.6.7->tf-models-official==2.15.*) (4.1.1)\n",
            "Requirement already satisfied: certifi>=2023.7.22 in /usr/local/lib/python3.10/dist-packages (from kaggle>=1.3.9->tf-models-official==2.15.*) (2024.8.30)\n",
            "Requirement already satisfied: python-dateutil in /usr/local/lib/python3.10/dist-packages (from kaggle>=1.3.9->tf-models-official==2.15.*) (2.8.2)\n",
            "Requirement already satisfied: requests in /usr/local/lib/python3.10/dist-packages (from kaggle>=1.3.9->tf-models-official==2.15.*) (2.32.3)\n",
            "Requirement already satisfied: tqdm in /usr/local/lib/python3.10/dist-packages (from kaggle>=1.3.9->tf-models-official==2.15.*) (4.66.5)\n",
            "Requirement already satisfied: python-slugify in /usr/local/lib/python3.10/dist-packages (from kaggle>=1.3.9->tf-models-official==2.15.*) (8.0.4)\n",
            "Requirement already satisfied: urllib3 in /usr/local/lib/python3.10/dist-packages (from kaggle>=1.3.9->tf-models-official==2.15.*) (2.0.7)\n",
            "Requirement already satisfied: bleach in /usr/local/lib/python3.10/dist-packages (from kaggle>=1.3.9->tf-models-official==2.15.*) (6.1.0)\n",
            "Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-packages (from pandas>=0.22.0->tf-models-official==2.15.*) (2024.1)\n",
            "Requirement already satisfied: tzdata>=2022.1 in /usr/local/lib/python3.10/dist-packages (from pandas>=0.22.0->tf-models-official==2.15.*) (2024.1)\n",
            "Requirement already satisfied: absl-py>=1.0.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow~=2.15.0->tf-models-official==2.15.*) (1.4.0)\n",
            "Requirement already satisfied: astunparse>=1.6.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow~=2.15.0->tf-models-official==2.15.*) (1.6.3)\n",
            "Requirement already satisfied: flatbuffers>=23.5.26 in /usr/local/lib/python3.10/dist-packages (from tensorflow~=2.15.0->tf-models-official==2.15.*) (24.3.25)\n",
            "Requirement already satisfied: gast!=0.5.0,!=0.5.1,!=0.5.2,>=0.2.1 in /usr/local/lib/python3.10/dist-packages (from tensorflow~=2.15.0->tf-models-official==2.15.*) (0.6.0)\n",
            "Requirement already satisfied: google-pasta>=0.1.1 in /usr/local/lib/python3.10/dist-packages (from tensorflow~=2.15.0->tf-models-official==2.15.*) (0.2.0)\n",
            "Requirement already satisfied: h5py>=2.9.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow~=2.15.0->tf-models-official==2.15.*) (3.11.0)\n",
            "Requirement already satisfied: libclang>=13.0.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow~=2.15.0->tf-models-official==2.15.*) (18.1.1)\n",
            "Requirement already satisfied: ml-dtypes~=0.3.1 in /usr/local/lib/python3.10/dist-packages (from tensorflow~=2.15.0->tf-models-official==2.15.*) (0.3.2)\n",
            "Requirement already satisfied: opt-einsum>=2.3.2 in /usr/local/lib/python3.10/dist-packages (from tensorflow~=2.15.0->tf-models-official==2.15.*) (3.3.0)\n",
            "Requirement already satisfied: packaging in /usr/local/lib/python3.10/dist-packages (from tensorflow~=2.15.0->tf-models-official==2.15.*) (24.1)\n",
            "Requirement already satisfied: protobuf!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.20.3 in /usr/local/lib/python3.10/dist-packages (from tensorflow~=2.15.0->tf-models-official==2.15.*) (3.20.3)\n",
            "Requirement already satisfied: setuptools in /usr/local/lib/python3.10/dist-packages (from tensorflow~=2.15.0->tf-models-official==2.15.*) (71.0.4)\n",
            "Requirement already satisfied: termcolor>=1.1.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow~=2.15.0->tf-models-official==2.15.*) (2.4.0)\n",
            "Requirement already satisfied: typing-extensions>=3.6.6 in /usr/local/lib/python3.10/dist-packages (from tensorflow~=2.15.0->tf-models-official==2.15.*) (4.12.2)\n",
            "Requirement already satisfied: wrapt<1.15,>=1.11.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow~=2.15.0->tf-models-official==2.15.*) (1.14.1)\n",
            "Requirement already satisfied: tensorflow-io-gcs-filesystem>=0.23.1 in /usr/local/lib/python3.10/dist-packages (from tensorflow~=2.15.0->tf-models-official==2.15.*) (0.37.1)\n",
            "Requirement already satisfied: grpcio<2.0,>=1.24.3 in /usr/local/lib/python3.10/dist-packages (from tensorflow~=2.15.0->tf-models-official==2.15.*) (1.64.1)\n",
            "Requirement already satisfied: tensorboard<2.16,>=2.15 in /usr/local/lib/python3.10/dist-packages (from tensorflow~=2.15.0->tf-models-official==2.15.*) (2.15.2)\n",
            "Requirement already satisfied: tensorflow-estimator<2.16,>=2.15.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow~=2.15.0->tf-models-official==2.15.*) (2.15.0)\n",
            "Requirement already satisfied: keras<2.16,>=2.15.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow~=2.15.0->tf-models-official==2.15.*) (2.15.0)\n",
            "Requirement already satisfied: tf-keras>=2.14.1 in /usr/local/lib/python3.10/dist-packages (from tensorflow-hub>=0.6.0->tf-models-official==2.15.*) (2.15.1)\n",
            "Requirement already satisfied: dm-tree~=0.1.1 in /usr/local/lib/python3.10/dist-packages (from tensorflow-model-optimization>=0.4.1->tf-models-official==2.15.*) (0.1.8)\n",
            "Requirement already satisfied: contourpy>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from matplotlib->tf-models-official==2.15.*) (1.3.0)\n",
            "Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.10/dist-packages (from matplotlib->tf-models-official==2.15.*) (0.12.1)\n",
            "Requirement already satisfied: fonttools>=4.22.0 in /usr/local/lib/python3.10/dist-packages (from matplotlib->tf-models-official==2.15.*) (4.53.1)\n",
            "Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from matplotlib->tf-models-official==2.15.*) (1.4.7)\n",
            "Requirement already satisfied: pyparsing>=2.3.1 in /usr/local/lib/python3.10/dist-packages (from matplotlib->tf-models-official==2.15.*) (3.1.4)\n",
            "Requirement already satisfied: pyasn1>=0.1.7 in /usr/local/lib/python3.10/dist-packages (from oauth2client->tf-models-official==2.15.*) (0.6.0)\n",
            "Requirement already satisfied: pyasn1-modules>=0.0.5 in /usr/local/lib/python3.10/dist-packages (from oauth2client->tf-models-official==2.15.*) (0.4.0)\n",
            "Requirement already satisfied: rsa>=3.1.4 in /usr/local/lib/python3.10/dist-packages (from oauth2client->tf-models-official==2.15.*) (4.9)\n",
            "Requirement already satisfied: portalocker in /usr/local/lib/python3.10/dist-packages (from sacrebleu->tf-models-official==2.15.*) (2.10.1)\n",
            "Requirement already satisfied: regex in /usr/local/lib/python3.10/dist-packages (from sacrebleu->tf-models-official==2.15.*) (2024.5.15)\n",
            "Requirement already satisfied: tabulate>=0.8.9 in /usr/local/lib/python3.10/dist-packages (from sacrebleu->tf-models-official==2.15.*) (0.9.0)\n",
            "Requirement already satisfied: colorama in /usr/local/lib/python3.10/dist-packages (from sacrebleu->tf-models-official==2.15.*) (0.4.6)\n",
            "Requirement already satisfied: lxml in /usr/local/lib/python3.10/dist-packages (from sacrebleu->tf-models-official==2.15.*) (4.9.4)\n",
            "Requirement already satisfied: scikit-learn>=0.21.3 in /usr/local/lib/python3.10/dist-packages (from seqeval->tf-models-official==2.15.*) (1.3.2)\n",
            "Requirement already satisfied: click in /usr/local/lib/python3.10/dist-packages (from tensorflow-datasets->tf-models-official==2.15.*) (8.1.7)\n",
            "Requirement already satisfied: promise in /usr/local/lib/python3.10/dist-packages (from tensorflow-datasets->tf-models-official==2.15.*) (2.3)\n",
            "Requirement already satisfied: pyarrow in /usr/local/lib/python3.10/dist-packages (from tensorflow-datasets->tf-models-official==2.15.*) (14.0.2)\n",
            "Requirement already satisfied: simple-parsing in /usr/local/lib/python3.10/dist-packages (from tensorflow-datasets->tf-models-official==2.15.*) (0.1.6)\n",
            "Requirement already satisfied: tensorflow-metadata in /usr/local/lib/python3.10/dist-packages (from tensorflow-datasets->tf-models-official==2.15.*) (1.15.0)\n",
            "Requirement already satisfied: toml in /usr/local/lib/python3.10/dist-packages (from tensorflow-datasets->tf-models-official==2.15.*) (0.10.2)\n",
            "Requirement already satisfied: array-record>=0.5.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow-datasets->tf-models-official==2.15.*) (0.5.1)\n",
            "Requirement already satisfied: etils>=1.6.0 in /usr/local/lib/python3.10/dist-packages (from etils[enp,epath,epy,etree]>=1.6.0; python_version < \"3.11\"->tensorflow-datasets->tf-models-official==2.15.*) (1.9.4)\n",
            "Requirement already satisfied: wheel<1.0,>=0.23.0 in /usr/local/lib/python3.10/dist-packages (from astunparse>=1.6.0->tensorflow~=2.15.0->tf-models-official==2.15.*) (0.44.0)\n",
            "Requirement already satisfied: fsspec in /usr/local/lib/python3.10/dist-packages (from etils[enp,epath,epy,etree]>=1.6.0; python_version < \"3.11\"->tensorflow-datasets->tf-models-official==2.15.*) (2024.6.1)\n",
            "Requirement already satisfied: importlib_resources in /usr/local/lib/python3.10/dist-packages (from etils[enp,epath,epy,etree]>=1.6.0; python_version < \"3.11\"->tensorflow-datasets->tf-models-official==2.15.*) (6.4.5)\n",
            "Requirement already satisfied: zipp in /usr/local/lib/python3.10/dist-packages (from etils[enp,epath,epy,etree]>=1.6.0; python_version < \"3.11\"->tensorflow-datasets->tf-models-official==2.15.*) (3.20.1)\n",
            "Requirement already satisfied: googleapis-common-protos<2.0.dev0,>=1.56.2 in /usr/local/lib/python3.10/dist-packages (from google-api-core!=2.0.*,!=2.1.*,!=2.2.*,!=2.3.0,<3.0.0.dev0,>=1.31.5->google-api-python-client>=1.6.7->tf-models-official==2.15.*) (1.65.0)\n",
            "Requirement already satisfied: proto-plus<2.0.0dev,>=1.22.3 in /usr/local/lib/python3.10/dist-packages (from google-api-core!=2.0.*,!=2.1.*,!=2.2.*,!=2.3.0,<3.0.0.dev0,>=1.31.5->google-api-python-client>=1.6.7->tf-models-official==2.15.*) (1.24.0)\n",
            "Requirement already satisfied: cachetools<6.0,>=2.0.0 in /usr/local/lib/python3.10/dist-packages (from google-auth!=2.24.0,!=2.25.0,<3.0.0.dev0,>=1.32.0->google-api-python-client>=1.6.7->tf-models-official==2.15.*) (5.5.0)\n",
            "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.10/dist-packages (from requests->kaggle>=1.3.9->tf-models-official==2.15.*) (3.3.2)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests->kaggle>=1.3.9->tf-models-official==2.15.*) (3.8)\n",
            "Requirement already satisfied: joblib>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from scikit-learn>=0.21.3->seqeval->tf-models-official==2.15.*) (1.4.2)\n",
            "Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.10/dist-packages (from scikit-learn>=0.21.3->seqeval->tf-models-official==2.15.*) (3.5.0)\n",
            "Requirement already satisfied: google-auth-oauthlib<2,>=0.5 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.16,>=2.15->tensorflow~=2.15.0->tf-models-official==2.15.*) (1.2.1)\n",
            "Requirement already satisfied: markdown>=2.6.8 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.16,>=2.15->tensorflow~=2.15.0->tf-models-official==2.15.*) (3.7)\n",
            "Requirement already satisfied: tensorboard-data-server<0.8.0,>=0.7.0 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.16,>=2.15->tensorflow~=2.15.0->tf-models-official==2.15.*) (0.7.2)\n",
            "Requirement already satisfied: werkzeug>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.16,>=2.15->tensorflow~=2.15.0->tf-models-official==2.15.*) (3.0.4)\n",
            "Requirement already satisfied: webencodings in /usr/local/lib/python3.10/dist-packages (from bleach->kaggle>=1.3.9->tf-models-official==2.15.*) (0.5.1)\n",
            "Requirement already satisfied: text-unidecode>=1.3 in /usr/local/lib/python3.10/dist-packages (from python-slugify->kaggle>=1.3.9->tf-models-official==2.15.*) (1.3)\n",
            "Requirement already satisfied: docstring-parser<1.0,>=0.15 in /usr/local/lib/python3.10/dist-packages (from simple-parsing->tensorflow-datasets->tf-models-official==2.15.*) (0.16)\n",
            "Requirement already satisfied: requests-oauthlib>=0.7.0 in /usr/local/lib/python3.10/dist-packages (from google-auth-oauthlib<2,>=0.5->tensorboard<2.16,>=2.15->tensorflow~=2.15.0->tf-models-official==2.15.*) (1.3.1)\n",
            "Requirement already satisfied: MarkupSafe>=2.1.1 in /usr/local/lib/python3.10/dist-packages (from werkzeug>=1.0.1->tensorboard<2.16,>=2.15->tensorflow~=2.15.0->tf-models-official==2.15.*) (2.1.5)\n",
            "Requirement already satisfied: oauthlib>=3.0.0 in /usr/local/lib/python3.10/dist-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib<2,>=0.5->tensorboard<2.16,>=2.15->tensorflow~=2.15.0->tf-models-official==2.15.*) (3.2.2)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import datetime\n",
        "import shutil\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "import tensorflow as tf\n",
        "import tensorflow_hub as hub\n",
        "import tensorflow_text as text\n",
        "from google.cloud import aiplatform\n",
        "from official.nlp import optimization  # to create AdamW optmizer\n",
        "\n",
        "tf.get_logger().setLevel(\"ERROR\")"
      ],
      "metadata": {
        "id": "nessWr1P8sM0"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(\"Num GPUs Available: \", len(tf.config.list_physical_devices(\"GPU\")))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UtCVylNH9i-E",
        "outputId": "a3b236cb-ae6f-464e-d256-9549dc5db276"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Num GPUs Available:  1\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "url = \"https://ai.stanford.edu/~amaas/data/sentiment/aclImdb_v1.tar.gz\"\n",
        "\n",
        "# Set a path to a folder outside the git repo. This is important so data won't get indexed by git on Jupyter lab\n",
        "path = \"/content/\"\n",
        "\n",
        "dataset = tf.keras.utils.get_file(\n",
        "    \"aclImdb_v1.tar.gz\", url, untar=True, cache_dir=path, cache_subdir=\"\"\n",
        ")\n",
        "\n",
        "dataset_dir = os.path.join(os.path.dirname(dataset), \"aclImdb\")\n",
        "\n",
        "train_dir = os.path.join(dataset_dir, \"train\")\n",
        "\n",
        "# remove unused folders to make it easier to load the data\n",
        "remove_dir = os.path.join(train_dir, \"unsup\")\n",
        "shutil.rmtree(remove_dir)"
      ],
      "metadata": {
        "id": "Gektuf5P9mDd"
      },
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "AUTOTUNE = tf.data.AUTOTUNE\n",
        "batch_size = 32\n",
        "seed = 42\n",
        "\n",
        "raw_train_ds = tf.keras.preprocessing.text_dataset_from_directory(\n",
        "    path + \"aclImdb/train\",\n",
        "    batch_size=batch_size,\n",
        "    validation_split=0.2,\n",
        "    subset=\"training\",\n",
        "    seed=seed,\n",
        ")\n",
        "\n",
        "class_names = raw_train_ds.class_names\n",
        "train_ds = raw_train_ds.cache().prefetch(buffer_size=AUTOTUNE)\n",
        "\n",
        "val_ds = tf.keras.preprocessing.text_dataset_from_directory(\n",
        "    path + \"aclImdb/train\",\n",
        "    batch_size=batch_size,\n",
        "    validation_split=0.2,\n",
        "    subset=\"validation\",\n",
        "    seed=seed,\n",
        ")\n",
        "\n",
        "val_ds = val_ds.cache().prefetch(buffer_size=AUTOTUNE)\n",
        "\n",
        "test_ds = tf.keras.preprocessing.text_dataset_from_directory(\n",
        "    path + \"aclImdb/test\", batch_size=batch_size\n",
        ")\n",
        "\n",
        "test_ds = test_ds.cache().prefetch(buffer_size=AUTOTUNE)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZShMLtOt9-mz",
        "outputId": "59791cd2-d781-4a0f-943c-e495324d654f"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Found 25000 files belonging to 2 classes.\n",
            "Using 20000 files for training.\n",
            "Found 25000 files belonging to 2 classes.\n",
            "Using 5000 files for validation.\n",
            "Found 25000 files belonging to 2 classes.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "for text_batch, label_batch in train_ds.take(1):\n",
        "    for i in range(3):\n",
        "        print(f\"Review: {text_batch.numpy()[i]}\")\n",
        "        label = label_batch.numpy()[i]\n",
        "        print(f\"Label : {label} ({class_names[label]})\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gIWIXCBa_hfX",
        "outputId": "6a065eff-1f4d-4151-8b8d-cecc6775b6f0"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Review: b'\"Pandemonium\" is a horror movie spoof that comes off more stupid than funny. Believe me when I tell you, I love comedies. Especially comedy spoofs. \"Airplane\", \"The Naked Gun\" trilogy, \"Blazing Saddles\", \"High Anxiety\", and \"Spaceballs\" are some of my favorite comedies that spoof a particular genre. \"Pandemonium\" is not up there with those films. Most of the scenes in this movie had me sitting there in stunned silence because the movie wasn\\'t all that funny. There are a few laughs in the film, but when you watch a comedy, you expect to laugh a lot more than a few times and that\\'s all this film has going for it. Geez, \"Scream\" had more laughs than this film and that was more of a horror film. How bizarre is that?<br /><br />*1/2 (out of four)'\n",
            "Label : 0 (neg)\n",
            "Review: b\"David Mamet is a very interesting and a very un-equal director. His first movie 'House of Games' was the one I liked best, and it set a series of films with characters whose perspective of life changes as they get into complicated situations, and so does the perspective of the viewer.<br /><br />So is 'Homicide' which from the title tries to set the mind of the viewer to the usual crime drama. The principal characters are two cops, one Jewish and one Irish who deal with a racially charged area. The murder of an old Jewish shop owner who proves to be an ancient veteran of the Israeli Independence war triggers the Jewish identity in the mind and heart of the Jewish detective.<br /><br />This is were the flaws of the film are the more obvious. The process of awakening is theatrical and hard to believe, the group of Jewish militants is operatic, and the way the detective eventually walks to the final violent confrontation is pathetic. The end of the film itself is Mamet-like smart, but disappoints from a human emotional perspective.<br /><br />Joe Mantegna and William Macy give strong performances, but the flaws of the story are too evident to be easily compensated.\"\n",
            "Label : 0 (neg)\n",
            "Review: b'Great documentary about the lives of NY firefighters during the worst terrorist attack of all time.. That reason alone is why this should be a must see collectors item.. What shocked me was not only the attacks, but the\"High Fat Diet\" and physical appearance of some of these firefighters. I think a lot of Doctors would agree with me that,in the physical shape they were in, some of these firefighters would NOT of made it to the 79th floor carrying over 60 lbs of gear. Having said that i now have a greater respect for firefighters and i realize becoming a firefighter is a life altering job. The French have a history of making great documentary\\'s and that is what this is, a Great Documentary.....'\n",
            "Label : 1 (pos)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# defining the URL of the smallBERT model to use\n",
        "tfhub_handle_encoder = (\n",
        "    \"https://tfhub.dev/tensorflow/small_bert/bert_en_uncased_L-4_H-512_A-8/1\"\n",
        ")\n",
        "\n",
        "# defining the corresponding preprocessing model for the BERT model above\n",
        "tfhub_handle_preprocess = (\n",
        "    \"https://tfhub.dev/tensorflow/bert_en_uncased_preprocess/3\"\n",
        ")\n",
        "\n",
        "print(f\"BERT model selected           : {tfhub_handle_encoder}\")\n",
        "print(f\"Preprocess model auto-selected: {tfhub_handle_preprocess}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EIXyjstcARx9",
        "outputId": "b08f2c88-f648-4d3c-cec9-f4dfd1b3e5bb"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "BERT model selected           : https://tfhub.dev/tensorflow/small_bert/bert_en_uncased_L-4_H-512_A-8/1\n",
            "Preprocess model auto-selected: https://tfhub.dev/tensorflow/bert_en_uncased_preprocess/3\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "bert_preprocess_model = hub.KerasLayer(tfhub_handle_preprocess)"
      ],
      "metadata": {
        "id": "reeEsYB2AYFk"
      },
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "text_test = [\"this is such an amazing movie!\"]\n",
        "text_preprocessed = bert_preprocess_model(text_test)\n",
        "\n",
        "# This print box will help you inspect the keys in the pre-processed dictionary\n",
        "print(f\"Keys       : {list(text_preprocessed.keys())}\")\n",
        "\n",
        "# 1. input_word_ids is the ids for the words in the tokenized sentence\n",
        "print(f'Shape      : {text_preprocessed[\"input_word_ids\"].shape}')\n",
        "print(f'Word Ids   : {text_preprocessed[\"input_word_ids\"][0, :12]}')\n",
        "\n",
        "# 2. input_mask is the tokens which we are masking (masked language model)\n",
        "print(f'Input Mask : {text_preprocessed[\"input_mask\"][0, :12]}')\n",
        "\n",
        "# 3. input_type_ids is the sentence id of the input sentence.\n",
        "print(f'Type Ids   : {text_preprocessed[\"input_type_ids\"][0, :12]}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "L7gtqGN2Adsa",
        "outputId": "a05efb52-ecdf-453d-f6ac-3a4410fbcb7b"
      },
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Keys       : ['input_type_ids', 'input_mask', 'input_word_ids']\n",
            "Shape      : (1, 128)\n",
            "Word Ids   : [ 101 2023 2003 2107 2019 6429 3185  999  102    0    0    0]\n",
            "Input Mask : [1 1 1 1 1 1 1 1 1 0 0 0]\n",
            "Type Ids   : [0 0 0 0 0 0 0 0 0 0 0 0]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "bert_model = hub.KerasLayer(tfhub_handle_encoder)"
      ],
      "metadata": {
        "id": "7uRLyOhlBd5i"
      },
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "bert_results = bert_model(text_preprocessed)\n",
        "\n",
        "print(f\"Loaded BERT: {tfhub_handle_encoder}\")\n",
        "print(f'Pooled Outputs Shape:{bert_results[\"pooled_output\"].shape}')\n",
        "print(f'Pooled Outputs Values:{bert_results[\"pooled_output\"][0, :12]}')\n",
        "print(f'Sequence Outputs Shape:{bert_results[\"sequence_output\"].shape}')\n",
        "print(f'Sequence Outputs Values:{bert_results[\"sequence_output\"][0, :12]}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1l9c2DXeBjZ3",
        "outputId": "32862fa1-db59-41da-a997-e1154205639d"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Loaded BERT: https://tfhub.dev/tensorflow/small_bert/bert_en_uncased_L-4_H-512_A-8/1\n",
            "Pooled Outputs Shape:(1, 512)\n",
            "Pooled Outputs Values:[ 0.76262885  0.99280983 -0.18611874  0.3667385   0.1523372   0.6550446\n",
            "  0.9681154  -0.94862705  0.0021616  -0.9877732   0.06842716 -0.97630596]\n",
            "Sequence Outputs Shape:(1, 128, 512)\n",
            "Sequence Outputs Values:[[-0.2894634   0.34321257  0.33231494 ...  0.21300837  0.7102077\n",
            "  -0.05771141]\n",
            " [-0.28742063  0.31981015 -0.23018579 ...  0.58455044 -0.21329722\n",
            "   0.72692055]\n",
            " [-0.66157013  0.6887687  -0.87433    ...  0.10877256 -0.26173246\n",
            "   0.47855335]\n",
            " ...\n",
            " [-0.2256114  -0.2892561  -0.0706443  ...  0.47565973  0.83277154\n",
            "   0.40025324]\n",
            " [-0.29824266 -0.27473152 -0.05450485 ...  0.48849776  1.0955359\n",
            "   0.18163376]\n",
            " [-0.4437817   0.00930699  0.07223739 ...  0.17290172  1.1833248\n",
            "   0.07898028]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def build_classifier_model(dropout_rate=0.1):\n",
        "    text_input = tf.keras.layers.Input(shape=(), dtype=tf.string, name=\"text\")\n",
        "\n",
        "    preprocessing_layer = hub.KerasLayer(tfhub_handle_preprocess, name=\"preprocessing\")\n",
        "    encoder_inputs = preprocessing_layer(text_input)\n",
        "\n",
        "    encoder = hub.KerasLayer(tfhub_handle_encoder, trainable=True, name=\"BERT_encoder\")\n",
        "    outputs = encoder(encoder_inputs)\n",
        "\n",
        "    net = outputs[\"pooled_output\"]\n",
        "    net = tf.keras.layers.Dropout(dropout_rate)(net)\n",
        "    net = tf.keras.layers.Dense(1, activation=\"sigmoid\", name=\"classifier\")(net)\n",
        "    return tf.keras.Model(text_input, net)\n",
        "\n",
        "\n",
        "# Let's check that the model runs with the output of the preprocessing model.\n",
        "dropout_rate = 0.15\n",
        "classifier_model = build_classifier_model(dropout_rate)\n",
        "bert_raw_result = classifier_model(tf.constant(text_test))\n",
        "print(bert_raw_result)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JVfszgsEBmPH",
        "outputId": "5213cf3a-a8a9-4dfc-a514-4901265ec9eb"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "tf.Tensor([[0.65321875]], shape=(1, 1), dtype=float32)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "tf.keras.utils.plot_model(classifier_model)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 466
        },
        "id": "i4vWGfOlGxdY",
        "outputId": "34a8b33e-b4b3-4d15-e3cb-b6ee01c50011"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPsAAAHBCAYAAACv5M3ZAAAABmJLR0QA/wD/AP+gvaeTAAAgAElEQVR4nO3deVRUZ5oG8OdSFFUUUAUoiLLJ4obiHIkaNZrGxCxqx1EWQSGKicalE5c2kQk6Dp24BDXiRKETW2MnZo6y6Lilk9iacemJIWqbwaCi0VaCiCCyWggFvPOHQ01KBIqtLvC9v3PqHL33u/e+33froe5Si0REBMZYd5dmJXcFjDHL4LAzJggOO2OC4LAzJghruQtoqzNnzmDz5s1yl8G6ubS0NLlLaLMu/8r+yy+/ID09Xe4y2iw9PR25ublyl8Eek5ub2y2eX0A3eGWv19X/8kqShGXLlmH69Olyl8J+JTU1FREREXKX0S66/Cs7Y8w8HHbGBMFhZ0wQHHbGBMFhZ0wQHHbGBMFhZ0wQHHbGBMFhZ0wQHHbGBMFhZ0wQHHbGBMFhZ0wQHHbGBMFh74K+//57DBo0CFZWVpAkCb169cKaNWvkLgv79u2Dr68vJEmCJElwc3NDdHS03GWx/9NtPs8uklGjRuHy5ct4+eWX8c033yA7OxuOjo5yl4XQ0FCEhobC398f9+7dQ35+vtwlsV8R8pW9srISY8aM6TLr7axE629XJ2TYd+7ciYKCgi6z3s5KtP52dcKFfenSpVi+fDmuX78OSZLg7+8PAKitrcXq1avh5eUFW1tbDB06FCkpKQCAP//5z7C3t4ckSXBycsKBAwdw7tw5eHt7Q6FQYObMmY2u15KSk5NhZ2cHjUaDgwcPYuLEidBqtfDw8MCePXsAAB999BHUajVcXV2xYMEC9O7dG2q1GmPGjEFGRgYAYPHixbCxsYGbm5tx3b/73e9gZ2cHSZJw7969dunv6dOnERAQAJ1OB7VajcDAQHzzzTcAgLlz5xrP/f38/HDhwgUAwJw5c6DRaKDT6XDo0KEm99uGDRug0Wjg4OCAgoICLF++HO7u7sjOzm7TOHdZ1MWlpKRQS7sRGhpKfn5+JtPefvttUqlUlJ6eTsXFxRQXF0dWVlZ09uxZIiK6dOkSaTQamj17tnGZd999l3bs2NHkes0FgFJSUlq0zEsvvUQAqLi42Dht5cqVBICOHz9OpaWlVFBQQOPGjSM7Ozuqrq4mIqL58+eTnZ0dXbp0iR4+fEhZWVk0YsQIcnBwoJycHCIiioqKol69eplsb+PGjQSACgsLm+yvn58f6XS6ZutPS0uj+Ph4un//PhUVFdGoUaOoR48exvmhoaGkUCjo9u3bJsvNnDmTDh06RETN77f68ViyZAlt3bqVQkJC6PLly83WVq81z69OKlW4V/YnefjwIZKTkzFt2jSEhobC0dERq1atglKpxK5duwAAgwYNQmJiIj777DP8x3/8B/bs2YOqqiq8/vrrMlf/ZGPGjIFWq4WLiwsiIyPx4MED5OTkGOdbW1tj0KBBUKlUCAgIQHJyMsrLy439tYSwsDD827/9G5ycnODs7IwpU6agqKgIhYWFAICFCxeitrbWpKaysjKcPXsWkyZNMmu/1fvggw/w5ptvYt++fRg4cKDF+tiZcNgBZGdnQ6/XY8iQIcZptra2cHNzw5UrV4zT3njjDYSFhWHBggVITU3Fhg0b5Ci3xWxsbAAABoOh0TbDhw+HRqMx6a+lKZVKAI9OqQDgueeeQ//+/fHpp5+C/u/3R/fu3YvIyEgoFAqz9xt7hMMO4MGDBwCAVatWGc8TJUnCrVu3oNfrTdquXbsWFRUV3fLClEqlMr6qWsKXX36J4OBguLi4QKVSYcWKFSbzJUnCggULcOPGDRw/fhwA8PnnnxuPplqy3xiHHQDg4uICAEhMTAQRmTzOnDljbGcwGLBkyRJs3rwZZ86c6RRvZGkvBoMBJSUl8PDw6NDtnDp1ComJicjJycG0adPg5uaGjIwMlJaWIiEhoUH7mJgYqNVq7NixA9nZ2dBqtfD29gZg/n5jj/CbagB4enpCrVbjxx9/bLLdW2+9hXnz5iEkJAS3b9/G+++/jxdffBGjR4+2UKUd58SJEyAijBo1CsCjc/qmDvtb6/z587Czs8PFixdhMBiwaNEi+Pr6Anj0Sv44JycnREREYO/evXBwcMC8efOM88zdb+wRIV/ZnZ2dkZeXh5s3b6K8vBwKhQJz5szBnj17kJycjLKyMtTW1iI3Nxd37twBACQlJcHd3R0hISEAgHXr1iEgIABRUVEoKyt74no7Iiztpa6uDsXFxaipqUFmZiaWLl0KLy8vxMTEAAD8/f1x//59HDhwAAaDAYWFhbh165bJOlrSX4PBgLt37+LEiROws7ODl5cXAODYsWN4+PAhrl27Zrz197iFCxeiqqoKR44cwSuvvGKcrlarm91v7Ffkug/QXlpza+Tvf/87eXt7k62tLY0dO5by8/OpqqqKYmNjycvLi6ytrcnFxYVCQ0MpKyuLXnnlFZIkiZydnem7774jIqJly5aRlZUVASCdTkfnzp174nrNhRbcevv+++9p8ODBxu27ubnR2rVrKSkpiTQaDQGgfv360fXr12n79u2k1WoJAHl7e9PVq1dp/vz5pFQqyd3dnaytrUmr1dLUqVPp+vXrxm0UFRXR+PHjSa1Wk4+PD7311lv0zjvvEADy9/ennJycBv394x//SH5+fgSgycf+/fuJiCg2NpacnZ3J0dGRwsPDadu2bQSA/Pz8jLcA6w0bNozefffdBmPR1H5LSEggW1tbAkCenp60e/dus/dHve50663L96K77IyWhL2t5s+fT87OzhbZVnuZNGkS3bhxw+Lb7S7PL+L77OKqv73VWf36lCAzMxNqtRo+Pj4yVtT18QU61inFxsZi4cKFICLMmTMHu3fvlrukLo9f2QUTFxeHXbt2obS0FD4+Pp32t8c1Gg0GDhyICRMmID4+HgEBAXKX1OVx2AWzbt06VFVVgYjwj3/8A2FhYXKX9ERr1qxBbW0tcnJyTK7As9bjsDMmCA47Y4LgsDMmCA47Y4LgsDMmCA47Y4LgsDMmCA47Y4LgsDMmCA47Y4LgsDMmCA47Y4LgsDMmiG7zefbw8HC5S2izxMREpKWlyV0G+5Xc3Fy5S2g3EtH/fft+F3XmzBls3rxZ7jK6jMLCQly+fBnPPvus3KV0Kd3gj3Balw87a5nU1FRERESAd7tw0vicnTFBcNgZEwSHnTFBcNgZEwSHnTFBcNgZEwSHnTFBcNgZEwSHnTFBcNgZEwSHnTFBcNgZEwSHnTFBcNgZEwSHnTFBcNgZEwSHnTFBcNgZEwSHnTFBcNgZEwSHnTFBcNgZEwSHnTFBcNgZEwSHnTFBcNgZEwSHnTFBcNgZEwSHnTFBcNgZEwSHnTFBcNgZEwSHnTFBWMtdAOs4ubm5mD17Nmpra43T7t27B2trawQHB5u0HTBgAD755BMLV8gsicPejXl4eODmzZu4ceNGg3knT540+f+4ceMsVRaTCR/Gd3OzZs2CUqlstl1kZKQFqmFy4rB3c1FRUTAYDE22CQgIwODBgy1UEZMLh72b8/f3x9ChQyFJ0hPnK5VKzJ4928JVMTlw2AUwa9YsKBSKJ86rqanB9OnTLVwRkwOHXQAzZsxAXV1dg+mSJOHpp59G3759LV8UszgOuwD69OmDMWPGwMrKdHcrFArMmjVLpqqYpXHYBfHqq682mEZECA0NlaEaJgcOuyDCw8NNXtkVCgUmTJgAV1dXGatilsRhF4STkxNefPFF44U6IkJ0dLTMVTFL4rALJDo62nihztraGlOmTJG5ImZJHHaBTJkyBSqVyvhvrVYrc0XMksx+b3xqampH1sEsJCgoCN999x18fHx4n3YDnp6eGD16tFltJSIisxo28g4sxph8wsLCkJaWZk7TtBYdxqekpICI+NGFH9XV1VixYkWj81NSUgBA9jr50fwjLCysRX8Y+JxdMEqlEvHx8XKXwWTAYReQra2t3CUwGXDYGRMEh50xQXDYGRMEh50xQXDYGRMEh50xQXDYGRMEh50xQXDYGRMEh50xQXDYGRMEh50xQXDYO5G//OUv0Ol0OHz4sNylmG3fvn3w9fWFJEmQJAmenp7YuXOncf7Jkyfh7u4OSZLg5uaG7du3d4o63dzchPsOPv4V106EyKzvEelUQkNDERoaCn9/f9y7dw+//PKLyfxnn30WkyZNgpWVFT7++GPZvgTl8Trz8/NlqUNOHPZOZPLkySgtLZW7jHZTV1eHuXPnQq1WIykpib/tSGZ8GN9NERHS0tJkO2yuq6vDa6+9Bo1Gg+TkZA56J9AhYf/oo4+gVqvh6uqKBQsWoHfv3lCr1RgzZgwyMjIAABs2bIBGo4GDgwMKCgqwfPlyuLu7Izs7G7W1tVi9ejW8vLxga2uLoUOHGr8uqa3rJiJs3rwZgwYNgkqlgpOTE6ZOnYorV66Y9GH37t0YPnw41Go17Ozs0LdvX7z//vsA0GR9wKPz1JEjR0Kj0UCr1SIwMBBlZWVNzvvb3/4GLy8vSJKEbdu2AQCSk5NhZ2cHjUaDgwcPYuLEidBqtfDw8MCePXuM26utrcW6deswYMAA2NraomfPnvDx8cG6detk+dHGuro6xMTEQKfTGfvyuKbGsKn9d/r0aQQEBECn00GtViMwMBDffPONcb1NjX1LNLWduXPnGs/9/fz8cOHCBQDAnDlzoNFooNPpcOjQoVb3scOQmQBQSkqKuc1p/vz5ZGdnR5cuXaKHDx9SVlYWjRgxghwcHCgnJ4eIiFauXEkAaMmSJbR161YKCQmhy5cv09tvv00qlYrS09OpuLiY4uLiyMrKis6ePdvmda9evZpsbGxo9+7dVFJSQpmZmRQUFEQ9e/ak/Px8IiJKTEwkALR+/XoqKiqi+/fv0yeffEJRUVFERE3WV1FRQVqtlhISEqiyspLy8/MpJCSECgsLm5xHRPTLL78QANq6datxHOv7cfz4cSotLaWCggIaN24c2dnZUXV1NRERrV27lhQKBR08eJD0ej2dP3+eevXqRcHBwWbvr3opKSnUgqeFkZ+fH+l0OqqpqaGoqChSKpWUnZ3daPvm9nFj+y8tLY3i4+Pp/v37VFRURKNGjaIePXoQETU7vr+uszlNbYeIKDQ0lBQKBd2+fdtkuZkzZ9KhQ4fa1EdzhYWFUVhYmLnNUzs07I8P6tmzZwkA/eEPfyCi/+9sZWWlsU1lZSVpNBqKjIw0TtPr9aRSqWjRokVtWrderyd7e3uTdRMR/fDDDwSA3nvvPaquriZHR0caP368SZuamhrasmVLs/X99NNPBICOHDnSYEyamkfUdNh/3Y+kpCQCQD///DMREY0YMYJGjhxpsq433niDrKysqKqq6onbakxbwu7g4EAzZsygoKAgAkCDBw+mioqKBm3N2cdP6veTrFu3jgBQQUFBs+NbX6c5YW9qO0REx44dIwC0Zs0aY5vS0lLq168f1dTUtGsfG9PSsFv0nH348OHQaDQNDpl/LTs7G3q9HkOGDDFOs7W1hZubW5PLmbPurKwsVFRUYPjw4SbTR4wYARsbG2RkZCAzMxMlJSV46aWXTNooFAosWbKk2fp8fX3h6uqK6OhoxMfH4+bNm8Z2Tc1rCRsbGwCAwWAAADx8+LDBlfza2loolcpGf5e9I+j1evzmN7/B+fPnMW3aNGRlZWHu3LkN2rV2Hz+JUqkE8Ki/7TW+zW0HAJ577jn0798fn376qXHs9+7di8jISCgUinbtY3ux+AU6lUqFwsLCRuc/ePAAALBq1SrjeZEkSbh16xb0en2b1l1SUgIAsLe3bzDP0dER5eXlxvM7R0fHVtVna2uLb7/9FmPHjsXatWvh6+uLyMhIVFZWNjmvLSZNmoTz58/j4MGDqKysxLlz53DgwAH89re/tWjY7e3tMX/+fADArl274Ovri7179yIxMdGkXVv28Zdffong4GC4uLhApVJhxYoVxnntOb5NbQd49DsKCxYswI0bN3D8+HEAwOeff47XX3+9zX3sKBYNu8FgQElJCTw8PBpt4+LiAgBITExs8D3ZZ86cadO66wNcXl7eYF79sn369AEA3Lt3r9X1DR48GIcPH0ZeXh5iY2ORkpKCTZs2NTuvteLj4/Hcc88hJiYGWq0WISEhmD59Ov70pz+1ab1todPpkJaWZgzKqVOnjPNau49zcnIwbdo0uLm5ISMjA6WlpUhISDBp05bxPXXqFBITE83aDgDExMRArVZjx44dyM7Ohlarhbe3d5v62JEsGvYTJ06AiDBq1KhG23h6ekKtVuPHH39s93UPGTIE9vb2OHfunMn0jIwMVFdX46mnnkLfvn3h7OyMo0ePtqq+vLw8XLp0CcCjHb5+/XoEBQXh0qVLTc5ri6ysLFy/fh2FhYUwGAzIyclBcnIynJyc2rTetgoKCkJiYiJqamowffp05OXlAWj9Pr548SIMBgMWLVoEX19fqNVqk1t6bR3f8+fPw87Ortnt1HNyckJERAQOHDiATZs2Yd68ecZ5re1jR+rQsNfV1aG4uBg1NTXIzMzE0qVL4eXlhZiYmEaXUavVmDNnDvbs2YPk5GSUlZWhtrYWubm5uHPnTpvXvXz5cuzfvx9ffPEFysrKcPHiRSxcuBC9e/fG/PnzoVKpEBcXh1OnTmHx4sW4ffs26urqUF5ejkuXLjVbX15eHhYsWIArV66guroaFy5cwK1btzBq1Kgm57XFm2++CS8vL1RUVLRpPR1h4cKFmDFjBu7evYvw8HAYDAaz9/HjvLy8AADHjh3Dw4cPce3aNePtVgCtHl+DwYC7d+/ixIkTsLOza3Y7j/evqqoKR44cwSuvvGKc3to+dihzL+WhFVfjlUolubu7k7W1NWm1Wpo6dSpdv36diIgSEhLI1taWAJCnpyft3r3buGxVVRXFxsaSl5cXWVtbk4uLC4WGhlJWVlab111XV0cbN26kfv36kVKpJCcnJ5o2bVqD20Tbtm2jwMBAUqvVpFaradiwYZSUlNRsfTdv3qQxY8aQk5MTKRQK6tOnD61cuZJqamqanLd161Zyc3MjAKTRaGjKlCmUlJREGo2GAFC/fv3o+vXrtH37dtJqtQSAvL296erVq/Ttt99Sjx49CIDxoVQqadCgQbRv3z6z9xlRy6/G79+/n/z8/Izb9fDwoLi4OJM25eXlNGDAAAJArq6utHPnzibHsKn9FxsbS87OzuTo6Ejh4eG0bds2AkB+fn50+vTpRsf38Tobe+zfv7/Z7dTf3q03bNgwevfddxuMTWv7aK5OdevN2dnZ7PYt0ZHr7oqSkpJo6dKlJtOqqqpo2bJlpFKpSK/Xm72u1t56E9mkSZPoxo0bFt9uS8Peoe+Nr79N0dXW3ZXk5+dj8eLFDc4NbWxs4OXlBYPBAIPBwD/51I4MBoPxVlxmZibUajV8fHxkrqp5/N74Ls7W1hZKpRI7d+7E3bt3YTAYkJeXhx07dmD16tWIjIyEVquVu8xuJTY2FteuXcPVq1cxZ84c49uoO7sOCXtcXBx27dqF0tJS+Pj4ID09vUusuyvS6XQ4evQofvrpJ/Tv3x+2trYICAjArl278MEHH+Czzz6Tu8RuR6PRYODAgZgwYQLi4+MREBAgd0lmkYjM+xC1JElISUmR5YMVzHJSU1MRERHRJT9bL5rw8HAAQFpamjnN0/gwnjFBcNgZEwSHnTFBcNgZEwSHnTFBcNgZEwSHnTFBcNgZEwSHnTFBcNgZEwSHnTFBcNgZEwSHnTFBtOjLK+T6VkxmOfX7ODU1VeZKWHNyc3Ob/Dblx7XoI66Msc4lLCzM7I+4mv3Kzp9v7h748+ri4nN2xgTBYWdMEBx2xgTBYWdMEBx2xgTBYWdMEBx2xgTBYWdMEBx2xgTBYWdMEBx2xgTBYWdMEBx2xgTBYWdMEBx2xgTBYWdMEBx2xgTBYWdMEBx2xgTBYWdMEBx2xgTBYWdMEBx2xgTBYWdMEBx2xgTBYWdMEBx2xgTBYWdMEBx2xgTBYWdMEBx2xgTBYWdMEBx2xgRhLXcBrOMUFhbiP//zP02mnTt3DgCwfft2k+n29vaYOXOmxWpjlicREcldBOsYVVVVcHFxwYMHD6BQKAAARAQigpXV/x/UGQwGzJo1C5999plcpbKOl8aH8d2YSqVCeHg4rK2tYTAYYDAYUFNTg9raWuP/DQYDAPCrugA47N3czJkzUV1d3WQbR0dHPP/88xaqiMmFw97NjR8/Hi4uLo3OVyqViI6OhrU1X77p7jjs3ZyVlRVmzpwJGxubJ843GAyYMWOGhaticuCwC2DGjBmNHsr37t0bo0ePtnBFTA4cdgE8/fTT8Pb2bjBdqVRi9uzZkCRJhqqYpXHYBfHqq69CqVSaTONDeLFw2AURFRVlvM1Wz9/fH0OHDpWpImZpHHZBDBw4EAEBAcZDdqVSiTlz5shcFbMkDrtAZs2aZXwnncFgwPTp02WuiFkSh10gkZGRqK2tBQA89dRT8Pf3l7kiZkkcdoF4e3tjxIgRAB69yjOxNPggTGpqKiIiIuSqhzHWDp7w+ba0Rt8jmZKS0rHVMFmUlZUhOTkZ//Iv/9Ki5SIiIrB06VJ+A04nd+bMGWzZsuWJ8xoNO1+86b5+85vfoF+/fi1aJiIiAqNHj+bnRRfQWNj5nF1ALQ066x447IwJgsPOmCA47IwJgsPOmCA47IwJgsPOmCA47IwJgsPOmCA47IwJgsPOmCA47IwJgsPOmCA47IwJos1h37dvH3x9fSFJksnD2toaPXv2xIQJE7B///5m2//60bdv3ybbqtVq+Pj44LXXXsM//vEPAI++cqmpdf76ceTIkbZ2W1Zz586Fg4MDJEnCjz/+KHc5DTy+3zw9PbFz507j/JMnT8Ld3R2SJMHNza3Bz0fLVaebmxuio6NlqcUi6DEpKSn0hMnN8vPzI51OZ/z//fv36dixYzRw4EACQHv37m2yfU1NDen1erp79y4NGjSo0ba1tbV09+5d+vzzz0mj0ZCrqyvdu3ePIiIi6OjRo1RSUkIGg4Hu3LlDAGjKlClUXV1NDx48oIKCApo3bx4dPny4xf3rbPbs2UMA6MKFCxbZHgBKSUlp0TKP7+N6dXV1NHfuXHrjjTeorq6uvUpstcbq7IqayG9qhx3GOzk54fnnn8e///u/A3j0dVdNUSgUsLW1haurK/r3799oOysrK7i6uuLVV1/Fm2++iYKCAhw7dgySJOGZZ56BTqcz+ZFCSZKgVCqh0Wjg4uKCp556qn06yFqlrq4Or7/+OpRKJT7++GP+NRoL6vCf7qw/JC8pKTF7mQMHDpjVrv7bUfPz87Fnzx6zlpk/f77ZdXRmXTEkdXV1eO2112Bvb49t27bJXY5wOvwCXWZmJoBHX4XU3q5duwYA+Kd/+qd2XzcA1NbWYvXq1fDy8oKtrS2GDh1q/G6+5ORk2NnZQaPR4ODBg5g4cSK0Wi08PDwa/OHZvXs3hg8fDrVaDTs7O/Tt2xfvv/8+gEdfDLh582YMGjQIKpUKTk5OmDp1Kq5cuWJcnoiwceNGDBgwACqVCjqdDu+8847ZtW7YsAEajQYODg4oKCjA8uXL4e7ujuzs7A4Ztyepq6tDTEwMdDpdo0FvbR9Onz6NgIAA6HQ6qNVqBAYG4ptvvjGu9+TJkxg5ciQ0Gg20Wi0CAwNRVlbW4j40tZ25c+caz/39/Pxw4cIFAMCcOXOg0Wig0+lw6NAhefdTC475m/T4eY9er6evvvqKvL296cUXX6SKioom2xMRLVmyhC5evNjsuouLi+nPf/4zaTQamjx58hPrqT9n/+d//ucW96Xe22+/TSqVitLT06m4uJji4uLIysqKzp49S0REK1euJAB0/PhxKi0tpYKCAho3bhzZ2dlRdXU1ERElJiYSAFq/fj0VFRXR/fv36ZNPPqGoqCgiIlq9ejXZ2NjQ7t27qaSkhDIzMykoKIh69uxJ+fn5xu1IkkQffvghFRcXk16vp6SkJJNzdnNrXbJkCW3dupVCQkLo8uXLZo8F2nDOXlNTQ1FRUaRUKik7O7vN4/14H9LS0ig+Pp7u379PRUVFNGrUKOrRowcREVVUVJBWq6WEhASqrKyk/Px8CgkJocLCwgZ1Nqep7RARhYaGkkKhoNu3b5ssN3PmTDp06FCb+miups7Z2zXsABo8AgMD6bPPPqOqqiqz2jcW9sfbSZJEa9asMYbqcW0Ne2VlJWk0GoqMjDRO0+v1pFKpaNGiRUT0/zumsrLS2KY+hD///DNVV1eTo6MjjR8/3mTdNTU1tGXLFtLr9WRvb2+yDSKiH374gQDQe++9R3q9njQaDb3wwgsmbX59ga61tbZEa8Pu4OBAM2bMoKCgIAJAgwcPbvCHn6j14/0k69atIwBUUFBAP/30EwGgI0eONFlnay7Q/Xo7RETHjh0jALRmzRpjm9LSUurXrx/V1NRYZD9Z7AKdTqcDEYGIYDAYkJubi2XLlmHx4sUYOnQo7t2712h7IsKSJUvMWvc777wDIoJOp2vwy6TtJTs7G3q9HkOGDDFOs7W1hZubm8kh9uNsbGwAPPp5pczMTJSUlOCll14yaaNQKLBkyRJkZWWhoqICw4cPN5k/YsQI2NjYICMjAz///DP0ej2ef/75dq/VEvR6PX7zm9/g/PnzmDZtGrKysjB37twG7dqzD/XPidraWvj6+sLV1RXR0dGIj4/HzZs329SfxrYDAM899xz69++PTz/91Pi97Xv37kVkZCQUCoXs+6nDztmtra3h7u6OOXPmYNOmTcjOzsb69eubXGbLli0mA9GYf/3Xf4Wbmxvi4uLwyy+/tFfJJh48eAAAWLVqlck9+lu3bkGv15u1jvrzQkdHxyfOr79oaW9v32Ceo6MjysvLkZubCwBwcXHp0Fo7ir29vfGi6K5du+Dr64u9e/ciMTHRpF1b+vDll18iODgYLi4uUKlUWLFihXGera0tvv32W4wdOxZr166Fr68vIiMjUVlZ2eK+NLUd4NFF0wULFuDGjRs4fvw4AODzz/K2WzMAABdeSURBVD/H66+/3uY+tgeLvIMuMDAQAHDp0qV2WZ+DgwM++OADlJeXY9GiRe2yzsfVhysxMdHk6IOIcObMGbPW0adPHwBocERTr/6PQHl5eYN5JSUl8PDwgFqtBgBUVVV1aK2WoNPpkJaWZgzKqVOnjPNa24ecnBxMmzYNbm5uyMjIQGlpKRISEkzaDB48GIcPH0ZeXh5iY2ORkpKCTZs2mVXzqVOnkJiYaNZ2ACAmJgZqtRo7duxAdnY2tFotvL2929TH9mKRsJ8/fx4AMGDAALPa37lzp9mfE541axaefvppHDlypNl7+K3h6ekJtVrdpneo9e3bF87Ozjh69OgT5w8ZMgT29vY4d+6cyfSMjAxUV1fjqaeewpAhQ2BlZYWTJ092aK2WEhQUhMTERNTU1GD69OnIy8sD0Po+XLx4EQaDAYsWLYKvry/UarXJbcm8vDzji4yLiwvWr1+PoKAgs194zp8/Dzs7u2a3U8/JyQkRERE4cOAANm3ahHnz5hnnyb2f2j3slZWVqKurAxEhLy8Pu3btwqpVq9CzZ08sW7asyWWJCJWVldi3bx+0Wm2TbSVJwkcffQRJkrB48WIUFxe3ZzegVqsxZ84c7NmzB8nJySgrK0NtbS1yc3Nx584ds9ahUqkQFxeHU6dOYfHixbh9+zbq6upQXl6OS5cuQa1WY/ny5di/fz+++OILlJWV4eLFi1i4cCF69+6N+fPnw8XFBWFhYUhPT8fOnTtRVlaGzMxMk7eYtketlrRw4ULMmDEDd+/eRXh4OAwGQ6v74OXlBQA4duwYHj58iGvXriEjI8M4Py8vDwsWLMCVK1dQXV2NCxcu4NatWxg1alSTNRoMBty9excnTpyAnZ1ds9t5vH9VVVU4cuQIXnnlFeN02fdTC67mPdH+/fsbvbKuUqmoX79+tGjRIsrJyWm2/a8fq1atov/+7/+m/v37G6f16dOHFixYYLL9mJgYAkCOjo60fv16Kisro2effZacnZ0JAFlZWZG/vz+tXbvW7D7Vq6qqotjYWPLy8iJra2tycXGh0NBQysrKoqSkJNJoNASA+vXrR9evX6ft27eTVqslAOTt7U1Xr14lIqJt27ZRYGAgqdVqUqvVNGzYMEpKSiKiR28d3bhxI/Xr14+USiU5OTnRtGnTTG5RlZeX07x586hHjx5kb29PY8eOpdWrVxMA8vDwoP/5n/9pstaEhASytbUlAOTp6Um7d+9u8VigBVfjH9/HHh4eFBcXZ9KmvLycBgwYQADI1dWVdu7c2eo+xMbGkrOzMzk6OlJ4eDht27aNAJCfnx+dPn2axowZQ05OTqRQKKhPnz60cuVKqqmpMfu5uH///ma3U//8rjds2DB69913G4xNR+8ni9x6Y91bS8LOiCZNmkQ3btyw+HZleW88YyIxGAzGf2dmZho/mdmZCBf2K1eumPUx2MjISLlLZV1IbGwsrl27hqtXr2LOnDnGt0N3Jh3+QZjOZuDAgU/6oXrG2kSj0WDgwIFwd3dHUlISAgIC5C6pAeFe2RnrCGvWrEFtbS1ycnJMrsB3Jhx2xgTBYWdMEBx2xgTBYWdMEBx2xgTBYWdMEBx2xgTBYWdMEBx2xgTBYWdMEBx2xgTBYWdMEBx2xgTR6Edcu+JvibGOFRERgYiICLnLYK3UIOxjxowx/vYU637OnDmDLVu28D4WkET8TQ5CSU1NRUREBH+Bh3jS+JydMUFw2BkTBIedMUFw2BkTBIedMUFw2BkTBIedMUFw2BkTBIedMUFw2BkTBIedMUFw2BkTBIedMUFw2BkTBIedMUFw2BkTBIedMUFw2BkTBIedMUFw2BkTBIedMUFw2BkTBIedMUFw2BkTBIedMUFw2BkTBIedMUFw2BkTBIedMUFw2BkTBIedMUFw2BkThLXcBbCOYzAYUFFRYTLtwYMHAIDi4mKT6ZIkwdHR0WK1McvjsHdjRUVF8PDwQG1tbYN5zs7OJv8PDg7Gf/3Xf1mqNCYDPozvxtzc3PDss8/Cyqrp3SxJEmbMmGGhqphcOOzd3KuvvgpJkppsY2VlhdDQUAtVxOTCYe/mQkNDoVAoGp2vUCjw8ssvo0ePHhasismBw97NabVavPzyy7C2fvLlGSJCdHS0haticuCwCyA6OvqJF+kAwMbGBr/97W8tXBGTA4ddAK+88go0Gk2D6dbW1pg2bRrs7e1lqIpZGoddAGq1GiEhIVAqlSbTa2pqEBUVJVNVzNI47IKYOXMmDAaDyTStVosXXnhBpoqYpXHYBTFhwgSTN9IolUpERkbCxsZGxqqYJXHYBWFtbY3IyEjjobzBYMDMmTNlropZEoddIDNmzDAeyvfq1Qvjxo2TuSJmSRx2gTzzzDPo06cPgEfvrGvubbSse+lyH4QJDw+Xu4QuzcHBAQBw4cIFHss2GD16NH7/+9/LXUaLdLk/7enp6cjNzZW7jC4nNzcX6enp8PLygoODA5ycnOQuqcv6/vvvcebMGbnLaLEu98oOAMuWLcP06dPlLqNLSU1NRUREBI4ePYrU1FQevzboqkdEXe6VnbUdB11MHHbGBMFhZ0wQHHbGBMFhZ0wQHHbGBMFhZ0wQHHbGBMFhZ0wQHHbGBMFhZ0wQHHbGBMFhZ0wQHHbGBCFc2OfOnQsHBwdIkoQff/xR7nI6rX379sHX1xeSJJk8bGxs4OrqiuDgYGzcuLHBTz+zzku4sO/YsQN/+tOf5C6j0wsNDcWNGzfg5+cHnU4HIkJdXR0KCgqQmpoKHx8fxMbGYvDgwTh37pzc5TIzCBf2rqyyshJjxoyRbfuSJMHR0RHBwcHYtWsXUlNTcffuXUyePBmlpaWy1dUaco+lHIQMe3M/YdxZ7dy5EwUFBXKXYRQWFoaYmBgUFBTg448/lrucFulsY2kJ3T7sRISNGzdiwIABUKlU0Ol0eOedd4zzN2zYAI1GAwcHBxQUFGD58uVwd3dHdnY2iAibN2/GoEGDoFKp4OTkhKlTp+LKlSsAgI8++ghqtRqurq5YsGABevfuDbVajTFjxiAjI8OkhqbWs3jxYtjY2MDNzc24zO9+9zvY2dlBkiTcu3cPS5cuxfLly3H9+nVIkgR/f38LjWDTYmJiAABfffUVj2VnR10MAEpJSTG7/cqVK0mSJPrwww+puLiY9Ho9JSUlEQC6cOGCsQ0AWrJkCW3dupVCQkLo8uXLtHr1arKxsaHdu3dTSUkJZWZmUlBQEPXs2ZPy8/OJiGj+/PlkZ2dHly5doocPH1JWVhaNGDGCHBwcKCcnh4jIrPVERUVRr169TGrfuHEjAaDCwkIiIgoNDSU/P79WjVtKSgq1Znf7+fmRTqdrdH5ZWRkBIE9PTyISYyzDwsIoLCysVcvKKLVbh12v15NGo6EXXnjBZPqePXueGPbKykqTZe3t7SkyMtJk2R9++IEA0HvvvUdEj56gj4fh7NmzBID+8Ic/mL2erhp2IiJJksjR0ZGIxBjLrhr2bn0Y//PPP0Ov1+P5559v8bJZWVmoqKjA8OHDTaaPGDECNjY2JoeWjxs+fDg0Gg2uXLnSpvV0BQ8ePAARQavVNtqGx7Jz6NZhr/9+eRcXlxYvW1JSAgBP/O1yR0dHlJeXN7m8SqVCYWFhm9fT2V29ehUAMHDgwEbb8Fh2Dt067Gq1GgBQVVXV4mUdHR0B4IlPoJKSEnh4eDS6rMFgMLZpy3q6gq+//hoAMHHixEbb8Fh2Dt067EOGDIGVlRVOnjzZqmXt7e0bvGEkIyMD1dXVeOqppxpd9sSJEyAijBo1yuz1WFtbN/j99M4uPz8fiYmJ8PDwwGuvvdZoOx7LzqFbh93FxQVhYWFIT0/Hzp07UVZWhszMTGzfvr3ZZdVqNZYvX479+/fjiy++QFlZGS5evIiFCxeid+/emD9/vrFtXV0diouLUVNTg8zMTCxduhReXl6IiYkxez3+/v64f/8+Dhw4AIPBgMLCQty6dcukJmdnZ+Tl5eHmzZsoLy+32BOaiFBRUYG6ujoQEQoLC5GSkoJnnnkGCoUCBw4caPKcnceyk5D1+mAroIW33srLy2nevHnUo0cPsre3p7Fjx9Lq1asJAHl4eFBUVBTZ2toabx/t3r3buGxdXR1t3LiR+vXrR0qlkpycnGjatGmUnZ1tbDN//nxSKpXk7u5O1tbWpNVqaerUqXT9+vUWraeoqIjGjx9ParWafHx86K233qJ33nmHAJC/vz/l5OTQ3//+d/L29iZbW1saO3as8VaTOVp6Nf7QoUM0dOhQ0mg0ZGNjQ1ZWVgTAeOV95MiR9N5771FRUZFxmYSEBCHGsqteje/2Ye9o8+fPJ2dnZ7nLaFZrb71ZUlcZy64a9m59GG8ptbW1cpfQbfBYdhwOO2OC4LC3QVxcHHbt2oXS0lL4+PggPT1d7pK6LB7Ljtclf5+9s1i3bh3WrVsndxndAo9lx+NXdsYEwWFnTBAcdsYEwWFnTBAcdsYEwWFnTBAcdsYEwWFnTBAcdsYEwWFnTBAcdsYEwWFnTBAcdsYE0SU/9ZaYmIi0tDS5y+hS6r9WOzw8XOZKur7vv/8eo0aNkruMFpOIiOQuoiX4ydo2hYWFuHz5Mp599lm5S+nSRo8ejd///vdyl9ESaV0u7KxtUlNTERERAd7twknjc3bGBMFhZ0wQHHbGBMFhZ0wQHHbGBMFhZ0wQHHbGBMFhZ0wQHHbGBMFhZ0wQHHbGBMFhZ0wQHHbGBMFhZ0wQHHbGBMFhZ0wQHHbGBMFhZ0wQHHbGBMFhZ0wQHHbGBMFhZ0wQHHbGBMFhZ0wQHHbGBMFhZ0wQHHbGBMFhZ0wQHHbGBMFhZ0wQHHbGBMFhZ0wQHHbGBGEtdwGs4+Tm5mL27Nmora01Trt37x6sra0RHBxs0nbAgAH45JNPLFwhsyQOezfm4eGBmzdv4saNGw3mnTx50uT/48aNs1RZTCZ8GN/NzZo1C0qlstl2kZGRFqiGyYnD3s1FRUXBYDA02SYgIACDBw+2UEVMLhz2bs7f3x9Dhw6FJElPnK9UKjF79mwLV8XkwGEXwKxZs6BQKJ44r6amBtOnT7dwRUwOHHYBzJgxA3V1dQ2mS5KEp59+Gn379rV8UcziOOwC6NOnD8aMGQMrK9PdrVAoMGvWLJmqYpbGYRfEq6++2mAaESE0NFSGapgcOOyCCA8PN3llVygUmDBhAlxdXWWsilkSh10QTk5OePHFF40X6ogI0dHRMlfFLInDLpDo6GjjhTpra2tMmTJF5oqYJXHYBTJlyhSoVCrjv7VarcwVMUvq8u+NT01NlbuELiUoKAjfffcdfHx8eOxawNPTE6NHj5a7jDaRiIjkLqItGntnGGPtKSwsDGlpaXKX0RZp3eIwPiUlBUTEDzMe1dXVWLFiRbPtUlJSAED2ejvDIywsTOZnePvoFmFn5lMqlYiPj5e7DCYDDruAbG1t5S6ByYDDzpggOOyMCYLDzpggOOyMCYLDzpggOOyMCYLDzpggOOyMCYLDzpggOOyMCYLDzpggOOyMCYLD/iubNm2Cq6srJEnCxx9/3OHb+8tf/gKdTofDhw+bTK+qqsKSJUvg5uYGjUaDr7/+utG2ncW+ffvg6+sLSZJMHjY2NnB1dUVwcDA2btyI4uJiuUsVFof9V95++2189913Ftse0ZO/N+TDDz/E119/jStXrmDLli2oqKhotG1nERoaihs3bsDPzw86nQ5EhLq6OhQUFCA1NRU+Pj6IjY3F4MGDce7cObnLFVKX/1qqrmzy5MkoLS1tMP3AgQMYPnw4HB0d8cYbbxinP6ltZyZJEhwdHREcHIzg4GBMnjwZERERmDx5Mq5evQqdTid3iULhV/ZOKDc316yfWe5qwsLCEBMTg4KCAoucJjFTQoZ99+7dGD58ONRqNezs7NC3b1+8//77jbY/ffo0AgICoNPpoFarERgYiG+++cY4/+TJkxg5ciQ0Gg20Wi0CAwNRVlbW5Ly//e1v8PLygiRJ2LZtGwDgr3/9K/z9/XHnzh189tlnkCQJ9vb2T2wLALW1tVi9ejW8vLxga2uLoUOHGr9OasOGDdBoNHBwcEBBQQGWL18Od3d3ZGdnd8SQmi0mJgYA8NVXXwFoug/Jycmws7ODRqPBwYMHMXHiRGi1Wnh4eGDPnj3GdTY1/k2tXzjUxQGglJQUs9snJiYSAFq/fj0VFRXR/fv36ZNPPqGoqCgiIrp27RoBoD/+8Y/GZdLS0ig+Pp7u379PRUVFNGrUKOrRowcREVVUVJBWq6WEhASqrKyk/Px8CgkJocLCwibnERH98ssvBIC2bt1qUmOvXr1o9uzZJtOe1Pbtt98mlUpF6enpVFxcTHFxcWRlZUVnz54lIqKVK1cSAFqyZAlt3bqVQkJC6PLly2aNU0pKCrXm6eHn50c6na7R+WVlZQSAPD09W9SH48ePU2lpKRUUFNC4cePIzs6Oqqurmx3j5tZvjrCwMAoLC2vxWHQyqUKFvbq6mhwdHWn8+PEm02tqamjLli1E9OSwP27dunUEgAoKCuinn34iAHTkyJEG7ZqaR9S2sFdWVpJGo6HIyEhjG71eTyqVihYtWkRE/x+UysrKRvvSmI4KOxGRJEnk6OjY6j4kJSURAPr555+bHGNz1m+O7hJ2oQ7jMzMzUVJSgpdeeslkukKhwJIlS8xeT/35dG1tLXx9feHq6oro6GjEx8fj5s2bxnZNzWur7Oxs6PV6DBkyxDjN1tYWbm5uuHLlSrttp709ePAARAStVtvqPtjY2AAADAZDk2PcVceoowgV9vrzOEdHxxYt9+WXXyI4OBguLi5QqVRYsWKFcZ6trS2+/fZbjB07FmvXroWvry8iIyNRWVnZ5Ly2evDgAQBg1apVJve1b926Bb1e3+b1d5SrV68CAAYOHNgufWhqjLvqGHUUocLep08fAMC9e/fMXiYnJwfTpk2Dm5sbMjIyUFpaioSEBJM2gwcPxuHDh5GXl4fY2FikpKRg06ZNzc5rCxcXFwBAYmJig+85P3PmTJvX31G+/vprAMDEiRPbrQ+NjXFXHaOOIlTY+/btC2dnZxw9etTsZS5evAiDwYBFixbB19cXarXa5Fdo8vLycOnSJQCPArh+/XoEBQXh0qVLTc5rK09PT6jVavz4449tXpel5OfnIzExER4eHnjttdfapQ9NjXFXHKOOJFTYVSoV4uLicOrUKSxevBi3b99GXV0dysvLGw2gl5cXAODYsWN4+PAhrl27hoyMDOP8vLw8LFiwAFeuXEF1dTUuXLiAW7duYdSoUU3Oayu1Wo05c+Zgz549SE5ORllZGWpra5Gbm4s7d+60ef1tQUSoqKhAXV0diAiFhYVISUnBM888A4VCgQMHDkCr1bZLH5oa4848RrKQ4apgu0ILb70REW3bto0CAwNJrVaTWq2mYcOGUVJSEn344YfUq1cvAkB2dnYUEhJCRESxsbHk7OxMjo6OFB4eTtu2bSMA5OfnR6dPn6YxY8aQk5MTKRQK6tOnD61cuZJqamro5s2bjc7bunUrubm5EQDSaDQ0ZcoUunnzJg0bNowAkLW1NQUFBVF6evoT2xIRVVVVUWxsLHl5eZG1tTW5uLhQaGgoZWVlUUJCAtna2hpvc+3evbtFY9TSq/GHDh2ioUOHkkajIRsbG7KysiIAxivvI0eOpPfee4+KiopMlmuqD0lJSaTRaAgA9evXj65fv07bt28nrVZLAMjb25v++te/NjrGza3fXN3lany3+GHHlJQUTJ8+Xe5SupXU1FRERER0+vfkW0J4eDgA8A87Msa6Bg47Y4LgsDMmCA47Y4LgsDMmCA47Y4LgsDMmCA47Y4LgsDMmCA47Y4LgsDMmCA47Y4LgsDMmCA47Y4LgsDMmCA47Y4LgsDMmiG7xw44iflNoR6sf09TUVJkrkV9ubi48PDzkLqPNusXXUjHW0cLCwrr811J1+Vf2Lv63ijGL4XN2xgTBYWdMEBx2xgTBYWdMEP8LboiQqT6OyD4AAAAASUVORK5CYII=\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "loss = tf.keras.losses.BinaryCrossentropy()\n",
        "metrics = tf.metrics.BinaryAccuracy()"
      ],
      "metadata": {
        "id": "GAOuglZUG4AY"
      },
      "execution_count": 16,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "epochs = 5\n",
        "steps_per_epoch = tf.data.experimental.cardinality(train_ds).numpy()\n",
        "num_train_steps = steps_per_epoch * epochs\n",
        "num_warmup_steps = int(0.1 * num_train_steps)\n",
        "\n",
        "init_lr = 3e-5\n",
        "optimizer = optimization.create_optimizer(\n",
        "    init_lr=init_lr,\n",
        "    num_train_steps=num_train_steps,\n",
        "    num_warmup_steps=num_warmup_steps,\n",
        "    optimizer_type=\"adamw\",\n",
        ")"
      ],
      "metadata": {
        "id": "wINmGpsUG8mZ"
      },
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "classifier_model.compile(optimizer=optimizer, loss=loss, metrics=metrics)"
      ],
      "metadata": {
        "id": "KDn5bwmHHAEz"
      },
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(f\"Training model with {tfhub_handle_encoder}\")\n",
        "history = classifier_model.fit(\n",
        "    x=train_ds, validation_data=val_ds, epochs=epochs\n",
        ")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "c7a5E93yHDZS",
        "outputId": "2f055add-2d19-494f-d1a0-603ef4525e71"
      },
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Training model with https://tfhub.dev/tensorflow/small_bert/bert_en_uncased_L-4_H-512_A-8/1\n",
            "Epoch 1/5\n",
            "625/625 [==============================] - 179s 273ms/step - loss: 0.4921 - binary_accuracy: 0.7469 - val_loss: 0.3730 - val_binary_accuracy: 0.8338\n",
            "Epoch 2/5\n",
            "625/625 [==============================] - 169s 270ms/step - loss: 0.3355 - binary_accuracy: 0.8532 - val_loss: 0.3743 - val_binary_accuracy: 0.8422\n",
            "Epoch 3/5\n",
            "625/625 [==============================] - 171s 274ms/step - loss: 0.2613 - binary_accuracy: 0.8917 - val_loss: 0.3956 - val_binary_accuracy: 0.8494\n",
            "Epoch 4/5\n",
            "625/625 [==============================] - 170s 271ms/step - loss: 0.2011 - binary_accuracy: 0.9225 - val_loss: 0.4372 - val_binary_accuracy: 0.8476\n",
            "Epoch 5/5\n",
            "625/625 [==============================] - 168s 268ms/step - loss: 0.1605 - binary_accuracy: 0.9406 - val_loss: 0.4665 - val_binary_accuracy: 0.8522\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "loss, accuracy = classifier_model.evaluate(test_ds)\n",
        "\n",
        "print(f\"Loss: {loss}\")\n",
        "print(f\"Accuracy: {accuracy}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7My4V3GkK0u9",
        "outputId": "c8d1b4df-7d5e-4ea9-bce9-b63327637836"
      },
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "782/782 [==============================] - 99s 127ms/step - loss: 0.4452 - binary_accuracy: 0.8546\n",
            "Loss: 0.44516298174858093\n",
            "Accuracy: 0.8546000123023987\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "history_dict = history.history\n",
        "print(history_dict.keys())\n",
        "\n",
        "acc = history_dict[\"binary_accuracy\"]\n",
        "val_acc = history_dict[\"val_binary_accuracy\"]\n",
        "loss = history_dict[\"loss\"]\n",
        "val_loss = history_dict[\"val_loss\"]\n",
        "\n",
        "epochs = range(1, len(acc) + 1)\n",
        "fig = plt.figure(figsize=(10, 6))\n",
        "fig.tight_layout()\n",
        "\n",
        "plt.subplot(2, 1, 1)\n",
        "# \"bo\" is for \"blue dot\"\n",
        "plt.plot(epochs, loss, \"r\", label=\"Training loss\")\n",
        "# b is for \"solid blue line\"\n",
        "plt.plot(epochs, val_loss, \"b\", label=\"Validation loss\")\n",
        "plt.title(\"Training and validation loss\")\n",
        "# plt.xlabel('Epochs')\n",
        "plt.ylabel(\"Loss\")\n",
        "plt.legend()\n",
        "\n",
        "plt.subplot(2, 1, 2)\n",
        "plt.plot(epochs, acc, \"r\", label=\"Training acc\")\n",
        "plt.plot(epochs, val_acc, \"b\", label=\"Validation acc\")\n",
        "plt.title(\"Training and validation accuracy\")\n",
        "plt.xlabel(\"Epochs\")\n",
        "plt.ylabel(\"Accuracy\")\n",
        "plt.legend(loc=\"lower right\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 600
        },
        "id": "bgDktttWMMDi",
        "outputId": "91f105b9-03b8-46fb-b608-956249d271c3"
      },
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "dict_keys(['loss', 'binary_accuracy', 'val_loss', 'val_binary_accuracy'])\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7c8e4b5c5720>"
            ]
          },
          "metadata": {},
          "execution_count": 21
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x600 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "dataset_name = \"imdb\"\n",
        "saved_model_path = \"./{}_bert\".format(dataset_name.replace(\"/\", \"_\"))\n",
        "TIMESTAMP = datetime.datetime.now().strftime(\"%Y%m%d%H%M%S\")\n",
        "\n",
        "EXPORT_PATH = os.path.join(saved_model_path, TIMESTAMP)\n",
        "\n",
        "classifier_model.save(EXPORT_PATH, include_optimizer=False)"
      ],
      "metadata": {
        "id": "9VSVDbUsMSyR"
      },
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "reloaded_model = tf.saved_model.load(EXPORT_PATH)"
      ],
      "metadata": {
        "id": "A7_aDX-kNBD5"
      },
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def print_my_examples(inputs, results):\n",
        "    result_for_printing = [\n",
        "        f\"input: {inputs[i]:<30} : score: {results[i][0]:.6f}\"\n",
        "        for i in range(len(inputs))\n",
        "    ]\n",
        "    print(*result_for_printing, sep=\"\\n\")\n",
        "    print()\n",
        "\n",
        "\n",
        "examples = [\n",
        "    \"this is such an amazing movie!\",  # this is the same sentence tried earlier\n",
        "    \"The movie was great!\",\n",
        "    \"The movie was meh.\",\n",
        "    \"The movie was okish.\",\n",
        "    \"The movie was terrible...\",\n",
        "]\n",
        "\n",
        "reloaded_results = reloaded_model(tf.constant(examples))\n",
        "original_results = classifier_model(tf.constant(examples))\n",
        "\n",
        "print(\"Results from the saved model:\")\n",
        "print_my_examples(examples, reloaded_results)\n",
        "print(\"Results from the model in memory:\")\n",
        "print_my_examples(examples, original_results)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JKlhjMlJNa8t",
        "outputId": "a49c5111-0db0-409f-90aa-2934d6cb5e30"
      },
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Results from the saved model:\n",
            "input: this is such an amazing movie! : score: 0.999349\n",
            "input: The movie was great!           : score: 0.995228\n",
            "input: The movie was meh.             : score: 0.982102\n",
            "input: The movie was okish.           : score: 0.036127\n",
            "input: The movie was terrible...      : score: 0.000800\n",
            "\n",
            "Results from the model in memory:\n",
            "input: this is such an amazing movie! : score: 0.999349\n",
            "input: The movie was great!           : score: 0.995228\n",
            "input: The movie was meh.             : score: 0.982102\n",
            "input: The movie was okish.           : score: 0.036127\n",
            "input: The movie was terrible...      : score: 0.000800\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [],
      "metadata": {
        "id": "K44yaTJnM_7g"
      }
    }
  ]
}