ICRS-OlmerV6.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## HEART-MET Assistive Robot Challenge with VideoMae Transformers\n",
    "\n",
    "Olmer Garcia-Bedoya, Jose Tomas Lorente , Sebastian Murcia\n",
    "\n",
    "**[Ekumens Inc](https://www.ekumenlabs.com/)**\n",
    "Copyright 2022- Ekumen Inc. All rights reserved.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 3526,
     "status": "ok",
     "timestamp": 1667574556192,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "cBO0vpkwXi2-",
    "outputId": "cedf09b6-3e1c-488b-abb4-e3cce7f3ccd0"
   },
   "outputs": [],
   "source": [
    "#!pip install torch torchvision torchaudio\n",
    "import torch\n",
    "print('gpu?'+str(torch.cuda.is_available()))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Required packages\n",
    "\n",
    "Package installed (just one time or in google colab (you do not need all of this packages)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# !pip install --upgrade pip\n",
    "# !pip install nvidia-pyindex\n",
    "# !pip install nvidia-tensorrt==8.4.3.1\n",
    "# !pip install torch-tensorrt==1.2.0 --find-links https://github.com/pytorch/TensorRT/releases/expanded_assets/v1.2.0\n",
    "# !pip install transformers\n",
    "# !pip install decord\n",
    "# !pip install opencv-python\n",
    "# !pip install pillow\n",
    "# !pip install torchvision\n",
    "# !pip install datasets\n",
    "# !pip install --upgrade requests\n",
    "# !pip install --upgrade scikit-learn scipy matplotlib"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "mPZJoSgn3_AN"
   },
   "source": [
    "## Load data from zip\n",
    "\n",
    "It is made the first time to unzip the data. Also work in colab sharing the zip a folder *ICSR 2022 HEART-MET*."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "executionInfo": {
     "elapsed": 904,
     "status": "ok",
     "timestamp": 1667574565895,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "XGxBdNc5z-YX"
   },
   "outputs": [],
   "source": [
    "# !mkdir training_set test_set validation_set\n",
    "# !ls"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 10002,
     "status": "ok",
     "timestamp": 1667574583346,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "Zb2C99M5zjyQ",
    "outputId": "f25df569-1d20-4605-c12f-16f5656fc04d"
   },
   "outputs": [],
   "source": [
    "# !pip install pyunpack\n",
    "# !pip install patool\n",
    "# !pip install py7zr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "## only google colab\n",
    "\n",
    "#from google.colab import drive\n",
    "#drive.mount('/content/drive', force_remount=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "executionInfo": {
     "elapsed": 101613,
     "status": "ok",
     "timestamp": 1667574721128,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "unNYWGc4yp2x"
   },
   "outputs": [],
   "source": [
    "# from pyunpack import Archive\n",
    "# Archive('training_set.7z').extractall(\"training_set\")\n",
    "# Archive('test_set.7z').extractall(\"test_set\")\n",
    "# Archive('validation_set.7z').extractall(\"validation_set\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "executionInfo": {
     "elapsed": 1323,
     "status": "ok",
     "timestamp": 1667574731965,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "JDX-wKANWi8o"
   },
   "outputs": [],
   "source": [
    "## only google colab\n",
    "#! cp \"/content/drive/MyDrive/ICSR 2022 HEART-MET Activity Recognition Challenge/training_labels.json\" /content/"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Configuration\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "executionInfo": {
     "elapsed": 2,
     "status": "ok",
     "timestamp": 1667574764430,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "FUj139BDY4Uy"
   },
   "outputs": [],
   "source": [
    "names=['Opening the door and walking in/out',\n",
    "        'Putting on a jacket',\n",
    "        'Touching a hot surface',\n",
    "        'Opening the fridge',\n",
    "        'Drinking water',\n",
    "        'Colliding against something',\n",
    "        'Eating food with a fork',\n",
    "        'Coughing or sneezing',\n",
    "        'Wiping a table',\n",
    "        'Reading a book',\n",
    "        'Neck roll exercise',\n",
    "        'Freehand exercise',\n",
    "        'Lying down',\n",
    "        'Limping',\n",
    "        'Talking on the phone',\n",
    "        'Using a computer',\n",
    "        'Falling down',\n",
    "        'Brushing teeth',\n",
    "        'Writing',\n",
    "        'Unknown activity']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "mY5TJSm4YRhs"
   },
   "source": [
    "The next code is adapted from (higgingFace->VideMAE-> VideoMAEForVideoClassification)[https://huggingface.co/docs/transformers/main/en/model_doc/videomae#transformers.VideoMAEForVideoClassification]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "executionInfo": {
     "elapsed": 4961,
     "status": "ok",
     "timestamp": 1667574772521,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "DB0vY8SrZGyS"
   },
   "outputs": [],
   "source": [
    "from decord import VideoReader, cpu\n",
    "from transformers import VideoMAEFeatureExtractor, VideoMAEForPreTraining,VideoMAEConfig,VideoMAEForVideoClassification\n",
    "import numpy as np\n",
    "import torch\n",
    "import os\n",
    "import json\n",
    "import csv #I make in a venv here more easy pandas..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 113,
     "referenced_widgets": [
      "60a8726cf933422db26f755d9c5623d1",
      "017e3f0a45c942e29b0fc432dca2ffc8",
      "373864a337c048778aacc3218f0aa5eb",
      "8f7ad298e78848c4918602274ddd45d1",
      "2e37ade3222b47b4bb3d9566dc862117",
      "55742a800669463d81d67456044285cb",
      "77d5ea8f038c4dd48d16cd885b00629a",
      "679fc76bd85144659d03c25ea1cf67b1",
      "e193535750a84ec0b98ab5077a6b5aa5",
      "5bcc6cfea7be4f3f8171f64ee9f44e17",
      "a533fd53359e42548409f60ac412c7ba",
      "bd9d32c5d158420990bf82ffb68f48c9",
      "a06160cdb0be42c2af5f55ec835e44ff",
      "295a086f553e46ecb8eb8b2c9b56916b",
      "17b6f0610cee47e1a33f2b54edc3631e",
      "9a9d479f3065407d88be6ca2b12abffa",
      "811ee525a96b49b1819473a5ccdc01a6",
      "6c590b31f3b043a4a2dcc3bacb742452",
      "fbdb8309e4654b6c839127fdb5e29cf0",
      "b698e60171154efabc766b47d2d7c486",
      "1a8775c21a7f466c944c790277a34326",
      "9b00cff2e12c4886b3f6cf1e6879d725",
      "9e90e00ea181485c8ac420e32b989f08",
      "fa4749f5fa94493a9ab3e9f218922106",
      "b7ff863729484b8fbb5fbc6a8c791e3e",
      "69f8d2bece1e4bc4987f81c63d82aded",
      "1af7e067011a4854a1e5f97624b230cd",
      "cafb90a4897b49028fbbcfe6b981e385",
      "622ef2865684418eb481e799c337c523",
      "87fea88df3494c8592b5de4485bcf8a7",
      "f04bb6ab61ee477c8a02719eeab23dff",
      "24fbf1047b6c4180815d2ea167d17252",
      "6687dc570dc74feba3bd2413c408eb76"
     ]
    },
    "executionInfo": {
     "elapsed": 30509,
     "status": "ok",
     "timestamp": 1667574811596,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "4ElqfGdNZBwO",
    "outputId": "5a9ec4ce-2667-4814-e1f0-a537176ec7b5"
   },
   "outputs": [],
   "source": [
    "# configuration = VideoMAEConfig({'patch_size':128})\n",
    "# configuration.return_dict=True\n",
    "\n",
    "name_model=\"MCG-NJU/videomae-base-finetuned-kinetics\"\n",
    "#TODO: uncomment to start from the last training model\n",
    "# name_model=\"./videomae-icsr-heart-v4\"\n",
    "\n",
    "feature_extractor = VideoMAEFeatureExtractor.from_pretrained(name_model)\n",
    "# model = VideoMAEForVideoClassification.from_pretrained(name_model)\n",
    "output=json.load(open('training_labels.json'))\n",
    "validation_labels = json.load(open('validation_labels.json'))\n",
    "path='training_set'\n",
    "validation_path='validation_set'\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "uUdrM7onc2Im"
   },
   "source": [
    "## Fine tuning VideoMAE\n",
    "\n",
    "Fine tuning is the gold of transformers, from few videos/images/sentences we can make a classifier.\n",
    "\n",
    "VideoMAE is based of a ImageMAE which is based in ViT (Vision Transformer), which is Based in Transformers, so let try.\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "4hXykNNbdm3F"
   },
   "source": [
    "### Create DataSet\n",
    "\n",
    "based in [image dataset](https://huggingface.co/docs/datasets/image_dataset). I think that it is not used in the best way the library [link text](https://huggingface.co/course/chapter5/4?fw=pt)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "id": "MDpKy2YZgUQm"
   },
   "outputs": [],
   "source": [
    "# !pip install --upgrade transformers decord datasets\n",
    "# !pip install --upgrade numpy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "executionInfo": {
     "elapsed": 1297,
     "status": "ok",
     "timestamp": 1667574841424,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "batqReKX57xm"
   },
   "outputs": [],
   "source": [
    "from datasets import Dataset\n",
    "import numpy as np\n",
    "\n",
    "my_set=[{'file_name':path+'/'+a,'labels':output[a][0],'description':names[output[a][0]]} for a in output.keys()]\n",
    "validation_set=[{'file_name':validation_path+'/'+a,'labels':validation_labels[a][0],'description':names[validation_labels[a][0]]} for a in validation_labels.keys()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "136"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(validation_set)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "792"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(my_set)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "executionInfo": {
     "elapsed": 8,
     "status": "ok",
     "timestamp": 1667574850782,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "_KPPhlYQ57xn"
   },
   "outputs": [],
   "source": [
    "\n",
    "full_augmented_set = {}\n",
    "full_augmented_set['train'] = my_set\n",
    "full_augmented_set['validation'] = validation_set\n",
    "\n",
    "with open(\"train_set.json\", \"w\") as final:\n",
    "    json.dump(full_augmented_set, final)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 249,
     "referenced_widgets": [
      "431c07b2e7ed4dbf9014c67d306945f8",
      "028465bf497d4737a5384f6f7c58200a",
      "d9b6dbfb87fd4c469058532892cf7a09",
      "cc31d15c0a584eb2af3b23c6db3d2503",
      "77033cd66ad340acbbdaeb82aa47591c",
      "462f437cc6144a7982a6dc89bd6f2a90",
      "21e17f19d8c84ababf07dd94864f53e5",
      "47c46c504521428d9bf84e798fde8937",
      "08301e86bd1745f28af465f1b9f9d050",
      "5f879ff41f624a7ba425b9e28646ab58",
      "824d814cdeea438798bd18426092fe41",
      "09e0443823d54f578af5cf635b7b3715",
      "ae9b118bc353457ab8e6c21208a8d57b",
      "bfe6a0d505be48b888ea13cdc0d56447",
      "bcea670e5b234125a4d801fafe8928f7",
      "045cd9695f3e479fb21048b39abfc18b",
      "24270d9757e247669e9ca2105a727cb8",
      "7e98903f8faf4030a3b46fb8b744b5e5",
      "743fdc0668b44aa2a79d9c64b1e2d08d",
      "7254dc86b6714a81b46997dc132360e0",
      "107b16ef81554781856cd67a6ae038da",
      "68141e3ab39b40169949a3576477a42b",
      "6372d32e9b8c4efb9e212a376a4c7947",
      "1144746deacc4c07aa06c81906f6ce77",
      "45af82f6eb18460784ab055d93f280ef",
      "c6d60b6c586f484ebc402f5d44cc2f7a",
      "7e0fd7d1189048f09ad63b0d6f5e7783",
      "4a2741e1b02749f5887bf798ee8e21d6",
      "52e6c6ed71eb427b88df2e3fbf5b87c1",
      "20fb65d2eee24bbb8d74fec87df215e6",
      "e6bcacc7468d430990aebd786bd5db3c",
      "3519c3abbb684bae90a994f8a089d4ff",
      "51379fdebbaf4587b5cabb85d4fb8faa",
      "1f6a6b3791e54b56bb5b7f2f0ba812d9",
      "1649a5747b8c48e58c4508a9906fea8f",
      "e2eafd1b4cfa4965bf05c9caa0ea4595",
      "e3ac5081836d40f797a5305f1429175e",
      "5ebfc1a0964a4af7a600afb7904c8e0b",
      "0dd258e5733e4584a0031bfa3a1e9392",
      "7e251276f132481e886901a9c0fa5562",
      "3197fad62a9d4d289c40772bfe021d7f",
      "71bfa349d37c4717bb34bb93e8779bd8",
      "5f2c37ea11f24e96864754cb94c6c5dd",
      "2f6043f91b71453ab2b7ac628c04e556",
      "fec838d6713444ff837f6f340502adb9",
      "21cf114acd51449da8bb1532a5f18f8b",
      "33da8b0d87a14575aa6f65b46f803bd2",
      "0e8d7313d9f24c49821bb81bb013a925",
      "7478c6eda4694875a8dcfb3b59676f26",
      "ddbbe0f725de4361bb71b756f749a215",
      "d251f651ebf24f6da07d1735559818b6",
      "13d7e978cba64a3cbb9bd0e9bb50ecbc",
      "20589c5d2372402e99f4c1406659c0ee",
      "5100f2875c724d10b1801c55084c33ab",
      "32ea79387ae845cea889c436ffabd9e1",
      "77c5c395d16247cf9bd2a0d4e7a216f9",
      "29aae51c7130465f97ce90491f0c9696",
      "bfa55478d7ed4d629c85c76d44c08a3c",
      "d1f5438cbf314ffaaecc473f9922ee70",
      "e9045bde9871448a863b44559d114a5a",
      "d20cd3c0c8484cf8a4581619bfd061fb",
      "c0b8d07504094fd3b4dc918fe80d5099",
      "fcaf3755cdc44224a46a32762fcae938",
      "c59ef23f2be94dcba663c347779c860a",
      "781aceb284cf4bb192dde0fd2b731079",
      "81aa9067dbd94f979c23bc64897a3eab"
     ]
    },
    "executionInfo": {
     "elapsed": 3621,
     "status": "ok",
     "timestamp": 1667574858629,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "KBRAccAV57xn",
    "outputId": "5a4bcc8b-f459-4c2e-ac0a-7bc3e5161209"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using custom data configuration default-8ba53836c2ef3b22\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading and preparing dataset json/default to /home/tomas/.cache/huggingface/datasets/json/default-8ba53836c2ef3b22/0.0.0...\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0881b372794d41ab827a15df80ccace1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading data files:   0%|          | 0/1 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d37f8794e30c4b2a96a94aab5bd4b781",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Extracting data files:   0%|          | 0/1 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "0 tables [00:00, ? tables/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using custom data configuration default-647a4d18c15a46f8\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dataset json downloaded and prepared to /home/tomas/.cache/huggingface/datasets/json/default-8ba53836c2ef3b22/0.0.0. Subsequent calls will reuse this data.\n",
      "Downloading and preparing dataset json/default to /home/tomas/.cache/huggingface/datasets/json/default-647a4d18c15a46f8/0.0.0...\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0bf340db9093446cb1065c6b2d954990",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading data files:   0%|          | 0/1 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4dd572404654467e9d38666e84d0b9e3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Extracting data files:   0%|          | 0/1 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "0 tables [00:00, ? tables/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dataset json downloaded and prepared to /home/tomas/.cache/huggingface/datasets/json/default-647a4d18c15a46f8/0.0.0. Subsequent calls will reuse this data.\n"
     ]
    }
   ],
   "source": [
    "# dataset = Dataset.from_json('train.json',field='train')\n",
    "# dataset_validation = Dataset.from_json('train.json',field='validation')\n",
    "# dataset_full = Dataset.from_json('full_train.json', field='train')\n",
    "dataset = Dataset.from_json('train_set.json',field='train')\n",
    "dataset_validation = Dataset.from_json('train_set.json',field='validation')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 194,
     "status": "ok",
     "timestamp": 1667421550894,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "ohfjncJwfSNz",
    "outputId": "89189f43-ecd5-4c17-9684-aa7ee9bc3a15"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'file_name': 'training_set/video0000.mp4',\n",
       " 'labels': 14,\n",
       " 'description': 'Talking on the phone'}"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset[0]\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "gZaUiZpT6cos"
   },
   "source": [
    "### Training and validation\n",
    "\n",
    "Based in [Fine tuning ViT](https://huggingface.co/blog/fine-tune-vit)\n",
    "\n",
    "####  sampling of the video"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# after many epochs 48 , it our validation give 0.71, but in the competition give 0.69\n",
    "# may be will be better to create a collate for test and other for training to get a better result during the simulation\n",
    "# this issu describe what todo https://github.com/huggingface/transformers/issues/13993, rewrite the function get_Data_allocation of the Trainer class (just inheritance) \n",
    "import torch\n",
    "def collate_with_path_batches(batch):\n",
    "    pixel_values = []\n",
    "    labels = []\n",
    "    for elem in batch:\n",
    "        file_path = elem['file_name']\n",
    "        videoreader = VideoReader(file_path, num_threads=1, ctx=cpu(0))\n",
    "        videoreader.seek(0)\n",
    "        frames = len(videoreader)\n",
    "        n_frames = 16\n",
    "        ##not improve nothing#send a fast motion video, like a crop!!\n",
    "        #random_speed = np.random.randint(0, 4)*n_frames\n",
    "        #frames=frames-random_speed\n",
    "        #initial_index = np.random.randint(0, n_frames+random_speed-1)\n",
    "        #indices = np.linspace(initial_index, frames-n_frames, num=n_frames)\n",
    "        \n",
    "        initial_index = np.random.randint(0, n_frames)        \n",
    "        \n",
    "        indices = np.linspace(initial_index, frames, num=n_frames)\n",
    "        indices = np.clip(indices, initial_index, frames- 1).astype(np.int64)\n",
    "        #TODO: add image augmentation\n",
    "        pixel_values.append(feature_extractor(list(videoreader.get_batch(indices).asnumpy()), return_tensors=\"pt\")['pixel_values'][0])\n",
    "        labels.append(elem['labels'])\n",
    "    return {\n",
    "        'pixel_values': torch.stack(pixel_values),\n",
    "        'labels': torch.tensor(labels)\n",
    "    }"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "executionInfo": {
     "elapsed": 2026,
     "status": "ok",
     "timestamp": 1667575074037,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "3G56RcBI6uOk"
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from datasets import load_metric\n",
    "\n",
    "metric = load_metric(\"accuracy\")\n",
    "def compute_metrics(p):\n",
    "    return metric.compute(predictions=np.argmax(p.predictions, axis=1), references=p.label_ids)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 4861,
     "status": "ok",
     "timestamp": 1667575087394,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "vsUZpL-66zPa",
    "outputId": "4b41f94c-11ee-4e14-c643-75bee01b3f8c"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "loading configuration file ./videomae-icsr-heart-v4/config.json\n",
      "Model config VideoMAEConfig {\n",
      "  \"_name_or_path\": \"MCG-NJU/videomae-base-finetuned-kinetics\",\n",
      "  \"architectures\": [\n",
      "    \"VideoMAEForVideoClassification\"\n",
      "  ],\n",
      "  \"attention_probs_dropout_prob\": 0.0,\n",
      "  \"decoder_hidden_size\": 384,\n",
      "  \"decoder_intermediate_size\": 1536,\n",
      "  \"decoder_num_attention_heads\": 6,\n",
      "  \"decoder_num_hidden_layers\": 4,\n",
      "  \"hidden_act\": \"gelu\",\n",
      "  \"hidden_dropout_prob\": 0.0,\n",
      "  \"hidden_size\": 768,\n",
      "  \"id2label\": {\n",
      "    \"0\": \"Opening the door and walking in/out\",\n",
      "    \"1\": \"Putting on a jacket\",\n",
      "    \"10\": \"Neck roll exercise\",\n",
      "    \"11\": \"Freehand exercise\",\n",
      "    \"12\": \"Lying down\",\n",
      "    \"13\": \"Limping\",\n",
      "    \"14\": \"Talking on the phone\",\n",
      "    \"15\": \"Using a computer\",\n",
      "    \"16\": \"Falling down\",\n",
      "    \"17\": \"Brushing teeth\",\n",
      "    \"18\": \"Writing\",\n",
      "    \"19\": \"Unknown activity\",\n",
      "    \"2\": \"Touching a hot surface\",\n",
      "    \"3\": \"Opening the fridge\",\n",
      "    \"4\": \"Drinking water\",\n",
      "    \"5\": \"Colliding against something\",\n",
      "    \"6\": \"Eating food with a fork\",\n",
      "    \"7\": \"Coughing or sneezing\",\n",
      "    \"8\": \"Wiping a table\",\n",
      "    \"9\": \"Reading a book\"\n",
      "  },\n",
      "  \"image_size\": 224,\n",
      "  \"initializer_range\": 0.02,\n",
      "  \"intermediate_size\": 3072,\n",
      "  \"label2id\": {\n",
      "    \"Brushing teeth\": \"17\",\n",
      "    \"Colliding against something\": \"5\",\n",
      "    \"Coughing or sneezing\": \"7\",\n",
      "    \"Drinking water\": \"4\",\n",
      "    \"Eating food with a fork\": \"6\",\n",
      "    \"Falling down\": \"16\",\n",
      "    \"Freehand exercise\": \"11\",\n",
      "    \"Limping\": \"13\",\n",
      "    \"Lying down\": \"12\",\n",
      "    \"Neck roll exercise\": \"10\",\n",
      "    \"Opening the door and walking in/out\": \"0\",\n",
      "    \"Opening the fridge\": \"3\",\n",
      "    \"Putting on a jacket\": \"1\",\n",
      "    \"Reading a book\": \"9\",\n",
      "    \"Talking on the phone\": \"14\",\n",
      "    \"Touching a hot surface\": \"2\",\n",
      "    \"Unknown activity\": \"19\",\n",
      "    \"Using a computer\": \"15\",\n",
      "    \"Wiping a table\": \"8\",\n",
      "    \"Writing\": \"18\"\n",
      "  },\n",
      "  \"layer_norm_eps\": 1e-12,\n",
      "  \"model_type\": \"videomae\",\n",
      "  \"norm_pix_loss\": false,\n",
      "  \"num_attention_heads\": 12,\n",
      "  \"num_channels\": 3,\n",
      "  \"num_frames\": 16,\n",
      "  \"num_hidden_layers\": 12,\n",
      "  \"patch_size\": 16,\n",
      "  \"problem_type\": \"single_label_classification\",\n",
      "  \"qkv_bias\": true,\n",
      "  \"torch_dtype\": \"float32\",\n",
      "  \"transformers_version\": \"4.24.0\",\n",
      "  \"tubelet_size\": 2,\n",
      "  \"use_mean_pooling\": true\n",
      "}\n",
      "\n",
      "loading weights file ./videomae-icsr-heart-v4/pytorch_model.bin\n",
      "All model checkpoint weights were used when initializing VideoMAEForVideoClassification.\n",
      "\n",
      "All the weights of VideoMAEForVideoClassification were initialized from the model checkpoint at ./videomae-icsr-heart-v4.\n",
      "If your task is similar to the task the model of the checkpoint was trained on, you can already use VideoMAEForVideoClassification for predictions without further training.\n"
     ]
    }
   ],
   "source": [
    "labels = names\n",
    "\n",
    "model = VideoMAEForVideoClassification.from_pretrained(\n",
    "    name_model,\n",
    "    num_labels=len(labels),\n",
    "    id2label={str(i): c for i, c in enumerate(labels)},\n",
    "    label2id={c: str(i) for i, c in enumerate(labels)},\n",
    "    ignore_mismatched_sizes=True,\n",
    "#     patch_size = 32,\n",
    "#     num_frames = 64\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "fc_norm.weight\n",
      "fc_norm.bias\n",
      "classifier.weight\n",
      "classifier.bias\n"
     ]
    }
   ],
   "source": [
    "for name, param in model.named_parameters():\n",
    "    if param.requires_grad:\n",
    "        if 'videomae' in name:\n",
    "            param.requires_grad = False\n",
    "            \n",
    "for name, param in model.named_parameters():\n",
    "    if param.requires_grad:\n",
    "        print(name)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 490,
     "status": "ok",
     "timestamp": 1667575094273,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "BmvmWI4a57xr",
    "outputId": "bafabebc-6568-4f83-acfd-14f46f7d57f0"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "PyTorch: setting up devices\n",
      "The default value for the training argument `--report_to` will change in v5 (from all installed integrations to none). In v5, you will need to use `--report_to all` to get the same behavior as now. You should start updating your code and make this info disappear :-).\n"
     ]
    }
   ],
   "source": [
    "from transformers import TrainingArguments\n",
    "\n",
    "training_args = TrainingArguments(\n",
    "  output_dir=\"./videomae-icsr-heart-v5\",\n",
    "  per_device_train_batch_size=16,\n",
    "  per_device_eval_batch_size=16,\n",
    "  evaluation_strategy=\"steps\",\n",
    "  num_train_epochs=48,\n",
    "  fp16=True,\n",
    "  save_steps=100,\n",
    "  eval_steps=100,\n",
    "  logging_steps=100,\n",
    "  weight_decay =0.05,\n",
    "  learning_rate=1e-4,\n",
    "  save_total_limit=5,\n",
    "  remove_unused_columns=False,\n",
    "  push_to_hub=False,\n",
    "  #report_to='tensorboard',\n",
    "  load_best_model_at_end=True,\n",
    "  optim=\"adamw_torch\"\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "# from transformers import TrainingArguments\n",
    "\n",
    "# training_args = TrainingArguments(\n",
    "#   output_dir=\"./videomae-icsr-heart\",\n",
    "#   per_device_train_batch_size=2,\n",
    "#   num_train_epochs=16,\n",
    "#   save_steps=198,\n",
    "#   logging_steps=198,\n",
    "#   save_total_limit=5,\n",
    "#   remove_unused_columns=False,\n",
    "# )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 378,
     "status": "ok",
     "timestamp": 1667575104288,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "WYSSr6k157xr",
    "outputId": "43f59bb8-5f1c-4d81-8f75-1ce809c54371"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using cuda_amp half precision backend\n"
     ]
    }
   ],
   "source": [
    "from transformers import Trainer\n",
    "\n",
    "trainer = Trainer(\n",
    "    model=model,\n",
    "    args=training_args,\n",
    "    data_collator=collate_with_path_batches,\n",
    "    compute_metrics=compute_metrics,\n",
    "    train_dataset=dataset,\n",
    "    eval_dataset=dataset_validation,\n",
    "    tokenizer=feature_extractor,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "id": "vfyxQlfx57xs"
   },
   "outputs": [],
   "source": [
    "torch.cuda.empty_cache()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "executionInfo": {
     "elapsed": 4359410,
     "status": "ok",
     "timestamp": 1667579468076,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "jxWG5dOA57xs",
    "outputId": "7cf52156-2971-4d4f-f8fc-dd4326645ea7"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "***** Running training *****\n",
      "  Num examples = 792\n",
      "  Num Epochs = 48\n",
      "  Instantaneous batch size per device = 16\n",
      "  Total train batch size (w. parallel, distributed & accumulation) = 16\n",
      "  Gradient Accumulation steps = 1\n",
      "  Total optimization steps = 2400\n",
      "  Number of trainable parameters = 16916\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "    <div>\n",
       "      \n",
       "      <progress value='2400' max='2400' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
       "      [2400/2400 1:10:31, Epoch 48/48]\n",
       "    </div>\n",
       "    <table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       " <tr style=\"text-align: left;\">\n",
       "      <th>Step</th>\n",
       "      <th>Training Loss</th>\n",
       "      <th>Validation Loss</th>\n",
       "      <th>Accuracy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>100</td>\n",
       "      <td>0.317400</td>\n",
       "      <td>0.989823</td>\n",
       "      <td>0.705882</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>200</td>\n",
       "      <td>0.299600</td>\n",
       "      <td>0.988221</td>\n",
       "      <td>0.713235</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>300</td>\n",
       "      <td>0.276200</td>\n",
       "      <td>0.982505</td>\n",
       "      <td>0.720588</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>400</td>\n",
       "      <td>0.266800</td>\n",
       "      <td>0.963977</td>\n",
       "      <td>0.713235</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>500</td>\n",
       "      <td>0.251600</td>\n",
       "      <td>0.966964</td>\n",
       "      <td>0.698529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>600</td>\n",
       "      <td>0.240900</td>\n",
       "      <td>0.958635</td>\n",
       "      <td>0.705882</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>700</td>\n",
       "      <td>0.232100</td>\n",
       "      <td>0.952057</td>\n",
       "      <td>0.705882</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>800</td>\n",
       "      <td>0.224700</td>\n",
       "      <td>0.939365</td>\n",
       "      <td>0.698529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>900</td>\n",
       "      <td>0.216100</td>\n",
       "      <td>0.938234</td>\n",
       "      <td>0.705882</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1000</td>\n",
       "      <td>0.208900</td>\n",
       "      <td>0.927718</td>\n",
       "      <td>0.705882</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1100</td>\n",
       "      <td>0.204400</td>\n",
       "      <td>0.913353</td>\n",
       "      <td>0.713235</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1200</td>\n",
       "      <td>0.198400</td>\n",
       "      <td>0.917358</td>\n",
       "      <td>0.713235</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1300</td>\n",
       "      <td>0.191700</td>\n",
       "      <td>0.915260</td>\n",
       "      <td>0.698529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1400</td>\n",
       "      <td>0.191400</td>\n",
       "      <td>0.923752</td>\n",
       "      <td>0.713235</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1500</td>\n",
       "      <td>0.184300</td>\n",
       "      <td>0.929873</td>\n",
       "      <td>0.698529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1600</td>\n",
       "      <td>0.181600</td>\n",
       "      <td>0.923015</td>\n",
       "      <td>0.698529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1700</td>\n",
       "      <td>0.176100</td>\n",
       "      <td>0.936443</td>\n",
       "      <td>0.691176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1800</td>\n",
       "      <td>0.173800</td>\n",
       "      <td>0.924801</td>\n",
       "      <td>0.691176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1900</td>\n",
       "      <td>0.175200</td>\n",
       "      <td>0.926001</td>\n",
       "      <td>0.705882</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2000</td>\n",
       "      <td>0.174100</td>\n",
       "      <td>0.926336</td>\n",
       "      <td>0.698529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2100</td>\n",
       "      <td>0.173100</td>\n",
       "      <td>0.923145</td>\n",
       "      <td>0.698529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2200</td>\n",
       "      <td>0.170000</td>\n",
       "      <td>0.920362</td>\n",
       "      <td>0.705882</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2300</td>\n",
       "      <td>0.167600</td>\n",
       "      <td>0.905428</td>\n",
       "      <td>0.691176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2400</td>\n",
       "      <td>0.168900</td>\n",
       "      <td>0.922342</td>\n",
       "      <td>0.705882</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table><p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "***** Running Evaluation *****\n",
      "  Num examples = 136\n",
      "  Batch size = 16\n",
      "Saving model checkpoint to ./videomae-icsr-heart-v5/checkpoint-100\n",
      "Configuration saved in ./videomae-icsr-heart-v5/checkpoint-100/config.json\n",
      "Model weights saved in ./videomae-icsr-heart-v5/checkpoint-100/pytorch_model.bin\n",
      "Feature extractor saved in ./videomae-icsr-heart-v5/checkpoint-100/preprocessor_config.json\n",
      "***** Running Evaluation *****\n",
      "  Num examples = 136\n",
      "  Batch size = 16\n",
      "Saving model checkpoint to ./videomae-icsr-heart-v5/checkpoint-200\n",
      "Configuration saved in ./videomae-icsr-heart-v5/checkpoint-200/config.json\n",
      "Model weights saved in ./videomae-icsr-heart-v5/checkpoint-200/pytorch_model.bin\n",
      "Feature extractor saved in ./videomae-icsr-heart-v5/checkpoint-200/preprocessor_config.json\n",
      "***** Running Evaluation *****\n",
      "  Num examples = 136\n",
      "  Batch size = 16\n",
      "Saving model checkpoint to ./videomae-icsr-heart-v5/checkpoint-300\n",
      "Configuration saved in ./videomae-icsr-heart-v5/checkpoint-300/config.json\n",
      "Model weights saved in ./videomae-icsr-heart-v5/checkpoint-300/pytorch_model.bin\n",
      "Feature extractor saved in ./videomae-icsr-heart-v5/checkpoint-300/preprocessor_config.json\n",
      "***** Running Evaluation *****\n",
      "  Num examples = 136\n",
      "  Batch size = 16\n",
      "Saving model checkpoint to ./videomae-icsr-heart-v5/checkpoint-400\n",
      "Configuration saved in ./videomae-icsr-heart-v5/checkpoint-400/config.json\n",
      "Model weights saved in ./videomae-icsr-heart-v5/checkpoint-400/pytorch_model.bin\n",
      "Feature extractor saved in ./videomae-icsr-heart-v5/checkpoint-400/preprocessor_config.json\n",
      "***** Running Evaluation *****\n",
      "  Num examples = 136\n",
      "  Batch size = 16\n",
      "Saving model checkpoint to ./videomae-icsr-heart-v5/checkpoint-500\n",
      "Configuration saved in ./videomae-icsr-heart-v5/checkpoint-500/config.json\n",
      "Model weights saved in ./videomae-icsr-heart-v5/checkpoint-500/pytorch_model.bin\n",
      "Feature extractor saved in ./videomae-icsr-heart-v5/checkpoint-500/preprocessor_config.json\n",
      "***** Running Evaluation *****\n",
      "  Num examples = 136\n",
      "  Batch size = 16\n",
      "Saving model checkpoint to ./videomae-icsr-heart-v5/checkpoint-600\n",
      "Configuration saved in ./videomae-icsr-heart-v5/checkpoint-600/config.json\n",
      "Model weights saved in ./videomae-icsr-heart-v5/checkpoint-600/pytorch_model.bin\n",
      "Feature extractor saved in ./videomae-icsr-heart-v5/checkpoint-600/preprocessor_config.json\n",
      "Deleting older checkpoint [videomae-icsr-heart-v5/checkpoint-100] due to args.save_total_limit\n",
      "***** Running Evaluation *****\n",
      "  Num examples = 136\n",
      "  Batch size = 16\n",
      "Saving model checkpoint to ./videomae-icsr-heart-v5/checkpoint-700\n",
      "Configuration saved in ./videomae-icsr-heart-v5/checkpoint-700/config.json\n",
      "Model weights saved in ./videomae-icsr-heart-v5/checkpoint-700/pytorch_model.bin\n",
      "Feature extractor saved in ./videomae-icsr-heart-v5/checkpoint-700/preprocessor_config.json\n",
      "Deleting older checkpoint [videomae-icsr-heart-v5/checkpoint-200] due to args.save_total_limit\n",
      "***** Running Evaluation *****\n",
      "  Num examples = 136\n",
      "  Batch size = 16\n",
      "Saving model checkpoint to ./videomae-icsr-heart-v5/checkpoint-800\n",
      "Configuration saved in ./videomae-icsr-heart-v5/checkpoint-800/config.json\n",
      "Model weights saved in ./videomae-icsr-heart-v5/checkpoint-800/pytorch_model.bin\n",
      "Feature extractor saved in ./videomae-icsr-heart-v5/checkpoint-800/preprocessor_config.json\n",
      "Deleting older checkpoint [videomae-icsr-heart-v5/checkpoint-300] due to args.save_total_limit\n",
      "***** Running Evaluation *****\n",
      "  Num examples = 136\n",
      "  Batch size = 16\n",
      "Saving model checkpoint to ./videomae-icsr-heart-v5/checkpoint-900\n",
      "Configuration saved in ./videomae-icsr-heart-v5/checkpoint-900/config.json\n",
      "Model weights saved in ./videomae-icsr-heart-v5/checkpoint-900/pytorch_model.bin\n",
      "Feature extractor saved in ./videomae-icsr-heart-v5/checkpoint-900/preprocessor_config.json\n",
      "Deleting older checkpoint [videomae-icsr-heart-v5/checkpoint-400] due to args.save_total_limit\n",
      "***** Running Evaluation *****\n",
      "  Num examples = 136\n",
      "  Batch size = 16\n",
      "Saving model checkpoint to ./videomae-icsr-heart-v5/checkpoint-1000\n",
      "Configuration saved in ./videomae-icsr-heart-v5/checkpoint-1000/config.json\n",
      "Model weights saved in ./videomae-icsr-heart-v5/checkpoint-1000/pytorch_model.bin\n",
      "Feature extractor saved in ./videomae-icsr-heart-v5/checkpoint-1000/preprocessor_config.json\n",
      "Deleting older checkpoint [videomae-icsr-heart-v5/checkpoint-500] due to args.save_total_limit\n",
      "***** Running Evaluation *****\n",
      "  Num examples = 136\n",
      "  Batch size = 16\n",
      "Saving model checkpoint to ./videomae-icsr-heart-v5/checkpoint-1100\n",
      "Configuration saved in ./videomae-icsr-heart-v5/checkpoint-1100/config.json\n",
      "Model weights saved in ./videomae-icsr-heart-v5/checkpoint-1100/pytorch_model.bin\n",
      "Feature extractor saved in ./videomae-icsr-heart-v5/checkpoint-1100/preprocessor_config.json\n",
      "Deleting older checkpoint [videomae-icsr-heart-v5/checkpoint-600] due to args.save_total_limit\n",
      "***** Running Evaluation *****\n",
      "  Num examples = 136\n",
      "  Batch size = 16\n",
      "Saving model checkpoint to ./videomae-icsr-heart-v5/checkpoint-1200\n",
      "Configuration saved in ./videomae-icsr-heart-v5/checkpoint-1200/config.json\n",
      "Model weights saved in ./videomae-icsr-heart-v5/checkpoint-1200/pytorch_model.bin\n",
      "Feature extractor saved in ./videomae-icsr-heart-v5/checkpoint-1200/preprocessor_config.json\n",
      "Deleting older checkpoint [videomae-icsr-heart-v5/checkpoint-700] due to args.save_total_limit\n",
      "***** Running Evaluation *****\n",
      "  Num examples = 136\n",
      "  Batch size = 16\n",
      "Saving model checkpoint to ./videomae-icsr-heart-v5/checkpoint-1300\n",
      "Configuration saved in ./videomae-icsr-heart-v5/checkpoint-1300/config.json\n",
      "Model weights saved in ./videomae-icsr-heart-v5/checkpoint-1300/pytorch_model.bin\n",
      "Feature extractor saved in ./videomae-icsr-heart-v5/checkpoint-1300/preprocessor_config.json\n",
      "Deleting older checkpoint [videomae-icsr-heart-v5/checkpoint-800] due to args.save_total_limit\n",
      "***** Running Evaluation *****\n",
      "  Num examples = 136\n",
      "  Batch size = 16\n",
      "Saving model checkpoint to ./videomae-icsr-heart-v5/checkpoint-1400\n",
      "Configuration saved in ./videomae-icsr-heart-v5/checkpoint-1400/config.json\n",
      "Model weights saved in ./videomae-icsr-heart-v5/checkpoint-1400/pytorch_model.bin\n",
      "Feature extractor saved in ./videomae-icsr-heart-v5/checkpoint-1400/preprocessor_config.json\n",
      "Deleting older checkpoint [videomae-icsr-heart-v5/checkpoint-900] due to args.save_total_limit\n",
      "***** Running Evaluation *****\n",
      "  Num examples = 136\n",
      "  Batch size = 16\n",
      "Saving model checkpoint to ./videomae-icsr-heart-v5/checkpoint-1500\n",
      "Configuration saved in ./videomae-icsr-heart-v5/checkpoint-1500/config.json\n",
      "Model weights saved in ./videomae-icsr-heart-v5/checkpoint-1500/pytorch_model.bin\n",
      "Feature extractor saved in ./videomae-icsr-heart-v5/checkpoint-1500/preprocessor_config.json\n",
      "Deleting older checkpoint [videomae-icsr-heart-v5/checkpoint-1000] due to args.save_total_limit\n",
      "***** Running Evaluation *****\n",
      "  Num examples = 136\n",
      "  Batch size = 16\n",
      "Saving model checkpoint to ./videomae-icsr-heart-v5/checkpoint-1600\n",
      "Configuration saved in ./videomae-icsr-heart-v5/checkpoint-1600/config.json\n",
      "Model weights saved in ./videomae-icsr-heart-v5/checkpoint-1600/pytorch_model.bin\n",
      "Feature extractor saved in ./videomae-icsr-heart-v5/checkpoint-1600/preprocessor_config.json\n",
      "Deleting older checkpoint [videomae-icsr-heart-v5/checkpoint-1200] due to args.save_total_limit\n",
      "***** Running Evaluation *****\n",
      "  Num examples = 136\n",
      "  Batch size = 16\n",
      "Saving model checkpoint to ./videomae-icsr-heart-v5/checkpoint-1700\n",
      "Configuration saved in ./videomae-icsr-heart-v5/checkpoint-1700/config.json\n",
      "Model weights saved in ./videomae-icsr-heart-v5/checkpoint-1700/pytorch_model.bin\n",
      "Feature extractor saved in ./videomae-icsr-heart-v5/checkpoint-1700/preprocessor_config.json\n",
      "Deleting older checkpoint [videomae-icsr-heart-v5/checkpoint-1300] due to args.save_total_limit\n",
      "***** Running Evaluation *****\n",
      "  Num examples = 136\n",
      "  Batch size = 16\n",
      "Saving model checkpoint to ./videomae-icsr-heart-v5/checkpoint-1800\n",
      "Configuration saved in ./videomae-icsr-heart-v5/checkpoint-1800/config.json\n",
      "Model weights saved in ./videomae-icsr-heart-v5/checkpoint-1800/pytorch_model.bin\n",
      "Feature extractor saved in ./videomae-icsr-heart-v5/checkpoint-1800/preprocessor_config.json\n",
      "Deleting older checkpoint [videomae-icsr-heart-v5/checkpoint-1400] due to args.save_total_limit\n",
      "***** Running Evaluation *****\n",
      "  Num examples = 136\n",
      "  Batch size = 16\n",
      "Saving model checkpoint to ./videomae-icsr-heart-v5/checkpoint-1900\n",
      "Configuration saved in ./videomae-icsr-heart-v5/checkpoint-1900/config.json\n",
      "Model weights saved in ./videomae-icsr-heart-v5/checkpoint-1900/pytorch_model.bin\n",
      "Feature extractor saved in ./videomae-icsr-heart-v5/checkpoint-1900/preprocessor_config.json\n",
      "Deleting older checkpoint [videomae-icsr-heart-v5/checkpoint-1500] due to args.save_total_limit\n",
      "***** Running Evaluation *****\n",
      "  Num examples = 136\n",
      "  Batch size = 16\n",
      "Saving model checkpoint to ./videomae-icsr-heart-v5/checkpoint-2000\n",
      "Configuration saved in ./videomae-icsr-heart-v5/checkpoint-2000/config.json\n",
      "Model weights saved in ./videomae-icsr-heart-v5/checkpoint-2000/pytorch_model.bin\n",
      "Feature extractor saved in ./videomae-icsr-heart-v5/checkpoint-2000/preprocessor_config.json\n",
      "Deleting older checkpoint [videomae-icsr-heart-v5/checkpoint-1600] due to args.save_total_limit\n",
      "***** Running Evaluation *****\n",
      "  Num examples = 136\n",
      "  Batch size = 16\n",
      "Saving model checkpoint to ./videomae-icsr-heart-v5/checkpoint-2100\n",
      "Configuration saved in ./videomae-icsr-heart-v5/checkpoint-2100/config.json\n",
      "Model weights saved in ./videomae-icsr-heart-v5/checkpoint-2100/pytorch_model.bin\n",
      "Feature extractor saved in ./videomae-icsr-heart-v5/checkpoint-2100/preprocessor_config.json\n",
      "Deleting older checkpoint [videomae-icsr-heart-v5/checkpoint-1700] due to args.save_total_limit\n",
      "***** Running Evaluation *****\n",
      "  Num examples = 136\n",
      "  Batch size = 16\n",
      "Saving model checkpoint to ./videomae-icsr-heart-v5/checkpoint-2200\n",
      "Configuration saved in ./videomae-icsr-heart-v5/checkpoint-2200/config.json\n",
      "Model weights saved in ./videomae-icsr-heart-v5/checkpoint-2200/pytorch_model.bin\n",
      "Feature extractor saved in ./videomae-icsr-heart-v5/checkpoint-2200/preprocessor_config.json\n",
      "Deleting older checkpoint [videomae-icsr-heart-v5/checkpoint-1800] due to args.save_total_limit\n",
      "***** Running Evaluation *****\n",
      "  Num examples = 136\n",
      "  Batch size = 16\n",
      "Saving model checkpoint to ./videomae-icsr-heart-v5/checkpoint-2300\n",
      "Configuration saved in ./videomae-icsr-heart-v5/checkpoint-2300/config.json\n",
      "Model weights saved in ./videomae-icsr-heart-v5/checkpoint-2300/pytorch_model.bin\n",
      "Feature extractor saved in ./videomae-icsr-heart-v5/checkpoint-2300/preprocessor_config.json\n",
      "Deleting older checkpoint [videomae-icsr-heart-v5/checkpoint-1100] due to args.save_total_limit\n",
      "***** Running Evaluation *****\n",
      "  Num examples = 136\n",
      "  Batch size = 16\n",
      "Saving model checkpoint to ./videomae-icsr-heart-v5/checkpoint-2400\n",
      "Configuration saved in ./videomae-icsr-heart-v5/checkpoint-2400/config.json\n",
      "Model weights saved in ./videomae-icsr-heart-v5/checkpoint-2400/pytorch_model.bin\n",
      "Feature extractor saved in ./videomae-icsr-heart-v5/checkpoint-2400/preprocessor_config.json\n",
      "Deleting older checkpoint [videomae-icsr-heart-v5/checkpoint-1900] due to args.save_total_limit\n",
      "\n",
      "\n",
      "Training completed. Do not forget to share your model on huggingface.co/models =)\n",
      "\n",
      "\n",
      "Loading best model from ./videomae-icsr-heart-v5/checkpoint-2300 (score: 0.9054275751113892).\n"
     ]
    }
   ],
   "source": [
    "train_results = trainer.train()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Saving the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 1433,
     "status": "ok",
     "timestamp": 1667579469478,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "sz6jG1Oi57xs",
    "outputId": "c0c43c9f-1371-4bcf-a8fd-ef3e26c79c59",
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Saving model checkpoint to ./videomae-icsr-heart-v5\n",
      "Configuration saved in ./videomae-icsr-heart-v5/config.json\n",
      "Model weights saved in ./videomae-icsr-heart-v5/pytorch_model.bin\n",
      "Feature extractor saved in ./videomae-icsr-heart-v5/preprocessor_config.json\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "***** train metrics *****\n",
      "  epoch                    =          48.0\n",
      "  total_flos               = 44124196877GF\n",
      "  train_loss               =         0.211\n",
      "  train_runtime            =    1:10:33.49\n",
      "  train_samples_per_second =          8.98\n",
      "  train_steps_per_second   =         0.567\n"
     ]
    }
   ],
   "source": [
    "trainer.save_model()\n",
    "trainer.log_metrics(\"train\", train_results.metrics)\n",
    "trainer.save_metrics(\"train\", train_results.metrics)\n",
    "trainer.save_state()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 211
    },
    "executionInfo": {
     "elapsed": 38754,
     "status": "ok",
     "timestamp": 1667579508228,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "UWK9TSk457xs",
    "outputId": "40156d0d-7f99-4068-b73b-801f5c039f2c"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "***** Running Evaluation *****\n",
      "  Num examples = 136\n",
      "  Batch size = 16\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "    <div>\n",
       "      \n",
       "      <progress value='9' max='9' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
       "      [9/9 00:12]\n",
       "    </div>\n",
       "    "
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "***** eval metrics *****\n",
      "  epoch                   =       48.0\n",
      "  eval_accuracy           =     0.6985\n",
      "  eval_loss               =     0.9059\n",
      "  eval_runtime            = 0:00:13.63\n",
      "  eval_samples_per_second =      9.976\n",
      "  eval_steps_per_second   =       0.66\n"
     ]
    }
   ],
   "source": [
    "metrics = trainer.evaluate(dataset_validation)\n",
    "trainer.log_metrics(\"eval\", metrics)\n",
    "trainer.save_metrics(\"eval\", metrics)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 187,
     "status": "ok",
     "timestamp": 1667426738132,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "AZXFxB-LQQ-7",
    "outputId": "5d2fa64a-bde2-4d04-cf87-eede2f5ad788"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'eval_loss': 0.9059243202209473,\n",
       " 'eval_accuracy': 0.6985294117647058,\n",
       " 'eval_runtime': 13.6332,\n",
       " 'eval_samples_per_second': 9.976,\n",
       " 'eval_steps_per_second': 0.66,\n",
       " 'epoch': 48.0}"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "executionInfo": {
     "elapsed": 1134,
     "status": "ok",
     "timestamp": 1667579509321,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "j4yIILWEt-1p"
   },
   "outputs": [],
   "source": [
    "kwargs = {\n",
    "    \"finetuned_from\": model.config._name_or_path,\n",
    "    \"tasks\": \"video-classification\",\n",
    "    \"dataset\": 'HEART-MET',\n",
    "    \"tags\": ['video-classification'],\n",
    "}\n",
    "\n",
    "trainer.create_model_card(**kwargs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "executionInfo": {
     "elapsed": 15981,
     "status": "ok",
     "timestamp": 1667579525293,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "Ev4EBAIBAdZa"
   },
   "outputs": [],
   "source": [
    "## only for google colab\n",
    "#!cp -R /content/videomae-icsr-heart/ /content/drive/MyDrive/ICSR 2022 HEART-MET Activity Recognition Challenge/models\n",
    "# !cp -R /content/videomae-icsr-heart/ /content/drive/MyDrive/models/"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "3QjKR9fwlq5h"
   },
   "source": [
    "## Testing (with validation)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 51665,
     "status": "ok",
     "timestamp": 1667582462627,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "yQGkGkfqmm4q",
    "outputId": "d5778354-97ff-44ac-8802-1ea55fa5dce4"
   },
   "outputs": [],
   "source": [
    "i=0\n",
    "predicted=[]\n",
    "\n",
    "for data in full_augmented_set['validation']:\n",
    "  #if i==2:\n",
    "  #  break\n",
    "  i+=1\n",
    "  videoreader = VideoReader(data['file_name'], num_threads=1, ctx=cpu(0))\n",
    "  videoreader.seek(0)\n",
    "  #indices = sample_frame_indices(clip_len=16, frame_sample_rate=4, seg_len=len(videoreader))\n",
    "  indices = np.linspace(0, len(videoreader), num=16)\n",
    "  indices = np.clip(indices, 0, len(videoreader)- 1).astype(np.int64)\n",
    "  pixel_values = feature_extractor(list(videoreader.get_batch(indices).asnumpy()), return_tensors=\"pt\")\n",
    "  pixel_values = pixel_values.to('cuda')\n",
    "  with torch.no_grad():\n",
    "        outputs = model(**pixel_values)#\n",
    "        pred=outputs.logits.argmax(-1).item()\n",
    "        predicted.append([data['file_name'],data['labels'],data['description'],pred])\n",
    "        #eval_results[] =\n",
    "        #print(data['labels'],pred)\n",
    "  \n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 424
    },
    "executionInfo": {
     "elapsed": 370,
     "status": "ok",
     "timestamp": 1667582567088,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "xJwAm11IqFzd",
    "outputId": "e2cd5c3d-3d4f-479e-cbd7-e5caded5d79e"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<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>file</th>\n",
       "      <th>label</th>\n",
       "      <th>description</th>\n",
       "      <th>predicted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>validation_set/video0000.mp4</td>\n",
       "      <td>8</td>\n",
       "      <td>Wiping a table</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>validation_set/video0001.mp4</td>\n",
       "      <td>18</td>\n",
       "      <td>Writing</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>validation_set/video0002.mp4</td>\n",
       "      <td>10</td>\n",
       "      <td>Neck roll exercise</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>validation_set/video0003.mp4</td>\n",
       "      <td>7</td>\n",
       "      <td>Coughing or sneezing</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>validation_set/video0004.mp4</td>\n",
       "      <td>18</td>\n",
       "      <td>Writing</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131</th>\n",
       "      <td>validation_set/video0131.mp4</td>\n",
       "      <td>10</td>\n",
       "      <td>Neck roll exercise</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>132</th>\n",
       "      <td>validation_set/video0132.mp4</td>\n",
       "      <td>14</td>\n",
       "      <td>Talking on the phone</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>133</th>\n",
       "      <td>validation_set/video0133.mp4</td>\n",
       "      <td>2</td>\n",
       "      <td>Touching a hot surface</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>134</th>\n",
       "      <td>validation_set/video0134.mp4</td>\n",
       "      <td>9</td>\n",
       "      <td>Reading a book</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>135</th>\n",
       "      <td>validation_set/video0135.mp4</td>\n",
       "      <td>13</td>\n",
       "      <td>Limping</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>136 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                             file  label             description  predicted\n",
       "0    validation_set/video0000.mp4      8          Wiping a table          8\n",
       "1    validation_set/video0001.mp4     18                 Writing         18\n",
       "2    validation_set/video0002.mp4     10      Neck roll exercise         10\n",
       "3    validation_set/video0003.mp4      7    Coughing or sneezing          4\n",
       "4    validation_set/video0004.mp4     18                 Writing         18\n",
       "..                            ...    ...                     ...        ...\n",
       "131  validation_set/video0131.mp4     10      Neck roll exercise          4\n",
       "132  validation_set/video0132.mp4     14    Talking on the phone         14\n",
       "133  validation_set/video0133.mp4      2  Touching a hot surface          2\n",
       "134  validation_set/video0134.mp4      9          Reading a book          9\n",
       "135  validation_set/video0135.mp4     13                 Limping         16\n",
       "\n",
       "[136 rows x 4 columns]"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "df = pd.DataFrame(predicted,columns=['file','label','description','predicted'])\n",
    "\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 357,
     "status": "ok",
     "timestamp": 1667582611044,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "9ekeTxX5qsQ6",
    "outputId": "50260d97-65d5-46b0-e325-22952daf0302"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                     precision    recall  f1-score   support\n",
      "\n",
      "Opening the door and walking in/out       0.83      0.83      0.83         6\n",
      "                Putting on a jacket       0.80      0.80      0.80         5\n",
      "             Touching a hot surface       0.50      0.50      0.50         6\n",
      "                 Opening the fridge       0.67      0.86      0.75         7\n",
      "                     Drinking water       0.60      0.90      0.72        10\n",
      "        Colliding against something       0.50      0.57      0.53         7\n",
      "            Eating food with a fork       0.64      0.88      0.74         8\n",
      "               Coughing or sneezing       0.25      0.33      0.29         6\n",
      "                     Wiping a table       1.00      0.50      0.67         6\n",
      "                     Reading a book       1.00      0.75      0.86         8\n",
      "                 Neck roll exercise       0.86      0.86      0.86         7\n",
      "                  Freehand exercise       0.00      0.00      0.00         7\n",
      "                         Lying down       0.33      0.33      0.33         3\n",
      "                            Limping       0.60      0.43      0.50         7\n",
      "               Talking on the phone       0.67      0.50      0.57         8\n",
      "                   Using a computer       0.89      1.00      0.94         8\n",
      "                       Falling down       0.58      1.00      0.74         7\n",
      "                     Brushing teeth       1.00      1.00      1.00         8\n",
      "                            Writing       1.00      0.88      0.93         8\n",
      "                   Unknown activity       0.00      0.00      0.00         4\n",
      "\n",
      "                           accuracy                           0.68       136\n",
      "                          macro avg       0.64      0.65      0.63       136\n",
      "                       weighted avg       0.67      0.68      0.66       136\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/tomas/.local/lib/python3.8/site-packages/sklearn/metrics/_classification.py:1334: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/tomas/.local/lib/python3.8/site-packages/sklearn/metrics/_classification.py:1334: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/tomas/.local/lib/python3.8/site-packages/sklearn/metrics/_classification.py:1334: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "\n",
    "print(classification_report(df['label'], df['predicted'], target_names=names))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 473
    },
    "executionInfo": {
     "elapsed": 2996,
     "status": "ok",
     "timestamp": 1667582709066,
     "user": {
      "displayName": "Olmer Garcia",
      "userId": "02129671575696093186"
     },
     "user_tz": 300
    },
    "id": "9s-30mn9rE4c",
    "outputId": "6c7b766b-6ad7-4914-fc5a-43ec39eea6ef"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay at 0x7fa07fb46c10>"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay,ConfusionMatrixDisplay\n",
    "cm = confusion_matrix(df['label'], df['predicted'])\n",
    "disp = ConfusionMatrixDisplay(confusion_matrix=cm,display_labels=names)\n",
    "disp.plot(xticks_rotation='vertical')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.to_csv(\"tes_fast_motion.csv\",index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "eval_results = {}\n",
    "for row in predicted:\n",
    "    eval_results[row[0]]=\"[\"+str(row[-1])+\"]\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(\"validation_result_o.json\", \"w\") as outfile:\n",
    "    json.dump(eval_results, outfile)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Test data\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "test_set/video0000.mp4\n",
      "test_set/video0001.mp4\n",
      "test_set/video0002.mp4\n",
      "test_set/video0003.mp4\n",
      "test_set/video0004.mp4\n",
      "test_set/video0005.mp4\n",
      "test_set/video0006.mp4\n",
      "test_set/video0007.mp4\n",
      "test_set/video0008.mp4\n",
      "test_set/video0009.mp4\n",
      "test_set/video0010.mp4\n",
      "test_set/video0011.mp4\n",
      "test_set/video0012.mp4\n",
      "test_set/video0013.mp4\n",
      "test_set/video0014.mp4\n",
      "test_set/video0015.mp4\n",
      "test_set/video0016.mp4\n",
      "test_set/video0017.mp4\n",
      "test_set/video0018.mp4\n",
      "test_set/video0019.mp4\n",
      "test_set/video0020.mp4\n",
      "test_set/video0021.mp4\n",
      "test_set/video0022.mp4\n",
      "test_set/video0023.mp4\n",
      "test_set/video0024.mp4\n",
      "test_set/video0025.mp4\n",
      "test_set/video0026.mp4\n",
      "test_set/video0027.mp4\n",
      "test_set/video0028.mp4\n",
      "test_set/video0029.mp4\n",
      "test_set/video0030.mp4\n",
      "test_set/video0031.mp4\n",
      "test_set/video0032.mp4\n",
      "test_set/video0033.mp4\n",
      "test_set/video0034.mp4\n",
      "test_set/video0035.mp4\n",
      "test_set/video0036.mp4\n",
      "test_set/video0037.mp4\n",
      "test_set/video0038.mp4\n",
      "test_set/video0039.mp4\n",
      "test_set/video0040.mp4\n",
      "test_set/video0041.mp4\n",
      "test_set/video0042.mp4\n",
      "test_set/video0043.mp4\n",
      "test_set/video0044.mp4\n",
      "test_set/video0045.mp4\n",
      "test_set/video0046.mp4\n",
      "test_set/video0047.mp4\n",
      "test_set/video0048.mp4\n",
      "test_set/video0049.mp4\n",
      "test_set/video0050.mp4\n",
      "test_set/video0051.mp4\n",
      "test_set/video0052.mp4\n",
      "test_set/video0053.mp4\n",
      "test_set/video0054.mp4\n",
      "test_set/video0055.mp4\n",
      "test_set/video0056.mp4\n",
      "test_set/video0057.mp4\n",
      "test_set/video0058.mp4\n",
      "test_set/video0059.mp4\n",
      "test_set/video0060.mp4\n",
      "test_set/video0061.mp4\n",
      "test_set/video0062.mp4\n",
      "test_set/video0063.mp4\n",
      "test_set/video0064.mp4\n",
      "test_set/video0065.mp4\n",
      "test_set/video0066.mp4\n",
      "test_set/video0067.mp4\n",
      "test_set/video0068.mp4\n",
      "test_set/video0069.mp4\n",
      "test_set/video0070.mp4\n",
      "test_set/video0071.mp4\n",
      "test_set/video0072.mp4\n",
      "test_set/video0073.mp4\n",
      "test_set/video0074.mp4\n",
      "test_set/video0075.mp4\n",
      "test_set/video0076.mp4\n",
      "test_set/video0077.mp4\n",
      "test_set/video0078.mp4\n",
      "test_set/video0079.mp4\n",
      "test_set/video0080.mp4\n",
      "test_set/video0081.mp4\n",
      "test_set/video0082.mp4\n",
      "test_set/video0083.mp4\n",
      "test_set/video0084.mp4\n",
      "test_set/video0085.mp4\n",
      "test_set/video0086.mp4\n",
      "test_set/video0087.mp4\n",
      "test_set/video0088.mp4\n",
      "test_set/video0089.mp4\n",
      "test_set/video0090.mp4\n",
      "test_set/video0091.mp4\n",
      "test_set/video0092.mp4\n",
      "test_set/video0093.mp4\n",
      "test_set/video0094.mp4\n",
      "test_set/video0095.mp4\n",
      "test_set/video0096.mp4\n",
      "test_set/video0097.mp4\n",
      "test_set/video0098.mp4\n",
      "test_set/video0099.mp4\n",
      "test_set/video0100.mp4\n",
      "test_set/video0101.mp4\n",
      "test_set/video0102.mp4\n",
      "test_set/video0103.mp4\n",
      "test_set/video0104.mp4\n",
      "test_set/video0105.mp4\n",
      "test_set/video0106.mp4\n",
      "test_set/video0107.mp4\n",
      "test_set/video0108.mp4\n",
      "test_set/video0109.mp4\n",
      "test_set/video0110.mp4\n",
      "test_set/video0111.mp4\n",
      "test_set/video0112.mp4\n",
      "test_set/video0113.mp4\n",
      "test_set/video0114.mp4\n",
      "test_set/video0115.mp4\n",
      "test_set/video0116.mp4\n",
      "test_set/video0117.mp4\n",
      "test_set/video0118.mp4\n",
      "test_set/video0119.mp4\n",
      "test_set/video0120.mp4\n",
      "test_set/video0121.mp4\n",
      "test_set/video0122.mp4\n",
      "test_set/video0123.mp4\n",
      "test_set/video0124.mp4\n",
      "test_set/video0125.mp4\n",
      "test_set/video0126.mp4\n",
      "test_set/video0127.mp4\n",
      "test_set/video0128.mp4\n",
      "test_set/video0129.mp4\n",
      "test_set/video0130.mp4\n",
      "test_set/video0131.mp4\n",
      "test_set/video0132.mp4\n",
      "test_set/video0133.mp4\n",
      "test_set/video0134.mp4\n",
      "test_set/video0135.mp4\n",
      "test_set/video0136.mp4\n",
      "test_set/video0137.mp4\n",
      "test_set/video0138.mp4\n",
      "test_set/video0139.mp4\n",
      "test_set/video0140.mp4\n",
      "test_set/video0141.mp4\n",
      "test_set/video0142.mp4\n",
      "test_set/video0143.mp4\n",
      "test_set/video0144.mp4\n",
      "test_set/video0145.mp4\n",
      "test_set/video0146.mp4\n",
      "test_set/video0147.mp4\n",
      "test_set/video0148.mp4\n",
      "test_set/video0149.mp4\n",
      "test_set/video0150.mp4\n",
      "test_set/video0151.mp4\n",
      "test_set/video0152.mp4\n",
      "test_set/video0153.mp4\n",
      "test_set/video0154.mp4\n",
      "test_set/video0155.mp4\n",
      "test_set/video0156.mp4\n",
      "test_set/video0157.mp4\n",
      "test_set/video0158.mp4\n",
      "test_set/video0159.mp4\n",
      "test_set/video0160.mp4\n",
      "test_set/video0161.mp4\n",
      "test_set/video0162.mp4\n",
      "test_set/video0163.mp4\n",
      "test_set/video0164.mp4\n",
      "test_set/video0165.mp4\n",
      "test_set/video0166.mp4\n",
      "test_set/video0167.mp4\n",
      "test_set/video0168.mp4\n",
      "test_set/video0169.mp4\n",
      "test_set/video0170.mp4\n",
      "test_set/video0171.mp4\n",
      "test_set/video0172.mp4\n",
      "test_set/video0173.mp4\n",
      "test_set/video0174.mp4\n",
      "test_set/video0175.mp4\n",
      "test_set/video0176.mp4\n",
      "test_set/video0177.mp4\n",
      "test_set/video0178.mp4\n",
      "test_set/video0179.mp4\n",
      "test_set/video0180.mp4\n",
      "test_set/video0181.mp4\n",
      "test_set/video0182.mp4\n",
      "test_set/video0183.mp4\n",
      "test_set/video0184.mp4\n",
      "test_set/video0185.mp4\n",
      "test_set/video0186.mp4\n",
      "test_set/video0187.mp4\n",
      "test_set/video0188.mp4\n",
      "test_set/video0189.mp4\n",
      "test_set/video0190.mp4\n",
      "test_set/video0191.mp4\n",
      "test_set/video0192.mp4\n",
      "test_set/video0193.mp4\n",
      "test_set/video0194.mp4\n",
      "test_set/video0195.mp4\n",
      "test_set/video0196.mp4\n",
      "test_set/video0197.mp4\n",
      "test_set/video0198.mp4\n",
      "test_set/video0199.mp4\n",
      "test_set/video0200.mp4\n",
      "test_set/video0201.mp4\n",
      "test_set/video0202.mp4\n",
      "test_set/video0203.mp4\n",
      "test_set/video0204.mp4\n",
      "test_set/video0205.mp4\n",
      "test_set/video0206.mp4\n",
      "test_set/video0207.mp4\n",
      "test_set/video0208.mp4\n",
      "test_set/video0209.mp4\n",
      "test_set/video0210.mp4\n",
      "test_set/video0211.mp4\n",
      "test_set/video0212.mp4\n",
      "test_set/video0213.mp4\n",
      "test_set/video0214.mp4\n",
      "test_set/video0215.mp4\n",
      "test_set/video0216.mp4\n",
      "test_set/video0217.mp4\n",
      "test_set/video0218.mp4\n",
      "test_set/video0219.mp4\n",
      "test_set/video0220.mp4\n",
      "test_set/video0221.mp4\n",
      "test_set/video0222.mp4\n",
      "test_set/video0223.mp4\n",
      "test_set/video0224.mp4\n",
      "test_set/video0225.mp4\n",
      "test_set/video0226.mp4\n",
      "test_set/video0227.mp4\n",
      "test_set/video0228.mp4\n",
      "test_set/video0229.mp4\n",
      "test_set/video0230.mp4\n",
      "test_set/video0231.mp4\n",
      "test_set/video0232.mp4\n",
      "test_set/video0233.mp4\n",
      "test_set/video0234.mp4\n",
      "test_set/video0235.mp4\n",
      "test_set/video0236.mp4\n",
      "test_set/video0237.mp4\n",
      "test_set/video0238.mp4\n",
      "test_set/video0239.mp4\n",
      "test_set/video0240.mp4\n",
      "test_set/video0241.mp4\n",
      "test_set/video0242.mp4\n",
      "test_set/video0243.mp4\n",
      "test_set/video0244.mp4\n",
      "test_set/video0245.mp4\n",
      "test_set/video0246.mp4\n",
      "test_set/video0247.mp4\n",
      "test_set/video0248.mp4\n",
      "test_set/video0249.mp4\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import glob\n",
    "videos = sorted(glob.glob('test_set' + '/*.mp4'))\n",
    "\n",
    "i=0\n",
    "predicted=[]\n",
    "\n",
    "for video in videos:\n",
    "  #if i==2:\n",
    "  #  break\n",
    "  i+=1\n",
    "  print(video)\n",
    "  try:\n",
    "      videoreader = VideoReader(video, num_threads=1, ctx=cpu(0))\n",
    "      videoreader.seek(0)\n",
    "      #indices = sample_frame_indices(clip_len=16, frame_sample_rate=4, seg_len=len(videoreader))\n",
    "      n_frames = 16        \n",
    "      initial_index = 0 #np.random.randint(0, n_frames)\n",
    "      indices = np.linspace(initial_index, len(videoreader), num=n_frames)\n",
    "      indices = np.clip(indices, 0, len(videoreader)- 1).astype(np.int64)\n",
    "      pixel_values = feature_extractor(list(videoreader.get_batch(indices).asnumpy()), return_tensors=\"pt\")\n",
    "      pixel_values = pixel_values.to('cuda')\n",
    "      with torch.no_grad():\n",
    "            outputs = model(**pixel_values)#\n",
    "            pred=outputs.logits.argmax(-1).item()\n",
    "            predicted.append([video,pred])\n",
    "  except:\n",
    "      predicted.append([video,19])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "## TODO: I remove by hand the folder and the \"\" \n",
    "eval_results = {}\n",
    "for row in predicted:\n",
    "    eval_results[row[0]]=\"[\"+str(row[-1])+\"]\"\n",
    "    \n",
    "with open(\"test_result_o.json\", \"w\") as outfile:\n",
    "    json.dump(eval_results, outfile)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "collapsed_sections": [],
   "provenance": []
  },
  "gpuClass": "standard",
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.8.10"
  },
  "vscode": {
   "interpreter": {
    "hash": "8faf4a258195431f720f804014a2ceef0cea3baee1483e687b2179ebac4a2d68"
   }
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "017e3f0a45c942e29b0fc432dca2ffc8": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_55742a800669463d81d67456044285cb",
      "placeholder": "​",
      "style": "IPY_MODEL_77d5ea8f038c4dd48d16cd885b00629a",
      "value": "Downloading: 100%"
     }
    },
    "028465bf497d4737a5384f6f7c58200a": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_462f437cc6144a7982a6dc89bd6f2a90",
      "placeholder": "​",
      "style": "IPY_MODEL_21e17f19d8c84ababf07dd94864f53e5",
      "value": "Downloading data files: 100%"
     }
    },
    "03b78f48e2ba4253b34d0946d5b40e15": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "045cd9695f3e479fb21048b39abfc18b": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "08301e86bd1745f28af465f1b9f9d050": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "09e0443823d54f578af5cf635b7b3715": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_ae9b118bc353457ab8e6c21208a8d57b",
       "IPY_MODEL_bfe6a0d505be48b888ea13cdc0d56447",
       "IPY_MODEL_bcea670e5b234125a4d801fafe8928f7"
      ],
      "layout": "IPY_MODEL_045cd9695f3e479fb21048b39abfc18b"
     }
    },
    "0dd258e5733e4584a0031bfa3a1e9392": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "0e8d7313d9f24c49821bb81bb013a925": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_5100f2875c724d10b1801c55084c33ab",
      "placeholder": "​",
      "style": "IPY_MODEL_32ea79387ae845cea889c436ffabd9e1",
      "value": " 1/1 [00:00&lt;00:00, 17.77it/s]"
     }
    },
    "107b16ef81554781856cd67a6ae038da": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "1144746deacc4c07aa06c81906f6ce77": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_4a2741e1b02749f5887bf798ee8e21d6",
      "placeholder": "​",
      "style": "IPY_MODEL_52e6c6ed71eb427b88df2e3fbf5b87c1",
      "value": ""
     }
    },
    "13d7e978cba64a3cbb9bd0e9bb50ecbc": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "1649a5747b8c48e58c4508a9906fea8f": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_0dd258e5733e4584a0031bfa3a1e9392",
      "placeholder": "​",
      "style": "IPY_MODEL_7e251276f132481e886901a9c0fa5562",
      "value": "Downloading data files: 100%"
     }
    },
    "17b6f0610cee47e1a33f2b54edc3631e": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_1a8775c21a7f466c944c790277a34326",
      "placeholder": "​",
      "style": "IPY_MODEL_9b00cff2e12c4886b3f6cf1e6879d725",
      "value": " 22.9k/22.9k [00:00&lt;00:00, 62.7kB/s]"
     }
    },
    "1a8775c21a7f466c944c790277a34326": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "1af7e067011a4854a1e5f97624b230cd": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "1f6a6b3791e54b56bb5b7f2f0ba812d9": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_1649a5747b8c48e58c4508a9906fea8f",
       "IPY_MODEL_e2eafd1b4cfa4965bf05c9caa0ea4595",
       "IPY_MODEL_e3ac5081836d40f797a5305f1429175e"
      ],
      "layout": "IPY_MODEL_5ebfc1a0964a4af7a600afb7904c8e0b"
     }
    },
    "20589c5d2372402e99f4c1406659c0ee": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "20fb65d2eee24bbb8d74fec87df215e6": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": "20px"
     }
    },
    "21cf114acd51449da8bb1532a5f18f8b": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_ddbbe0f725de4361bb71b756f749a215",
      "placeholder": "​",
      "style": "IPY_MODEL_d251f651ebf24f6da07d1735559818b6",
      "value": "Extracting data files: 100%"
     }
    },
    "21e17f19d8c84ababf07dd94864f53e5": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "24270d9757e247669e9ca2105a727cb8": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "2436895f138e4ae59b4ab936287554ed": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "24fbf1047b6c4180815d2ea167d17252": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "28828f3642a24b7984ed84748aa83c1d": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_cb03eff0750d46bca00a21f58e81f841",
      "max": 1,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_dc0071c2fe704b7e8042069d1b6db79d",
      "value": 1
     }
    },
    "295a086f553e46ecb8eb8b2c9b56916b": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_fbdb8309e4654b6c839127fdb5e29cf0",
      "max": 22865,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_b698e60171154efabc766b47d2d7c486",
      "value": 22865
     }
    },
    "29aae51c7130465f97ce90491f0c9696": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_d20cd3c0c8484cf8a4581619bfd061fb",
      "placeholder": "​",
      "style": "IPY_MODEL_c0b8d07504094fd3b4dc918fe80d5099",
      "value": ""
     }
    },
    "2e37ade3222b47b4bb3d9566dc862117": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "2f6043f91b71453ab2b7ac628c04e556": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "3197fad62a9d4d289c40772bfe021d7f": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "32ea79387ae845cea889c436ffabd9e1": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "33da8b0d87a14575aa6f65b46f803bd2": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_13d7e978cba64a3cbb9bd0e9bb50ecbc",
      "max": 1,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_20589c5d2372402e99f4c1406659c0ee",
      "value": 1
     }
    },
    "3519c3abbb684bae90a994f8a089d4ff": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "35929e7e4214489b8559a583fd28d171": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "373864a337c048778aacc3218f0aa5eb": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_679fc76bd85144659d03c25ea1cf67b1",
      "max": 271,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_e193535750a84ec0b98ab5077a6b5aa5",
      "value": 271
     }
    },
    "431c07b2e7ed4dbf9014c67d306945f8": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_028465bf497d4737a5384f6f7c58200a",
       "IPY_MODEL_d9b6dbfb87fd4c469058532892cf7a09",
       "IPY_MODEL_cc31d15c0a584eb2af3b23c6db3d2503"
      ],
      "layout": "IPY_MODEL_77033cd66ad340acbbdaeb82aa47591c"
     }
    },
    "45af82f6eb18460784ab055d93f280ef": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "info",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_20fb65d2eee24bbb8d74fec87df215e6",
      "max": 1,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_e6bcacc7468d430990aebd786bd5db3c",
      "value": 1
     }
    },
    "462f437cc6144a7982a6dc89bd6f2a90": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "47c46c504521428d9bf84e798fde8937": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "4a2741e1b02749f5887bf798ee8e21d6": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "4a4bd6bca11747dab469e933548c31c8": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_03b78f48e2ba4253b34d0946d5b40e15",
      "placeholder": "​",
      "style": "IPY_MODEL_ba01c404bddf435eba574cd35aa411e0",
      "value": " 1/1 [00:00&lt;00:00,  9.30ba/s]"
     }
    },
    "5100f2875c724d10b1801c55084c33ab": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "51379fdebbaf4587b5cabb85d4fb8faa": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "52e6c6ed71eb427b88df2e3fbf5b87c1": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "55742a800669463d81d67456044285cb": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "5bcc6cfea7be4f3f8171f64ee9f44e17": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "5ebfc1a0964a4af7a600afb7904c8e0b": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "5f2c37ea11f24e96864754cb94c6c5dd": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "5f879ff41f624a7ba425b9e28646ab58": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "60a8726cf933422db26f755d9c5623d1": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_017e3f0a45c942e29b0fc432dca2ffc8",
       "IPY_MODEL_373864a337c048778aacc3218f0aa5eb",
       "IPY_MODEL_8f7ad298e78848c4918602274ddd45d1"
      ],
      "layout": "IPY_MODEL_2e37ade3222b47b4bb3d9566dc862117"
     }
    },
    "622ef2865684418eb481e799c337c523": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "6372d32e9b8c4efb9e212a376a4c7947": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_1144746deacc4c07aa06c81906f6ce77",
       "IPY_MODEL_45af82f6eb18460784ab055d93f280ef",
       "IPY_MODEL_c6d60b6c586f484ebc402f5d44cc2f7a"
      ],
      "layout": "IPY_MODEL_7e0fd7d1189048f09ad63b0d6f5e7783"
     }
    },
    "6687dc570dc74feba3bd2413c408eb76": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "679fc76bd85144659d03c25ea1cf67b1": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "68141e3ab39b40169949a3576477a42b": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "69f8d2bece1e4bc4987f81c63d82aded": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_24fbf1047b6c4180815d2ea167d17252",
      "placeholder": "​",
      "style": "IPY_MODEL_6687dc570dc74feba3bd2413c408eb76",
      "value": " 346M/346M [00:19&lt;00:00, 19.2MB/s]"
     }
    },
    "6c590b31f3b043a4a2dcc3bacb742452": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "7121d94ad31140a5a9d6b88edd690aa9": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "71bfa349d37c4717bb34bb93e8779bd8": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "7254dc86b6714a81b46997dc132360e0": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "73578108cdb24b4586b123f9271e2fdf": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "743fdc0668b44aa2a79d9c64b1e2d08d": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "7478c6eda4694875a8dcfb3b59676f26": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "77033cd66ad340acbbdaeb82aa47591c": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "77c5c395d16247cf9bd2a0d4e7a216f9": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_29aae51c7130465f97ce90491f0c9696",
       "IPY_MODEL_bfa55478d7ed4d629c85c76d44c08a3c",
       "IPY_MODEL_d1f5438cbf314ffaaecc473f9922ee70"
      ],
      "layout": "IPY_MODEL_e9045bde9871448a863b44559d114a5a"
     }
    },
    "77d5ea8f038c4dd48d16cd885b00629a": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "781aceb284cf4bb192dde0fd2b731079": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "7b96d51d2819479c9adf009d0d1b6710": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_e6b09c270dac4dc19263fe6596985c0c",
      "placeholder": "​",
      "style": "IPY_MODEL_cae454ab61954479a24dad6620b3d937",
      "value": " 1/1 [00:00&lt;00:00, 10.45ba/s]"
     }
    },
    "7e0fd7d1189048f09ad63b0d6f5e7783": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": "hidden",
      "width": null
     }
    },
    "7e251276f132481e886901a9c0fa5562": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "7e98903f8faf4030a3b46fb8b744b5e5": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "811ee525a96b49b1819473a5ccdc01a6": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "81aa9067dbd94f979c23bc64897a3eab": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "824d814cdeea438798bd18426092fe41": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "87fea88df3494c8592b5de4485bcf8a7": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "8f7ad298e78848c4918602274ddd45d1": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_5bcc6cfea7be4f3f8171f64ee9f44e17",
      "placeholder": "​",
      "style": "IPY_MODEL_a533fd53359e42548409f60ac412c7ba",
      "value": " 271/271 [00:00&lt;00:00, 2.61kB/s]"
     }
    },
    "9024ea625a284800811b19d7a0e26079": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_bfbbfff4a70e4d7183742f5e5118f6bb",
      "max": 1,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_c74dd5761ca349ed9eba8b7073484ae7",
      "value": 1
     }
    },
    "95889cc921824cfd90c53b7454175378": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "9a9d479f3065407d88be6ca2b12abffa": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "9b00cff2e12c4886b3f6cf1e6879d725": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "9e90e00ea181485c8ac420e32b989f08": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_fa4749f5fa94493a9ab3e9f218922106",
       "IPY_MODEL_b7ff863729484b8fbb5fbc6a8c791e3e",
       "IPY_MODEL_69f8d2bece1e4bc4987f81c63d82aded"
      ],
      "layout": "IPY_MODEL_1af7e067011a4854a1e5f97624b230cd"
     }
    },
    "a06160cdb0be42c2af5f55ec835e44ff": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_811ee525a96b49b1819473a5ccdc01a6",
      "placeholder": "​",
      "style": "IPY_MODEL_6c590b31f3b043a4a2dcc3bacb742452",
      "value": "Downloading: 100%"
     }
    },
    "a07073612892430e94915351ad1c32e2": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "a533fd53359e42548409f60ac412c7ba": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "ae9b118bc353457ab8e6c21208a8d57b": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_24270d9757e247669e9ca2105a727cb8",
      "placeholder": "​",
      "style": "IPY_MODEL_7e98903f8faf4030a3b46fb8b744b5e5",
      "value": "Extracting data files: 100%"
     }
    },
    "b56c2b116f4e43a89dcb1597f69deea2": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_95889cc921824cfd90c53b7454175378",
      "placeholder": "​",
      "style": "IPY_MODEL_35929e7e4214489b8559a583fd28d171",
      "value": "Creating json from Arrow format: 100%"
     }
    },
    "b698e60171154efabc766b47d2d7c486": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "b7ff863729484b8fbb5fbc6a8c791e3e": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_87fea88df3494c8592b5de4485bcf8a7",
      "max": 346198861,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_f04bb6ab61ee477c8a02719eeab23dff",
      "value": 346198861
     }
    },
    "b9ecdecdc0af4ed5a8b14fe484eff014": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_fa915a0ccb534634b3b580fb3cb09c16",
       "IPY_MODEL_9024ea625a284800811b19d7a0e26079",
       "IPY_MODEL_7b96d51d2819479c9adf009d0d1b6710"
      ],
      "layout": "IPY_MODEL_a07073612892430e94915351ad1c32e2"
     }
    },
    "ba01c404bddf435eba574cd35aa411e0": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "bcea670e5b234125a4d801fafe8928f7": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_107b16ef81554781856cd67a6ae038da",
      "placeholder": "​",
      "style": "IPY_MODEL_68141e3ab39b40169949a3576477a42b",
      "value": " 1/1 [00:00&lt;00:00, 22.97it/s]"
     }
    },
    "bd9d32c5d158420990bf82ffb68f48c9": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_a06160cdb0be42c2af5f55ec835e44ff",
       "IPY_MODEL_295a086f553e46ecb8eb8b2c9b56916b",
       "IPY_MODEL_17b6f0610cee47e1a33f2b54edc3631e"
      ],
      "layout": "IPY_MODEL_9a9d479f3065407d88be6ca2b12abffa"
     }
    },
    "bfa55478d7ed4d629c85c76d44c08a3c": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "info",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_fcaf3755cdc44224a46a32762fcae938",
      "max": 1,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_c59ef23f2be94dcba663c347779c860a",
      "value": 1
     }
    },
    "bfbbfff4a70e4d7183742f5e5118f6bb": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "bfe6a0d505be48b888ea13cdc0d56447": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_743fdc0668b44aa2a79d9c64b1e2d08d",
      "max": 1,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_7254dc86b6714a81b46997dc132360e0",
      "value": 1
     }
    },
    "c0b8d07504094fd3b4dc918fe80d5099": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "c59ef23f2be94dcba663c347779c860a": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "c6d60b6c586f484ebc402f5d44cc2f7a": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_3519c3abbb684bae90a994f8a089d4ff",
      "placeholder": "​",
      "style": "IPY_MODEL_51379fdebbaf4587b5cabb85d4fb8faa",
      "value": " 0/? [00:00&lt;?, ? tables/s]"
     }
    },
    "c74dd5761ca349ed9eba8b7073484ae7": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "cae454ab61954479a24dad6620b3d937": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "cafb90a4897b49028fbbcfe6b981e385": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "cb03eff0750d46bca00a21f58e81f841": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "cc31d15c0a584eb2af3b23c6db3d2503": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_5f879ff41f624a7ba425b9e28646ab58",
      "placeholder": "​",
      "style": "IPY_MODEL_824d814cdeea438798bd18426092fe41",
      "value": " 1/1 [00:00&lt;00:00, 34.65it/s]"
     }
    },
    "d1f5438cbf314ffaaecc473f9922ee70": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_781aceb284cf4bb192dde0fd2b731079",
      "placeholder": "​",
      "style": "IPY_MODEL_81aa9067dbd94f979c23bc64897a3eab",
      "value": " 0/? [00:00&lt;?, ? tables/s]"
     }
    },
    "d20cd3c0c8484cf8a4581619bfd061fb": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "d251f651ebf24f6da07d1735559818b6": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "d9b6dbfb87fd4c469058532892cf7a09": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_47c46c504521428d9bf84e798fde8937",
      "max": 1,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_08301e86bd1745f28af465f1b9f9d050",
      "value": 1
     }
    },
    "dc0071c2fe704b7e8042069d1b6db79d": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "ddbbe0f725de4361bb71b756f749a215": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "e193535750a84ec0b98ab5077a6b5aa5": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "e2eafd1b4cfa4965bf05c9caa0ea4595": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_3197fad62a9d4d289c40772bfe021d7f",
      "max": 1,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_71bfa349d37c4717bb34bb93e8779bd8",
      "value": 1
     }
    },
    "e3ac5081836d40f797a5305f1429175e": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_5f2c37ea11f24e96864754cb94c6c5dd",
      "placeholder": "​",
      "style": "IPY_MODEL_2f6043f91b71453ab2b7ac628c04e556",
      "value": " 1/1 [00:00&lt;00:00, 31.83it/s]"
     }
    },
    "e6b09c270dac4dc19263fe6596985c0c": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "e6bcacc7468d430990aebd786bd5db3c": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "e79a955f009e447ea03841ebd51f5475": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_b56c2b116f4e43a89dcb1597f69deea2",
       "IPY_MODEL_28828f3642a24b7984ed84748aa83c1d",
       "IPY_MODEL_4a4bd6bca11747dab469e933548c31c8"
      ],
      "layout": "IPY_MODEL_73578108cdb24b4586b123f9271e2fdf"
     }
    },
    "e9045bde9871448a863b44559d114a5a": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": "hidden",
      "width": null
     }
    },
    "f04bb6ab61ee477c8a02719eeab23dff": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "fa4749f5fa94493a9ab3e9f218922106": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_cafb90a4897b49028fbbcfe6b981e385",
      "placeholder": "​",
      "style": "IPY_MODEL_622ef2865684418eb481e799c337c523",
      "value": "Downloading: 100%"
     }
    },
    "fa915a0ccb534634b3b580fb3cb09c16": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_7121d94ad31140a5a9d6b88edd690aa9",
      "placeholder": "​",
      "style": "IPY_MODEL_2436895f138e4ae59b4ab936287554ed",
      "value": "Creating json from Arrow format: 100%"
     }
    },
    "fbdb8309e4654b6c839127fdb5e29cf0": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "fcaf3755cdc44224a46a32762fcae938": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": "20px"
     }
    },
    "fec838d6713444ff837f6f340502adb9": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_21cf114acd51449da8bb1532a5f18f8b",
       "IPY_MODEL_33da8b0d87a14575aa6f65b46f803bd2",
       "IPY_MODEL_0e8d7313d9f24c49821bb81bb013a925"
      ],
      "layout": "IPY_MODEL_7478c6eda4694875a8dcfb3b59676f26"
     }
    }
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}