Roberta_base.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "source": [
        "!pip install transformers\n",
        "!pip install torch\n",
        "!pip install sklearn\n",
        "!pip install wget\n",
        "!pip install sentencepiece"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "f0woS0NjNZ2t",
        "outputId": "8eecdd5a-705f-4350-be03-a91abf7208e6"
      },
      "id": "f0woS0NjNZ2t",
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Collecting transformers\n",
            "  Downloading transformers-4.18.0-py3-none-any.whl (4.0 MB)\n",
            "\u001b[K     |████████████████████████████████| 4.0 MB 10.1 MB/s \n",
            "\u001b[?25hRequirement already satisfied: regex!=2019.12.17 in /usr/local/lib/python3.7/dist-packages (from transformers) (2019.12.20)\n",
            "Requirement already satisfied: filelock in /usr/local/lib/python3.7/dist-packages (from transformers) (3.6.0)\n",
            "Requirement already satisfied: tqdm>=4.27 in /usr/local/lib/python3.7/dist-packages (from transformers) (4.63.0)\n",
            "Requirement already satisfied: requests in /usr/local/lib/python3.7/dist-packages (from transformers) (2.23.0)\n",
            "Collecting sacremoses\n",
            "  Downloading sacremoses-0.0.49-py3-none-any.whl (895 kB)\n",
            "\u001b[K     |████████████████████████████████| 895 kB 39.3 MB/s \n",
            "\u001b[?25hRequirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.7/dist-packages (from transformers) (21.3)\n",
            "Collecting huggingface-hub<1.0,>=0.1.0\n",
            "  Downloading huggingface_hub-0.5.1-py3-none-any.whl (77 kB)\n",
            "\u001b[K     |████████████████████████████████| 77 kB 9.2 MB/s \n",
            "\u001b[?25hCollecting pyyaml>=5.1\n",
            "  Downloading PyYAML-6.0-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl (596 kB)\n",
            "\u001b[K     |████████████████████████████████| 596 kB 64.7 MB/s \n",
            "\u001b[?25hCollecting tokenizers!=0.11.3,<0.13,>=0.11.1\n",
            "  Downloading tokenizers-0.11.6-cp37-cp37m-manylinux_2_12_x86_64.manylinux2010_x86_64.whl (6.5 MB)\n",
            "\u001b[K     |████████████████████████████████| 6.5 MB 53.8 MB/s \n",
            "\u001b[?25hRequirement already satisfied: importlib-metadata in /usr/local/lib/python3.7/dist-packages (from transformers) (4.11.3)\n",
            "Requirement already satisfied: numpy>=1.17 in /usr/local/lib/python3.7/dist-packages (from transformers) (1.21.5)\n",
            "Requirement already satisfied: typing-extensions>=3.7.4.3 in /usr/local/lib/python3.7/dist-packages (from huggingface-hub<1.0,>=0.1.0->transformers) (3.10.0.2)\n",
            "Requirement already satisfied: pyparsing!=3.0.5,>=2.0.2 in /usr/local/lib/python3.7/dist-packages (from packaging>=20.0->transformers) (3.0.7)\n",
            "Requirement already satisfied: zipp>=0.5 in /usr/local/lib/python3.7/dist-packages (from importlib-metadata->transformers) (3.7.0)\n",
            "Requirement already satisfied: chardet<4,>=3.0.2 in /usr/local/lib/python3.7/dist-packages (from requests->transformers) (3.0.4)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.7/dist-packages (from requests->transformers) (2021.10.8)\n",
            "Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /usr/local/lib/python3.7/dist-packages (from requests->transformers) (1.24.3)\n",
            "Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.7/dist-packages (from requests->transformers) (2.10)\n",
            "Requirement already satisfied: click in /usr/local/lib/python3.7/dist-packages (from sacremoses->transformers) (7.1.2)\n",
            "Requirement already satisfied: six in /usr/local/lib/python3.7/dist-packages (from sacremoses->transformers) (1.15.0)\n",
            "Requirement already satisfied: joblib in /usr/local/lib/python3.7/dist-packages (from sacremoses->transformers) (1.1.0)\n",
            "Installing collected packages: pyyaml, tokenizers, sacremoses, huggingface-hub, transformers\n",
            "  Attempting uninstall: pyyaml\n",
            "    Found existing installation: PyYAML 3.13\n",
            "    Uninstalling PyYAML-3.13:\n",
            "      Successfully uninstalled PyYAML-3.13\n",
            "Successfully installed huggingface-hub-0.5.1 pyyaml-6.0 sacremoses-0.0.49 tokenizers-0.11.6 transformers-4.18.0\n",
            "Requirement already satisfied: torch in /usr/local/lib/python3.7/dist-packages (1.10.0+cu111)\n",
            "Requirement already satisfied: typing-extensions in /usr/local/lib/python3.7/dist-packages (from torch) (3.10.0.2)\n",
            "Requirement already satisfied: sklearn in /usr/local/lib/python3.7/dist-packages (0.0)\n",
            "Requirement already satisfied: scikit-learn in /usr/local/lib/python3.7/dist-packages (from sklearn) (1.0.2)\n",
            "Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.7/dist-packages (from scikit-learn->sklearn) (3.1.0)\n",
            "Requirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.7/dist-packages (from scikit-learn->sklearn) (1.1.0)\n",
            "Requirement already satisfied: scipy>=1.1.0 in /usr/local/lib/python3.7/dist-packages (from scikit-learn->sklearn) (1.4.1)\n",
            "Requirement already satisfied: numpy>=1.14.6 in /usr/local/lib/python3.7/dist-packages (from scikit-learn->sklearn) (1.21.5)\n",
            "Collecting wget\n",
            "  Downloading wget-3.2.zip (10 kB)\n",
            "Building wheels for collected packages: wget\n",
            "  Building wheel for wget (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for wget: filename=wget-3.2-py3-none-any.whl size=9675 sha256=e5dd9e318a574bf4c19cdd7e2393693a5d14fe8e9d915401f575d85afe4e68ab\n",
            "  Stored in directory: /root/.cache/pip/wheels/a1/b6/7c/0e63e34eb06634181c63adacca38b79ff8f35c37e3c13e3c02\n",
            "Successfully built wget\n",
            "Installing collected packages: wget\n",
            "Successfully installed wget-3.2\n",
            "Collecting sentencepiece\n",
            "  Downloading sentencepiece-0.1.96-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.2 MB)\n",
            "\u001b[K     |████████████████████████████████| 1.2 MB 10.0 MB/s \n",
            "\u001b[?25hInstalling collected packages: sentencepiece\n",
            "Successfully installed sentencepiece-0.1.96\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from transformers import AdamW, RobertaConfig\n",
        "from transformers import RobertaTokenizer, AutoModelForSequenceClassification, RobertaModel\n",
        "from torch.utils.data import DataLoader, RandomSampler, SequentialSampler, TensorDataset\n",
        "from sklearn.metrics import precision_recall_fscore_support, confusion_matrix, ConfusionMatrixDisplay\n",
        "from sklearn.model_selection import train_test_split\n",
        "import matplotlib.pyplot as plt\n",
        "%matplotlib inline\n",
        "import seaborn as sns\n",
        "import pandas as pd, numpy as np\n",
        "import random, torch, wget, os\n",
        "from tqdm import tqdm"
      ],
      "metadata": {
        "id": "8q723WwUNaWI"
      },
      "id": "8q723WwUNaWI",
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "id": "controversial-diversity",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "controversial-diversity",
        "outputId": "f0a56da6-0eed-4fe3-97d3-3f2e1c9951ae"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "There are 1 GPU(s) available.\n",
            "We will use the GPU: Tesla P100-PCIE-16GB\n"
          ]
        }
      ],
      "source": [
        "pd.set_option('precision', 2)\n",
        "\n",
        "local = False\n",
        "gpu = True\n",
        "epochs = 4\n",
        "batch_size = 32\n",
        "\n",
        "# If there's a GPU available...\n",
        "if torch.cuda.is_available() and not local:\n",
        "    # Tell PyTorch to use the GPU.    \n",
        "    device = torch.device(\"cuda\")\n",
        "\n",
        "    print('There are %d GPU(s) available.' % torch.cuda.device_count())\n",
        "\n",
        "    print('We will use the GPU:', torch.cuda.get_device_name(0))\n",
        "\n",
        "# If not...\n",
        "else:\n",
        "    print('No GPU available, using the CPU instead.')\n",
        "    device = torch.device(\"cpu\")\n",
        "    gpu = False"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "automatic-denial",
      "metadata": {
        "id": "automatic-denial"
      },
      "source": [
        "### Dowloading dataset"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "id": "transparent-classics",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "id": "transparent-classics",
        "outputId": "8e704ae0-ca83-46de-f2cd-a2c352a4deb5"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'./webis_test.csv'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 4
        }
      ],
      "source": [
        "wget.download(\"https://ml-coding-test.s3.eu-west-1.amazonaws.com/webis_train.csv\", './webis_train.csv')\n",
        "wget.download(\"https://ml-coding-test.s3.eu-west-1.amazonaws.com/webis_test.csv\", './webis_test.csv')"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "variable-sellers",
      "metadata": {
        "id": "variable-sellers"
      },
      "source": [
        "### Reading dataset and splitting training data"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "id": "miniature-modeling",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "miniature-modeling",
        "outputId": "684b9d5d-501d-46a5-f5e3-2766c58d3415"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of training sentences: 17,535\n",
            "\n",
            "Number of validation sentences: 1,949\n",
            "\n",
            "Number of test sentences: 18,913\n",
            "\n",
            "Max length: 124\n",
            "\n"
          ]
        }
      ],
      "source": [
        "def cast_labels(data: pd.core.series.Series):\n",
        "    data[data==\"no-clickbait\"] = 0\n",
        "    data[data==\"clickbait\"] = 1\n",
        "    return data\n",
        "\n",
        "df = pd.read_csv(\"webis_train.csv\", delimiter=',', header=0)\n",
        "# Removing nan values to simplify training process / I will only use postText to classify each instance\n",
        "df = df[[\"postText\", \"truthClass\"]].dropna()\n",
        "training_sentences = df.postText\n",
        "training_labels = cast_labels(df.truthClass)\n",
        "\n",
        "training_sentences, validation_sentences, training_labels, validation_labels = train_test_split(training_sentences, \n",
        "                                                                                                training_labels, test_size=0.1, random_state=42, stratify=training_labels)\n",
        "\n",
        "df = pd.read_csv(\"webis_test.csv\", delimiter=',', header=0)\n",
        "# Removing nan values to simplify training process / I will only use postText to classify each instance\n",
        "df = df[[\"postText\", \"truthClass\"]].dropna()\n",
        "test_sentences = df.postText\n",
        "test_labels = cast_labels(df.truthClass)\n",
        "\n",
        "del df\n",
        "print('Number of training sentences: {:,}\\n'.format(training_sentences.shape[0]))\n",
        "print('Number of validation sentences: {:,}\\n'.format(validation_sentences.shape[0]))\n",
        "print('Number of test sentences: {:,}\\n'.format(test_sentences.shape[0]))\n",
        "max_length = max([max(map(len, training_sentences)), max(map(len, validation_sentences)), max(map(len, test_sentences))])\n",
        "print('Max length: {:,}\\n'.format(max_length))\n",
        "# manual setting max_length to 200 to ensure compatibility after tokenization.\n",
        "max_length = 200"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "id": "hydraulic-belarus",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 113,
          "referenced_widgets": [
            "2632ab6bcdfa45118c88be284d090974",
            "182c8fd393ea49c88b51cdc5bfc2c4b6",
            "efb7368315e741f7a04b3c5d8cf21207",
            "d9579312d2304bfe860a52bd0c9974c3",
            "39045994c6374eacbfbe865e2b34413e",
            "92eb503b7ecd40d1978499bd8c237387",
            "306b5202004d428e82b284949f3184f8",
            "b574da97c5cb4ef88b4afb462ccdb64a",
            "b97e708f893a4ef3b134da80cf8fe38a",
            "f4e0395c167f4fe0b64a5fc3f2e33e86",
            "7b388c3fc35e4ab6b319e61a99ad751d",
            "c553ccead5cd4b38b4a17f4bdf51aa2b",
            "fc23f28dd62946529026a6ca1a95b43d",
            "24f3563c09a24f86802037d0cf21d1d2",
            "58942574809344e982d50d737a03af70",
            "4a5d8ce1bf9447b0984998e2c4d0d6c0",
            "caf6cfd34547420f97f390bb18ddb16b",
            "b71b46ec3194408caa0a2fd162b11113",
            "fbb59c6455e546018a1d1598304b65e8",
            "230d7222e05141ba8334e6320735598d",
            "9d55736fea7d4dc28c7e8743c1572b55",
            "075742bef8fd44b38bab0b1bd9aed704",
            "e400aeac856d4da08911d160d883ee8d",
            "5692363d2eee4026a54538c9c717386b",
            "40738dd844bf4bc98ba65b8c9fa1dcb5",
            "d3e5da3dda2d4c24ac0952b4a6aab536",
            "5ffb7a848f4e40ada514dfe10babfede",
            "3635c3611abc4d309e9bfdd7085cddfc",
            "3d6390bfe23e48d8be5099644aa57c08",
            "9a32e88b09a543d4ae7eca046e33fd03",
            "38ea274fb15441b2bd20e429a99e4916",
            "aee79de1e60c40c1a9aaf24c31ca963e",
            "0f31eca582fe4288b5bc20ff89336036"
          ]
        },
        "id": "hydraulic-belarus",
        "outputId": "f3e60684-c0fb-40e1-e08d-815a7b3e0b42"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading:   0%|          | 0.00/878k [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "2632ab6bcdfa45118c88be284d090974"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading:   0%|          | 0.00/446k [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "c553ccead5cd4b38b4a17f4bdf51aa2b"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading:   0%|          | 0.00/481 [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "e400aeac856d4da08911d160d883ee8d"
            }
          },
          "metadata": {}
        }
      ],
      "source": [
        "if local:\n",
        "    tokenizer = RobertaTokenizer.from_pretrained(\"./bert_models/roberta-base/\")\n",
        "else:\n",
        "    tokenizer = RobertaTokenizer.from_pretrained('roberta-base')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "id": "saved-james",
      "metadata": {
        "id": "saved-james"
      },
      "outputs": [],
      "source": [
        "def cast_data(data: pd.core.series.Series, max_length):\n",
        "    input_ids = []\n",
        "    attention_masks = []\n",
        "    for sent in data:\n",
        "        encoded_dict = tokenizer.encode_plus(\n",
        "                            sent,                      \n",
        "                            add_special_tokens = False, \n",
        "                            truncation=True,\n",
        "                            max_length = max_length,\n",
        "                            padding = \"max_length\",\n",
        "                            return_attention_mask = True,\n",
        "                            return_tensors = 'pt'\n",
        "                       )\n",
        "        input_ids.append(encoded_dict['input_ids'])\n",
        "        attention_masks.append(encoded_dict['attention_mask'])\n",
        "    input_ids = torch.cat(input_ids, dim=0)\n",
        "    attention_masks = torch.cat(attention_masks, dim=0)\n",
        "    return input_ids, attention_masks "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "id": "korean-maria",
      "metadata": {
        "id": "korean-maria"
      },
      "outputs": [],
      "source": [
        "training_ids, training_attention = cast_data(training_sentences, max_length)\n",
        "training_labels = torch.tensor(training_labels.to_list())\n",
        "\n",
        "validation_ids, validation_attention = cast_data(validation_sentences, max_length)\n",
        "validation_labels = torch.tensor(validation_labels.to_list())\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "id": "deadly-caution",
      "metadata": {
        "id": "deadly-caution"
      },
      "outputs": [],
      "source": [
        "training_dataset = TensorDataset(training_ids, training_attention, training_labels)\n",
        "train_dataloader = DataLoader(training_dataset,  sampler = RandomSampler(training_dataset),  batch_size = batch_size)\n",
        "validation_dataset = TensorDataset(validation_ids, validation_attention, validation_labels)\n",
        "validation_dataloader = DataLoader(validation_dataset, sampler = SequentialSampler(validation_dataset), batch_size = batch_size)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "id": "arranged-motor",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 156,
          "referenced_widgets": [
            "44eeed1d20544f9d8d38f5bdb7d6fc43",
            "b230a0c2a9f44791833a66e93748126a",
            "2954ccc1d45a4dcdb3fafe485aa8567a",
            "caebdbd785bd4df5b401ad9ab0c8895d",
            "f452fed41d18459d9748415e60fb45e2",
            "7641147695624277b82ab446201d4394",
            "2c0b4662ac594c099401727e733cf19b",
            "6d68b056bd5e4708a2ca3fe851128c6e",
            "ec4da00f8d2d47c595c26cbfefe4cdb8",
            "92a49ae2030443f7ba56bd772ee9196a",
            "f35d37da113749328b77a8c69027f535"
          ]
        },
        "id": "arranged-motor",
        "outputId": "ca139334-b773-49a7-e970-33fb2f5355d0"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading:   0%|          | 0.00/478M [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "44eeed1d20544f9d8d38f5bdb7d6fc43"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Some weights of the model checkpoint at roberta-base were not used when initializing RobertaForSequenceClassification: ['roberta.pooler.dense.bias', 'lm_head.layer_norm.weight', 'roberta.pooler.dense.weight', 'lm_head.layer_norm.bias', 'lm_head.dense.weight', 'lm_head.decoder.weight', 'lm_head.dense.bias', 'lm_head.bias']\n",
            "- This IS expected if you are initializing RobertaForSequenceClassification 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 RobertaForSequenceClassification 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 RobertaForSequenceClassification were not initialized from the model checkpoint at roberta-base and are newly initialized: ['classifier.dense.bias', 'classifier.out_proj.weight', 'classifier.out_proj.bias', 'classifier.dense.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": [
        "if local:\n",
        "    model = AutoModelForSequenceClassification.from_pretrained(\"./bert_models/roberta-base/\", \n",
        "                                                          num_labels = 2, \n",
        "                                                          output_attentions = False, output_hidden_states = False)\n",
        "else:\n",
        "    model = AutoModelForSequenceClassification.from_pretrained(\"roberta-base\", \n",
        "                                                          num_labels = 2, \n",
        "                                                          output_attentions = False, output_hidden_states = False)\n",
        "if gpu:\n",
        "    model.cuda()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "id": "descending-barcelona",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "descending-barcelona",
        "outputId": "591cd4fa-2696-4e14-f63a-2c937ea3dd16"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/transformers/optimization.py:309: 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",
            "  FutureWarning,\n"
          ]
        }
      ],
      "source": [
        "optimizer = AdamW(model.parameters(), lr = 2e-5)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "id": "accompanied-administrator",
      "metadata": {
        "id": "accompanied-administrator"
      },
      "outputs": [],
      "source": [
        "# Function to calculate the accuracy of our predictions vs labels\n",
        "def flat_accuracy(preds, labels):\n",
        "    labels_flat = labels.flatten()\n",
        "    return np.sum(np.argmax(preds, axis=1).flatten() == labels_flat) / len(labels_flat)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "id": "black-branch",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "black-branch",
        "outputId": "9a8760e1-7836-4f77-b28a-f7a266376938"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Training epoch 0: 548it [05:41,  1.60it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "  Average training loss: 0.38\n",
            "\n",
            "Running Validation...\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Processing validation dataset: 100%|██████████| 61/61 [00:12<00:00,  4.92it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "  Accuracy: 0.84\n",
            "  Validation Loss: 0.36\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Training epoch 1: 548it [05:32,  1.65it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "  Average training loss: 0.30\n",
            "\n",
            "Running Validation...\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Processing validation dataset: 100%|██████████| 61/61 [00:12<00:00,  4.95it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "  Accuracy: 0.85\n",
            "  Validation Loss: 0.35\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Training epoch 2: 548it [05:32,  1.65it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "  Average training loss: 0.25\n",
            "\n",
            "Running Validation...\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Processing validation dataset: 100%|██████████| 61/61 [00:12<00:00,  4.95it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "  Accuracy: 0.85\n",
            "  Validation Loss: 0.35\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Training epoch 3: 548it [05:32,  1.65it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "  Average training loss: 0.19\n",
            "\n",
            "Running Validation...\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Processing validation dataset: 100%|██████████| 61/61 [00:12<00:00,  4.95it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "  Accuracy: 0.85\n",
            "  Validation Loss: 0.46\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\n"
          ]
        }
      ],
      "source": [
        "seed_val = 42\n",
        "\n",
        "random.seed(seed_val)\n",
        "np.random.seed(seed_val)\n",
        "torch.manual_seed(seed_val)\n",
        "torch.cuda.manual_seed_all(seed_val)\n",
        "\n",
        "history = []\n",
        "\n",
        "for epoch_i in range(0, epochs):\n",
        "    total_train_loss = 0\n",
        "    \n",
        "    model.train()\n",
        "\n",
        "    progress_bar = tqdm(enumerate(train_dataloader), desc=\"Training epoch \"+str(epoch_i), miniters=20)\n",
        "    for step, batch in progress_bar:\n",
        "        b_input_ids = batch[0].to(device)\n",
        "        b_input_mask = batch[1].to(device)\n",
        "        b_labels = batch[2].to(device)\n",
        "\n",
        "        model.zero_grad()        \n",
        "\n",
        "        loss, logits = model(b_input_ids, \n",
        "                             token_type_ids=None, \n",
        "                             attention_mask=b_input_mask, \n",
        "                             labels=b_labels).to_tuple()\n",
        "\n",
        "        total_train_loss += loss.item()\n",
        "        loss.backward()\n",
        "        torch.nn.utils.clip_grad_norm_(model.parameters(), 1.0)\n",
        "        optimizer.step()\n",
        "\n",
        "    avg_train_loss = total_train_loss / len(train_dataloader)            \n",
        "    print(\"\")\n",
        "    print(\"  Average training loss: {0:.2f}\".format(avg_train_loss))\n",
        "    \n",
        "    \n",
        "    print(\"\")\n",
        "    print(\"Running Validation...\")\n",
        "\n",
        "\n",
        "    model.eval()\n",
        "\n",
        "    total_eval_accuracy = 0\n",
        "    total_eval_loss = 0\n",
        "    nb_eval_steps = 0\n",
        "    progress_bar = tqdm(validation_dataloader, desc=\"Processing validation dataset\", miniters=20)\n",
        "\n",
        "    # Evaluate data for one epoch\n",
        "    for batch in progress_bar:        \n",
        "        b_input_ids = batch[0].to(device)\n",
        "        b_input_mask = batch[1].to(device)\n",
        "        b_labels = batch[2].to(device)\n",
        "        \n",
        "        with torch.no_grad():        \n",
        "            loss, logits = model(b_input_ids, \n",
        "                                   token_type_ids=None, \n",
        "                                   attention_mask=b_input_mask,\n",
        "                                   labels=b_labels).to_tuple()\n",
        "            \n",
        "        total_eval_loss += loss.item()\n",
        "\n",
        "        logits = logits.detach().cpu().numpy()\n",
        "        label_ids = b_labels.to('cpu').numpy()\n",
        "\n",
        "        total_eval_accuracy += flat_accuracy(logits, label_ids)\n",
        "    avg_val_accuracy = total_eval_accuracy / len(validation_dataloader)\n",
        "    print(\"  Accuracy: {0:.2f}\".format(avg_val_accuracy))\n",
        "\n",
        "\n",
        "    avg_val_loss = total_eval_loss / len(validation_dataloader)\n",
        "    \n",
        "    print(\"  Validation Loss: {0:.2f}\".format(avg_val_loss))\n",
        "\n",
        "    \n",
        "    history.append(\n",
        "        {\n",
        "            'epoch': epoch_i + 1,\n",
        "            'Training Loss': avg_train_loss,\n",
        "            'Valid. Loss': avg_val_loss,\n",
        "            'Valid. Accur.': avg_val_accuracy,\n",
        "\n",
        "\n",
        "        }\n",
        "    )"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "through-cambridge",
      "metadata": {
        "id": "through-cambridge"
      },
      "source": [
        "### Training stats"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "id": "compound-backup",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "compound-backup",
        "outputId": "548e0e8e-72e1-44db-e0f9-7df3eab6caa8"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "       Training Loss  Valid. Loss  Valid. Accur.\n",
              "epoch                                           \n",
              "1               0.38         0.36           0.84\n",
              "2               0.30         0.35           0.85\n",
              "3               0.25         0.35           0.85\n",
              "4               0.19         0.46           0.85"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-7b43ceb1-4af7-4b01-9c76-0b859efac141\">\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>Training Loss</th>\n",
              "      <th>Valid. Loss</th>\n",
              "      <th>Valid. Accur.</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>epoch</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.38</td>\n",
              "      <td>0.36</td>\n",
              "      <td>0.84</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.30</td>\n",
              "      <td>0.35</td>\n",
              "      <td>0.85</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.25</td>\n",
              "      <td>0.35</td>\n",
              "      <td>0.85</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.19</td>\n",
              "      <td>0.46</td>\n",
              "      <td>0.85</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-7b43ceb1-4af7-4b01-9c76-0b859efac141')\"\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-7b43ceb1-4af7-4b01-9c76-0b859efac141 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-7b43ceb1-4af7-4b01-9c76-0b859efac141');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 14
        }
      ],
      "source": [
        "df_stats = pd.DataFrame(data=history).set_index('epoch')\n",
        "df_stats"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "id": "exact-latex",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 427
        },
        "id": "exact-latex",
        "outputId": "49c6517a-6c89-4072-a613-ed8f06d669b3"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 864x432 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "# Use plot styling from seaborn.\n",
        "sns.set(style='darkgrid')\n",
        "\n",
        "# Increase the plot size and font size.\n",
        "sns.set(font_scale=1.5)\n",
        "plt.rcParams[\"figure.figsize\"] = (12,6)\n",
        "\n",
        "# Plot the learning curve.\n",
        "plt.plot(df_stats['Training Loss'], 'b-o', label=\"Training\")\n",
        "plt.plot(df_stats['Valid. Loss'], 'g-o', label=\"Validation\")\n",
        "\n",
        "# Label the plot.\n",
        "plt.title(\"Training & Validation Loss\")\n",
        "plt.xlabel(\"Epoch\")\n",
        "plt.ylabel(\"Loss\")\n",
        "plt.legend()\n",
        "plt.xticks([1, 2, 3, 4])\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "grateful-depression",
      "metadata": {
        "id": "grateful-depression"
      },
      "source": [
        "## Testing the model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "id": "invisible-attack",
      "metadata": {
        "id": "invisible-attack"
      },
      "outputs": [],
      "source": [
        "test_ids, test_attention = cast_data(test_sentences, max_length)\n",
        "test_labels = torch.tensor(test_labels.to_list())\n",
        "prediction_data = TensorDataset(test_ids, test_attention, test_labels)\n",
        "prediction_sampler = SequentialSampler(prediction_data)\n",
        "prediction_dataloader = DataLoader(prediction_data, sampler=prediction_sampler, batch_size=32)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "id": "opposite-python",
      "metadata": {
        "id": "opposite-python"
      },
      "outputs": [],
      "source": [
        "model.eval()\n",
        "predictions , true_labels = [], []\n",
        "for batch in prediction_dataloader:\n",
        "    batch = tuple(t.to(device) for t in batch)\n",
        "    b_input_ids, b_input_mask, b_labels = batch\n",
        "  \n",
        "    with torch.no_grad():\n",
        "        outputs = model(b_input_ids, token_type_ids=None, \n",
        "                          attention_mask=b_input_mask).to_tuple()\n",
        "    logits = outputs[0]\n",
        "    logits = logits.detach().cpu().numpy()\n",
        "    label_ids = b_labels.to('cpu').numpy()\n",
        "    predictions+=[np.argmax(x) for x in logits.tolist()]\n",
        "    true_labels+=label_ids.tolist()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "id": "european-integral",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "european-integral",
        "outputId": "900c04bd-a409-4eed-eb3c-717a3d9b5a41"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Precision 0.8138637379705129\n",
            "Recall 0.7923429467140097\n",
            "Fscore 0.802142024605851\n"
          ]
        }
      ],
      "source": [
        "precision, recall, fscore, _ = precision_recall_fscore_support(true_labels, predictions, average=\"macro\")\n",
        "print(\"Precision\", precision)\n",
        "print(\"Recall\", recall)\n",
        "print(\"Fscore\", fscore)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "id": "acoustic-utility",
      "metadata": {
        "id": "acoustic-utility",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 407
        },
        "outputId": "2c3a717c-294d-4720-b3a7-f68b896994ce"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 864x432 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay\n",
        "cm = confusion_matrix(true_labels, predictions)\n",
        "disp = ConfusionMatrixDisplay(confusion_matrix=cm)\n",
        "disp.plot()\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Analyzing false positives\n",
        "limit_to_print = 20\n",
        "for i, (true, pred) in enumerate(zip(true_labels, predictions)):\n",
        "  if not true and pred and limit_to_print:\n",
        "    print(test_sentences[i])\n",
        "    print()\n",
        "    print()\n",
        "    limit_to_print-=1"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ehFlmpzpiWls",
        "outputId": "615cfdcf-8fdf-4b1b-ecb4-5376206d50c3"
      },
      "id": "ehFlmpzpiWls",
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Don't abandon the snot if you want to be healthy. #EwwButTrue\n",
            "\n",
            "\n",
            "United Takes Added Steps to Win Back Customers and Avoid More Ugly Events\n",
            "\n",
            "\n",
            "Shea Moisture put out—and promptly pulled—(another) controversial ad.\n",
            "\n",
            "\n",
            "Here's where weed is legal in the U.S.:\n",
            "\n",
            "\n",
            "The fascinating hidden history of London's lost rivers\n",
            "\n",
            "\n",
            "21 really cute tattoos that are perfect for couples\n",
            "\n",
            "\n",
            "How to get Lady Gaga's mesmerizing eyes from #Coachella\n",
            "\n",
            "\n",
            "As per latest leak coming out of #China, the #iPhone8 may sport some truly unique #features\n",
            "\n",
            "\n",
            "Asteroid study explains how disaster movies get it wrong  via @NBCNewsMACH\n",
            "\n",
            "\n",
            "Big ideas: the airline mess, provocative views from Bridgewater's Dalio. Plus @domino's win!\n",
            "\n",
            "\n",
            "This cyclist just conquered the art of aerodynamics\n",
            "\n",
            "🎥:\n",
            "\n",
            "\n",
            "27 stylish bathing suits you can get on Amazon\n",
            "\n",
            "\n",
            "'Poppers' can cause irreversible damage to your eyes\n",
            "\n",
            "\n",
            "Elon Musk has an awkward problem at Tesla: employee parking\n",
            "\n",
            "\n",
            "State of emergency declared in Egypt following Palm Sunday church attacks\n",
            "\n",
            "\n",
            "MTV gets rid of gender-specific categories for the MTV Movie &amp; TV Awards.\n",
            "\n",
            "\n",
            "Senate expected to confirm Neil Gorsuch as Supreme Court justice\n",
            "\n",
            "\n",
            "The National Organization for Women call for Bill O'Reilly to be fired as 12 companies pull their ads\n",
            "\n",
            "\n",
            "What is the nuclear option? Here is how the maneuver would work:\n",
            "\n",
            "\n",
            "Explorer: Along the Mekong River, a Timeless Place Where Time Marches On\n",
            "\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Analyzing false negative\n",
        "limit_to_print = 20\n",
        "for i, (true, pred) in enumerate(zip(true_labels, predictions)):\n",
        "  if true and not pred and limit_to_print:\n",
        "    print(test_sentences[i])\n",
        "    limit_to_print-=1"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NJqheBHujEY0",
        "outputId": "02782ea6-656e-456c-d014-7dce3d2971b9"
      },
      "id": "NJqheBHujEY0",
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "He worked in a factory floor. He now helps build the robots that are replacing human labor\n",
            "Congratulations, Kalpit Veerval 👏\n",
            "Jealous wife chops off her husband's PENIS with a pair of scissors after growing 'tired of his womanising'\n",
            "Latest Uber controversy sheds light on how companies use your emails:\n",
            "\"Why is Bill Clinton given a pass over the other Bills?\"\n",
            "Guys, the new “Zelda” game is so freakin’ fun\n",
            "What the French elections mean for Brexit\n",
            "Did Von Miller wear a better Coachella outfit than Cam Newton? 🤔\n",
            "How the shale boom turned the world's energy markets upside down:\n",
            "Opinions: A young prince is reimagining Saudi Arabia. Can he make his vision come true?\n",
            "What to watch for in the first round of the French presidential election\n",
            "Chris Brown's new documentary is coming to theaters and shit gets real\n",
            "Dana White reveals just how much money Conor McGregor could bank for Floyd Mayweather fight\n",
            "Riding the wave: Meet the surfer who helped 7 million people get clean water\n",
            "Here's what the 'Game of Thrones' stars look like in real life\n",
            "The internet has found Tom Hardy's doppelganger. 👀\n",
            "Check out the full credits for Harry Styles' #SignoftheTimes single\n",
            "REVEALED: Kim’s plan to kidnap Western tourists from South Korea\n",
            "A roundup of top #entertainment stories from the day that was\n",
            "America's airlines are shameless. But United just set a new low | Arwa Mahdawi\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        ""
      ],
      "metadata": {
        "id": "T5A8rXRkjIwt"
      },
      "id": "T5A8rXRkjIwt",
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.8.0"
    },
    "colab": {
      "name": "Roberta_base.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "accelerator": "GPU",
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "2632ab6bcdfa45118c88be284d090974": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_182c8fd393ea49c88b51cdc5bfc2c4b6",
              "IPY_MODEL_efb7368315e741f7a04b3c5d8cf21207",
              "IPY_MODEL_d9579312d2304bfe860a52bd0c9974c3"
            ],
            "layout": "IPY_MODEL_39045994c6374eacbfbe865e2b34413e"
          }
        },
        "182c8fd393ea49c88b51cdc5bfc2c4b6": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_92eb503b7ecd40d1978499bd8c237387",
            "placeholder": "​",
            "style": "IPY_MODEL_306b5202004d428e82b284949f3184f8",
            "value": "Downloading: 100%"
          }
        },
        "efb7368315e741f7a04b3c5d8cf21207": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_b574da97c5cb4ef88b4afb462ccdb64a",
            "max": 898823,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_b97e708f893a4ef3b134da80cf8fe38a",
            "value": 898823
          }
        },
        "d9579312d2304bfe860a52bd0c9974c3": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_f4e0395c167f4fe0b64a5fc3f2e33e86",
            "placeholder": "​",
            "style": "IPY_MODEL_7b388c3fc35e4ab6b319e61a99ad751d",
            "value": " 878k/878k [00:00&lt;00:00, 1.36MB/s]"
          }
        },
        "39045994c6374eacbfbe865e2b34413e": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "92eb503b7ecd40d1978499bd8c237387": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "306b5202004d428e82b284949f3184f8": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "b574da97c5cb4ef88b4afb462ccdb64a": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "b97e708f893a4ef3b134da80cf8fe38a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "f4e0395c167f4fe0b64a5fc3f2e33e86": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "7b388c3fc35e4ab6b319e61a99ad751d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "c553ccead5cd4b38b4a17f4bdf51aa2b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_fc23f28dd62946529026a6ca1a95b43d",
              "IPY_MODEL_24f3563c09a24f86802037d0cf21d1d2",
              "IPY_MODEL_58942574809344e982d50d737a03af70"
            ],
            "layout": "IPY_MODEL_4a5d8ce1bf9447b0984998e2c4d0d6c0"
          }
        },
        "fc23f28dd62946529026a6ca1a95b43d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_caf6cfd34547420f97f390bb18ddb16b",
            "placeholder": "​",
            "style": "IPY_MODEL_b71b46ec3194408caa0a2fd162b11113",
            "value": "Downloading: 100%"
          }
        },
        "24f3563c09a24f86802037d0cf21d1d2": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_fbb59c6455e546018a1d1598304b65e8",
            "max": 456318,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_230d7222e05141ba8334e6320735598d",
            "value": 456318
          }
        },
        "58942574809344e982d50d737a03af70": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_9d55736fea7d4dc28c7e8743c1572b55",
            "placeholder": "​",
            "style": "IPY_MODEL_075742bef8fd44b38bab0b1bd9aed704",
            "value": " 446k/446k [00:00&lt;00:00, 1.22MB/s]"
          }
        },
        "4a5d8ce1bf9447b0984998e2c4d0d6c0": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "caf6cfd34547420f97f390bb18ddb16b": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "b71b46ec3194408caa0a2fd162b11113": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "fbb59c6455e546018a1d1598304b65e8": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "230d7222e05141ba8334e6320735598d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "9d55736fea7d4dc28c7e8743c1572b55": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "075742bef8fd44b38bab0b1bd9aed704": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "e400aeac856d4da08911d160d883ee8d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_5692363d2eee4026a54538c9c717386b",
              "IPY_MODEL_40738dd844bf4bc98ba65b8c9fa1dcb5",
              "IPY_MODEL_d3e5da3dda2d4c24ac0952b4a6aab536"
            ],
            "layout": "IPY_MODEL_5ffb7a848f4e40ada514dfe10babfede"
          }
        },
        "5692363d2eee4026a54538c9c717386b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_3635c3611abc4d309e9bfdd7085cddfc",
            "placeholder": "​",
            "style": "IPY_MODEL_3d6390bfe23e48d8be5099644aa57c08",
            "value": "Downloading: 100%"
          }
        },
        "40738dd844bf4bc98ba65b8c9fa1dcb5": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_9a32e88b09a543d4ae7eca046e33fd03",
            "max": 481,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_38ea274fb15441b2bd20e429a99e4916",
            "value": 481
          }
        },
        "d3e5da3dda2d4c24ac0952b4a6aab536": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_aee79de1e60c40c1a9aaf24c31ca963e",
            "placeholder": "​",
            "style": "IPY_MODEL_0f31eca582fe4288b5bc20ff89336036",
            "value": " 481/481 [00:00&lt;00:00, 14.1kB/s]"
          }
        },
        "5ffb7a848f4e40ada514dfe10babfede": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "3635c3611abc4d309e9bfdd7085cddfc": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "3d6390bfe23e48d8be5099644aa57c08": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "9a32e88b09a543d4ae7eca046e33fd03": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "38ea274fb15441b2bd20e429a99e4916": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "aee79de1e60c40c1a9aaf24c31ca963e": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "0f31eca582fe4288b5bc20ff89336036": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "44eeed1d20544f9d8d38f5bdb7d6fc43": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_b230a0c2a9f44791833a66e93748126a",
              "IPY_MODEL_2954ccc1d45a4dcdb3fafe485aa8567a",
              "IPY_MODEL_caebdbd785bd4df5b401ad9ab0c8895d"
            ],
            "layout": "IPY_MODEL_f452fed41d18459d9748415e60fb45e2"
          }
        },
        "b230a0c2a9f44791833a66e93748126a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_7641147695624277b82ab446201d4394",
            "placeholder": "​",
            "style": "IPY_MODEL_2c0b4662ac594c099401727e733cf19b",
            "value": "Downloading: 100%"
          }
        },
        "2954ccc1d45a4dcdb3fafe485aa8567a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_6d68b056bd5e4708a2ca3fe851128c6e",
            "max": 501200538,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_ec4da00f8d2d47c595c26cbfefe4cdb8",
            "value": 501200538
          }
        },
        "caebdbd785bd4df5b401ad9ab0c8895d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_92a49ae2030443f7ba56bd772ee9196a",
            "placeholder": "​",
            "style": "IPY_MODEL_f35d37da113749328b77a8c69027f535",
            "value": " 478M/478M [00:11&lt;00:00, 25.8MB/s]"
          }
        },
        "f452fed41d18459d9748415e60fb45e2": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "7641147695624277b82ab446201d4394": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "2c0b4662ac594c099401727e733cf19b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "6d68b056bd5e4708a2ca3fe851128c6e": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "ec4da00f8d2d47c595c26cbfefe4cdb8": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "92a49ae2030443f7ba56bd772ee9196a": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "f35d37da113749328b77a8c69027f535": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        }
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 5
}