notebooks/6_llm_finetuning.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tpOyl9AW9hO9"
      },
      "source": [
        "# **Finetuning a large language model**\n",
        "\n",
        "[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sekhansen/text_algorithms_econ/blob/main/notebooks/6_llm_finetuning.ipynb)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PS7QjJCs3Pfo"
      },
      "source": [
        "# 0. Setup"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "yFyax9i13Pfr"
      },
      "outputs": [],
      "source": [
        "# Instructions for Colab: \n",
        "# 1. First CHANGE RUNTIME TYPE to GPU \n",
        "# 2. Run install commands\n",
        "# 3. You might need to RESTART RUNTIME\n",
        "# 4. Run the rest of the cells below"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gc7f-6ml3Pfv",
        "outputId": "b8459b76-23ba-4bcd-9e74-0fa20f447b2e"
      },
      "outputs": [],
      "source": [
        "%%capture\n",
        "\n",
        "# install required libraries\n",
        "!pip3 install transformers                  # HuggingFace library for interacting with BERT (and multiple other models)\n",
        "!pip3 install datasets                      # HuggingFace library to process dataframes\n",
        "!pip3 install sentence-transformers         # library to use Sentence Similarity BERT\n",
        "!pip3 install bertviz                       # visualize BERT's attention weigths\n",
        "!pip3 install annoy                         # Spotify's library for finding nearest neighbours                     \n",
        "!pip3 install ipywidgets"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "VZZjP70q3Pfx"
      },
      "outputs": [],
      "source": [
        "# (COLAB) you might need to restart RUNTIME after installing packages!"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TljxViEZ3Pfz",
        "outputId": "82d5242f-55e0-4287-8f4f-3c2a6eff6d5a"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "GPU: True\n"
          ]
        }
      ],
      "source": [
        "# import libraries\n",
        "import gdown\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import gdown\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt\n",
        "import torch\n",
        "\n",
        "from transformers import AutoModel, BertModel, AutoTokenizer, AutoModelForSequenceClassification, pipeline, TrainingArguments, Trainer, utils\n",
        "from transformers.pipelines.base import KeyDataset\n",
        "from datasets import load_dataset, load_metric, Dataset, DatasetDict\n",
        "\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.model_selection import cross_val_score, train_test_split\n",
        "from sklearn.metrics import f1_score, precision_score, recall_score, accuracy_score\n",
        "\n",
        "from google.colab import output\n",
        "output.enable_custom_widget_manager()\n",
        "\n",
        "# test GPU\n",
        "print(f\"GPU: {torch.cuda.is_available()}\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "id": "5pV-LpDu3Pf3"
      },
      "outputs": [],
      "source": [
        "# define dictionary with paths to data in Google Drive\n",
        "urls_dict = {\"10k_sent_2019_firms\":        (\"https://drive.google.com/uc?id=1eQB8rwSklyVD3u8sZImFII74b7jBeUIL\", \"parquet\"),\n",
        "            }"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TnLAzjq43Pf6",
        "outputId": "3bca381d-a6bc-4171-835c-65bd221584d3"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "Downloading...\n",
            "From: https://drive.google.com/uc?id=1eQB8rwSklyVD3u8sZImFII74b7jBeUIL\n",
            "To: /content/10k_sent_2019_firms.parquet\n",
            "100%|██████████| 720k/720k [00:00<00:00, 82.0MB/s]\n"
          ]
        }
      ],
      "source": [
        "# download all files\n",
        "for file_name, attributes in urls_dict.items():\n",
        "    url = attributes[0]\n",
        "    extension = attributes[1]\n",
        "    gdown.download(url, f\"./{file_name}.{extension}\", quiet=False) "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tdvAYAtu3Pf-"
      },
      "source": [
        "# 1. Load and prepare the data\n",
        "\n",
        "This tutorial uses text data from the **10-K reports** filed by publicly-traded firms in the U.S. in 2019. 10-K reports are a very rich source of data since firms include information regarding their organizational structure, financial performance and risk factors. We will use a version of the data where the risk factors section of each report has been splitted into sentences and each sentence has been assigned an ID that combines the firm identifier (i.e. **CIK**) and a sentence number. The raw data we use has a total of 1,744,131 sentences for 4,033 firms.\n",
        "\n",
        "More on the 10-K reports [here](https://www.investor.gov/introduction-investing/getting-started/researching-investments/how-read-10-k)."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "id": "hQ1Y68xb3PgB",
        "outputId": "5c3105a2-209e-4b42-8a6f-873698f546d3"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-abe73c4e-3681-4850-9e76-3f07c6b3aa70\">\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>sentences</th>\n",
              "      <th>cik</th>\n",
              "      <th>year</th>\n",
              "      <th>sent_no</th>\n",
              "      <th>sent_id</th>\n",
              "      <th>naics2</th>\n",
              "      <th>naics2_name</th>\n",
              "      <th>sentence_len</th>\n",
              "      <th>keep_sent</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>The following discussion sets forth the materi...</td>\n",
              "      <td>19617</td>\n",
              "      <td>2019</td>\n",
              "      <td>0</td>\n",
              "      <td>19617_0</td>\n",
              "      <td>52</td>\n",
              "      <td>Finance and Insurance</td>\n",
              "      <td>18</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Readers should not consider any descriptions o...</td>\n",
              "      <td>19617</td>\n",
              "      <td>2019</td>\n",
              "      <td>1</td>\n",
              "      <td>19617_1</td>\n",
              "      <td>52</td>\n",
              "      <td>Finance and Insurance</td>\n",
              "      <td>23</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Any of the risk factors discussed below could ...</td>\n",
              "      <td>19617</td>\n",
              "      <td>2019</td>\n",
              "      <td>2</td>\n",
              "      <td>19617_2</td>\n",
              "      <td>52</td>\n",
              "      <td>Finance and Insurance</td>\n",
              "      <td>53</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>JPMorgan Chase's businesses are highly regulat...</td>\n",
              "      <td>19617</td>\n",
              "      <td>2019</td>\n",
              "      <td>4</td>\n",
              "      <td>19617_4</td>\n",
              "      <td>52</td>\n",
              "      <td>Finance and Insurance</td>\n",
              "      <td>25</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>JPMorgan Chase is a financial services firm wi...</td>\n",
              "      <td>19617</td>\n",
              "      <td>2019</td>\n",
              "      <td>5</td>\n",
              "      <td>19617_5</td>\n",
              "      <td>52</td>\n",
              "      <td>Finance and Insurance</td>\n",
              "      <td>10</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6793</th>\n",
              "      <td>With the completion of the TFCF acquisition, o...</td>\n",
              "      <td>1744489</td>\n",
              "      <td>2019</td>\n",
              "      <td>194</td>\n",
              "      <td>1744489_194</td>\n",
              "      <td>51</td>\n",
              "      <td>Information</td>\n",
              "      <td>19</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6794</th>\n",
              "      <td>The increased indebtedness could have the effe...</td>\n",
              "      <td>1744489</td>\n",
              "      <td>2019</td>\n",
              "      <td>195</td>\n",
              "      <td>1744489_195</td>\n",
              "      <td>51</td>\n",
              "      <td>Information</td>\n",
              "      <td>22</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6795</th>\n",
              "      <td>The increased levels of indebtedness could als...</td>\n",
              "      <td>1744489</td>\n",
              "      <td>2019</td>\n",
              "      <td>196</td>\n",
              "      <td>1744489_196</td>\n",
              "      <td>51</td>\n",
              "      <td>Information</td>\n",
              "      <td>14</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6796</th>\n",
              "      <td>repurchases and dividends, and other activitie...</td>\n",
              "      <td>1744489</td>\n",
              "      <td>2019</td>\n",
              "      <td>197</td>\n",
              "      <td>1744489_197</td>\n",
              "      <td>51</td>\n",
              "      <td>Information</td>\n",
              "      <td>21</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6797</th>\n",
              "      <td>Our financial flexibility may be further const...</td>\n",
              "      <td>1744489</td>\n",
              "      <td>2019</td>\n",
              "      <td>198</td>\n",
              "      <td>1744489_198</td>\n",
              "      <td>51</td>\n",
              "      <td>Information</td>\n",
              "      <td>23</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>6798 rows × 9 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-abe73c4e-3681-4850-9e76-3f07c6b3aa70')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-abe73c4e-3681-4850-9e76-3f07c6b3aa70 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-abe73c4e-3681-4850-9e76-3f07c6b3aa70');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "                                              sentences      cik  year  \\\n",
              "0     The following discussion sets forth the materi...    19617  2019   \n",
              "1     Readers should not consider any descriptions o...    19617  2019   \n",
              "2     Any of the risk factors discussed below could ...    19617  2019   \n",
              "3     JPMorgan Chase's businesses are highly regulat...    19617  2019   \n",
              "4     JPMorgan Chase is a financial services firm wi...    19617  2019   \n",
              "...                                                 ...      ...   ...   \n",
              "6793  With the completion of the TFCF acquisition, o...  1744489  2019   \n",
              "6794  The increased indebtedness could have the effe...  1744489  2019   \n",
              "6795  The increased levels of indebtedness could als...  1744489  2019   \n",
              "6796  repurchases and dividends, and other activitie...  1744489  2019   \n",
              "6797  Our financial flexibility may be further const...  1744489  2019   \n",
              "\n",
              "      sent_no      sent_id  naics2            naics2_name  sentence_len  \\\n",
              "0           0      19617_0      52  Finance and Insurance            18   \n",
              "1           1      19617_1      52  Finance and Insurance            23   \n",
              "2           2      19617_2      52  Finance and Insurance            53   \n",
              "3           4      19617_4      52  Finance and Insurance            25   \n",
              "4           5      19617_5      52  Finance and Insurance            10   \n",
              "...       ...          ...     ...                    ...           ...   \n",
              "6793      194  1744489_194      51            Information            19   \n",
              "6794      195  1744489_195      51            Information            22   \n",
              "6795      196  1744489_196      51            Information            14   \n",
              "6796      197  1744489_197      51            Information            21   \n",
              "6797      198  1744489_198      51            Information            23   \n",
              "\n",
              "      keep_sent  \n",
              "0          True  \n",
              "1          True  \n",
              "2          True  \n",
              "3          True  \n",
              "4          True  \n",
              "...         ...  \n",
              "6793       True  \n",
              "6794       True  \n",
              "6795       True  \n",
              "6796       True  \n",
              "6797       True  \n",
              "\n",
              "[6798 rows x 9 columns]"
            ]
          },
          "execution_count": 7,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# read data\n",
        "df = pd.read_parquet(\"10k_sent_2019_firms.parquet\")\n",
        "df"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "w9F0a0ZQAyFy"
      },
      "source": [
        "# 2. Load model and tokenizer\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 180,
          "referenced_widgets": [
            "de668fb1cc5f439d946b6bd1f3a931a0",
            "4a86d561ef9b43b08675a602f974b5c7",
            "c954a9f610134ffdb06f715c6e182355",
            "d860afde15d64cb78aa8241864de8943",
            "e5b48e63274b48b3b4d4bf72a38de4e6",
            "242d12e1eb8d4ad58bdbfa5fc0594992",
            "b954033668c042439cbc0343d54f6077",
            "eaca405716e44ccaaac7ebb310961db9",
            "58e700dc905b4ba184ab3ddcfcf12ae1",
            "3a680ac3a0fd464aae169df14dad71ba",
            "edeb6c73a5c942f1b7d3fb3135975e3e",
            "2565844bb5954ede9486fc77b9a10d6e",
            "f60b0837e18047599db2d0e61c403ecf",
            "d01f03a55dd3496a884d93665538e94a",
            "68fe3becf7fe4e9b9d39e2c063419613",
            "9e93f72a338246a3a2b5ca412e3e7179",
            "7298a266a32d442c83cfb019eb5effcd",
            "5b96a22825374ff38fc608de5b1da421",
            "bb15b94101bf4c6387ca570c38fcc11c",
            "14b904a24b1f4062a5173071c154f6da",
            "160e3fc88ee046c39c3c808e389579c1",
            "c88aa53a291d4916b0eb6c60efd76831",
            "873bad729f734567ac5846208ce4d307",
            "721680ddd918482182a8b10320d6d26e",
            "9447c105d697407a80be0ceaf304c41c",
            "5bd8a4b2a6d94ae1b3df250d3ea32af2",
            "0556cb05494f4739929e7c5c4b0cd6d6",
            "874f25e654e1447abbd889832dce4b6d",
            "2f3fbefb77e84e2eab62c2688ab1844b",
            "506f0d79186b4229aa55fdb4d5a5d713",
            "edaaf3ac4ba74c56a5b8bfee73715a61",
            "8612628549974ba489b5925ef67a2a08",
            "7fba5f474ad94ad28039d1ac636cb9c2",
            "31127dc74e004eb594716fc22628c801",
            "7c456a0ba5614f0f88c245de2e1f37c7",
            "fea3e9e4b9c64164a20162542815273c",
            "38234c5f6b314d61a6dfe4e9adb06173",
            "bf4c986e9bad46b5901ab0e9ba17d6aa",
            "1b35f3444a4342168b422234f7af6cb4",
            "52abcb64ed404e43a3634ca496cb0615",
            "85b9ca31525e4a6bb6d3680193066c49",
            "4425d62c6ef6404db9248051c5c79b2a",
            "00e242f22fe34e4995e7887887db1bea",
            "2ac4cdad2cb54c599dd71d5fd242ceb1"
          ]
        },
        "id": "ymzx9vCSA0Wm",
        "outputId": "e929b01a-34ac-4279-9d5a-7e0314babbb0"
      },
      "outputs": [
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "de668fb1cc5f439d946b6bd1f3a931a0",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "Downloading (…)okenizer_config.json:   0%|          | 0.00/263 [00:00<?, ?B/s]"
            ]
          },
          "metadata": {
            "application/vnd.jupyter.widget-view+json": {
              "colab": {
                "custom_widget_manager": {
                  "url": "https://ssl.gstatic.com/colaboratory-static/widgets/colab-cdn-widget-manager/b3e629b1971e1542/manager.min.js"
                }
              }
            }
          },
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "c88aa53a291d4916b0eb6c60efd76831",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "Downloading (…)lve/main/config.json:   0%|          | 0.00/568 [00:00<?, ?B/s]"
            ]
          },
          "metadata": {
            "application/vnd.jupyter.widget-view+json": {
              "colab": {
                "custom_widget_manager": {
                  "url": "https://ssl.gstatic.com/colaboratory-static/widgets/colab-cdn-widget-manager/b3e629b1971e1542/manager.min.js"
                }
              }
            }
          },
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "7fba5f474ad94ad28039d1ac636cb9c2",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "Downloading (…)solve/main/vocab.txt:   0%|          | 0.00/221k [00:00<?, ?B/s]"
            ]
          },
          "metadata": {
            "application/vnd.jupyter.widget-view+json": {
              "colab": {
                "custom_widget_manager": {
                  "url": "https://ssl.gstatic.com/colaboratory-static/widgets/colab-cdn-widget-manager/b3e629b1971e1542/manager.min.js"
                }
              }
            }
          },
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "2ac4cdad2cb54c599dd71d5fd242ceb1",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "Downloading (…)cial_tokens_map.json:   0%|          | 0.00/112 [00:00<?, ?B/s]"
            ]
          },
          "metadata": {
            "application/vnd.jupyter.widget-view+json": {
              "colab": {
                "custom_widget_manager": {
                  "url": "https://ssl.gstatic.com/colaboratory-static/widgets/colab-cdn-widget-manager/b3e629b1971e1542/manager.min.js"
                }
              }
            }
          },
          "output_type": "display_data"
        },
        {
          "data": {
            "text/plain": [
              "BertTokenizerFast(name_or_path='nlpaueb/sec-bert-base', vocab_size=30000, model_max_length=512, is_fast=True, padding_side='right', truncation_side='right', special_tokens={'unk_token': '[UNK]', 'sep_token': '[SEP]', 'pad_token': '[PAD]', 'cls_token': '[CLS]', 'mask_token': '[MASK]'}, clean_up_tokenization_spaces=True)"
            ]
          },
          "execution_count": 8,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# load a tokenizer using the name of the model we want to use\n",
        "sec_tokenizer = AutoTokenizer.from_pretrained(\"nlpaueb/sec-bert-base\")\n",
        "\n",
        "# inspect the configuration of the tokenizer\n",
        "sec_tokenizer"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 121,
          "referenced_widgets": [
            "b53f89fce60d4dbc9048742bafc4eb7d",
            "358f974f1a654790a9f7f23b0d207573",
            "6b155526877740c28ed1d29fd5b3a831",
            "09a6ac632a2242e9b1964e012055f9ca",
            "578bc08e2a1845e699ab42b7262cbb5f",
            "9ba7061f647a439c984b60561c76843d",
            "9757ceb6a0814b7f88d49079fa4fdd87",
            "0319768795434433b34e7bed353f539f",
            "ac3ea102011e4037a464dac24f598f2d",
            "e40f78a2a6254fecab2083545144d8a7",
            "158e5eed736c42f2901816bf3475bcb5"
          ]
        },
        "id": "a1VslaZoA509",
        "outputId": "f40381d4-e8b5-49d4-ef80-fb5f364a3671"
      },
      "outputs": [
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "b53f89fce60d4dbc9048742bafc4eb7d",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "Downloading pytorch_model.bin:   0%|          | 0.00/439M [00:00<?, ?B/s]"
            ]
          },
          "metadata": {
            "application/vnd.jupyter.widget-view+json": {
              "colab": {
                "custom_widget_manager": {
                  "url": "https://ssl.gstatic.com/colaboratory-static/widgets/colab-cdn-widget-manager/b3e629b1971e1542/manager.min.js"
                }
              }
            }
          },
          "output_type": "display_data"
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "Some weights of the model checkpoint at nlpaueb/sec-bert-base were not used when initializing BertModel: ['cls.predictions.decoder.weight', 'cls.predictions.transform.LayerNorm.bias', 'cls.predictions.decoder.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.transform.dense.weight', 'cls.predictions.bias', 'cls.seq_relationship.weight', 'cls.predictions.transform.dense.bias']\n",
            "- This IS expected if you are initializing BertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
            "- This IS NOT expected if you are initializing BertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n"
          ]
        }
      ],
      "source": [
        "# load the model trained on 10-K report using its name\n",
        "sec_model = AutoModel.from_pretrained(\"nlpaueb/sec-bert-base\")\n",
        "\n",
        "# put model in evaluation model (we will not do any training)\n",
        "sec_model = sec_model.eval()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0jKbLBT58L9n"
      },
      "source": [
        "# 3. Finetuning a model\n",
        "\n",
        "We will show how we can finetune a BERT model for any classification task we want. Finetuning involves training (with a very small learning rate) all of the parameters of the model for a particular task. This is one of the most powerful ways of using modern language models."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SuiFu1SG8L9u"
      },
      "source": [
        "![image.png](data:image/png;base64,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)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ikmYxue88L97"
      },
      "source": [
        "*Source: Alammar (2018)*"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "id": "TjmwAPLZAbk4"
      },
      "outputs": [],
      "source": [
        "# get only the NAICS2 code for each sentence to use as the labels for our regression\n",
        "labels = df[[\"naics2\"]]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pW1B7OjA3Phh",
        "outputId": "b36596c9-0851-4a71-ac22-05fa74678d5a"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "6798"
            ]
          },
          "execution_count": 11,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# create list with all the indexes of available sentences\n",
        "sent_idxs = list(range(0, len(labels)))\n",
        "len(sent_idxs)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gNwso_hw3Phi",
        "outputId": "01c3d62d-1920-4a95-baa4-00f75001d91c"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            " Train sentences: 5438\n",
            " Test sentences: 1360\n"
          ]
        }
      ],
      "source": [
        "# perform a train/test split\n",
        "train_idxs, test_idxs = train_test_split(sent_idxs, test_size=0.2, random_state=92)\n",
        "print(f\" Train sentences: {len(train_idxs)}\\n\", f\"Test sentences: {len(test_idxs)}\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "id": "tyvIc0QK8L9-",
        "outputId": "3a1df9db-de8f-43c6-e91b-d35b5f4e9f97"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-04fa306c-9a7f-4b1c-9c6a-21824eac283f\">\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>sentences</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>3639</th>\n",
              "      <td>We experienced a work stoppage in 2008 when a ...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2680</th>\n",
              "      <td>Finally, holders of the Tesla Convertible Note...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1507</th>\n",
              "      <td>There can be significant differences between o...</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>911</th>\n",
              "      <td>We also rely on other companies to maintain re...</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>621</th>\n",
              "      <td>The techniques used for attacks by third parti...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5007</th>\n",
              "      <td>Our revenues and cash requirements are affecte...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>710</th>\n",
              "      <td>As is common in our industry, our advertisers ...</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6162</th>\n",
              "      <td>Global markets for the Company's products and ...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4138</th>\n",
              "      <td>Longer payment cycles in some countries, incre...</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4218</th>\n",
              "      <td>The concentration of our stock ownership limit...</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5438 rows × 2 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-04fa306c-9a7f-4b1c-9c6a-21824eac283f')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-04fa306c-9a7f-4b1c-9c6a-21824eac283f 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-04fa306c-9a7f-4b1c-9c6a-21824eac283f');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "                                              sentences  label\n",
              "3639  We experienced a work stoppage in 2008 when a ...      0\n",
              "2680  Finally, holders of the Tesla Convertible Note...      0\n",
              "1507  There can be significant differences between o...      3\n",
              "911   We also rely on other companies to maintain re...      2\n",
              "621   The techniques used for attacks by third parti...      0\n",
              "...                                                 ...    ...\n",
              "5007  Our revenues and cash requirements are affecte...      1\n",
              "710   As is common in our industry, our advertisers ...      2\n",
              "6162  Global markets for the Company's products and ...      0\n",
              "4138  Longer payment cycles in some countries, incre...      2\n",
              "4218  The concentration of our stock ownership limit...      2\n",
              "\n",
              "[5438 rows x 2 columns]"
            ]
          },
          "execution_count": 13,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# format the train data adequately\n",
        "df_finetune = df.loc[train_idxs].copy()\n",
        "\n",
        "df_finetune = df_finetune[[\"sentences\", \"naics2\"]]\n",
        "df_finetune.columns = [\"sentences\", \"label\"]\n",
        "\n",
        "# transform labels into integers\n",
        "df_finetune[\"label\"] = df_finetune[\"label\"].astype(int)\n",
        "df_finetune\n",
        "\n",
        "# map labels from original sector code to ints from 0 to num_sectors\n",
        "num_sectors = len(df_finetune.groupby('label').size())\n",
        "label2id_label = {k:v for k,v in zip(df_finetune.groupby('label').size().index.values, range(0, num_sectors))}\n",
        "df_finetune[\"label\"] = df_finetune[\"label\"].apply(lambda x: label2id_label[x])\n",
        "df_finetune"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "id": "L-XywoGw8L-A",
        "outputId": "cacad728-124a-4460-a69c-7e68b1760bfa"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-febb72bf-430c-4f76-9d92-81873bcd7747\">\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>sentences</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1236</th>\n",
              "      <td>•  integration of the acquired company's accou...</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>474</th>\n",
              "      <td>Competition for qualified personnel within the...</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3418</th>\n",
              "      <td>Any reduction in our and our subsidiaries' cre...</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6564</th>\n",
              "      <td>Natural disasters or other catastrophes could ...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2646</th>\n",
              "      <td>For the battery and drive unit on our current ...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1950</th>\n",
              "      <td>Regulatory requirements in the U.S. and in non...</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6648</th>\n",
              "      <td>If personal information of our customers or em...</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6529</th>\n",
              "      <td>The evolution of retailing in online and mobil...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6452</th>\n",
              "      <td>Our success depends, in part, on our ability t...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1040</th>\n",
              "      <td>•  recruiting and retaining talented and capab...</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>1360 rows × 2 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-febb72bf-430c-4f76-9d92-81873bcd7747')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-febb72bf-430c-4f76-9d92-81873bcd7747 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-febb72bf-430c-4f76-9d92-81873bcd7747');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "                                              sentences  label\n",
              "1236  •  integration of the acquired company's accou...      2\n",
              "474   Competition for qualified personnel within the...      3\n",
              "3418  Any reduction in our and our subsidiaries' cre...      3\n",
              "6564  Natural disasters or other catastrophes could ...      1\n",
              "2646  For the battery and drive unit on our current ...      0\n",
              "...                                                 ...    ...\n",
              "1950  Regulatory requirements in the U.S. and in non...      3\n",
              "6648  If personal information of our customers or em...      2\n",
              "6529  The evolution of retailing in online and mobil...      1\n",
              "6452  Our success depends, in part, on our ability t...      1\n",
              "1040  •  recruiting and retaining talented and capab...      2\n",
              "\n",
              "[1360 rows x 2 columns]"
            ]
          },
          "execution_count": 14,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# format the test data adequately\n",
        "df_test = df.loc[test_idxs].copy()\n",
        "\n",
        "df_test = df_test[[\"sentences\", \"naics2\"]]\n",
        "df_test.columns = [\"sentences\", \"label\"]\n",
        "\n",
        "# transform labels into integers\n",
        "df_test[\"label\"] = df_test[\"label\"].astype(int)\n",
        "df_test\n",
        "\n",
        "# map labels from original sector code to ints from 0 to num_sectors\n",
        "df_test[\"label\"] = df_test[\"label\"].apply(lambda x: label2id_label[x])\n",
        "df_test"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "id": "SEZAK1uZ8L-D"
      },
      "outputs": [],
      "source": [
        "# transform data into Dataset class\n",
        "finetune_dataset = Dataset.from_pandas(df_finetune)\n",
        "test_dataset = Dataset.from_pandas(df_test)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 86,
          "referenced_widgets": [
            "713f1759ddda435fadbbf4cb4c1bc3f8",
            "3781b22edb2c44e3ab46724a593e64d3",
            "3dd6681a688a4d32b3f3ff031df5c70c",
            "28f52e9ca172414eb8e85cdcf84a0d66",
            "ebefda73df4443078d53c6328b49f8b8",
            "2a1a7f2314d848a8bd502e655f5394c9",
            "8286cd7cc74c4b49948a38f116860fb2",
            "9e86ec1e1ddc4e3c887b24c6ba8cb2ea",
            "2c3e6de690ad4611a99467a037f0b5f1",
            "f61ce8d981984eda93f91e6f82b242f1",
            "2c9fcae81a134ccab4fffd04b5178e77",
            "f6c9190f977245ab859e1a677c0f2af6",
            "142df8ca09c64c998eae98b9438b8c4c",
            "66fcec54203a4aea8cb588dfa4c775f7",
            "f55f19231630409298a7bf6a350523c0",
            "931189a93cda49cfa4e46e0db48f3371",
            "6a973d02543c4b91a94b85c57e8a8aec",
            "ff5c37f9068c46cf866812330d4abf6f",
            "c0f24bbfc27241da8ec14e280868b3e0",
            "a231b2d8cece4b899891de7af3d89320",
            "95d1236a3a8b4c3d9193c5ce8356a1af",
            "2504621b0890488ca27145eeb4cb60bd"
          ]
        },
        "id": "bA5CAANV8L-G",
        "outputId": "bff7ae52-da8f-4dd5-95b4-b50e6e2e0145"
      },
      "outputs": [
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "713f1759ddda435fadbbf4cb4c1bc3f8",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "Map:   0%|          | 0/5438 [00:00<?, ? examples/s]"
            ]
          },
          "metadata": {
            "application/vnd.jupyter.widget-view+json": {
              "colab": {
                "custom_widget_manager": {
                  "url": "https://ssl.gstatic.com/colaboratory-static/widgets/colab-cdn-widget-manager/b3e629b1971e1542/manager.min.js"
                }
              }
            }
          },
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "2504621b0890488ca27145eeb4cb60bd",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "Map:   0%|          | 0/1360 [00:00<?, ? examples/s]"
            ]
          },
          "metadata": {
            "application/vnd.jupyter.widget-view+json": {
              "colab": {
                "custom_widget_manager": {
                  "url": "https://ssl.gstatic.com/colaboratory-static/widgets/colab-cdn-widget-manager/b3e629b1971e1542/manager.min.js"
                }
              }
            }
          },
          "output_type": "display_data"
        },
        {
          "data": {
            "text/plain": [
              "Dataset({\n",
              "    features: ['sentences', 'label', '__index_level_0__', 'input_ids', 'token_type_ids', 'attention_mask'],\n",
              "    num_rows: 5438\n",
              "})"
            ]
          },
          "execution_count": 16,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# tokenize the dataset\n",
        "def tokenize_function(examples):\n",
        "    return sec_tokenizer(examples[\"sentences\"], max_length=60, padding=\"max_length\", truncation=True)\n",
        "\n",
        "tokenized_ft = finetune_dataset.map(tokenize_function, batched=True)    # batched=True is key for training\n",
        "tokenized_test = test_dataset.map(tokenize_function, batched=True)\n",
        "\n",
        "tokenized_ft"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 25,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tYTgaplP8L-J",
        "outputId": "d6844c57-40a1-402d-d972-e203bd371c3b"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "Some weights of the model checkpoint at nlpaueb/sec-bert-base were not used when initializing BertForSequenceClassification: ['cls.predictions.decoder.weight', 'cls.predictions.transform.LayerNorm.bias', 'cls.predictions.decoder.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.transform.dense.weight', 'cls.predictions.bias', 'cls.seq_relationship.weight', 'cls.predictions.transform.dense.bias']\n",
            "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
            "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
            "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at nlpaueb/sec-bert-base and are newly initialized: ['classifier.bias', 'classifier.weight']\n",
            "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n"
          ]
        }
      ],
      "source": [
        "# load the model for finetunning. \n",
        "# NOTE that we use a different class from the transformers library:\n",
        "# AutoModel vs. AutoModelForSequenceClassification\n",
        "num_labels = len(df_finetune.groupby('label').size())\n",
        "model_ft = AutoModelForSequenceClassification.from_pretrained(\"nlpaueb/sec-bert-base\", \n",
        "                                                              num_labels=num_labels,\n",
        "                                                              output_hidden_states=False)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "njc2MjRY8L-P"
      },
      "source": [
        "You can find a complete description of all the available parameters of the ```TrainingArguments()``` class [here](https://huggingface.co/docs/transformers/v4.21.2/en/main_classes/trainer#transformers.TrainingArguments)."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 26,
      "metadata": {
        "id": "R6C0EEaA8L-S"
      },
      "outputs": [],
      "source": [
        "# define the main arguments for training\n",
        "training_args = TrainingArguments(\"./\",                               # path to save model\n",
        "                                  learning_rate=3e-5,                 # we use a very small learning rate\n",
        "                                  num_train_epochs=5,                 # number of iterations through the corpus\n",
        "                                  per_device_train_batch_size=8,\n",
        "                                  per_device_eval_batch_size=1,\n",
        "                                  evaluation_strategy=\"no\",\n",
        "                                  save_strategy=\"no\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 27,
      "metadata": {
        "id": "N6eRzu0E8L-V"
      },
      "outputs": [],
      "source": [
        "# define the set of metrics to be computed through the training process\n",
        "def compute_metrics(eval_pred):\n",
        "    metric1 = load_metric(\"precision\")\n",
        "    metric2 = load_metric(\"recall\")\n",
        "    metric3 = load_metric(\"f1\")\n",
        "    metric4 = load_metric(\"accuracy\")\n",
        "    \n",
        "    logits, labels = eval_pred\n",
        "    predictions = np.argmax(logits, axis=-1)\n",
        "    \n",
        "    precision = metric1.compute(predictions=predictions, references=labels, average=\"micro\")[\"precision\"]\n",
        "    recall = metric2.compute(predictions=predictions, references=labels, average=\"micro\")[\"recall\"]\n",
        "    f1 = metric3.compute(predictions=predictions, references=labels, average=\"micro\")[\"f1\"]\n",
        "    accuracy = metric4.compute(predictions=predictions, references=labels)[\"accuracy\"]\n",
        "    \n",
        "    return {\"precision\": precision, \"recall\": recall, \n",
        "            \"f1\": f1, \"accuracy\": accuracy}\n",
        "\n",
        "# by default the Trainer will use MSEloss from (torch.nn) for regression and\n",
        "# CrossEntropy loss for classification\n",
        "trainer = Trainer(\n",
        "    model=model_ft,\n",
        "    args=training_args,\n",
        "    train_dataset=tokenized_ft,\n",
        "    eval_dataset=tokenized_ft,  # in-sample evaluation\n",
        "    compute_metrics=compute_metrics\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 28,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 353
        },
        "id": "2-Of70Ji8L-W",
        "outputId": "6bcc559d-035d-453b-c905-4bc778e5d987"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.9/dist-packages/transformers/optimization.py:391: FutureWarning: This implementation of AdamW is deprecated and will be removed in a future version. Use the PyTorch implementation torch.optim.AdamW instead, or set `no_deprecation_warning=True` to disable this warning\n",
            "  warnings.warn(\n"
          ]
        },
        {
          "data": {
            "text/html": [
              "\n",
              "    <div>\n",
              "      \n",
              "      <progress value='3400' max='3400' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
              "      [3400/3400 06:20, Epoch 5/5]\n",
              "    </div>\n",
              "    <table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              " <tr style=\"text-align: left;\">\n",
              "      <th>Step</th>\n",
              "      <th>Training Loss</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <td>500</td>\n",
              "      <td>0.831300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1000</td>\n",
              "      <td>0.524400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1500</td>\n",
              "      <td>0.376100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2000</td>\n",
              "      <td>0.246800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2500</td>\n",
              "      <td>0.133800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3000</td>\n",
              "      <td>0.077700</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table><p>"
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "text/plain": [
              "TrainOutput(global_step=3400, training_loss=0.32789999905754535, metrics={'train_runtime': 380.6943, 'train_samples_per_second': 71.422, 'train_steps_per_second': 8.931, 'total_flos': 838373210251200.0, 'train_loss': 0.32789999905754535, 'epoch': 5.0})"
            ]
          },
          "execution_count": 28,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# train model (should take around 5 minutes with GPU)\n",
        "trainer.train()\n",
        "\n",
        "# save final version of the model\n",
        "#trainer.save_model(\"./models/\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 29,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        },
        "id": "KOhHBi4J8L-Y",
        "outputId": "c787503c-fd0f-497a-8aed-ea234a0aa85c"
      },
      "outputs": [
        {
          "data": {
            "text/html": [],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "{'test_loss': 1.2546815872192383, 'test_precision': 0.7941176470588235, 'test_recall': 0.7941176470588235, 'test_f1': 0.7941176470588235, 'test_accuracy': 0.7941176470588235, 'test_runtime': 18.7918, 'test_samples_per_second': 72.372, 'test_steps_per_second': 72.372}\n"
          ]
        }
      ],
      "source": [
        "# evaluate final model on the test dataset\n",
        "results = trainer.predict(tokenized_test)\n",
        "final_metrics = results[2]\n",
        "print(final_metrics)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 30,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        },
        "id": "pdpx7gzV8L-Z",
        "outputId": "5759c8d0-5f26-4466-b9a2-5e5666c18b59"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-47a600af-c6b2-42b4-9347-01b0bd692d9a\">\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>accuracy</th>\n",
              "      <th>model</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.616176</td>\n",
              "      <td>word_embeddings</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.627206</td>\n",
              "      <td>BERT</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.708088</td>\n",
              "      <td>SEC BERT</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.794118</td>\n",
              "      <td>Finetuned BERT</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-47a600af-c6b2-42b4-9347-01b0bd692d9a')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-47a600af-c6b2-42b4-9347-01b0bd692d9a 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-47a600af-c6b2-42b4-9347-01b0bd692d9a');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "   accuracy            model\n",
              "0  0.616176  word_embeddings\n",
              "1  0.627206             BERT\n",
              "2  0.708088         SEC BERT\n",
              "3  0.794118   Finetuned BERT"
            ]
          },
          "execution_count": 30,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# compare finetuning to other methods\n",
        "df_compare = pd.DataFrame({\"accuracy\": [0.616176, 0.627206, 0.708088, final_metrics[\"test_accuracy\"]],\n",
        "                           \"model\": [\"word_embeddings\", \"BERT\", \"SEC BERT\", \"Finetuned BERT\"]})\n",
        "df_compare"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 31,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 720
        },
        "id": "0QLUSIf28L-c",
        "outputId": "2edba130-915b-440d-e929-b781409cf7e6"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1200x800 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.figure(figsize=(12,8))\n",
        "sns.barplot(data=df_compare, x=\"model\", y=\"accuracy\")\n",
        "plt.title(\"Accuracy in test set across methods\", fontsize=16)\n",
        "plt.show()"
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "provenance": []
    },
    "gpuClass": "standard",
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.10.2"
    },
    "orig_nbformat": 4,
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "00e242f22fe34e4995e7887887db1bea": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "0319768795434433b34e7bed353f539f": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "0556cb05494f4739929e7c5c4b0cd6d6": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "09a6ac632a2242e9b1964e012055f9ca": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "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_e40f78a2a6254fecab2083545144d8a7",
            "placeholder": "​",
            "style": "IPY_MODEL_158e5eed736c42f2901816bf3475bcb5",
            "value": " 439M/439M [00:04&lt;00:00, 103MB/s]"
          }
        },
        "142df8ca09c64c998eae98b9438b8c4c": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "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": ""
          }
        },
        "14b904a24b1f4062a5173071c154f6da": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "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_9e93f72a338246a3a2b5ca412e3e7179",
            "placeholder": "​",
            "style": "IPY_MODEL_7298a266a32d442c83cfb019eb5effcd",
            "value": " 568/568 [00:00&lt;00:00, 24.1kB/s]"
          }
        },
        "158e5eed736c42f2901816bf3475bcb5": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "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": ""
          }
        },
        "160e3fc88ee046c39c3c808e389579c1": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "1b35f3444a4342168b422234f7af6cb4": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "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": ""
          }
        },
        "242d12e1eb8d4ad58bdbfa5fc0594992": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "2504621b0890488ca27145eeb4cb60bd": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "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_ff5c37f9068c46cf866812330d4abf6f",
              "IPY_MODEL_c0f24bbfc27241da8ec14e280868b3e0",
              "IPY_MODEL_a231b2d8cece4b899891de7af3d89320"
            ],
            "layout": "IPY_MODEL_95d1236a3a8b4c3d9193c5ce8356a1af"
          }
        },
        "2565844bb5954ede9486fc77b9a10d6e": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "28f52e9ca172414eb8e85cdcf84a0d66": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "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_f61ce8d981984eda93f91e6f82b242f1",
            "placeholder": "​",
            "style": "IPY_MODEL_2c9fcae81a134ccab4fffd04b5178e77",
            "value": " 5000/5438 [00:00&lt;00:00, 6999.56 examples/s]"
          }
        },
        "2a1a7f2314d848a8bd502e655f5394c9": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "2ac4cdad2cb54c599dd71d5fd242ceb1": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "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_52abcb64ed404e43a3634ca496cb0615",
              "IPY_MODEL_85b9ca31525e4a6bb6d3680193066c49",
              "IPY_MODEL_4425d62c6ef6404db9248051c5c79b2a"
            ],
            "layout": "IPY_MODEL_00e242f22fe34e4995e7887887db1bea"
          }
        },
        "2c3e6de690ad4611a99467a037f0b5f1": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "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": ""
          }
        },
        "2c9fcae81a134ccab4fffd04b5178e77": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "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": ""
          }
        },
        "2f3fbefb77e84e2eab62c2688ab1844b": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "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_9447c105d697407a80be0ceaf304c41c",
            "placeholder": "​",
            "style": "IPY_MODEL_5bd8a4b2a6d94ae1b3df250d3ea32af2",
            "value": "Downloading (…)solve/main/vocab.txt: 100%"
          }
        },
        "31127dc74e004eb594716fc22628c801": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "358f974f1a654790a9f7f23b0d207573": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "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_9ba7061f647a439c984b60561c76843d",
            "placeholder": "​",
            "style": "IPY_MODEL_9757ceb6a0814b7f88d49079fa4fdd87",
            "value": "Downloading pytorch_model.bin: 100%"
          }
        },
        "3781b22edb2c44e3ab46724a593e64d3": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "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_2a1a7f2314d848a8bd502e655f5394c9",
            "placeholder": "​",
            "style": "IPY_MODEL_8286cd7cc74c4b49948a38f116860fb2",
            "value": "Map:  92%"
          }
        },
        "38234c5f6b314d61a6dfe4e9adb06173": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "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": ""
          }
        },
        "3a680ac3a0fd464aae169df14dad71ba": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "3dd6681a688a4d32b3f3ff031df5c70c": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "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": "",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_9e86ec1e1ddc4e3c887b24c6ba8cb2ea",
            "max": 5438,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_2c3e6de690ad4611a99467a037f0b5f1",
            "value": 5438
          }
        },
        "4425d62c6ef6404db9248051c5c79b2a": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "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_bf4c986e9bad46b5901ab0e9ba17d6aa",
            "placeholder": "​",
            "style": "IPY_MODEL_1b35f3444a4342168b422234f7af6cb4",
            "value": " 112/112 [00:00&lt;00:00, 7.57kB/s]"
          }
        },
        "4a86d561ef9b43b08675a602f974b5c7": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "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_242d12e1eb8d4ad58bdbfa5fc0594992",
            "placeholder": "​",
            "style": "IPY_MODEL_b954033668c042439cbc0343d54f6077",
            "value": "Downloading (…)okenizer_config.json: 100%"
          }
        },
        "506f0d79186b4229aa55fdb4d5a5d713": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "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_873bad729f734567ac5846208ce4d307",
            "max": 221438,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_721680ddd918482182a8b10320d6d26e",
            "value": 221438
          }
        },
        "52abcb64ed404e43a3634ca496cb0615": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "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_fea3e9e4b9c64164a20162542815273c",
            "placeholder": "​",
            "style": "IPY_MODEL_38234c5f6b314d61a6dfe4e9adb06173",
            "value": "Downloading (…)cial_tokens_map.json: 100%"
          }
        },
        "578bc08e2a1845e699ab42b7262cbb5f": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "58e700dc905b4ba184ab3ddcfcf12ae1": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "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": ""
          }
        },
        "5b96a22825374ff38fc608de5b1da421": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "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_d01f03a55dd3496a884d93665538e94a",
            "placeholder": "​",
            "style": "IPY_MODEL_68fe3becf7fe4e9b9d39e2c063419613",
            "value": "Downloading (…)lve/main/config.json: 100%"
          }
        },
        "5bd8a4b2a6d94ae1b3df250d3ea32af2": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "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": ""
          }
        },
        "66fcec54203a4aea8cb588dfa4c775f7": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "68fe3becf7fe4e9b9d39e2c063419613": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "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": ""
          }
        },
        "6a973d02543c4b91a94b85c57e8a8aec": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "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": ""
          }
        },
        "6b155526877740c28ed1d29fd5b3a831": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "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_0319768795434433b34e7bed353f539f",
            "max": 438908790,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_ac3ea102011e4037a464dac24f598f2d",
            "value": 438908790
          }
        },
        "713f1759ddda435fadbbf4cb4c1bc3f8": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "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_3781b22edb2c44e3ab46724a593e64d3",
              "IPY_MODEL_3dd6681a688a4d32b3f3ff031df5c70c",
              "IPY_MODEL_28f52e9ca172414eb8e85cdcf84a0d66"
            ],
            "layout": "IPY_MODEL_ebefda73df4443078d53c6328b49f8b8"
          }
        },
        "721680ddd918482182a8b10320d6d26e": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "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": ""
          }
        },
        "7298a266a32d442c83cfb019eb5effcd": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "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": ""
          }
        },
        "7c456a0ba5614f0f88c245de2e1f37c7": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "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": ""
          }
        },
        "7fba5f474ad94ad28039d1ac636cb9c2": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "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_2f3fbefb77e84e2eab62c2688ab1844b",
              "IPY_MODEL_506f0d79186b4229aa55fdb4d5a5d713",
              "IPY_MODEL_edaaf3ac4ba74c56a5b8bfee73715a61"
            ],
            "layout": "IPY_MODEL_8612628549974ba489b5925ef67a2a08"
          }
        },
        "8286cd7cc74c4b49948a38f116860fb2": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "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": ""
          }
        },
        "85b9ca31525e4a6bb6d3680193066c49": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "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_31127dc74e004eb594716fc22628c801",
            "max": 112,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_7c456a0ba5614f0f88c245de2e1f37c7",
            "value": 112
          }
        },
        "8612628549974ba489b5925ef67a2a08": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "873bad729f734567ac5846208ce4d307": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "874f25e654e1447abbd889832dce4b6d": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "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": ""
          }
        },
        "931189a93cda49cfa4e46e0db48f3371": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "9447c105d697407a80be0ceaf304c41c": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "95d1236a3a8b4c3d9193c5ce8356a1af": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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": "hidden",
            "width": null
          }
        },
        "9757ceb6a0814b7f88d49079fa4fdd87": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "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": ""
          }
        },
        "9ba7061f647a439c984b60561c76843d": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "9e86ec1e1ddc4e3c887b24c6ba8cb2ea": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "9e93f72a338246a3a2b5ca412e3e7179": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "a231b2d8cece4b899891de7af3d89320": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "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_931189a93cda49cfa4e46e0db48f3371",
            "placeholder": "​",
            "style": "IPY_MODEL_6a973d02543c4b91a94b85c57e8a8aec",
            "value": " 1000/1360 [00:00&lt;00:00, 5502.26 examples/s]"
          }
        },
        "ac3ea102011e4037a464dac24f598f2d": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "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": ""
          }
        },
        "b53f89fce60d4dbc9048742bafc4eb7d": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "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_358f974f1a654790a9f7f23b0d207573",
              "IPY_MODEL_6b155526877740c28ed1d29fd5b3a831",
              "IPY_MODEL_09a6ac632a2242e9b1964e012055f9ca"
            ],
            "layout": "IPY_MODEL_578bc08e2a1845e699ab42b7262cbb5f"
          }
        },
        "b954033668c042439cbc0343d54f6077": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "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": ""
          }
        },
        "bb15b94101bf4c6387ca570c38fcc11c": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "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_2565844bb5954ede9486fc77b9a10d6e",
            "max": 568,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_f60b0837e18047599db2d0e61c403ecf",
            "value": 568
          }
        },
        "bf4c986e9bad46b5901ab0e9ba17d6aa": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "c0f24bbfc27241da8ec14e280868b3e0": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "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": "",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_f6c9190f977245ab859e1a677c0f2af6",
            "max": 1360,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_142df8ca09c64c998eae98b9438b8c4c",
            "value": 1360
          }
        },
        "c88aa53a291d4916b0eb6c60efd76831": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "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_5b96a22825374ff38fc608de5b1da421",
              "IPY_MODEL_bb15b94101bf4c6387ca570c38fcc11c",
              "IPY_MODEL_14b904a24b1f4062a5173071c154f6da"
            ],
            "layout": "IPY_MODEL_160e3fc88ee046c39c3c808e389579c1"
          }
        },
        "c954a9f610134ffdb06f715c6e182355": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "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_eaca405716e44ccaaac7ebb310961db9",
            "max": 263,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_58e700dc905b4ba184ab3ddcfcf12ae1",
            "value": 263
          }
        },
        "d01f03a55dd3496a884d93665538e94a": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "d860afde15d64cb78aa8241864de8943": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "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_3a680ac3a0fd464aae169df14dad71ba",
            "placeholder": "​",
            "style": "IPY_MODEL_edeb6c73a5c942f1b7d3fb3135975e3e",
            "value": " 263/263 [00:00&lt;00:00, 4.78kB/s]"
          }
        },
        "de668fb1cc5f439d946b6bd1f3a931a0": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "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_4a86d561ef9b43b08675a602f974b5c7",
              "IPY_MODEL_c954a9f610134ffdb06f715c6e182355",
              "IPY_MODEL_d860afde15d64cb78aa8241864de8943"
            ],
            "layout": "IPY_MODEL_e5b48e63274b48b3b4d4bf72a38de4e6"
          }
        },
        "e40f78a2a6254fecab2083545144d8a7": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "e5b48e63274b48b3b4d4bf72a38de4e6": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "eaca405716e44ccaaac7ebb310961db9": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "ebefda73df4443078d53c6328b49f8b8": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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": "hidden",
            "width": null
          }
        },
        "edaaf3ac4ba74c56a5b8bfee73715a61": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "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_0556cb05494f4739929e7c5c4b0cd6d6",
            "placeholder": "​",
            "style": "IPY_MODEL_874f25e654e1447abbd889832dce4b6d",
            "value": " 221k/221k [00:00&lt;00:00, 768kB/s]"
          }
        },
        "edeb6c73a5c942f1b7d3fb3135975e3e": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "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": ""
          }
        },
        "f55f19231630409298a7bf6a350523c0": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "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": ""
          }
        },
        "f60b0837e18047599db2d0e61c403ecf": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "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": ""
          }
        },
        "f61ce8d981984eda93f91e6f82b242f1": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "f6c9190f977245ab859e1a677c0f2af6": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "fea3e9e4b9c64164a20162542815273c": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "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
          }
        },
        "ff5c37f9068c46cf866812330d4abf6f": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "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_66fcec54203a4aea8cb588dfa4c775f7",
            "placeholder": "​",
            "style": "IPY_MODEL_f55f19231630409298a7bf6a350523c0",
            "value": "Map:  74%"
          }
        }
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}