{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BL8OGgI9CmQy"
      },
      "source": [
        "# Práctica 8\n",
        "\n",
        "Ahora es tu turno de trabajar con modelos de lenguaje y self-supervised learning.\n",
        "\n",
        "## Ejercicio obligatorio (5 puntos)\n",
        "\n",
        "El ejercicio que tenéis que realizar obligatoriamente consiste en usar el dataset de clasificación de texto que empleaste en la práctica anterior para construir un modelo siguiendo las instrucciones proporcionadas en el notebook de instrucciones. Además debes comparar los resultados obtenidos con los que lograste en la práctica anterior. \n",
        "\n",
        "## Ejercicio opcional 1 (0.5 puntos)\n",
        "\n",
        "Crea un nuevo espacio en HuggingFace con el nuevo modelo que has creado.\n",
        "\n",
        "## Ejercicio opcional 2 (4.5 puntos)\n",
        "\n",
        "El ejercicio opcional está centrado en investigar cómo utilizar self-supervised learning para clasificación de imágenes. Para ello crea un nuevo notebook donde reproduzcas los resultados de la parte obligatoria de la Práctica 1. A continuación realiza los siguientes ejercicios:\n",
        "- Entrena el modelo desde cero (mira qué hace el parámetro `pretrained` del método `cnn_learner`. \n",
        "- Usando la librería [Self Supervised Learning Fastai Extension](https://github.com/KeremTurgutlu/self_supervised) entrena tres modelos de self-supervised usando los algoritmos SimCLR, BYOL, y SwAV disponibles en dicha librería.\n",
        "- A partir de los modelos de self-supervised entrenados en el paso anterior, crea nuevos modelos de clasificación. \n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tijRfybvCmQ2"
      },
      "source": [
        "Al finalizar, recuerda guardar los cambios en GitHub utilizando la opción Archivo -> Guardar una copia en GitHub.\n"
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "N3DDTYS45C9h"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5ZDQ6rKcC2kq"
      },
      "source": [
        "## Librerías\n",
        "\n",
        "Comenzamos actualizando la librería FastAI y descargando la librería datasets de HuggingFace. Al finalizar la instalación deberás reiniciar el kernel (menú Entorno de ejecución -> Reiniciar Entorno de ejecución)."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "y_vFVAFzC2kq",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "8063282b-ccae-44b3-9357-9a88f23a2727"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m468.7/468.7 kB\u001b[0m \u001b[31m7.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m110.5/110.5 kB\u001b[0m \u001b[31m5.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.0/1.0 MB\u001b[0m \u001b[31m24.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m224.2/224.2 kB\u001b[0m \u001b[31m10.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m212.2/212.2 kB\u001b[0m \u001b[31m10.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m132.9/132.9 kB\u001b[0m \u001b[31m1.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m158.8/158.8 kB\u001b[0m \u001b[31m10.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m269.3/269.3 kB\u001b[0m \u001b[31m11.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m114.2/114.2 kB\u001b[0m \u001b[31m4.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25h"
          ]
        }
      ],
      "source": [
        "!pip install fastai -Uqq\n",
        "!pip install datasets -Uqq"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JAo2DOLlC2kr"
      },
      "source": [
        "Cargamos a continuación las librerías que necesitaremos en esta práctica que son la parte de procesado de lenguaje natural de la librería fastAI, la librería pandas, y la funcionalidad para cargar datasets de HuggingFace."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "id": "-5sb8kBaC2ks"
      },
      "outputs": [],
      "source": [
        "import pandas as pd\n",
        "from fastai.text.all import *\n",
        "from datasets import load_dataset"
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "p1aBx-Sk62AB"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Dataset\n",
        "\n",
        "Para este ejemplo vamos a usar el dataset [clinc_oos](https://huggingface.co/datasets/clinc_oos), un dataset para detectar 150 clases de intención en 10 dominios, el conjunto de datos contiene una etiqueta para la intención \"fuera del ámbito\".( Label Id | Label name | |--- |--- | | 0 | restaurant_reviews | | 1 | nutrition_info | | 2 | account_blocked | | 3 | oil_change_how | | 4 | time | | 5 | weather | | 6 | redeem_rewards | | 7 | interest_rate | | 8 | gas_type | | 9 | accept_reservations | | 10 | smart_home | | 11 | user_name | | 12 | report_lost_card | | 13 | repeat | | 14 | whisper_mode | | 15 | what_are_your_hobbies | | 16 | order | | 17 | jump_start | | 18 | schedule_meeting | | 19 | meeting_schedule | | 20 | freeze_account | | 21 | what_song | | 22 | meaning_of_life | | 23 | restaurant_reservation | | 24 | traffic | | 25 | make_call | | 26 | text | | 27 | bill_balance ..)\n",
        "Descarga el dataset usando el siguiente comando. "
      ],
      "metadata": {
        "id": "V3SGfSUaEHv0"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "clinc_oos_dataset = load_dataset(\"clinc_oos\",'small')"
      ],
      "metadata": {
        "id": "ephhRD5rDnCd",
        "outputId": "999b8928-7ceb-4f17-bd1c-f94d92193fd4",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 244,
          "referenced_widgets": [
            "acea7c14a43a4d98ac3fa7d55bcf6ee6",
            "2da90ab6b0f54d49bd1c97944d7aa83e",
            "bc81e5b43dac454a9b043edcbe8bdd22",
            "a8520126760f425cb07c030d6ea2c273",
            "dc6a959e6b694c5cbc8a539272f29ad7",
            "b52ea6a60f84468d982c195927cbb7ef",
            "11e54992e4fd4d40bae190ab927514b7",
            "50b6f98995b940c0881aef1870ade8ab",
            "ffdd49432e9744f7bd5c23bff0244b54",
            "ba2477e9ee9f48c8be6f2f74f8bc2cd3",
            "0b28d318f69844d88afd25bd867d827e",
            "a03bc89197ea4573b0f6784ca5b4e80b",
            "1b0a75b3b78b468aa8e75f3074d58dfb",
            "603fffb705344e41bbf2cb31fc1e5bfc",
            "e74351f03d53422789bbdbbfeb764efa",
            "c65151b08bec4cf48d0e0d6ee1250f59",
            "78fabf12226a4f9cab33f1885b963534",
            "e34d7ab3f4834d17ae863d288445d9be",
            "ff14ace405e4409589bc3531a789dda1",
            "08c58056296849fdbab23840a9196979",
            "84ff998cb0f8461daa8deaf55dc17db5",
            "63cd9ad6473b43f0856596c29c6a0cf3",
            "a4f85e3bf0eb45c4857f6173b1adebcd",
            "8edf734aa6d64f5f89906b464d512650",
            "94dba14a5bc342bb8711f6843b6d7ed1",
            "fc6d1aa44bd94a3ab9ca40c6a66c8ffb",
            "a15bf77b783d4833b9c71ad2b2d9a9a4",
            "76f7f82fe6134040a90c7413538aafe5",
            "30d05e576f3747e08d932ea00fed01be",
            "ffda192e0bc24c40aa272eea8011e1a1",
            "89692d1d3c914672a458832e51be6d2d",
            "1ebc784efa24427b8af27c9fdda2686b",
            "f7d8ef7ca553491c9fba3ab382af530c",
            "fd342b9fb4a84009b32ab4c7bcb377a4",
            "daf05dc94f98455684bc96752c216e9c",
            "575724ca3bb24bc79d64ed65544bc043",
            "2f8aa3e45a55423d83606705f5010348",
            "7e55727bedd94e849a125e4406f6a980",
            "f595b20d8db44b9998732a1cc713a1e7",
            "00d5b27795f448d485da956dbd8a73d4",
            "3d5308aafb4744508c3390d59f8c6992",
            "cbcbf6c645a14f61b625154537e9e528",
            "0bc1a29b59384aec931b7d0c611c5ec2",
            "8ff6ad225eba4c1b9baf01aa9db1a7d3",
            "18935a837f49426dac45432e83a5913c",
            "c6a107ffcc754e82921ad75bc6643898",
            "2e382b16620c4bb09534050dae7a9d6e",
            "997bd650baad41819aa3a713fa4341f3",
            "29e85d0ef3754c7f8366469cd530cacb",
            "5bee187efff34f8eab2c434eb21211ea",
            "275b01de8e6444a69420369d2a96b76d",
            "71ab8921857b496791db2a2286fab6ba",
            "225e43ff721745c5b01dd7fe5dfe7e4d",
            "a6900524efc04d8d8019c415196d4a46",
            "a682df09134f4c74aef12a2e294e2cae",
            "b329b0dc353247f58bf8c38b24c4e051",
            "2ac54f2315934367ae8bd34aba136f21",
            "579357f9baec40f2915d8f3021fc20eb",
            "4e1c1bd8b4d74978bff0cfa348a1bff0",
            "68bae7a3d76f4e68afe3e6fac15ccfb4",
            "720d5c006f2f46d982be76cb50eb8f17",
            "d2a91625d68f43479aa16f24b386af74",
            "19601448d0704949aaa8459fae71b878",
            "cdbef121730444b8abd227f4dbd42d3c",
            "c24578c84bae4ab5ac0145c742f0747d",
            "843ad42380cb40b989659ac691e67d65",
            "f2d9294ee6f44814a8e8d9a87cf08c64",
            "532efa9cd0d8459eac8b03bba83aad51",
            "f6d159342ce047df904907d04cceb9ec",
            "035514d9dad34ea6b645018dad177693",
            "09394797973941e68705559ddcd9b802",
            "571f0233507544148b23a0dcf251daa3",
            "8fcd79171edb43b28b3291c06440b5cb",
            "cf801abb3ce94c9b8bf91b4d1efa39db",
            "116121b6f33b4883884fa748e515d969",
            "a2320a4c9d0e492c93cc9b06e758e0d5",
            "10e14d896714494da3d8d37dd8ed9282",
            "c8aca95b4c1d4734a86f912b7ea43c6e",
            "55a5af05b8234f938ff6104af90f2460",
            "f66bcff001e04b199a1bbc9d92f42367",
            "ff9eb64ef78b4722b373285f8bce32ed",
            "70aad33e2cd043c08a94c8b8a93a3342",
            "892450aebc2b4611aec22ffb88a8cba9",
            "1e6314ba4f4a445b9b68928c6debb77c",
            "ef7d3624b5fe44eea8d40f6b6a02898b",
            "e652c021449c485484fb899b17acd0ef",
            "0661f3b9e7bc408f82030c5b7c164fcb",
            "ab50e73bf5404380baba92cf1aa40976"
          ]
        }
      },
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading builder script:   0%|          | 0.00/8.57k [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "acea7c14a43a4d98ac3fa7d55bcf6ee6"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading metadata:   0%|          | 0.00/14.4k [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "a03bc89197ea4573b0f6784ca5b4e80b"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading readme:   0%|          | 0.00/23.4k [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "a4f85e3bf0eb45c4857f6173b1adebcd"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading and preparing dataset clinc_oos/small to /root/.cache/huggingface/datasets/clinc_oos/small/1.0.0/abcc41d382f8137f039adc747af44714941e8196e845dfbdd8ae7a7e020e6ba1...\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading data:   0%|          | 0.00/217k [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "fd342b9fb4a84009b32ab4c7bcb377a4"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Generating train split:   0%|          | 0/7600 [00:00<?, ? examples/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "18935a837f49426dac45432e83a5913c"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Generating validation split:   0%|          | 0/3100 [00:00<?, ? examples/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "b329b0dc353247f58bf8c38b24c4e051"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Generating test split:   0%|          | 0/5500 [00:00<?, ? examples/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "f2d9294ee6f44814a8e8d9a87cf08c64"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Dataset clinc_oos downloaded and prepared to /root/.cache/huggingface/datasets/clinc_oos/small/1.0.0/abcc41d382f8137f039adc747af44714941e8196e845dfbdd8ae7a7e020e6ba1. Subsequent calls will reuse this data.\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "  0%|          | 0/3 [00:00<?, ?it/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "c8aca95b4c1d4734a86f912b7ea43c6e"
            }
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "clinc_oos_dataset"
      ],
      "metadata": {
        "id": "Du4cdYWVI0J7",
        "outputId": "cbe0b060-b88a-474e-efc6-d2a5de555f2b",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "DatasetDict({\n",
              "    train: Dataset({\n",
              "        features: ['text', 'intent'],\n",
              "        num_rows: 7600\n",
              "    })\n",
              "    validation: Dataset({\n",
              "        features: ['text', 'intent'],\n",
              "        num_rows: 3100\n",
              "    })\n",
              "    test: Dataset({\n",
              "        features: ['text', 'intent'],\n",
              "        num_rows: 5500\n",
              "    })\n",
              "})"
            ]
          },
          "metadata": {},
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Podemos ver que el dataset es una estructura DatasetDict que puede verse como un diccionario. El diccionario tiene tres claves que son train, validation y test que indican respectivamente los conjuntos de entrenamiento, validación y test (estas claves pueden variar dependiendo del dataset). Cada uno de estos subconjuntos es un Dataset que puede verse como una lista. Podemos ver por ejemplo la primera frase del conjunto de entrenamiento del siguiente modo."
      ],
      "metadata": {
        "id": "ZVJsMgNUJe4T"
      }
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0XB8Dl8jjwbN"
      },
      "source": [
        "## Carga de datos\n",
        "\n",
        "Cargamos a continuación el dataset en distintos dataframes de pandas (el formato que puede leer la librería de FastAI)."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "id": "mePgxpeLjwbO"
      },
      "outputs": [],
      "source": [
        "train_df = clinc_oos_dataset[\"train\"].to_pandas()\n",
        "valid_df = clinc_oos_dataset[\"validation\"].to_pandas()\n",
        "test_df = clinc_oos_dataset[\"test\"].to_pandas()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FN-QG9CljwbO"
      },
      "source": [
        "Podemos ver el contenido del dataset usando el siguiente comando."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "id": "veoRjZOWjwbP",
        "outputId": "e790ac73-5a62-461e-cfe4-3bebe7b1b25f"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                                                                           text  \\\n",
              "0     can you walk me through setting up direct deposits to my bank of internet savings account   \n",
              "1                                                            i want to switch to direct deposit   \n",
              "2                                                                  set up direct deposit for me   \n",
              "3                                                   how do i go about setting up direct deposit   \n",
              "4                                i need to get my paycheck direct deposited to my chase account   \n",
              "...                                                                                         ...   \n",
              "7595                                     what percentage of species display cold blooded traits   \n",
              "7596                                                      what does it mean to be an alpha male   \n",
              "7597                                                              what animals have alpha males   \n",
              "7598                                                              why do males want to be alpha   \n",
              "7599                                        what's the average battery life of an android phone   \n",
              "\n",
              "      intent  \n",
              "0        108  \n",
              "1        108  \n",
              "2        108  \n",
              "3        108  \n",
              "4        108  \n",
              "...      ...  \n",
              "7595      42  \n",
              "7596      42  \n",
              "7597      42  \n",
              "7598      42  \n",
              "7599      42  \n",
              "\n",
              "[7600 rows x 2 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-8a5e6fef-dcbf-4625-87dd-26672978d545\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>text</th>\n",
              "      <th>intent</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>can you walk me through setting up direct deposits to my bank of internet savings account</td>\n",
              "      <td>108</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>i want to switch to direct deposit</td>\n",
              "      <td>108</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>set up direct deposit for me</td>\n",
              "      <td>108</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>how do i go about setting up direct deposit</td>\n",
              "      <td>108</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>i need to get my paycheck direct deposited to my chase account</td>\n",
              "      <td>108</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7595</th>\n",
              "      <td>what percentage of species display cold blooded traits</td>\n",
              "      <td>42</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7596</th>\n",
              "      <td>what does it mean to be an alpha male</td>\n",
              "      <td>42</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7597</th>\n",
              "      <td>what animals have alpha males</td>\n",
              "      <td>42</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7598</th>\n",
              "      <td>why do males want to be alpha</td>\n",
              "      <td>42</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7599</th>\n",
              "      <td>what's the average battery life of an android phone</td>\n",
              "      <td>42</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>7600 rows × 2 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-8a5e6fef-dcbf-4625-87dd-26672978d545')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-8a5e6fef-dcbf-4625-87dd-26672978d545 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-8a5e6fef-dcbf-4625-87dd-26672978d545');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ],
      "source": [
        "train_df"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sLJE_AtyjwbP"
      },
      "source": [
        "Del dataset nos interesan dos campos: `text` (que contiene el poema) y `intent` ."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ivLo2dIHjwbP",
        "outputId": "f5120ac6-b4ff-4535-92f9-b3a8b4557ebe"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0       can you walk me through setting up direct deposits to my bank of internet savings account\n",
              "1                                                              i want to switch to direct deposit\n",
              "2                                                                    set up direct deposit for me\n",
              "3                                                     how do i go about setting up direct deposit\n",
              "4                                  i need to get my paycheck direct deposited to my chase account\n",
              "                                                  ...                                            \n",
              "7595                                       what percentage of species display cold blooded traits\n",
              "7596                                                        what does it mean to be an alpha male\n",
              "7597                                                                what animals have alpha males\n",
              "7598                                                                why do males want to be alpha\n",
              "7599                                          what's the average battery life of an android phone\n",
              "Name: text, Length: 7600, dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 11
        }
      ],
      "source": [
        "train_df['text']"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Dataset\n",
        "\n",
        "Para este ejemplo vamos a usar el dataset [clinc_oos](https://huggingface.co/datasets/clinc_oos), un dataset para detectar 150 clases de intención en 10 dominios, el conjunto de datos contiene una etiqueta para la intención \"fuera del ámbito\".( Label Id | Label name | |--- |--- | | 0 | restaurant_reviews | | 1 | nutrition_info | | 2 | account_blocked | | 3 | oil_change_how | | 4 | time | | 5 | weather | | 6 | redeem_rewards | | 7 | interest_rate | | 8 | gas_type | | 9 | accept_reservations | | 10 | smart_home | | 11 | user_name | | 12 | report_lost_card | | 13 | repeat | | 14 | whisper_mode | | 15 | what_are_your_hobbies | | 16 | order | | 17 | jump_start | | 18 | schedule_meeting | | 19 | meeting_schedule | | 20 | freeze_account | | 21 | what_song | | 22 | meaning_of_life | | 23 | restaurant_reservation | | 24 | traffic | | 25 | make_call | | 26 | text | | 27 | bill_balance ..)\n",
        "Descarga el dataset usando el siguiente comando. "
      ],
      "metadata": {
        "id": "W-giW-0s7Hsp"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "clinc_oos_dataset = load_dataset(\"clinc_oos\",'small')"
      ],
      "metadata": {
        "outputId": "2a45d8bc-cb19-476c-82f8-f866cd035942",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 148,
          "referenced_widgets": [
            "d2b46b2844f3474899145a75ffb4c571",
            "b10eccc8dc2c429abdb627a62d6863e7",
            "765efc5a4bce4307bd4540d8c1f56a0e",
            "02d90990072842078d64e09f72176f8f",
            "a1884aac1bd9483ab57f0b928c63832b",
            "1be048cecdb14c2799b0ece8c6c9c9e4",
            "3d3ff51a288540fc8cba1d6b4b831a14",
            "bb970639d5144a3f92ac875377642655",
            "6e7e5d794a8c40ecbda3cd7ad568f153",
            "29f81fe476cd47089f5617bc7276d65c",
            "8fc2e123da594603987eb98660b87d94",
            "10d5cd30a0e54c5ba7b6df6b8ef580a0",
            "b80735bbbcfa4b589a77d4eb34b23d04",
            "4ccaa9c7811a4cb6829eedc1bca8ab34",
            "1fe255143a2547719eabda362cc73069",
            "1d37657c44d24109975288f069e35383",
            "89ea37b2f176438c94b7273d18999424",
            "a72de9da781c40d194194af2b03ce2fa",
            "2939a6c917544effb6d30290e02750c2",
            "b4e7b22a14ab4de39a6be5b9ac87a859",
            "81514c49084c4699acf2a19f176905fe",
            "a7f35632035245c7a808facbcde1faeb",
            "c0ad3dcde07c46dc9026973722d53b97",
            "e6dd5c7de93f45bcb5f8966d7df979a2",
            "2e78b25a5d5d4674b1d1f61708a7f8bb",
            "b97a5dca9f6141b08b81ff1a58d68919",
            "45957dbb65994da793d90e4632a7e0d9",
            "d15ffa6b1da143e8baf62041b23662cc",
            "817890f3aef0421a83b33b3531c8ac1b",
            "9e49685857a045fbbcc7378a5fdbaa50",
            "018fe2a8ea194860934fe72fd989b067",
            "98162db40afa4484a4321d58fd9d2848",
            "71cba57f6c4c4e53b2c390ed3fa0bcbc",
            "1cedd010d23c4709ad213689552198de",
            "f60a867c6dd445e39103a549f176bb10",
            "6e2e2c2a86014b85af276d407cd32ed6",
            "c21f1c59e8c04770b2952383125dec27",
            "d5875aca8911496b9b96367b6a8e8852",
            "146d51dcd77f41518c064c23941569c3",
            "670ffe1337604111be17198d8b087155",
            "7fd60f1fbc0042198decd1113f697fad",
            "14d341830ed042219b6bd40b748a994c",
            "88871391967848f18068333f6a8a6530",
            "309e721a2df84da39090d74931655bc2",
            "cd54965515904b2dacb44113735df90d",
            "ffcf7b7524be47ec9f50726299d22d9e",
            "b3637041f9ab41fdb0bc409c33c8f921",
            "8682ad4b47de444b968ba2e1992ddb65",
            "d9a8ed1d734f49ffaf5e59138272abfd",
            "afe03557c5df47afb4c3483ca18f3de2",
            "d883e3b74b8343c78642b355e51b590c",
            "41a3d446c21b40c28f5ad0328ef75adc",
            "bc73d68234e54b83a053797d64ba24fd",
            "e95a9bed6ee24ab4916528200a033764",
            "ec7c526f8e3f43a094eb2d3bf87fe0df"
          ]
        },
        "id": "p7psPv0w7Hsr"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading and preparing dataset clinc_oos/small to /root/.cache/huggingface/datasets/clinc_oos/small/1.0.0/abcc41d382f8137f039adc747af44714941e8196e845dfbdd8ae7a7e020e6ba1...\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading data:   0%|          | 0.00/217k [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "d2b46b2844f3474899145a75ffb4c571"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Generating train split:   0%|          | 0/7600 [00:00<?, ? examples/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "10d5cd30a0e54c5ba7b6df6b8ef580a0"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Generating validation split:   0%|          | 0/3100 [00:00<?, ? examples/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "c0ad3dcde07c46dc9026973722d53b97"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Generating test split:   0%|          | 0/5500 [00:00<?, ? examples/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "1cedd010d23c4709ad213689552198de"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Dataset clinc_oos downloaded and prepared to /root/.cache/huggingface/datasets/clinc_oos/small/1.0.0/abcc41d382f8137f039adc747af44714941e8196e845dfbdd8ae7a7e020e6ba1. Subsequent calls will reuse this data.\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "  0%|          | 0/3 [00:00<?, ?it/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "cd54965515904b2dacb44113735df90d"
            }
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "clinc_oos_dataset"
      ],
      "metadata": {
        "outputId": "5f63daec-d372-4acc-aad8-d660d528ff3c",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vO3Al_D27Hst"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "DatasetDict({\n",
              "    train: Dataset({\n",
              "        features: ['text', 'intent'],\n",
              "        num_rows: 7600\n",
              "    })\n",
              "    validation: Dataset({\n",
              "        features: ['text', 'intent'],\n",
              "        num_rows: 3100\n",
              "    })\n",
              "    test: Dataset({\n",
              "        features: ['text', 'intent'],\n",
              "        num_rows: 5500\n",
              "    })\n",
              "})"
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Podemos ver que el dataset es una estructura DatasetDict que puede verse como un diccionario. El diccionario tiene tres claves que son train, validation y test que indican respectivamente los conjuntos de entrenamiento, validación y test (estas claves pueden variar dependiendo del dataset). Cada uno de estos subconjuntos es un Dataset que puede verse como una lista. Podemos ver por ejemplo la primera frase del conjunto de entrenamiento del siguiente modo."
      ],
      "metadata": {
        "id": "K_S0SmHQ7Hsu"
      }
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-Q9BiqO-7Hs1"
      },
      "source": [
        "## Carga de datos\n",
        "\n",
        "Cargamos a continuación el dataset en distintos dataframes de pandas (el formato que puede leer la librería de FastAI)."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "U-rv9q8V7Hs2"
      },
      "outputs": [],
      "source": [
        "train_df = clinc_oos_dataset[\"train\"].to_pandas()\n",
        "valid_df = clinc_oos_dataset[\"validation\"].to_pandas()\n",
        "test_df = clinc_oos_dataset[\"test\"].to_pandas()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "v5b9b13Y7Hs3"
      },
      "source": [
        "Podemos ver el contenido del dataset usando el siguiente comando."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HG8ruoAv7Hs4"
      },
      "source": [
        "Del dataset nos interesan dos campos: `text` (que contiene el poema) y `intent` ."
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "XMxR9DgE7XV4"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Dataset\n",
        "\n",
        "Para este ejemplo vamos a usar el dataset [clinc_oos](https://huggingface.co/datasets/clinc_oos), un dataset para detectar 150 clases de intención en 10 dominios, el conjunto de datos contiene una etiqueta para la intención \"fuera del ámbito\".( Label Id | Label name | |--- |--- | | 0 | restaurant_reviews | | 1 | nutrition_info | | 2 | account_blocked | | 3 | oil_change_how | | 4 | time | | 5 | weather | | 6 | redeem_rewards | | 7 | interest_rate | | 8 | gas_type | | 9 | accept_reservations | | 10 | smart_home | | 11 | user_name | | 12 | report_lost_card | | 13 | repeat | | 14 | whisper_mode | | 15 | what_are_your_hobbies | | 16 | order | | 17 | jump_start | | 18 | schedule_meeting | | 19 | meeting_schedule | | 20 | freeze_account | | 21 | what_song | | 22 | meaning_of_life | | 23 | restaurant_reservation | | 24 | traffic | | 25 | make_call | | 26 | text | | 27 | bill_balance ..)\n",
        "Descarga el dataset usando el siguiente comando. "
      ],
      "metadata": {
        "id": "mhgyRRLq7Xjs"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "clinc_oos_dataset = load_dataset(\"clinc_oos\",'small')"
      ],
      "metadata": {
        "outputId": "2a45d8bc-cb19-476c-82f8-f866cd035942",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 148,
          "referenced_widgets": [
            "d2b46b2844f3474899145a75ffb4c571",
            "b10eccc8dc2c429abdb627a62d6863e7",
            "765efc5a4bce4307bd4540d8c1f56a0e",
            "02d90990072842078d64e09f72176f8f",
            "a1884aac1bd9483ab57f0b928c63832b",
            "1be048cecdb14c2799b0ece8c6c9c9e4",
            "3d3ff51a288540fc8cba1d6b4b831a14",
            "bb970639d5144a3f92ac875377642655",
            "6e7e5d794a8c40ecbda3cd7ad568f153",
            "29f81fe476cd47089f5617bc7276d65c",
            "8fc2e123da594603987eb98660b87d94",
            "10d5cd30a0e54c5ba7b6df6b8ef580a0",
            "b80735bbbcfa4b589a77d4eb34b23d04",
            "4ccaa9c7811a4cb6829eedc1bca8ab34",
            "1fe255143a2547719eabda362cc73069",
            "1d37657c44d24109975288f069e35383",
            "89ea37b2f176438c94b7273d18999424",
            "a72de9da781c40d194194af2b03ce2fa",
            "2939a6c917544effb6d30290e02750c2",
            "b4e7b22a14ab4de39a6be5b9ac87a859",
            "81514c49084c4699acf2a19f176905fe",
            "a7f35632035245c7a808facbcde1faeb",
            "c0ad3dcde07c46dc9026973722d53b97",
            "e6dd5c7de93f45bcb5f8966d7df979a2",
            "2e78b25a5d5d4674b1d1f61708a7f8bb",
            "b97a5dca9f6141b08b81ff1a58d68919",
            "45957dbb65994da793d90e4632a7e0d9",
            "d15ffa6b1da143e8baf62041b23662cc",
            "817890f3aef0421a83b33b3531c8ac1b",
            "9e49685857a045fbbcc7378a5fdbaa50",
            "018fe2a8ea194860934fe72fd989b067",
            "98162db40afa4484a4321d58fd9d2848",
            "71cba57f6c4c4e53b2c390ed3fa0bcbc",
            "1cedd010d23c4709ad213689552198de",
            "f60a867c6dd445e39103a549f176bb10",
            "6e2e2c2a86014b85af276d407cd32ed6",
            "c21f1c59e8c04770b2952383125dec27",
            "d5875aca8911496b9b96367b6a8e8852",
            "146d51dcd77f41518c064c23941569c3",
            "670ffe1337604111be17198d8b087155",
            "7fd60f1fbc0042198decd1113f697fad",
            "14d341830ed042219b6bd40b748a994c",
            "88871391967848f18068333f6a8a6530",
            "309e721a2df84da39090d74931655bc2",
            "cd54965515904b2dacb44113735df90d",
            "ffcf7b7524be47ec9f50726299d22d9e",
            "b3637041f9ab41fdb0bc409c33c8f921",
            "8682ad4b47de444b968ba2e1992ddb65",
            "d9a8ed1d734f49ffaf5e59138272abfd",
            "afe03557c5df47afb4c3483ca18f3de2",
            "d883e3b74b8343c78642b355e51b590c",
            "41a3d446c21b40c28f5ad0328ef75adc",
            "bc73d68234e54b83a053797d64ba24fd",
            "e95a9bed6ee24ab4916528200a033764",
            "ec7c526f8e3f43a094eb2d3bf87fe0df"
          ]
        },
        "id": "9v9oMoJX7Xjs"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading and preparing dataset clinc_oos/small to /root/.cache/huggingface/datasets/clinc_oos/small/1.0.0/abcc41d382f8137f039adc747af44714941e8196e845dfbdd8ae7a7e020e6ba1...\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading data:   0%|          | 0.00/217k [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "d2b46b2844f3474899145a75ffb4c571"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Generating train split:   0%|          | 0/7600 [00:00<?, ? examples/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "10d5cd30a0e54c5ba7b6df6b8ef580a0"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Generating validation split:   0%|          | 0/3100 [00:00<?, ? examples/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "c0ad3dcde07c46dc9026973722d53b97"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Generating test split:   0%|          | 0/5500 [00:00<?, ? examples/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "1cedd010d23c4709ad213689552198de"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Dataset clinc_oos downloaded and prepared to /root/.cache/huggingface/datasets/clinc_oos/small/1.0.0/abcc41d382f8137f039adc747af44714941e8196e845dfbdd8ae7a7e020e6ba1. Subsequent calls will reuse this data.\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "  0%|          | 0/3 [00:00<?, ?it/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "cd54965515904b2dacb44113735df90d"
            }
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "clinc_oos_dataset"
      ],
      "metadata": {
        "outputId": "5f63daec-d372-4acc-aad8-d660d528ff3c",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Nd71pRe97Xjt"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "DatasetDict({\n",
              "    train: Dataset({\n",
              "        features: ['text', 'intent'],\n",
              "        num_rows: 7600\n",
              "    })\n",
              "    validation: Dataset({\n",
              "        features: ['text', 'intent'],\n",
              "        num_rows: 3100\n",
              "    })\n",
              "    test: Dataset({\n",
              "        features: ['text', 'intent'],\n",
              "        num_rows: 5500\n",
              "    })\n",
              "})"
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Podemos ver que el dataset es una estructura DatasetDict que puede verse como un diccionario. El diccionario tiene tres claves que son train, validation y test que indican respectivamente los conjuntos de entrenamiento, validación y test (estas claves pueden variar dependiendo del dataset). Cada uno de estos subconjuntos es un Dataset que puede verse como una lista. Podemos ver por ejemplo la primera frase del conjunto de entrenamiento del siguiente modo."
      ],
      "metadata": {
        "id": "iys3O7jz7Xjt"
      }
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "e2OkF7097Xju"
      },
      "source": [
        "## Carga de datos\n",
        "\n",
        "Cargamos a continuación el dataset en distintos dataframes de pandas (el formato que puede leer la librería de FastAI)."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Sv7ZtBqs7Xju"
      },
      "outputs": [],
      "source": [
        "train_df = clinc_oos_dataset[\"train\"].to_pandas()\n",
        "valid_df = clinc_oos_dataset[\"validation\"].to_pandas()\n",
        "test_df = clinc_oos_dataset[\"test\"].to_pandas()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "go0-ENcw7Xju"
      },
      "source": [
        "Podemos ver el contenido del dataset usando el siguiente comando."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Y1KMKZgN7Xjv"
      },
      "source": [
        "Del dataset nos interesan dos campos: `text` (que contiene el poema) y `intent` ."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "uw3h-1zPC2kw"
      },
      "source": [
        "## Modelo de lenguaje\n",
        "\n",
        "El proceso a seguir para hacer fine-tuning sobre el modelo de lenguaje de FastAI es análogo al visto en prácticas anteriores. Comenzamos creando un `DataBlock` a partir de nuestro dataframe. "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "id": "AW74KDBJC2kw"
      },
      "outputs": [],
      "source": [
        "db_lm = DataBlock(\n",
        "    blocks=TextBlock.from_df('text', is_lm=True,max_vocab=100000), # Indicamos que vamos a trabajar con un modelo de lenguaje\n",
        "    get_items=ColReader('text'), # Indicamos donde estará el texto dentro del dataframe\n",
        "    splitter=RandomSplitter(0.1) # Partimos el dataset en entrenamiento y validación\n",
        ")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "leP76qI4C2kw"
      },
      "source": [
        "Creamos ahora nuestro `dataloader` (esto puede llevar varios segundos)."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "metadata": {
        "id": "e_Uzg_gRC2kx"
      },
      "outputs": [],
      "source": [
        "dls_lm = db_lm.dataloaders(train_valid_df, bs=128, seq_len=80)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ESmIQ-pkC2kx"
      },
      "source": [
        "Podemos ahora mostrar un batch de este `dataloader`. Como podemos apreciar, la entrada del modelo es una frase, y la salida es dicha frase desplazada una posición a la derecha."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "id": "WLkzmtyYC2kx",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 216
        },
        "outputId": "d778b76f-edb1-4058-aab6-3b1d1e5274c2"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>text</th>\n",
              "      <th>text_</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>xxbos xxunk when is my flight going to land xxbos do i have meetings with anyone today xxbos how soon does my visa card expire xxbos set a new timer for ten xxunk xxbos what can i get for 10 usd in cad xxbos when was the last date that i was paid xxbos can you help me find someone to look at my car 's engine because the light is on xxbos tell the restaurant to cancel my reservation</td>\n",
              "      <td>xxunk when is my flight going to land xxbos do i have meetings with anyone today xxbos how soon does my visa card expire xxbos set a new timer for ten xxunk xxbos what can i get for 10 usd in cad xxbos when was the last date that i was paid xxbos can you help me find someone to look at my car 's engine because the light is on xxbos tell the restaurant to cancel my reservation xxbos</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>beef and xxunk xxbos what can i do if i lost my luggage xxbos can you get me a recipe for mashed potatoes xxbos share the nutrition info for pizza with me xxbos can my 401k rollover or not xxbos what is the lowest amount i can pay for my cable bill xxbos are there a lot of calories in muffins xxbos cancel that xxbos how long until my oil needs to be changed xxbos what month does my card</td>\n",
              "      <td>and xxunk xxbos what can i do if i lost my luggage xxbos can you get me a recipe for mashed potatoes xxbos share the nutrition info for pizza with me xxbos can my 401k rollover or not xxbos what is the lowest amount i can pay for my cable bill xxbos are there a lot of calories in muffins xxbos cancel that xxbos how long until my oil needs to be changed xxbos what month does my card stop</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>"
            ]
          },
          "metadata": {}
        }
      ],
      "source": [
        "dls_lm.show_batch(max_n=2)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LJlEUlkcC2kx"
      },
      "source": [
        "Creamos ahora nuestro `Learner`."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "metadata": {
        "id": "rOARW1odC2kx",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 37
        },
        "outputId": "84ce2684-d0c6-4a8e-ca96-265cf5ee401c"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "<style>\n",
              "    /* Turns off some styling */\n",
              "    progress {\n",
              "        /* gets rid of default border in Firefox and Opera. */\n",
              "        border: none;\n",
              "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
              "        background-size: auto;\n",
              "    }\n",
              "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
              "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
              "    }\n",
              "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
              "        background: #F44336;\n",
              "    }\n",
              "</style>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "    <div>\n",
              "      <progress value='105070592' class='' max='105067061' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
              "      100.00% [105070592/105067061 00:01&lt;00:00]\n",
              "    </div>\n",
              "    "
            ]
          },
          "metadata": {}
        }
      ],
      "source": [
        "learn = language_model_learner(\n",
        "    dls_lm, # El dataloader que usamos\n",
        "    AWD_LSTM, # La arquitectura que es la misma usada en la práctica anterior\n",
        "    drop_mult=0.3, # Aplicamos dropout para evitar el sobreajuste\n",
        "    metrics=[accuracy, Perplexity()] # Como métricas usamos la accuracy y la perplexity.\n",
        ").to_fp16()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GZeI2zGAC2ky"
      },
      "source": [
        "Y por último entrenamos el modelo. "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "metadata": {
        "id": "wRXPjZB9C2ky",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 566
        },
        "outputId": "17be9a39-4039-4989-d74d-74bdb28a8ab9"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.9/dist-packages/torch/amp/autocast_mode.py:204: UserWarning: User provided device_type of 'cuda', but CUDA is not available. Disabling\n",
            "  warnings.warn('User provided device_type of \\'cuda\\', but CUDA is not available. Disabling')\n",
            "/usr/local/lib/python3.9/dist-packages/torch/cuda/amp/grad_scaler.py:120: UserWarning: torch.cuda.amp.GradScaler is enabled, but CUDA is not available.  Disabling.\n",
            "  warnings.warn(\"torch.cuda.amp.GradScaler is enabled, but CUDA is not available.  Disabling.\")\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "<style>\n",
              "    /* Turns off some styling */\n",
              "    progress {\n",
              "        /* gets rid of default border in Firefox and Opera. */\n",
              "        border: none;\n",
              "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
              "        background-size: auto;\n",
              "    }\n",
              "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
              "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
              "    }\n",
              "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
              "        background: #F44336;\n",
              "    }\n",
              "</style>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: left;\">\n",
              "      <th>epoch</th>\n",
              "      <th>train_loss</th>\n",
              "      <th>valid_loss</th>\n",
              "      <th>accuracy</th>\n",
              "      <th>perplexity</th>\n",
              "      <th>time</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <td>0</td>\n",
              "      <td>5.696906</td>\n",
              "      <td>4.259863</td>\n",
              "      <td>0.198017</td>\n",
              "      <td>70.800308</td>\n",
              "      <td>04:09</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.9/dist-packages/torch/amp/autocast_mode.py:204: UserWarning: User provided device_type of 'cuda', but CUDA is not available. Disabling\n",
            "  warnings.warn('User provided device_type of \\'cuda\\', but CUDA is not available. Disabling')\n",
            "/usr/local/lib/python3.9/dist-packages/torch/cuda/amp/grad_scaler.py:120: UserWarning: torch.cuda.amp.GradScaler is enabled, but CUDA is not available.  Disabling.\n",
            "  warnings.warn(\"torch.cuda.amp.GradScaler is enabled, but CUDA is not available.  Disabling.\")\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "<style>\n",
              "    /* Turns off some styling */\n",
              "    progress {\n",
              "        /* gets rid of default border in Firefox and Opera. */\n",
              "        border: none;\n",
              "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
              "        background-size: auto;\n",
              "    }\n",
              "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
              "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
              "    }\n",
              "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
              "        background: #F44336;\n",
              "    }\n",
              "</style>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: left;\">\n",
              "      <th>epoch</th>\n",
              "      <th>train_loss</th>\n",
              "      <th>valid_loss</th>\n",
              "      <th>accuracy</th>\n",
              "      <th>perplexity</th>\n",
              "      <th>time</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <td>0</td>\n",
              "      <td>3.862475</td>\n",
              "      <td>3.561986</td>\n",
              "      <td>0.299479</td>\n",
              "      <td>35.233109</td>\n",
              "      <td>04:13</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1</td>\n",
              "      <td>3.606975</td>\n",
              "      <td>3.192473</td>\n",
              "      <td>0.342748</td>\n",
              "      <td>24.348566</td>\n",
              "      <td>04:14</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2</td>\n",
              "      <td>3.388048</td>\n",
              "      <td>2.945242</td>\n",
              "      <td>0.369992</td>\n",
              "      <td>19.015263</td>\n",
              "      <td>04:16</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3</td>\n",
              "      <td>3.182660</td>\n",
              "      <td>2.770793</td>\n",
              "      <td>0.400341</td>\n",
              "      <td>15.971290</td>\n",
              "      <td>04:20</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>4</td>\n",
              "      <td>3.000447</td>\n",
              "      <td>2.687873</td>\n",
              "      <td>0.407853</td>\n",
              "      <td>14.700380</td>\n",
              "      <td>04:18</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>5</td>\n",
              "      <td>2.841043</td>\n",
              "      <td>2.651250</td>\n",
              "      <td>0.414864</td>\n",
              "      <td>14.171744</td>\n",
              "      <td>04:02</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>6</td>\n",
              "      <td>2.706536</td>\n",
              "      <td>2.625234</td>\n",
              "      <td>0.423878</td>\n",
              "      <td>13.807803</td>\n",
              "      <td>04:02</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>7</td>\n",
              "      <td>2.591934</td>\n",
              "      <td>2.609441</td>\n",
              "      <td>0.425581</td>\n",
              "      <td>13.591455</td>\n",
              "      <td>04:03</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>8</td>\n",
              "      <td>2.498425</td>\n",
              "      <td>2.608087</td>\n",
              "      <td>0.427284</td>\n",
              "      <td>13.573062</td>\n",
              "      <td>04:03</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>9</td>\n",
              "      <td>2.420030</td>\n",
              "      <td>2.607645</td>\n",
              "      <td>0.427784</td>\n",
              "      <td>13.567060</td>\n",
              "      <td>04:02</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>"
            ]
          },
          "metadata": {}
        }
      ],
      "source": [
        "learn.fine_tune(10,base_lr=2e-2)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Pt2h96oNC2ky"
      },
      "source": [
        "Una vez entrenado el modelo guardamos el `encoder` que usaremos luego para nuestro modelo de clasificación (esto es análogo a lo que vimos para los modelos de clasificación de imágenes)."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "metadata": {
        "id": "KHT9LdeBC2ky"
      },
      "outputs": [],
      "source": [
        "learn.save_encoder('finetuned')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MuF9fJQeC2ky"
      },
      "source": [
        "## Entrenando un modelo de clasificación\n",
        "\n",
        "Pasamos ahora a crear nuestro modelo de clasificación de texto. El proceso será el mismo que el que vimos en la práctica anterior con la diferencia de que antes de empezar a entrenar el modelo cargaremos el `encoder` guardado en el paso anterior.\n",
        "\n",
        "Comenzamos definiendo un `DataBlock` que se creará a partir de nuestro dataframe `df`.  "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 25,
      "metadata": {
        "id": "GWQt6jF0C2ky"
      },
      "outputs": [],
      "source": [
        "sentiment_clas = DataBlock(\n",
        "    blocks=(TextBlock.from_df('text', vocab=dls_lm.vocab), # La entrada del modelo es texto usando el mismo \n",
        "                                                           # vocabulario que en el modelo de lenguaje \n",
        "            CategoryBlock), #, y la salida una clase \n",
        "    get_x=ColReader('text'),  # Indicamos donde estará el texto dentro del dataframe\n",
        "    get_y=ColReader('intent'), # Indicamos cómo extraer la clase del dataframe\n",
        "    splitter=ColSplitter('set') # Partimos el dataset en entrenamiento y validación\n",
        ")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "A_yKEURkC2kz"
      },
      "source": [
        "Ahora definimos nuestro dataloader a partir del DataBlock que acabamos de crear. "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 26,
      "metadata": {
        "id": "9ORL1uwgC2kz",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 17
        },
        "outputId": "a3222cb5-d572-497d-d4ab-c8b4b937b1b1"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "<style>\n",
              "    /* Turns off some styling */\n",
              "    progress {\n",
              "        /* gets rid of default border in Firefox and Opera. */\n",
              "        border: none;\n",
              "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
              "        background-size: auto;\n",
              "    }\n",
              "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
              "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
              "    }\n",
              "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
              "        background: #F44336;\n",
              "    }\n",
              "</style>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": []
          },
          "metadata": {}
        }
      ],
      "source": [
        "dls = sentiment_clas.dataloaders(train_valid_df, bs=64)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Hy2jPfJjC2kz"
      },
      "source": [
        "Podemos mostrar un batch de nuestro dataloader. "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 27,
      "metadata": {
        "id": "GB8nwXKbC2kz",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 112
        },
        "outputId": "59e115b9-68b5-4960-de61-8462cc3411df"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>text</th>\n",
              "      <th>category</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>xxbos can you tell me what to do as i am in the airport and have been for some time and there is still no xxunk of my xxunk</td>\n",
              "      <td>113</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>xxbos i xxunk like a rental car in denver xxunk between january 1st and january 3rd and i xxunk like a ford if possible</td>\n",
              "      <td>40</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>"
            ]
          },
          "metadata": {}
        }
      ],
      "source": [
        "dls.show_batch(max_n=2)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WARPAlm5C2kz"
      },
      "source": [
        "Pasamos ahora a crear nuestro `learner` usando el método `text_classifier_learner` al que pasamos como arquitectura de red la arquitectura [AWD_LSTM](https://arxiv.org/abs/1708.02182), además aplicamos dropout a nuestro modelo. "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 28,
      "metadata": {
        "id": "07dPozDGC2kz"
      },
      "outputs": [],
      "source": [
        "callbacks = [ShowGraphCallback(),\n",
        "             SaveModelCallback()]\n",
        "\n",
        "learnClass = text_classifier_learner(dls, AWD_LSTM, drop_mult=0.5, metrics=accuracy,cbs=callbacks).to_fp16()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UE76n_MkC2kz"
      },
      "source": [
        "Cargamos a continuación el `encoder` del modelo de lenguaje. "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 29,
      "metadata": {
        "id": "mUHHd338C2k0"
      },
      "outputs": [],
      "source": [
        "learnClass = learnClass.load_encoder('finetuned')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lWPtCffDC2k0"
      },
      "source": [
        "Ahora podemos utilizar toda la funcionalidad que ya vimos para clasificación de imágenes. Por ejemplo, podemos buscar un learning rate adecuado para entrenar nuestro modelo. "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 30,
      "metadata": {
        "id": "mF9qM3rgC2k0",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 541
        },
        "outputId": "31592431-2195-41cd-c129-f18a0f097263"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.9/dist-packages/torch/amp/autocast_mode.py:204: UserWarning: User provided device_type of 'cuda', but CUDA is not available. Disabling\n",
            "  warnings.warn('User provided device_type of \\'cuda\\', but CUDA is not available. Disabling')\n",
            "/usr/local/lib/python3.9/dist-packages/torch/cuda/amp/grad_scaler.py:120: UserWarning: torch.cuda.amp.GradScaler is enabled, but CUDA is not available.  Disabling.\n",
            "  warnings.warn(\"torch.cuda.amp.GradScaler is enabled, but CUDA is not available.  Disabling.\")\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "<style>\n",
              "    /* Turns off some styling */\n",
              "    progress {\n",
              "        /* gets rid of default border in Firefox and Opera. */\n",
              "        border: none;\n",
              "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
              "        background-size: auto;\n",
              "    }\n",
              "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
              "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
              "    }\n",
              "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
              "        background: #F44336;\n",
              "    }\n",
              "</style>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": []
          },
          "metadata": {}
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "SuggestedLRs(valley=0.004365158267319202)"
            ]
          },
          "metadata": {},
          "execution_count": 30
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "learnClass.lr_find()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mRUbbfWYC2k0"
      },
      "source": [
        "Y a continuación aplicar fine tuning. "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 31,
      "metadata": {
        "id": "wqN1ap8uC2k0",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "a5e241da-de4c-4773-dfa3-229c7f2be2d5"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.9/dist-packages/torch/amp/autocast_mode.py:204: UserWarning: User provided device_type of 'cuda', but CUDA is not available. Disabling\n",
            "  warnings.warn('User provided device_type of \\'cuda\\', but CUDA is not available. Disabling')\n",
            "/usr/local/lib/python3.9/dist-packages/torch/cuda/amp/grad_scaler.py:120: UserWarning: torch.cuda.amp.GradScaler is enabled, but CUDA is not available.  Disabling.\n",
            "  warnings.warn(\"torch.cuda.amp.GradScaler is enabled, but CUDA is not available.  Disabling.\")\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "<style>\n",
              "    /* Turns off some styling */\n",
              "    progress {\n",
              "        /* gets rid of default border in Firefox and Opera. */\n",
              "        border: none;\n",
              "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
              "        background-size: auto;\n",
              "    }\n",
              "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
              "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
              "    }\n",
              "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
              "        background: #F44336;\n",
              "    }\n",
              "</style>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: left;\">\n",
              "      <th>epoch</th>\n",
              "      <th>train_loss</th>\n",
              "      <th>valid_loss</th>\n",
              "      <th>accuracy</th>\n",
              "      <th>time</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <td>0</td>\n",
              "      <td>2.663182</td>\n",
              "      <td>1.729634</td>\n",
              "      <td>0.542903</td>\n",
              "      <td>01:56</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Better model found at epoch 0 with valid_loss value: 1.7296338081359863.\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x400 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.9/dist-packages/torch/amp/autocast_mode.py:204: UserWarning: User provided device_type of 'cuda', but CUDA is not available. Disabling\n",
            "  warnings.warn('User provided device_type of \\'cuda\\', but CUDA is not available. Disabling')\n",
            "/usr/local/lib/python3.9/dist-packages/torch/cuda/amp/grad_scaler.py:120: UserWarning: torch.cuda.amp.GradScaler is enabled, but CUDA is not available.  Disabling.\n",
            "  warnings.warn(\"torch.cuda.amp.GradScaler is enabled, but CUDA is not available.  Disabling.\")\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "<style>\n",
              "    /* Turns off some styling */\n",
              "    progress {\n",
              "        /* gets rid of default border in Firefox and Opera. */\n",
              "        border: none;\n",
              "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
              "        background-size: auto;\n",
              "    }\n",
              "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
              "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
              "    }\n",
              "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
              "        background: #F44336;\n",
              "    }\n",
              "</style>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: left;\">\n",
              "      <th>epoch</th>\n",
              "      <th>train_loss</th>\n",
              "      <th>valid_loss</th>\n",
              "      <th>accuracy</th>\n",
              "      <th>time</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <td>0</td>\n",
              "      <td>0.986069</td>\n",
              "      <td>0.732839</td>\n",
              "      <td>0.802258</td>\n",
              "      <td>04:17</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1</td>\n",
              "      <td>0.844675</td>\n",
              "      <td>0.916558</td>\n",
              "      <td>0.787419</td>\n",
              "      <td>04:15</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2</td>\n",
              "      <td>0.681875</td>\n",
              "      <td>0.846208</td>\n",
              "      <td>0.808387</td>\n",
              "      <td>04:17</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3</td>\n",
              "      <td>0.512396</td>\n",
              "      <td>0.730708</td>\n",
              "      <td>0.838064</td>\n",
              "      <td>04:19</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>4</td>\n",
              "      <td>0.372664</td>\n",
              "      <td>0.731234</td>\n",
              "      <td>0.854839</td>\n",
              "      <td>04:17</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>5</td>\n",
              "      <td>0.289928</td>\n",
              "      <td>0.596011</td>\n",
              "      <td>0.877419</td>\n",
              "      <td>04:20</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>6</td>\n",
              "      <td>0.168818</td>\n",
              "      <td>0.548993</td>\n",
              "      <td>0.897419</td>\n",
              "      <td>04:26</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>7</td>\n",
              "      <td>0.100016</td>\n",
              "      <td>0.543601</td>\n",
              "      <td>0.898387</td>\n",
              "      <td>04:22</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>8</td>\n",
              "      <td>0.064139</td>\n",
              "      <td>0.524489</td>\n",
              "      <td>0.905806</td>\n",
              "      <td>04:21</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>9</td>\n",
              "      <td>0.046383</td>\n",
              "      <td>0.506498</td>\n",
              "      <td>0.907097</td>\n",
              "      <td>04:22</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Better model found at epoch 0 with valid_loss value: 0.7328385710716248.\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x400 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAg0AAAFfCAYAAADNtv/1AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAABaIElEQVR4nO3dd3wUdf7H8ddukt000oA0SCD0HnoEGyoa0UPBhogCtrvz0FMRC+dP7GI/LChiw7tTURRREVGMFJEmJTQhdBIgCTWd1J3fHwtLlgTIhiSbLO/n4zEPMt/5zuxn5gHsJ9/5FpNhGAYiIiIiZ2B2dwAiIiLSMChpEBERkSpR0iAiIiJVoqRBREREqkRJg4iIiFSJkgYRERGpEiUNIiIiUiXe7g6gJthsNvbt20ejRo0wmUzuDkdERKTBMAyD3NxcoqOjMZtP35bgEUnDvn37iImJcXcYIiIiDVZaWhrNmzc/bR2PSBoaNWoEQLN7pvHni0PdHI2IiEjDkZOTQ0xMjOO79HQ8Imk4/krCbPXHLyAQHy911RAREXFFVV7ve9y3a0FRmbtDEBER8UgelzTkFZe6OwQRERGP5HFJQ36RkgYREZHa4BF9GsrLU9IgIuJxysrKKCkpcXcYDZaPjw9eXl5nfR2PSxrUp0FExHMYhkFGRgZZWVnuDqXBCwkJITIy8qzmM/K8pEF9GkREPMbxhCE8PBx/f39N4FcNhmFQUFDA/v37AYiKiqr2tTwuaSgpM9wdgoiI1ICysjJHwtC4cWN3h9Og+fn5AbB//37Cw8Or/arC4zpClpTZ3B2CiIjUgON9GPz9/d0ciWc4/hzPpm+IxyUNxUoaREQ8il5J1IyaeI4elzSopUFERKR2eF7SUKqkQUREpDZ4XNKg1xMiIuJJWrZsyaRJk9wdBqDREyIiIjVuwIABdO/evUa+7P/44w8CAgLOPqga4HFJQ7FeT4iISD1nGAZlZWV4e5/5a7hp06Z1EFHVeNzrCXWEFBHxXIZhUFBcWuebYVS9FXv06NEsXLiQN954A5PJhMlkYtq0aZhMJn788Ud69eqF1Wpl8eLFbN++nWuvvZaIiAgCAwPp06cPv/zyi9P1Tn49YTKZ+OCDDxg6dCj+/v60bduW7777rqYe8Wl5XEuDkgYREc91tKSMThN+qvPP/fOZRPwtVfvKfOONN9iyZQtdunThmWeeAWDjxo0APPbYY7z66qu0atWK0NBQ0tLSuOqqq3j++eexWq385z//YfDgwaSkpBAbG3vKz3j66ad5+eWXeeWVV3jrrbcYMWIEu3fvJiws7Oxv9jQ8sKVBfRpERMR9goODsVgs+Pv7ExkZSWRkpGMGxmeeeYbLL7+c1q1bExYWRnx8PH/729/o0qULbdu25dlnn6V169ZnbDkYPXo0w4cPp02bNrzwwgvk5eWxYsWKWr83j2tp0OgJERHP5efjxZ/PJLrlc2tC7969nfbz8vJ46qmn+OGHH0hPT6e0tJSjR4+Smpp62ut069bN8XNAQABBQUGOtSVqk+clDeoIKSLisUwmU5VfE9RHJ4+CGDduHPPmzePVV1+lTZs2+Pn5ccMNN1BcXHza6/j4+Djtm0wmbLba//5ruE/+FNSnQURE3M1isVBWVnbGer///jujR49m6NChgL3lYdeuXbUcXfV5YJ8GJQ0iIuJeLVu2ZPny5ezatYuDBw+eshWgbdu2zJw5k+TkZNauXcstt9xSJy0G1eVxSYNeT4iIiLuNGzcOLy8vOnXqRNOmTU/ZR+H1118nNDSU/v37M3jwYBITE+nZs2cdR1t1Hvd6okhJg4iIuFm7du1YunSpU9no0aMr1GvZsiW//vqrU9mYMWOc9k9+XVHZnBFZWVnVitNVHtfScLT4zO+QRERExHWelzSUKGkQERGpDR6XNBQqaRAREakVHpg0qE+DiIhIbfDApEEtDSIiIrXBpaRh4sSJ9OnTh0aNGhEeHs6QIUNISUk543kzZsygQ4cO+Pr60rVrV+bMmeN03DAMJkyYQFRUFH5+fgwcOJCtW7e6difHqE+DiIhI7XApaVi4cCFjxoxh2bJlzJs3j5KSEq644gry8/NPec6SJUsYPnw4d955J2vWrGHIkCEMGTKEDRs2OOq8/PLLvPnmm0yZMoXly5cTEBBAYmIihYWFLt/Q0ZIyl5YwFRERkaoxGWfxDXvgwAHCw8NZuHAhF110UaV1hg0bRn5+PrNnz3aUnXfeeXTv3p0pU6ZgGAbR0dE89NBDjBs3DoDs7GwiIiKYNm0aN998c4VrFhUVUVRU5NjPyckhJiaGmAe+xGz1J+W5K7F618ziIiIi4h6FhYXs3LmTuLg4fH193R1Og3eq55mTk0NwcDDZ2dkEBQWd9hpn1achOzsb4LTrdy9dupSBAwc6lSUmJjomvdi5cycZGRlOdYKDg0lISKgwMcZxEydOJDg42LHFxMQ4HS8sVmdIERFpuFq2bMmkSZMc+yaTiVmzZp2y/q5duzCZTCQnJ9dqXNVOGmw2Gw888ADnn38+Xbp0OWW9jIwMIiIinMoiIiLIyMhwHD9edqo6Jxs/fjzZ2dmOLS0tzel4fnGpy/cjIiJSX6WnpzNo0CB3h1H9aaTHjBnDhg0bWLx4cU3GUyVWqxWr1XrK4/lFShpERMRzREZGujsEoJotDffeey+zZ89m/vz5NG/e/LR1IyMjyczMdCrLzMx0PIDjf56ujqtylTSIiIibTJ06lejo6AqrVV577bXccccdbN++nWuvvZaIiAgCAwPp06cPv/zyy2mvefLriRUrVtCjRw98fX3p3bs3a9asqY1bqcClpMEwDO69916++eYbfv31V+Li4s54Tr9+/UhKSnIqmzdvHv369QMgLi6OyMhIpzo5OTksX77cUcdVeYVKGkREPJJhQHF+3W8ujBm48cYbOXToEPPnz3eUHT58mLlz5zJixAjy8vK46qqrSEpKYs2aNVx55ZUMHjz4lCthniwvL4+//OUvdOrUiVWrVvHUU085BhLUNpdeT4wZM4bPPvuMb7/9lkaNGjn6HAQHB+Pn5wfAyJEjadasGRMnTgTg/vvv5+KLL+a1117j6quvZvr06axcuZKpU6cC9uzpgQce4LnnnqNt27bExcXxxBNPEB0dzZAhQ6p1U3lqaRAR8UwlBfBCdN1/7r/2gSWgSlVDQ0MZNGgQn332GZdddhkAX331FU2aNOGSSy7BbDYTHx/vqP/ss8/yzTff8N1333Hvvfee8fqfffYZNpuNDz/8EF9fXzp37syePXu45557qndvLnCppeHdd98lOzubAQMGEBUV5di++OILR53U1FTS09Md+/379+ezzz5j6tSpxMfH89VXXzFr1iynzpOPPPII9913H3/961/p06cPeXl5zJ07t9pDbNTSICIi7jRixAi+/vprx/QAn376KTfffDNms5m8vDzGjRtHx44dCQkJITAwkE2bNlW5pWHTpk1069bN6Tuyui3zrnKppaEqUzosWLCgQtmNN97IjTfeeMpzTCYTzzzzDM8884wr4ZyS+jSIiHgoH3/7b/3u+FwXDB48GMMw+OGHH+jTpw+//fYb//73vwEYN24c8+bN49VXX6VNmzb4+flxww03UFxcXBuR16hqj56ozzR6QkTEQ5lMVX5N4E6+vr5cd911fPrpp2zbto327dvTs2dPAH7//XdGjx7N0KFDAXsfhV27dlX52h07duS///0vhYWFjtaGZcuW1fg9VMbjFqwCKCrV+hMiIuJeI0aM4IcffuCjjz5ixIgRjvK2bdsyc+ZMkpOTWbt2LbfcckuFkRanc8stt2Aymbj77rv5888/mTNnDq+++mpt3EIFnpk0aHlsERFxs0svvZSwsDBSUlK45ZZbHOWvv/46oaGh9O/fn8GDB5OYmOhohaiKwMBAvv/+e9avX0+PHj14/PHHeemll2rjFirwyNcTRaVKGkRExL3MZjP79lXsf9GyZUt+/fVXp7IxY8Y47Z/8uuLkPoXnnXdehSmj62KxRs9sadDrCRERkRrnkUlDsVoaREREapxHJg25haWUlClxEBERqUkemTQkbd7Ppa8toLBErylERERqikcmDQBph4+yaMsBd4chIiLiMTw2aQDYfiDf3SGIiMhZcmUOAzm1mniOHjXk0ttsovwjKSjWzJAiIg2VxWJxDFts2rQpFosFk8nk7rAaHMMwKC4u5sCBA5jNZiwWS7Wv5VFJg9XHzNFy+/lF6tMgItJQmc1m4uLiSE9Pr3S+A3GNv78/sbGxmM3Vf8ngUUmDr7eZo+UaF9TSICLSsFksFmJjYyktLaWsTL8IVpeXlxfe3t5n3VLjUUmDv9WbI+XyhPxi/QUTEWnoTCYTPj4++Pj4uDuUc55HdYQM8XN+T1Og1S5FRERqjEclDaEBzllovl5PiIiI1BiPShr8vL2c9gv0ekJERKTGeFTSwEn9O/L1ekJERKTGeFTSYAKaBJ7o16CWBhERkZrjWUmDycS3917A/13dEVBLg4iISE3yrKQBaBbix1+6RQP2lgbDMNwblIiIiIfwqKTheJ8Gf6u9Q2SpzaBYS2SLiIjUCI9KGo73g/T3OTGK4qj6NYiIiNQIl5OGRYsWMXjwYKKjozGZTMyaNeu09UePHo3JZKqwde7c2VHnqaeeqnC8Q4cOLt/M8ekxvb3MWL3tt6ZZIUVERGqGy0lDfn4+8fHxTJ48uUr133jjDdLT0x1bWloaYWFh3HjjjU71Onfu7FRv8eLFrobmNOIywGqfIVuzQoqIiNQMl9eeGDRoEIMGDapy/eDgYIKDgx37s2bN4siRI9x+++3OgXh7ExkZ6Wo4zsplDX7HXlGopUFERKRm1Hmfhg8//JCBAwfSokULp/KtW7cSHR1Nq1atGDFiBKmpqae8RlFRETk5OU4bgKlc1hBwrDOkWhpERERqRp0mDfv27ePHH3/krrvucipPSEhg2rRpzJ07l3fffZedO3dy4YUXkpubW+l1Jk6c6GjBCA4OJiYmBoDyK376W+yNKGppEBERqRl1mjR88sknhISEMGTIEKfyQYMGceONN9KtWzcSExOZM2cOWVlZfPnll5VeZ/z48WRnZzu2tLQ04OQ+DcdaGrRolYiISI1wuU9DdRmGwUcffcRtt92GxWI5bd2QkBDatWvHtm3bKj1utVqxWq0VyittaShSS4OIiEhNqLOWhoULF7Jt2zbuvPPOM9bNy8tj+/btREVFufQZTn0aLGppEBERqUkuJw15eXkkJyeTnJwMwM6dO0lOTnZ0XBw/fjwjR46scN6HH35IQkICXbp0qXBs3LhxLFy4kF27drFkyRKGDh2Kl5cXw4cPd+1myt2Nv1UtDSIiIjXJ5dcTK1eu5JJLLnHsjx07FoBRo0Yxbdo00tPTK4x8yM7O5uuvv+aNN96o9Jp79uxh+PDhHDp0iKZNm3LBBRewbNkymjZt6mJ0amkQERGpLS4nDQMGDDjtIlDTpk2rUBYcHExBQcEpz5k+fbqrYVSq8tETShpERERqgkeuPQHQyNeeNGQVlLgnGBEREQ/jWUlDuayhVdMAALZkVj7Xg4iIiLjGs5KGcm0NnaLsU1dvP5BPYYk6Q4qIiJwtz0oayrU0RARZCfX3ocxmsDUzz31BiYiIeAjPShrK/2wy0Sk6CIBNGTnuCUhERMSDeFbSUL6pAYgK9gPgka/WsSb1iDtCEhER8RgelTScLNB6YkTp0HeWuDESERGRhs+zkgbnhgaCfJ2nocjXMtkiIiLV5lFJg+mkrKGRr4/T/u/bDtZlOCIiIh7Fs5KGk1oaGp3U0rDrUH4dRiMiIuJZPCtpOGk/8KSk4WBecd0FIyIi4mE8K2mo0NLg/HriYG5RHUYjIiLiWTwsaTi5T4NzS8PMNXspKbPVZUgiIiIew7OShpP2Tx49AfDHzsN1E4yIiIiH8ayk4aSWhkCrT4U6G/dpdkgREZHq8Kyk4aT98q8n2oYHArBxX3YdRiQiIuI5PDpp8Ld4OX5uF9kIgFnJ+/hmzZ46jEpERMQzeFbSYDp5/0RBu/BGjp8f/GItmTmFdRWWiIiIR/CspKFCW8MJfVqGOu1ruWwRERHXVBxe0ICd3NIAMPMf/UnJyKVf68YkxIWx/NjoiT1HCuo4OhERkYbNo1oaOkYHVSjrGRvK8L6xmEwm3rutl6ND5O7DShpERERc4VFJw8COEac9HuJvYWT/lgBs2KtRFCIiIq7wqKShKnrEhACwNi0LwzDcG4yIiEgD4nLSsGjRIgYPHkx0dDQmk4lZs2adtv6CBQswmUwVtoyMDKd6kydPpmXLlvj6+pKQkMCKFStcDa1K2kU0wsfLRE5hKXuOHK2VzxAREfFELicN+fn5xMfHM3nyZJfOS0lJIT093bGFh4c7jn3xxReMHTuWJ598ktWrVxMfH09iYiL79+93NbwzsnibaXNs+OWmdM0OKSIiUlUuJw2DBg3iueeeY+jQoS6dFx4eTmRkpGMzm0989Ouvv87dd9/N7bffTqdOnZgyZQr+/v589NFHroZXJR2j7EnDn0oaREREqqzO+jR0796dqKgoLr/8cn7//XdHeXFxMatWrWLgwIEngjKbGThwIEuXLq30WkVFReTk5DhtrugUZR9loc6QIiIiVVfrSUNUVBRTpkzh66+/5uuvvyYmJoYBAwawevVqAA4ePEhZWRkREc4jHyIiIir0ezhu4sSJBAcHO7aYmBiXYjqvVWMAftt6kNzCkmrclYiIyLmn1pOG9u3b87e//Y1evXrRv39/PvroI/r378+///3val9z/PjxZGdnO7a0tDSXzu8cHUTrpgEUldp4f9GOaschIiJyLnHLkMu+ffuybds2AJo0aYKXlxeZmZlOdTIzM4mMjKz0fKvVSlBQkNPmCpPJxM19YgH46PddGnopIiJSBW5JGpKTk4mKigLAYrHQq1cvkpKSHMdtNhtJSUn069ev1mIYdWySp7yiUjK0eJWIiMgZubz2RF5enqOVAGDnzp0kJycTFhZGbGws48ePZ+/evfznP/8BYNKkScTFxdG5c2cKCwv54IMP+PXXX/n5558d1xg7diyjRo2id+/e9O3bl0mTJpGfn8/tt99eA7dYOYu3mVZNA9hxIJ8tmXlEBfvV2meJiIh4ApeThpUrV3LJJZc49seOHQvAqFGjmDZtGunp6aSmpjqOFxcX89BDD7F37178/f3p1q0bv/zyi9M1hg0bxoEDB5gwYQIZGRl0796duXPnVugcWdN6xoay40A+v205wMXtmtbqZ4mIiDR0JsMDXujn5OQQHBxMdna2S/0bZq7ew9gv15IQF8YXf6u9VyEiIiL1lSvfoefc2hPltWjsD8DeLE0nLSIicibndNLQPNSeNKRnF1JaZnNzNCIiIvXbOZ00NA20Emj1psxm8N3afe4OR0REpF47p5MGs9nEsD722SQn/rhZ8zWIiIicxjmdNACMu6I9XmYTB3KLWJ2a5e5wRERE6q1zPmnws3hxeUf70M43kra6ORoREZH665xPGgAeubI9AIu3HiBTs0OKiIhUSkkD0KppIL1bhGIzYM76dHeHIyIiUi8paThmQHv7jJBr1K9BRESkUkoajomPCQFg7Z4st8YhIiJSXylpOKZbsxAAdh8q4Eh+sXuDERERqYeUNBwT7O9DXJMAQK0NIiIilVHSUE5882AA1qZluzkSERGR+kdJQzldm4cAsHGfkgYREZGTKWkop11EIADbDuS5ORIREZH6R0lDOa2b2pOG1EMFlGjVSxERESdKGsqJCvYlwOJFqc1g96F8d4cjIiJSryhpKMdkMtE6/Ngriv16RSEiIlKekoaTdIhsBMDKXUfcHImIiEj9oqThJAPahwMwe126+jWIiIiUo6ThJJd1DKdxgIWMnEK1NoiIiJSjpOEkVm8vercMBTRfg4iISHlKGirRJdo+M6RWvBQRETnB5aRh0aJFDB48mOjoaEwmE7NmzTpt/ZkzZ3L55ZfTtGlTgoKC6NevHz/99JNTnaeeegqTyeS0dejQwdXQakz/No0BWLztIGU2w21xiIiI1CcuJw35+fnEx8czefLkKtVftGgRl19+OXPmzGHVqlVccsklDB48mDVr1jjV69y5M+np6Y5t8eLFroZWY+KbhxDk60320RItXiUiInKMt6snDBo0iEGDBlW5/qRJk5z2X3jhBb799lu+//57evTocSIQb28iIyNdDadWeHuZOa9VY37+M5PVu4/QMzbU3SGJiIi4XZ33abDZbOTm5hIWFuZUvnXrVqKjo2nVqhUjRowgNTX1lNcoKioiJyfHaatpxyd5+mrVHsbPXEdOYUmNf4aIiEhDUudJw6uvvkpeXh433XSToywhIYFp06Yxd+5c3n33XXbu3MmFF15Ibm5updeYOHEiwcHBji0mJqbG44xrHADA5oxcPl+RxnOz/6zxzxAREWlI6jRp+Oyzz3j66af58ssvCQ8Pd5QPGjSIG2+8kW7dupGYmMicOXPIysriyy+/rPQ648ePJzs727GlpaXVeKwtGvs77f++7VCNf4aIiEhD4nKfhuqaPn06d911FzNmzGDgwIGnrRsSEkK7du3Ytm1bpcetVitWq7U2wnRo2STAaX9v1lEWbTnARe2a1urnioiI1Fd10tLw+eefc/vtt/P5559z9dVXn7F+Xl4e27dvJyoqqg6iq1x4Iyu+Ps6PZ8xnq90UjYiIiPu5nDTk5eWRnJxMcnIyADt37iQ5OdnRcXH8+PGMHDnSUf+zzz5j5MiRvPbaayQkJJCRkUFGRgbZ2SdmWxw3bhwLFy5k165dLFmyhKFDh+Ll5cXw4cPP8vaqz2Qy0T6ikVNZbmEphqF5G0RE5NzkctKwcuVKevTo4RguOXbsWHr06MGECRMASE9Pdxr5MHXqVEpLSxkzZgxRUVGO7f7773fU2bNnD8OHD6d9+/bcdNNNNG7cmGXLltG0qXtfBXRtHlyh7GBesRsiERERcT+X+zQMGDDgtL9tT5s2zWl/wYIFZ7zm9OnTXQ2jTrQr19LQOMDCofxiUg8X0LRR7fanEBERqY+09sRp9Ig5MalTx6ggAK5/d4mmlhYRkXNSnY2eaIi6Ng/mqcGdCPG3UFJmY/G2gwAkp2XRq4VmiRQRkXOLWhrOYPT5cQzp0YwbejXHbLKXLd+pORtEROTco6ShikwmE89c2wWA/y7djU2vKERE5ByjpMEFN/ZuTqDVm/TsQj5dvtvd4YiIiNQpJQ0usHp7kdjZvhLnE99uJCO70M0RiYiI1B0lDS568PK2jp9/WJ/uxkhERETqlpIGFzUP9eepwZ0A+H7tPjdHIyIiUneUNFTDVV2j8DKbSE7LouVjP7Dk2FBMERERT6akoRrCg3y5uNxql7d8sNyN0YiIiNQNJQ3VdEWnCKf9nQfz3RSJiIhI3VDSUE2D46OJDPJ17C/boQmfRETEsylpqKYAqzc/3n8hw/vGAjBxzib2HClwc1QiIiK1R0nDWQgNsHDnBXEA5BSWct/na9wckYiISO1R0nCW2oQH8reLWgGwJjWLNv+aw3+X7nJvUCIiIrVASUMNGNqzmePnUpvBE99uZG1aFiVlNjdGJSIiUrOUNNSADpFB/N/VHZ3Krp38Ow/PWOumiERERGqekoYaMrp/SwZ1iXQqm5W8jy9XprkpIhERkZqlpKGGeHuZeffWXky5tadT+SNfrSPtsEZViIhIw6ekoYb1jA2tUHbrh5oxUkREGj4lDTUsPMgXPx8vp7LdhwrYfUgzRoqISMOmpKEWLHh4AK/c0I0dL1xF/9aNAZj+h/o2iIhIw6akoRZEBPlyY+8YzGYTI/u1BGDqoh38uS/HvYGJiIicBSUNtSyxcwQXtWtKmc3gn9PXMPrjFcxP2e/usERERFzmctKwaNEiBg8eTHR0NCaTiVmzZp3xnAULFtCzZ0+sVitt2rRh2rRpFepMnjyZli1b4uvrS0JCAitWrHA1tHrJZDLxz0vbYDbBtv15LEg5wF2frNTETyIi0uC4nDTk5+cTHx/P5MmTq1R/586dXH311VxyySUkJyfzwAMPcNddd/HTTz856nzxxReMHTuWJ598ktWrVxMfH09iYiL793vGb+S9W4bx6JUdHPtlNoOFKQfcGJGIiIjrTIZhGNU+2WTim2++YciQIaes8+ijj/LDDz+wYcMGR9nNN99MVlYWc+fOBSAhIYE+ffrw9ttvA2Cz2YiJieG+++7jscceq3DNoqIiioqKHPs5OTnExMSQnZ1NUFBQdW+nVhUUl/LPz5P5ZVMmAPHNg5k15nxMJpObIxMRkXNZTk4OwcHBVfoOrfU+DUuXLmXgwIFOZYmJiSxduhSA4uJiVq1a5VTHbDYzcOBAR52TTZw4keDgYMcWExNTezdQQ/wt3nwwqjcr/28gVm8za/dkc9cnK9lxIM/doYmIiFRJrScNGRkZREREOJVFRESQk5PD0aNHOXjwIGVlZZXWycjIqPSa48ePJzs727GlpTWc4YxNAq3cfr59Oe2kzfu57PWFLN9xyM1RiYiInFmDHD1htVoJCgpy2hqShxPbM7p/SwAMA17+KcW9AYmIiFRBrScNkZGRZGZmOpVlZmYSFBSEn58fTZo0wcvLq9I6kZHOC0B5Ci+ziaeu6czChwcAsGr3EdKzj7o3KBERkTOo9aShX79+JCUlOZXNmzePfv36AWCxWOjVq5dTHZvNRlJSkqOOp2rROICesSEA/LrZM0aKiIiI53I5acjLyyM5OZnk5GTAPqQyOTmZ1NRUwN7fYOTIkY76f//739mxYwePPPIImzdv5p133uHLL7/kwQcfdNQZO3Ys77//Pp988gmbNm3innvuIT8/n9tvv/0sb6/+u6yjvS/H3A2V998QERGpL7xdPWHlypVccskljv2xY8cCMGrUKKZNm0Z6erojgQCIi4vjhx9+4MEHH+SNN96gefPmfPDBByQmJjrqDBs2jAMHDjBhwgQyMjLo3r07c+fOrdA50hNd1TWK135O4betB9m4L5vO0cHuDklERKRSZzVPQ33hyhjT+ujez1Yze106t54Xy3NDuro7HBEROYfUq3ka5Mxu6RsLwKw1+ygoLnVzNCIiIpVT0lAPnNeqMS0a+5NXVMrsdenuDkdERKRSShrqAbPZxM197K0Nn69IPUPteqKsBDLWw9Ej7o5ERETqiMsdIaV23NCrOa/9nMKa1Cw2Z+TQIbKe9s0wDPjzW0h6Bg5vt5eFtYZmPSG6p/3PyG5g8XdvnCIiUuOUNNQTTRtZubxTBD9uyOCrlXv4v790cndIFe36HeZNgL0r7fvevlBaaE8eDm+H9TPs5SYvCO8I0T1OJBMRncHLx32xi4jIWVPSUI9c2z2aHzdkMG9TJo9f3bH+rICZuRF+eRq2HlvO3CcA+t8H/e+1v6bYtxr2rjn252rIy4DMDfZtzX/t53hZIbKrc4tE47Zg1hsyEZGGQkMu65H8olI6P2n/Yr6+Z3NeuynevQFlpcGCiZD8GWCA2Rt6jYaLHoFGp5lDI2efPXk4nkTsWw2F2RXrWRpBdHfnFomQWKgvyZKIyDnAle9QJQ31zIgPlvH7tkP4eJlY/1Qivj5edR/E0SPw2+uw/D0oK7KXdRoCl02Axq1dv55hwOEdsG+NPYnYuwrS10JpJett+Dc+0RJx/M/A8LO6HREROTUlDQ3Y0eIyOk6YC8A7I3pyVdeouvvwkqOwYir89tqJloEWF8Dlz0DzXjX7WWWlcGCzc2tE5kawVTJPRVBzaNajXDLRA3w1c6aISE1Q0tDAvfjjZqYs3E6flqF8+bd+td+3wVYGaz+H+S9Azl57WXgnGPg0tL287l4XlBTa+0GUf7VxcAtQyV/Rxm2cWySiuoGPX93EKSLiQZQ0NHCbM3IY9MZvGAa8fUsP/tItunY+yDBg68/wy1Ow/097WVBzuPRx6DYMzBVfjRSVlrE2LZteLULxMtdBMlGYY3+VUb5FIquSuSxMXhDRyTmRCO+oERsiImegpMEDPPntBj5Zupt+rRrz+V/Pq/kP2LPSPnxy9+/2fd8QuPAh6PtX8PE95Wkf/LaD537YxA29mvPqjW7qqJl/8ET/iOPJRH4lS4t7+9rnjCjfPyKstUZsiIiU48p3qIZc1lN/vbg1/1m2m6U7DrHjQB6tmgbWzIUPbrVPzLTpO/u+lxXO+ztc8CD4hZ7x9P8u2w3AV6v2EBXsy0NXtK+ZuFwR0MT+2qTt5fZ9w7C/VnEasZEMRdmwZ4V9O84aDNHx9r4anYdC03Z1H7+ISAOlloZ67M5pf5C0eT+j+7fkqWs6n93FcjNg4Uuw6hMwysBkhvhb4JLxENz8jKcXFJfi5+PF7dP+YEHKAUf5nH9eSKfoevjMbbZjIzbKvdZIX2ufjKq8iC725KHLdRDWyj2xioi4kV5PeIiFWw4w6qMVNPL1ZtX/XY7FuxrN6oU5sOQtWPo2lBTYy9oNsg+fjKjarJPbD+Qx6I3f6NosmFW77WtNmE1gM+DSDuF8OKp3/ZmI6nTKSuHAJvurmZQ5sP1X59EaUd3tyUPnofb5IkREzgFaGttDXNCmCeGNrOQWlvLjBhdXvywtts+z8GYPWPSyPWFo3gdGz4Fbplc5YQD4bcsBikttjoQB4JUb4vE2m/h1836m/5HmWmzu4uVtn5Wy9+0wYgaM2wrXvAWtLrF3pExPtvfzmNQVPhgIS9+xT1QlIiKAWhrqvTeTtvL6vC2EBVhY9MglBFrP0A3FZoONM+HXZ+HILntZ4zZw2ZPQcbDLwydnrt7D2C/XVihf/9QVfL4ilRfmbAbgonZN+WBk7+q1htQH+QftC3Ft/AZ2LebEME8TxPazt0B0ulYTTYmIx9HrCQ9yOL+Yvs//QqnNoFmIH7+MvRg/yylmidyxAOY9af+NGSAwAgY8Bj1uq9bQw6PFZXR56ifKbPa/Ig9d3o7QAAtDezQjwOqNzWZw5yd/MP9YH4eJ13VleF8PaNbPzbAnEBtmQtqyE+UmM7S8ADpfBx2vgYDG7otRRKSGKGnwMFMWbufFHzc79rs0C+LhxA5c3K6pvSB9HfzypP0dPdjXdDj/fuj3D7AEVOszUw8VcNEr8x37l7Rvyoej+mA+aW6Gw/nF9Hx2HgC9WoTy9T39q/V59Vb2Hnvrw4aZ9s6Ux5m8oNUAewtEh7+AX4i7IhQROStKGjzQh4t38uzsP53KpvylCVdkfoB5/ZcAlJm82d5iGD6XPspBWyMa+Xrz4o+b+Uu3aG7odeYREuW1fOwHx8/vj+zN5Z1OvUBVZk4hCS8kAfbXFP+5o69Ln9VgHN5pTyA2zoSM9SfKvSzQ+jJ7AtF+EFgbuS9GEREXaZ4GD3TH+S2Ja+LPv+dtZc/eNO71/pZL5s3DbLL3/t8dfRW37ryctM0RsHmD07kLUg5g8TZzTbzzzJL7so4ybckubk1oQWxjf0e5zXYij/Qym06bMABEBPnSt2UYK3YdZtGWA2zcl03naA9cGyIsDi4ca98ObrMnDxtm2kdkbPnRvnn72ueP6HwdtEusdkuPiEh9pJaGhqS4AJa9g23xJMzFuQD8VtaFQ/0e56mV3mQVlJz29Es7hPPOiJ4czi/m3QXbScnMZcXOw4B9cazLOoZj9fYiI7uQ8ybaWw5m33cBXZqdOQHYfiCPy15bCEBEkJWFD1/inhU63WH/JnvysHEmHNp2otzHH9pdaW+BaHP5aWfaFBFxF72e8DRlpZD8P5g/EfIyADAiu/G21228tr2Zo5rV28wXf+vHA9PXsOtQAT5eJiYM7szLczeTW2hvkbjj/Dgycwv5YV3lQzjXP3UFb/yylQ8W7ySuSQDzxw2ocpgH84q49NUF5BSW8t5tvUjsHFn9e26IDMP+2uJ4C0TW7hPHLI2gw1X2FojWl4K3xX1xioiUU+tJw+TJk3nllVfIyMggPj6et956i759K3+PPWDAABYuXFih/KqrruKHH+zvzUePHs0nn3zidDwxMZG5c+dWKR6PTRoMAzb/AElPH1vtEfukQ5dOgC7XY8PEpa8tYNch+6RNl3eK4P2RvStc5kBuEfdPX8OS7Ydc+vhJw7ozpEezM1cs5/kf/uT933YyOD6at4b3cOlcj2IY9o6TG2bCxlmQs+fEMd9g6DAYugyFuIu1qJaIuFWtJg1ffPEFI0eOZMqUKSQkJDBp0iRmzJhBSkoK4eEVx7AfPnyY4uJix/6hQ4eIj4/ngw8+YPTo0YA9acjMzOTjjz921LNarYSGnnktBPDQpCF1mX2iobTl9n2/MLj4Eeh9B3hbHdW27c/lye82YsLE1JG98LecupvKwzPWMmOV/csrKtiXrIISru/VjIhGvrw2b4tT3bGXt+Ofl7V1OezktCyGTLYvgvXrQxfX3JoZDZnNBnv+gA1fw5+zIC/zxDG/MOh0jb0FouUFla4sKiJSm2o1aUhISKBPnz68/fbbANhsNmJiYrjvvvt47LHHznj+pEmTmDBhAunp6QQE2DuJjR49mqysLGbNmuVKKA4elTTs32xvWUiZY9/39oN+Y+D8f9p/Qz0LhmHw44YMrN5mLu0QTqnNwMfrxGRMS7Yf5KtVe7igTROG9mhWramhDcNgwKsL2H2oAF8fM6ufuPy0icw5x1YGu5fYX2H8+S0UlGv9CQi3TyDV5TqIOU+rcYpInai1pKG4uBh/f3+++uorhgwZ4igfNWoUWVlZfPvtt2e8RteuXenXrx9Tp051lI0ePZpZs2ZhsVgIDQ3l0ksv5bnnnqNx48onzykqKqKoqMixn5OTQ0xMTMNOGnL2wfwXIPlTMGz2eQB63gYXPwZBUe6OziVLtx9i+Pv2SZEuad+UqSN7OyUnckxZKexaZH+Fsel7KMw6caxRNHQeAl2uh2a9XJ7JU0Skqmotadi3bx/NmjVjyZIl9OvXz1H+yCOPsHDhQpYvX37a81esWEFCQgLLly936gMxffp0/P39iYuLY/v27fzrX/8iMDCQpUuX4uVVsbn2qaee4umnn65Q3iCThqNZ8PskWPbuiRUYO/zFPu1zA162+bu1+/jn52uA6s0U+d3afWzLzKVDVBClNoM9RwrIPlrCr5v20yM2hOF9Y+keE9IwFsqqitJi+4yeG2fa+7EU5Zw4FhJrX0Sr83UQFa8EQkRqVL1NGv72t7+xdOlS1q1bd9p6O3bsoHXr1vzyyy9cdtllFY57TEvD3tXwv+vg6LGFoGLOg8ufgdgE98ZVQ576biPTluwiwOLFokcuoXGg9bT1S8tsTFm4nV2HCvhq1Z7T1gV7K8ZHo/t4TuJwXEkhbPvFnkCkzIWS/BPHgmPB79hrKse/3HL/hB3/nCsrK1d+NmVn/JzyZeVOD4uDniPta6B4n/7vgojUnVqb3KlJkyZ4eXmRmZnpVJ6ZmUlk5OmH1+Xn5zN9+nSeeeaZM35Oq1ataNKkCdu2bas0abBarVitHvCfTnhH+1j+gHAY+JR9NkEP+gL8v6s78seuw2zcl8PURTsY1DWK+ObBp/ySn59ygFd/3lKh3Gyyf/f4+XhRUFzmVL/1v+bw9T396RFbtU6zDYKPL3T8i30rLoCtP9lfYWz9GbJTIdvdAVZTdirsXGjv/Nn9Fug5qkG3pomci1xKGiwWC7169SIpKcnRp8Fms5GUlMS999572nNnzJhBUVERt9566xk/Z8+ePRw6dIioqIb1Lt9lPn4w6nsIaWFfttnDeHuZGd2/JQ9/tY73Fu3gvUU7eOTK9vxjQJsKdctsBpPnn5gYKTLIl3GJ7ekcHUTHqCCneu8t2s7Lc1MAsBkwbOoyujULJiLIl9duiievqJTX521h3Z4shvWO4dbzWjTc1giL/7FXE0OhKBf2rgJbabkKx+7L6f5cLHMqP5uy03yOUQbbkmDNfyFnLyx92761OB96jbYvAKbJr0TqvWoNuRw1ahTvvfceffv2ZdKkSXz55Zds3ryZiIgIRo4cSbNmzZg4caLTeRdeeCHNmjVj+vTpTuV5eXk8/fTTXH/99URGRrJ9+3YeeeQRcnNzWb9+fZVaFDxq9ISHKSgupc9zv5BfroVg+b8uIyLI+Qvi3QXbeWmufVGul67vyo29YiosjlWeYRi8NDeFKQu3O5WHBVg4nF/sVNYk0ML7I3t7VmtEQ1VWan/1smqavQXFsNnLfUNOtD6Ed3BnhCLnHFe+Q13u0j5s2DBeffVVJkyYQPfu3UlOTmbu3LlERNjXJ0hNTSU93Xm2wZSUFBYvXsydd95Z4XpeXl6sW7eOa665hnbt2nHnnXfSq1cvfvvtN894BXGO87d48+yQLk5lH/++q0K9mavtfRiG943lhjMkDAAmk4nHBnXghaFdncpPThgADuYVM/SdJcxP2e9i9FLjvLyh/ZVwy3R4YANc8jgEx9hHjix7B95JgA8TIflzKDnq7mhF5CSaRlrqxJz16azYeZhpS3bhb/Fi/rgBjtaG4lIbHSfMpcxmsGz8ZUQGV72ZuqTMxpQF9taGqYt2kFtUyuD4aJ69tjPBfj6M/GgFv209CECovw/vjOhFl2ZBNPLVLIz1hq3Mvqz7qmmQ8qP9VQbY5yXpdrP99UVEJ3dGKOLRtPaE1EsFxaVcOek3Ug8X8NeLWvGvqzoCsCUzlyv+vYgAixcbnk6sdv+DHQfyOJxfTO+WYY6y4lIbaUcKGP3xCtIO239z7dUilK/+3q/h9nPwZDnp9nVWVv3H3nHyuOZ97clD56H2fh4iUmNq9fWESHX5W7yZ8Bf7b4yfLtvN9gN5AMxeuw+ATtFBZ/VF3qppoFPCAGDxNtO6aSD/uePEMNZVu484dbqUeiQoCi56GO5fC7d+bR+eafaGPSvg23/Aax3gh3H2hcFEpM4paZA6dUmHcDpHB5FfXMZlry1kw95svkneC+DyBFCuiGsSwO+PXerYn/5HGh7QyOa5zGZoMxCG/Q8e3AiXTYDQllCUDX+8D1MugPcvg9X/heL8M15ORGqGXk9IndufW8iAVxY4zbkAsOJflxEeVLvD7vKKSun57DyKS230bhHKp3cnYPXWIlENgs1mn+dh1TTYPPvE0FNLI+h2o/31RVS8OyMUaZD0ekLqtfBGvky9zXkJ77GXt6v1hAEg0OrNw1e0B2Dl7iPMWrO31j9TaojZDK0vgZs+gbGbYODTEBoHxbmw8iN47yKYOsCeVBTlujtaEY+klgZxm69X7eGhGWsZHB/Nmzd3r9OOicfnhYgK9uXN4T3oc1JfCGkgbDbY9Zs9Udj0PdhK7OWWQOh6g731IbqHOyMUqfc0ekLkDLKPljDglfkcKSjBbIK5D1xEu4hG7g5Lzkb+QUj+zJ5AHC436VdUvD156HID+Or/B5GT6fWEyBkE+/nw8e32lVZtBnyj1xQNX0ATOP+fcN8qGDXbniR4WSB9Lcx+0D7y4rv77FNxN/zflUTcQi0Nck6bsz6df3y6miaBVn4ZexEh/hZ3hyQ1Kf8QrJtub304WG4xtIiu0GsUdLvJPomUyDlMLQ0iVXRph3Biw/w5mFfEtCW73B2O1LSAxtBvDIxZAbf/CN2GgZcVMtfDnHH21odZYyDtD7U+iFSBWhrknPflH2k88vU6LF5m5j5wIa2aBro7JKlNBYdh3Rf21ocDm0+Uh3e2933odhP4hbgpOJG6p46QIi4oLbNx64fLWbbjMHFNAogO8eWyDhHccUGcu0OT2mQYkLbcnjxs/AZKC+3l3n726ap7jYKYhJOW+RbxPEoaRFyUkpHLoDcWYSv3r+HhxPb8Y0BrrVFxLjh6BNZ9aU8g9v95orxpB+g5EqJ7QlgrCAxXEiEeR0mDSDU8+tU6vliZ5lT2jwGteeTKDm6KSOqcYcCelfbkYcPXUHrS8tw+ARAWZ5/SOqzVsZ/j7D8HNwezZheVhkdJg0g1ZOYUkvBCEgBXd43ih/XpeJlN/PzgRbRWP4dzT2G2vfVh82w4vAOy94BhO3V9sw+EtjiRRIQd+zM0zl7uba272EVcoKRBpJo2Z+SQX1RGrxah3PXJSn7ZlMmVnSOZclsvd4cm7lZaDFmp9gTiyE77n4d32n8+sgvKik9zssneEuFomSiXUITFgVUTi4n7KGkQqQFbM3NJnGTv5/D4VR25+6JW7g5J6itbGeTsq5hQHE8qivNOf35A03JJxEmtFP5h6kchtUpJg0gNeeOXrfz7F/ukQHecH8cjV7bH4mXGbNZ/4lJFhmGf4rpCQnFsv+DQ6c+3Bjn3nSifUDSKsi/kJXIWlDSI1KA7pv3Br5v3O/a7NAvi0zvPI9jfx41RiccozD7RInFyC0XOGaY39/a1d8p0SiiOJRghseClv6NyZkoaRGpQSkYuIz5YxsG8yt9Z331hHI9f3amOo5JzQslROLK78haKrFSwlZ76XJOXfZIqs/exzevYnz4n7Z98/FT7p6njVdVrVvVzfez9PKxB9j+9Nb17bVLSIFIL/rtsN0/M2lDpsf/dmcAFbZvUcURyTisrhey0cq89yrVQHN5ZcbhoQ+bt65xE+Aad+Nmp7Ph+ZWWNNCT2FJQ0iNQSwzD4+c9M5m7IYMn2g2TmFAHQNy6ML//Wr0Y+41BeET9tzOTKLpGEBeg3LKkGw4DcDCjMsrdG2ErtnTUdP5+irOwMx097jZIqnlPJflnJSfvF9s6jJQU1+1wsgceSCFeSjZPKLIEe1zFVSYNIHUnPPkr/F3/FMOCLv55HQqvG1b5WTmEJoz9awerULAAaB1j452VtuSUhFh8vdXaTc1BZKRTlQFFuuT9zoTDn2P7JZbnO5cfLyopqLiaTGSzlE4tKEhDL8fLAk/ZPKvPyrrm4zkKtJw2TJ0/mlVdeISMjg/j4eN566y369u1bad1p06Zx++23O5VZrVYKCwsd+4Zh8OSTT/L++++TlZXF+eefz7vvvkvbtm2rFI+SBnGnh2esZcaqPQRavfnXVR25uU+MS6MrZq7eQ9Lm/fywLv2Udf7v6o6MSGiBn0XNqyIuKy06kVAUVpaEZJ/4+eSE43gSUpgDRlnNxuXtVy6JCCz3yuXksvL7jZy342Vn8eqlVpOGL774gpEjRzJlyhQSEhKYNGkSM2bMICUlhfDw8Ar1p02bxv33309KSsqJDzWZiIiIcOy/9NJLTJw4kU8++YS4uDieeOIJ1q9fz59//omvr+8ZY1LSIO6UX1TKsKlL2bA3B7C3EPSNC2Pdnmz8LV68P7I3LZsEVHru+j3ZDH57sVPZ8L4xjLuiPbd9uII/03Ocjl3crikv39CN8EZWrYkhUpcMw94x1ZFcZJ+UXJRv5ciz7xfnlTuWe6KstPDMn+cqH/8zJxbWY0nISfs5xSaCW/eunaQhISGBPn368PbbbwNgs9mIiYnhvvvu47HHHqtQf9q0aTzwwANkZWVVej3DMIiOjuahhx5i3LhxAGRnZxMREcG0adO4+eabzxiTkgZxt/JTUJ8s1N+H0f3juPuiOPwtJ5ojp/2+k6e+P7E4UveYEB5ObM95rRrjZTZRXGoj9XA+QyYvIa/IuZf8refF8sw1XTRfhEhDVFbinERUlliccj/PnpgcLzvtTKRVk1NkEPxibpW+Q116oVJcXMyqVasYP368o8xsNjNw4ECWLl16yvPy8vJo0aIFNpuNnj178sILL9C5c2cAdu7cSUZGBgMHDnTUDw4OJiEhgaVLl1aaNBQVFVFUdOIdVU5OToU6InUpIsiXP59J5PaP/2D5zsNOx44UlPDvX7bw9eo9jO7fkuF9Y3loRjJz1mc46vztolaMv6qj03kWbzNtwhuR9NDFTFm4nY9/3+U49r9lqfyx8whzH7hQLQ4iDY2Xj32mT/+ws79WaVHFRKLS/dzKy4pzITsbyK3Sx7mUNBw8eJCysjKnVwsAERERbN68udJz2rdvz0cffUS3bt3Izs7m1VdfpX///mzcuJHmzZuTkZHhuMbJ1zx+7GQTJ07k6aefdiV0kVrnb/Hmkzv68uvm/fRr1ZjQAAu7D+Uzc/VePvhtB6mHC3hm9p88M/tPp/M6RDbipj4xp7xuRJAvTw7uzBNXd+LLlWk8NnM9ACmZuXy3dh/Xdm9Wq/clIvWYt9W+BVS/EzY5OfBUcJWq1nqX7H79+jFy5Ei6d+/OxRdfzMyZM2natCnvvfdeta85fvx4srOzHVtaWtqZTxKpA74+XlzVNYrQY0MlWzQO4MHL2zH3gYu4vmfzCvU3Pp3I3Aeqtoqm2Wzi5r6xzB83wFH24o+bKSk7zcqLIiI1yKWkoUmTJnh5eZGZmelUnpmZSWRkZJWu4ePjQ48ePdi2bRuA4zxXrmm1WgkKCnLaROqzmDB/Xrspnqm39SI62JdbEmLZ8twgAqyuD7mKaxLA5mevpEmghfTsQn7aWHmLnIhITXMpabBYLPTq1YukpBMdvmw2G0lJSfTrV7WJbcrKyli/fj1RUVEAxMXFERkZ6XTNnJwcli9fXuVrijQUV3SOZMn4y3hhaFcs3tVv6PP18eL6XvaWi4e+XMuq3UdqKkQRkVNy+X+tsWPH8v777/PJJ5+wadMm7rnnHvLz8x1zMYwcOdKpo+QzzzzDzz//zI4dO1i9ejW33noru3fv5q677gLswy8feOABnnvuOb777jvWr1/PyJEjiY6OZsiQITVzlyIe6MGB7TivVRhFpTbu/s/KCiMsRERqmstto8OGDePAgQNMmDCBjIwMunfvzty5cx0dGVNTUzGXW6r1yJEj3H333WRkZBAaGkqvXr1YsmQJnTqdWODnkUceIT8/n7/+9a9kZWVxwQUXMHfu3CrN0SByrvL18eLdEb24/N+LOJhXxPdr9zG8b6y7wxIRD6ZppEUauPcX7eD5OZuIDfPn07sSiAnzd3dIItKAuPIdqgntRRq4G3s3J9jPh9TDBVz48nxmr9vn7pBExEMpaRBp4EL8LUy6ubujY+W9n62h93O/kLQp8wxnioi4Rq8nRDxESZmNe/63il827Xcqv/W8WP51VUenKaxFRI7T6wmRc5CPl5m3b+nJTb2dJ5H637JUznshCZutwf9+ICJupqRBxIP4+njx8g3xbH1+EG/f0sNRnlNYytPfb2T9nmwKS2p4eV8ROWfo9YSIB8suKOEvb/9G2uGjjrK4JgF884/+hPhb3BiZiNQXej0hIgAE+/uw6OFLnOZv2Hkwnwtfms/k+ds4WqxWBxGpOiUNIh7OZDIx8bqubH72Sl6/KR5fHzO5RaW88lMKz8zeWGufm1NYwsQ5m5i7Ib3WPkNE6paSBpFzhK+PF9f1bM6Xf+tH00ZWAD5fkcaq3Ydr7DNW7T7Cwi0HSM8+yvQVqby3aAd//99q3kzaWmOfISLuoz4NIueoR75ay5cr9wDQqmkA/70zgehgX37amEFUsB/xMSEuXe/XzZncMW3lKY+/NbwHg+OjzyZkEakF6tMgImf02KCOmE32n3ccyOe2D5bz1ao9/P1/q7l28u/cP30NS7YfpKTMxtHiMsoqGbK540AehSVlFJfamLJwx2k/76nvNnI4v7g2bkVE6ohaGkTOYenZRxnz6WpWp2adsa6vj5l2EY0IC7AQGeRLSZnB16v3EBPmR2mZQXp2IQDf/KM/jXy9efHHzZhNJl65MZ4b3l3C1v15BPv58MkdfenuYiuGiNQeV75DlTSICCkZudzw7hJyjy2v3SM2hDKbwYa92bgyJ9TDie0Zc0mbCuVbMnMZOvl38ovLaNU0gLn3X+SY9lpE3EtJg4i4LCUjl/cWbifraAmv3xRPiL+FvVlH2bY/jzB/C49+vY7dh/KJjwlh/d5sAizeNPL1pl/rxixIOUBxqY2fx15EkK9PpdffcSCPS19b6NiPbx7Mo1d2oH+bJnV1iyJSCSUNIlKrymwGXsc7RBzbL7MZZ2w9+HF9Ovd8utqx7+NlYu4DF9G6aWCtxSoip6ekQUTqrfkp+xn/9Xoycgqdyls29mdoj+Z0jg4ioVUYjU7RYiEiNUtJg4jUe/tzCrn5/WXsOJBf4VjLxv78764Emof6uyEykXOLhlyKSL0XHuTLLw9ezL+HxdO6aYDTsV2HCrjo5fm8u2A7Ow7kuSlCETmZWhpEpN4wDPvQzb+8tdhpToerukbyxF864WUyER7k68YIRTyPXk+ISIO282A+P23M4PV5WygutTnKTSYYkRDLk4M74+OlhlKRmqCkQUQ8Ql5RKV+v2sPT32+sMF/E9T2b89yQLvhZvFy+bkZ2IfNT9tOvVWNaNgmocPxf36zns+WpgH3uiX8MaI3JZKpQT8QTKGkQEY+Snn0UL7OJd+ZvZ9qSXU7HujQL4r3betMsxO+01/h8RSrjZ64nyNebnMJSR/nwvrFMvK6rYz8ju5DzJiY5nXtjr+Y8P7SrJqQSj6SkQUQ81v6cQj5ZuouPf99FQXEZABFBVi7tEA6Y+MeA1sSEOY+6yCksodtTP5/ymgEWL8YltmfD3hy+Xr2n0jqNrN788M8LiW2sER3iWWp99MTkyZNp2bIlvr6+JCQksGLFilPWff/997nwwgsJDQ0lNDSUgQMHVqg/evRoTCaT03bllVdWJzQR8XDhQb48nNiBz+4+j9vOa4GfjxeZOUV8viKNz1ekcuHL8+k/MYmxXyTzbfJeNu7L5o+dJ5b/7tsyjP/c0Zdl4y9jzCWtAcgvLuPp7/90ShgevbIDu168mqm39cLbbCK3qJSh7/zOkm0H6/yeReoLl1savvjiC0aOHMmUKVNISEhg0qRJzJgxg5SUFMLDwyvUHzFiBOeffz79+/fH19eXl156iW+++YaNGzfSrFkzwJ40ZGZm8vHHHzvOs1qthIaGVikmtTSInLv2Zh1l2u87OZBbxE8bMzlaUnbKusP7xjDxum6OfcMw+G7tPr5evZf9OYXkFpZSUmajR2wIL18fT7C/fYKp+Zv389CMtY4RHdf1bMZzQ7rgb/Gu3ZsTqQO1+noiISGBPn368PbbbwNgs9mIiYnhvvvu47HHHjvj+WVlZYSGhvL2228zcuRIwJ40ZGVlMWvWLFdCcVDSICIAq3Yf5qPFu8AE3mYT3ybvczo++ZaeXN0tqlrXPlpcxuPfrGfmmr0AxIb588Go3rSLaHS2YYu4lSvfoS6lycXFxaxatYrx48c7ysxmMwMHDmTp0qVVukZBQQElJSWEhYU5lS9YsIDw8HBCQ0O59NJLee6552jcuHGl1ygqKqKoqMixn5OT48ptiIiH6tUijF4tTvzf8u+buh9rSdhDt+bBDOoSWe1r+1m8eO2meKJD/Hh7/jZSDxdw3TtLeGt4Dwa0b6rRFXJOcKlPw8GDBykrKyMiIsKpPCIigoyMjCpd49FHHyU6OpqBAwc6yq688kr+85//kJSUxEsvvcTChQsZNGgQZWWVNzNOnDiR4OBgxxYTE+PKbYjIOcJsNjGkRzP+e2cCDyd2wGw+uy92k8nEuMT2rPy/gXSIbEReUSm3T/uD+z5f4zSfhIinqtPxQy+++CLTp0/nm2++wdf3xKxuN998M9dccw1du3ZlyJAhzJ49mz/++IMFCxZUep3x48eTnZ3t2NLS0uroDkREoEmglel/PY9hvWPwNpuYvS6dOz/5g9IyJQ7i2VxKGpo0aYKXlxeZmZlO5ZmZmURGnr7Z79VXX+XFF1/k559/plu3bqet26pVK5o0acK2bdsqPW61WgkKCnLaRETqUoi/hZdu6MYHo3rj5+PFb1sP0ubxH/nXN+uZuyFDLQ/ikVxKGiwWC7169SIp6cTEJzabjaSkJPr163fK815++WWeffZZ5s6dS+/evc/4OXv27OHQoUNERVWvw5KISF0Z0D6cF67r4tj/bHkqf//fKkZ/vIK8otLTnCnS8Lj8emLs2LG8//77fPLJJ2zatIl77rmH/Px8br/9dgBGjhzp1FHypZde4oknnuCjjz6iZcuWZGRkkJGRQV6efeW6vLw8Hn74YZYtW8auXbtISkri2muvpU2bNiQmJtbQbYqI1J6hPZrzxV/P4/JOEViOrYmxZPshbnh3CWmHC9wcnUjNcXmQ8bBhwzhw4AATJkwgIyOD7t27M3fuXEfnyNTUVMzmE7nIu+++S3FxMTfccIPTdZ588kmeeuopvLy8WLduHZ988glZWVlER0dzxRVX8Oyzz2K1Ws/y9kRE6kZCq8YktLKP+Fq3J4vhU5exOSOXC1+eT6DVm/FXdWBEQgunc/bnFPL3/60iMtiXV2+M17wPUu9pGmkRkVqwbX8e932+hk3pzkPC/3pRK+67tA2NfH2Y+OMm3lu4A4COUUG8eXN32kY04t0F2zlSUMxVXaOweJnpFG3/fy0zp5Dfth5kYMdwgv18NMxTaoTWnhARqQdyCkt48tuNfHNsQqjjGvl6c3nHCMdEUcdZvc2M7t+S9xbtcCrvFBXEpowcyv9v3a15MK/f1J024YFkFRRTWGIjMtgXEVcpaRARqWe27c/lg9928uvm/ezPLXI69skdfflw8U4WbTng0jWP958oPjbU84pOEVzTPZqru0apFUKqTEmDiEg9lXus9WHmmr1EBftyx/lx3H1RK2w2g4+X7OK5H/7EMOCRK9tzY68YDuQW8c2aPWzOyOX6ns25oG0TflyfzrfJ+1i5+0iln3Fdj2Y8M6QLgVb1kZAzU9IgIlLPlZbZ8DKbKrQI7DlSQKi/hYAzfOHbbAZr0o6wOSOXrs2CMZtMvPJTCguPtVY08vXmrgtace+lbfA6y5kwxbMpaRAROUct2nKAp77fyI4D+QBc2LYJL9/QjahgP35Yl07akQLuuiAOb6+KI+6PFpexdk8WnaODaOTrU9ehi5soaRAROYfZbAYz1+zl8W/WU3RsZspOUUH8eWwkR9vwQP7vL51YvfsIuYWl3HdpG0IDLIz5bDU/rEsn1N+HxwZ1YEiPZli9vdx5K1IHlDSIiAjb9ufx0Iy1rE3LOm29iCArr94Yz20frnAqb900gCcHd+b8Nk30isODKWkQEREADMNgfsp+nvthEzsO5PPOiJ6s3HWET5buosxW8b//ZiF+jOrf4thcESWOsvsubcM13aM1AZUHUtIgIiJOSstsZOYW0SzED7DPRulv9abMZnDP/1axZPshwD7y4vVh3dlxII9nZ//Joq0HnZKLQKs391/WlqE9m/HHzsN0iwlxXFMaJiUNIiJSZWU2g0VbDrB+bzbX92rulATkFZXy2s8pfLVyD7mVLMBl8TYzrHcM13SPpl14Ixr5emPWq4wGRUmDiIjUqNIyG9+s2cvq1CwWbTnA3qyjldZrFuLHFZ0j6NY8mN4twogJ86/jSMVVShpERKTWFJWW8V3yPoL8fAi0evPVqj38sC7dMTPlcRZvM5e0b8oFbZvSKzaUdhGBlQ71FPdS0iAiInVqX9ZRFqQcYH9uIVsz89iSmcvW/XkV6rWLCOS2fi25tns0Qb4+7M8p5HBBMTsO5PPWr9toFuLLNd2b8ZeuUXrNUUeUNIiIiFsZhsFPGzP4dfN+9hw5ytq0LPKLy6p8fnxMCE9f05nuMSHV+vwym8HuQ/k0D/XHbEItHKehpEFEROqVMpvBzoP5/LIpk69W7WFbuVYIkwnCG1kpLTOIbezP1sw88o51uhzUJZJbEmI5v3WTKrU8HMorYm/WUV6em8LibQcBMJvgso4R/GNAa3rEhtbODTZgShpERKTeMgyDLZl5pB4uoFvzYJoGWp0Sgv25hbw8N4WvVu1xlFm8zdzUuzmj+7ckPMgXPx8v/th5mCaNrLRpGsjerKMUldq45u3FFJymRaNXi1CG9YmhddMAVu46QpvwQNbtySbYz4dLO4TTsklArd57faSkQUREGrzNGTl8vjyVz1akUlJ26q8qi5e5QidMgB6xIfz7pu4cKSjm0+WpzFqzl9JKJrQqr3N0EPExISTEhXFxu6aE+FvO+j7qOyUNIiLiMQpLyli+8zD/W7abXzZlcvxby+JlxsCokFC8dH1XujYLITLYl7CAE1/6mTmFzFy9l/8u3cW+7EKs3mYCrN4EWL1oFuLH8p2HKf+NaDZB7xZhXNIhnFB/H9bvzcbf4oWPl5nCEhsh/j6kZxeSV1SKzWbgZTZh9TbTISqItuGBBFi9KS614etjplmIH5+vSMNmGPRsEUqflqEkbdrPwbwi+rVuTEQjX6w+ZrfMuKmkQUREPFJmTiEAuYUlhPpb8DabOVJQTPNQPzZn5OJlNtEhslGFJcfLO3l2zOMO5xfzw/p0dh7IZ8n2g2zOyK3Ve6lMkK83If4WLmjbhC7RwVi8zVi8zRSWlHEkvxhvLzONfL0J9bdgGAbhQb4E+Xrj6+NFVLCv474LS8pYkLKfrZl5NAv1o8xmkF9USudmwXRtFoyvz4mFyJQ0iIiInKU9RwqYv3k/SZv3sz+niNAAH1o2DmDPkaME+/mw+3ABHSIaEejrTVFpGUG+PmQdLSEzu5At+3M5WmwjwOrFgdwiCorL6BDZiNZNA0lOy3KaHMvHy3Ta1y9VFeLvQ9dmwUQG+TKjXH+Qk1m8zfRuEUrvlmE0CbTQxGrj6l6tlTSIiIi4W5nNILewhGA/H0wmE4ZhsGr3EQ7kFnF5pwjyi8ootdkosxnszTpKZk4hi7cdJDOniOJSG8WlNkrKbJhMUFRqI8DizcG8IorLbBSWlHEorxjj2OecrHN0EF5mE1kFJYQGWNh75CgH84qc6tiKCkibdFOVvkO1XJmIiEgt8jKbnDpUmkwmercMc+wH+5+YQyI8yBeAK7tEufQZxaU21u7JYuPebPZmHWVfViFdmwfzt4taOb2qMQyD7QfyWLr9EBv35XCkoJj1OzNIq+LnqKVBRETkHObKd6imyBIREZEqqVbSMHnyZFq2bImvry8JCQmsWLHitPVnzJhBhw4d8PX1pWvXrsyZM8fpuGEYTJgwgaioKPz8/Bg4cCBbt26tTmgiIiJSS1xOGr744gvGjh3Lk08+yerVq4mPjycxMZH9+/dXWn/JkiUMHz6cO++8kzVr1jBkyBCGDBnChg0bHHVefvll3nzzTaZMmcLy5csJCAggMTGRwsLC6t+ZiIiI1CiX+zQkJCTQp08f3n77bQBsNhsxMTHcd999PPbYYxXqDxs2jPz8fGbPnu0oO++88+jevTtTpkzBMAyio6N56KGHGDduHADZ2dlEREQwbdo0br755grXLCoqoqjoRO/P7OxsYmNjSUtLU58GERERF+Tk5BATE0NWVhbBwcGnr2y4oKioyPDy8jK++eYbp/KRI0ca11xzTaXnxMTEGP/+97+dyiZMmGB069bNMAzD2L59uwEYa9ascapz0UUXGf/85z8rveaTTz5pANq0adOmTZu2GtrS0tLOmAe4NOTy4MGDlJWVERER4VQeERHB5s2bKz0nIyOj0voZGRmO48fLTlXnZOPHj2fs2LGO/aysLFq0aEFqauqZsyRxcjzDVCuN6/Tsqk/Prvr07KpPz65yhmGQm5tLdHT0Ges2yHkarFYrVqu1QnlwcLD+IlRTUFCQnl016dlVn55d9enZVZ+eXUVV/YXbpY6QTZo0wcvLi8zMTKfyzMxMIiMjKz0nMjLytPWP/+nKNUVERKTuuZQ0WCwWevXqRVJSkqPMZrORlJREv379Kj2nX79+TvUB5s2b56gfFxdHZGSkU52cnByWL19+ymuKiIhI3XP59cTYsWMZNWoUvXv3pm/fvkyaNIn8/Hxuv/12AEaOHEmzZs2YOHEiAPfffz8XX3wxr732GldffTXTp09n5cqVTJ06FbBPp/nAAw/w3HPP0bZtW+Li4njiiSeIjo5myJAhVYrJarXy5JNPVvrKQk5Pz6769OyqT8+u+vTsqk/P7uxVaxrpt99+m1deeYWMjAy6d+/Om2++SUJCAgADBgygZcuWTJs2zVF/xowZ/N///R+7du2ibdu2vPzyy1x11VWO44Zh8OSTTzJ16lSysrK44IILeOedd2jXrt3Z36GIiIjUCI9Ye0JERERqn9aeEBERkSpR0iAiIiJVoqRBREREqkRJg4iIiFSJRyQNri7V7ekmTpxInz59aNSoEeHh4QwZMoSUlBSnOoWFhYwZM4bGjRsTGBjI9ddfX2GCrdTUVK6++mr8/f0JDw/n4YcfprS0tC5vxe1efPFFx7Dg4/TsTm3v3r3ceuutNG7cGD8/P7p27crKlSsdxw3DYMKECURFReHn58fAgQPZunWr0zUOHz7MiBEjCAoKIiQkhDvvvJO8vLy6vpU6VVZWxhNPPEFcXBx+fn60bt2aZ599lvL91PXs7BYtWsTgwYOJjo7GZDIxa9Ysp+M19ZzWrVvHhRdeiK+vLzExMbz88su1fWsNwxlXp6jnpk+fblgsFuOjjz4yNm7caNx9991GSEiIkZmZ6e7Q3CYxMdH4+OOPjQ0bNhjJycnGVVddZcTGxhp5eXmOOn//+9+NmJgYIykpyVi5cqVx3nnnGf3793ccLy0tNbp06WIMHDjQWLNmjTFnzhyjSZMmxvjx491xS26xYsUKo2XLlka3bt2M+++/31GuZ1e5w4cPGy1atDBGjx5tLF++3NixY4fx008/Gdu2bXPUefHFF43g4GBj1qxZxtq1a41rrrnGiIuLM44ePeqoc+WVVxrx8fHGsmXLjN9++81o06aNMXz4cHfcUp15/vnnjcaNGxuzZ882du7cacyYMcMIDAw03njjDUcdPTu7OXPmGI8//rgxc+ZMA6iwgGJNPKfs7GwjIiLCGDFihLFhwwbj888/N/z8/Iz33nuvrm6z3mrwSUPfvn2NMWPGOPbLysqM6OhoY+LEiW6Mqn7Zv3+/ARgLFy40DMMwsrKyDB8fH2PGjBmOOps2bTIAY+nSpYZh2P9hms1mIyMjw1Hn3XffNYKCgoyioqK6vQE3yM3NNdq2bWvMmzfPuPjiix1Jg57dqT366KPGBRdccMrjNpvNiIyMNF555RVHWVZWlmG1Wo3PP//cMAzD+PPPPw3A+OOPPxx1fvzxR8NkMhl79+6tveDd7OqrrzbuuOMOp7LrrrvOGDFihGEYenancnLSUFPP6Z133jFCQ0Od/r0++uijRvv27Wv5juq/Bv16ori4mFWrVjFw4EBHmdlsZuDAgSxdutSNkdUv2dnZAISFhQGwatUqSkpKnJ5bhw4diI2NdTy3pUuX0rVrV6fVRxMTE8nJyWHjxo11GL17jBkzhquvvtrpGYGe3el899139O7dmxtvvJHw8HB69OjB+++/7zi+c+dOMjIynJ5dcHAwCQkJTs8uJCSE3r17O+oMHDgQs9nM8uXL6+5m6lj//v1JSkpiy5YtAKxdu5bFixczaNAgQM+uqmrqOS1dupSLLroIi8XiqJOYmEhKSgpHjhypo7upnxrkKpfHVWep7nONzWbjgQce4Pzzz6dLly6AfTlyi8VCSEiIU92Tlyyv7LkeP+bJpk+fzurVq/njjz8qHNOzO7UdO3bw7rvvMnbsWP71r3/xxx9/8M9//hOLxcKoUaMc917Zsyn/7MLDw52Oe3t7ExYW5tHP7rHHHiMnJ4cOHTrg5eVFWVkZzz//PCNGjADQs6uimnpOGRkZxMXFVbjG8WOhoaG1En9D0KCTBjmzMWPGsGHDBhYvXuzuUBqEtLQ07r//fubNm4evr6+7w2lQbDYbvXv35oUXXgCgR48ebNiwgSlTpjBq1Cg3R1e/ffnll3z66ad89tlndO7cmeTkZB544AGio6P17KReadCvJ6qzVPe55N5772X27NnMnz+f5s2bO8ojIyMpLi4mKyvLqf7JS5ZX9lyPH/NUq1atYv/+/fTs2RNvb2+8vb1ZuHAhb775Jt7e3kREROjZnUJUVBSdOnVyKuvYsSOpqanAiXs/3b/XyMhI9u/f73S8tLSUw4cPe/Sze/jhh3nssce4+eab6dq1K7fddhsPPvigY+E/PbuqqanndK7+G66KBp00VGep7nOBYRjce++9fPPNN/z6668Vmtl69eqFj4+P03NLSUkhNTXV8dz69evH+vXrnf5xzZs3j6CgoApfDJ7ksssuY/369SQnJzu23r17M2LECMfPenaVO//88ysM7d2yZQstWrQAIC4ujsjISKdnl5OTw/Lly52eXVZWFqtWrXLU+fXXX7HZbI5F8TxRQUEBZrPzf8deXl7YbDZAz66qauo59evXj0WLFlFSUuKoM2/ePNq3b39Ov5oAPGPIpdVqNaZNm2b8+eefxl//+lcjJCTEqef6ueaee+4xgoODjQULFhjp6emOraCgwFHn73//uxEbG2v8+uuvxsqVK41+/foZ/fr1cxw/PmzwiiuuMJKTk425c+caTZs29fhhg5UpP3rCMPTsTmXFihWGt7e38fzzzxtbt241Pv30U8Pf39/43//+56jz4osvGiEhIca3335rrFu3zrj22msrHQ7Xo0cPY/ny5cbixYuNtm3betywwZONGjXKaNasmWPI5cyZM40mTZoYjzzyiKOOnp1dbm6usWbNGmPNmjUGYLz++uvGmjVrjN27dxuGUTPPKSsry4iIiDBuu+02Y8OGDcb06dMNf39/Dbk0PGDIpWEYxltvvWXExsYaFovF6Nu3r7Fs2TJ3h+RWQKXbxx9/7Khz9OhR4x//+IcRGhpq+Pv7G0OHDjXS09OdrrNr1y5j0KBBhp+fn9GkSRPjoYceMkpKSur4btzv5KRBz+7Uvv/+e6NLly6G1Wo1OnToYEydOtXpuM1mM5544gkjIiLCsFqtxmWXXWakpKQ41Tl06JAxfPhwIzAw0AgKCjJuv/12Izc3ty5vo87l5OQY999/vxEbG2v4+voarVq1Mh5//HGnIX96dnbz58+v9P+3UaNGGYZRc89p7dq1xgUXXGBYrVajWbNmxosvvlhXt1ivaWlsERERqZIG3adBRERE6o6SBhEREakSJQ0iIiJSJUoaREREpEqUNIiIiEiVKGkQERGRKlHSICIiIlWipEFERESqREmDiIiIVImSBhEREakSJQ0iIiJSJf8PCovuhUdz1/MAAAAASUVORK5CYII=\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Better model found at epoch 3 with valid_loss value: 0.7307079434394836.\n",
            "Better model found at epoch 5 with valid_loss value: 0.5960114002227783.\n",
            "Better model found at epoch 6 with valid_loss value: 0.5489932298660278.\n",
            "Better model found at epoch 7 with valid_loss value: 0.5436006188392639.\n",
            "Better model found at epoch 8 with valid_loss value: 0.5244889855384827.\n",
            "Better model found at epoch 9 with valid_loss value: 0.5064975023269653.\n"
          ]
        }
      ],
      "source": [
        "learnClass.fine_tune(10, 6e-2)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "j6lus3EmC2k0"
      },
      "source": [
        "Ahora podemos usar nuestro modelo para predecir la clase de una nueva frase. "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 32,
      "metadata": {
        "id": "SsaEoSsLC2k1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 573
        },
        "outputId": "604541f0-3205-49b6-fcb1-538c70567bd5"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.9/dist-packages/torch/amp/autocast_mode.py:204: UserWarning: User provided device_type of 'cuda', but CUDA is not available. Disabling\n",
            "  warnings.warn('User provided device_type of \\'cuda\\', but CUDA is not available. Disabling')\n",
            "/usr/local/lib/python3.9/dist-packages/torch/cuda/amp/grad_scaler.py:120: UserWarning: torch.cuda.amp.GradScaler is enabled, but CUDA is not available.  Disabling.\n",
            "  warnings.warn(\"torch.cuda.amp.GradScaler is enabled, but CUDA is not available.  Disabling.\")\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "<style>\n",
              "    /* Turns off some styling */\n",
              "    progress {\n",
              "        /* gets rid of default border in Firefox and Opera. */\n",
              "        border: none;\n",
              "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
              "        background-size: auto;\n",
              "    }\n",
              "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
              "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
              "    }\n",
              "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
              "        background: #F44336;\n",
              "    }\n",
              "</style>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": []
          },
          "metadata": {}
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "('42',\n",
              " tensor(42),\n",
              " tensor([6.1113e-05, 3.4140e-06, 5.3438e-08, 6.2474e-07, 1.7562e-07, 1.8319e-04,\n",
              "         7.3557e-09, 7.5409e-09, 1.4656e-06, 1.7566e-04, 1.3840e-05, 4.2550e-09,\n",
              "         1.3802e-09, 8.7871e-10, 9.8407e-08, 2.7898e-07, 1.5222e-04, 9.9282e-07,\n",
              "         3.3463e-07, 2.7594e-05, 1.4694e-08, 1.3287e-07, 3.0450e-07, 5.2692e-07,\n",
              "         7.8050e-07, 6.2470e-08, 3.3111e-08, 1.5844e-07, 1.2301e-06, 2.4226e-06,\n",
              "         4.8855e-07, 6.1708e-07, 3.2115e-09, 9.2302e-09, 1.1573e-07, 1.0924e-08,\n",
              "         2.7457e-07, 2.5876e-07, 1.5623e-07, 6.7291e-07, 1.2329e-06, 2.6838e-09,\n",
              "         9.9911e-01, 3.3957e-07, 8.0291e-10, 3.5584e-08, 6.4111e-07, 1.7399e-07,\n",
              "         1.3064e-07, 5.9979e-08, 9.4901e-08, 5.9597e-08, 5.8993e-08, 1.6217e-07,\n",
              "         5.0505e-07, 1.2821e-07, 7.6527e-09, 5.9862e-09, 1.3971e-06, 9.2649e-08,\n",
              "         2.1495e-09, 8.4540e-07, 1.0286e-05, 1.6611e-07, 3.6453e-08, 1.8690e-07,\n",
              "         5.4414e-07, 3.1302e-08, 1.5286e-08, 3.6581e-10, 4.6303e-08, 8.9668e-07,\n",
              "         6.2288e-06, 2.1385e-07, 2.5370e-09, 2.3735e-09, 8.2600e-10, 6.3578e-06,\n",
              "         3.8083e-06, 1.5287e-09, 4.0596e-08, 5.1849e-09, 7.8767e-07, 4.0321e-08,\n",
              "         9.0285e-08, 2.1779e-07, 3.1756e-06, 1.9532e-08, 1.4984e-07, 8.1456e-10,\n",
              "         3.6222e-07, 3.3713e-08, 7.8861e-08, 6.9834e-09, 6.5178e-08, 1.5851e-08,\n",
              "         2.0026e-06, 3.3722e-09, 1.6185e-06, 4.3474e-07, 4.8947e-10, 2.3182e-10,\n",
              "         1.6654e-05, 7.4820e-09, 4.9210e-08, 4.9979e-08, 2.3749e-05, 1.2189e-05,\n",
              "         4.4972e-08, 4.6015e-08, 6.5330e-08, 1.4203e-08, 2.9564e-08, 2.1334e-07,\n",
              "         4.5185e-07, 1.5453e-08, 4.6652e-06, 2.5597e-07, 3.2864e-09, 4.6629e-07,\n",
              "         3.6141e-06, 1.7922e-08, 1.0121e-07, 4.1776e-08, 3.1856e-06, 3.4923e-07,\n",
              "         1.8673e-09, 4.2030e-09, 6.8361e-05, 9.9169e-07, 2.8033e-08, 2.7790e-08,\n",
              "         1.8843e-09, 4.5917e-08, 3.7620e-05, 2.1628e-05, 4.6948e-08, 1.2228e-06,\n",
              "         6.4407e-06, 3.4788e-06, 2.7080e-09, 2.4151e-07, 2.4032e-06, 2.2485e-09,\n",
              "         2.5496e-07, 6.6574e-06, 1.7707e-09, 2.9897e-09, 2.0567e-07, 3.3097e-06,\n",
              "         2.0779e-07]))"
            ]
          },
          "metadata": {},
          "execution_count": 32
        }
      ],
      "source": [
        "learnClass.predict('with pale blue berries. in these peaceful shades--.')"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Por último, podemos validar nuestro modelo en el conjunto de test, para lo cuál hay que combinar los dataframes y construir un nuevo dataloader. "
      ],
      "metadata": {
        "id": "O5IXQsJbEp0o"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "test_df['set']=True\n",
        "train_test_df = pd.concat([train_df,test_df])\n",
        "dls_test = sentiment_clas.dataloaders(train_test_df, bs=64)"
      ],
      "metadata": {
        "id": "8IhEtk84Enuo",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 17
        },
        "outputId": "ec8343c6-3198-4e7b-8c38-0482cbc9e69b"
      },
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "<style>\n",
              "    /* Turns off some styling */\n",
              "    progress {\n",
              "        /* gets rid of default border in Firefox and Opera. */\n",
              "        border: none;\n",
              "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
              "        background-size: auto;\n",
              "    }\n",
              "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
              "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
              "    }\n",
              "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
              "        background: #F44336;\n",
              "    }\n",
              "</style>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": []
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Modificamos ahora el dataloader de nuestro learner, y procedemos a validar."
      ],
      "metadata": {
        "id": "JNU-NRgVE4jY"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "learnClass.dls = dls_test\n",
        "learnClass.validate()"
      ],
      "metadata": {
        "id": "mJkfXRBKE3OP",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 121
        },
        "outputId": "91a9c0d7-d8b9-4f57-9fb1-75ab50c066df"
      },
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.9/dist-packages/torch/amp/autocast_mode.py:204: UserWarning: User provided device_type of 'cuda', but CUDA is not available. Disabling\n",
            "  warnings.warn('User provided device_type of \\'cuda\\', but CUDA is not available. Disabling')\n",
            "/usr/local/lib/python3.9/dist-packages/torch/cuda/amp/grad_scaler.py:120: UserWarning: torch.cuda.amp.GradScaler is enabled, but CUDA is not available.  Disabling.\n",
            "  warnings.warn(\"torch.cuda.amp.GradScaler is enabled, but CUDA is not available.  Disabling.\")\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "<style>\n",
              "    /* Turns off some styling */\n",
              "    progress {\n",
              "        /* gets rid of default border in Firefox and Opera. */\n",
              "        border: none;\n",
              "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
              "        background-size: auto;\n",
              "    }\n",
              "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
              "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
              "    }\n",
              "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
              "        background: #F44336;\n",
              "    }\n",
              "</style>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": []
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Better model found at epoch 0 with valid_loss value: 0.8236363530158997.\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(#2) [1.1376911401748657,0.8236363530158997]"
            ]
          },
          "metadata": {},
          "execution_count": 35
        }
      ]
    }
  ],
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.6.7"
    },
    "colab": {
      "name": "Practica8.ipynb",
      "provenance": []
    },
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "acea7c14a43a4d98ac3fa7d55bcf6ee6": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_2da90ab6b0f54d49bd1c97944d7aa83e",
              "IPY_MODEL_bc81e5b43dac454a9b043edcbe8bdd22",
              "IPY_MODEL_a8520126760f425cb07c030d6ea2c273"
            ],
            "layout": "IPY_MODEL_dc6a959e6b694c5cbc8a539272f29ad7"
          }
        },
        "2da90ab6b0f54d49bd1c97944d7aa83e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_b52ea6a60f84468d982c195927cbb7ef",
            "placeholder": "​",
            "style": "IPY_MODEL_11e54992e4fd4d40bae190ab927514b7",
            "value": "Downloading builder script: 100%"
          }
        },
        "bc81e5b43dac454a9b043edcbe8bdd22": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_50b6f98995b940c0881aef1870ade8ab",
            "max": 8567,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_ffdd49432e9744f7bd5c23bff0244b54",
            "value": 8567
          }
        },
        "a8520126760f425cb07c030d6ea2c273": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_ba2477e9ee9f48c8be6f2f74f8bc2cd3",
            "placeholder": "​",
            "style": "IPY_MODEL_0b28d318f69844d88afd25bd867d827e",
            "value": " 8.57k/8.57k [00:00&lt;00:00, 224kB/s]"
          }
        },
        "dc6a959e6b694c5cbc8a539272f29ad7": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "b52ea6a60f84468d982c195927cbb7ef": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "11e54992e4fd4d40bae190ab927514b7": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "50b6f98995b940c0881aef1870ade8ab": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "ffdd49432e9744f7bd5c23bff0244b54": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "ba2477e9ee9f48c8be6f2f74f8bc2cd3": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "0b28d318f69844d88afd25bd867d827e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "a03bc89197ea4573b0f6784ca5b4e80b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_1b0a75b3b78b468aa8e75f3074d58dfb",
              "IPY_MODEL_603fffb705344e41bbf2cb31fc1e5bfc",
              "IPY_MODEL_e74351f03d53422789bbdbbfeb764efa"
            ],
            "layout": "IPY_MODEL_c65151b08bec4cf48d0e0d6ee1250f59"
          }
        },
        "1b0a75b3b78b468aa8e75f3074d58dfb": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_78fabf12226a4f9cab33f1885b963534",
            "placeholder": "​",
            "style": "IPY_MODEL_e34d7ab3f4834d17ae863d288445d9be",
            "value": "Downloading metadata: 100%"
          }
        },
        "603fffb705344e41bbf2cb31fc1e5bfc": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_ff14ace405e4409589bc3531a789dda1",
            "max": 14391,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_08c58056296849fdbab23840a9196979",
            "value": 14391
          }
        },
        "e74351f03d53422789bbdbbfeb764efa": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_84ff998cb0f8461daa8deaf55dc17db5",
            "placeholder": "​",
            "style": "IPY_MODEL_63cd9ad6473b43f0856596c29c6a0cf3",
            "value": " 14.4k/14.4k [00:00&lt;00:00, 765kB/s]"
          }
        },
        "c65151b08bec4cf48d0e0d6ee1250f59": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "78fabf12226a4f9cab33f1885b963534": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "e34d7ab3f4834d17ae863d288445d9be": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "ff14ace405e4409589bc3531a789dda1": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "08c58056296849fdbab23840a9196979": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "84ff998cb0f8461daa8deaf55dc17db5": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "63cd9ad6473b43f0856596c29c6a0cf3": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "a4f85e3bf0eb45c4857f6173b1adebcd": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_8edf734aa6d64f5f89906b464d512650",
              "IPY_MODEL_94dba14a5bc342bb8711f6843b6d7ed1",
              "IPY_MODEL_fc6d1aa44bd94a3ab9ca40c6a66c8ffb"
            ],
            "layout": "IPY_MODEL_a15bf77b783d4833b9c71ad2b2d9a9a4"
          }
        },
        "8edf734aa6d64f5f89906b464d512650": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_76f7f82fe6134040a90c7413538aafe5",
            "placeholder": "​",
            "style": "IPY_MODEL_30d05e576f3747e08d932ea00fed01be",
            "value": "Downloading readme: 100%"
          }
        },
        "94dba14a5bc342bb8711f6843b6d7ed1": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_ffda192e0bc24c40aa272eea8011e1a1",
            "max": 23440,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_89692d1d3c914672a458832e51be6d2d",
            "value": 23440
          }
        },
        "fc6d1aa44bd94a3ab9ca40c6a66c8ffb": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_1ebc784efa24427b8af27c9fdda2686b",
            "placeholder": "​",
            "style": "IPY_MODEL_f7d8ef7ca553491c9fba3ab382af530c",
            "value": " 23.4k/23.4k [00:00&lt;00:00, 985kB/s]"
          }
        },
        "a15bf77b783d4833b9c71ad2b2d9a9a4": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "76f7f82fe6134040a90c7413538aafe5": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "30d05e576f3747e08d932ea00fed01be": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "ffda192e0bc24c40aa272eea8011e1a1": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "89692d1d3c914672a458832e51be6d2d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "1ebc784efa24427b8af27c9fdda2686b": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "f7d8ef7ca553491c9fba3ab382af530c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "fd342b9fb4a84009b32ab4c7bcb377a4": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_daf05dc94f98455684bc96752c216e9c",
              "IPY_MODEL_575724ca3bb24bc79d64ed65544bc043",
              "IPY_MODEL_2f8aa3e45a55423d83606705f5010348"
            ],
            "layout": "IPY_MODEL_7e55727bedd94e849a125e4406f6a980"
          }
        },
        "daf05dc94f98455684bc96752c216e9c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_f595b20d8db44b9998732a1cc713a1e7",
            "placeholder": "​",
            "style": "IPY_MODEL_00d5b27795f448d485da956dbd8a73d4",
            "value": "Downloading data: "
          }
        },
        "575724ca3bb24bc79d64ed65544bc043": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_3d5308aafb4744508c3390d59f8c6992",
            "max": 216610,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_cbcbf6c645a14f61b625154537e9e528",
            "value": 216610
          }
        },
        "2f8aa3e45a55423d83606705f5010348": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_0bc1a29b59384aec931b7d0c611c5ec2",
            "placeholder": "​",
            "style": "IPY_MODEL_8ff6ad225eba4c1b9baf01aa9db1a7d3",
            "value": " 1.70M/? [00:00&lt;00:00, 26.9MB/s]"
          }
        },
        "7e55727bedd94e849a125e4406f6a980": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "f595b20d8db44b9998732a1cc713a1e7": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "00d5b27795f448d485da956dbd8a73d4": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "3d5308aafb4744508c3390d59f8c6992": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "cbcbf6c645a14f61b625154537e9e528": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "0bc1a29b59384aec931b7d0c611c5ec2": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "8ff6ad225eba4c1b9baf01aa9db1a7d3": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "18935a837f49426dac45432e83a5913c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_c6a107ffcc754e82921ad75bc6643898",
              "IPY_MODEL_2e382b16620c4bb09534050dae7a9d6e",
              "IPY_MODEL_997bd650baad41819aa3a713fa4341f3"
            ],
            "layout": "IPY_MODEL_29e85d0ef3754c7f8366469cd530cacb"
          }
        },
        "c6a107ffcc754e82921ad75bc6643898": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_5bee187efff34f8eab2c434eb21211ea",
            "placeholder": "​",
            "style": "IPY_MODEL_275b01de8e6444a69420369d2a96b76d",
            "value": "Generating train split: 100%"
          }
        },
        "2e382b16620c4bb09534050dae7a9d6e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_71ab8921857b496791db2a2286fab6ba",
            "max": 7600,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_225e43ff721745c5b01dd7fe5dfe7e4d",
            "value": 7600
          }
        },
        "997bd650baad41819aa3a713fa4341f3": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_a6900524efc04d8d8019c415196d4a46",
            "placeholder": "​",
            "style": "IPY_MODEL_a682df09134f4c74aef12a2e294e2cae",
            "value": " 7600/7600 [00:18&lt;00:00, 24647.06 examples/s]"
          }
        },
        "29e85d0ef3754c7f8366469cd530cacb": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": "hidden",
            "width": null
          }
        },
        "5bee187efff34f8eab2c434eb21211ea": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "275b01de8e6444a69420369d2a96b76d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "71ab8921857b496791db2a2286fab6ba": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "225e43ff721745c5b01dd7fe5dfe7e4d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "a6900524efc04d8d8019c415196d4a46": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "a682df09134f4c74aef12a2e294e2cae": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "b329b0dc353247f58bf8c38b24c4e051": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_2ac54f2315934367ae8bd34aba136f21",
              "IPY_MODEL_579357f9baec40f2915d8f3021fc20eb",
              "IPY_MODEL_4e1c1bd8b4d74978bff0cfa348a1bff0"
            ],
            "layout": "IPY_MODEL_68bae7a3d76f4e68afe3e6fac15ccfb4"
          }
        },
        "2ac54f2315934367ae8bd34aba136f21": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_720d5c006f2f46d982be76cb50eb8f17",
            "placeholder": "​",
            "style": "IPY_MODEL_d2a91625d68f43479aa16f24b386af74",
            "value": "Generating validation split: 100%"
          }
        },
        "579357f9baec40f2915d8f3021fc20eb": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_19601448d0704949aaa8459fae71b878",
            "max": 3100,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_cdbef121730444b8abd227f4dbd42d3c",
            "value": 3100
          }
        },
        "4e1c1bd8b4d74978bff0cfa348a1bff0": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_c24578c84bae4ab5ac0145c742f0747d",
            "placeholder": "​",
            "style": "IPY_MODEL_843ad42380cb40b989659ac691e67d65",
            "value": " 3100/3100 [00:18&lt;00:00, 21874.39 examples/s]"
          }
        },
        "68bae7a3d76f4e68afe3e6fac15ccfb4": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": "hidden",
            "width": null
          }
        },
        "720d5c006f2f46d982be76cb50eb8f17": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "d2a91625d68f43479aa16f24b386af74": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "19601448d0704949aaa8459fae71b878": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "cdbef121730444b8abd227f4dbd42d3c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "c24578c84bae4ab5ac0145c742f0747d": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "843ad42380cb40b989659ac691e67d65": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "f2d9294ee6f44814a8e8d9a87cf08c64": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_532efa9cd0d8459eac8b03bba83aad51",
              "IPY_MODEL_f6d159342ce047df904907d04cceb9ec",
              "IPY_MODEL_035514d9dad34ea6b645018dad177693"
            ],
            "layout": "IPY_MODEL_09394797973941e68705559ddcd9b802"
          }
        },
        "532efa9cd0d8459eac8b03bba83aad51": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_571f0233507544148b23a0dcf251daa3",
            "placeholder": "​",
            "style": "IPY_MODEL_8fcd79171edb43b28b3291c06440b5cb",
            "value": "Generating test split: 100%"
          }
        },
        "f6d159342ce047df904907d04cceb9ec": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_cf801abb3ce94c9b8bf91b4d1efa39db",
            "max": 5500,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_116121b6f33b4883884fa748e515d969",
            "value": 5500
          }
        },
        "035514d9dad34ea6b645018dad177693": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_a2320a4c9d0e492c93cc9b06e758e0d5",
            "placeholder": "​",
            "style": "IPY_MODEL_10e14d896714494da3d8d37dd8ed9282",
            "value": " 5500/5500 [00:18&lt;00:00, 20476.87 examples/s]"
          }
        },
        "09394797973941e68705559ddcd9b802": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": "hidden",
            "width": null
          }
        },
        "571f0233507544148b23a0dcf251daa3": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "8fcd79171edb43b28b3291c06440b5cb": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "cf801abb3ce94c9b8bf91b4d1efa39db": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "116121b6f33b4883884fa748e515d969": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "a2320a4c9d0e492c93cc9b06e758e0d5": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "10e14d896714494da3d8d37dd8ed9282": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "c8aca95b4c1d4734a86f912b7ea43c6e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_55a5af05b8234f938ff6104af90f2460",
              "IPY_MODEL_f66bcff001e04b199a1bbc9d92f42367",
              "IPY_MODEL_ff9eb64ef78b4722b373285f8bce32ed"
            ],
            "layout": "IPY_MODEL_70aad33e2cd043c08a94c8b8a93a3342"
          }
        },
        "55a5af05b8234f938ff6104af90f2460": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_892450aebc2b4611aec22ffb88a8cba9",
            "placeholder": "​",
            "style": "IPY_MODEL_1e6314ba4f4a445b9b68928c6debb77c",
            "value": "100%"
          }
        },
        "f66bcff001e04b199a1bbc9d92f42367": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_ef7d3624b5fe44eea8d40f6b6a02898b",
            "max": 3,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_e652c021449c485484fb899b17acd0ef",
            "value": 3
          }
        },
        "ff9eb64ef78b4722b373285f8bce32ed": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_0661f3b9e7bc408f82030c5b7c164fcb",
            "placeholder": "​",
            "style": "IPY_MODEL_ab50e73bf5404380baba92cf1aa40976",
            "value": " 3/3 [00:00&lt;00:00, 76.76it/s]"
          }
        },
        "70aad33e2cd043c08a94c8b8a93a3342": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "892450aebc2b4611aec22ffb88a8cba9": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "1e6314ba4f4a445b9b68928c6debb77c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "ef7d3624b5fe44eea8d40f6b6a02898b": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "e652c021449c485484fb899b17acd0ef": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "0661f3b9e7bc408f82030c5b7c164fcb": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "ab50e73bf5404380baba92cf1aa40976": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        }
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}