Omega_Transformer_0_2.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/URLbug/Omega_Transformers/blob/main/Omega_Transformer_0_2.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "g7i1B9Eww8Mp",
        "outputId": "6859b384-185a-4aba-edec-8a434dae9e8d"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Reading package lists... Done\n",
            "Building dependency tree... Done\n",
            "Reading state information... Done\n",
            "The following packages will be REMOVED:\n",
            "  libcudnn8-dev\n",
            "The following held packages will be changed:\n",
            "  libcudnn8\n",
            "The following packages will be DOWNGRADED:\n",
            "  libcudnn8\n",
            "0 upgraded, 0 newly installed, 1 downgraded, 1 to remove and 7 not upgraded.\n",
            "Need to get 463 MB of archives.\n",
            "After this operation, 1,290 MB disk space will be freed.\n",
            "Get:1 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  libcudnn8 8.7.0.84-1+cuda11.8 [463 MB]\n",
            "Fetched 463 MB in 9s (49.1 MB/s)\n",
            "(Reading database ... 120493 files and directories currently installed.)\n",
            "Removing libcudnn8-dev (8.9.0.131-1+cuda11.8) ...\n",
            "update-alternatives: removing manually selected alternative - switching libcudnn to auto mode\n",
            "\u001b[1mdpkg:\u001b[0m \u001b[1;33mwarning:\u001b[0m downgrading libcudnn8 from 8.9.0.131-1+cuda11.8 to 8.7.0.84-1+cuda11.8\n",
            "(Reading database ... 120460 files and directories currently installed.)\n",
            "Preparing to unpack .../libcudnn8_8.7.0.84-1+cuda11.8_amd64.deb ...\n",
            "Unpacking libcudnn8 (8.7.0.84-1+cuda11.8) over (8.9.0.131-1+cuda11.8) ...\n",
            "Setting up libcudnn8 (8.7.0.84-1+cuda11.8) ...\n",
            "\u001b[33mWARNING: Skipping tensorflow-text as it is not installed.\u001b[0m\u001b[33m\n",
            "\u001b[0mCollecting tensorflow-text\n",
            "  Downloading tensorflow_text-2.13.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (6.5 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m6.5/6.5 MB\u001b[0m \u001b[31m19.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: tensorflow-hub>=0.8.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow-text) (0.14.0)\n",
            "Collecting tensorflow<2.14,>=2.13.0 (from tensorflow-text)\n",
            "  Downloading tensorflow-2.13.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (524.1 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m524.1/524.1 MB\u001b[0m \u001b[31m3.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: absl-py>=1.0.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.14,>=2.13.0->tensorflow-text) (1.4.0)\n",
            "Requirement already satisfied: astunparse>=1.6.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.14,>=2.13.0->tensorflow-text) (1.6.3)\n",
            "Requirement already satisfied: flatbuffers>=23.1.21 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.14,>=2.13.0->tensorflow-text) (23.5.26)\n",
            "Requirement already satisfied: gast<=0.4.0,>=0.2.1 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.14,>=2.13.0->tensorflow-text) (0.4.0)\n",
            "Requirement already satisfied: google-pasta>=0.1.1 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.14,>=2.13.0->tensorflow-text) (0.2.0)\n",
            "Requirement already satisfied: grpcio<2.0,>=1.24.3 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.14,>=2.13.0->tensorflow-text) (1.56.2)\n",
            "Requirement already satisfied: h5py>=2.9.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.14,>=2.13.0->tensorflow-text) (3.8.0)\n",
            "Collecting keras<2.14,>=2.13.1 (from tensorflow<2.14,>=2.13.0->tensorflow-text)\n",
            "  Downloading keras-2.13.1-py3-none-any.whl (1.7 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.7/1.7 MB\u001b[0m \u001b[31m74.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: libclang>=13.0.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.14,>=2.13.0->tensorflow-text) (16.0.6)\n",
            "Requirement already satisfied: numpy<=1.24.3,>=1.22 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.14,>=2.13.0->tensorflow-text) (1.22.4)\n",
            "Requirement already satisfied: opt-einsum>=2.3.2 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.14,>=2.13.0->tensorflow-text) (3.3.0)\n",
            "Requirement already satisfied: packaging in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.14,>=2.13.0->tensorflow-text) (23.1)\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<2.14,>=2.13.0->tensorflow-text) (3.20.3)\n",
            "Requirement already satisfied: setuptools in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.14,>=2.13.0->tensorflow-text) (67.7.2)\n",
            "Requirement already satisfied: six>=1.12.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.14,>=2.13.0->tensorflow-text) (1.16.0)\n",
            "Collecting tensorboard<2.14,>=2.13 (from tensorflow<2.14,>=2.13.0->tensorflow-text)\n",
            "  Downloading tensorboard-2.13.0-py3-none-any.whl (5.6 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m5.6/5.6 MB\u001b[0m \u001b[31m113.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hCollecting tensorflow-estimator<2.14,>=2.13.0 (from tensorflow<2.14,>=2.13.0->tensorflow-text)\n",
            "  Downloading tensorflow_estimator-2.13.0-py2.py3-none-any.whl (440 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m440.8/440.8 kB\u001b[0m \u001b[31m39.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: termcolor>=1.1.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.14,>=2.13.0->tensorflow-text) (2.3.0)\n",
            "Collecting typing-extensions<4.6.0,>=3.6.6 (from tensorflow<2.14,>=2.13.0->tensorflow-text)\n",
            "  Downloading typing_extensions-4.5.0-py3-none-any.whl (27 kB)\n",
            "Requirement already satisfied: wrapt>=1.11.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.14,>=2.13.0->tensorflow-text) (1.14.1)\n",
            "Requirement already satisfied: tensorflow-io-gcs-filesystem>=0.23.1 in /usr/local/lib/python3.10/dist-packages (from tensorflow<2.14,>=2.13.0->tensorflow-text) (0.32.0)\n",
            "Requirement already satisfied: wheel<1.0,>=0.23.0 in /usr/local/lib/python3.10/dist-packages (from astunparse>=1.6.0->tensorflow<2.14,>=2.13.0->tensorflow-text) (0.41.0)\n",
            "Requirement already satisfied: google-auth<3,>=1.6.3 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.14,>=2.13->tensorflow<2.14,>=2.13.0->tensorflow-text) (2.17.3)\n",
            "Requirement already satisfied: google-auth-oauthlib<1.1,>=0.5 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.14,>=2.13->tensorflow<2.14,>=2.13.0->tensorflow-text) (1.0.0)\n",
            "Requirement already satisfied: markdown>=2.6.8 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.14,>=2.13->tensorflow<2.14,>=2.13.0->tensorflow-text) (3.4.4)\n",
            "Requirement already satisfied: requests<3,>=2.21.0 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.14,>=2.13->tensorflow<2.14,>=2.13.0->tensorflow-text) (2.27.1)\n",
            "Requirement already satisfied: tensorboard-data-server<0.8.0,>=0.7.0 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.14,>=2.13->tensorflow<2.14,>=2.13.0->tensorflow-text) (0.7.1)\n",
            "Requirement already satisfied: werkzeug>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.14,>=2.13->tensorflow<2.14,>=2.13.0->tensorflow-text) (2.3.6)\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<2.14,>=2.13->tensorflow<2.14,>=2.13.0->tensorflow-text) (5.3.1)\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<2.14,>=2.13->tensorflow<2.14,>=2.13.0->tensorflow-text) (0.3.0)\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<2.14,>=2.13->tensorflow<2.14,>=2.13.0->tensorflow-text) (4.9)\n",
            "Requirement already satisfied: requests-oauthlib>=0.7.0 in /usr/local/lib/python3.10/dist-packages (from google-auth-oauthlib<1.1,>=0.5->tensorboard<2.14,>=2.13->tensorflow<2.14,>=2.13.0->tensorflow-text) (1.3.1)\n",
            "Requirement already satisfied: urllib3<1.27,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests<3,>=2.21.0->tensorboard<2.14,>=2.13->tensorflow<2.14,>=2.13.0->tensorflow-text) (1.26.16)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests<3,>=2.21.0->tensorboard<2.14,>=2.13->tensorflow<2.14,>=2.13.0->tensorflow-text) (2023.7.22)\n",
            "Requirement already satisfied: charset-normalizer~=2.0.0 in /usr/local/lib/python3.10/dist-packages (from requests<3,>=2.21.0->tensorboard<2.14,>=2.13->tensorflow<2.14,>=2.13.0->tensorflow-text) (2.0.12)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests<3,>=2.21.0->tensorboard<2.14,>=2.13->tensorflow<2.14,>=2.13.0->tensorflow-text) (3.4)\n",
            "Requirement already satisfied: MarkupSafe>=2.1.1 in /usr/local/lib/python3.10/dist-packages (from werkzeug>=1.0.1->tensorboard<2.14,>=2.13->tensorflow<2.14,>=2.13.0->tensorflow-text) (2.1.3)\n",
            "Requirement already satisfied: pyasn1<0.6.0,>=0.4.6 in /usr/local/lib/python3.10/dist-packages (from pyasn1-modules>=0.2.1->google-auth<3,>=1.6.3->tensorboard<2.14,>=2.13->tensorflow<2.14,>=2.13.0->tensorflow-text) (0.5.0)\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<1.1,>=0.5->tensorboard<2.14,>=2.13->tensorflow<2.14,>=2.13.0->tensorflow-text) (3.2.2)\n",
            "Installing collected packages: typing-extensions, tensorflow-estimator, keras, tensorboard, tensorflow, tensorflow-text\n",
            "  Attempting uninstall: typing-extensions\n",
            "    Found existing installation: typing_extensions 4.7.1\n",
            "    Uninstalling typing_extensions-4.7.1:\n",
            "      Successfully uninstalled typing_extensions-4.7.1\n",
            "  Attempting uninstall: keras\n",
            "    Found existing installation: keras 2.12.0\n",
            "    Uninstalling keras-2.12.0:\n",
            "      Successfully uninstalled keras-2.12.0\n",
            "  Attempting uninstall: tensorboard\n",
            "    Found existing installation: tensorboard 2.12.3\n",
            "    Uninstalling tensorboard-2.12.3:\n",
            "      Successfully uninstalled tensorboard-2.12.3\n",
            "  Attempting uninstall: tensorflow\n",
            "    Found existing installation: tensorflow 2.12.0\n",
            "    Uninstalling tensorflow-2.12.0:\n",
            "      Successfully uninstalled tensorflow-2.12.0\n",
            "Successfully installed keras-2.13.1 tensorboard-2.13.0 tensorflow-2.13.0 tensorflow-estimator-2.13.0 tensorflow-text-2.13.0 typing-extensions-4.5.0\n",
            "Collecting datasets\n",
            "  Downloading datasets-2.14.2-py3-none-any.whl (518 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m518.9/518.9 kB\u001b[0m \u001b[31m8.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: numpy>=1.17 in /usr/local/lib/python3.10/dist-packages (from datasets) (1.22.4)\n",
            "Requirement already satisfied: pyarrow>=8.0.0 in /usr/local/lib/python3.10/dist-packages (from datasets) (9.0.0)\n",
            "Collecting dill<0.3.8,>=0.3.0 (from datasets)\n",
            "  Downloading dill-0.3.7-py3-none-any.whl (115 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m115.3/115.3 kB\u001b[0m \u001b[31m13.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: pandas in /usr/local/lib/python3.10/dist-packages (from datasets) (1.5.3)\n",
            "Requirement already satisfied: requests>=2.19.0 in /usr/local/lib/python3.10/dist-packages (from datasets) (2.27.1)\n",
            "Requirement already satisfied: tqdm>=4.62.1 in /usr/local/lib/python3.10/dist-packages (from datasets) (4.65.0)\n",
            "Collecting xxhash (from datasets)\n",
            "  Downloading xxhash-3.3.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (194 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m194.1/194.1 kB\u001b[0m \u001b[31m13.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hCollecting multiprocess (from datasets)\n",
            "  Downloading multiprocess-0.70.15-py310-none-any.whl (134 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m134.8/134.8 kB\u001b[0m \u001b[31m15.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: fsspec[http]>=2021.11.1 in /usr/local/lib/python3.10/dist-packages (from datasets) (2023.6.0)\n",
            "Requirement already satisfied: aiohttp in /usr/local/lib/python3.10/dist-packages (from datasets) (3.8.5)\n",
            "Collecting huggingface-hub<1.0.0,>=0.14.0 (from datasets)\n",
            "  Downloading huggingface_hub-0.16.4-py3-none-any.whl (268 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m268.8/268.8 kB\u001b[0m \u001b[31m13.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: packaging in /usr/local/lib/python3.10/dist-packages (from datasets) (23.1)\n",
            "Requirement already satisfied: pyyaml>=5.1 in /usr/local/lib/python3.10/dist-packages (from datasets) (6.0.1)\n",
            "Requirement already satisfied: attrs>=17.3.0 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets) (23.1.0)\n",
            "Requirement already satisfied: charset-normalizer<4.0,>=2.0 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets) (2.0.12)\n",
            "Requirement already satisfied: multidict<7.0,>=4.5 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets) (6.0.4)\n",
            "Requirement already satisfied: async-timeout<5.0,>=4.0.0a3 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets) (4.0.2)\n",
            "Requirement already satisfied: yarl<2.0,>=1.0 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets) (1.9.2)\n",
            "Requirement already satisfied: frozenlist>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets) (1.4.0)\n",
            "Requirement already satisfied: aiosignal>=1.1.2 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets) (1.3.1)\n",
            "Requirement already satisfied: filelock in /usr/local/lib/python3.10/dist-packages (from huggingface-hub<1.0.0,>=0.14.0->datasets) (3.12.2)\n",
            "Requirement already satisfied: typing-extensions>=3.7.4.3 in /usr/local/lib/python3.10/dist-packages (from huggingface-hub<1.0.0,>=0.14.0->datasets) (4.5.0)\n",
            "Requirement already satisfied: urllib3<1.27,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests>=2.19.0->datasets) (1.26.16)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests>=2.19.0->datasets) (2023.7.22)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests>=2.19.0->datasets) (3.4)\n",
            "Requirement already satisfied: python-dateutil>=2.8.1 in /usr/local/lib/python3.10/dist-packages (from pandas->datasets) (2.8.2)\n",
            "Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-packages (from pandas->datasets) (2022.7.1)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.10/dist-packages (from python-dateutil>=2.8.1->pandas->datasets) (1.16.0)\n",
            "Installing collected packages: xxhash, dill, multiprocess, huggingface-hub, datasets\n",
            "Successfully installed datasets-2.14.2 dill-0.3.7 huggingface-hub-0.16.4 multiprocess-0.70.15 xxhash-3.3.0\n"
          ]
        }
      ],
      "source": [
        "!apt install --allow-change-held-packages libcudnn8=8.7.0.84-1+cuda11.8\n",
        "!pip uninstall -y -q tensorflow-estimator tensorflow-text\n",
        "!pip install -U tensorflow-text\n",
        "!pip install datasets"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 30,
      "metadata": {
        "id": "ibZi5O5MAScm"
      },
      "outputs": [],
      "source": [
        "import keras\n",
        "\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import tensorflow as tf\n",
        "import tensorflow_text as text\n",
        "import tensorflow_datasets as tfds\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "from keras.models import Sequential, Model\n",
        "from keras.layers import Dense, Dropout, Embedding, Input, Flatten\n",
        "from keras.optimizers import Adam\n",
        "\n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "from datasets import load_dataset\n",
        "\n",
        "from datetime import datetime"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 241,
          "referenced_widgets": [
            "1c66e15c3fed416e9ffcced39429c276",
            "4053125e3122462f91ea021c341f972c",
            "4186f779cb134e8eb16bbb973a993148",
            "97ed262e06f94929a1dc231911455520",
            "fe647113430c4f198f057a96fe423b47",
            "2afeb66749b843c59a0e1c1d4e0dce68",
            "7e08f1f170064973aeccf720ac731cf5",
            "73baa2ba932d4e709a07fdb53d443771",
            "77e7acbcca6a4aa087aace7e00fc7528",
            "3c04b77febf94713bf6b832b2a090be5",
            "64238f56a0fe4697b350b18458b69f4a",
            "e30bbdcbf73745aabcba4d2016554a41",
            "2c6ca44cd391410a8c7a635d6828d2b0",
            "6f918e5e466b4ae68b313147a5959dc6",
            "24b31e5f85b14df89200707ff906c845",
            "d60ad06322a241b09ddb281d3736279c",
            "9d13d00e10f34df0b5ded3df36cdbd54",
            "166309fbc8ec4c23b9e001400aa1bbe8",
            "fdf6ee5be70e4cf09a9d81ad17170e62",
            "06b8d904df6d4cceaea939060c0fbae8",
            "ba236a64db5a4a76956a891f7ab5150a",
            "cef1d158396a42fd913a65858019e127",
            "c70b5856b43843ebb1e24f3cc6102cd2",
            "f3ab84e712044d1cbe81f3419eea88cc",
            "940b488d575144a3bc5c41b9f9e3e987",
            "e8e7911774424baba4779e83ab71cd1e",
            "a0c97271414140df800aa7330f9b84ac",
            "96556008bf65479b8220d457c943ccb4",
            "f911a922fccf4c309f09803cd5afe842",
            "d60e0b9c71784a7a87c1a782b5cfd224",
            "4cd6b62bc70c42178e56a0c2e01cc76a",
            "f6c0447a883945b8973fe4c6063e5934",
            "4ca9979c31394412af517eee639ddcfb",
            "00b78a0c7ab94b04a80dab81f08c4e88",
            "59c1006b93824b88b042782591913eea",
            "2202dc3cf22649dc910862d37ec28cb6",
            "8087c2f73cc64ac5a66fb319263b4e10",
            "5f0901624f4041b1994b49a3ab1be82f",
            "2deee584b165405281ae402675f22f85",
            "6c2c786c8fbe4256baff7d98c1bab07f",
            "433fc96f3609478c8c6aa48873d741a4",
            "d17d116b5ba34a468a35d80216384ca6",
            "f663b50655314ea4bbf8884250fbd9fe",
            "8e036dd9a4494a538bdcbe83ee3c3d06",
            "1d182909914842e2a8e06c0e7ca436df",
            "e4c3d15c5cd94e3c9489de48702e6848",
            "673a81ac011846f7a855d46716a56fd5",
            "8db3fb7178b6409eb75c271bc888081e",
            "5cf7dbd4edfa4c5a83ed0521d9fc5eb1",
            "a5ba108402e941dbb64dd1099bc8c1c3",
            "594aa85647454693a058ab738b6c6268",
            "1a53cae95c3e44bcb8c39e3c47552740",
            "326510b412224d4d979a6009fab2e683",
            "53f26580a4d34b08952a02bdd8154f8d",
            "546683616f114b05992da9a9cc95a55f",
            "01aa81ca076840569cb688c74e62dfbe",
            "1d905f36c22447d59b0c88ffdb51ab78",
            "29bd043d91574b45833d2cc04da61e33",
            "fb580cce9f60417c8e3d4230e85eb3ac",
            "71ab27c69c474a1ab522b667729e3863",
            "93bb467242184554a304b1131ea67000",
            "6cf8cf4a93634a539b6c9fada0fcad93",
            "d726f4bae06f43aebe6b504bcd6df903",
            "adaef6d1f85f4f6c9b587ba722912a80",
            "5e8d9b92347d463ab728f9754fc17d91",
            "71a64f0a447744f482765d3ceee8080f",
            "3fa98e32b6a74df5bb5551004538416b",
            "ee7e3b7262ed4e2d8031f0bc8d74b709",
            "e81d78fdea704c33b5f34bcfc112367f",
            "365fd3166ad6467d80b7948c2ededee0",
            "6dff9c660a774ff98660baa642a0f1b2",
            "dc8c1d9b67f043d1a7fd180bb6d2346a",
            "4860b9dd54094d6bab65bc3757a392f1",
            "a2d8e873e7d049fdbc0da753eb03df0e",
            "b934f02abbe0414db742cac592379be6",
            "9c5daa52339346bc95e8363504e0494f",
            "e888753994de40c4b05806301b72e9bd"
          ]
        },
        "id": "hFPQf09xq28n",
        "outputId": "cedb9dd0-5aa4-414d-ad70-eb08e6444b18"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading builder script:   0%|          | 0.00/6.36k [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "1c66e15c3fed416e9ffcced39429c276"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading metadata:   0%|          | 0.00/2.63k [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "e30bbdcbf73745aabcba4d2016554a41"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading readme:   0%|          | 0.00/7.81k [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "c70b5856b43843ebb1e24f3cc6102cd2"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading data:   0%|          | 0.00/118M [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "00b78a0c7ab94b04a80dab81f08c4e88"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Generating train split:   0%|          | 0/38969 [00:00<?, ? examples/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "1d182909914842e2a8e06c0e7ca436df"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Generating test split:   0%|          | 0/7500 [00:00<?, ? examples/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "01aa81ca076840569cb688c74e62dfbe"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Generating validation split:   0%|          | 0/6327 [00:00<?, ? examples/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "3fa98e32b6a74df5bb5551004538416b"
            }
          },
          "metadata": {}
        }
      ],
      "source": [
        "crd3 = load_dataset(\"crd3\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "id": "EdyoQogjq62H"
      },
      "outputs": [],
      "source": [
        "df = [i.to_pandas() for i in crd3.values()]\n",
        "\n",
        "df = pd.concat(df, axis=0)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 337
        },
        "id": "x70Db67ztaH9",
        "outputId": "40ea9085-e2d3-4319-b14c-403dc92dfa1c"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                               chunk  chunk_id  turn_start  \\\n",
              "0  Matthew Mercer introduces himself and the conc...         0           0   \n",
              "1  Matt explains how their eight-player game has ...         1           0   \n",
              "2  Donations will be accepted in the following we...         2          21   \n",
              "3  Marisha Ray explains how 826LA helps children ...         3          22   \n",
              "4  The cast introduce themselves and their roles:...         4          23   \n",
              "\n",
              "   turn_end  alignment_score  \\\n",
              "0         0         0.000000   \n",
              "1        21         0.498612   \n",
              "2        22         1.078358   \n",
              "3        23         1.280000   \n",
              "4        24         0.196787   \n",
              "\n",
              "                                               turns  \n",
              "0  [{'names': ['MATT'], 'utterances': ['Hello eve...  \n",
              "1  [{'names': ['MATT'], 'utterances': ['Hello eve...  \n",
              "2  [{'names': ['MATT'], 'utterances': ['Well, gre...  \n",
              "3  [{'names': ['MARISHA'], 'utterances': ['Yes! 8...  \n",
              "4  [{'names': ['MATT'], 'utterances': ['Yes. Also...  "
            ],
            "text/html": [
              "\n",
              "\n",
              "  <div id=\"df-16ef7931-905a-45f5-b158-e86f140db701\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>chunk</th>\n",
              "      <th>chunk_id</th>\n",
              "      <th>turn_start</th>\n",
              "      <th>turn_end</th>\n",
              "      <th>alignment_score</th>\n",
              "      <th>turns</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Matthew Mercer introduces himself and the conc...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>[{'names': ['MATT'], 'utterances': ['Hello eve...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Matt explains how their eight-player game has ...</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>21</td>\n",
              "      <td>0.498612</td>\n",
              "      <td>[{'names': ['MATT'], 'utterances': ['Hello eve...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Donations will be accepted in the following we...</td>\n",
              "      <td>2</td>\n",
              "      <td>21</td>\n",
              "      <td>22</td>\n",
              "      <td>1.078358</td>\n",
              "      <td>[{'names': ['MATT'], 'utterances': ['Well, gre...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Marisha Ray explains how 826LA helps children ...</td>\n",
              "      <td>3</td>\n",
              "      <td>22</td>\n",
              "      <td>23</td>\n",
              "      <td>1.280000</td>\n",
              "      <td>[{'names': ['MARISHA'], 'utterances': ['Yes! 8...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>The cast introduce themselves and their roles:...</td>\n",
              "      <td>4</td>\n",
              "      <td>23</td>\n",
              "      <td>24</td>\n",
              "      <td>0.196787</td>\n",
              "      <td>[{'names': ['MATT'], 'utterances': ['Yes. Also...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-16ef7931-905a-45f5-b158-e86f140db701')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "\n",
              "\n",
              "\n",
              "    <div id=\"df-92284e05-442d-401d-94c3-af374716c7d3\">\n",
              "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-92284e05-442d-401d-94c3-af374716c7d3')\"\n",
              "              title=\"Suggest charts.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "      </button>\n",
              "    </div>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "    background-color: #E8F0FE;\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: #1967D2;\n",
              "    height: 32px;\n",
              "    padding: 0 0 0 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: #E2EBFA;\n",
              "    box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: #174EA6;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "    background-color: #3B4455;\n",
              "    fill: #D2E3FC;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart:hover {\n",
              "    background-color: #434B5C;\n",
              "    box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "    filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "    fill: #FFFFFF;\n",
              "  }\n",
              "</style>\n",
              "\n",
              "    <script>\n",
              "      async function quickchart(key) {\n",
              "        const containerElement = document.querySelector('#' + key);\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      }\n",
              "    </script>\n",
              "\n",
              "      <script>\n",
              "\n",
              "function displayQuickchartButton(domScope) {\n",
              "  let quickchartButtonEl =\n",
              "    domScope.querySelector('#df-92284e05-442d-401d-94c3-af374716c7d3 button.colab-df-quickchart');\n",
              "  quickchartButtonEl.style.display =\n",
              "    google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "}\n",
              "\n",
              "        displayQuickchartButton(document);\n",
              "      </script>\n",
              "      <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-16ef7931-905a-45f5-b158-e86f140db701 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-16ef7931-905a-45f5-b158-e86f140db701');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ],
      "source": [
        "df.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zTr8nP47tnLM",
        "outputId": "89c16072-6422-4147-eea0-a82f35e9cf20"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "chunk              0\n",
              "chunk_id           0\n",
              "turn_start         0\n",
              "turn_end           0\n",
              "alignment_score    0\n",
              "turns              0\n",
              "dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 6
        }
      ],
      "source": [
        "df.isna().sum()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 507
        },
        "id": "wIpR_8nhfeJ2",
        "outputId": "a21729f4-b513-4b14-ff66-a91851ced21a"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "<ipython-input-7-f9bae3447ea4>:3: FutureWarning: The default value of numeric_only in DataFrame.corr is deprecated. In a future version, it will default to False. Select only valid columns or specify the value of numeric_only to silence this warning.\n",
            "  sns.heatmap(df[cols].corr(),\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<Axes: >"
            ]
          },
          "metadata": {},
          "execution_count": 7
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "cols = ['chunk',\t'chunk_id',\t'turn_start',\t'turn_end',\t'alignment_score',\t'turns']\n",
        "\n",
        "sns.heatmap(df[cols].corr(),\n",
        "            cbar=True,\n",
        "            annot=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "id": "kQIXsoT_yvmo"
      },
      "outputs": [],
      "source": [
        "turns = [f\"{' '.join(i[0]['utterances'])}\" for i in df[\"turns\"]]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "id": "NeijVpJ2zhyB"
      },
      "outputs": [],
      "source": [
        "chunk = [i for i in df[\"chunk\"]]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "id": "EQ3cDBfudVUQ"
      },
      "outputs": [],
      "source": [
        "X = chunk\n",
        "y = turns"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "id": "lLTWyWRqUSNz"
      },
      "outputs": [],
      "source": [
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=.2)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "id": "mVLz1_AWTHMZ"
      },
      "outputs": [],
      "source": [
        "dataset = tf.data.Dataset.from_tensor_slices((X_train, y_train))\n",
        "\n",
        "dataset_2 = tf.data.Dataset.from_tensor_slices((X_test, y_test))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 90
        },
        "id": "F5bmB4s5Weyu",
        "outputId": "ae2f9e5e-fabf-4c0e-d94d-7773e4fae743"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading data from https://storage.googleapis.com/download.tensorflow.org/models/ted_hrlr_translate_pt_en_converter.zip\n",
            "184801/184801 [==============================] - 0s 0us/step\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'./ted_hrlr_translate_pt_en_converter.zip'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 13
        }
      ],
      "source": [
        "model_name = \"ted_hrlr_translate_pt_en_converter\"\n",
        "tf.keras.utils.get_file(\n",
        "    f\"{model_name}.zip\",\n",
        "    f\"https://storage.googleapis.com/download.tensorflow.org/models/{model_name}.zip\",\n",
        "    cache_dir='.', cache_subdir='', extract=True\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "id": "fBRPDxQk6nv2"
      },
      "outputs": [],
      "source": [
        "tokenizers = tf.saved_model.load(model_name)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "id": "_BYApmRlsLHo"
      },
      "outputs": [],
      "source": [
        "def tokenize_pairs(en, en_2):\n",
        "    en = tokenizers.en.tokenize(en)\n",
        "    # Convert from ragged to dense, padding with zeros.\n",
        "    en = en.to_tensor()\n",
        "\n",
        "    en_2 = tokenizers.en.tokenize(en_2)\n",
        "    # Convert from ragged to dense, padding with zeros.\n",
        "    en_2 = en_2.to_tensor()\n",
        "    return en, en_2"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "id": "RU-QnMiw8GVy"
      },
      "outputs": [],
      "source": [
        "MAX_LENGTH = 128\n",
        "BUFFER_SIZE = 20000\n",
        "BATCH_SIZE = 64\n",
        "\n",
        "def prepare_batch(en, en_2):\n",
        "    en = tokenizers.en.tokenize(en)      # Output is ragged.\n",
        "    en = en[:, :MAX_LENGTH]    # Trim to MAX_TOKENS.\n",
        "    en = en.to_tensor()  # Convert to 0-padded dense Tensor\n",
        "\n",
        "    en_2 = tokenizers.en.tokenize(en_2)\n",
        "    en_2 = en_2[:, :(MAX_LENGTH+1)]\n",
        "    en_inputs = en_2[:, :-1].to_tensor()  # Drop the [END] tokens\n",
        "    en_labels = en_2[:, 1:].to_tensor()   # Drop the [START] tokens\n",
        "\n",
        "    return (en, en_inputs), en_labels\n",
        "\n",
        "def make_batches(ds):\n",
        "  return ds.cache()\\\n",
        "  .shuffle(BUFFER_SIZE)\\\n",
        "  .batch(BATCH_SIZE)\\\n",
        "  .map(prepare_batch, num_parallel_calls=tf.data.AUTOTUNE)\\\n",
        "  .prefetch(tf.data.AUTOTUNE)\n",
        "\n",
        "\n",
        "train_dataset = make_batches(dataset)\n",
        "test_dataset = make_batches(dataset_2)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ld6l-YwOUgdw",
        "outputId": "7914caaf-1adb-4ad4-fd85-3014115bc3a2"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "((<tf.Tensor: shape=(64, 128), dtype=int64, numpy=\n",
            "array([[   2,   71,  518, ...,    0,    0,    0],\n",
            "       [   2,   58, 3895, ...,    0,    0,    0],\n",
            "       [   2,   83, 2190, ...,    0,    0,    0],\n",
            "       ...,\n",
            "       [   2,   43, 1548, ...,    0,    0,    0],\n",
            "       [   2,   76,  546, ...,    0,    0,    0],\n",
            "       [   2,   83,  731, ...,    0,    0,    0]])>, <tf.Tensor: shape=(64, 128), dtype=int64, numpy=\n",
            "array([[   2,   81,  431, ...,   71,   43,  690],\n",
            "       [   2,  679, 3074, ...,    0,    0,    0],\n",
            "       [   2, 1287,   88, ...,    0,    0,    0],\n",
            "       ...,\n",
            "       [   2,  525,   13, ...,    0,    0,    0],\n",
            "       [   2,   37, 5180, ...,    0,    0,    0],\n",
            "       [   2,  221,  887, ..., 3284, 2788,  171]])>), <tf.Tensor: shape=(64, 128), dtype=int64, numpy=\n",
            "array([[  81,  431,   92, ...,   43,  690,   13],\n",
            "       [ 679, 3074,  459, ...,    0,    0,    0],\n",
            "       [1287,   88,   15, ...,    0,    0,    0],\n",
            "       ...,\n",
            "       [ 525,   13,   45, ...,    0,    0,    0],\n",
            "       [  37, 5180, 4292, ...,    0,    0,    0],\n",
            "       [ 221,  887,  147, ..., 2788,  171, 1081]])>)\n",
            "((<tf.Tensor: shape=(64, 128), dtype=int64, numpy=\n",
            "array([[   2,  117,  887, ...,    0,    0,    0],\n",
            "       [   2,   77,  233, ...,    0,    0,    0],\n",
            "       [   2,  154,  103, ...,    0,    0,    0],\n",
            "       ...,\n",
            "       [   2,   58, 3419, ...,    0,    0,    0],\n",
            "       [   2,  217,  604, ...,    0,    0,    0],\n",
            "       [   2, 3639, 3934, ...,    0,    0,    0]])>, <tf.Tensor: shape=(64, 128), dtype=int64, numpy=\n",
            "array([[   2,    1,  154, ...,    0,    0,    0],\n",
            "       [   2,  101,  169, ...,   13,   72, 3310],\n",
            "       [   2,   87,   55, ...,    0,    0,    0],\n",
            "       ...,\n",
            "       [   2,  100,   79, ...,    0,    0,    0],\n",
            "       [   2,   79,  149, ...,    0,    0,    0],\n",
            "       [   2,   45,  324, ...,    0,    0,    0]])>), <tf.Tensor: shape=(64, 128), dtype=int64, numpy=\n",
            "array([[   1,  154,   13, ...,    0,    0,    0],\n",
            "       [ 101,  169,   15, ...,   72, 3310, 5108],\n",
            "       [  87,   55,  840, ...,    0,    0,    0],\n",
            "       ...,\n",
            "       [ 100,   79,  942, ...,    0,    0,    0],\n",
            "       [  79,  149,  586, ...,    0,    0,    0],\n",
            "       [  45,  324,   89, ...,    0,    0,    0]])>)\n",
            "((<tf.Tensor: shape=(64, 128), dtype=int64, numpy=\n",
            "array([[   2,  117,   43, ...,    0,    0,    0],\n",
            "       [   2,   49, 6392, ...,    0,    0,    0],\n",
            "       [   2,   46,  841, ...,    0,    0,    0],\n",
            "       ...,\n",
            "       [   2,  175,   80, ...,    0,    0,    0],\n",
            "       [   2,   58, 3895, ...,    0,    0,    0],\n",
            "       [   2,  118,  110, ...,    3,    0,    0]])>, <tf.Tensor: shape=(64, 128), dtype=int64, numpy=\n",
            "array([[   2,    1,  154, ...,    0,    0,    0],\n",
            "       [   2, 4006,   13, ..., 1803,   15,   10],\n",
            "       [   2,  193,   73, ...,   89, 2308,   92],\n",
            "       ...,\n",
            "       [   2, 4006,  506, ...,    0,    0,    0],\n",
            "       [   2,  191,   76, ...,    0,    0,    0],\n",
            "       [   2,  175,  558, ...,   45,    9,  142]])>), <tf.Tensor: shape=(64, 128), dtype=int64, numpy=\n",
            "array([[   1,  154,   13, ...,    0,    0,    0],\n",
            "       [4006,   13,   72, ...,   15,   10,   39],\n",
            "       [ 193,   73,  247, ..., 2308,   92,   81],\n",
            "       ...,\n",
            "       [4006,  506,   13, ...,    0,    0,    0],\n",
            "       [ 191,   76,   15, ...,    0,    0,    0],\n",
            "       [ 175,  558, 3614, ...,    9,  142,  179]])>)\n"
          ]
        }
      ],
      "source": [
        "for i in train_dataset.take(3):\n",
        "  print(i)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "id": "monMgQvcZzP7"
      },
      "outputs": [],
      "source": [
        "def mask(seq):\n",
        "  return tf.cast(tf.math.equal(seq, 0), tf.float32)[:, tf.newaxis, tf.newaxis, :]\n",
        "\n",
        "def mask_final(size):\n",
        "  return 1 - tf.linalg.band_part(tf.ones((size, size)), -1, 0)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FqA7c65ohQLT",
        "outputId": "f0064b27-4869-44cc-e8d0-a5ebfbf36bb5"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(<tf.Tensor: shape=(3, 1, 1, 5), dtype=float32, numpy=\n",
              " array([[[[0., 0., 1., 1., 0.]]],\n",
              " \n",
              " \n",
              "        [[[0., 0., 0., 1., 1.]]],\n",
              " \n",
              " \n",
              "        [[[1., 1., 1., 0., 0.]]]], dtype=float32)>,\n",
              " <tf.Tensor: shape=(3, 3), dtype=float32, numpy=\n",
              " array([[0., 1., 1.],\n",
              "        [0., 0., 1.],\n",
              "        [0., 0., 0.]], dtype=float32)>)"
            ]
          },
          "metadata": {},
          "execution_count": 19
        }
      ],
      "source": [
        "x = tf.constant([[7, 6, 0, 0, 1], [1, 2, 3, 0, 0], [0, 0, 0, 4, 5]])\n",
        "x_ = tf.random.uniform((1, 3))\n",
        "\n",
        "mask(x), mask_final(x_.shape[1])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "metadata": {
        "id": "fA9vK6-CAS8w"
      },
      "outputs": [],
      "source": [
        "class PositionalFF(keras.layers.Layer):\n",
        "  def __init__(self, d_model, diff):\n",
        "    super(PositionalFF, self).__init__()\n",
        "\n",
        "    self.model = keras.Sequential([\n",
        "        Dense(diff, activation='relu'),\n",
        "        Dense(d_model)\n",
        "    ])\n",
        "\n",
        "  def call(self, x):\n",
        "    return self.model(x)\n",
        "\n",
        "def get_angles(pos, i, d_model):\n",
        "  angle_rates = 1 / np.power(10000, (2 * (i//2)) / np.float32(d_model))\n",
        "  return pos * angle_rates\n",
        "\n",
        "def positional_encoder(i, d_model):\n",
        "  angle_rads = get_angles(np.arange(i)[:, np.newaxis],\n",
        "                          np.arange(d_model)[np.newaxis, :],\n",
        "                          d_model)\n",
        "\n",
        "  angle_rads[:, 0::2] = np.sin(angle_rads[:, 0::2])\n",
        "\n",
        "  angle_rads[:, 1::2] = np.cos(angle_rads[:, 1::2])\n",
        "\n",
        "  pos_encoding = angle_rads[np.newaxis, ...]\n",
        "\n",
        "  return tf.cast(pos_encoding, dtype=tf.float32)\n",
        "\n",
        "class Add_and_Norm(keras.layers.Layer):\n",
        "  def __init__(self, dropout, epsilon=1e-6):\n",
        "    super(Add_and_Norm, self).__init__()\n",
        "\n",
        "    self.add = Dropout(dropout)\n",
        "    self.norm = keras.layers.LayerNormalization(epsilon=epsilon)\n",
        "\n",
        "  def call(self, x, y):\n",
        "    return self.norm(self.add(x) + y)\n",
        "\n",
        "class Encoder_Layer(keras.layers.Layer):\n",
        "  def __init__(self, d_model, num_heads, dff, rate=0.1):\n",
        "    super(Encoder_Layer, self).__init__()\n",
        "\n",
        "    self.multihead = keras.layers.MultiHeadAttention(num_heads, d_model)\n",
        "    self.add_norm = Add_and_Norm(rate)\n",
        "    self.positional_ff = PositionalFF(d_model ,dff)\n",
        "    self.add_norm_2 = Add_and_Norm(rate)\n",
        "\n",
        "  def call(self, x):\n",
        "    multiheads, _ = self.multihead(x, x, x, use_causal_mask = True, return_attention_scores=True)\n",
        "    addnorm = self.add_norm(multiheads, x)\n",
        "    pos_ff = self.positional_ff(addnorm)\n",
        "    x = self.add_norm_2(pos_ff, addnorm)\n",
        "\n",
        "    return x\n",
        "\n",
        "class Transformers_Encoder(keras.layers.Layer):\n",
        "  def __init__(self, num_layers, d_model, num_heads, dff, input_vocab_size,\n",
        "               maximum_position_encoding, rate=0.1):\n",
        "    super(Transformers_Encoder, self).__init__()\n",
        "\n",
        "    self.d_model = d_model\n",
        "    self.num_layers = num_layers\n",
        "\n",
        "    self.embeding = Embedding(input_vocab_size, d_model)\n",
        "    self.pos_encoding = positional_encoder(maximum_position_encoding,\n",
        "                                            self.d_model)\n",
        "\n",
        "    self.encoder_layer = [Encoder_Layer(d_model, num_heads, dff, rate) for _ in range(num_layers)]\n",
        "    self.dropout = tf.keras.layers.Dropout(rate)\n",
        "\n",
        "  def call(self, x):\n",
        "    b = tf.shape(x)[1]\n",
        "    x = self.embeding(x)\n",
        "    x *= tf.math.sqrt(tf.cast(self.d_model, dtype=tf.float32))\n",
        "    x = x + self.pos_encoding[:, :b, :]\n",
        "\n",
        "    x = self.dropout(x)\n",
        "\n",
        "    for i in range(self.num_layers):\n",
        "      x = self.encoder_layer[i](x)\n",
        "\n",
        "    return x"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "metadata": {
        "id": "jvxCZK9sEkwy"
      },
      "outputs": [],
      "source": [
        "class Decoder_Layer(keras.layers.Layer):\n",
        "  def __init__(self, d_model, num_heads, dff, rate=0.1):\n",
        "    super(Decoder_Layer, self).__init__()\n",
        "\n",
        "    self.multihead = keras.layers.MultiHeadAttention(num_heads, d_model)\n",
        "    self.multihead_2 = keras.layers.MultiHeadAttention(num_heads, d_model)\n",
        "\n",
        "    self.add_norm = Add_and_Norm(rate)\n",
        "    self.positional_ff = PositionalFF(d_model, dff)\n",
        "    self.add_norm_2 = Add_and_Norm(rate)\n",
        "    self.add_norm_3 = Add_and_Norm(rate)\n",
        "\n",
        "  def call(self, x):\n",
        "    multiheads, attention = self.multihead(x[0], x[0], x[0], return_attention_scores=True, use_causal_mask = True)\n",
        "\n",
        "    addnorm = self.add_norm(multiheads, x[0])\n",
        "\n",
        "    multihead_2, attention_2 = self.multihead_2(addnorm, x[1], x[1], return_attention_scores=True, use_causal_mask = True)\n",
        "\n",
        "    addnorm_2 = self.add_norm_2(multihead_2, addnorm)\n",
        "\n",
        "    pos_ff = self.positional_ff(addnorm_2)\n",
        "\n",
        "    x[0] = self.add_norm_3(pos_ff, addnorm_2)\n",
        "\n",
        "    return x[0], attention, attention_2\n",
        "\n",
        "class Transformer_Decoder(keras.layers.Layer):\n",
        "  def __init__(self, num_layers, d_model, num_heads, dff, target_vocab_size,\n",
        "               maximum_position_encoding, rate=0.1):\n",
        "    super(Transformer_Decoder, self).__init__()\n",
        "\n",
        "    self.d_model = d_model\n",
        "    self.num_layers = num_layers\n",
        "\n",
        "    self.pos_encoding = positional_encoder(maximum_position_encoding, d_model)\n",
        "\n",
        "    self.embeding = Embedding(maximum_position_encoding, d_model)\n",
        "\n",
        "    self.decoder_layer = [Decoder_Layer(d_model, num_heads, dff, rate) for _ in range(num_layers)]\n",
        "\n",
        "    self.dropout = tf.keras.layers.Dropout(rate)\n",
        "\n",
        "  def call(self, x):\n",
        "    b = tf.shape(x[0])[1]\n",
        "\n",
        "    attention_w = {}\n",
        "\n",
        "    x_ = self.embeding(x[0])\n",
        "    x_ *= tf.math.sqrt(tf.cast(self.d_model, dtype=tf.float32))\n",
        "    x_ = x_ + self.pos_encoding[:, :b, :]\n",
        "\n",
        "    x_ =  self.dropout(x_)\n",
        "\n",
        "    for i in range(self.num_layers):\n",
        "      x_, block1, block2 = self.decoder_layer[i]([x_, x[1]])\n",
        "\n",
        "      attention_w[f'decoder_layer{i+1}_block1'] = block1\n",
        "      attention_w[f'decoder_layer{i+1}_block2'] = block2\n",
        "\n",
        "    return x_, attention_w"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 22,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Ksa9KdMvyEtU",
        "outputId": "18f8b145-a1d7-4c1e-fe6b-c3309f077827"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(64, 62, 512)\n"
          ]
        }
      ],
      "source": [
        "sample_encoder = Transformers_Encoder(num_layers=2, d_model=512, num_heads=8,\n",
        "                         dff=2048, input_vocab_size=8500,\n",
        "                         maximum_position_encoding=10000)\n",
        "temp_input = tf.random.uniform((64, 62), dtype=tf.int64, minval=0, maxval=200)\n",
        "\n",
        "sample_encoder_output = sample_encoder(temp_input, training=False)\n",
        "\n",
        "print(sample_encoder_output.shape)  # (batch_size, input_seq_len, d_model)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "188oWFFDHz7Y",
        "outputId": "b4aa0643-1ed7-4752-fc47-522c928ff07f"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(64, 26, 512) (64, 8, 26, 62)\n"
          ]
        }
      ],
      "source": [
        "sample_decoder = Transformer_Decoder(num_layers=2, d_model=512, num_heads=8,\n",
        "                         dff=2048, target_vocab_size=8000,\n",
        "                         maximum_position_encoding=5000)\n",
        "temp_input = tf.random.uniform((64, 26), dtype=tf.int64, minval=0, maxval=200)\n",
        "\n",
        "output, attn = sample_decoder([temp_input, sample_encoder_output],\n",
        "                              training=False)\n",
        "\n",
        "print(output.shape, attn['decoder_layer2_block2'].shape)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 24,
      "metadata": {
        "id": "9gRlVBNsOW9E"
      },
      "outputs": [],
      "source": [
        "num_layers = 4\n",
        "d_model = 128\n",
        "dff = 512\n",
        "num_heads = 8\n",
        "dropout_rate = .1\n",
        "\n",
        "class Transformer(Model):\n",
        "  def __init__(self, num_layers, d_model, num_heads, dff, input_vocab_size,\n",
        "               target_vocab_size, pe_input, pe_target, rate=0.1):\n",
        "    super(Transformer, self).__init__()\n",
        "    self.encoder = Transformers_Encoder(num_layers, d_model, num_heads, dff,\n",
        "                             input_vocab_size, pe_input, rate)\n",
        "\n",
        "    self.decoder = Transformer_Decoder(num_layers, d_model, num_heads, dff,\n",
        "                           target_vocab_size, pe_target, rate)\n",
        "\n",
        "    self.dense = Dense(target_vocab_size)\n",
        "\n",
        "\n",
        "  def create_masks(self, inp, tar):\n",
        "    enc_padding_mask = mask(inp)\n",
        "\n",
        "    dec_padding_mask = mask(inp)\n",
        "\n",
        "    look_ahead_mask = mask_final(tf.shape(tar)[1])\n",
        "    dec_target_padding_mask = mask(tar)\n",
        "    look_ahead_mask = tf.maximum(dec_target_padding_mask, look_ahead_mask)\n",
        "\n",
        "    return enc_padding_mask, look_ahead_mask, dec_padding_mask\n",
        "\n",
        "  def call(self, inputs):\n",
        "    context, x  = inputs\n",
        "\n",
        "    context = self.encoder(context)  # (batch_size, context_len, d_model)\n",
        "\n",
        "    x, _ = self.decoder([x, context])  # (batch_size, target_len, d_model)\n",
        "\n",
        "    logits = self.dense(x)  # (batch_size, target_len, target_vocab_size)\n",
        "\n",
        "    try:\n",
        "      del logits._keras_mask\n",
        "    except AttributeError:\n",
        "      pass\n",
        "\n",
        "    return logits\n",
        "\n",
        "transformer = Transformer(\n",
        "    num_layers=num_layers,\n",
        "    d_model=d_model,\n",
        "    num_heads=num_heads,\n",
        "    dff=dff,\n",
        "    input_vocab_size=tokenizers.en.get_vocab_size().numpy(),\n",
        "    target_vocab_size=tokenizers.en.get_vocab_size().numpy(),\n",
        "    pe_input=1000,\n",
        "    pe_target=1000,\n",
        "    rate=dropout_rate)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 25,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Pxp1eyPUAkOU",
        "outputId": "705afb3b-e392-481d-9b31-5b0b4bb22a8a"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tf.Tensor: shape=(), dtype=string, numpy=b'cheap sustainedridge pantsges sure skull race race counted counted race race raceographicographicographic administration counted counted counted race race newton newton newton nanopatch officer race race race race newton newton newton produce hat full bought full bought newton newton newtonically fullically produce produce produce'>"
            ]
          },
          "metadata": {},
          "execution_count": 25
        }
      ],
      "source": [
        "class Translator(tf.Module):\n",
        "  def __init__(self, tokenizers, transformer):\n",
        "    self.tokenizers = tokenizers\n",
        "    self.transformer = transformer\n",
        "\n",
        "  def __call__(self, sentence, max_length=50):\n",
        "    assert isinstance(sentence, tf.Tensor)\n",
        "    if len(sentence.shape) == 0:\n",
        "      sentence = sentence[tf.newaxis]\n",
        "\n",
        "    sentence = self.tokenizers.en.tokenize(sentence).to_tensor()\n",
        "\n",
        "    encoder_input = sentence\n",
        "\n",
        "    start_end = self.tokenizers.en.tokenize([''])[0]\n",
        "    start = start_end[0][tf.newaxis]\n",
        "    end = start_end[1][tf.newaxis]\n",
        "\n",
        "    output_array = tf.TensorArray(dtype=tf.int64, size=0, dynamic_size=True)\n",
        "    output_array = output_array.write(0, start)\n",
        "\n",
        "    for i in tf.range(max_length):\n",
        "      output = tf.transpose(output_array.stack())\n",
        "      predictions = self.transformer([encoder_input, output], training=False)\n",
        "\n",
        "      predictions = predictions[:, -1:, :]  # Shape `(batch_size, 1, vocab_size)`.\n",
        "\n",
        "      predicted_id = tf.argmax(predictions, axis=-1)\n",
        "\n",
        "      output_array = output_array.write(i+1, predicted_id[0])\n",
        "\n",
        "      if predicted_id == end:\n",
        "        break\n",
        "\n",
        "    output = tf.transpose(output_array.stack())\n",
        "    text = tokenizers.en.detokenize(output)[0]  # Shape: `()`.\n",
        "\n",
        "    tokens = tokenizers.en.lookup(output)[0]\n",
        "\n",
        "    self.transformer([encoder_input, output[:,:-1]], training=False)\n",
        "\n",
        "    return text, tokens\n",
        "\n",
        "\n",
        "translator = Translator(tokenizers, transformer)\n",
        "\n",
        "translated_text, translated_tokens = translator(\n",
        "    tf.constant('sentence'))\n",
        "translated_text"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 26,
      "metadata": {
        "id": "cKuKXXd-GmE5"
      },
      "outputs": [],
      "source": [
        "class CustomSchedule(tf.keras.optimizers.schedules.LearningRateSchedule):\n",
        "  def __init__(self, d_model, warmup_steps=4000):\n",
        "    super(CustomSchedule, self).__init__()\n",
        "\n",
        "    self.d_model = d_model\n",
        "    self.d_model = tf.cast(self.d_model, tf.float32)\n",
        "\n",
        "    self.warmup_steps = warmup_steps\n",
        "\n",
        "  def __call__(self, step):\n",
        "    step = tf.cast(step, tf.float32)\n",
        "\n",
        "    arg1 = tf.math.rsqrt(step)\n",
        "    arg2 = step * (self.warmup_steps ** -1.5)\n",
        "\n",
        "    return tf.math.rsqrt(self.d_model) * tf.math.minimum(arg1, arg2)\n",
        "\n",
        "  def get_config(self):\n",
        "    config = {\n",
        "    'd_model': self.d_model,\n",
        "    'warmup_steps': self.warmup_steps,\n",
        "\n",
        "     }\n",
        "    return config\n",
        "\n",
        "lr = CustomSchedule(d_model, 4000)\n",
        "\n",
        "optimizer = tf.keras.optimizers.Adam(lr, beta_1=0.9, beta_2=0.98, epsilon=1e-9) # 3e-5"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 27,
      "metadata": {
        "id": "wH8KXLj8R7jQ"
      },
      "outputs": [],
      "source": [
        "def masked_loss(label, pred):\n",
        "  mask = label != 0\n",
        "  loss_object = tf.keras.losses.SparseCategoricalCrossentropy(\n",
        "    from_logits=True, reduction='none')\n",
        "  loss = loss_object(label, pred)\n",
        "\n",
        "  mask = tf.cast(mask, dtype=loss.dtype)\n",
        "  loss *= mask\n",
        "\n",
        "  loss = tf.reduce_sum(loss)/tf.reduce_sum(mask)\n",
        "  return loss\n",
        "\n",
        "\n",
        "def masked_accuracy(label, pred):\n",
        "  pred = tf.argmax(pred, axis=2)\n",
        "  label = tf.cast(label, pred.dtype)\n",
        "  match = label == pred\n",
        "\n",
        "  mask = label != 0\n",
        "\n",
        "  match = match & mask\n",
        "\n",
        "  match = tf.cast(match, dtype=tf.float32)\n",
        "  mask = tf.cast(mask, dtype=tf.float32)\n",
        "  return tf.reduce_sum(match)/tf.reduce_sum(mask)\n",
        "\n",
        "transformer.compile(\n",
        "    loss=masked_loss,\n",
        "    optimizer=optimizer,\n",
        "    metrics=[masked_accuracy])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 28,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8DNzJgIySJqq",
        "outputId": "63303411-10ac-4ea7-8b02-c9c2d3ed3be9"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/50\n",
            "660/660 [==============================] - 357s 477ms/step - loss: 6.8644 - masked_accuracy: 0.0704 - val_loss: 5.2971 - val_masked_accuracy: 0.1489\n",
            "Epoch 2/50\n",
            "660/660 [==============================] - 267s 404ms/step - loss: 4.8442 - masked_accuracy: 0.1776 - val_loss: 4.4004 - val_masked_accuracy: 0.2137\n",
            "Epoch 3/50\n",
            "660/660 [==============================] - 262s 397ms/step - loss: 4.2175 - masked_accuracy: 0.2280 - val_loss: 3.8762 - val_masked_accuracy: 0.2676\n",
            "Epoch 4/50\n",
            "660/660 [==============================] - 259s 393ms/step - loss: 3.7901 - masked_accuracy: 0.2729 - val_loss: 3.4731 - val_masked_accuracy: 0.3212\n",
            "Epoch 5/50\n",
            "660/660 [==============================] - 259s 392ms/step - loss: 3.4700 - masked_accuracy: 0.3152 - val_loss: 3.1847 - val_masked_accuracy: 0.3655\n",
            "Epoch 6/50\n",
            "660/660 [==============================] - 242s 366ms/step - loss: 3.2461 - masked_accuracy: 0.3481 - val_loss: 2.9821 - val_masked_accuracy: 0.3987\n",
            "Epoch 7/50\n",
            "660/660 [==============================] - 241s 365ms/step - loss: 3.0673 - masked_accuracy: 0.3757 - val_loss: 2.7953 - val_masked_accuracy: 0.4338\n",
            "Epoch 8/50\n",
            "660/660 [==============================] - 241s 365ms/step - loss: 2.8813 - masked_accuracy: 0.4055 - val_loss: 2.6255 - val_masked_accuracy: 0.4651\n",
            "Epoch 9/50\n",
            "660/660 [==============================] - 239s 362ms/step - loss: 2.7343 - masked_accuracy: 0.4298 - val_loss: 2.4867 - val_masked_accuracy: 0.4939\n",
            "Epoch 10/50\n",
            "660/660 [==============================] - 258s 391ms/step - loss: 2.6163 - masked_accuracy: 0.4495 - val_loss: 2.3844 - val_masked_accuracy: 0.5121\n",
            "Epoch 11/50\n",
            "660/660 [==============================] - 259s 392ms/step - loss: 2.5192 - masked_accuracy: 0.4660 - val_loss: 2.2975 - val_masked_accuracy: 0.5313\n",
            "Epoch 12/50\n",
            "660/660 [==============================] - 259s 393ms/step - loss: 2.4403 - masked_accuracy: 0.4798 - val_loss: 2.2301 - val_masked_accuracy: 0.5449\n",
            "Epoch 13/50\n",
            "660/660 [==============================] - 243s 368ms/step - loss: 2.3699 - masked_accuracy: 0.4920 - val_loss: 2.1838 - val_masked_accuracy: 0.5541\n",
            "Epoch 14/50\n",
            "660/660 [==============================] - 241s 365ms/step - loss: 2.3099 - masked_accuracy: 0.5027 - val_loss: 2.1502 - val_masked_accuracy: 0.5634\n",
            "Epoch 15/50\n",
            "660/660 [==============================] - 259s 392ms/step - loss: 2.2566 - masked_accuracy: 0.5125 - val_loss: 2.0774 - val_masked_accuracy: 0.5782\n",
            "Epoch 16/50\n",
            "660/660 [==============================] - 258s 391ms/step - loss: 2.2076 - masked_accuracy: 0.5212 - val_loss: 2.0458 - val_masked_accuracy: 0.5854\n",
            "Epoch 17/50\n",
            "660/660 [==============================] - 241s 364ms/step - loss: 2.1639 - masked_accuracy: 0.5291 - val_loss: 2.0071 - val_masked_accuracy: 0.5928\n",
            "Epoch 18/50\n",
            "660/660 [==============================] - 258s 390ms/step - loss: 2.1238 - masked_accuracy: 0.5364 - val_loss: 1.9835 - val_masked_accuracy: 0.5983\n",
            "Epoch 19/50\n",
            "660/660 [==============================] - 240s 364ms/step - loss: 2.0888 - masked_accuracy: 0.5427 - val_loss: 1.9547 - val_masked_accuracy: 0.6057\n",
            "Epoch 20/50\n",
            "660/660 [==============================] - 259s 392ms/step - loss: 2.0559 - masked_accuracy: 0.5493 - val_loss: 1.9345 - val_masked_accuracy: 0.6100\n",
            "Epoch 21/50\n",
            "660/660 [==============================] - 258s 391ms/step - loss: 2.0260 - masked_accuracy: 0.5547 - val_loss: 1.9101 - val_masked_accuracy: 0.6145\n",
            "Epoch 22/50\n",
            "660/660 [==============================] - 258s 391ms/step - loss: 1.9947 - masked_accuracy: 0.5604 - val_loss: 1.8908 - val_masked_accuracy: 0.6181\n",
            "Epoch 23/50\n",
            "660/660 [==============================] - 258s 391ms/step - loss: 1.9668 - masked_accuracy: 0.5657 - val_loss: 1.8652 - val_masked_accuracy: 0.6235\n",
            "Epoch 24/50\n",
            "660/660 [==============================] - 258s 391ms/step - loss: 1.9413 - masked_accuracy: 0.5705 - val_loss: 1.8551 - val_masked_accuracy: 0.6279\n",
            "Epoch 25/50\n",
            "660/660 [==============================] - 259s 392ms/step - loss: 1.9192 - masked_accuracy: 0.5749 - val_loss: 1.8324 - val_masked_accuracy: 0.6330\n",
            "Epoch 26/50\n",
            "660/660 [==============================] - 241s 365ms/step - loss: 1.8931 - masked_accuracy: 0.5797 - val_loss: 1.8162 - val_masked_accuracy: 0.6355\n",
            "Epoch 27/50\n",
            "660/660 [==============================] - 258s 391ms/step - loss: 1.8727 - masked_accuracy: 0.5834 - val_loss: 1.8083 - val_masked_accuracy: 0.6413\n",
            "Epoch 28/50\n",
            "660/660 [==============================] - 240s 364ms/step - loss: 1.8531 - masked_accuracy: 0.5876 - val_loss: 1.7857 - val_masked_accuracy: 0.6417\n",
            "Epoch 29/50\n",
            "660/660 [==============================] - 258s 391ms/step - loss: 1.8319 - masked_accuracy: 0.5914 - val_loss: 1.7672 - val_masked_accuracy: 0.6461\n",
            "Epoch 30/50\n",
            "660/660 [==============================] - 240s 364ms/step - loss: 1.8120 - masked_accuracy: 0.5951 - val_loss: 1.7644 - val_masked_accuracy: 0.6460\n",
            "Epoch 31/50\n",
            "660/660 [==============================] - 240s 364ms/step - loss: 1.7953 - masked_accuracy: 0.5985 - val_loss: 1.7561 - val_masked_accuracy: 0.6494\n",
            "Epoch 32/50\n",
            "660/660 [==============================] - 239s 363ms/step - loss: 1.7789 - masked_accuracy: 0.6017 - val_loss: 1.7381 - val_masked_accuracy: 0.6531\n",
            "Epoch 33/50\n",
            "660/660 [==============================] - 258s 390ms/step - loss: 1.7625 - masked_accuracy: 0.6049 - val_loss: 1.7274 - val_masked_accuracy: 0.6555\n",
            "Epoch 34/50\n",
            "660/660 [==============================] - 265s 401ms/step - loss: 1.7459 - masked_accuracy: 0.6083 - val_loss: 1.7222 - val_masked_accuracy: 0.6572\n",
            "Epoch 35/50\n",
            "660/660 [==============================] - 259s 393ms/step - loss: 1.7316 - masked_accuracy: 0.6106 - val_loss: 1.7098 - val_masked_accuracy: 0.6604\n",
            "Epoch 36/50\n",
            "660/660 [==============================] - 258s 391ms/step - loss: 1.7167 - masked_accuracy: 0.6138 - val_loss: 1.6998 - val_masked_accuracy: 0.6616\n",
            "Epoch 37/50\n",
            "660/660 [==============================] - 258s 391ms/step - loss: 1.7036 - masked_accuracy: 0.6163 - val_loss: 1.6926 - val_masked_accuracy: 0.6662\n",
            "Epoch 38/50\n",
            "660/660 [==============================] - 258s 390ms/step - loss: 1.6883 - masked_accuracy: 0.6193 - val_loss: 1.6769 - val_masked_accuracy: 0.6678\n",
            "Epoch 39/50\n",
            "660/660 [==============================] - 259s 392ms/step - loss: 1.6757 - masked_accuracy: 0.6217 - val_loss: 1.6736 - val_masked_accuracy: 0.6697\n",
            "Epoch 40/50\n",
            "660/660 [==============================] - 258s 390ms/step - loss: 1.6645 - masked_accuracy: 0.6241 - val_loss: 1.6641 - val_masked_accuracy: 0.6705\n",
            "Epoch 41/50\n",
            "660/660 [==============================] - 240s 363ms/step - loss: 1.6521 - masked_accuracy: 0.6266 - val_loss: 1.6557 - val_masked_accuracy: 0.6714\n",
            "Epoch 42/50\n",
            "660/660 [==============================] - 257s 390ms/step - loss: 1.6387 - masked_accuracy: 0.6290 - val_loss: 1.6498 - val_masked_accuracy: 0.6724\n",
            "Epoch 43/50\n",
            "660/660 [==============================] - 258s 391ms/step - loss: 1.6260 - masked_accuracy: 0.6317 - val_loss: 1.6405 - val_masked_accuracy: 0.6752\n",
            "Epoch 44/50\n",
            "660/660 [==============================] - 240s 363ms/step - loss: 1.6156 - masked_accuracy: 0.6337 - val_loss: 1.6313 - val_masked_accuracy: 0.6783\n",
            "Epoch 45/50\n",
            "660/660 [==============================] - 257s 390ms/step - loss: 1.6054 - masked_accuracy: 0.6357 - val_loss: 1.6179 - val_masked_accuracy: 0.6797\n",
            "Epoch 46/50\n",
            "660/660 [==============================] - 258s 390ms/step - loss: 1.5952 - masked_accuracy: 0.6375 - val_loss: 1.6273 - val_masked_accuracy: 0.6808\n",
            "Epoch 47/50\n",
            "660/660 [==============================] - 242s 366ms/step - loss: 1.5843 - masked_accuracy: 0.6400 - val_loss: 1.6164 - val_masked_accuracy: 0.6820\n",
            "Epoch 48/50\n",
            "660/660 [==============================] - 258s 391ms/step - loss: 1.5762 - masked_accuracy: 0.6415 - val_loss: 1.6146 - val_masked_accuracy: 0.6801\n",
            "Epoch 49/50\n",
            "660/660 [==============================] - 259s 392ms/step - loss: 1.5663 - masked_accuracy: 0.6432 - val_loss: 1.5970 - val_masked_accuracy: 0.6852\n",
            "Epoch 50/50\n",
            "660/660 [==============================] - 258s 391ms/step - loss: 1.5571 - masked_accuracy: 0.6453 - val_loss: 1.5973 - val_masked_accuracy: 0.6865\n"
          ]
        }
      ],
      "source": [
        "model = transformer.fit(train_dataset,\n",
        "                epochs=50, validation_data=test_dataset)"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(model.history['masked_accuracy'],\n",
        "         label='Точность на тренеровочном наборе')\n",
        "plt.plot(model.history['val_masked_accuracy'],\n",
        "         label='Точность на тестовом наборе')\n",
        "plt.xlabel('Эпоха обучения')\n",
        "plt.ylabel('Точность')\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 449
        },
        "id": "r-y-JyUgjgyo",
        "outputId": "7334c7e9-073d-4c1b-abe8-60270596a12c"
      },
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 32,
      "metadata": {
        "id": "LCVsA2-XaKO-",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "76555fe6-f112-4c55-a8c6-b05dde5438ed"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Ответ нейросети: b\"i ' m going to flip out my bottle and co out my hand and try to give it some love the keys .\"\n",
            "Ожидаемый ответ: Loot Crate has donated two Loot Crate DX boxes to give away to you. We will be doing this at the break and at the end of the night, random in chat. You must be active! Stay tuned for that! Exclamation point!\n",
            "Ответ нейросети: b\"i ' m going to be at emconond .\"\n",
            "Ожидаемый ответ: Well, how much of them lived? It's Barovia, to be fair. Anyway, welcome. Before we hop into tonight's game, we have some announcements to get through. First and foremost, I believe, if Sam is ready, it looks like he's stretching himself. We have our fantastic sponsor, for campaign two since day one, our friends at D&D Beyond.\n",
            "Ответ нейросети: b\"i ' m going to take a quick and i ' m going to take a big break and i ' m going to use my action to try and get him in the la .\"\n",
            "Ожидаемый ответ: A giant glob of magma pours down and splashes down onto the ground, hitting everyone in the vicinity. Everyone down here takes 12 points of heat damage.\n",
            "Ответ нейросети: b\"i ' m going to try and roll him out of the way .\"\n",
            "Ожидаемый ответ: So go ahead and mark that. Okay. And this guy over here at the very top moves over as well and is going to go ahead and take two more shots at Scanlan. Since there's currently not a whole lot he can see in the area that's a threat to him, near this gleaming beacon of pink and purple up on the rooftop another friend of his is currently whaling on, so. Two attacks against you. That's going to be a 19.\n",
            "Ответ нейросети: b\"i ' m going to try and get out of the way .\"\n",
            "Ожидаемый ответ: All right, as you turn around the corner and look in, the interior is very low-lit.\n"
          ]
        }
      ],
      "source": [
        "for i in range(5):\n",
        "  translated_text, _ = translator(tf.constant(X_test[i]), 100)\n",
        "\n",
        "  print(f'Ответ нейросети: {translated_text}')\n",
        "  print('Ожидаемый ответ: ' + y_test[i])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 34,
      "metadata": {
        "id": "vskHQEb2jPQ9",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "f6775669-3202-453a-88fa-ae2fb8eef6eb"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "You - hello\n",
            "tf.Tensor(b\"i ' m going to take a quick . i ' m going to go straight and take a quick on the door and i sit out of the door and i sit out my hand out of the door and i ' m going to walk out my hand\", shape=(), dtype=string)\n",
            "You - break\n"
          ]
        }
      ],
      "source": [
        "while True:\n",
        "  inp = input('You - ')\n",
        "\n",
        "  if inp.lower() == 'break':\n",
        "    break\n",
        "\n",
        "  print(translator(tf.constant(inp), 50)[0])"
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "machine_shape": "hm",
      "provenance": [],
      "include_colab_link": true
    },
    "gpuClass": "standard",
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    },
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "1c66e15c3fed416e9ffcced39429c276": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_4053125e3122462f91ea021c341f972c",
              "IPY_MODEL_4186f779cb134e8eb16bbb973a993148",
              "IPY_MODEL_97ed262e06f94929a1dc231911455520"
            ],
            "layout": "IPY_MODEL_fe647113430c4f198f057a96fe423b47"
          }
        },
        "4053125e3122462f91ea021c341f972c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_2afeb66749b843c59a0e1c1d4e0dce68",
            "placeholder": "​",
            "style": "IPY_MODEL_7e08f1f170064973aeccf720ac731cf5",
            "value": "Downloading builder script: 100%"
          }
        },
        "4186f779cb134e8eb16bbb973a993148": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_73baa2ba932d4e709a07fdb53d443771",
            "max": 6364,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_77e7acbcca6a4aa087aace7e00fc7528",
            "value": 6364
          }
        },
        "97ed262e06f94929a1dc231911455520": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_3c04b77febf94713bf6b832b2a090be5",
            "placeholder": "​",
            "style": "IPY_MODEL_64238f56a0fe4697b350b18458b69f4a",
            "value": " 6.36k/6.36k [00:00&lt;00:00, 325kB/s]"
          }
        },
        "fe647113430c4f198f057a96fe423b47": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "2afeb66749b843c59a0e1c1d4e0dce68": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "7e08f1f170064973aeccf720ac731cf5": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "73baa2ba932d4e709a07fdb53d443771": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "77e7acbcca6a4aa087aace7e00fc7528": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "3c04b77febf94713bf6b832b2a090be5": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "64238f56a0fe4697b350b18458b69f4a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "e30bbdcbf73745aabcba4d2016554a41": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_2c6ca44cd391410a8c7a635d6828d2b0",
              "IPY_MODEL_6f918e5e466b4ae68b313147a5959dc6",
              "IPY_MODEL_24b31e5f85b14df89200707ff906c845"
            ],
            "layout": "IPY_MODEL_d60ad06322a241b09ddb281d3736279c"
          }
        },
        "2c6ca44cd391410a8c7a635d6828d2b0": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_9d13d00e10f34df0b5ded3df36cdbd54",
            "placeholder": "​",
            "style": "IPY_MODEL_166309fbc8ec4c23b9e001400aa1bbe8",
            "value": "Downloading metadata: 100%"
          }
        },
        "6f918e5e466b4ae68b313147a5959dc6": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_fdf6ee5be70e4cf09a9d81ad17170e62",
            "max": 2635,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_06b8d904df6d4cceaea939060c0fbae8",
            "value": 2635
          }
        },
        "24b31e5f85b14df89200707ff906c845": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_ba236a64db5a4a76956a891f7ab5150a",
            "placeholder": "​",
            "style": "IPY_MODEL_cef1d158396a42fd913a65858019e127",
            "value": " 2.63k/2.63k [00:00&lt;00:00, 57.0kB/s]"
          }
        },
        "d60ad06322a241b09ddb281d3736279c": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "9d13d00e10f34df0b5ded3df36cdbd54": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "166309fbc8ec4c23b9e001400aa1bbe8": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "fdf6ee5be70e4cf09a9d81ad17170e62": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "06b8d904df6d4cceaea939060c0fbae8": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "ba236a64db5a4a76956a891f7ab5150a": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "cef1d158396a42fd913a65858019e127": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "c70b5856b43843ebb1e24f3cc6102cd2": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_f3ab84e712044d1cbe81f3419eea88cc",
              "IPY_MODEL_940b488d575144a3bc5c41b9f9e3e987",
              "IPY_MODEL_e8e7911774424baba4779e83ab71cd1e"
            ],
            "layout": "IPY_MODEL_a0c97271414140df800aa7330f9b84ac"
          }
        },
        "f3ab84e712044d1cbe81f3419eea88cc": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_96556008bf65479b8220d457c943ccb4",
            "placeholder": "​",
            "style": "IPY_MODEL_f911a922fccf4c309f09803cd5afe842",
            "value": "Downloading readme: 100%"
          }
        },
        "940b488d575144a3bc5c41b9f9e3e987": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_d60e0b9c71784a7a87c1a782b5cfd224",
            "max": 7812,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_4cd6b62bc70c42178e56a0c2e01cc76a",
            "value": 7812
          }
        },
        "e8e7911774424baba4779e83ab71cd1e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_f6c0447a883945b8973fe4c6063e5934",
            "placeholder": "​",
            "style": "IPY_MODEL_4ca9979c31394412af517eee639ddcfb",
            "value": " 7.81k/7.81k [00:00&lt;00:00, 458kB/s]"
          }
        },
        "a0c97271414140df800aa7330f9b84ac": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "96556008bf65479b8220d457c943ccb4": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "f911a922fccf4c309f09803cd5afe842": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "d60e0b9c71784a7a87c1a782b5cfd224": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "4cd6b62bc70c42178e56a0c2e01cc76a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "f6c0447a883945b8973fe4c6063e5934": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "4ca9979c31394412af517eee639ddcfb": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "00b78a0c7ab94b04a80dab81f08c4e88": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_59c1006b93824b88b042782591913eea",
              "IPY_MODEL_2202dc3cf22649dc910862d37ec28cb6",
              "IPY_MODEL_8087c2f73cc64ac5a66fb319263b4e10"
            ],
            "layout": "IPY_MODEL_5f0901624f4041b1994b49a3ab1be82f"
          }
        },
        "59c1006b93824b88b042782591913eea": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_2deee584b165405281ae402675f22f85",
            "placeholder": "​",
            "style": "IPY_MODEL_6c2c786c8fbe4256baff7d98c1bab07f",
            "value": "Downloading data: 100%"
          }
        },
        "2202dc3cf22649dc910862d37ec28cb6": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_433fc96f3609478c8c6aa48873d741a4",
            "max": 117519820,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_d17d116b5ba34a468a35d80216384ca6",
            "value": 117519820
          }
        },
        "8087c2f73cc64ac5a66fb319263b4e10": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_f663b50655314ea4bbf8884250fbd9fe",
            "placeholder": "​",
            "style": "IPY_MODEL_8e036dd9a4494a538bdcbe83ee3c3d06",
            "value": " 118M/118M [00:02&lt;00:00, 43.9MB/s]"
          }
        },
        "5f0901624f4041b1994b49a3ab1be82f": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "2deee584b165405281ae402675f22f85": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "6c2c786c8fbe4256baff7d98c1bab07f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "433fc96f3609478c8c6aa48873d741a4": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "d17d116b5ba34a468a35d80216384ca6": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "f663b50655314ea4bbf8884250fbd9fe": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "8e036dd9a4494a538bdcbe83ee3c3d06": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "1d182909914842e2a8e06c0e7ca436df": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_e4c3d15c5cd94e3c9489de48702e6848",
              "IPY_MODEL_673a81ac011846f7a855d46716a56fd5",
              "IPY_MODEL_8db3fb7178b6409eb75c271bc888081e"
            ],
            "layout": "IPY_MODEL_5cf7dbd4edfa4c5a83ed0521d9fc5eb1"
          }
        },
        "e4c3d15c5cd94e3c9489de48702e6848": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_a5ba108402e941dbb64dd1099bc8c1c3",
            "placeholder": "​",
            "style": "IPY_MODEL_594aa85647454693a058ab738b6c6268",
            "value": "Generating train split: 100%"
          }
        },
        "673a81ac011846f7a855d46716a56fd5": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_1a53cae95c3e44bcb8c39e3c47552740",
            "max": 38969,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_326510b412224d4d979a6009fab2e683",
            "value": 38969
          }
        },
        "8db3fb7178b6409eb75c271bc888081e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_53f26580a4d34b08952a02bdd8154f8d",
            "placeholder": "​",
            "style": "IPY_MODEL_546683616f114b05992da9a9cc95a55f",
            "value": " 38969/38969 [00:42&lt;00:00, 1188.03 examples/s]"
          }
        },
        "5cf7dbd4edfa4c5a83ed0521d9fc5eb1": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "a5ba108402e941dbb64dd1099bc8c1c3": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "594aa85647454693a058ab738b6c6268": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "1a53cae95c3e44bcb8c39e3c47552740": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "326510b412224d4d979a6009fab2e683": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "53f26580a4d34b08952a02bdd8154f8d": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "546683616f114b05992da9a9cc95a55f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "01aa81ca076840569cb688c74e62dfbe": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_1d905f36c22447d59b0c88ffdb51ab78",
              "IPY_MODEL_29bd043d91574b45833d2cc04da61e33",
              "IPY_MODEL_fb580cce9f60417c8e3d4230e85eb3ac"
            ],
            "layout": "IPY_MODEL_71ab27c69c474a1ab522b667729e3863"
          }
        },
        "1d905f36c22447d59b0c88ffdb51ab78": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_93bb467242184554a304b1131ea67000",
            "placeholder": "​",
            "style": "IPY_MODEL_6cf8cf4a93634a539b6c9fada0fcad93",
            "value": "Generating test split: 100%"
          }
        },
        "29bd043d91574b45833d2cc04da61e33": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_d726f4bae06f43aebe6b504bcd6df903",
            "max": 7500,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_adaef6d1f85f4f6c9b587ba722912a80",
            "value": 7500
          }
        },
        "fb580cce9f60417c8e3d4230e85eb3ac": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_5e8d9b92347d463ab728f9754fc17d91",
            "placeholder": "​",
            "style": "IPY_MODEL_71a64f0a447744f482765d3ceee8080f",
            "value": " 7500/7500 [00:04&lt;00:00, 1290.37 examples/s]"
          }
        },
        "71ab27c69c474a1ab522b667729e3863": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "93bb467242184554a304b1131ea67000": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "6cf8cf4a93634a539b6c9fada0fcad93": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "d726f4bae06f43aebe6b504bcd6df903": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "adaef6d1f85f4f6c9b587ba722912a80": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "5e8d9b92347d463ab728f9754fc17d91": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "71a64f0a447744f482765d3ceee8080f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "3fa98e32b6a74df5bb5551004538416b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_ee7e3b7262ed4e2d8031f0bc8d74b709",
              "IPY_MODEL_e81d78fdea704c33b5f34bcfc112367f",
              "IPY_MODEL_365fd3166ad6467d80b7948c2ededee0"
            ],
            "layout": "IPY_MODEL_6dff9c660a774ff98660baa642a0f1b2"
          }
        },
        "ee7e3b7262ed4e2d8031f0bc8d74b709": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_dc8c1d9b67f043d1a7fd180bb6d2346a",
            "placeholder": "​",
            "style": "IPY_MODEL_4860b9dd54094d6bab65bc3757a392f1",
            "value": "Generating validation split: 100%"
          }
        },
        "e81d78fdea704c33b5f34bcfc112367f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_a2d8e873e7d049fdbc0da753eb03df0e",
            "max": 6327,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_b934f02abbe0414db742cac592379be6",
            "value": 6327
          }
        },
        "365fd3166ad6467d80b7948c2ededee0": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_9c5daa52339346bc95e8363504e0494f",
            "placeholder": "​",
            "style": "IPY_MODEL_e888753994de40c4b05806301b72e9bd",
            "value": " 6327/6327 [00:06&lt;00:00, 902.17 examples/s]"
          }
        },
        "6dff9c660a774ff98660baa642a0f1b2": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "dc8c1d9b67f043d1a7fd180bb6d2346a": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "4860b9dd54094d6bab65bc3757a392f1": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "a2d8e873e7d049fdbc0da753eb03df0e": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "b934f02abbe0414db742cac592379be6": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "9c5daa52339346bc95e8363504e0494f": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "e888753994de40c4b05806301b72e9bd": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        }
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}