multi_label_text_classification.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MZ4OdPWGbqku"
      },
      "source": [
        "### Import & Install Libraries ⭐"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_oQgqG-PYLhk",
        "outputId": "f43fc710-6f91-45cf-eb9c-4724951cfad8"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "[nltk_data] Downloading package wordnet to /root/nltk_data...\n",
            "[nltk_data] Downloading package stopwords to /root/nltk_data...\n",
            "[nltk_data]   Unzipping corpora/stopwords.zip.\n",
            "[nltk_data] Downloading package omw-1.4 to /root/nltk_data...\n"
          ]
        },
        {
          "data": {
            "text/plain": [
              "True"
            ]
          },
          "execution_count": 1,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "import os, shutil, random, re, nltk, torch\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import seaborn as sns\n",
        "\n",
        "from tqdm import tqdm\n",
        "from sklearn.metrics import (\n",
        "    accuracy_score, precision_score, recall_score, f1_score, confusion_matrix,\n",
        "    ConfusionMatrixDisplay\n",
        ")\n",
        "from sklearn.feature_extraction.text import TfidfVectorizer\n",
        "from nltk.corpus import stopwords\n",
        "from nltk.stem import WordNetLemmatizer\n",
        "from tabulate import tabulate\n",
        "from sklearn.linear_model import LogisticRegression, SGDClassifier\n",
        "from sklearn.svm import LinearSVC\n",
        "from transformers import DistilBertTokenizer, DistilBertModel, RobertaTokenizer, RobertaModel\n",
        "from torch.utils.data import Dataset, DataLoader\n",
        "from tabulate import tabulate\n",
        "\n",
        "nltk.download('wordnet')\n",
        "nltk.download('stopwords')\n",
        "nltk.download('omw-1.4')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "K7Dl7DFKTmcX",
        "outputId": "4c8a3791-9223-417e-e2bc-4d5638bdb980"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Collecting contractions\n",
            "  Downloading contractions-0.1.73-py2.py3-none-any.whl.metadata (1.2 kB)\n",
            "Collecting iterative-stratification\n",
            "  Downloading iterative_stratification-0.1.9-py3-none-any.whl.metadata (1.3 kB)\n",
            "Collecting scikit-multilearn\n",
            "  Downloading scikit_multilearn-0.2.0-py3-none-any.whl.metadata (6.0 kB)\n",
            "Collecting textsearch>=0.0.21 (from contractions)\n",
            "  Downloading textsearch-0.0.24-py2.py3-none-any.whl.metadata (1.2 kB)\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.11/dist-packages (from iterative-stratification) (1.26.4)\n",
            "Requirement already satisfied: scipy in /usr/local/lib/python3.11/dist-packages (from iterative-stratification) (1.13.1)\n",
            "Requirement already satisfied: scikit-learn in /usr/local/lib/python3.11/dist-packages (from iterative-stratification) (1.6.1)\n",
            "Collecting anyascii (from textsearch>=0.0.21->contractions)\n",
            "  Downloading anyascii-0.3.2-py3-none-any.whl.metadata (1.5 kB)\n",
            "Collecting pyahocorasick (from textsearch>=0.0.21->contractions)\n",
            "  Downloading pyahocorasick-2.1.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (13 kB)\n",
            "Requirement already satisfied: joblib>=1.2.0 in /usr/local/lib/python3.11/dist-packages (from scikit-learn->iterative-stratification) (1.4.2)\n",
            "Requirement already satisfied: threadpoolctl>=3.1.0 in /usr/local/lib/python3.11/dist-packages (from scikit-learn->iterative-stratification) (3.5.0)\n",
            "Downloading contractions-0.1.73-py2.py3-none-any.whl (8.7 kB)\n",
            "Downloading iterative_stratification-0.1.9-py3-none-any.whl (8.5 kB)\n",
            "Downloading scikit_multilearn-0.2.0-py3-none-any.whl (89 kB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m89.4/89.4 kB\u001b[0m \u001b[31m6.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading textsearch-0.0.24-py2.py3-none-any.whl (7.6 kB)\n",
            "Downloading anyascii-0.3.2-py3-none-any.whl (289 kB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m289.9/289.9 kB\u001b[0m \u001b[31m13.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading pyahocorasick-2.1.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (118 kB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m118.3/118.3 kB\u001b[0m \u001b[31m9.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hInstalling collected packages: scikit-multilearn, pyahocorasick, anyascii, textsearch, iterative-stratification, contractions\n",
            "Successfully installed anyascii-0.3.2 contractions-0.1.73 iterative-stratification-0.1.9 pyahocorasick-2.1.0 scikit-multilearn-0.2.0 textsearch-0.0.24\n"
          ]
        }
      ],
      "source": [
        "! pip install contractions iterative-stratification scikit-multilearn\n",
        "\n",
        "from iterstrat.ml_stratifiers import MultilabelStratifiedKFold\n",
        "from contractions import fix\n",
        "from skmultilearn.ensemble import RakelO"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Pw7jwoXpGuUG",
        "outputId": "7f28180c-418c-4c4c-b90e-bdce4654a27d"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ],
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TjdT2-CAd0jJ"
      },
      "source": [
        "### Data Preparation ⭐"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ULXV-yaIYIWl"
      },
      "source": [
        "##### Load Source Data & Load into Dataframe"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "id": "o5vcSKa0YH4b"
      },
      "outputs": [],
      "source": [
        "# Raw URL files (credit: https://github.com/google-research/google-research/tree/master/goemotions)\n",
        "file_urls = [\n",
        "    'https://storage.googleapis.com/gresearch/goemotions/data/full_dataset/goemotions_1.csv',\n",
        "    'https://storage.googleapis.com/gresearch/goemotions/data/full_dataset/goemotions_2.csv',\n",
        "    'https://storage.googleapis.com/gresearch/goemotions/data/full_dataset/goemotions_3.csv'\n",
        "]\n",
        "\n",
        "# Source data locations\n",
        "source_data_dir = 'source_data'\n",
        "source_data_files = [\n",
        "    'goemotions_1.csv',\n",
        "    'goemotions_2.csv',\n",
        "    'goemotions_3.csv'\n",
        "]\n",
        "\n",
        "# Remove directory and all its contents if it exists\n",
        "if os.path.exists(source_data_dir):\n",
        "    shutil.rmtree(source_data_dir)\n",
        "\n",
        "# Download files into source_data_dir folder\n",
        "for url in file_urls:\n",
        "    os.system(f\"wget -P {source_data_dir} {url}\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BpOky_lVaz2U",
        "outputId": "83bea232-990e-45c4-cc76-82e756bd4767"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 211225 entries, 0 to 211224\n",
            "Data columns (total 37 columns):\n",
            " #   Column                Non-Null Count   Dtype  \n",
            "---  ------                --------------   -----  \n",
            " 0   text                  211225 non-null  object \n",
            " 1   id                    211225 non-null  object \n",
            " 2   author                211225 non-null  object \n",
            " 3   subreddit             211225 non-null  object \n",
            " 4   link_id               211225 non-null  object \n",
            " 5   parent_id             211225 non-null  object \n",
            " 6   created_utc           211225 non-null  float64\n",
            " 7   rater_id              211225 non-null  int64  \n",
            " 8   example_very_unclear  211225 non-null  bool   \n",
            " 9   admiration            211225 non-null  int64  \n",
            " 10  amusement             211225 non-null  int64  \n",
            " 11  anger                 211225 non-null  int64  \n",
            " 12  annoyance             211225 non-null  int64  \n",
            " 13  approval              211225 non-null  int64  \n",
            " 14  caring                211225 non-null  int64  \n",
            " 15  confusion             211225 non-null  int64  \n",
            " 16  curiosity             211225 non-null  int64  \n",
            " 17  desire                211225 non-null  int64  \n",
            " 18  disappointment        211225 non-null  int64  \n",
            " 19  disapproval           211225 non-null  int64  \n",
            " 20  disgust               211225 non-null  int64  \n",
            " 21  embarrassment         211225 non-null  int64  \n",
            " 22  excitement            211225 non-null  int64  \n",
            " 23  fear                  211225 non-null  int64  \n",
            " 24  gratitude             211225 non-null  int64  \n",
            " 25  grief                 211225 non-null  int64  \n",
            " 26  joy                   211225 non-null  int64  \n",
            " 27  love                  211225 non-null  int64  \n",
            " 28  nervousness           211225 non-null  int64  \n",
            " 29  optimism              211225 non-null  int64  \n",
            " 30  pride                 211225 non-null  int64  \n",
            " 31  realization           211225 non-null  int64  \n",
            " 32  relief                211225 non-null  int64  \n",
            " 33  remorse               211225 non-null  int64  \n",
            " 34  sadness               211225 non-null  int64  \n",
            " 35  surprise              211225 non-null  int64  \n",
            " 36  neutral               211225 non-null  int64  \n",
            "dtypes: bool(1), float64(1), int64(29), object(6)\n",
            "memory usage: 58.2+ MB\n"
          ]
        }
      ],
      "source": [
        "# Loop through each file, read into dataframe, and append to list\n",
        "dfs = []\n",
        "for file in source_data_files:\n",
        "    file_path = os.path.join(source_data_dir, file)\n",
        "    df = pd.read_csv(file_path)\n",
        "    dfs.append(df)\n",
        "\n",
        "# Concatenate all dataframe into single dataframe\n",
        "df = pd.concat(dfs, ignore_index=True)\n",
        "\n",
        "df.info()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WKnv8gLTcFkX"
      },
      "source": [
        "##### Condense Dataframe & Check Unique Values"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 26,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hHvW-W9LcFAb",
        "outputId": "1e974c04-cfae-41e3-ee41-fcacb2200cb4"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "example_very_unclear\n",
            "[False  True]\n",
            "--------------------------------------------------\n",
            "neutral\n",
            "[0 1]\n",
            "--------------------------------------------------\n",
            "sadness\n",
            "[1 0]\n",
            "--------------------------------------------------\n",
            "grief\n",
            "[0 1]\n",
            "--------------------------------------------------\n",
            "fear\n",
            "[0 1]\n",
            "--------------------------------------------------\n",
            "nervousness\n",
            "[0 1]\n",
            "--------------------------------------------------\n",
            "remorse\n",
            "[0 1]\n",
            "--------------------------------------------------\n",
            "anger\n",
            "[0 1]\n",
            "--------------------------------------------------\n",
            "annoyance\n",
            "[0 1]\n",
            "--------------------------------------------------\n",
            "confusion\n",
            "[0 1]\n",
            "--------------------------------------------------\n",
            "disappointment\n",
            "[0 1]\n",
            "--------------------------------------------------\n",
            "disapproval\n",
            "[0 1]\n",
            "--------------------------------------------------\n",
            "disgust\n",
            "[0 1]\n",
            "--------------------------------------------------\n",
            "joy\n",
            "[0 1]\n",
            "--------------------------------------------------\n",
            "excitement\n",
            "[0 1]\n",
            "--------------------------------------------------\n",
            "optimism\n",
            "[0 1]\n",
            "--------------------------------------------------\n",
            "love\n",
            "[0 1]\n",
            "--------------------------------------------------\n",
            "caring\n",
            "[0 1]\n",
            "--------------------------------------------------\n",
            "pride\n",
            "[0 1]\n",
            "--------------------------------------------------\n",
            "gratitude\n",
            "[0 1]\n",
            "--------------------------------------------------\n",
            "relief\n",
            "[0 1]\n",
            "--------------------------------------------------\n",
            "curiosity\n",
            "[0 1]\n",
            "--------------------------------------------------\n",
            "admiration\n",
            "[0 1]\n",
            "--------------------------------------------------\n",
            "amusement\n",
            "[0 1]\n",
            "--------------------------------------------------\n"
          ]
        }
      ],
      "source": [
        "# Keep only target labels, text, and indicator if example is unclear\n",
        "condensed_df = df[['text',\n",
        "                   'example_very_unclear',\n",
        "                   'neutral',\n",
        "                   'sadness',\n",
        "                   'grief',\n",
        "                   'fear',\n",
        "                   'nervousness',\n",
        "                   'remorse',\n",
        "                   'anger',\n",
        "                   'annoyance',\n",
        "                   'confusion',\n",
        "                   'disappointment',\n",
        "                   'disapproval',\n",
        "                   'disgust',\n",
        "                   'joy',\n",
        "                   'excitement',\n",
        "                   'optimism',\n",
        "                   'love',\n",
        "                   'caring',\n",
        "                   'pride',\n",
        "                   'gratitude',\n",
        "                   'relief',\n",
        "                   'curiosity',\n",
        "                   'admiration',\n",
        "                   'amusement']\n",
        "                  ]\n",
        "\n",
        "# Check unique values in each column to make sure no invalid entries\n",
        "for col in condensed_df.columns[1:]:\n",
        "    unique_values = condensed_df[col].unique()\n",
        "    print(col)\n",
        "    print(unique_values)\n",
        "    print(\"-\" * 50)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ewIjghW8hdfF"
      },
      "source": [
        "##### Remove Unclear & 'Empty' Rows,  Print Samples"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 27,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tXH-0meQgMxE",
        "outputId": "2fb8a786-c73a-4693-da57-be875c53b6bb"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Number of unclear examples removed: 3411\n",
            "Number of \"empty\" examples removed: 24230\n",
            "\n",
            "Text: Sweet, I've put that as location and took a note of that route. You fancy it?\n",
            "Emotions: ['neutral']\n",
            "\n",
            "Text: your cat has the cutest face!\n",
            "Emotions: ['admiration']\n",
            "\n",
            "Text: She's a good chief stew and very good at her JOB but a shitty people person and a shitty leader.\n",
            "Emotions: ['admiration']\n",
            "\n",
            "Number of rows before dropping: 211225\n",
            "Number of rows after dropping: 183584\n",
            "Number dropped: 27641\n",
            "\n",
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 183584 entries, 0 to 183583\n",
            "Data columns (total 24 columns):\n",
            " #   Column          Non-Null Count   Dtype \n",
            "---  ------          --------------   ----- \n",
            " 0   text            183584 non-null  object\n",
            " 1   neutral         183584 non-null  int64 \n",
            " 2   sadness         183584 non-null  int64 \n",
            " 3   grief           183584 non-null  int64 \n",
            " 4   fear            183584 non-null  int64 \n",
            " 5   nervousness     183584 non-null  int64 \n",
            " 6   remorse         183584 non-null  int64 \n",
            " 7   anger           183584 non-null  int64 \n",
            " 8   annoyance       183584 non-null  int64 \n",
            " 9   confusion       183584 non-null  int64 \n",
            " 10  disappointment  183584 non-null  int64 \n",
            " 11  disapproval     183584 non-null  int64 \n",
            " 12  disgust         183584 non-null  int64 \n",
            " 13  joy             183584 non-null  int64 \n",
            " 14  excitement      183584 non-null  int64 \n",
            " 15  optimism        183584 non-null  int64 \n",
            " 16  love            183584 non-null  int64 \n",
            " 17  caring          183584 non-null  int64 \n",
            " 18  pride           183584 non-null  int64 \n",
            " 19  gratitude       183584 non-null  int64 \n",
            " 20  relief          183584 non-null  int64 \n",
            " 21  curiosity       183584 non-null  int64 \n",
            " 22  admiration      183584 non-null  int64 \n",
            " 23  amusement       183584 non-null  int64 \n",
            "dtypes: int64(23), object(1)\n",
            "memory usage: 33.6+ MB\n"
          ]
        }
      ],
      "source": [
        "num_rows_before = condensed_df.shape[0]\n",
        "\n",
        "# Drop unclear examples\n",
        "condensed_df = condensed_df.loc[condensed_df['example_very_unclear'] != 1]\n",
        "condensed_df = condensed_df.drop(columns=['example_very_unclear'])\n",
        "\n",
        "num_rows_after_drop_unclear = condensed_df.shape[0]\n",
        "print(f'Number of unclear examples removed: {num_rows_before - num_rows_after_drop_unclear}')\n",
        "\n",
        "# Drop 'empty' examples ('empty' meaning row has no positive labels)\n",
        "condensed_df = condensed_df.loc[condensed_df.iloc[:, 1:].sum(axis=1) > 0]\n",
        "\n",
        "num_rows_after_drop_empty = condensed_df.shape[0]\n",
        "print(f'Number of \"empty\" examples removed: {num_rows_after_drop_unclear - num_rows_after_drop_empty}\\n')\n",
        "\n",
        "# Reset indexes after dropping rows\n",
        "condensed_df = condensed_df.reset_index(drop=True)\n",
        "\n",
        "num_rows_after = condensed_df.shape[0]\n",
        "\n",
        "# Print three examples and corresponding emotions\n",
        "num_rows = len(condensed_df.index)\n",
        "indexes = random.sample(range(0, num_rows), 3)\n",
        "\n",
        "for i in indexes:\n",
        "    row = condensed_df.loc[i]\n",
        "    text, emotions = row['text'], row[row == 1].index.tolist()\n",
        "    print(f\"Text: {text}\\nEmotions: {emotions}\\n\")\n",
        "\n",
        "# Print num of rows before and after dropping\n",
        "print(f'Number of rows before dropping: {num_rows_before}\\nNumber of rows after dropping: {num_rows_after}\\nNumber dropped: {num_rows_before - num_rows_after}\\n')\n",
        "\n",
        "condensed_df.info()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YoGZCZJLtYEz"
      },
      "source": [
        "#####  Plot Label Value Counts"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0qPJp0a_tdJP",
        "outputId": "20073890-1bc2-4c47-e2d7-e504f1cf3e94"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 900x400 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1100x1100 with 23 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Calculate positive counts for each label\n",
        "label_col_names = condensed_df.columns[1:]\n",
        "positive_counts = (condensed_df[label_col_names] == 1).sum()\n",
        "\n",
        "# Plot counts for each column\n",
        "plt.figure(figsize=(9, 4))\n",
        "positive_counts.plot(kind='bar')\n",
        "plt.title('Total Postive Counts / Label')\n",
        "plt.xlabel('Label')\n",
        "plt.ylabel('Count')\n",
        "plt.xticks(rotation=45, ha='right')  # Rotate x-axis labels for better visibility\n",
        "plt.tight_layout()\n",
        "plt.show()\n",
        "print()\n",
        "\n",
        "# Calculate negative count for each label\n",
        "negative_counts = (condensed_df[label_col_names] == 0).sum()\n",
        "\n",
        "# Create larger plot with subplots\n",
        "num_labels = len(label_col_names)\n",
        "fig, axes = plt.subplots(nrows=int(np.ceil(num_labels / 3)), ncols=3, figsize=(11, 11))\n",
        "axes = axes.flatten()\n",
        "\n",
        "# Plot each label's positive and negative counts in separate subplots\n",
        "for i, label_name in enumerate(label_col_names):\n",
        "    axes[i].bar(['Positive', 'Negative'], [positive_counts[label_name], negative_counts[label_name]], color=['green', 'red'])\n",
        "    axes[i].set_title(label_name)\n",
        "    axes[i].set_ylabel('Count')\n",
        "\n",
        "# Remove empty subplots (if any)\n",
        "for j in range(i + 1, len(axes)):\n",
        "    fig.delaxes(axes[j])\n",
        "\n",
        "plt.suptitle('Counts of Positive and Negative Values / Label')\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ivDG0XuOslrM"
      },
      "source": [
        "### Data Preprocessing ⭐"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Y0azMy_UweYM"
      },
      "source": [
        "##### Text Cleaning (Baseline)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "gn55osaisp31"
      },
      "outputs": [],
      "source": [
        "def clean_text(text):\n",
        "    # Convert to lowercase\n",
        "    text = text.lower()\n",
        "\n",
        "    # Expand contractions (e.g., \"can't\" -> \"cannot\")\n",
        "    text = fix(text)\n",
        "\n",
        "    # -------------------------------------------------------------------\n",
        "    # (Comment out or remove for baseline - reintroduce later to test impact)\n",
        "    # Replace emoticons with descriptive tokens\n",
        "    # emoticons = {\n",
        "    #     r':\\)': '[SMILEYFACE]',\n",
        "    #     r';\\)': '[WINKYFACE]',\n",
        "    #     r':d': '[SMILEYFACE]',\n",
        "    #     r';d': '[WINKYFACE]',\n",
        "    #     r':\\(': '[FROWNYFACE]',\n",
        "    #     r':o': '[SURPRISEDFACE]'\n",
        "    # }\n",
        "    # for pattern, replacement in emoticons.items():\n",
        "    #     text = re.sub(pattern, replacement, text)\n",
        "\n",
        "    # Replace exclamation marks and question marks with placeholders\n",
        "    # text = re.sub(r'!', ' [EXCLAMATIONMARK]', text)\n",
        "    # text = re.sub(r'\\?', ' [QUESTIONMARK]', text)\n",
        "    # -------------------------------------------------------------------\n",
        "\n",
        "    # 3) Replace URLs with [URL]\n",
        "    text = re.sub(r'(https?://\\S+|www\\.\\S+)', '[URL]', text)\n",
        "\n",
        "    # 4) Replace @handles with [USER]\n",
        "    text = re.sub(r'@\\w+', '[USER]', text)\n",
        "\n",
        "    # -------------------------------------------------------------------\n",
        "    # (Comment out or remove for baseline - reintroduce later to test impact)\n",
        "    # Convert variations of [name] to [NAME]\n",
        "    # text = re.sub(r'\\[name\\]', '[NAME]', text, flags=re.IGNORECASE')\n",
        "\n",
        "    # Replace ellipses (two or more dots) with [ELLIPSIS]\n",
        "    # text = re.sub(r'\\.{2,}', ' [ELLIPSIS]', text)\n",
        "\n",
        "    # Remove excessive repeated characters (e.g., \"mmeEEEE\" -> \"me\")\n",
        "    # text = re.sub(r'([a-z])\\1{2,}', r'\\1', text)\n",
        "\n",
        "    # Remove all non-alphabetical characters except spaces and square brackets\n",
        "    # text = re.sub(r'[^a-zA-Z\\s\\[\\]]', '', text)\n",
        "    # -------------------------------------------------------------------\n",
        "\n",
        "    # 5) Remove extra spaces\n",
        "    text = re.sub(r'\\s+', ' ', text).strip()\n",
        "\n",
        "    return text\n",
        "\n",
        "condensed_df['cleaned_text'] = condensed_df['text'].apply(clean_text)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "D9MKyyJxRhBb",
        "outputId": "89928342-36e7-45ce-98da-059c6ede783a"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Average length before cleaning: 69.13 chars\n",
            "Average length after cleaning: 69.88 chars\n",
            "\n",
            "Dirty Review: I’ll do that. Thank you\n",
            "Clean Review: i will do that. thank you\n",
            "\n",
            "Dirty Review: Go to your room.\n",
            "Clean Review: go to your room.\n",
            "\n",
            "Dirty Review: Cold approach is a huge waste of time due to the ratio of success-failure\n",
            "Clean Review: cold approach is a huge waste of time due to the ratio of success-failure\n",
            "\n",
            "Dirty Review: Please seek medical help\n",
            "Clean Review: please seek medical help\n",
            "\n",
            "Dirty Review: Me too, my sister and I thought we became idiots, we even googled shapes and their sides\n",
            "Clean Review: me too, my sister and i thought we became idiots, we even googled shapes and their sides\n",
            "\n"
          ]
        }
      ],
      "source": [
        "avg_length_before = condensed_df['text'].str.len().mean()\n",
        "avg_length_after = condensed_df['cleaned_text'].str.len().mean()\n",
        "\n",
        "print(f\"Average length before cleaning: {avg_length_before:.2f} chars\")\n",
        "print(f\"Average length after cleaning: {avg_length_after:.2f} chars\\n\")\n",
        "\n",
        "# Print few reviews before and after cleaning\n",
        "num_rows = len(condensed_df.index)\n",
        "indexes = random.sample(range(0, num_rows), 5)\n",
        "\n",
        "for i in indexes:\n",
        "    review_dirty, review_clean = condensed_df['text'].loc[i], condensed_df['cleaned_text'].loc[i]\n",
        "    print(f\"Dirty Review: {review_dirty}\\nClean Review: {review_clean}\\n\")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "EfYVilooLFKV"
      },
      "source": [
        "##### Stop Word Removal & Lemmatization (Baseline)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "sFdNZQhNlZQo"
      },
      "outputs": [],
      "source": [
        "# # Initialize stop words and lemmatizer\n",
        "# stop_words = set(stopwords.words('english'))\n",
        "# lemmatizer = WordNetLemmatizer()\n",
        "\n",
        "# # Removes stopwords and performs lemmatization\n",
        "# # Stopwords: such as \"the,\" \"is,\" \"at,\" and \"which\" | Lemmatization: reducing words to their base or dictionary form, known as the lemma (e.g.,  \"running\" and \"ran\" would both be reduced to \"run\")\n",
        "# def remove_stopWords_lemmatize(text):\n",
        "#     # Tokenize text into words\n",
        "#     words = text.split()\n",
        "\n",
        "#     # Remove stopwords and lemmatize\n",
        "#     processed_words = [lemmatizer.lemmatize(word) for word in words if word not in stop_words]\n",
        "\n",
        "#     # Join words back into single string\n",
        "#     return ' '.join(processed_words)\n",
        "\n",
        "# # Apply preprocessing to cleaned_reviewBody column\n",
        "# avg_length_before = condensed_df['cleaned_text'].str.len().mean()\n",
        "# condensed_df['cleaned_text'] = condensed_df['cleaned_text'].apply(remove_stopWords_lemmatize)\n",
        "\n",
        "# # Print avg char length before and after cleaning + stop word removal & lemmatization\n",
        "# avg_length_after = condensed_df['cleaned_text'].str.len().mean()\n",
        "\n",
        "# print(f\"Average length before cleaning: {avg_length_before:.2f} chars\")\n",
        "# print(f\"Average length after cleaning: {avg_length_after:.2f} chars\\n\")\n",
        "\n",
        "# # Print sample reviews before and after cleaning + stop word removal & lemmatization\n",
        "# num_rows = len(condensed_df.index)\n",
        "# indexes = random.sample(range(0, num_rows), 3)\n",
        "\n",
        "# for i in indexes:\n",
        "#     review_dirty, review_clean = condensed_df['text'].loc[i], condensed_df['cleaned_text'].loc[i]\n",
        "#     print(f\"Dirty Review: {review_dirty}\\nClean Review: {review_clean}\\n\")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NiravP_Vsn7d"
      },
      "source": [
        "##### Vocabulary Size & Words Per Sample Stats"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5ih0wk9ZsnVM",
        "outputId": "c4ce9272-9c27-4ab4-da7e-21aa8273f6b9"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Vocabulary size: 59060\n",
            "Average number of words per sample: 13.49\n",
            "Maximum words in a sample: 34\n",
            "Minimum words in a sample: 1\n"
          ]
        }
      ],
      "source": [
        "# Flatten all text into one string and split into words\n",
        "all_text = \" \".join(condensed_df['cleaned_text'])\n",
        "\n",
        "# Create set of all unqiue words & count\n",
        "vocabulary = set(all_text.split())\n",
        "vocab_size = len(vocabulary)\n",
        "\n",
        "print(f'Vocabulary size: {vocab_size}')\n",
        "\n",
        "# Calculate number of words in each sample\n",
        "word_counts = condensed_df['cleaned_text'].apply(lambda x: len(x.split()))\n",
        "\n",
        "# Compute the average number of words per sample\n",
        "avg_words = word_counts.mean()\n",
        "\n",
        "print(f'Average number of words per sample: {avg_words:.2f}')\n",
        "\n",
        "# Compute the highest and lowest word counts in the samples\n",
        "max_words = word_counts.max()\n",
        "min_words = word_counts.min()\n",
        "\n",
        "print(f'Maximum words in a sample: {max_words}')\n",
        "print(f'Minimum words in a sample: {min_words}')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_Zhl6VFCwZb1"
      },
      "source": [
        "##### Train-Test Split & Visualization"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sTDfTznoruAc",
        "outputId": "1958b344-e5d8-4b9f-d558-df5d7925af43"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Total number of examples in dataset: 183584\n",
            "Total number of examples in train split: 146830 (79.98%)\n",
            "Total number of examples in test split: 36754 (20.02%)\n"
          ]
        }
      ],
      "source": [
        "# Initialize multi-label k-Fold splitter (https://github.com/trent-b/iterative-stratification/tree/master)\n",
        "mskf = MultilabelStratifiedKFold(n_splits=5, shuffle=True, random_state=42)\n",
        "\n",
        "# Generate train-test splits and retrieve first split (change later when running kfold)\n",
        "for train_idxs, test_idxs in mskf.split(condensed_df, condensed_df):\n",
        "    train_df = condensed_df.iloc[train_idxs].reset_index(drop=True)\n",
        "    test_df = condensed_df.iloc[test_idxs].reset_index(drop=True)\n",
        "    break\n",
        "\n",
        "dataset_size = condensed_df.shape[0]\n",
        "train_size = train_df.shape[0]\n",
        "test_size = test_df.shape[0]\n",
        "print(f'Total number of examples in dataset: {dataset_size}\\n'\n",
        "      f'Total number of examples in train split: {train_size} ({(train_size / dataset_size) * 100:.2f}%)\\n'\n",
        "      f'Total number of examples in test split: {test_size} ({(test_size / dataset_size) * 100:.2f}%)')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "13yR7H1Lq9GN",
        "outputId": "b01282fb-a16f-47a5-c2e3-a8cc6c9b753b"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1000x600 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Extract target label col names (assumes all int64 columns are labels)\n",
        "label_col_names = [col for col in train_df.columns if train_df[col].dtype == 'int64']\n",
        "\n",
        "# Count positive instances / label in each split\n",
        "train_counts = train_df[label_col_names].sum()\n",
        "test_counts = test_df[label_col_names].sum()\n",
        "\n",
        "x = np.arange(len(label_col_names))\n",
        "bar_width = 0.4\n",
        "\n",
        "# Side-by-side bar chart\n",
        "plt.figure(figsize=(10, 6))\n",
        "plt.bar(x - bar_width/2, train_counts, bar_width, label='Train', color='blue')\n",
        "plt.bar(x + bar_width/2, test_counts, bar_width, label='Test', color='orange')\n",
        "\n",
        "plt.xticks(x, label_col_names, rotation=45, ha='right')\n",
        "plt.ylabel('Number of Positive Examples')\n",
        "plt.title('Distribution of Labels in Train vs Test')\n",
        "plt.legend()\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "y9EsRiNIR3yA"
      },
      "source": [
        "### Modeling & Evaluation ⭐"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SsBJX0Ho3hfl"
      },
      "source": [
        "#### Shared Functions"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "6FqpofIt3bhm"
      },
      "outputs": [],
      "source": [
        "def eval_clf_preds(y_true, y_preds, binary_clf_evals=False):\n",
        "    multiLabel_metrics = {\n",
        "        \"Accuracy\": accuracy_score(y_true, y_preds),\n",
        "        \"Precision\": precision_score(y_true, y_preds, average='micro', zero_division=0),\n",
        "        \"Recall\": recall_score(y_true, y_preds, average='micro', zero_division=0),\n",
        "        \"F1-Score\": f1_score(y_true, y_preds, average='micro', zero_division=0)\n",
        "    }\n",
        "\n",
        "    eval_results = {\"Multi-label\": multiLabel_metrics}\n",
        "\n",
        "    if binary_clf_evals:\n",
        "        individual_evals = []\n",
        "        num_labels = y_true.shape[1]\n",
        "        for i in range(num_labels):\n",
        "            acc = accuracy_score(y_true.iloc[:, i], y_preds.iloc[:, i])\n",
        "            prec = precision_score(y_true.iloc[:, i], y_preds.iloc[:, i], zero_division=0)\n",
        "            rec = recall_score(y_true.iloc[:, i], y_preds.iloc[:, i], zero_division=0)\n",
        "            f1 = f1_score(y_true.iloc[:, i], y_preds.iloc[:, i], zero_division=0)\n",
        "            individual_evals.append({\n",
        "                \"Target Label\": f\"{y_true.columns[i]}\",\n",
        "                \"Accuracy\": acc,\n",
        "                \"Precision\": prec,\n",
        "                \"Recall\": rec,\n",
        "                \"F1-Score\": f1\n",
        "            })\n",
        "        eval_results[\"Binary\"] = individual_evals\n",
        "\n",
        "    return eval_results"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sNn8KnPxxvMh"
      },
      "source": [
        "#### Using Manually Extracted Features"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "r_pJsvi-oTKt"
      },
      "source": [
        "##### Extract Features from Text using TF-ID"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4p_z9PgMoYKQ",
        "outputId": "2b391bfa-eb5e-47c8-9aae-04f34774b4bd"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "TF-IDF Features for Training Data: (146830, 100)\n",
            "TF-IDF Features for Testing Data: (36754, 100)\n"
          ]
        }
      ],
      "source": [
        "# Initialize TF-ID feature vectorizer\n",
        "vectorizer = TfidfVectorizer(max_features=100)\n",
        "\n",
        "# Extract features from text instances in train and test splits\n",
        "X_train = vectorizer.fit_transform(train_df['cleaned_text'])\n",
        "X_test = vectorizer.transform(test_df['cleaned_text'])\n",
        "\n",
        "# Define train and test multi-label targets\n",
        "y_train = train_df.iloc[ : ,1:24]\n",
        "y_test = test_df.iloc[ : ,1:24]\n",
        "\n",
        "# (Number of documents, Number of featurs)\n",
        "print(f\"TF-IDF Features for Training Data: {X_train.shape}\")\n",
        "print(f\"TF-IDF Features for Testing Data: {X_test.shape}\")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "iVJhrM6pnCxI"
      },
      "source": [
        "##### Logistic Regression & SVM Models for Multi-Label Targets (Binary Relevance)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GkVSUAsRxUo-"
      },
      "source": [
        "###### Train One Logistic Regression Model & One SVM Model Per Label, Evaluate Performance"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3gZC0cX5xUo_",
        "outputId": "cad8aa5d-78ef-40f4-8a61-f6b7caf3324c"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "Training Logistic Regression Classifiers: 100%|██████████| 22/22 [00:13<00:00,  1.59it/s]\n",
            "Training SVM Classifiers: 100%|██████████| 22/22 [00:10<00:00,  2.13it/s]\n"
          ]
        }
      ],
      "source": [
        "# Train one classifier per emotion\n",
        "lr_classifiers = []\n",
        "for i in tqdm(range(y_train.shape[1]), desc=\"Training Logistic Regression Classifiers\"):\n",
        "    clf = LogisticRegression()\n",
        "    clf.fit(X_train, y_train.iloc[:, i])\n",
        "    lr_classifiers.append(clf)\n",
        "\n",
        "# Train one SVM linear model per emotion\n",
        "svm_classifiers = []\n",
        "for i in tqdm(range(y_train.shape[1]), desc=\"Training SVM Classifiers\"):\n",
        "    clf = LinearSVC(max_iter=20)  # Specify linear kernel for SVM\n",
        "    clf.fit(X_train, y_train.iloc[:, i])\n",
        "    svm_classifiers.append(clf)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HDbCLaEWxUo_",
        "outputId": "a3c3530d-a4d9-46c3-9603-b0d6b7781ca9"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Overall Multi-label Classification Metrics for Logistic Regression on train dataset:\n",
            "Accuracy: 0.0958\n",
            "Precision: 0.7467\n",
            "Recall: 0.0703\n",
            "F1-Score: 0.1284\n",
            "(Samples tested: 146830)\n",
            "--------------------------------------------------\n",
            "Overall Multi-label Classification Metrics for Logistic Regression on test dataset:\n",
            "Accuracy: 0.0978\n",
            "Precision: 0.7489\n",
            "Recall: 0.0714\n",
            "F1-Score: 0.1304\n",
            "(Samples tested: 36754)\n",
            "--------------------------------------------------\n",
            "Overall Multi-label Classification Metrics for SVM on train dataset:\n",
            "Accuracy: 0.0914\n",
            "Precision: 0.7809\n",
            "Recall: 0.0661\n",
            "F1-Score: 0.1218\n",
            "(Samples tested: 146830)\n",
            "--------------------------------------------------\n",
            "Overall Multi-label Classification Metrics for SVM on test dataset:\n",
            "Accuracy: 0.0922\n",
            "Precision: 0.7753\n",
            "Recall: 0.0662\n",
            "F1-Score: 0.1219\n",
            "(Samples tested: 36754)\n",
            "--------------------------------------------------\n"
          ]
        }
      ],
      "source": [
        "# Store predictions of individual logistic regression classifiers on train and tests sets\n",
        "csfs_y_preds_train = pd.DataFrame({i: clf.predict(X_train) for i, clf in enumerate(lr_classifiers)})\n",
        "csfs_y_preds_test = pd.DataFrame({i: clf.predict(X_test) for i, clf in enumerate(lr_classifiers)})\n",
        "\n",
        "# Evaluate performance on multi-label task\n",
        "results_train_lr = evaluate_clf_models(y_train, csfs_y_preds_train, dataset_type=\"train\", model_type=\"Logistic Regression\")\n",
        "results_test_lr = evaluate_clf_models(y_test, csfs_y_preds_test, dataset_type=\"test\", model_type=\"Logistic Regression\")\n",
        "\n",
        "# Store predictions of individual svm classifiers on train and tests sets\n",
        "csfs_y_preds_train = pd.DataFrame({i: clf.predict(X_train) for i, clf in enumerate(svm_classifiers)})\n",
        "csfs_y_preds_test = pd.DataFrame({i: clf.predict(X_test) for i, clf in enumerate(svm_classifiers)})\n",
        "\n",
        "# Evaluate performance on multi-label task\n",
        "results_train_svm = evaluate_clf_models(y_train, csfs_y_preds_train, dataset_type=\"train\", model_type=\"SVM\")\n",
        "results_test_svm = evaluate_clf_models(y_test, csfs_y_preds_test, dataset_type=\"test\", model_type=\"SVM\")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "W7izZznkWyWt"
      },
      "source": [
        "##### RakelO (Ensemble Method)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kccdUkTICpRe"
      },
      "source": [
        "###### Train RakeIo Classifier, Evaluate Performance"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "X01qnBZ-WxZy"
      },
      "outputs": [],
      "source": [
        "# Note: Model reuses TFID-feats extracted previously\n",
        "\n",
        "# Docs: http://scikit.ml/api/skmultilearn.ensemble.rakelo.html#skmultilearn.ensemble.RakelO\n",
        "cfr = RakelO(\n",
        "    base_classifier=LogisticRegression(max_iter=1000),\n",
        "    model_count=(2 * y_train.shape[1])\n",
        ")\n",
        "\n",
        "# Train classifier on training data\n",
        "cfr.fit(X_train, y_train)\n",
        "\n",
        "# Store predictions of classifier\n",
        "y_pred_train = cfr.predict(X_train)\n",
        "y_pred_test = cfr.predict(X_test)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sND9XypMZQIm",
        "outputId": "254d90a0-8bc0-4c77-b5cc-5cb349dcf99c"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Overall Multi-label Classification Metrics for RakelO (LR Ensemble) on train dataset:\n",
            "Accuracy: 0.1013\n",
            "Precision: 0.7198\n",
            "Recall: 0.0757\n",
            "F1-Score: 0.1370\n",
            "(Samples tested: 146830)\n",
            "--------------------------------------------------\n",
            "Overall Multi-label Classification Metrics for RakelO (LR Ensemble) on test dataset:\n",
            "Accuracy: 0.1028\n",
            "Precision: 0.7228\n",
            "Recall: 0.0763\n",
            "F1-Score: 0.1381\n",
            "(Samples tested: 36754)\n",
            "--------------------------------------------------\n"
          ]
        }
      ],
      "source": [
        "# Evaluate performance\n",
        "results_train_rakeio = evaluate_clf_model(y_train, y_pred_train, 'train', 'RakelO (LR Ensemble)')\n",
        "results_test_rakeio = evaluate_clf_model(y_test, y_pred_test, 'test', 'RakelO (LR Ensemble)')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sgGFjqOzzs_c"
      },
      "source": [
        "#### Using Learned Features"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DFZr31_oV0UJ"
      },
      "source": [
        "##### Extracting Feature Embeddings using DistilBERT & RoBERTa"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pvcj_X8snEt-"
      },
      "source": [
        "###### Load Instances of Tokenizers and Model Classes with Pre-Trained Models"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424,
          "referenced_widgets": [
            "94fcde6fbb4748569ae0b65fa32dd45a",
            "2d9b38d1a35149bfadc1a304015cba92",
            "171f40bc89d6437eaa8534a743038349",
            "b5d78abe651d4873b4293edb1c910675",
            "e952f73bb71b42e2ae358eef9a519db0",
            "0e1e9f21aca74539b8b610e6b4d5c4b0",
            "e4a4eb51118848649ba38369178f9905",
            "4c6d013dd817420780e7cf4c379c667d",
            "7ebc447c3dd44b9b97aee5a6e969c199",
            "cf801b8a1fcc455da2306242d9f54136",
            "85bc01eff9e045b7a5dda7055e17f0de",
            "a44c195268324772ab260d14f1d38dd9",
            "deffb02b2090469e88a0fd3330a04557",
            "9d219091b8cc4be990cffa479edbaf64",
            "8d4f25d91cd54cf39d6ed414431e7255",
            "32fdaba87bd44637901f1d8c0c73e513",
            "ac33694f66994c3e9b6a2e5e078fbfe5",
            "9f597c67fcda441c9037cf69b25ee4ac",
            "f7ef09bb64fe4a0d9be14cd2eece8f65",
            "bbc946d030cc4bdeb828edca824356b5",
            "530e04ed3ab24eb8bb6f852d2fafffae",
            "89bcbe0eb4084684a5f4c498e0ff26c7",
            "210f48c9e624498d8bdbaafa19510a2c",
            "cf4cdf07224147b59ea813fc1ce479af",
            "b2b9eea953a541f3a0c7719d9d4bee54",
            "fe9a348c3f0b4226901813b32905f517",
            "e434200356ce43a7bca5b5cf1fbc53b6",
            "c547a419af514f2ab7034a127be355d4",
            "83ce9dac48454bdebf6a9f336e7eca68",
            "83a0cbc2f5304587a4877ef4e4c202d9",
            "8fa0e4639c524185ad701c68d7c9ef4c",
            "f9b8cbed4a664e9aad77deea6f2b8c39",
            "72a7784ab9b644b6901a421afb1a19e5",
            "67708facdfe64ce587170066efc51f8a",
            "4a0fb05c031e42c4b9b1194a30c5da9f",
            "5b3d5600f74449a2ae434e9c8551ee41",
            "2b944f0e05ee40f8b1985bc57dfc351e",
            "5fdb52f0866c4fd2a1aef95b773beace",
            "832d655d892a4b6b9f0f07655bd281fd",
            "53ef66d4635a42c28b6d52ff76f5e099",
            "4079af81b66744b991c307ca0850f3f8",
            "c44a9de67b66454f8e6027f933e9c055",
            "1b7b474aca0b484baa8d31c29b33a927",
            "24eb9a1543994ebabc78fc21b38395f1",
            "cd7c8ce702ab4cc2bcea50e61eb7cf89",
            "0eb6efd617a04be782ecffda4c386669",
            "5ef2208a369245579dd9c64f3904003c",
            "32a3e0e82d6b4e4d89622f2b1d825487",
            "e8b4e83df5e74fbba199a65bcabda497",
            "4ed1d2955e8b4512aacdb235bd70ef07",
            "9d1e2268fe7148df924023500d4d22f2",
            "959373a8b1f14477b18fa48d23ee10b2",
            "7656f56c662746f7b9d03813e37e3656",
            "751b9fc0ab0043ac9971411e7302ad89",
            "725b9f3a838945df92f7b316b67f3927",
            "36986e412bea480b943b0f8c38ddb4a3",
            "1b68b31a65a349ae843369b8a0b03323",
            "d007c33743d6486aa84b0a62fa4e89a9",
            "37a216a4c7e745f095b788eb4753cebd",
            "fc5c525a02c247e6b438d5cd00d7efa7",
            "9d9b592726bf40638ac4515ee8b04ec9",
            "b0579c20e1dd45baa4eeb0af17dd920f",
            "5bbbe2ebd32c4d82909ef543d5008caa",
            "221ac8bc9583414584c39796d4ee7ab8",
            "27418a91eee94ac193258ee194f4412d",
            "21540d448e6c460cbaad965a3659dc99",
            "0854e273d3154a78a00a685854e6e7a4",
            "a3f67146220f400c82ff0236d07162f4",
            "6cf7f72efe974f33b8849f3ab049daec",
            "cf856c7982c94828af0e1e7d8e1e5285",
            "e56544d71940410fa293f3888654c634",
            "cc4ec11b351c48d78c8b573fd40301df",
            "645f4bb400e34efab18ffe63946f5329",
            "9e188051f4c44e3f81b473843f762610",
            "5f3b86b081f54f889a7e50ad4223a734",
            "909fa9050b804169a443223310fd6dc9",
            "56a31ac940be454da77c66bd2d9e2e2e",
            "7bd2fc7d573a4bbaad2699066f9e4b1d",
            "538d773cee224c3cb16064fcc851ac59",
            "29d4741a47a146d4be5a98914c3c3ef4",
            "44da2b9729b84af5b22533c92867f824",
            "955d93cae7b04cb38b75e937e1c6b025",
            "0dcee213b9c944ddad9459cbc08d2084",
            "af259a9b6d064bada3350a45da3961d6",
            "fe03e0bba3ba4413879e0aab1acfa73b",
            "328b0579a6e740cdb7f4c36625c9a063",
            "baf77f9862bc43728f1a0e1f42fdf2fd",
            "55331618b0ef41ff8476ea8f8f035cce",
            "fcb33bb07eae44189c50b95c90988fa8",
            "32c49056d3d5465f8bdfe1b8fd5e22a2",
            "c7de73c1144343269cfd7f0786267981",
            "a3377ac61e3b42ccb632dee2091795b0",
            "c7ce951e4e1f4e93b6e27b5d40376060",
            "6e32e3bd91c049b88833ad41ff4d0ab3",
            "09013d348dcc44669b6c4da46fc4e6bf",
            "716e17e0b8604bf68c43764fcbb8633a",
            "ff86746a739a47c98af4e72810c20177",
            "dc95fea471be407b9972db4702b1a089",
            "8765f04caa284ec2bca71aa96a57adac",
            "e822f7a637d449ff96e29c4349454baa",
            "98562d71d83141599a7b481afdc05e88",
            "82eb5391297a4c1993258c7e816642d9",
            "6669401af59949558d684420fbe80715",
            "d58dba0481b147e1a55b243c0b130253",
            "bf96e6e5ca6e439fa12b22913d7c0ff3",
            "969eadd452e045da9ea282ed2f968d3b",
            "d7ef0074ac4a40db880ff1446ee08087",
            "4b6bef99bd0a463cacb1988e6a0f7223",
            "09af5f6f152b4b908810b4284bd937a6",
            "93232b4e658e4555a984a512ac929de1",
            "bf6c4f5b56594db7bdf19b488d75660f",
            "9c5173c7f40d4d2d9e9667d58009920c",
            "042b7cd400c241f794f901805aaa1cd3",
            "f2b199aadfb94784bca9834ed79f5af9",
            "4b973a3d4cef46efbc3b472462f8f2b0",
            "4292de59507a45f1a1948e1dbc48f096",
            "8c77c28c6e5f470eb0fe51875a70a9fe",
            "f866b3e489ed4bf49c8a75b321612d0d",
            "1fc331390e7748438981f98c4a09f88b",
            "f926396796ca4406a2230aa4632c02a8",
            "2bc953c0e4324f9b99ca60e4896a0b17"
          ]
        },
        "id": "HIQaJO2zwHd4",
        "outputId": "88af882b-b1a8-4575-850a-07fe0080e854"
      },
      "outputs": [
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "94fcde6fbb4748569ae0b65fa32dd45a",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "tokenizer_config.json:   0%|          | 0.00/48.0 [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "a44c195268324772ab260d14f1d38dd9",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "vocab.txt:   0%|          | 0.00/232k [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "210f48c9e624498d8bdbaafa19510a2c",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "tokenizer.json:   0%|          | 0.00/466k [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "67708facdfe64ce587170066efc51f8a",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "config.json:   0%|          | 0.00/483 [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "cd7c8ce702ab4cc2bcea50e61eb7cf89",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "model.safetensors:   0%|          | 0.00/268M [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "36986e412bea480b943b0f8c38ddb4a3",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "tokenizer_config.json:   0%|          | 0.00/25.0 [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "0854e273d3154a78a00a685854e6e7a4",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "vocab.json:   0%|          | 0.00/899k [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "7bd2fc7d573a4bbaad2699066f9e4b1d",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "merges.txt:   0%|          | 0.00/456k [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "fcb33bb07eae44189c50b95c90988fa8",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "tokenizer.json:   0%|          | 0.00/1.36M [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "e822f7a637d449ff96e29c4349454baa",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "config.json:   0%|          | 0.00/481 [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "bf6c4f5b56594db7bdf19b488d75660f",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "model.safetensors:   0%|          | 0.00/499M [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "Some weights of RobertaModel were not initialized from the model checkpoint at roberta-base and are newly initialized: ['roberta.pooler.dense.bias', 'roberta.pooler.dense.weight']\n",
            "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n"
          ]
        }
      ],
      "source": [
        "def load_tokenizer_model_instances(class_pairs, model_names):\n",
        "    \"\"\"\n",
        "    Loads tokenizer and model instances for each provided model name using the corresponding classes.\n",
        "    Args:\n",
        "        class_pairs (List[Tuple[TokenizerClass, ModelClass]]):\n",
        "            A list of tuples where each tuple pairs a tokenizer class and model class from Hugging Face's Transformers library.\n",
        "        model_names (List[str]):\n",
        "            A list of pre-trained model names. Where model name i is associated to class pair i in class_pairs.\n",
        "    Returns:\n",
        "        List[Tuple[tokenizer, model]]:\n",
        "            A list of tuples, where each tuple contains a tokenizer instance and a model instance loaded with the pre-trained model\n",
        "            and set to evaluation mode.\n",
        "    \"\"\"\n",
        "    loaded_tokenizer_model_pairs = []\n",
        "\n",
        "    for i in range(len(model_names)):\n",
        "        tokenizer_class, model_class = classPairs[i]\n",
        "        tokenizer = tokenizer_class.from_pretrained(model_names[i])\n",
        "        model = model_class.from_pretrained(model_names[i])\n",
        "        model.eval()\n",
        "        loaded_tokenizer_model_pairs.append((tokenizer, model))\n",
        "\n",
        "    return loaded_tokenizer_model_pairs\n",
        "\n",
        "# Define Transformers model and tokenizer classes and selected pre-trained model names\n",
        "tokenizer_model_class_pairs = [(DistilBertTokenizer, DistilBertModel), (RobertaTokenizer, RobertaModel)]\n",
        "embedding_model_names = ['distilbert-base-uncased', 'roberta-base']\n",
        "\n",
        "# Load pre-trained tokenizers and models\n",
        "loaded_tokenizer_model_class_pairs = load_tokenizer_model_instances(tokenizer_model_class_pairs, embedding_model_names)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HQJcZJNPx4-n"
      },
      "source": [
        "###### Create Tokenizing Datasets (TokenizerDataset) and Setup DataLoaders for Inferencing Embedding Models"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "wqUSkeoNV77o"
      },
      "outputs": [],
      "source": [
        "class TokenizerDataset(Dataset):\n",
        "    \"\"\"\n",
        "    A PyTorch Dataset that tokenizes text data and prepares model inputs along with corresponding labels.\n",
        "    Args:\n",
        "        df (pandas.DataFrame): DataFrame containing text and label data.\n",
        "        tokenizer (transformers.PreTrainedTokenizer): Tokenizer instance with a `from_pretrained` method.\n",
        "        text_column (str, optional): Name of the column in the DataFrame that contains text data. Defaults to 'cleaned_text'.\n",
        "    \"\"\"\n",
        "    def __init__(self, df, tokenizer, text_column='cleaned_text'):\n",
        "        self.df = df\n",
        "        self.tokenizer = tokenizer\n",
        "        self.text_col = text_column\n",
        "\n",
        "    def __len__(self):\n",
        "        return self.df.shape[0]\n",
        "\n",
        "    def __getitem__(self, idx):\n",
        "        \"\"\"\n",
        "        Retrieves tokenized data and labels for a given index.\n",
        "\n",
        "        This method extracts the text from the specified column in the DataFrame, tokenizes it with padding and truncation, and\n",
        "        converts the tokenized outputs to flattened tensors. It also extracts label data from a slice of the row, converts it\n",
        "        to a float tensor, and returns the index for reference.\n",
        "        \"\"\"\n",
        "        row = self.df.iloc[idx]\n",
        "        text = str(row[self.text_col])\n",
        "\n",
        "        # Tokenizers inherit from base class: https://huggingface.co/docs/transformers/v4.48.2/en/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__\n",
        "        encoding = self.tokenizer(\n",
        "            text,\n",
        "            max_length=128, # Max number of words to tokenize per text entry\n",
        "            padding='max_length', # Pad (with 0s) tokenized sequences of length < max_length\n",
        "            truncation=True, # Drop all tokenized words after max_length (if applicable)\n",
        "            return_tensors='pt' # Return tensors of type torch.Tensor instead of python integers\n",
        "        )\n",
        "\n",
        "        # Note: see 'Returns' section in tokenizer base class (linked above) for additional descriptions of data returned and accessed through 'encoding'\n",
        "        # 'flatten()' is used to ensure the output is a 1D sequence of integers; originally they are of shape (1, sequence_length) rather than just (sequence_length); 1 represents the a batch of one sample\n",
        "        item = {\n",
        "            'input_ids': encoding['input_ids'].flatten(), # List of integers, where each integer corresponds to a token produced by the tokenizer\n",
        "            'attention_mask': encoding['attention_mask'].flatten(), # List of integers that tells the model which tokens in the input_ids should be actively considered during processing (0 or 1)\n",
        "        }\n",
        "\n",
        "        if 'token_type_ids' in encoding:\n",
        "            item['token_type_ids'] = encoding['token_type_ids'].flatten() # List of integers used to indicate the segment of a sentence each token in the input_ids belongs to\n",
        "\n",
        "        labels_series = row[1:24].astype(np.int64)\n",
        "        item['labels'] = torch.tensor(labels_series.to_numpy(), dtype=torch.float32)\n",
        "        item['index'] = idx\n",
        "\n",
        "        return item\n",
        "\n",
        "def create_tokenizer_dataLoaders(tokenizer_model_pairs, device):\n",
        "    \"\"\"\n",
        "    Creates pairs train and test TokenizerDataset DataLoaders used for batch inferencing embedding models.\n",
        "\n",
        "    **Note:** This function expects that `train_df` and `test_df` are defined as pandas DataFrames\n",
        "    accessible in the global scope.\n",
        "\n",
        "    Args:\n",
        "        tokenizer_model_pairs (List[Tuple[tokenizer, model]]):\n",
        "            Pairs of tokenizer instances and model instances loaded with a pre-trained model\n",
        "        device (torch.device):\n",
        "            The device on which the model should be placed.\n",
        "\n",
        "    Returns:\n",
        "        List[Tuple[DataLoader, DataLoader]]:\n",
        "            A list of tuples, where each tuple contains a DataLoader for the training dataset a DataLoader for the testing dataset, set\n",
        "            to load tokenized text data.\n",
        "    \"\"\"\n",
        "    tokenizer_dataLoaders = []\n",
        "\n",
        "    for tokenizer, model in tokenizer_model_pairs:\n",
        "        model.to(device)\n",
        "        train_dataset = TokenizerDataset(train_df, tokenizer)\n",
        "        test_dataset  = TokenizerDataset(test_df, tokenizer)\n",
        "\n",
        "        # https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader\n",
        "        train_dataLoader = DataLoader(train_dataset, batch_size=16) # Shuffle data at every epoch while training\n",
        "        test_dataLoader  = DataLoader(test_dataset, batch_size=16)\n",
        "\n",
        "        tokenizer_dataLoaders.append((train_dataLoader, test_dataLoader))\n",
        "\n",
        "    return tokenizer_dataLoaders\n",
        "\n",
        "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
        "# Setup train and dataLoader pairs for each model\n",
        "tokenizer_DataLoader_pairs = create_tokenizer_dataLoaders(loaded_tokenizer_model_class_pairs, device)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HLxuJ-QiyOEu"
      },
      "source": [
        "###### Inference Embedding Models & Create Embedding Datasets (FeatureEmbeddingDataset)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "CQQ75V27VsCj"
      },
      "outputs": [],
      "source": [
        "class FeatureEmbeddingDataset(Dataset):\n",
        "    def __init__(self, embeddings, labels):\n",
        "        self.embeddings = embeddings\n",
        "        self.labels = labels\n",
        "\n",
        "    def __len__(self):\n",
        "        return self.embeddings.shape[0]\n",
        "\n",
        "    def __getitem__(self, idx):\n",
        "        return self.embeddings[idx], self.labels[idx], idx"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kjhLPhks7AbM",
        "outputId": "afd860ce-b347-4d3c-856e-199ed8ee1f23"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "Extracting feature embeddings from train split using distilbert-base-uncased: 100%|██████████| 9177/9177 [10:20<00:00, 14.79it/s]\n",
            "Extracting feature embeddings from test split using distilbert-base-uncased: 100%|██████████| 2298/2298 [02:28<00:00, 15.45it/s]\n",
            "Extracting feature embeddings from train split using roberta-base: 100%|██████████| 9177/9177 [16:58<00:00,  9.01it/s]\n",
            "Extracting feature embeddings from test split using roberta-base: 100%|██████████| 2298/2298 [04:05<00:00,  9.35it/s]\n"
          ]
        }
      ],
      "source": [
        "def extract_feature_embeddings(dataloader, model, device, model_name, data_split_name):\n",
        "    \"\"\"\n",
        "    Extracts feature embeddings and labels from a DataLoader using a given model. The extracted embeddings and labels from\n",
        "    each batch are concatenated and wrapped into a FeatureEmbeddingDataset.\n",
        "\n",
        "    Args:\n",
        "        dataloader (torch.utils.data.DataLoader):\n",
        "            DataLoader providing batches of tokenized inputs and labels.\n",
        "        model (torch.nn.Module):\n",
        "            The loaded model to use to compute the embeddings. The model should output either a\n",
        "            `pooler_output` or a `last_hidden_state` attribute.\n",
        "        device (torch.device):\n",
        "            The device (CPU or GPU) on which the model inference should run.\n",
        "        model_name (str):\n",
        "            A string identifier for the model. Used for descriptive progress output.\n",
        "        data_split_name (str):\n",
        "            A string indicating the data split (e.g., 'train', 'test') for progress display.\n",
        "\n",
        "    Returns:\n",
        "        FeatureEmbeddingDataset:\n",
        "            A dataset containing the concatenated feature embeddings and corresponding labels.\n",
        "    \"\"\"\n",
        "    embedding_batches = []\n",
        "    label_batches = []\n",
        "\n",
        "    # inference_mode() helps model get better performance by disabling training layers in model; https://pytorch.org/docs/stable/generated/torch.autograd.grad_mode.inference_mode.html\n",
        "    with torch.inference_mode():\n",
        "        for batch in tqdm(dataloader, desc=f\"Extracting feature embeddings from {data_split_name} split using {model_name}\"):\n",
        "            input_ids = batch['input_ids'].to(device)\n",
        "            attention_mask = batch['attention_mask'].to(device)\n",
        "\n",
        "            outputs = model(input_ids=input_ids, attention_mask=attention_mask)\n",
        "\n",
        "            # Check if model includes 'pooler_output' (designed to summarize the entire input sequence)\n",
        "            if hasattr(outputs, 'pooler_output') and outputs.pooler_output is not None:\n",
        "                embeddings = outputs.pooler_output\n",
        "            else:\n",
        "                embeddings = outputs.last_hidden_state[:, 0, :] # 'outputs' is of shape (batch_size, sequence_length, hidden_size); here we select all samples in the batch, the first dimension (i.e., the [CLS] token that represents the overall sequence), all features for that token\n",
        "\n",
        "            embedding_batches.append(embeddings.cpu())\n",
        "            label_batches.append(batch['labels'])\n",
        "\n",
        "        # Combines separate tensors created from each batch into a single tensor that represents the embeddings / labels for the entire dataset\n",
        "        embeddings = torch.cat(embedding_batches, dim=0)\n",
        "        labels = torch.cat(label_batches, dim=0)\n",
        "\n",
        "        return FeatureEmbeddingDataset(embeddings, labels)\n",
        "\n",
        "def create_feature_embedding_datasets(tokenizer_dataLoader_pairs, loaded_tokenizer_model_class_pairs, device):\n",
        "    \"\"\"\n",
        "    Creates feature embedding datasets for both training and testing splits for each tokenizer/model pair.\n",
        "\n",
        "    Args:\n",
        "        tokenizer_dataLoader_pairs (List[Tuple[DataLoader, DataLoader]]):\n",
        "            A list of tuples, each containing a training DataLoader and a testing DataLoader for loading tokenized data.\n",
        "        loaded_tokenizer_model_class_pairs (List[Tuple[tokenizer, model]]):\n",
        "            A list of tuples where each tuple consists of a tokenizer instance and a model instance. The model instance is used to extract feature\n",
        "            embeddings.\n",
        "        device (torch.device):\n",
        "            The device on which model inference should run (e.g., CPU or GPU).\n",
        "\n",
        "    Returns:\n",
        "        List[Tuple[FeatureEmbeddingDataset, FeatureEmbeddingDataset]]:\n",
        "            A list of tuples, where each tuple contains a FeatureEmbeddingDataset for the training split and a FeatureEmbeddingDataset for the\n",
        "            testing split.\n",
        "    \"\"\"\n",
        "    feature_embedding_dataset_pairs = []\n",
        "\n",
        "    for i, pair in enumerate(tokenizer_dataLoader_pairs):\n",
        "        train_dataset  = extract_feature_embeddings(pair[0], loaded_tokenizer_model_class_pairs[i][1], device, embedding_model_names[i], 'train')\n",
        "        test_dataset = extract_feature_embeddings(pair[1], loaded_tokenizer_model_class_pairs[i][1], device, embedding_model_names[i], 'test')\n",
        "\n",
        "        feature_embedding_dataset_pairs.append((train_dataset, test_dataset))\n",
        "\n",
        "    return feature_embedding_dataset_pairs\n",
        "\n",
        "# Create pairs of train and test PyTorch datasets which contain feature embedding representations of the input data (1 pair per embedding model)\n",
        "feature_embedding_dataset_pairs = create_feature_embedding_datasets(tokenizer_DataLoader_pairs, loaded_tokenizer_model_class_pairs, device)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "b6XPXLa9FJc5"
      },
      "source": [
        "###### Save Embedding Datasets to Google Drive"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "2xz904c0uhlt"
      },
      "outputs": [],
      "source": [
        "def prepare_folder(folder_path, clear_folder=True):\n",
        "    if os.path.exists(folder_path):\n",
        "        if clear_folder:\n",
        "            shutil.rmtree(folder_path)\n",
        "            os.makedirs(folder_path)\n",
        "    else:\n",
        "        os.makedirs(folder_path)\n",
        "\n",
        "def save_dataset_pairs(dataset_pairs, model_names, dataset_name):\n",
        "    base_path = '/content/drive/MyDrive/multiLabelClassification'\n",
        "\n",
        "    for i, (train_dataset, test_dataset) in enumerate(dataset_pairs):\n",
        "        folder_path = os.path.join(base_path, model_names[i])\n",
        "        prepare_folder(folder_path, clear_folder=True)\n",
        "\n",
        "        torch.save(train_dataset, os.path.join(folder_path, f'train_{dataset_name}.pt'))\n",
        "        torch.save(test_dataset, os.path.join(folder_path, f'test_{dataset_name}.pt'))\n",
        "\n",
        "save_dataset_pairs(feature_embedding_dataset_pairs, embedding_model_names, dataset_name='feature_embeddings_dataset')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HffH_qpjHqZe"
      },
      "source": [
        "###### Load Embedding Datasets & Setup DataLoaders for Training + Evaluating Models"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rNSMJ6NmGfCc",
        "outputId": "30ff9ca4-f05b-41a4-bd6c-88999fb10b0f"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "<ipython-input-23-3b779d00896c>:6: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.\n",
            "  train_dataset = torch.load(f'/content/drive/MyDrive/multiLabelClassification/{name}/train_{dataset_name}.pt')\n",
            "<ipython-input-23-3b779d00896c>:7: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.\n",
            "  test_dataset = torch.load(f'/content/drive/MyDrive/multiLabelClassification/{name}/test_{dataset_name}.pt')\n"
          ]
        }
      ],
      "source": [
        "def load_dataset_pairs(model_names, dataset_name):\n",
        "    embedding_dataset_pairs = []\n",
        "\n",
        "    for name in model_names:\n",
        "        # Note: need to have 'FeatureEmbeddingDataset' defined to load datasets\n",
        "        train_dataset = torch.load(f'/content/drive/MyDrive/multiLabelClassification/{name}/train_{dataset_name}.pt')\n",
        "        test_dataset = torch.load(f'/content/drive/MyDrive/multiLabelClassification/{name}/test_{dataset_name}.pt')\n",
        "        embedding_dataset_pairs.append((train_dataset, test_dataset))\n",
        "\n",
        "    return embedding_dataset_pairs\n",
        "\n",
        "def create_dataLoaders(model_names):\n",
        "    embedding_dataset_pairs = load_dataset_pairs(model_names, dataset_name='feature_embeddings_dataset')\n",
        "\n",
        "    embedding_dataLoader_pairs = []\n",
        "\n",
        "    for train_dataset, test_dataset in embedding_dataset_pairs:\n",
        "        train_dataLoader = DataLoader(train_dataset, batch_size=256, shuffle=True) # shuffle=True since we are using this dataLoader for training\n",
        "        test_dataLoader = DataLoader(test_dataset, batch_size=256)\n",
        "\n",
        "        embedding_dataLoader_pairs.append((train_dataLoader, test_dataLoader))\n",
        "\n",
        "    return embedding_dataLoader_pairs\n",
        "\n",
        "pretrained_model_names = ['distilbert-base-uncased', 'roberta-base']\n",
        "\n",
        "# Loads pairs of DataLoaders where each pair contains a train and test dataLoader, which load batches of data with feature embeddings as input values\n",
        "embedding_dataLoader_pairs = create_dataLoaders(pretrained_model_names)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Ofi95mM-yWmq"
      },
      "source": [
        "##### Train \"Simple\" Classifiers using Embeddings"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uLqU74wZwjiY",
        "outputId": "af48168b-9372-4b6b-8075-059aed444e12"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "Using distilbert-base-uncased's embeddings to train binary classifiers using log_loss (loss function): 100%|██████████| 23/23 [00:58<00:00,  2.56s/it]\n",
            "Using distilbert-base-uncased's embeddings to train binary classifiers using hinge (loss function): 100%|██████████| 23/23 [01:06<00:00,  2.89s/it]\n",
            "Using distilbert-base-uncased's embeddings to train binary classifiers using modified_huber (loss function): 100%|██████████| 23/23 [00:52<00:00,  2.27s/it]\n",
            "Using distilbert-base-uncased's embeddings to train binary classifiers using squared_hinge (loss function): 100%|██████████| 23/23 [00:51<00:00,  2.24s/it]\n",
            "Using distilbert-base-uncased's embeddings to train binary classifiers using perceptron (loss function): 100%|██████████| 23/23 [01:12<00:00,  3.14s/it]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "--------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "Using roberta-base's embeddings to train binary classifiers using log_loss (loss function): 100%|██████████| 23/23 [01:18<00:00,  3.43s/it]\n",
            "Using roberta-base's embeddings to train binary classifiers using hinge (loss function): 100%|██████████| 23/23 [01:10<00:00,  3.06s/it]\n",
            "Using roberta-base's embeddings to train binary classifiers using modified_huber (loss function): 100%|██████████| 23/23 [00:59<00:00,  2.58s/it]\n",
            "Using roberta-base's embeddings to train binary classifiers using squared_hinge (loss function): 100%|██████████| 23/23 [00:59<00:00,  2.57s/it]\n",
            "Using roberta-base's embeddings to train binary classifiers using perceptron (loss function): 100%|██████████| 23/23 [00:51<00:00,  2.24s/it]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "--------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        }
      ],
      "source": [
        "def train_clfs(embedding_model_names, dataLoader_pairs, num_labels, clf_loss_fns):\n",
        "    \"\"\"\n",
        "    Trains binary relevance classifiers using embeddings produced by each embedding model.\n",
        "\n",
        "    Args:\n",
        "        embedding_model_names (list):\n",
        "            Names of the pre-trained embedding models used for creating embeddings.\n",
        "        dataLoader_pairs (list):\n",
        "            Pairs of dataLoaders (train_dataLoader, test_dataLoader) for each embedding model.\n",
        "        num_labels (int):\n",
        "            Number of target labels in multi-label classificaiton task.\n",
        "        clf_loss_fns (list):\n",
        "            List of loss functions to use for the SGDClassifier.\n",
        "\n",
        "    Returns:\n",
        "        trained_clfs:\n",
        "            A nested dictionary where: trained_clfs[embedding_model_name][clf_loss_fn][label_index] = trained binary classifier.\n",
        "            trained_clfs[embedding_model_name][clf_loss_fn] = set of binary classifiers to be used for multi-label task.\n",
        "    \"\"\"\n",
        "    # Holds all trained classifiers\n",
        "    trained_clfs = {}\n",
        "\n",
        "    # Retrieve train dataLoader for each embedding model\n",
        "    for i, embedding_model_name in enumerate(embedding_model_names):\n",
        "        train_dataLoader, _ = dataLoader_pairs[i]\n",
        "        trained_clfs[embedding_model_name] = {}\n",
        "\n",
        "        # Train a classifier (using binary relevance) per loss function\n",
        "        for clf_loss_fn in clf_loss_fns:\n",
        "            trained_clfs[embedding_model_name][clf_loss_fn] = []\n",
        "            # Train one binary classifier per label and add to set\n",
        "            for i in tqdm(range(num_labels), desc=f\"Using {embedding_model_name}'s embeddings to train binary classifiers using {clf_loss_fn} (loss function)\"):\n",
        "                clf = SGDClassifier(loss=clf_loss_fn)\n",
        "                # Train the classifier with mini-batch updates via partial_fit\n",
        "                for embeddings_batch, labels_batch, _ in train_dataLoader:\n",
        "                    # Convert tensors to NumPy arrays to ensure compatibility with partial_fit\n",
        "                    X_batch = embeddings_batch.numpy()\n",
        "                    y_batch = labels_batch.numpy()[:, i]\n",
        "\n",
        "                    # Initialize binary classifier classes if not already set (only need to set first time)\n",
        "                    if not hasattr(clf, 'classes_'):\n",
        "                        clf.partial_fit(X_batch, y_batch, classes=[0, 1])\n",
        "                    else:\n",
        "                        clf.partial_fit(X_batch, y_batch)\n",
        "                trained_clfs[embedding_model_name][clf_loss_fn].append(clf) # Add trained binary classifier to set of classifiers for current embedding model and loss function\n",
        "\n",
        "        print(\"-\" * 50)\n",
        "\n",
        "    return trained_clfs\n",
        "\n",
        "embedding_model_names = ['distilbert-base-uncased', 'roberta-base']\n",
        "SGDClassifier_loss_functions = ['log_loss', 'hinge', 'modified_huber', 'squared_hinge', 'perceptron']\n",
        "num_labels = train_df.shape[1] - 2 # -2 to account for text columns\n",
        "\n",
        "# Train every classifier type (using binary relevance) using all model ebeddings as training data\n",
        "trained_clfs = train_clfs(embedding_model_names, embedding_dataLoader_pairs, num_labels, SGDClassifier_loss_functions)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HLvQBcMgU8VT"
      },
      "source": [
        "###### Evaluate \"Simple\" Classifiers"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NigfnZygyuEV",
        "outputId": "c9640039-b1ac-4dbe-87f4-6461f86dc3ca"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "Evaluating classifiers: 100%|██████████| 10/10 [03:24<00:00, 20.44s/it]\n"
          ]
        }
      ],
      "source": [
        "def get_binary_clfs_preds(binary_clfs, dataLoader):\n",
        "    \"\"\"\n",
        "    Obtains predictions from each binary classifiers over all batches of data.\n",
        "    **Note:** Since the dataLoader may shuffle samples, the function uses the original indices to sort the predictions back to the\n",
        "    original order of the dataset.\n",
        "\n",
        "    Args:\n",
        "        binary_clfs (List of 'SGDClassifier' objects):\n",
        "            A list of binary classifier objects.\n",
        "        dataLoader (PyTorch dataLoader of dataset type FeatureEmbeddingDataset):\n",
        "            Yields batches of embedding data in the form: (embeddings_batch, labels_batch, original_indices_batch).\n",
        "\n",
        "    Returns:\n",
        "        binary_clfs_preds (dict):\n",
        "            A dictionary where: binary_clfs_preds[i] = NumPy array containing the predictions for binary classifier i, sorted in the same order as the original dataset (think of it as a column of predictions for a label)\n",
        "    \"\"\"\n",
        "    # Holds all predictions for each binary classifier\n",
        "    binary_clfs_preds = {i: [] for i in range(len(binary_clfs))}\n",
        "    # Used to track original indicies of samples (needed when loading data from train_dataLoader since it shuffles each full batch of data)\n",
        "    original_indices = []\n",
        "\n",
        "    # Iterate over all batches of data provided by dataLoader\n",
        "    for embeddings_batch, labels_batch, original_indices_batch in dataLoader:\n",
        "        # .numpy() converts PyTorch tensor to a NumPy array\n",
        "        original_indices.append(original_indices_batch.numpy())\n",
        "        X_batch = embeddings_batch.numpy()\n",
        "\n",
        "        # Retrieve and save predictions made by each binary classifier\n",
        "        for i, clf in enumerate(binary_clfs):\n",
        "            preds = clf.predict(X_batch)\n",
        "            binary_clfs_preds[i].append(preds)\n",
        "\n",
        "    # Combine each batch of predictions into a single NumPy array and add save to 'binary_clfs_preds'\n",
        "    for i in binary_clfs_preds:\n",
        "        binary_clfs_preds[i] = np.concatenate(binary_clfs_preds[i])\n",
        "\n",
        "    original_indices = np.concatenate(original_indices) # Combine each batch of indicies into single NumPy array\n",
        "    sorted_idx = np.argsort(original_indices) # Produces mapping that reorders the shuffled indices back to their original order (e.g., concatenated indices were [3, 1, 2], then np.argsort would return [1, 2, 0] (index 1 (which is 1) should come first, then index 2 (which is 2), followed by index 0 (which is 3))\n",
        "\n",
        "    # Reorder predictions to ensure that each prediction lines up with its corresponding sample from the original dataset\n",
        "    for i in binary_clfs_preds:\n",
        "        binary_clfs_preds[i] = binary_clfs_preds[i][sorted_idx]\n",
        "\n",
        "    return binary_clfs_preds\n",
        "\n",
        "def evaluate_clfs(trained_clfs, embedding_model_names, dataLoader_pairs, y_train, y_test):\n",
        "    \"\"\"\n",
        "    Args:\n",
        "        embedding_model_names (List of 'SGDClassifier' objects):\n",
        "            A list of binary classifier objects.\n",
        "        dataLoader_pairs (PyTorch dataLoader of dataset type FeatureEmbeddingDataset):\n",
        "            Yields batches of embedding data in the form: (embeddings_batch, labels_batch, original_indices_batch).\n",
        "    Returns:\n",
        "    eval_results:\n",
        "        A nested dictionary where:\n",
        "        eval_results[embedding_model_name][loss] =  {\n",
        "            \"train\": train_results,\n",
        "            \"test\": test_results\n",
        "        }\n",
        "\n",
        "        train_results & test_results = {\n",
        "            \"Multi-label\": {\n",
        "                \"Accuracy\": [float],\n",
        "                \"Precision\": [float],\n",
        "                \"Recall\": [float],\n",
        "                \"F1-Score\": [float]\n",
        "            }\n",
        "        }\n",
        "    \"\"\"\n",
        "    # Holds all evaluation results\n",
        "    eval_results = {}\n",
        "\n",
        "    # Calculate total number clfs to be evaluated\n",
        "    total_steps = len(embedding_model_names) * len(SGDClassifier_loss_functions)\n",
        "\n",
        "    with tqdm(total=total_steps, desc=\"Evaluating classifiers\") as pbar:\n",
        "        # Retrieve train and test embedding dataLoaders for each embedding model\n",
        "        for i, embedding_model_name in enumerate(embedding_model_names):\n",
        "            train_dataLoader, test_dataLoader = dataLoader_pairs[i]\n",
        "\n",
        "            eval_results[embedding_model_name] = {}\n",
        "\n",
        "            # For each trained classifier, get its predictions on train and test sets\n",
        "            for loss_fn in trained_clfs[embedding_model_name]:\n",
        "                binary_clfs = trained_clfs[embedding_model_name][loss_fn] # Set of binary classifiers\n",
        "\n",
        "                # Retrieve predictions of each binary classifier for both data splits\n",
        "                binary_clf_preds_train_dict = get_binary_clfs_preds(binary_clfs, train_dataLoader)\n",
        "                binary_clf_preds_test_dict  = get_binary_clfs_preds(binary_clfs, test_dataLoader)\n",
        "\n",
        "                # Convert dicitonary (containing predictions (value) of each binary classifier i (key)) to pandas df for evaluation\n",
        "                binary_clf_preds_train_df = pd.DataFrame(binary_clf_preds_train_dict)\n",
        "                binary_clf_preds_test_df  = pd.DataFrame(binary_clf_preds_test_dict)\n",
        "\n",
        "                # Evaluate binary relevance classifiers' predictions\n",
        "                train_results = eval_clf_preds(\n",
        "                    y_train,\n",
        "                    binary_clf_preds_train_df,\n",
        "                )\n",
        "                test_results = eval_clf_preds(\n",
        "                    y_test,\n",
        "                    binary_clf_preds_test_df,\n",
        "                )\n",
        "\n",
        "                # Save evaluations for clf model on multi-label task\n",
        "                eval_results[embedding_model_name][loss_fn] = {\n",
        "                    \"train\": train_results,\n",
        "                    \"test\": test_results\n",
        "                }\n",
        "\n",
        "                pbar.update(1)\n",
        "\n",
        "    return eval_results\n",
        "\n",
        "# Collect ground truth values from original dataframes\n",
        "y_train = train_df.iloc[ : ,1:24]\n",
        "y_test = test_df.iloc[ : ,1:24]\n",
        "\n",
        "# Retrieve evaluations of each binary relevance classifier on train and tests data splits\n",
        "eval_results = evaluate_clfs(trained_clfs, embedding_model_names, embedding_dataLoader_pairs, y_train, y_test)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Zq6PDAulALNy"
      },
      "source": [
        "###### Visualize Results"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "zT2OT3mBAKyw",
        "outputId": "22600dd8-2855-451a-f50e-3845d1c91793"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Evaluation Results:\n",
            "+-------------------------+---------------------+--------------+------------+-------------+-----------+------------+\n",
            "| Embedding Model         | Clf Loss Function   | Data Split   |   Accuracy |   Precision |    Recall |   F1-Score |\n",
            "|-------------------------+---------------------+--------------+------------+-------------+-----------+------------|\n",
            "| distilbert-base-uncased | log_loss            | train        | 0.103991   |   0.43098   | 0.26975   |  0.331816  |\n",
            "| distilbert-base-uncased | log_loss            | test         | 0.101567   |   0.427043  | 0.266539  |  0.328219  |\n",
            "| distilbert-base-uncased | hinge               | train        | 0.164701   |   0.211447  | 0.437283  |  0.285056  |\n",
            "| distilbert-base-uncased | hinge               | test         | 0.162921   |   0.20718   | 0.430581  |  0.279753  |\n",
            "| distilbert-base-uncased | modified_huber      | train        | 0.103228   |   0.139235  | 0.44184   |  0.211744  |\n",
            "| distilbert-base-uncased | modified_huber      | test         | 0.102465   |   0.136519  | 0.431618  |  0.207428  |\n",
            "| distilbert-base-uncased | squared_hinge       | train        | 0.212709   |   0.300025  | 0.399729  |  0.342774  |\n",
            "| distilbert-base-uncased | squared_hinge       | test         | 0.208739   |   0.294264  | 0.391641  |  0.33604   |\n",
            "| distilbert-base-uncased | perceptron          | train        | 0.243356   |   0.350878  | 0.387235  |  0.368161  |\n",
            "| distilbert-base-uncased | perceptron          | test         | 0.23913    |   0.345946  | 0.38107   |  0.36266   |\n",
            "| roberta-base            | log_loss            | train        | 0.0459239  |   0.0471551 | 0.668119  |  0.0880928 |\n",
            "| roberta-base            | log_loss            | test         | 0.0466616  |   0.0473993 | 0.653081  |  0.0883839 |\n",
            "| roberta-base            | hinge               | train        | 0.108043   |   0.102432  | 0.597344  |  0.174876  |\n",
            "| roberta-base            | hinge               | test         | 0.104941   |   0.100052  | 0.585677  |  0.170908  |\n",
            "| roberta-base            | modified_huber      | train        | 0.0505687  |   0.455374  | 0.213265  |  0.290487  |\n",
            "| roberta-base            | modified_huber      | test         | 0.0508516  |   0.453334  | 0.212212  |  0.289095  |\n",
            "| roberta-base            | squared_hinge       | train        | 0.00361643 |   0.0633802 | 0.0647753 |  0.0640702 |\n",
            "| roberta-base            | squared_hinge       | test         | 0.00380911 |   0.0649581 | 0.0660761 |  0.0655123 |\n",
            "| roberta-base            | perceptron          | train        | 0.00683103 |   0.198209  | 0.101222  |  0.134008  |\n",
            "| roberta-base            | perceptron          | test         | 0.00625782 |   0.197643  | 0.100786  |  0.133496  |\n",
            "+-------------------------+---------------------+--------------+------------+-------------+-----------+------------+\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 2400x600 with 4 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAACVoAAAJSCAYAAAA4KezDAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjAsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvlHJYcgAAAAlwSFlzAAAPYQAAD2EBqD+naQAAvpdJREFUeJzs3Xd8jef/x/F3EgkJkkhEEKKiGiuxR8xIbWrX3rM1a5ToSGu01K7QUrVqVVujKDF/pbU6bEXtFTNkECHj/P7wyPk6Mk4OkaCv5+Ph0Z77vu7r+ty3k+R2532uy8pgMBgEAAAAAAAAAAAAAAAAAEiRdWYXAAAAAAAAAAAAAAAAAAAvOoJWAAAAAAAAAAAAAAAAAGAGQSsAAAAAAAAAAAAAAAAAMIOgFQAAAAAAAAAAAAAAAACYQdAKAAAAAAAAAAAAAAAAAMwgaAUAAAAAAAAAAAAAAAAAZhC0AgAAAAAAAAAAAAAAAAAzCFoBAAAAAAAAAAAAAAAAgBlZMrsAAAAAAAAAAAAAAAAAvFri4+MVGxub2WUAqbK1tZWNjU2a2xO0AgAAAAAAAAAAAAAAQLowGAy6du2awsPDM7sUIE2cnZ2VN29eWVlZmW1L0AoAAAAAAAAAAAAAAADpIjFklSdPHjk4OKQpvAJkBoPBoOjoaN24cUOSlC9fPrPHELQCAAAAAAAAAAAAAADAM4uPjzeGrFxdXTO7HMAse3t7SdKNGzeUJ08es8sIWmdEUQAAAAAAAAAAAAAAAHi1xcbGSpIcHBwyuRIg7RLfr4nv39QQtAIAAAAA4AUXEBAgb29vkz+lSpWSv7+/3nvvPf31118ZWk9kZKRGjx6t2rVrq1SpUvL29lbnzp0ztAY8vcDAQOP7qFmzZqm2PXz4sMn7LqPfa+Ykfm1cvnw5s0sxcfr0afXr109+fn4qXry4vL29FRwcnNllvTLOnz+vMWPGqFGjRipTpox8fHxUs2ZNtWrVSmPGjNGmTZvSbazEr5dVq1aZbF+1apW8vb0VGBiYbmMBAAAAwKuE5QLxMrHk/UrQCgAAAACAl0S5cuXUokULtWjRQjVr1lRCQoI2btyoTp06acGCBRlWx8cff6xly5bJyspKdevWVYsWLVSjRo0MGx/p58SJEzp69GiK+3/66afnMu6LGpBKD9HR0erTp4+2bdum/Pnzq0mTJmrRooWKFy+eqXV17txZ3t7e2rdv30tdx+bNm/XWW29p6dKlCgsLU7ly5VSvXj15e3vr+vXrWrp0qYKCgtK56rS7fPmyvL29FRAQkGk1AAAAAACA5ydLZhcAAAAAAADS5u2331bLli2Nrx88eKCgoCCtWbNGkyZNkr+/vwoXLvxca4iNjdXWrVuVNWtWrV27Vjly5Hiu4+H5KVWqlI4ePaqVK1eqVKlSSfbHxMRow4YNcnNzk42Nja5du5YJVaZu4cKFio2Nlbu7e2aXYnTkyBFduXJFZcuW1ffff5/Z5bxSbt26pZEjR+rhw4fq0aOH3nvvPWXNmtWkzdGjR9N1RquU1K1bV6VLl1bOnDmf+1gAAAAAgMwRGBioo0ePav369f+JcZE2zGgFAAAAAMBLKmvWrAoKCpKDg4Pi4+O1ZcuW5z7mzZs3FRcXp9y5cxOyesn5+/srd+7c+uWXX/TgwYMk+0NCQhQVFaXmzZvLxsYmEyo0z9PTU0WKFJGtrW1ml2J09epVSdJrr72WuYW8gv7v//5P0dHRypMnj0aOHJkkZCU9ChAOGzbsudeSM2dOFSlSRHny5HnuYwEAAADAqyIhwfCfGhevJma0AgAAAADgJZY9e3YVLlxYx44dS7IM27lz57RgwQLt3r1b169fl52dnYoVK6Y2bdqoWbNmSfrq3Lmz/vjjD3333XeysbHRt99+q4MHDyo8PFyff/65Ro0aZWx75coVeXt7G19/9913qly5siQpLi5OP/74o37++WedOnVKDx8+VL58+VSzZk317t072dmHEvs6efKkVq5cqR9++EGnT5/W3bt3tW3bNknSm2++KQ8PD23dulVLlizRjz/+qIsXLypnzpwKCAjQ0KFD5ezsrIcPH2ru3Llat26dQkND5ejoqAYNGmjo0KFycHAwGffu3bvasGGDdu7cqX///Vc3btyQJBUsWFABAQHq2bOnHB0dk9QbEBCgK1euaNu2bbp8+bK++eYbHTlyRA8ePFCRIkXUtWtXNW/ePNm/M4PBoC1btmjlypU6evSoIiIi5OTkJE9PT9WuXVtdunRRtmzZTI45evSoFixYoL///lu3bt2Sg4ODfHx81KVLF9WqVSvZccyxsbFR06ZNNX/+fG3ZskVNmjQx2b9y5UpJUqtWrbRhw4ZU+9qzZ4+WLl1qfL/kzJlT5cqVU69evVS2bFlju1WrVpm8j958802TfhLfR/v27VOXLl1UqVIlffPNN/rmm28UEhKi0NBQubq6avv27ZJM/x4KFChg0pcl1zkhIUE//vijVq9erdOnT+v+/ftydHSUm5ubKlasqO7duyfp/0mJNSdavXq1Vq9ebXx98uRJ4/+Hh4dr/vz5xvePtbW1ChcurIYNG6pz585J/v7Tej3SUtfj/y9J48ePN5kpLyIiQosWLdK2bdt08eJFJSQkyNPTUw0bNlT37t1lb29vcnxar52ldSQnLCxMkuTi4pJqu+Q8/l45fvy45s+fr5MnT8pgMKhUqVLq1auXRV9Lie/lFi1aaMKECZIefeI48e/8ye+Rkul7AAAAAAD+i6ytrTRr+S5duRGRYWN65HFS//bVMmw8vPoIWgEAAAAA8JK7e/euJMnOzs64bePGjRo5cqQePHggLy8v1apVS1FRUTp8+LBGjBihvXv3avz48cn2FxISou+//15eXl6qWrWqIiIiZGdnpxYtWig6OlqbNm2Sg4OD6tevbzwmd+7ckqSHDx+qb9++2r17t7JmzarKlSsrR44cOnDggBYvXqz169dr3rx5KlmyZLJjjx07VsuWLVPZsmXl7++vS5cuycrKSgbD/z55+P7772vr1q2qWLGiPD09deDAAa1YsUJHjhzR0qVL1atXL508eVKVKlXSa6+9pr/++kuLFy/WhQsXNHfuXJPxTpw4oY8//lguLi4qXLiwSpYsqcjISB09elSzZ8/Wxo0btWLFCuXKlSvZeleuXKmvv/5aJUqUUI0aNXTlyhUdPHhQI0eOVHh4uLp162bSPjY2VkOHDtXmzZtlbW0tX19fValSRXfu3NHp06c1ZcoUNWrUyCTUs2jRIk2YMEEJCQkqXry4fH19devWLe3bt0+///67Bg4cqAEDBiRbnzmtW7fW/PnztXLlSpOg1cWLF/Xnn3+qXLlyZpej/OKLLzR//nxZW1urVKlSKl++vK5evapt27bp//7v/zR27Fi1atVK0qMZqFq0aKFNmzYpOjpa9evXNwm/Jb6PEj148ECdO3fWmTNnVKFCBRUrVkzh4eFmz8vS6/zhhx9q1apVypo1q8qXLy8XFxeFh4fr8uXLWrJkifz8/MwGrXLnzq0WLVrowoUL2r9/vzw9PVW+fPkk7S5duqSuXbvqypUrcnFxUa1atRQbG6t9+/Zp8uTJ2rhxoxYsWCAnJ6ckxz7N9Uis67ffftOtW7dUvXp1ubm5Gfd7enoa///06dPq1auXrl69Kjc3N5UvX15ZsmTRkSNH9OWXX2rz5s1avHixyXJ5ab12ltSRknz58kmSTp06pT179sjPz8/sMU9avHixFi5cqFKlSql27dq6ePGi/vjjD/3xxx/66KOP1LlzZ4v7TFS+fPkUv0cCAAAAAB65ciNC56/cyewynsrJkyc1ceJE/f3337KxsVG1atUUGBio/PnzG9tERUVp9OjR2rZtm7Jly6a3335bzs7O+uKLL57pAzhpGfunn37SggULdOnSJdnb28vLy0ujRo2Sr69vmvYjbQhaAQAAAADwEjtx4oQuXbokSSpevLikRw9eRowYISsrKwUHB6tevXrG9leuXNE777yjVatWqXLlysnOurRs2TIFBQWpY8eOJtubNGmiy5cva9OmTcqVK5dxFpfHzZgxQ7t375anp6cWLFhgDKfExsbq008/1U8//aRBgwZp48aNJsGwRGvWrNHy5ctVpkwZk+2Js3VduXJFNjY22rBhgzw8PCRJd+7cUbt27fTPP/+obdu2ypYtm7Zu3WoMR126dEmtWrXSzp079ffff5uEXwoUKKCFCxeqcuXKsra2Nm6/f/++Pv30U61Zs0YzZszQJ598kuz1nzt3rr7++mvVrl3buC1xppuZM2eqXbt2JrMTTZkyRZs3b5aHh4e++uorFStWzLjPYDBo7969JgGb3377TePHj5ezs7OCg4NVsWJF476TJ0+qT58+Cg4OVqVKlVSpUqVka0xNkSJFVLZsWe3du1ehoaHGh3MrV66UwWBQ69atUz3+hx9+0Pz581WoUCHNmDHD5Hz+/PNP9e3bV5988onKly+v1157TRUqVFCFChX0xx9/KDo6WiNGjEg1wHTo0CF5e3tr8+bNJqEccyy5zqGhoVq1apXy5s2rn376Kck4Z86cSTKLU3KKFCmiCRMmaNWqVdq/f7/Kly+f7NfIsGHDdOXKFQUEBGjKlCnGoNnt27fVq1cvHTt2TGPGjNGUKVPS5Xok1tW5c2fdunVLffr0Mc4+97iYmBi9++67unr1qt59913169fP+DV6//59ffTRR1q/fr0+//xzY0jTkmuX1jpSU6dOHbm7u+v69evq3r27KlasKD8/P5UsWVI+Pj5pmulq0aJFmjRpkpo2bWrctmHDBg0dOlQTJkxQ5cqV9cYbb1hUV6K3335bfn5+qX6PBAAAAAC8nK5evapOnTqpYMGCmjRpkh48eKBp06apU6dOWrt2rXLkyCFJGjVqlPbu3av3339fHh4e+uGHH3Ts2LHnPvaff/6pDz/8UD169FCtWrUUExOjw4cPKyoqSpLM7kfaWZtvAgAAAAAAXjRRUVHasWOHBg4cqISEBOXJk0cNGzaUJM2ePVsPHz7Ue++9ZxKykiQPDw999tlnkh4t05acKlWqJAlZpcWDBw+0dOlSSY8eKj0eoLG1tdVHH32k3LlzG8NayenRo0eSkNWTPvroI2PISpJy5cql9u3bS3o0081nn31mMgNVwYIFjaGKPXv2mPSVN29e+fn5mYSsJMne3l6ffvqpsmTJopCQkBRr6dSpk0nISpJatmwpLy8vRUVF6ejRo8btYWFhWrJkiSQlCSVJkpWVlfz8/ExmCwoODpbBYNDo0aNNQlbSo+UWAwMDJcnY79No1aqVEhIStGrVKkmPloJbs2aNHBwcjO+p5CQkJCg4OFiSNHXq1CTnU7FiRfXr10+xsbFasWLFU9cXFBRkUcjK0ut869YtSVKJEiWSHadIkSImnw59Fn/99ZcOHToke3t7jR071mQ2LxcXF40ZM0bSo+DPtWvXku3D0uuRVqtXr9bFixdVu3ZtvffeeyZBSHt7e40ZM0aurq5au3atIiIeLfGQkddOerRU6sKFC1W6dGkZDAb98ccf+vLLL9WnTx/5+fmpefPmWr58ueLj41Ps48033zQJWUlSo0aNVK9ePcXFxWnx4sXpVi8AAAAA4NWxcOFCxcXFaf78+apbt66aNGmib775RqGhocZl5E+fPq0tW7boo48+UocOHVSrVi0FBwcra9asz33sw4cPy9nZWSNHjlSVKlXk7++vQYMGqVq1amnaj7QjaAUAAAAAwEti1KhR8vb2lre3typUqKA+ffro4sWL8vT01Ny5c+Xg4KCEhATt3LlT0qPwQHJ8fHzk4OCg48eP68GDB0n2P+1yV0eOHFF0dLScnZ0VEBCQZL+9vb2xpn379iXbR4MGDVIdI0uWLMk+ACpUqJAkKX/+/MnORpO4/8aNG8n2u3//fn3zzTcaPXq0Ro0apcDAQI0ePVq2tra6ffu2MVjypCdDVomKFCkiSbp+/bpx2759+xQbG6uSJUuqVKlSqZzlI7dv39bhw4eVLVu2FMdJnBFo//79ZvtLScOGDeXg4KBVq1bJYDDot99+07Vr14zbU/LPP//oxo0b8vT0TPF8EmfZOnDgwFPV5urqqgoVKlh0jKXX2cvLS9mzZ9fOnTv19ddfG2eIex7++OMPSVKNGjWSLJMoSaVKlVKxYsWUkJBgbPu4p7keabVjxw5JSjFclz17dpUqVUpxcXE6cuSIpIy9dom8vLz0ww8/6Mcff1T//v1VvXp140xWx48f16effqpevXrp4cOHyR7fokWLZLcnzu6X3HUHAAAAAOCvv/5S5cqV5ezsbNxWpEgRFStWTH///bckGf+9/OabbxrbWFtbp/hcJz3HLlGihMLDwxUYGKhdu3bp/v37Jn2Y24+0Y+lAAAAAAABeEuXKlTMGhmxtbeXi4qIyZcqoRo0aypLl0T/xw8PDdffuXUlSrVq1zPYZHh4ud3d3k22PzxZlicQQU2rHe3p6SjINIFkytpubm/FcH5cYCMqXL1+yx2XPnl2SkoQvwsLCNHDgQONDqZTcvXvXZEm/RCnN1pM4XfzjQbYrV65IehQUSYvLly/LYDAoJiZGPj4+qba9c+dOmvpMTo4cOVS/fn2tXr1ae/fu1cqVKyU9mukqNYmhmosXL8rb2zvVtrdv336q2p7mvWjpdc6RI4fGjx+vUaNGafr06Zo+fbrc3NyMX1tNmjQxvn+eVeL7PrXlEj09PXXixIlkv0ae9mszLRL/PkeMGKERI0ak2jbx7zMjr92TfH195evrK+nRcpD//POP5s2bp19++UW7d+/Wd999p169eiU5LqVrn7g9pZnEAAAAAAD/bZGRkSpevHiS7a6ursYP6N28eVO2trYms5VLStNS9886tp+fnyZOnKjvvvtOPXv2VNasWVW/fn198MEHcnZ2NrsfaUfQCgAAAACAl8Tbb7+tli1bptomISHB+P8pzdzyOFtb2yTbsmXLZnlx6cTc2E8u8Wfp/id9+OGH+vvvv1W2bFkNHDhQxYoVk6Ojo/G6VK9eXTdv3pTBYEj2eCsrK4vGs0TimA4ODk89y1hatWrVSqtXr9a3336rffv2qXDhwipfvnya6nNzc1P16tVTbfv4Uo6WyKj3Yv369VW1alVt27ZNf//9t/bv368tW7Zoy5YtmjFjhubPn282TJYRnuf1SPzekdJsW497PGD4Ilw7KysrlSxZUlOnTtX9+/e1fft2bd26NdmglTkpfa0DAAAAAP7bnJycFBYWlmR7WFiYXnvtNUmPnpHExsYqKirKJGz1tB9As2RsSWrWrJmaNWum27dva9u2bRo/fryyZMmizz//PE37kTYErQAAAAAAeIXkypVL2bJlU0xMjEaMGPHMn5izRJ48eST9b0ah5CTOmvPkLFqZITo6Wjt37pS1tbW++eYbOTo6Jtl/69atdBsvMZxy9uzZNLXPmzevpEchks8//9ziEJklKlasqEKFCun333+XJLOBvsfrc3Z21oQJE55bbZay9Donypkzp5o3b25cQu7q1asaO3astm3bprFjx2rJkiXPXFvi+z61JfYy62skX758Onv2rFq3bm12Cc8nZcS1S6vq1atr+/btKc7ydvnyZRUrVizJ9sTvW4nvawAAAAAAHle+fHn98MMPioiIMM56fvbsWZ08edI4K3ipUqUkSdu2bTP+GzkhIUH/93//99zHfpyLi4vefvtt7dy5M9nnI+b2I3XP7wkdAAAAAADIcDY2NqpataokaePGjRk6to+PjxwcHBQeHq5t27Yl2R8TE6MNGzZIkipXrpyhtSUnKipK8fHxypEjR5KQlSStXbs2XWe3qVKlimxtbXXs2DEdO3bMbHt3d3d5e3vr3r17+u2339KtjpS0a9dOzs7OcnV1NT4MTI2Pj49y5cql06dP69SpUxaNlThjWHx8/NOUmipLr3NK8uXLp0GDBkmSjh8/ni61VapUSZL022+/JRvi++eff3T8+HFZW1urYsWK6TJmInPXvGbNmpLS5/tGatfuWf7u0/L1GBoaKinlwNTPP/+c7PY1a9ZI+t/f0dNKPL+4uLhn6gcAAAAA8GLp1q2bsmTJoh49emjr1q365Zdf1LdvX+XLl884q3zRokVVt25djRs3TsuXL9eOHTs0cOBAxcTEPNOs6GkZe8aMGRozZoxCQkL0559/aunSpfrtt99UpUqVNO1H2hG0AgAAAADgFTNgwADZ2tpq0qRJWr16tclygon+/fdfbd68OV3HzZo1qzp27ChJ+uKLL0xmtoqNjdVnn32mmzdvqkCBAs99Kby0yJ07t5ycnBQZGWkMWSQ6ePCgpk6dmq7jubq6qn379pKkwYMH699//zXZbzAYtGfPHkVFRRm3vffee5KkUaNGafv27Un6NBgMOnTokHEmqmfRo0cP7du3T7t37zbOTpYaW1tbDRgwQAaDQQMGDNBff/2VpE18fLz27NmjgwcPmmxPnK3J0oBWWlh6nf/55x9t2LBBMTExSfpKvOaPL5X3LCpUqKDSpUsrJiZGQUFBun//vnHf7du3FRQUJElq1KiR8uXLly5jJjJ3zdu0aSMPDw+FhIRo0qRJunv3bpI2N2/e1A8//GB8/TTX7ln+7pctW6aRI0dq//79SfYZDAZt3rxZS5culSQ1btw42T62bNmiX375xWRbSEiINm/erCxZsqhTp04W1/U4FxcX2dra6tatWwoPD3+mvgAAAADgVeSRx0mveeTKsD8eeZzSpe58+fJp8eLFcnJy0vDhw/Xxxx+rWLFiWrx4sXLkyGFs9/nnn8vf318TJ07UiBEjVLBgQbVo0cJkKcHnMbaPj4/Onj2r0aNHq0ePHlq4cKF69uypAQMGpGk/0o6lAwEAAAAAeMWULFlSkyZN0qhRoxQYGKjp06fr9ddfV65cuRQREaF///1X165dU6NGjVSvXr10HXvQoEE6evSo9uzZo0aNGqly5crKnj27Dh48qNDQUDk7O+vLL7+UnZ1duo77NGxsbNSvXz+NHz9eI0eO1LJly1SwYEGFhobqwIEDatq0qf76669Ul0K01Pvvv6/Lly9r+/btatasmUqXLi0PDw+Fh4fr1KlTun79urZt22Z8+BYQEKAPP/xQX3zxhd59910VKlRIhQsXVo4cOXTnzh2dOHFCYWFh6t27t6pXr55udaZVp06dFBoaqnnz5qljx44qWrSoPD09lS1bNt28eVMnTpxQZGSkPv30U5UpU8Z4XP369bVv3z69//77ql69unFGsZ49e8rLy+uZ67LkOoeGhmrIkCHKli2bSpQooXz58ikuLk7//vuvzp07J1tbW73//vvPXFOiKVOmqGvXrtq2bZvefPNNVahQQXFxcdq3b5/u3r2rkiVLGgNX6al+/fpatWqVJk2apD179sjFxUVWVlZq1aqVypUrJwcHB82ZM0d9+/bVt99+qx9++EHe3t5yd3dXTEyMzp8/rzNnzsjV1VVt2rSRpKe6dubqSE1sbKzWrFmjNWvWyMXFRSVKlJCzs7OioqJ0+vRp49dq06ZN1bp162T76NKli4YOHaoFCxaoUKFCunTpkg4dOiRJGjlyZLLLClrC1tZWAQEB2rRpk5o3b67y5csrW7ZskqTPPvvsmfoGAAAAgJddQoJB/dtXy5Rxra0tn1FqwoQJJq+LFSum+fPnp3qMo6OjJk+ebLKtY8eOFv1788lx0zJ27dq1Vbt27afej7QjaAUAAAAAwCuoYcOG8vHx0eLFi7V7927t379f8fHxyp07tzw9PdWxY0c1aNAg3ce1s7MzhjR+/vln/fXXX3r48KHy5cunzp07q3fv3sYZbV4E3bp1U4ECBfTtt9/qzJkzOnXqlLy8vBQUFKT27dvrzTffTNfx7Ozs9NVXX+mXX37R6tWrdfToUR09elTOzs4qVKiQunbtKjc3N5NjunTpoipVqmjJkiXat2+f9uzZI2tra+XOnVvFixeXv79/ugfmLDFixAjVqVNHy5Yt0/79+/Xbb7/J1tZWbm5uqlSpUrL1tW/fXvfu3dPatWu1Y8cOPXjwQNKjgEx6BK0suc6lS5fWsGHD9Ndff+nMmTM6fvy4bGxslDdvXnXs2FGdOnVKl5oSFSxYUKtWrdL8+fO1detW/frrr7K2tlbhwoXVsGFDdenSxRjMSU/+/v7GpQv27t1rnE2rfPnyxoBT0aJFtXbtWn3//ffaunWrTp48qYMHD8rZ2Vl58+ZVjx49VLduXWOfT3Pt0lJHSlq3bq0CBQpo7969OnTokE6fPq2wsDDZ2NgoT548atKkiZo1a2ZcBjE5Xbp0UdmyZbVo0SLjrFsVKlRQr1690u2B85gxY+Ts7KzffvtNmzZtUmxsrCSCVgAAAADwNGGnl23cTZs26erVq3rjjTd0//59rV+/Xn/99ZdmzZqVYTXg+bIyGAyGzC4CAAAAAAAAAJ6XgIAAXblyRdu2bVOBAgUyuxwAAAAAeGXFxMTo3LlzKly48HP5INGL7rffftPUqVN1/vx5xcbGysvLS7169VLTpk0lSfHx8UotppMlC/MlZQZL3rf8DQEAAAAAAAAAAAAAAADPqEaNGqpRo0aK+7t166Y//vgjxf18QOjFR9AKAAAAAAAAAAAAAAAAeM5Gjx6te/fupbg/T548GVgNngZBKwAAAAAAAAAAAAAAAOA58/LyyuwS8IwIWgEAAAAAAAB4pW3fvj2zSwAAAAAAAK8A68wuAAAAAAAAAAAAAAAAAABedAStAAAAAAAAAAAAAAAAAMAMglYAAAAAAAAAAAAAAAAAYAZBKwAAAAAAAAAAAAAAAAAwg6AVAAAAAAAAAAAAAAAAAJhB0AoAAAAAAAAAAAAAAAB4TGBgoJo0aZJqG29vb82bNy+DKsKLIEtmFwAAAAAAAAAAAAAAAIBXmyEhQVbWGT8f0PMcd8WKFcqfP/9z6RsvJoJWAAAAAAAAAAAAAAAAeK6srK11bv1c3Q+7mmFj2rvmU+EmvZ9b/2XKlHlufePFRNAKAAAAAAAAAAAAAAAAz939sKu6f/1iZpdhkX379mn8+PE6f/68Xn/9dX366acqVaqUpEdLB44YMUI9e/aUJHXu3FkODg5q0aKFpk2bphs3bsjHx0fjxo2Tp6ensc9r167pk08+0d69e+Xs7Kzu3bvrypUr2rZtm7Zv327SbvLkyfrtt990//59+fj4aNSoUcbxkfEIWgEAAAAAAAAAAAAAAABPuHnzpsaNG6c+ffooZ86cmjJligYMGKAtW7bI1tY22WOOHz+u27dva/jw4YqPj9eECRP0/vvva8WKFZIkg8Ggfv366datWxo9erRy5sypefPmKTQ0VNaPLXEYERGhDh06yMHBQR9//LFy5sypxYsXq2vXrtq8ebNcXV0z5BrAFEErAAAAAAAAAAAAAAAA4AkRERFasmSJihYtKkmyt7dXly5ddOjQIVWoUCHZY6KiorRmzRq5uLhIkqKjozVq1Chdu3ZNefPm1c6dO3Xs2DEtXbrU2EeVKlVUq1YtOTo6GvtZtGiRIiMj9eOPPxpDVX5+fqpfv77mzZunESNGPM9TRwqszTcBAAAAAAAAAAAAAAAA/lvy5MljDFlJ0uuvvy5Jun79eorHFCtWzBiyevyYa9euSZKOHDkiR0dHk6BW9uzZ5efnZ9LPrl27VLlyZTk5OSkuLk5xcXGytrZWxYoVdeTIkWc/OTwVZrQCAAAAAAAAAAAAAAAAnvD4DFOSjMsFPnjw4KmPuXHjhkkQK9GT2+7cuaODBw+qZMmSSdp6enqmoXo8DwStAAAAAAAAAAAAAAAAgAyQJ08e3b59O8n2J7c5OTmpRo0aGjx4cJK2dnZ2z60+pI6gFQAAAAAAAAAAAAAAAJABfHx8FBkZqT///FMVK1aUJN27d0979uwxmQ2ratWqWrt2rYoUKSIHB4fMKhdPIGgFAAAAAAAAAAAAAAAAZICaNWuqZMmSGjZsmIYOHSpHR0d9++23yp49u6ysrIztunXrpnXr1qlTp07q0qWL8ufPr9u3b+vQoUNyd3dXt27dMu8k/sMIWgEAAAAAAAAAAAAAAOC5s3fN90qPlxZWVlb66quvFBQUpKCgIDk6OqpLly46d+6cjh8/bmyXK1curVixQtOnT9fkyZMVHh4uV1dXlS5dWnXr1s3EM/hvszIYDIbMLgIAAAAAAAAAAAAAAAAvt5iYGJ07d06FCxdWtmzZTPYZEhJkZW2d4TVl1riWePjwoRo3bqwKFSpo/PjxmV3Of05q79snMaMVAAAAAAAAAAAAAAAAnqvMCju9iCGrFStWKCEhQYULF1ZkZKSWL1+uK1euaOrUqZldGswgaAUAAAAAAAAAAAAAAABkkKxZs+qbb77RlStXJEnFihXTnDlz5OPjk8mVwRyCVgAAAAAAAAAAAAAAAEAGad68uZo3b57ZZeApvHjzowEAAAAAAAAAAAAAAADAC4agFQAAAAAAAAAAAAAAAACYQdAKAAAAAAAAAAAAAAAAAMwgaAUAAAAAAAAAAAAAAAAAZhC0AgAAAAAAAAAAAAAAAAAzCFoBAAAAAAAAAAAAAAAAgBkErQAAAAAAAAAAAAAAAADADIJWAAAAAAAAAAAAAAAAQCZr1qyZAgMDja8DAwPVpEkTkza7du3SW2+9pVKlSqlChQqSJG9vb82bNy9daujXr586d+6c5vYBAQEaM2ZMuoyd3Pm+aLJkdgEAAAAAAAAAAAAAAAB4tSUkJMjaOuPnA8qscdNDv379FB0dbbJt1KhR8vb21ieffKKsWbNKklasWKH8+fNnRon/OQStAAAAAAAAAAAAAAAA8FxZW1trzo7vFBpxPcPGzO/krr61umTYeOnN09PT5PW9e/d0/fp1DRo0yDiblSSVKVMmgyt7OcXExChbtmzP1MfLGdkDAAAAAAAAAAAAAADASyU04rouhF3OsD/PEupKXMZu9+7deuutt+Tr66tOnTrp8uXLCg8P1+DBg1WuXDnVqVNHGzZsMDn2+++/V/369VWqVCkFBAToq6++UkJCgkmb/fv3q2XLlvLx8VGTJk20Y8eOFGuQpFWrVqlcuXKSpA8//FDe3t7GZQaTWzrw119/1dtvvy1fX19VqVJFn3zySZLZsc6cOaNOnTrJx8dHderU0erVq5/6ei1dulS1a9dW+fLl1a9fP92+fdu4b9WqVfL29jbZJiVdKjHRjh071KRJE/n4+Khly5Y6ePBgkjarVq3SW2+9JR8fH9WoUUPTpk1TfHx8kjEPHDig7t27q0yZMpo4ceJTn18iZrQCAAAAAAAAAAAAAAAAnnDz5k1NmDBB7777rrJkyaJx48Zp+PDhsre3V4UKFdSmTRv98MMPev/991W6dGl5eHho8eLFGjdunDp37ix/f38dOHBAM2fOVFRUlEaOHGnst2fPnvL29tb06dMVGRmp0aNHKzo6WsWLF0+2Fn9/fy1YsEDdu3fXu+++K39/f7m4uCTbNiQkREOGDFHLli01cOBA3bx5U1OmTFFkZKSmTZsmSXrw4IF69Oghe3t7YwBpxowZunv3rl577TWLrtP27dt14cIFBQUF6c6dOxo/frzGjh1rHMsSN2/e1OjRozVw4EA5Ojpq7ty56tmzpzZv3ixXV1dJ0oIFCzRp0iR17dpVgYGBOnPmjDFoNXz4cJP+hg0bprZt26pv376yt7e3uJ4nEbQCAAAAAAAAAAAAAAAAnhAREaElS5aoaNGikqQbN25o7Nix6t27t/r37y9J8vHx0ZYtW7R161Z16tRJs2bNUuPGjfXRRx9JkqpXr67Y2FjNnz9fffr0Ua5cubRo0SJZWVlp7ty5ypkzpyQpb9686tatW4q1uLi4qFSpUpIeLSmY0nKBBoNBEydOVKNGjfTZZ58Zt7u5ualPnz7q16+fihYtqlWrVunGjRvauHGjMVhVokQJNWjQwOKglcFg0Ndffy07OztJ0pUrVzRnzhwlJCTI2tqyxfbCw8M1ffp0+fn5SZIqVaqkWrVqaeHChRo2bJju3r2rGTNmqFevXho6dKgkqVq1arK1tdWECRPUs2dP5cqVy9hfu3bt1KdPH4tqSA1LBwIAAAAAAAAAAAAAAABPyJMnjzFkJckYQKpatapxm6Ojo1xcXHTt2jWdPXtWd+7cUYMGDUz6adSokWJjY3X48GFJ0qFDh1S5cmVjyEqS/Pz85Ozs/Mw1nzt3TleuXFHDhg0VFxdn/FOpUiVZW1vr6NGjkqTDhw+raNGiJqGqQoUKqVixYhaPWbFiRWPISpKKFCmi2NhYhYWFWdxXzpw5jSGrxNdVq1bVoUOHJEkHDhxQdHS0GjRoYHJ+VatWVUxMjE6dOmXSn7+/v8U1pIYZrQAAAAAAAAAAAAAAAIAnODo6mry2tbWVJJOAlCTZ2dnpwYMHioiIkCTjEneJEl8n7r9586YKFSqUZLyUlgK0xJ07dyTJOOPWk65evSrp0excT9aZWOuDBw8sGvPJ65QYurK0Hyn5a+Dq6qozZ85I+t/5tWjRItnjE88vUe7cuS2uITUErQAAAAAAAAAAAAAAAIBnlDgj1e3bt022J87s5OTkJOnRMn7Jzfb05HHPUkNQUJB8fX2T7M+TJ4/xv8eOHUuyPywsTDly5HjmOh6XNWtWSVJsbKzJ9sjIyCRtk7sGYWFhcnNzk/S/azhz5kzlzZs3SdsCBQo8c72pYelAAAAAAAAAAAAAAAAA4BkVLlxYLi4uCgkJMdm+ceNG2draGoNPvr6+2rdvn6Kiooxt9uzZo/Dw8GeuwcvLS3nz5tWlS5fk4+OT5I+7u7skycfHR6dOndKFCxeMx164cEEnTpx45hqelDjm2bNnjdvOnDmTZPYpSYqKitKePXtMXu/evVulS5eWJJUtW1b29va6du1asueXK1eudK//ccxoBQAAAAAAAAAAAAAAgOcuv5P7Kz2ejY2N+vXrp3HjxsnFxUW1atXSwYMHNXfuXHXt2tUYAuratauWLVum3r17q3fv3oqMjFRwcLBxNqpnYWVlpcDAQA0fPlzR0dHy9/eXvb29QkNDtWPHDg0ZMkSFCxdWy5Yt9fXXX6tv374aPHiwJGnGjBnpvtSeJJUuXVr58uXT559/rmHDhunu3bv65ptvkj1fZ2dnffjhhxo0aJBy5sypuXPnymAwqGvXrpIeLVM4aNAgTZo0SdeuXVOlSpVkY2OjS5cuadu2bQoODpa9vX26n0MiglYAAAAAAAAAAAAAAAB4rhISEtS3VpdMGdfaOuMWfOvcubOyZMmihQsXavny5XJzc9OAAQP0zjvvGNvkyZNHc+fO1bhx4zR48GB5enoqKChI06ZNS5caGjZsKEdHR82ePVvr1q2TJHl4eKhGjRrGIFW2bNk0f/58ffrpp3r//ffl7u6ufv36adu2bSYzbaUHW1tbzZw5U59++qnxfD/44ANNmDAhSVs3NzcNHz5cEydO1MWLF1W0aFHNmzfPJADWo0cPubu7a8GCBVqyZImyZMkiT09P+fv7y9bWNl1rf5KVwWAwPNcRAAAAAAAAAAAAAAAA8MqLiYnRuXPnVLhwYWXLli2zywHSxJL3bcZF9gAAAAAAAAAAAAAAAADgJcXSgQAAAAAAAAAAAAAAAABMxMXFpbjPyspKNjY2GVjNi4GgFQAAAAAAAAAAAAAAAAATJUuWTHGfh4eHtm/fnoHVvBgIWgEAAAAAAAAAAAAAAAAw8dNPP6W4z87OLgMreXEQtAIAAAAAAAAAAAAAAABgwsfHJ7NLeOFYZ3YBAAAAAAAAAAAAAAAAAPCiI2gFAAAAAAAAAAAAAAAAAGYQtAIAAAAAAAAAAAAAAAAAMwhaAQAAAAAAAAAAAAAAAIAZBK0AAAAAAAAAAAAAAAAAwAyCVgAAAAAAAAAAAAAAAABgBkErAAAAAAAAAAAAAAAAPFeGhIT/1Lgvg2bNmikwMDDN7b29vTVv3rxU26xatUre3t66ffv2s5b3QsqS2QUAAAAAwPMQGBioP/74Q9u3b0/zMfv27VOXLl303XffqXLlys+xOgAA8Cry9vbWgAEDNHDgQEmPHi6PGjVK27ZtU4ECBTK5OgAAAADIXFbW1jr49RzdDb2aYWPmyJ9PZd7tm2HjQfL399eKFSvk6OiY2aU8FwStAKTZ0qVLNWbMGPn6+urHH3/M7HIAAMALKPGXiYns7OyUP39+VatWTf369VPu3LkzsToAAPCye/Jew8bGRq6urqpWrZqGDBkid3f3TKwOAAA8T0/eBzyud+/eGj58uH7//Xdt2LBBhw8f1pkzZ5QvXz6LPoAlSSdPntSsWbN05MgR3bp1S87Oznr99dcVEBCgzp07p8epAMB/2t3Qq4q8cCGzy3glPHz4UFmyZJG19Yu1mJ2Li4tcXFwyu4znhqAVgDRbt26dPDw8dPjwYV24cEGFChXK7JIAAMALatCgQSpQoIAePnyov//+W8uXL9eOHTu0fv162dvbZ0gNY8eOlcFgsOiYihUr6vDhw7K1tX1OVQEAgPTw+L3GwYMHtXr1av39999av369smbNmtnlAQCA5yjxPuBxb7zxhiRp/fr12rBhg0qUKKE8efJY3Pf+/fvVpUsX5c+fX2+//bbc3Nx09epVHTp0SN999x1BKwD4jzl16pQmTpyow4cPKyYmRnnz5lXr1q3Vu3dvSdIPP/yg2bNnKywsTGXKlNHw4cPVunVrjR8/Xi1btpT0aObjESNGqGfPnsZ+Fy5cqPHjx+vkyZOSpOjoaE2ePFm7du3StWvX5OrqqurVq+v9999Xzpw5jccFBATI399f+fLl07Jly3T16lXt3r1bLi4uWrVqlRYsWKDz58/L2dlZLVu21KBBg2RjY2M8fv/+/Ro3bpxOnTqlQoUK6f3333+q65KQkKDg4GAtX75c8fHxCggI0McffywHBwdJ/wtH79mzRy4uLrp8+bLefPNNTZw4UYcOHdK6deuUNWtWvfXWWxo2bJiyZPlfdGnLli2aPHmyQkNDVbx4cX3yySfq2rWrunbtapw9WpJ+/fVXzZo1SydPnpSDg4Pq16+vkSNHGmt4nghaAUiTS5cu6cCBA5o5c6aCgoK0bt06DRgwILPLSiI6OjpDvnkCAIDU1axZUz4+PpKkt99+W87OzlqwYIG2bdumJk2aJGn/PH6GP01Yytraml/OAgDwEnjyXiNXrlyaO3eutm3bpkaNGmVydQAA4Hl6/D7gSUOGDNHYsWNla2urvn376tSpUxb1PXv2bOXMmVM//fRTkuWOwsLCnrrmp3H//v0M+7AaACB577zzjnLnzq3PPvtMOXLk0MWLF3Xt2jVJ0v/93//p448/VsuWLdWoUSMdO3ZMgwcPfqpxYmJiFB8fryFDhsjFxUVXr17V7Nmz1a9fPy1evNik7ebNm1WoUCF9+OGHsra2loODgxYsWKBJkyapa9euCgwM1JkzZzRt2jTFx8dr+PDhkqSbN2+qZ8+e8vb21vTp0xUZGanRo0crOjpaxYsXt6jepUuXqnz58powYYLOnz+viRMnytXV1ThWSqZPn64333xT06dP14EDBxQcHCxPT0+1b99ekvTPP/9o8ODBql27tj744ANduXJFQ4YM0cOHD036CQkJ0ZAhQ9SyZUsNHDhQN2/e1JQpUxQZGalp06ZZdC5P48WaPwzAC2vdunVycnJSrVq1VL9+fa1bty5Jm8jISH3++ecKCAhQqVKlVLNmTY0YMUK3b982tnnw4IGCg4NVv359+fj4qHr16howYIAuXrwoSdq3b5+8vb21b98+k74vX74sb29vrVq1yrgtMDBQZcuW1cWLF9W7d2+VLVvW+M37r7/+0qBBg+Tv769SpUqpVq1a+vzzzxUTE5Ok7jNnzmjw4MGqUqWKfH19Vb9+feM34L1798rb21tbtmxJ9pp4e3vrwIEDT3FFAQD4b6lSpYqkRz/TU/sZnpCQoIULF6px48by8fFR1apVFRQUpIiIiCR97tixQ506dVLZsmVVrlw5tWrVyuQeJTAwUAEBASbH/PLLL2rZsqXxmLfeekuLFi0y7k/pXmTjxo1q2bKlfH19VblyZQ0fPlzXr183aZN4XtevX1e/fv1UtmxZValSRV988YXi4+Of7QICAIBUVahQQdKjD4olOnPmjAYNGqRKlSrJx8dHLVu21LZt25Ica+55xsOHD/Xll1+qZcuWKl++vMqUKaMOHTpo7969GXNyAAAgzdzd3Z9pluqLFy/q9ddfTxKykiRXV9ck237++We1bt1apUuXVsWKFdWxY0f9/vvvJm2WLl2qxo0bq1SpUqpevbpGjx6tyMhIkzadO3dWkyZNdPToUXXs2FGlS5fW1KlTJT26F5kxY4bq1q1r/H3HxIkTk/zSGQCQvm7fvq3Lly+rb9++qlOnjqpUqaI2bdpo0KBBkqSvv/5aFSpU0Pjx41WjRg298847atas2VON5eLiotGjR6tBgwaqVKmSGjdurHHjxumPP/7QuXPnTNrGxsZq7ty5qlOnjgICAhQXF6cZM2aoV69eGjlypKpVq6YuXbpoxIgRWrx4se7cuSNJWrRokaysrDR37ly9+eabatGihT777DPjfku4ublpypQpqlmzprp06aImTZpo06ZNZo/z9fXVRx99pGrVqmnAgAGqVKmSyXFz5sxRgQIFFBwcrFq1aqlDhw7q37+/Hjx4YGxjMBg0ceJENWrUSJ999plq1qypVq1aacKECdq4caPFIeunwYxWANJk3bp1qlu3ruzs7NSkSRMtX75chw8flq+vryTp3r176tixo86cOaNWrVqpRIkSunPnjrZv367r16/LxcVF8fHx6tu3r/bs2aPGjRurS5cuunfvnnbt2qV///1Xnp6eFtcVFxennj17qnz58ho5cqSyZcsm6VGKNSYmRu3bt5ezs7MOHz6sJUuW6Nq1a5oxY4bx+BMnTqhjx47KkiWL2rZtKw8PD128eFHbt2/XkCFDVLlyZeXLl894/k9eE09PT5UtW/YZriwAAP8NiaFqZ2dnSSn/DA8KCtLq1avVsmVLde7cWZcvX9bSpUv1zz//aPny5caHpatWrdIHH3ygokWLqm/fvsqZM6eOHz+u3377TW+99VayNezatUtDhw6Vn5+fMdh19uxZ7d+/X127dk2x9sRpjn18fDR06FCFhYXpu+++0/79+7VmzRqTh6/x8fHq2bOnfH19NWLECO3Zs0fz589XwYIF1aFDh2e+jgAAIHlXrlyRJOPP5VOnTql9+/Zyd3dX79695eDgoI0bN6p///4KDg42/hs/Lc8z7t69qx9//FFNmjTR22+/rXv37umnn35Sr1699OOPP1r8yV8AAPBs7t69a/IBb+nRL6jTg4eHhw4cOKB///3XuBxhSmbOnKng4GCVLVtWgwYNkq2trQ4dOqS9e/eqevXqkqTg4GDNnDlTVatWVfv27XXu3DktX75cR44cMXnOIUnh4eHq3bu3GjdurKZNm8rV1VUJCQl699139ffff6tNmzYqUqSI/v33Xy1atEjnz5/XV199lS7nDQBIKleuXPLw8NDUqVMVEREhPz8/5c2bV9Kj58DHjh1LsvRe/fr1n/p785o1a7Rw4UJduHBB0dHRxu3nz59X4cKFja8rV65ssjrEgQMHFB0drQYNGiguLs64vWrVqoqJidGpU6dUqVIlHTp0SJUrVzZZitDPz8/4zN4SVatWNXldpEgR/fLLL2aPS/z5+Phxj3+I6ciRI6pTp46srf83Z9Sbb75pcsy5c+d05coVffDBBybnW6lSJVlbW+vo0aMqWrSoRedjKYJWAMw6evSozp49q48//liSVL58eeXNm1fr1q0zBq3mzZunf//9VzNnzjQJJPXr108Gg0HSox8Oe/bs0ahRo9StWzdjmz59+hjbWOrhw4dq0KCBhg0bZrJ9+PDhxl/YSlLbtm1VqFAhTZ06VaGhocqfP78kady4cTIYDFq9erVxW+LxkmRlZaWmTZtqwYIFioqKMv7guX37tnbt2qV33nnnqeoGAOBVl/jQ8+HDh9q/f79mzZqlbNmyqXbt2jp48GCyP8P/+usv/fjjj5o8ebJJWKpy5crq1auXQkJC9NZbbykqKkrjxo2Tr6+vFi9ebLLUX2r3FL/++qty5MihefPmmaxLn5rY2FhNnjxZb7zxhpYuXWocq3z58urbt68WLlxo/AST9Gj2zoYNG6p///6SpPbt26tFixb66aefCFoBAJCOHr/XOHTokGbOnCk7OzvVrl1bkvTZZ58pX758Wrlypezs7CRJHTp0UPv27TV58mTjs4u0PM9wcnLS9u3bjf1IUps2bdSwYUMtXrxYn3/+eUadNgAAkEx+v5Do5MmT6dJ3jx491Lt3bzVv3ly+vr4qX768/Pz8VLlyZZNQ1IULFzRr1izVrVtXM2bMMPmFcOI9xO3btzVnzhxVr15dc+fONbbx8vLSmDFjtHbtWrVq1cp43M2bNzV69Gi1a9fOuO3nn3/W7t27tXjxYuMMnpJUtGhRffLJJ9q/f7/KlSuXLucOADBlZWWlefPmadq0aRozZoyio6NVsmRJjRo1Sq+99pri4uKSBH1z5879VGNt2bJFI0eOVNu2bTVkyBA5Ozvr5s2bSWZzkpLOsJg4I1WLFi2S7fvq1auSHv2cKVSoUJL9TxNWfnLmR1tb2zTNtPh4yCu5427evJmknhw5cpj8DiDxfBOfwT8p8XyfJ4JWAMxat26dcufOrcqVK0t69EOlUaNGWrt2rQIDA2VjY6PNmzerWLFiSWZ9SmwvPVovNleuXOrUqVOKbZ5G4pqtj3s8ZBUdHa2YmBiVLVtWBoNB//zzj/Lnz6/bt2/rzz//VJcuXUxCVk/W06xZM82ZM0chISF6++23JUkbNmxQXFycmjZt+tR1AwDwKnvyoaeHh4cmT54sd3d347Ynf4aHhIQoZ86cqlatmsknU0uWLCkHBwft27dPb731lnbt2qV79+6pT58+Jv/AklK/p3B0dNT9+/e1a9cu1axZM03ncfToUYWFhWnAgAEmY/n7+8vLy0u//vqrSdAqufMqX7681q5dm6bxAABA2iR3rzFp0iTlzZtX4eHh2rt3rwYNGqS7d++atKtevbqCg4N1/fp1ubu7p+l5ho2NjTGknZCQoMjISCUkJKhUqVL6559/ns8JAgCAFAUFBZnM7JGeqlWrpu+//17ffPONfv/9dx04cEDffvutXFxcNG7cOOOsGlu3blVCQoL69+9vErKS/ncPsXv3bsXGxqpLly4mbd5++21NnTpVO3bsMAla2dnZqWXLliZ9hYSEqEiRIvLy8jJ5VlKlShVJ0r59+whaAcBzVLhwYc2YMUOxsbE6cOCApk6dqnfeeUe//vqrsmTJkmSGxVu3biXpw87OTrGxsSbbnlxCNiQkRMWLF9eYMWOM2/74449ka3ryGbiTk5OkRzMtJs649bgCBQpIerTcX1hYWJL9T55DZnJzc0tSz927d03CZokzcAUFBRknhXlcnjx5nmuNEkErAGbEx8frl19+UeXKlXX58mXjdl9fX82fP1979uxR9erVdfHiRdWrVy/Vvi5evKjChQsrS5b0+9aTJUuWZH9ghIaGasaMGdq+fbsiIiJM9iU+ZL106ZIkmZ3+t0iRIvLx8dG6deuMQat169apTJkyyaZ+AQDA/x562tjYKHfu3CpcuLDJQ8XkfoZfuHBBUVFR8vPzS7bPxH8EJi5DaOn0vx06dNDGjRvVu3dvubu7q1q1amrYsGGqoavQ0FBJSvYBrpeXl/7++2+TbVmzZk3yiRsnJ6ck9yMAAODZJN5rREVFaeXKlfrzzz+NM05dvHhRBoNBX375pb788stkjw8LC5O7u3uanmdI0urVqzV//nydO3fO5AF54gNrAACQcXx9feXj4/PUx8fHxyf5Ja6Tk5PxXsLX11czZ87Uw4cPdeLECW3dulULFy7U4MGDtWbNGr3++uu6ePGirK2tVaRIkRTHSXym4OXlZbLdzs5OBQsWNC59nMjd3d1kBk3p0bOSM2fOmH1WAgB4vmxtbVWpUiX16dNH7777rm7duqUSJUpoy5YtJh8E2rRpU5Jj8+bNqzNnzphs2717t8nrmJgYk5kTpUe/j06LsmXLyt7eXteuXUv2Q0SJfH19tXz5cpNVnPbs2aPw8PA0jZMRfHx89OuvvyowMND4+4StW7eatPHy8lLevHl16dIldezYMTPKJGgFIHV79+7VzZs39csvvyS7ruq6deuSrKX6LFKahSIhISHZ7XZ2dkk+LRIfH6/u3bsrIiJCvXr1kpeXlxwcHHT9+nUFBgam2Fdqmjdvrs8++0zXrl3Tw4cPdfDgQQUFBVncDwAA/xXmHnom9zM8ISFBrq6umjx5crLHPM0Uxo9zdXXVmjVr9Pvvv2vnzp3auXOnVq1apebNm+uLL754pr4TpXVJQgAA8Gwev9eoU6eOOnTooGHDhikkJMT47/4ePXqoRo0ayR7v6emZ5rF+/vlnBQYGqk6dOurZs6dcXV1lY2OjOXPmGD/EBQAAXh5Xr141zkyV6LvvvjOu6pHIzs5Ovr6+8vX11WuvvaZRo0YpJCREAwYMeC51Pb5SR6KEhAS98cYbGjVqVLLHJPdBdAB40eXIn++lGO/EiRP64osv1KhRIxUsWFB3797VnDlz5OHhIU9PT73zzjvq16+fRo0apUaNGunYsWP6+eefk/RTv359LVq0SD4+PipcuLDWrl2r69evm7SpWrWqxowZo1mzZqls2bLasWOH9uzZk6Y6HR0dNWjQIE2aNEnXrl1TpUqVZGNjo0uXLmnbtm0KDg6Wvb29unbtqmXLlql3797q3bu3IiMjFRwcbJwh6kXQt29ftW7dWgMHDlSbNm0UGhqq+fPnK2vWrMYcgZWVlQIDAzV8+HBFR0fL399f9vb2Cg0N1Y4dOzRkyJDnNvNlIoJWAFK1bt06ubq6Jhsq2rJli7Zs2aLRo0fL09NTp06dSrUvT09PHTp0SLGxsUkSuYkS13ONiooy2f7kJztS8++//+r8+fP64osv1Lx5c+P2Xbt2mbQrWLCgsb05jRo10oQJE7R+/Xpjorhhw4ZprgkAAJjn6empPXv2qFy5csk+XHy8nSSdOnXK4tkl7ezsFBAQoICAACUkJOjTTz/VihUr1K9fv2T7Slxe+Ny5c0k+PXru3Lkkyw8DAICMZ2Njo6FDh6pLly5aunSpcQkeW1tbVa1aNdVj0/I8Y9OmTSpYsKBmzpxp8gGxGTNmPHvxAAAgw7m5uWnBggUm24oVK5bqMaVKlZIk3bhxQ9Kje4iEhASdOXNGxYsXT/aYxGcGZ8+eNf4+QpIePnyoy5cvm71PSRznxIkT8vPzS/GD6gDwMjEkJKjMu30zZVyrJz74a46bm5ty586tOXPm6Pr168qZM6cqVKigSZMmycbGRm+++aZGjx6t2bNn65dfflHp0qU1ffp04wpJifr166ewsDDNmjVLVlZWatu2rbp06aIJEyYY27Rr106XL1/WkiVLNG/ePFWvXl1TpkxRmzZt0lRrjx495O7urgULFmjJkiXKkiWLPD095e/vb/y9fJ48eTR37lyNGzdOgwcPlqenp4KCgjRt2jSLrsvzVKJECU2fPl1TpkzRgAEDVLRoUU2YMEFdunQxzsIlSQ0bNpSjo6Nmz55tnPnLw8NDNWrUUO7cuZ97nQStAKQoJiZGmzdvVoMGDdSgQYMk+/PkyaP169dr+/btqlevnmbNmqUtW7YkmZLQYDDIyspK9erV06+//qqlS5eaTKH4eBsPDw/Z2Njozz//VJ06dYz7ly9fnua6E2fHMBgMJv1/9913Ju1cXFxUsWJFrVy5Ut27dzf5RWliPY+3rVGjhtauXasHDx6oevXqzzyrBgAAMNWwYUMtW7ZMX331lYYOHWqyLy4uTtHR0XJ0dFT16tWVPXt2zZkzRzVq1FDWrFmN7Z78Gf64O3fuKFeuXMbX1tbW8vb2lvToIWdySpUqJVdXV33//fdq3bq1cQr/HTt26MyZM+rfv/8znTMAAEgflStXlq+vrxYtWqSuXbuqUqVKWrFihTp16qQ8efKYtL19+7bx3/RpeZ6ROGPl4/cZhw4d0sGDBwldAwDwEsqaNWuKIae9e/eqcuXKSZ4t7NixQ9L/lgGsU6eOJk+erFmzZmnGjBkms3Yn3jNUrVpVtra2Wrx4sWrWrGns86efflJUVJRq1aplttaGDRtqx44d+uGHH9S2bVuTfTExMUpISJCDg0PaTx4AMpmlYafMHNfV1VWTJk1KtU27du3Url074+vIyMgkbRwcHDR+/Pgk27t37278fxsbG40cOVIjR440aXPy5EmT19u3b0+xlsaNG6tx48ap1luhQgWtWbPGZJu/v3+qxzzpyZokqVu3bia//2/ZsqVatmxpfF2gQIFkj/vwww/14YcfmmyrV6+e6tWrZ3y9Z88excXFJQk2V6tWTdWqVbOo9vRC0ApAirZv36579+4pICAg2f1lypSRi4uL1q5dqylTpmjTpk0aPHiwWrVqpZIlSyoiIkLbt2/X6NGjVaxYMTVv3lxr1qzR+PHjdfjwYZUvX17379/Xnj171L59e9WpU0c5c+ZUgwYNtGTJEllZWalgwYL69ddfLVpn3MvLS56envriiy90/fp15ciRQ5s2bUr2B9tHH32k9u3bq0WLFmrbtq0KFCigK1eu6Ndff00ytWPz5s01aNAgSdLgwYMtuJIAACAtKlWqpLZt22rOnDk6fvy4qlWrJltbW50/f14hISH68MMP1aBBA+XIkUOjRo3SRx99pNatW6tJkyZydHTUiRMnFBMTk+IygB999JEiIiJUpUoVubu7KzQ0VEuWLFHx4sVVpEiRZI+xtbXV8OHDNWrUKHXq1EmNGzdWWFiYvvvuO3l4eCQJjwMAgMzTs2dPDR48WKtWrdInn3yiDh066K233lKbNm1UsGBB3bp1SwcPHtS1a9e0du1a4zHmnmf4+/tr8+bN6t+/v/z9/XX58mV9//33ev311xUdHZ3JZw0AAB534sQJ4y+hL1y4oKioKH311VeSHs1aldLvOxKNGzdO9+/fV926deXl5aXY2Fjt379fGzdulIeHh/GXxoUKFdI777yjr776Sh06dFC9evVkZ2enI0eOKE+ePBo2bJhcXFzUt29fzZw5U7169VJAQIDOnTunZcuWycfHR02bNjV7Ps2aNdPGjRv1ySefaN++fSpXrpzi4+N19uxZhYSE6NtvvzUupwwAwKvg008/lZ+fn5ydnXX69Gl99dVXKlGihCpUqJDZpRkRtAKQorVr1ypr1qwpJkGtra3l7++vdevW6eHDh1q6dKmCg4O1ZcsWrV69Wq6urvLz85O7u7ukR0ncuXPn6uuvv9b69eu1efNmOTs7q1y5csbZJKRHvwSNi4vT999/Lzs7OzVo0EAjRoxQkyZN0lS3ra2tZs+erXHjxmnOnDnKmjWr6tatq44dO6pZs2YmbYsVK6YffvhBX375pZYvX64HDx4of/78yS4LWLt2bTk5OSkhISHJ+u0AACB9jBkzRqVKldL333+vadOmycbGRh4eHmratKnKlStnbPf222/L1dVV33zzjb766itlyZJFXl5eqQafmjZtqh9++EHLli1TZGSk3Nzc1LBhQw0cONDkk6dPatmypbJly6a5c+dq8uTJcnBwUJ06dfT+++8blz0GAACZr169evL09NT8+fPVpk0brVy5UjNnztTq1asVHh4uFxcXlShRwmRGyuzZs5t9ntGyZUvdunVLK1as0O+//67XX39dkyZNUkhIiP7444/MOl0AAJCMf/75R19++aXJtsTXLVq0MBu0GjFihEJCQrRjxw6tWLFCsbGxyp8/vzp06KB3333X5DnA4MGDVaBAAS1ZskTTpk2Tvb29vL29TX4PMXDgQLm4uGjJkiUaP368nJyc1KZNGw0dOtS4lFNqrK2tNWvWLC1cuFA///yztmzZInt7exUoUECdO3dW4cKFLbk8AACkKi4uLsV9j8/4/DxFRkZq7NixCg8PV44cOVSjRg2NHDky1Wf4Gc3K8PjaWgCAFMXFxalGjRqqXbu2Pv/888wuBwAAAAAAAAAAAABeKDExMTp37pwKFy6sbNmyZXY5SKPLly+nOtlIpUqVtHjx4gysKGNZ8r5lRisASKOtW7fq9u3bat68eWaXAgAAAAAAAAAAAABAusiTJ49++umnFPdnz549A6t5sRG0AgAzDh06pJMnTxrXf61UqVJmlwQAAAAAAAAAAAAAQLqws7OTj49PZpfxUrB4EcMLFy4oKChIzZo1U4kSJdSkSZM0HWcwGPTNN9/I399fvr6+atu2rQ4ePGjp8ACQ4ZYvX65PP/1ULi4u+uKLLzK7HAAAAAAAAAAAAAAAkAksDlqdOnVKO3bsUKFChVSkSJE0Hzd37lzNmDFD3bp105w5c+Tm5qYePXro0qVLlpYAABlqwoQJ+ueff7Rq1Sq98cYbmV0OAAAAAAAAAAAAAADIBBYHrQICArRjxw7NmDFDJUuWTNMxDx480Jw5c9SjRw9169ZNfn5+mjp1qpydnTVv3jyLiwYAAAAAAAAAAAAAAACAjGRx0Mra2uJDtH//ft29e1cNGzY0brOzs1PdunW1c+dOi/sDAAAAAAAAAAAAAAAAgIyUJSMGOXv2rCTJy8vLZHuRIkW0aNEixcTEKFu2bBb1eeDAARkMBtna2qZbnQAA4MUTGxsrKysrlS1b9pn64d4BAIBXH/cNAAAgrbhvAAAAaZVe9w0AXg0ZErSKjIyUnZ2dsmbNarLd0dFRBoNBERERFgetDAaDDAaDHj58mJ6lAgCAVxT3DgAAIK24bwAAAGnFfQMAAABeFlu3btX169fVsWPHzC7lpZYhQavnwdbWVgaDQa+//npmlwIAAJ6j06dPy8rK6pn74d4BAIBXH/cNAAAgrbhvAAAAaZVe9w2QEhIMsrbO+GuZWeO+aLZu3aqjR48StHpGGRK0cnR01MOHD/XgwQOTWa0iIyNlZWUlJyenp+rXyspKDg4O6VUmAAB4AaXnP164dwAA4NXGfQMAAEgr7hsAAEBaEbJKP9bWVvr5h326dTMqw8bM7ZZTzdpUzrDx0iomJsbild8yisFgUGxsrOzs7DK7lBeSdUYM4uXlJUk6d+6cyfazZ88qf/78L+ybBwAAAAAAAAAAAAAAAOnj1s0oXQ8Nz7A/zxLqCgwMVJMmTbRjxw41adJEPj4+atmypQ4ePGjSbtWqVXrrrbfk4+OjGjVqaNq0aYqPjzfZ7+3trQMHDqh79+4qU6aMJk6cKEm6fv26RowYoapVq8rX11cNGjTQokWLnqr/gwcPqkuXLipdurQCAgL0008/mZzL6tWrderUKXl7e8vb21uBgYFJzrNp06by8fHR9u3bJUmbN29Ws2bN5OPjo+rVq2v8+PF68OCBsd99+/bJ29tbu3bt0rBhw1S2bFnVrl1bc+fOferr/qLLkBmtypUrpxw5cmjjxo0qVqyYJCk2NlabN29WzZo1M6IEAAAAAAAAAAAAAAAAIM1u3ryp0aNHa+DAgXJ0dNTcuXPVs2dPbd68Wa6urlqwYIEmTZqkrl27KjAwUGfOnDEGoYYPH27S17Bhw9S2bVv17dtX9vb2unPnjtq2bStJGjJkiAoUKKALFy7o4sWLxmMs6X/o0KFq27atevfurQ0bNujDDz9Unjx5VLNmTfXr10+3b9/W2bNnNXnyZEmSi4uL8dgbN25o3Lhxevfdd5UvXz7lz59f27Zt06BBg9S4cWMNGzZMZ8+e1bRp03T16lXNmDHDZOxPPvlEzZo106xZs7R161ZNnjxZ3t7er2QmyOKg1f3797Vjxw5J0pUrV3T37l2FhIRIkipVqiQXFxd17dpVoaGh2rJliyQpa9as6tu3r4KDg+Xi4qI33nhDy5cvV3h4uHr27JmOpwMAAAAAAAAAAAAAAAA8u/DwcE2fPl1+fn6SHuViatWqpYULF6pv376aMWOGevXqpaFDh0qSqlWrJltbW02YMEE9e/ZUrly5jH21a9dOffr0Mb6eNm2awsLCtHHjRhUoUECSjONI0t27dy3qv1mzZurbt68kqUaNGrp06ZJmzZqlmjVrytPTUy4uLgoNDVWZMmWSnGdERITmzp2r0qVLG7e99957KlOmjKZMmSJJqlmzpuzt7RUUFKSTJ0/K29vb2LZevXoaOHCg8Rx+/fVXbdq06ZUMWlm8dGBYWJgGDx6swYMH648//tDVq1eNr0+dOiVJSkhIMJmmTJJ69+6tAQMGaP78+erTp4+uXbumefPmqWDBgulzJgAAAAAAAAAAAAAAAEA6yZkzp0n4KWfOnKpataoOHTqkAwcOKDo6Wg0aNFBcXJzxT9WqVRUTE2PM0CTy9/c3eb1nzx5VqVLFGLJ6kqX9161b1+R1vXr1dOzYsST5neQ4OzubhKzu3bun48ePq379+ibtGjVqJEn6+++/TbZXr17d+P9WVlYqUqSIrl27Znbcl5HFM1oVKFBAJ0+eTLXN4sWLk2yzsrJS3759jek5AAAAAAAAAAAAAAAA4EX1+PJ6iVxdXXXmzBnduXNHktSiRYtkj7169arJ69y5c5u8Dg8PV9GiRVMc29L+XV1dk4wXGxurO3fuJBn7SU/uj4qKksFgSNJnzpw5ZWdnp4iIiCTbH2dra6uoqKhUx3xZWRy0AgAAAAAAAAAAAAAAAF51t2/fTrItLCxMbm5ucnJykiTNnDlTefPmTdIupZmqEjk7O+vGjRsp7re0/7CwMLm7uxtf37p1S7a2tibLC6bEysrK5HXOnDllZWWV5PyjoqL08OFDY23/RQStAAAAAAAAAAAAAAAAgCdERUVpz549xuUDo6KitHv3bnXs2FFly5aVvb29rl27lmTZvrTw8/PT/PnzFRoaqvz58yfZb2n/W7ZsUYkSJYyvN2/erJIlS8rGxkbSo1mmHjx4kKbasmfPruLFiyskJETdunUzbt+4caMkqXz58mnq51VE0AoAAAAAAAAAAAAAAAB4grOzsz788EMNGjRIOXPm1Ny5c2UwGNS1a1c5Ojpq0KBBmjRpkq5du6ZKlSrJxsZGly5d0rZt2xQcHCx7e/sU++7WrZt+/vlnderUSe+++64KFiyoS5cu6fz583r//fct7v/nn39WtmzZVKJECW3YsEF//vmnvvnmG+P+IkWKaOXKlVq/fr0KFSqkXLlypTrr1oABA9S/f38NHz5cTZs21blz5zRt2jTVr19f3t7e6XOBX0IErQAAAAAAAAAAAAAAAPDc5XbL+VKN5+bmpuHDh2vixIm6ePGiihYtqnnz5il37tySpB49esjd3V0LFizQkiVLlCVLFnl6esrf31+2trap9p0rVy4tX75cU6ZM0eTJk3X//n15eHioQ4cOxjaW9D9lyhRNnTpVs2bNkqurq8aOHatatWoZ97du3VqHDx/W2LFjFR4erhYtWmjChAkp1vfmm2/qyy+/1KxZs9SvXz85OzurTZs2GjZs2NNcyleGlcFgMGR2EU/jyJEjkiQfH59MrgQAADxP6fUzn3sHAABefdw3AACAtOK+AQAApBU/7y0TExOjc+fOqXDhwsqWLZvJvoQEg6ytrTK8pqcdNzAwUEePHtX69eufQ1XpZ9WqVRo1apT27NkjFxeXzC7npZTa+/ZJ1hlUEwAAAAAAAAAAAAAAAP6jMiNklZnj4tVE0AoAAAAAAAAAAAAAAAAAzMiS2QUAAAAAAAAAAAAAAAAAL5IJEyZkdglp0rJlS7Vs2TKzy/jPYEYrAAAAAAAAAAAAAAAAADCDoBUAAAAAAAAAAAAAAAAAmEHQCgAAAAAAAAAAAAAAAOnGYDBkdglAmlnyfiVoBQAAAAAAAAAAAAAAgGdma2srSYqOjs7kSoC0S3y/Jr5/U5PleRcDAAAAAAAAAAAAAACAV5+NjY2cnZ1148YNSZKDg4OsrKwyuSogeQaDQdHR0bpx44acnZ1lY2Nj9hiCVgAAAAAAAAAAAAAAAEgXefPmlSRj2Ap40Tk7Oxvft+YQtAIAAAAAAAAAAAAAAEC6sLKyUr58+ZQnTx7FxsZmdjlAqmxtbdM0k1UiglYAAAAAAAAAAAAAAABIVzY2NhYFWICXgXVmFwAAAAAAAAAAAAAAAAAALzqCVgAAAAAAAAAAAAAAAABgBkErAAAAAAAAAAAAAAAAADCDoBUAAAAAAAAAAAAAAAAAmEHQCgAAAAAAAAAAAAAAAADMIGgFAAAAAAAAAAAAAAAAAGYQtAIAAAAAAAAAAAAAAAAAMwhaAQAAAAAAAAAAAAAAAIAZBK0AAAAAAAAAAAAAAAAAwAyCVgAAAAAAAAAAAAAAAABgBkErAAAAAAAAAAAAAAAAADCDoBUAAAAAAAAAAAAAAAAAmEHQCgAAAAAAAAAAAAAAAADMIGgFAAAAAAAAAAAAAAAAAGYQtAIAAAAAAAAAAAAAAAAAMwhaAQAAAAAAAAAAAAAAAIAZBK0AAAAAAAAAAAAAAAAAwAyCVgAAAAAAAAAAAAAAAABgBkErAAAAAAAAAAAAAAAAADCDoBUAAAAAAAAAAAAAAAAAmEHQCgAAAAAAAAAAAAAAAADMIGgFAAAAAAAAAAAAAAAAAGYQtAIAAAAAAAAAAAAAAAAAMwhaAQAAAAAAAAAAAAAAAIAZBK0AAAAAAAAAAAAAAAAAwAyCVgAAAAAAAAAAAAAAAABgBkErAAAAAAAAAAAAAAAAADCDoBUAAAAAAAAAAAAAAAAAmEHQCgAAAAAAAAAAAAAAAADMIGgFAAAAAAAAAAAAAAAAAGYQtAIAAAAAAAAAAAAAAAAAMwhaAQAAAAAAAAAAAAAAAIAZBK0AAAAAAAAAAAAAAAAAwAyCVgAAAAAAAAAAAAAAAABgBkErAAAAAAAAAAAAAAAAADDD4qDVmTNn1L17d5UpU0bVqlXTxIkT9fDhQ7PH3blzR0FBQfL391eZMmXUpEkTLV++/KmKBgAAAAAAAAAAAAAAAICMlMWSxhEREeratatee+01BQcH6/r165owYYJiYmIUFBSU6rGDBw/W2bNnNXToUOXLl087d+7Up59+KhsbG7Vp0+aZTgIAAAAAAAAAAAAAAAAAnieLglbff/+97t27p5kzZ8rZ2VmSFB8fr9GjR6tv375yd3dP9ribN29q3759Gj9+vFq2bClJ8vPz05EjR/TLL78QtAIAAAAAAAAAAAAAAADwQrNo6cCdO3fKz8/PGLKSpIYNGyohIUG7du1K8bi4uDhJUs6cOU2258iRQwaDwZISAAAAAAAAAAAAAAAAACDDWRS0Onv2rLy8vEy2OTo6ys3NTWfPnk3xuHz58ql69eqaPXu2Tp8+rbt372rDhg3atWuXOnbs+HSVAwAAAAAAAAAAAAAAAEAGsWjpwMjISDk6OibZ7uTkpIiIiFSPDQ4O1pAhQ9S4cWNJko2NjT766CPVr1/fkhJMGAwGRUdHP/XxAADgxWcwGGRlZZVufXHvAADAq4v7BgAAkFbcNwAAgLRKz/sGAC8/i4JWT8tgMGjUqFE6f/68pkyZIjc3N+3evVuff/65nJycjOErS8XGxur48ePpXC0AAHjR2NnZpUs/3DsAAPDq474BAACkFfcNAAAgrdLrvgHAy8+ioJWjo6OioqKSbI+IiJCTk1OKx/36668KCQnR2rVr5e3tLUmqXLmywsLCNGHChKcOWtna2ur1119/qmMBAMDL4fTp0+nWF/cOAAC82rhvAAAAacV9AwAASKv0vG8A8PKzKGjl5eWls2fPmmyLiorSzZs35eXlleJxp0+flo2Njd544w2T7cWLF9ePP/6o+/fvy97e3pJSJElWVlZycHCw+DgAAPDySM/peLl3AADg1cZ9AwAASCvuGwAAQFqxbCCAx1lb0rhmzZravXu3IiMjjdtCQkJkbW2tatWqpXich4eH4uPjdfLkSZPtx44dk6ur61OFrAAAAAAAAAAAAAAAAAAgo1gUtGrXrp2yZ8+u/v376/fff9fKlSs1ceJEtWvXTu7u7sZ2Xbt2Vd26dY2va9asqfz582vQoEH6+eeftWfPHk2aNEmrV69Wp06d0u9sAAAAAAAAAAAAAAAAAOA5sGjpQCcnJy1atEhjx45V//79lT17drVu3VpDhgwxaZeQkKD4+Hjj6xw5cmjhwoWaNm2aJk+erKioKBUoUECBgYEErQAAAAAAAAAAAAAAAAC88CwKWklSkSJFtHDhwlTbLF68OMm2QoUKafr06ZYOBwAAAAAAAAAAAAAAAACZzqKlAwEAAAAAAAAAAAAAAADgv4igFQAAAAAAAAAAAAAAAACYQdAKAAAAAAAAAAAAAAAAAMwgaAUAAAAAAAAAAAAAAAAAZhC0AgAAAAAAAAAAAAAAAAAzCFoBAAAAAAAAAAAAAAAAgBkErQAAAAAAAAAAAAAAAADADIJWAAAAAAAAAAAAAAAAAGAGQSsAAAAAAAAAAAAAAAAAMIOgFQAAAAAAAAAAAAAAAACYQdAKAAAAAAAAAAAAAAAAAMwgaAUAAAAAAAAAAAAAAAAAZhC0AgAAAAAAAAAAAAAAAAAzCFoBAAAAAAAAAAAAAAAAgBkErQAAAAAAAAAAAAAAAADADIJWAAAAAAAAAAAAAAAAAGAGQSsAAAAAAAAAAAAAAAAAMIOgFQAAAAAAAAAAAAAAAACYQdAKAAAAAAAAAAAAAAAAAMwgaAUAAAAAAAAAAAAAAAAAZhC0AgAAAAAAAAAAAAAAAAAzCFoBAAAAAAAAAAAAAAAAgBkErQAAAAAAAAAAAAAAAADADIJWAAAAAAAAAAAAAAAAAGAGQSsAAAAAAAAAAAAAAAAAMIOgFQAAAAAAAAAAAAAAAACYQdAKAAAAAAAAAAAAAAAAAMwgaAUAAAAAAAAAAAAAAAAAZhC0AgAAAAAAAAAAAAAAAAAzCFoBAAAAAAAAAAAAAAAAgBkErQAAAAAAAAAAAAAAAADADIJWAAAAAAAAAAAAAAAAAGAGQSsAAAAAAAAAAAAAAAAAMIOgFQAAAAAAAAAAAAAAAACYQdAKAAAAAAAAAAAAAAAAAMwgaAUAAAAAAAAAAAAAAAAAZhC0AgAAAAAAAAAAAAAAAAAzCFoBAAAAAAAAAAAAAAAAgBkErQAAAAAAAAAAAAAAAADADIJWAAAAAAAAAAAAAAAAAGAGQSsAAAAAAAAAyAAJCYbMLsGsl6FGAAAAAAAyS5bMLgAAAAAAAAAA/gusra00a/kuXbkRkdmlJMsjj5P6t6+W2WUAAAAAAPDCImgFAAAAAAAAABnkyo0Inb9yJ7PLAAAAAAAAT4GlAwEAAAAAAAAAAAAAAADADIJWAAAAAAAAAAAAAAAAAGAGQSsAAAAAAAAAAAAAAAAAMIOgFQAAAAAAAAAAAAAAAACYQdAKAAAAAAAAAAAAAAAAAMwgaAUAAAAAAAAAAAAAAAAAZhC0AgAAAAAAAAAAAAAAAAAzCFoBAAAAAAAAAAAAAAAAgBkErQAAAAAAAAAAAAAAAADADIuDVmfOnFH37t1VpkwZVatWTRMnTtTDhw/TdOz169c1cuRIValSRb6+vmrYsKHWrl1rcdEAAAAAAAAAAAAAAAAAkJGyWNI4IiJCXbt21Wuvvabg4GBdv35dEyZMUExMjIKCglI99saNG2rbtq0KFy6ssWPHKkeOHDp16lSaQ1oAAAAAAAAAAAAAAAAAkFksClp9//33unfvnmbOnClnZ2dJUnx8vEaPHq2+ffvK3d09xWMnTZqkvHnz6ttvv5WNjY0kyc/P7+krBwAAAAAAAAAAAAAAAIAMYtHSgTt37pSfn58xZCVJDRs2VEJCgnbt2pXicXfv3tXGjRvVoUMHY8gKAAAAAAAAAAAAAAAAAF4WFs1odfbsWbVq1cpkm6Ojo9zc3HT27NkUjzt27JhiY2OVJUsWderUSQcOHJCzs7OaN2+u9957T7a2tk9VvMFgUHR09FMdCwAAXg4Gg0FWVlbp1hf3DgAAvLq4bwDwIrOyspK9vX1ml5Em9+/fl8FgyOwygOeK+wYAAJBW6XnfAODlZ1HQKjIyUo6Ojkm2Ozk5KSIiIsXjbt26JUn66KOP1KZNGw0YMECHDx/WjBkzZG1trWHDhllY9iOxsbE6fvz4Ux0LAABeHnZ2dunSD/cOAAC8+rhvAPCisre3V4kSJTK7jDQ5d+6c7t+/n9llAM8d9w0AACCt0uu+AcDLz6Kg1dNKSEiQJFWtWlWBgYGSpCpVqujevXuaP3+++vfvr2zZslncr62trV5//fV0rRUAALxYTp8+nW59ce8AAMCrjfsGAC+yl+kT8IULF2ZGK7zyuG8AAABplZ73DQBefhYFrRwdHRUVFZVke0REhJycnFI9TnoUrnqcn5+fZs+erQsXLsjb29uSUiQ9ejjh4OBg8XEAAODlkZ6/jODeAQCAVxv3DQCQPl6WJQ6BZ8F9AwAASKuX6UMTAJ4/a0sae3l56ezZsybboqKidPPmTXl5eaV4nLlPcjx48MCSMgAAAAAAAAAAAAAAAAAgQ1kUtKpZs6Z2796tyMhI47aQkBBZW1urWrVqKR7n4eGhN954Q7t37zbZvnv3bmXLlo0pdQEAAAAAAAAAAAAAAAC80CwKWrVr107Zs2dX//799fvvv2vlypWaOHGi2rVrJ3d3d2O7rl27qm7duibHDhkyRNu3b9dnn32mXbt2afbs2Zo/f766devGlLoAAAAAAAAAkMmccmaTISEhs8sw62WoEQAAAADwaspiSWMnJyctWrRIY8eOVf/+/ZU9e3a1bt1aQ4YMMWmXkJCg+Ph4k20BAQGaOnWqvvrqKy1fvlx58uTRwIED1adPn2c/CwAAAAAAAADAM8mezU5W1tY6t36u7oddzexykmXvmk+Fm/TO7DIAAAAAAP9RFgWtJKlIkSJauHBhqm0WL16c7PZGjRqpUaNGlg4JAAAAAAAAAMgg98Ou6v71i5ldBgAAAAAALxyLlg4EAAAAAAAAAAAAAAAAgP8iglYAAAAAAAAAAAAAAAAAYAZBKwAAAAAAAAAAAAAAAAAwg6AVAAAAAAAAAAAAAAAAAJhB0AoAAAAAAAAAAAAAAAAAzCBoBQAAAAAAAAAAAAAAAABmELQCAAAAAAAAAAAAAAAAADMIWgEAAAAAAAAAAAAAAACAGQStAAAAAAAAAAAAAAAAAMAMglYAAAAAAAAAAAAAAAAAYAZBKwAAAAAAAAAAAAAAAAAwg6AVAAAAAAAAAAAAAAAAAJhB0AoAAAAAAAAAAAAAAAAAzCBoBQAAAAAAAAAAAAAAAABmELQCAAAAAAAAAAAAAAAAADMIWgEAAAAAAAAAAAAAAACAGQStAAAAAAAAAAAAAAAAAMAMglYAAAAAAAAAAAAAAAAAYAZBKwAAAAAAAAAAAAAAAAAwg6AVAAAAAAAAAAAAAAAAAJhB0AoAAAAAAAAAAAAAAAAAzCBoBQAAAAAAAAAAAAAAAABmELQCAAAAAAAAAAAAAAAAADMIWgEAAAAAAAAAAAAAAACAGQStAAAAAAAAAAAAAAAAAMAMglYAAAAAAAAAAAAAAAAAYAZBKwAAAAAAAAAAAAAAAAAwg6AVAAAAAAAAAAAAAAAAAJhB0AoAAAAAAAAAAAAAAAAAzCBoBQAAAAAAAAAAAAAAAABmELQCAAAAAAAAAAAAAAAAADMIWgEAAAAAAAAAAAAAAACAGQStAAAAAAAAAAAAAAAAAMAMglYAAAAAAAAAAAAAAAAAYAZBKwAAAAAAAAAAAAAAAAAwg6AVAAAAAAAAAAAAAAAAAJhB0AoAAAAAAAAAAAAAAAAAzCBoBQAAAAAAAAAAAAAAAABmELQCAAAAAAAAAAAAAAAAADMIWgEAAAAAAAAAAAAAAACAGQStAAAAAACZKiEhIbNLMOtlqBEAAAAAAAAA8HxlyewCAAAAAAD/bdbW1pqz4zuFRlzP7FKSld/JXX1rdcnsMgAAAAAAAAAAmYygFQAAAAAg04VGXNeFsMuZXQYAAAAAAAAAACli6UAAAAAAAAAAAAAAAAAAMIOgFQAAAAAAAAAAAAAAAACYQdAKAAAAAAAAAAAAAAAAAMwgaAUAAAAAAAAAAAAAAAAAZhC0AgAAAAAAQLIMCQmZXUKqEl7w+qSXo0YAAAAAAACkTZbMLgAAAOBVkpBgkLW1VWaXkaqXoUYAAPBisLK21rn1c3U/7Gpml5KEU+FS8qjZUnN2fKfQiOuZXU6y8ju5q2+tLpldBgAAAAAAANIJQSsAAIB0ZG1tpVnLd+nKjYjMLiVZHnmc1L99tcwuAwAAvETuh13V/esXM7uMJLK55JUkhUZc14Wwy5lcDQAAAAAAAP4LCFoBAACksys3InT+yp3MLgMAAAAAAAAAAABAOrK29IAzZ86oe/fuKlOmjKpVq6aJEyfq4cOHFvWxcOFCeXt7q2/fvpYODwAAAAAAAAAAAAAAAAAZzqIZrSIiItS1a1e99tprCg4O1vXr1zVhwgTFxMQoKCgoTX3cvHlTs2bNkqur61MVDAAAAAAAAAAAAAAAAAAZzaKg1ffff6979+5p5syZcnZ2liTFx8dr9OjR6tu3r9zd3c32MWnSJAUEBCg0NPSpCgYAAAAAAAAAAAAAAACAjGbR0oE7d+6Un5+fMWQlSQ0bNlRCQoJ27dpl9vi//vpLW7du1bBhwywuFAAAAAAAAAAAAAAAAAAyi0VBq7Nnz8rLy8tkm6Ojo9zc3HT27NlUj42Pj9fYsWP1zjvvKE+ePJZXCgAAAAAAAAAAAAAAAACZxKKlAyMjI+Xo6Jhku5OTkyIiIlI9dtmyZbp//766detmUYGpMRgMio6OTrf+AADAi8dgMMjKyird+nqe9w5WVlayt7d/bv2np/v378tgMGR2GQDA906kq5fpvuFl8DJ9fb7o+P4Bia+p9MbXFZ4V9w0AACCt0vO+AcDLz6Kg1dMKCwvTjBkz9MUXX8jOzi7d+o2NjdXx48fTrT8AAPBiSq/7h+d972Bvb68SJUo8t/7T07lz53T//v3MLgMA+N6JdPey3De8DF6mr88XHd8/IPE1ld74ukJ64L4BAACkVXrmHAC83CwKWjk6OioqKirJ9oiICDk5OaV43Jdffilvb29VqFBBkZGRkqS4uDjFxcUpMjJSDg4OypLF8syXra2tXn/9dYuPAwAAL4/Tp0+nW1/P+97hZfpES+HChfn0N4AXAt87kZ5epvuGl8HL9PX5ouP7ByS+ptIbX1d4Vtw3AACAtErP+wYALz+L0k1eXl46e/asybaoqCjdvHlTXl5eKR537tw5/fnnn6pYsWKSfRUrVtTcuXNVs2ZNS0qR9OjhhIODg8XHAQCAl0d6/jKCe4f/YckSALAc3ztffNw34EXF9w8g/fF1hWfFfQMAAEgrPjQB4HEWBa1q1qyp2bNnKzIyUo6OjpKkkJAQWVtbq1q1aike98EHHxhnskr0+eefK1u2bBo6dKi8vb2fonQAAAAAAADgxeVkn1OGhARZWVtndimpehlqBAAAAAAAeBFYFLRq166dFi9erP79+6tv3766fv26Jk6cqHbt2snd3d3YrmvXrgoNDdWWLVskScWLF0/Sl6OjoxwcHFS5cuVnPAUAAAAAAADgxeNgZy8ra2sd/HqO7oZezexykpUjfz6VebdvZpcBAAAAAADwUrAoaOXk5KRFixZp7Nix6t+/v7Jnz67WrVtryJAhJu0SEhIUHx+froUCAAAAAAAAL6O7oVcVeeFCZpcBAAAAAACAZ2RR0EqSihQpooULF6baZvHixWb7SUsbAAAAAAAAAM9PVicnJSQYZG1tldmlpOhFrw8AAAAAAPx3WBy0AgAAAAAAAPBqyOLgIGtrK/38wz7duhmV2eUkkdstp5q1qZzZZQAAAAAAAEgiaAUAAAAAAAD85926GaXroeGZXQYAAAAAAMALzTqzCwAAAAAAAAAAAAAAAACAFx1BKwAAAAAAAAAAAAAAAAAwg6AVAAAAAAAAAAAAAAAAAJhB0AoA/r+9O4+zsqD7//+eFZBhkUAEdzHFNdfcul1SS61cwC2zXO7MjDQ1c63cN3LHJXeNXDKXLNPUytyyzNBySxNUEBRlH3aYOb8//DlfvKUYBOY6Z+b5fDzuxx2HOfC5Ys7hddHnXBcAAAAAALBIzc2lokdYpEqYEQCoXLVFDwAAAAAAAACUv+rqqtx3518z4f3GokdZqN59umXP/bYsegwAoB2zaAUAAAAAAAC0yoT3GzN+3JSixwAAKIRbBwIAAAAAAAAAACyCRSsAAAAAAAAAAIBFsGgFAADtUHNzqegR/qtynw8AAAAAAOD/qi16AAAAYOmrrq7KfXf+NRPebyx6lI/p3adb9txvy6LHAAAAAAAAWCwWrQAAoJ2a8H5jxo+bUvQYAAAAAAAA7YJbBwIAAAAAAEDBSs3NRY8AAMAiuKIVAAAAAAAAFKyqujrPX31Npo97p+hRFqrPRhtmnX0HFz0GAEChLFoBAAAAAABAGZg+7p1Me+utosdYqK79+hU9AgBA4dw6EAAAAAAAoIw1N5eKHmGRKmFGAABYUq5oBQAAAAAAUMaqq6ty5e1PZex7U4seZaFWWqFHhnx126LHAACAZc6iFQAAAAAAQJkb+97UvDl2ctFjAABAh+bWgQAAAAAAAAAAAItg0QoAAAAAAAAAAGARLFoBAAAAAAAAAAAsgkUrAAAAAAAAAACARbBoBQAAAAAAAAAAsAgWrQAAAAAAAAAAABbBohUAAAAAAAAAAMAiWLQCAAAAAAAAAABYBItWAACwmErNzUWPAAAAAAAAQBurLXoAAACoNFXV1Xn+6msyfdw7RY+yUH022jDr7Du46DEAAAAAAADaFYtWAADwCUwf906mvfVW0WMsVNd+/YoeAQAAAAAK19zcnOrq8r3JU7nPB8DHWbQCAAAAAAAAoN2prq7ONY/9LOOmji96lI/p36Nvjtj+G0WPAcBismgFAAAAAAAAQLs0bur4vDXx7aLHAKCdcB1CAAAAAAAAPrEe3Tqn1Nxc9BiLVAkzAgBQ3lzRCgAAAAAAgE+sa+f6VFVX5437r8usie8UPc5CdflUv6zx5cOLHgMAgApn0QoAAAAAAIAlNmviO5k1fnTRYwAAwDLj1oEAAAAAAAAAAACLYNEKAAAAoADNzaWiRwAAAAAAFoNbBwIAAAAUoLq6Klfe/lTGvje16FEW6jPr9M/+u25c9BgAAAAAUDYsWgEAAAAUZOx7U/Pm2MlFj7FQ/ft0L3oEAAAAACgrbh0IAAAAAAAAAACwCBatAAAAAAAAAAAAFsGiFQAAAAAAAAAAwCJYtAIAAAAAAAAAAFgEi1YAAAAAAAAAAACLYNEKAAAAAAAAAABgESxaQYVobi4VPcIiVcKMAAAAAAAAAACfRG3RAwCtU11dlfvu/GsmvN9Y9CgL1btPt+y535ZFjwEAAAAAAAAAsExYtIIKMuH9xowfN6XoMQAAAAAAAAAAOhy3DgQAAAAAAAAAAFgEi1YAAAAAAAAAAACLYNEKAAAAAAAAAABgESxaAQAAAAAAAAAALELt4j5h5MiROfvss/Pcc8+la9eu2XPPPXPMMcekvr7+Pz7nvffey80335ynnnoqo0ePTrdu3bLFFlvkuOOOy0orrbREBwAAAAAAAAAAALCsLdai1dSpU3PwwQdn9dVXz7BhwzJ+/Picf/75mT17dn784x//x+e99NJLeeSRRzJ48OB85jOfyeTJk3P11Vdn3333zf33359evXot8YEAAAAAANC+1Xbtnubm5lRXl+/NGsp9PgAAAD65xVq0uuOOOzJjxoxcccUV6dmzZ5KkqakpZ5xxRo444oj07dt3oc/bbLPN8uCDD6a29v/9dptuuml22GGH/OpXv8phhx32yY8AAAAAAIAOobbTcqmurs41j/0s46aOL3qcj+nfo2+O2P4bRY8BAADAMrJYi1aPP/54tt5665YlqyTZbbfdctppp+Wpp57KoEGDFvq87t27f+yxFVdcMb169cp77723eBPDMlJqbk6VT5oB0M716Na57P/OK/f5AACA4o2bOj5vTXy76DGAClIJV8QDAKD8Ldai1ahRozJ48OCPPNa9e/f06dMno0aNWqzf+I033sjEiRMzYMCAxXregkqlUmbOnPmJnw8fqqqqSpcuXfL81ddk+rh3ih7nY/pstGHW2Xfwor+wDMyaNSulUqnoMYB2pFQqpaqqaqn9WsuyHT78+6Scde1cn6rq6rxx/3WZNbH8/s7r8ql+WePLh5f13yeV8OdcKcr5z5m2U0mvKd+z5U83QPvVHt6Dve47lvbwPdve6YaOp9yviLfRSutm8GZfLnqMdsP7MEnlvH/6fi1/S7MbgMq3WItW06ZNW+jVqXr06JGpU6e2+tcplUo5++yzs8IKK+RLX/rS4ozwEfPmzcsrr7zyiZ8PH+rSpUvWW2+9TB/3Tqa99VbR43xM1379ih6h1d54443MmjWr6DGAdqa+vn6p/DrLuh0+/PukEsya+E5mjR9d9Bj/UTn/fVJJf87lrpz/nGk7lfSaKvfv2bq6utTWLtZpfpuaP39+5s2bt8x/H90A7VO5vwe3htd9x9Ievmc7At3QMZXrFfH69Vih6BHaFe/DJJXz/un7tTIsrW4AKl8h/wI7bNiw/OUvf8n111+f5ZZb7hP/OnV1dVlrrbWW4mR0VDaQl5411ljD1j2wVL3++utL7dda1u3g75Olp5z/PvHnvPSU858zbaeSXlPl/D1bVVWVTp06lfVtUJqbmzNnzpxl+t+hboD2q5zfg1vL675jaQ/fs+2dboD2zfswSeW8f/p+LX9LsxuAyrdYi1bdu3dPY2Pjxx6fOnVqevTo0apf484778yVV16Zc845J1tvvfXi/PYfU1VVtUSLWsDSVwmXYAUqy9I8GdYOlcPfJx2DP2cqTSV8z953518z4f2Pn7cXrXefbtlzvy2X+X+HugHar0p4D4YF+Z4tf7oB2jfvw1QS36/lr1KW9oC2sViLVmuuuWZGjRr1kccaGxvz/vvvZ80111zk8x955JGcfvrpOfroo7PPPvss3qQAAADAfzXh/caMHzel6DEAAAAAANqlxbqnwHbbbZc///nPmTZtWstjv/vd71JdXZ1tt932vz73r3/9a4477rjsu+++GTJkyCebFgAAAAAAAAAAoACLtWh1wAEHpGvXrhkyZEiefPLJ3H333Rk6dGgOOOCA9O3bt+XrDj744Oyyyy4tPx45cmSGDBmS1VdfPXvuuWeef/75lv8bPXr00jsaAAAAAAAAAACAZWCxbh3Yo0eP3HLLLTnrrLMyZMiQdO3aNfvss0+OPfbYj3xdc3NzmpqaWn78j3/8I42NjWlsbMxXv/rVj3zt3nvvnfPPP38JDgEAAAAAAAAAAGDZWqxFqyQZMGBAbr755v/6NcOHD//IjwcNGpRBgwYt7m8FAAAAAAAAAABQFhbr1oEAAAAAAAAAAAAdkUUrAAAAAAAAAACARbBoBQAAAAAAAMBiKTU3Fz0CALS52qIHAAAAAAAAAKCyVFVX5437r8usie8UPcpC9Vhjg6y03aCixwCgnbFo1U6UmptTVV2+Fygr9/kAAAAAAACAxTNr4juZNX500WMsVOdeKxY9AgDtkEWrdqKcN8a7fKpf1vjy4UWPAQAAAAAAAAAAn5hFq3aknDfGAQAAAAAAAACgkrmXGwAAAAAAAAAAwCJYtAIAAAAAAAAAAFgEi1YAAAAAAAAAAACLYNEKAAAAAAAAAABgESxaAQAAAAAAAAAALIJFKwAAAAAAAAAAgEWwaAUAQNmo7do9zc3NRY8BAAAAAAAAH1Nb9AAAAPCh2k7Lpbq6Otc89rOMmzq+6HEWaqOV1s3gzb5c9BgAAAAAAMtcc3Mp1dVVRY/xX1XCjED7YdEKAICyM27q+Lw18e2ix1iofj1WKHoEAAAAAIA2UV1dlfvu/GsmvN9Y9CgL1btPt+y535ZFjwF0IBatAAAAAAAAAICFmvB+Y8aPm1L0GABlobroAQAAAAAAAAAAAMqdRSsAAAAAAAAAAIBFsGgFAAAAAAAAAACwCBatAAAAAAAAAAAAFsGiFQAAAAAAAAAAwCJYtAIAAAAAAAAAAFgEi1YAFarU3Fz0CP9Vuc8HAAAAAAAAAIujtugBKkFzcynV1VVFj1Gxart2T3Nzc6qr7fXB0lRVXZ037r8usya+U/QoH9PlU/2yxpcPL3oMAAAAABai3P/Nu9znAwAAOi6LVq1QXV2VK29/KmPfm1r0KAv1mXX6Z/9dNy56jP+ottNyqa6uzjWP/Szjpo4vepyF2mildTN4sy8XPQYstlkT38ms8aOLHgMAKHOl5uZU+eADAAD/v+rqqtx3518z4f3Gokf5mN59umXP/bYsegwAAICFsmjVSmPfm5o3x04ueoyF6t+ne9EjtMq4qePz1sS3ix5jofr1WKHoEQAAYJkp5yth9lhjg6y03aCixwAA6HAmvN+Y8eOmFD0GAABARbFoBQAA0AGU65UwO/dasegRAAAAAACgVdw7AgAAAAAAAKCMNDeXih4BAFgIV7QCAAAAAAAAKCPV1VW58vanMva9qUWPslCfWad/9t9146LHAIA2Z9EKAAAAAAAAoMyMfW9q3hw7uegxFqp/n+5FjwAAhXDrQAAAAAAAAAAAgEWwaAUAAAAAAAAAALAIFq0AAAAAAAAAAAAWwaIVAAAAAAAAAADAIli0AgAAAAAAAAAAWASLVgAAAPBf9OjSLaXm5qLHAAAqgG4AAFpLNwBUptqiBwAAAIBytlx9l1RVV+f5q6/J9HHvFD3OQvXZaMOss+/goscAgA5PNwAAraUbACqTRSsAAABohenj3sm0t94qeoyF6tqvX9EjAAAL0A0AQGvpBoDK4taBAAAAAAAAAAAAi2DRCoClrrZr9zRXwH3FK2FGAAAAAAAAAMqDWwcCsNTVdlou1dXVueaxn2Xc1PFFj7NQ/Xv0zRHbf6PoMQAAAAAAAACoEBatABaiubmU6uqqoseoeOOmjs9bE98uegwAWKZ0AwAAAAAAdAwWrQAWorq6Klfe/lTGvje16FEW6jPr9M/+u25c9BgAQHQDAAAAAAB0FBatAP6Dse9NzZtjJxc9xkL179O96BEAgAXoBgAAAAAAaP+qix4AAAAAAAAAAACg3Fm0AgAAAAAAAAAAWASLVgAAAAAAAAAAAItg0QoAAAAAAAAAAGARLFoBAAAAAAAAAAAsgkUrAAAAAAAAAACARVjsRauRI0fm0EMPzcYbb5xtt902Q4cOzdy5cxf5vFKplGuvvTY77LBDNtpoo+y///55/vnnP8nMAAAAAAAAAAAAbWqxFq2mTp2agw8+OPPmzcuwYcNy7LHH5s4778z555+/yOded911ufzyy3PIIYfkmmuuSZ8+fXLYYYdlzJgxn3h4AAAAAAAAAACAtlC7OF98xx13ZMaMGbniiivSs2fPJElTU1POOOOMHHHEEenbt+9Cnzdnzpxcc801Oeyww3LIIYckSTbbbLPsuuuuueGGG3L66acvyTEAAAAAAAAAAAAsU4t1RavHH388W2+9dcuSVZLstttuaW5uzlNPPfUfnzdixIhMnz49u+22W8tj9fX12WWXXfL4448v/tQAAAAAAAAAAABtaLEWrUaNGpU111zzI4917949ffr0yahRo/7r85J87LkDBgzIuHHjMnv27MUZAwA6jObmUtEj/FflPh8AAAAAAADA0lJVKpVa/b+Qrr/++vne976Xb33rWx95/Mtf/nI22WSTnHXWWQt93tVXX52rrroqL7zwwkce/93vfpfvfe97efzxx//jbQf/kxEjRqRUKqWurm6xnvdJVFVVZdr02Wlqbl7mv9cnUV9Xm65d6jN/ZmNKzU1Fj/Mx1bV1qencNY2zp2d+Gc6XJPW1delav1zmTmtMc9P8osf5mJr6+tR17ZqZM+akqak8vw9raqqzXNdOWYy3lLLmdb9kKuF1X1dTm4ZOXYseo1Vmz56X5jL8Xqyurk7nznXL/HU/b968VFVVZdNNN12iX6et2sH7x5KphPePcu+GpPzbQTe0La/7Jed1v+Ta6nWvG5Yu7x9LzvvHktMNbavcX/dJ+b/2ve6XnG5YOO8fS877x5Ir9/ePpH21g9f9kvO6X3Je9x9YWt0AtA+1RQ/wSVVVVX3k/y9r3Rs6t8nvsyRql+tW9Aj/VbfODUWPsEj13cv7v8PlunYqeoRFaqvXZFvwul9ylfC6rwSdOy/7peIlsaxf91VVVUvl92jLdvD+seQq4f2j3LshKf920A1ty+t+yXndLznd8HHeP5ac94+lo6O/f7Qlr/ulo9xf+173S043fJz3j6XD+8eSK/f3j6T9tIPX/dLhdb/kOvrrfml1A9A+LNaiVffu3dPY2Pixx6dOnZoePXr81+fNnTs3c+bMSadO/+9NeNq0aamqqvqvz/1PNtlkk8V+DgDQcWkHAKC1dAMA0Fq6AQAAOpbqxfniNddcM6NGjfrIY42NjXn//fez5ppr/tfnJckbb7zxkcdHjRqV/v37p3Pn8t/GBgAAAAAAAAAAOq7FWrTabrvt8uc//znTpk1reex3v/tdqqurs+222/7H52266aZpaGjIgw8+2PLYvHnz8vDDD2e77bb7BGMDAAAAAAAAAAC0ncW6deABBxyQ4cOHZ8iQITniiCMyfvz4DB06NAcccED69u3b8nUHH3xwxo0bl0ceeSRJ0qlTpxxxxBEZNmxYevXqlbXXXju33357pkyZkv/93/9dukcEAAAAAAAAAACwlC3WolWPHj1yyy235KyzzsqQIUPStWvX7LPPPjn22GM/8nXNzc1pamr6yGOHH354SqVSbrzxxkyaNCnrrrtubrjhhqyyyipLfhQAAAAAAAAAAADLUFWpVCoVPQQAAAAAAAAAAEA5qy56AAAAAAAAAAAAgHJn0QoAAAAAAAAAAGARLFoBAAAAAAAAAAAsgkUrAAAAAAAAAACARbBoBQAAAAAAAAAAsAgWrQAAAAAAAAAAABbBohUAAAAAAAAAAMAiWLSCMlUqlYoeAQCoELoBAGgt3QAAtJZuAAD4OItWUIZmz56dH/7wh/nHP/5R9CgAQJnTDQBAa+kGAKC1dAMAwMJZtIIy9NRTT+WJJ57IxRdfnJdeeqnocQCAMqYbAIDW0g0AQGvpBgCAhasque4nlKVf/epXufnmm9OtW7eceOKJ2WCDDYoeCQAoU7oBAGgt3QAAtJZuAAD4OFe0gjIzd+7cJMlee+2VAw88MDNmzMgFF1yQV199teDJoONobm7+2GNNTU0FTALw3+kGKJ5uACqFboDyoB2ASqAboDzoBoDyZNEKykhzc3Pq6+uTJJdffnlefPHFvPPOO/nb3/6Ws846K6+88krBE0L719TUlOrqD/56HDt2bEaNGpX58+enpqam4MkAPko3QPF0A1ApdAOUB+0AVALdAOVBNwCUL7cOhDJ0wgkn5G9/+1uGDBmS/v3754knnshvf/vbrLbaajn55JOz3nrrFT0itEvNzc0tJy4nn3xy/v73v2fSpEnp1q1bDj/88Gy//fZZaaWVCp4S4KN0AxRDNwCVSDdAcbQDUGl0AxRHNwCUN4tWUGbGjBmTAw44IEceeWQOOuiglsfvvPPOXH/99VlhhRXy4x//OGuvvXaBU0L7dvLJJ+evf/1rhgwZkn79+uXpp5/OL3/5y2yzzTY55ZRT0rt376JHBEiiG6Ac6AagUugGKA/aAagEugHKg24AKE9uHQhlZubMmZk4cWJWXnnlJMmcOXOSJPvtt1/22GOPPPvsszn33HPzz3/+s8gxod0aNWpUnnvuuRxzzDH5yle+km222SaHHnpopkyZkhVWWCHdu3cvekSAFroBiqUbgEqiG6B42gGoFLoBiqcbAMqXRSsoE83NzUmSddZZJ6uttlruueeeJEmnTp1aTmIOPfTQ9OvXL6+//nquuuqqzJ07t7B5ob0aP3583nrrray//vqpr6/P66+/nt122y1f/OIX873vfS/19fV5+eWXW16XAEXQDVAedANQCXQDlA/tAJQ73QDlQzcAlC+LVlCQpqamj/z4w3stz58/P/vss09efPHFXHnllUk+OIlJkn//+99ZeeWV881vfjM//OEPU19f37ZDQzvz4T8cLKhPnz7p2bNnxo0blzfffDMHHnhgtt5665x77rnp0qVL7r///tx0002ZMmVK2w8MdFi6AYqnG4BKoRugPGgHoBLoBigPugGgstQWPQB0RE1NTampqUmS3HzzzXn77bcze/bsfPGLX8z//M//ZL/99ssrr7ySe++9N2PHjs0xxxyTMWPG5N57701VVVX233//dOnSpeCjgMq24OvwmWeeyZprrpnevXund+/eWX755XPppZfm7bffzlZbbZULL7wwNTU1mTx5cp588slMnz7daxBoM7oBiqcbgEqhG6A8aAegEugGKA+6AaDyVJVKpVLRQ0BHdfTRR+eFF15Iv379Ul1dnWeffTYHHHBAvvWtb6Vz58655ZZbcvfdd2fChAlpaGhIp06dct1112W99dYrenSoaAueuJx88sl54403suOOO+aQQw5Jp06d8s9//jOHH3545s+fn8svvzzbbrttXn311dxyyy159NFHM3z48Ky11loFHwXQ0egGKIZuACqRboDiaAeg0ugGKI5uAKhMFq2gILfccktuvPHGXHLJJVl77bXT0NCQK664IldccUXOP//87LXXXpk9e3ZmzpyZJ598Mj169MinP/3p9O/fv+jRod047rjj8vzzz+eUU07JhhtumL59+6a5ubnlHxSOPvrodO/ePTNmzEifPn3S2NiYYcOGZeDAgUWPDnQwugGKpxuASqEboDxoB6AS6AYoD7oBoLK4dSAUZOTIkdlggw2y4YYbpq6uLm+88UbuuOOO7L777tl1112TJHV1denVq1f22GOPgqeF9uf+++/P3//+91x00UXZbLPNUlVVlenTp2fSpEmpqqrK5ptvnrvvvjsvvPBCRo0alXXXXTfrrLNOVlxxxaJHBzog3QDF0g1AJdENUDztAFQK3QDF0w0AlceiFbSBUqmUqqqqj/x4ypQpmTFjRurq6jJmzJjsv//+2XbbbXP22Wenc+fOuf7667PBBhtkq622KnByaL8mTZqUzp07Z80110xzc3NGjBiRM888M1OnTs3EiRNzzjnnZK+99kq/fv2KHhXoYHQDlB/dAJQr3QDlSTsA5Ug3QHnSDQCVx6IVLGML3l95+vTpaWhoSFVVVVZeeeU89NBD+cMf/pBTTjkl22yzTc4666wst9xyefvtt/PUU0+lqqoqn/3sZ1NdXV3wUUD707t377z11lu56KKLMnXq1Dz22GPZaaedsu2222bEiBE566yz8rnPfS69e/cuelSgA9ENUJ50A1COdAOUL+0AlBvdAOVLNwBUHotWsAwtePJy5ZVXpqmpKbvvvnvWWmutfPvb387vf//7DBkyJDvssEPOP//8dO7cORMmTMhVV12Vd955J1/84hedvMASWvB1uKDdd989Y8aMyb333pt11lknP/7xj7Pvvvsm+eDTXM8///xCnwewrOgGKJ5uACqFboDyoB2ASqAboDzoBoD2o6pUKpWKHgLaowUvw3vMMcfkhRdeyNe//vXstttu6du3b5Lk2WefzY9+9KMkyQEHHJDZs2fn+eefz9///vf87Gc/y8CBAwubH9qDBU9c7r///rz//vupqqrK9ttvnzXWWCNJ0tjYmE6dOqW+vj5JMmXKlJx//vkZN25crrrqqjQ0NBQ2P9Bx6AYonm4AKoVugPKgHYBKoBugPOgGgPbFohUsY+ecc07++Mc/5uKLL85aa62Vrl27prm5ueUTIG+//XbOOOOMTJo0KXPnzs16662Xww8/PGuttVbBk0NlW/B19v3vfz8jRoxInz598tZbb2X11VfP/vvvn0GDBn3kOU888UQefPDBPPLII7n11luz9tprFzE60IHpBiiGbgAqkW6A4mgHoNLoBiiObgBof9w6EJah6dOn55VXXsmee+6Zz3zmM0mSMWPG5LbbbsuUKVOy2WabZZ999sl1112X6dOnp7a2NjU1Namrqyt4cqh8H564nH766Xn++edz4YUXZrPNNstll12Wq6++OjNnzkxVVVX23nvvJMlDDz2USy65JF27ds3Pf/5zJy5Am9MNUBzdAFQa3QDF0g5AJdENUCzdAND+WLSCZaipqSmjRo3Keuutl9deey1/+ctfcskll6Rv376pq6vLvffem9ra2uy1114u+QnLwIgRIzJq1Kicdtpp2WyzzfLTn/401157bU499dT8/Oc/z2WXXZYk2XvvvbPTTjulW7duGTBgQMtlswHakm6AYukGoJLoBiiedgAqhW6A4ukGgPbFrQNhKVnw0p8L3vf8jjvuyOmnn54ePXqkU6dOGTRoUI488shMmDAhQ4YMyZZbbpmTTz65yNGh3Zo4cWJuueWWHHHEEXnsscdy2mmn5eSTT86gQYPyyiuv5MADD8zaa6+dPfbYI1/72teKHhfoQHQDlB/dAJQr3QDlSTsA5Ug3QHnSDQDtiytawVLQ1NSUmpqaJMnUqVPT2NiYlVdeOUlywAEHZODAgZk2bVp69eqVDTbYIEnSuXPn1NXVpXfv3oXNDe3Jwv4R4VOf+lSGDBmSTp065Y9//GO23HLL7LTTTkmSfv36Zfnll8+rr76a3/zmN9ljjz3SrVu3Ig8B6CB0AxRPNwCVQjdAedAOQCXQDVAedANA+2fRCpbQgicv55xzTv72t79lzJgxWXPNNXP00Udns802y8Ybb/yR54wdOzZXX3113nvvvey6664FTA3ty9y5c1NfX58kefvttzNnzpwMGDAgzc3N6dSpU2bMmJHXXnstq622Wnr06JEkmTx5cjbeeOMceuihWX755Z24AG1CN0DxdANQKXQDlAftAFQC3QDlQTcAdAzVRQ8Ale7Dk5fjjjsuf/rTn7LvvvvmzjvvzNixY3PxxRfn4Ycfzpw5c1q+/tprr83JJ5+cJ554Itdcc01WWWWVokaHijZ79uz84Q9/SJKWE5cTTzwxBxxwQL785S9n9913z80335z3338/Xbt2zZe+9KU88sgjufPOO/OXv/wl119/fV5++eWsscYaLZ/sAljWdAMUQzcAlUg3QHG0A1BpdAMURzcAdDyuaAVLwc0335yRI0dm6NCh2WSTTfKzn/0sU6dOzXLLLZcLLrggSbLrrrumqakpXbp0yeqrr54zzzwzq6++erGDQ4UqlUo588wz8/jjj2fatGnZe++9c9lll+WZZ57Jt771rayyyiq55557cscdd+TFF1/Mqaeemr333jtvvvlmfvzjH6dnz56pq6vLddddl4aGhqIPB+hgdAO0Ld0AVDLdAG1POwCVSjdA29MNAB1TValUKhU9BFSyUqmUK664IjNmzMhJJ52U4cOHZ+jQobn00kuz7bbbZvDgwZkzZ06++93vZvfdd099fX3mzJmTTp06FT06VLTnnnsuF198caZMmZKDDjooL7/8cjbaaKMMHjy45WuuvPLK/PrXv86XvvSlHHXUUZk5c2ZGjhyZ6dOnZ8CAAenbt2+BRwB0RLoBiqEbgEqkG6A42gGoNLoBiqMbADoei1awmBa81/mH91qeN29e3n///VRVVeVrX/tavva1r+XAAw9Mly5dcsMNN+QnP/lJunXrljPOOCO77757wUcAla9UKqWqqiovvPBCzj///MyYMSPjx4/PxRdfnK233voj/0hwyimn5PHHH89vf/vblnueA7QV3QDF0w1ApdANUB60A1AJdAOUB90A0DFVFz0AVJIFT15uueWW3HfffRkzZkzq6urSv3//vPHGG5k6dWq22WabdOnSJUlSXV2dPffcMxtvvHEGDhxY5PjQblRVVSVJNtxww5x44onp2rVrJk+enCeffDJJ0qlTp8yZMydJ8r3vfS8TJkzIU089Vdi8QMekG6A86AagEugGKB/aASh3ugHKh24A6JgsWkErlUqllpOXY445JrfccksmTJiQzp07t3xNr1690tTU1BJJEyZMyIsvvpiBAwfmuuuuy5prrlnI7NCebbTRRvnRj36UTTfdNPfdd19+8YtfJEnLp0TGjBmThoaGdO/evcgxgQ5GN0B50g1AOdINUL60A1BudAOUL90A0HHUFj0AVIoPt9IvuOCCPP/887nooosycODAdO3ateVr+vfvnz322CM//elP88ADD6S+vj5vvPFGjjzyyKLGhnavVCpl4MCBOfXUU3Peeefl2muvzYQJE3LooYfm3//+d371q1+lc+fOGTBgQNGjAh2IboDypBuAcqQboHxpB6Dc6AYoX7oBoOOoKpVKpaKHgEoxY8aMHH744fnMZz6TY489NvX19S0/9+F9mMeMGZNHH300Tz75ZHr16pVvfvObWWuttQqcGtqvDy+TPXfu3NTX1+fFF1/Meeedl7///e/p169fVl555STJySefnPXWW6/gaYGORjdAedENQDnTDVB+tANQrnQDlB/dANCxWLSCxfDuu+/mC1/4Qk444YQcdNBBLSctC2psbEy3bt2SpCWogKXvwxOXUaNG5bjjjsuwYcOyyiqr5F//+lfOPffcjBw5Mocffnj233//dOnSpehxgQ5IN0D50A1AudMNUF60A1DOdAOUF90A0PFUFz0AlKuF7SA2NDRkwIABeeaZZzJ58uRUVVV95Ov+8Ic/5N5778306dOTxMkLLKHm5uaFPj5//vzU1NRk9OjROeigg7LCCitk+eWXT5IMHDgwxx13XDbYYIPsvPPOTlyANqEboHi6AagUugHKg3YAKoFugPKgGwBYkEUrWIimpqaWT4AsGE8NDQ3Zaqut8uijj+ZPf/pTGhsbW75u8uTJue+++/Liiy+mtra2kLmhPWlqakp19Qd/Tb377rt58803M378+CRJbW1tJkyYkEGDBmXrrbfOxRdfnIaGhiQf/OPDxhtvnMsvv7zlcrwAy5JugOLpBqBS6AYoD9oBqAS6AcqDbgDg/3LrQPg/mpubW4Lp2muvzauvvprVV189W2yxRbbaaqskyXe+85089dRTOeCAA7Lrrrtm4sSJefDBB/Pkk0/m1ltvda9zWEIfXmo3SX784x/nlVdeyeuvv571118/e+yxR/bbb78kyZ133pndd9+95cRlQQu7ZDbA0qYboHi6AagUugHKg3YAKoFugPKgGwBYGItWsIAFY+f73/9+nnzyyay55pp59dVXs8oqq2T//ffPgQcemCT50Y9+lEcffTQTJkxInz59svzyy2fo0KEZOHBgkYcA7crxxx+fZ599Nj/4wQ/yqU99Kj/5yU/y0ksv5bjjjsu3vvWtJB/9RweAtqQboLzoBqCc6QYoP9oBKFe6AcqPbgBgQa4bCv+/BU9eRo4cmffeey/Dhg3LZz/72bz66qu56KKLMnz48DQ1NeXrX/96zjrrrIwcOTKTJ0/O8ssvn169erXcdxlYcrfffnv+9a9/5aKLLspmm22W4cOH57XXXsuWW26ZSy+9NPX19TnkkENSXV39kU+VALQF3QDlRTcA5Uw3QPnRDkC50g1QfnQDAP+XRSv4/3148vKjH/0ozc3N6dGjRzbYYIMkyTrrrJPjjjsul156aW699daUSqV84xvfyIABA4ocGdqVuXPnpr6+vuXHtbW12WuvvVpOXC644IJcfPHFWXPNNXPKKafk/PPPT11dXb72ta85cQHanG6AYukGoJLoBiiedgAqhW6A4ukGABbF9Qvp8Ba8e+b06dMzZ86c3H333Rk7dmxmzZrV8nMDBw7Msccem9VXXz2//OUvc8MNNxQxLrQrTU1NefbZZzNjxoyWE5fbbrstkydPztZbb5299tor48aNy2233ZZjjjkmO+ywQ9Zaa60ceOCBqa+vz1lnnZWbb7652IMAOhTdAMXRDUCl0Q1QLO0AVBLdAMXSDQAsDotWdGgLXoY3SRoaGnL88cfnf//3f/PKK6/kt7/9bZqbm1t+/sNPjPTs2TO/+93vMnXq1CLGhnZj5MiRufHGG3PCCSckSY488sjcdtttmT17dlZeeeX07t07EyZMyMSJE7Pqqqu2nOBMnjw5m2yySY444oj8z//8T5GHAHQgugGKpRuASqIboHjaAagUugGKpxsAWBxuHUiH9uHJy0knnZTll18+J554YlZYYYUceuihmTlzZi644IJ06dIl++yzT8vXrr322jnttNPStWvX9OjRo8jxoeKtssoqWX/99XPPPfdkhx12SKlUyk033ZS+ffu2fM28efMyZ86cvP3225k0aVJqa2vz2muvZdNNN80RRxyRzp07F3gEQEeiG6BYugGoJLoBiqcdgEqhG6B4ugGAxWHRig5v2rRpqa+vz0033ZSGhoYMGTIkvXv3zne/+92USqWcfvrpKZVK2XfffVtOYtZaa62Cp4bKN2/evHTp0iVDhgzJn/70p7zwwgvZaqutstJKK6W6ujrz589PbW1tNttssxx44IEZOnRo7rvvvtTU1GT06NG54447nLgAbU43QDF0A1CJdAMURzsAlUY3QHF0AwCLq6q04I2foQP4v5fhTZLx48dn+PDhuf766/Pd73433/3ud5MkEydOzLBhw/KrX/0q3//+93PQQQd97LnA4mtubk519Qd3rx06dGjmzp2bUqmUP/3pT1l//fVbPqU1d+7clkvwDh8+PM8//3w6d+6cww47LAMGDCjyEIAOQjdA8XQDUCl0A5QH7QBUAt0A5UE3APBJWLSiQ1nw5GXq1KkfuaTu+PHj87Of/Sw33HDDR05iJk2alPPPPz9PPPFEHnrooXTv3r2Q2aG9WPB1eMopp+SPf/xjbrzxxgwYMCDXX3997rnnnqy33nq54IILstxyy2XevHmpqalJqVRKTU3NR058AJYl3QDF0w1ApdANUB60A1AJdAOUB90AwCdl0YoO6aSTTkpdXV2GDBmSFVdcseXx8ePH5+abb85NN92U448/Pt/85jeTfHAS09TUlD59+hQ1MrQLC564vPTSS7n44otzyCGHZIsttkjnzp0ze/bsXH/99bn33nuz/vrrZ+jQoamvr8/QoUMzf/78nHjiiamtrfWJLaBN6QYohm4AKpFugOJoB6DS6AYojm4AYEnUFj0AFKFnz565+eab061bt3zjG99oOYnp27dvDjjggPzxj3/MhRdemMbGxhx77LHp1atXwRND+7Dgp0Pee++9NDY2ZuONN07nzp0zb968dO7cOd/85jdTVVWVu+++O3vuuWfWWmutPPbYY7nrrrtSV1dX8BEAHZFugGLoBqAS6QYojnYAKo1ugOLoBgCWhEUrOpQPN9RPOumkNDQ05IorrkipVMrBBx/cchKz2mqrZcMNN0ynTp1yxx135OCDD87yyy9vKx2Woj59+uSee+5JbW1t3nrrrWywwQapq6tLU1NTywnMCiuskCeeeCJNTU2599578+lPf7rosYEORjdAedANQCXQDVA+tANQ7nQDlA/dAMAn4daBdDgL3jN52LBhufLKK3PooYfmoIMOykorrZTx48fnnHPOye67756tt976I/dHBxbfgpfgnT9/fmprP9jxHT58eM4555zsueeeOeqoo7LyyisnSZqamlruc15VVZU5c+akU6dOhc0PdGy6AdqWbgAqmW6AtqcdgEqlG6Dt6QYAlhZXtKLDWfAE5qijjkqpVMq1116bN954I+uss05Gjx6dF154IaeeeqqTF1hCH56IJMncuXMzc+bM9OzZM0ny9a9/PTNmzMill16a7t2755BDDslKK62Umpqaj7xOnbgARdIN0HZ0A1DpdAO0Le0AVDLdAG1LNwCwNFm0okNpampKbW1tRo0alauuuioXXnhhjj766PTp0yf33Xdf7rvvvvTp0ydXX311+vbtW/S4UNEWPHE5++yz8+KLL+aNN97Idtttl5133jlf/OIX8+1vfztJcumll6ZUKuWwww5L//79W05cAIqkG6Dt6Aag0ukGaFvaAahkugHalm4AYGlz60DanQW3yxf04WVAR48enf322y/rr79+LrvssjQ0NCRJJk6cmKqqqtTU1PiECCxFxx13XJ577rnsueee6dmzZx566KFMnz49n//853PssccmSa699tpccsklGTRoUI466qisuOKKBU8NdBS6AcqLbgDKmW6A8qMdgHKlG6D86AYAlhZXtKJdWXArfdSoUamrq0ttbW369euX2travPPOO/nCF76QPfbYIz/+8Y9bTl6SpFevXi33ZgaWjqeffjrPPfdczjrrrGyxxRbp1KlT1l133Rx88MHZfvvtM3v27HTu3Dnf+ta3MmPGjAwfPrzlhAZgWdMNUF50A1DOdAOUH+0AlCvdAOVHNwCwNLmiFe3Ggp8QOfnkk/Pcc89l8uTJ6dKlSw455JAMHjw4VVVVufLKK3PkkUeme/fuBU8M7d+9996bCy+8ML/97W/Ts2fPjBw5MgcddFC22mqrnHvuuenSpUtee+21rL322kmSyZMnZ/nlly94aqAj0A1QfnQDUK50A5Qn7QCUI90A5Uk3ALA0uaIV7caCJy9PP/10jj/++FRXV2fy5Mk566yz8uqrr+a0007LiSeeWPCk0HF06dIljY2Nqaqqyvjx4/PVr3412267bc4+++x06dIl99xzT0aPHp0VVlghPXv2TM+ePYseGeggdAOUH90AlCvdAOVJOwDlSDdAedINACxNFq1oV15//fW8+OKLOeGEE7LLLrukrq4u48ePz1lnnZXOnTsv9J7owJJb8JNaC1pxxRXTu3fvnH/++fnjH/+Yz33ucznzzDPTtWvXTJw4MX/605/S0NCQTp06JYnLYgNtSjdAMXQDUIl0AxRHOwCVRjdAcXQDAG1BzdGuTJkyJaNGjUqfPn1SV1eXN998M1/5yley22675cQTT0x9fX1effXVoseEdqWpqanlxGXEiBEZMWJEXnvttSTJxhtvnN133z333ntvVlhhhQwZMiQNDQ0ZM2ZMLr744owYMSLf/OY306VLlyIPAeigdAO0Pd0AVCrdAMXQDkAl0g1QDN0AQFtxRSsq1sK20qurq9O1a9d07tw5Y8aMyX777Zdtttmm5ZMiv/vd7/Kzn/0sF110Ufr161fQ5FD5Zs6cmcceeyy77LJLams/+KvkmGOOybPPPpsJEyakd+/e+exnP5vzzjsvxx9/fEqlUu6+++6ccMIJ6datW2bOnJl333031113XdZcc82CjwboCHQDFEc3AJVGN0CxtANQSXQDFEs3AFCEqlKpVCp6CFhcTU1NqampSZL8/e9/zworrJBVVlklSbLffvtl6tSpmTRpUrbbbruce+656dSpUyZOnJgLL7wwkydPzgUXXJAePXoUeQhQ0a677rpcdNFFOe200/LVr341V155Ze655558//vfT7du3fLqq6/mpptuykorrZTrrrsuPXr0yG9/+9uMHj06I0eOzMYbb5ztt9++5XULsCzpBiiWbgAqiW6A4mkHoFLoBiiebgCgCBatqDgLnrycfPLJef311zN48OB8+ctfTkNDQ5599tmcd955GTlyZG677basueaaGTNmTG644YY89thjGT58eNZaa62CjwIq2+TJk3P11Vfn1ltvzbnnnptRo0ale/fuOfTQQ1NdXZ158+bl5ZdfzlFHHZX11lsvP/3pT4seGeigdAMUTzcAlUI3QHnQDkAl0A1QHnQDAEWwaEXFOu644/Liiy/mmGOOyRZbbJE+ffokSebMmZO//e1vOffcczN58uR07do13bt3T2NjY4YNG5aBAwcWPDm0D1OnTs1ll12W22+/PTU1NTnppJNy0EEHtfx8c3Nzfv3rX+eMM87IxRdfnB133LHAaYGOTjdAsXQDUEl0AxRPOwCVQjdA8XQDAG2ttugB4JN4+OGH88wzz+TCCy/MFltskZqamkybNi0TJ05Mc3NzPve5z+WBBx7IXXfdlZkzZ2aVVVbJeuutl759+xY9OlS8UqmUqqqq9OjRI0cddVS6deuWa665JiNGjMhXvvKVlstdV1dXZ/PNN0+SjB8/vsiRgQ5ON0BxdANQaXQDFEs7AJVEN0CxdAMARbFoRUV69913U1NTk6222ipz5szJc889lzPOOCNTpkzJnDlzcsghh+Q73/lO9tlnn6JHhXajubk51dXVqaqqanls+eWXz8EHH5w5c+bklltuyWc+85nsu+++WW655ZIk8+fPT8+ePdPU1JTk/534ALQl3QBtTzcAlUo3QDG0A1CJdAMUQzcAUDSLVlSUD8NnvfXWy/jx43PUUUelVCrlT3/6U77whS9k6623zj/+8Y/cdddd2W233bL66qsLJVgKmpqaUlNTkyT529/+lpkzZyZJtt9++/Tq1StHH3105s6dmwsuuCBjxozJ5z//+STJb37zm0yfPj3bbbddkng9Am1KN0AxdANQiXQDFEc7AJVGN0BxdAMA5cCiFWVtwWBK/l/4DBw4MGeeeWZ+/vOf59Of/nROP/30lk+FlEqlPPPMM+nWrZtQgqWgubm55XX4/e9/PyNGjMj777+fJNl8881z+OGHZ5tttskPfvCD1NTUZPjw4bn99tuz0047Zfr06Rk+fHhWWWWVIg8B6CB0AxRPNwCVQjdAedAOQCXQDVAedAMA5aKqVCqVih4CFmbBk5df/epXef/99zNt2rQccsgh+dSnPpUkmTVrVkqlUsulPydNmpSf/OQnGTt2bK644op07969sPmhvTn99NPz6KOP5vTTT0/Pnj3T2NiY0047LfX19TnppJOy4447ZvLkyRk+fHiuuuqqnHvuudl9993TuXPnokcHOgDdAOVFNwDlTDdA+dEOQLnSDVB+dAMARXNFK8rSglvpxx13XF566aX06NEjM2fOzMMPP5yTTjopW2+9dbp06dLynEcffTQPP/xwfv/73+fnP/+5kxdYQvPnz09tbW2am5szadKk/Otf/8p+++2X7bbbruX1effdd2f//ffPVVddlS222CLLL798DjzwwMyePTuf+cxnnLgAbUI3QPF0A1ApdAOUB+0AVALdAOVBNwBQbqqLHgAWprr6g2/NM844I88//3zOO++83HnnnfniF7+Yt956K2eccUaefPLJzJ07N0ny4IMP5sYbb8xrr72Wn//851lnnXWKHB/ahdraD3Zxr7766vztb3/Lv/71r9TX17ecuMybNy+9evXKpZdemldeeSX3339/kqR37975wQ9+kAEDBhQ2O9Cx6AYonm4AKoVugPKgHYBKoBugPOgGAMqNRSvK1j/+8Y+89NJLOe2007LpppvmuuuuyzXXXJNzzjknq6yySs4888w8/vjjaW5uzqabbprvfve7ufrqq528wBJqampq+c/nnHNObrrppvTs2TMDBgzIs88+m/HjxydJ6urqkiQrr7xyPvWpT+Wdd95peV5VVVXbDg10eLoBiqEbgEqkG6A42gGoNLoBiqMbAChXFq0oW+uss0522223bLzxxnnggQdy9dVX5+yzz87gwYNzzDHHZPr06bn88svzu9/9Lp/61Key5ZZbZoUVVih6bKh4H34K5IUXXkhdXV3OOOOMbL311tlrr73y2GOP5de//nUmTpzY8vXTp09P165d06tXryRJqVQqZG6gY9MNUAzdAFQi3QDF0Q5ApdENUBzdAEC5qi16AEg+uNf5h5fh/VDnzp1z0EEHpa6uLn/605+y3XbbZeedd07ywcnNSiutlFGjRuXiiy/O9ttv33LpUGDJXXfddbnqqqtSXV2dq666Kkny9a9/PaNHj85FF12UN998M7vuumtqa2tzzz33ZMqUKfn85z+fxCdEgGVPN0B50Q1AOdMNUH60A1CudAOUH90AQDlSfBRu7ty5qa+vT5KMHDky1dXV6dGjR3r16pW6urqUSqW8/fbbqa6uTkNDQ5LknXfeyaqrrpphw4alS5cu6dq1a5GHAO3O5z//+Tz99NP5y1/+krfeeitbbrllkuTUU09Nr169cscdd+Tuu+/OCiuskC5duuTGG2/MKqusUvDUQEegG6D86AagXOkGKE/aAShHugHKk24AoBxVlVw3kQLMmjUrI0aMyOabb55OnTolSX7wgx/k6aefzqxZs9LQ0JBvf/vb2XHHHbPiiivmmmuuyS9+8Yvstdde2WqrrXLvvffmxRdfzC233NJyCVBg6RozZkyOP/74jBo1KkOHDs2OO+7Y8nNvvfVWJkyYkPr6+qy00kpeh8AypRug/OkGoFzoBqgM2gEoB7oBKoNuAKDcWLSiEN/73vfy+9//PhdddFF23XXXXHnllbnrrrty5JFHpr6+Ps8991x+8YtfZJ999slxxx2XefPmZejQoXniiSeSJF27ds3VV1+dgQMHFnwk0L6NGTMmp5xySt57772cfPLJ2WGHHZJ8cG9zl90F2opugMqgG4ByoBugcmgHoGi6ASqHbgCgnFi0ohBz5szJAQcckIkTJ+aUU07JI488kk022SQHHXRQy9fcfvvtOfPMM3P00UfnyCOPzKRJkzJ27NhMnjw566yzTvr27VvgEUDHMXr06Jx66qkZP358Tj311Gy//fZFjwR0MLoBKoduAIqmG6CyaAegSLoBKotuAKBcWLSizc2fPz+1tbWZO3duBg0alClTpqRLly4566yzstVWW2XevHmpq6tLklxyySW55ZZbcu+992aNNdYoeHLouEaPHp3TTjstL7/8ci666KJ87nOfK3okoIPQDVB5dANQFN0AlUk7AEXQDVCZdAMA5aC66AHoeGprazN//vzU19fnnnvuSf/+/TNmzJg8+OCDmTVrVurq6jJv3rwkye67757q6uq88sorBU8NHduqq66aH//4x9lkk02yyiqrFD0O0IHoBqg8ugEoim6AyqQdgCLoBqhMugGAcmDRikIseBLz85//PBtuuGEeeOCBPPzwwy0nMckHl+6tra1NbW1twRMDa6yxRoYNG5bVVlut6FGADkY3QOXRDUBRdANUJu0AFEE3QGXSDQAUzaIVhamtrc28efNSX1+fW2+9NSuuuGLOP//8XH/99ZkyZUr+/ve/5xe/+EXq6+uzwQYbFD0ukLT84wJAW9MNUHl0A1AU3QCVSTsARdANUJl0AwBFqiqVSqWih6B9a25uTnX1x3f6mpqaUlNTk8bGxnTr1i1z5szJV7/61bz88svp379/+vfvn9ra2px44olZd911C5gcAGhrugEAaC3dAAC0lm4AAGBpcUUrlommpqbMmjUrSVpOXkaPHv2Rn6+pqclbb72VwYMH55///Gc6deqUO+64I5tvvnnGjRuX/fbbL5dffrmTFwBo53QDANBaugEAaC3dAADAsmDRimXiiSeeyPHHH59JkyYlSb7zne9k6NChmT17dkqlUmpqavLmm2/mwAMPzLrrrpsBAwa03Av9xhtvzMYbb5xNNtkk3bt3L/hIAIBlTTcAAK2lGwCA1tINAAAsC24dyDLx9NNP54wzzsj8+fOz2mqr5fXXX8+wYcOy0UYbJUkaGxuz7bbbZvvtt895552XhoaGJMm8efPcVxkAOhjdAAC0lm4AAFpLNwAAsCxYtGKZefzxxzNkyJAkybBhw7LDDjt85D7ojz76aDbffPN069atyDEBgDKgGwCA1tINAEBr6QYAAJY2tw5kqWpqamr5z2PHjs1aa62VPn365Oyzz867776b6urqzJ8/P0my4447tpy82PcDgI5HNwAAraUbAIDW0g0AACxLrmjFUlMqlVJVVZUkOeywwzJz5sxcdtlleeGFF3LZZZdl1qxZ+dnPfpb+/funqakpNTU1H3kOANBx6AYAoLV0AwDQWroBAIBlzRWtWCqamppaTkQee+yxTJkyJd/5znfSt2/f7Lzzzvne976XLl265Bvf+Ebee++91NTUZM6cObnpppvyyiuvFDw9ANCWdAMA0Fq6AQBoLd0AAEBbqC16ANqHmpqaJMkdd9yRf/zjH1l11VWz1VZbtXwSZOedd06SXHbZZdlnn31y/PHH5+mnn85DDz2UHXfcscjRAYA2phsAgNbSDQBAa+kGAADagkUrlpqXXnopZ555Zrp06ZJdd9019fX1SZL58+entrY2O++8c+rq6nLNNdfk3HPPTZ8+fXLrrbdmjTXWKHhyAKCt6QYAoLV0AwDQWroBAIBlrapUKpWKHoL247HHHsspp5ySWbNm5eKLL84OO+yQ5P+dxCTJ1KlTM27cuPTp0ye9e/cucFoAoEi6AQBoLd0AALSWbgAAYFmyaMUn0tTU1HIZ3v/r0UcfzY9+9KOsvfba+fa3v53Pfvazi3wOANB+6QYAoLV0AwDQWroBAIAiWLRisS14InL77bfn3XffTU1NTTbffPNsscUWqaury0MPPZSzzz47a6+9do444oiWk5gP74UOAHQMugEAaC3dAAC0lm4AAKAoFq1YLAuegBx11FEZMWJEVl111YwZMya9evXKeuutlzPOOCOdOnXKH/7wh5x++ulZd911c8ghh2SbbbYpeHoAoC3pBgCgtXQDANBaugEAgCLVFj0AleXDk5frrrsuL730Uq688sqst956qa+vz5lnnpnbbrstm266afbbb7/stNNOSZJjjjkmnTt3zqabbprOnTsXOT4A0IZ0AwDQWroBAGgt3QAAQJEsWvGJ/Pvf/84GG2yQtdZaK/X19Rk/fnx+85vfZO+9985XvvKVJB9cunennXbKFVdckdVWW83JCwB0ULoBAGgt3QAAtJZuAACgCNVFD0BlKZVKmT9/fv7973+nvr4+DQ0NefPNN/OVr3wl2267bX70ox+lS5cuueuuu/KHP/whpVIp22+/fVZfffWiRwcA2phuAABaSzcAAK2lGwAAKJJFK/6rUqn0kR9XVVWltrY2O+ywQ1566aX89re/zf77759tttkmZ511VpZbbrn8+9//ziOPPJKpU6d+7PkAQPulGwCA1tINAEBr6QYAAMqJRSv+o3nz5rXc63zWrFmZO3duy89tvvnmmTt3bk444YSsv/76ufTSS9OtW7dMmTIlN998c8aOHZttttkm1dW+xQCgI9ANAEBr6QYAoLV0AwAA5aaqZJWfBcydOzfTpk1L7969Wx4799xz89xzz6Wqqiprrrlmjj/++PTu3Tu/+c1vct5552W11VbLPvvsk1mzZuWZZ57J008/neHDh2fgwIEFHgkAsKzpBgCgtXQDANBaugEAgHJm0YoWTU1N2XfffbPqqqvm5JNPTt++ffPDH/4wf/zjH7PLLrtk7ty5+fOf/5yampqcdtpp2X777fPwww/noYceylNPPZU+ffpkjTXWyNFHH5211lqr6MMBAJYh3QAAtJZuAABaSzcAAFDuLFrxEbfeemvOP//87Lnnntl///1z4YUX5tBDD80OO+yQJBk/fnyOP/74vPXWW7nllluyxhprZP78+Zk+fXq6deuW+fPnp1OnTsUeBADQJnQDANBaugEAaC3dAABAObNoRebOnZv6+vqWH99zzz059dRTs/3222fChAm5+uqr06dPnzQ3N6e6ujrTpk3L/vvvn379+uXGG2/8yK9VKpVa7pcOALQ/ugEAaC3dAAC0lm4AAKBSVBc9AMWaP39+vv71r+d///d/Wx4bNGhQzj333DzxxBN58cUX8/LLLydJqqurM3/+/HTv3j0HHHBAXn/99YwePfojv56TFwBov3QDANBaugEAaC3dAABAJbFo1cHNnj07u+++e55//vkcd9xxLY/vvffeGTp0aKqrq3P33XdnzJgxSZLa2tokSefOnTNnzhwnLADQgegGAKC1dAMA0Fq6AQCASlJb9AAUY/bs2WlqakpDQ0MGDx6c5ZZbLuecc06OPfbYXHLJJUmSL33pS5k7d25OPfXUdOrUKV/96lez6aabZsyYMfnrX/+aPn36pKGhoeAjAQCWNd0AALSWbgAAWks3AABQiapKpVKp6CFoW3Pnzs2hhx6aTTfdNN///veTJI2NjXnwwQdz7rnnZscdd2w5iUk+uBf6D3/4w3Tu3DkbbbRRqqurM3r06AwbNizrrrtuUYcBALQB3QAAtJZuAABaSzcAAFCpLFp1QHPnzs0DDzyQXXbZJV27ds2cOXPSqVOnTJ06NQ899NBCT2Luv//+/PCHP0z37t1z2mmnZYMNNkjfvn0LPAoAoC3oBgCgtXQDANBaugEAgEpl0aqDaWpqSk1NTcuPhw4dmhEjRuT6669PQ0PDfz2J+eUvf5kLL7wwv/vd77L88ssXMT4A0IZ0AwDQWroBAGgt3QAAQCWrLnoA2lZVVdVHftypU6dMmDAhJ554YqZPn54ePXrki1/8Yk455ZQ8+uijOfbYY1u+dt99980jjzzi5AUAOgjdAAC0lm4AAFpLNwAAUMksWnUAzc3N+f3vf58//OEPqa7+4I/87LPPzlNPPZXvfOc72XffffPqq6/mBz/4wcdOYp544okcfvjhLb9W9+7dizoMAKAN6AYAoLV0AwDQWroBAID2orboAVj2pk6dmmeeeSYPPPBA5syZkwceeCD//Oc/s88++6Suri6HHnpoSqVS7rrrrvzgBz/IT37yk5aTmDlz5uTaa6/N+PHj3escADoA3QAAtJZuAABaSzcAANBeVJVKpVLRQ7Dsvfjii7nqqqvy5z//OV26dMkvfvGLrLrqqi33Qp87d25uvPHG3HXXXfn0pz+dn/zkJ2loaMi0adOS+IQIAHQkugEAaC3dAAC0lm4AAKA9cOvAdu7DPboNNtggnTp1yvz581NTU5MRI0YkSWpqajJv3rzU19fnsMMOyz777JM33ngjRx55ZKZPn57u3bs7eQGADkI3AACtpRsAgNbSDQAAtCduHdiOffgpkA/tvPPO2XHHHfPAAw/ksssuS3NzcwYNGpS6urqPnMTMmzcvjzzySBobG9PQ0FDgEQAAbUU3AACtpRsAgNbSDQAAtDduHdhOLXjyctttt6Vnz57ZfffdkyT//Oc/c9VVV+XVV1/Nd7/73QwePDhJMmvWrIwdOzZrrbVWpkyZkp49exY1PgDQhnQDANBaugEAaC3dAABAe+TWge1Qc3Nzy8nLsccem+HDh+fVV1/Ne++9lyTZaKONcsQRR2SdddbJlVdemV/+8peZMWNGzj///Jx00klpbGx08gIAHYRuAABaSzcAAK2lGwAAaK9c0aqdKZVKqaqqSpL86Ec/ytNPP51zzjkn6667brp3757m5uZUV3+wXzdixIjccMMNefTRR7P66qtn4sSJufHGG7P++usXeQgAQBvRDQBAa+kGAKC1dAMAAO1ZbdEDsORmz56dkSNHZv311285eXnzzTfzzDPP5Mgjj8yWW26ZJHnvvffy5z//OZMmTcqgQYOy6aabplu3btl5550zduzYfOUrX8lqq61W5KEAAMuYbgAAWks3AACtpRsAAOgoLFpVuPnz5+fAAw/MNttsk3XXXbflUyCNjY15991306NHj4wbNy7PP/98zjzzzNTW1mbChAm58847c/fdd+fTn/50Pv3pTxd8FABAW9ANAEBr6QYAoLV0AwAAHYlbB1awpqam1NTU5Nlnn83AgQPT0NCQ0aNHZ9VVV02SHHjggXnllVeywgorZNKkSdl7771zwAEHZNq0aTn44INz0kkn5atf/WrBRwEAtAXdAAC0lm4AAFpLNwAA0NG4olWFmjt3bg466KDsueeeGTRoULp06ZJzzjknL7zwQo4//vhsvvnmufXWW3PZZZelT58+WXnllbP99tsnSf71r3+lX79+6d+/f8FHAQC0Bd0AALSWbgAAWks3AADQEVm0qlD19fVZbrnlcumll6ZLly4ZNGhQNthggzzwwAO54YYbkiSbb755jjnmmI88b/z48bntttuSJOuss05bjw0AFEA3AACtpRsAgNbSDQAAdEQWrSpQqVRKVVVVbr755hx99NE5/fTTUyqVMnjw4Cy33HI57bTTcv3116e6ujqbbrppy/Ouv/76vPTSS3n66adz8803Z8UVVyzwKACAtqAbAIDW0g0AQGvpBgAAOqrqogdg8TU1NbX856FDh2adddbJlVdembvvvju77LJLTj/99Pzzn//Mtddem+eeey7JB58Q+de//pVZs2bl1ltvzcCBA4saHwBoQ7oBAGgt3QAAtJZuAACgo6oqlUqlooeg9T78lEiSnHDCCWlsbMzbb7+d0aNHp1OnTjnxxBMzePDgPPzwwzn99NOz0UYb5Tvf+U422mijzJw5M01NTenWrVvBRwEAtAXdAAC0lm4AAFpLNwAA0JG5dWCF+fDk5eyzz85TTz2Vc889N6uuumomTZqUn/70pznrrLOSJIMHD275ugsuuCAnnnhiNtpoo8LmBgDanm4AAFpLNwAAraUbAADoyCxaVaBp06blxRdfzBe+8IVsv/32SZI11lgj11xzTYYMGZLzzjsvyQcnMXPnzs2wYcPSp0+fIkcGAAqiGwCA1tINAEBr6QYAADoqi1YVqHPnzpk9e3amTJnS8lhTU1Nqampywgkn5NBDD821116bOXPm5MADD8wOO+yQhoaG4gYGAAqjGwCA1tINAEBr6QYAADqq6qIHYPFVV1dnww03zCuvvJIXXnghSVJTU5MkWWmlldK3b9+MGzcuV111VRobG528AEAHphsAgNbSDQBAa+kGAAA6KotWFai2tjaHHHJIxo8fn5/+9Kd59dVXW35u4sSJ6du3b37961/n3nvvTbdu3QqcFAAomm4AAFpLNwAAraUbAADoqNw6sEINGDAgl19+eY466qi8++672WmnndK/f/88+uijef7557Pccsu53zkAkEQ3AACtpxsAgNbSDQAAdERVpVKpVPQQfHKvvvpqzj333IwcOTLz58/P8ssvn0suuSQDBw4sejQAoMzoBgCgtXQDANBaugEAgI7EolU7MGPGjEyfPj2NjY3p3bt3evbsWfRIAECZ0g0AQGvpBgCgtXQDAAAdhUUrAAAAAAAAAACARaguegAAAAAAAAAAAIByZ9EKAAAAAAAAAABgESxaAQAAAAAAAAAALIJFKwAAAAAAAAAAgEWwaAUAAAAAAAAAALAIFq0AAAAAAAAAAAAWwaIVAAAAAAAAAADAIli0AgAAAAAAAAAAWASLVgAAAAAAAAAAAItg0QoAAAAAAAAAAGARLFoBAAAAAAAAAAAsgkUrAAAAAAAAAACARfj/ALzT+qM6hbllAAAAAElFTkSuQmCC",
            "text/plain": [
              "<Figure size 2400x600 with 4 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "def display_eval_results(eval_results):\n",
        "    rows = []\n",
        "    for embedding_model, loss_fns in eval_results.items():\n",
        "        for loss_fn, splits in loss_fns.items():\n",
        "            for split, metrics in splits.items():\n",
        "                multi_label_metrics = metrics.get(\"Multi-label\", {})\n",
        "                rows.append([\n",
        "                    embedding_model,\n",
        "                    loss_fn,\n",
        "                    split,\n",
        "                    multi_label_metrics.get(\"Accuracy\", None),\n",
        "                    multi_label_metrics.get(\"Recall\", None),\n",
        "                    multi_label_metrics.get(\"Precision\", None),\n",
        "                    multi_label_metrics.get(\"F1-Score\", None)\n",
        "                ])\n",
        "\n",
        "    columns = [\"Embedding Model\", \"Clf Loss Function\", \"Data Split\", \"Accuracy\", \"Precision\", \"Recall\", \"F1-Score\"]\n",
        "    results_df = pd.DataFrame(rows, columns=columns)\n",
        "\n",
        "    print(\"Evaluation Results:\")\n",
        "    print(tabulate(results_df, headers='keys', tablefmt='grid', showindex=False))\n",
        "\n",
        "    sns.set(style=\"whitegrid\")\n",
        "    metrics_to_plot = [\"Accuracy\", \"Precision\", \"Recall\", \"F1-Score\"]\n",
        "    splits = results_df[\"Data Split\"].unique()\n",
        "\n",
        "    for split in splits:\n",
        "        split_df = results_df[results_df[\"Data Split\"] == split]\n",
        "        num_metrics = len(metrics_to_plot)\n",
        "        fig, axes = plt.subplots(1, num_metrics, figsize=(6 * num_metrics, 6), sharey=True)\n",
        "\n",
        "        fig.suptitle(f\"Performance Metrics for {split} Split\", fontsize=16)\n",
        "\n",
        "        if num_metrics == 1:\n",
        "            axes = [axes]\n",
        "\n",
        "        for i, (ax, metric) in enumerate(zip(axes, metrics_to_plot)):\n",
        "            current_ax = sns.barplot(\n",
        "                data=split_df,\n",
        "                x=\"Embedding Model\",\n",
        "                y=metric,\n",
        "                hue=\"Clf Loss Function\",\n",
        "                ax=ax,\n",
        "                dodge=True\n",
        "            )\n",
        "            ax.set_title(f\"{metric}\")\n",
        "            ax.set_ylim(0, 1)\n",
        "            ax.set_ylabel(\"\")\n",
        "            ax.set_xlabel(\"\")\n",
        "            ax.tick_params(axis='x', rotation=45)\n",
        "\n",
        "            if i >= 0:\n",
        "                if ax.get_legend() is not None:\n",
        "                    ax.get_legend().remove()\n",
        "\n",
        "        handles, labels = axes[0].get_legend_handles_labels()\n",
        "        fig.legend(handles, labels, loc='upper right', bbox_to_anchor=(1, 1))\n",
        "\n",
        "        fig.tight_layout(rect=[0, 0, 0.9, 0.95])\n",
        "        plt.show()\n",
        "\n",
        "display_eval_results(eval_results)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IPGsCYiPvgFG"
      },
      "source": [
        "##### Train MLP using Embeddings (ToDo)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "VhWmdM5Zvvzc"
      },
      "outputs": [],
      "source": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "R8PfFHHWMaY8"
      },
      "source": [
        "##### Fine-Tuning RoBERTa Model (ToDo)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Y4EDnJOani8E"
      },
      "outputs": [],
      "source": []
    }
  ],
  "metadata": {
    "colab": {
      "collapsed_sections": [
        "MZ4OdPWGbqku",
        "TjdT2-CAd0jJ",
        "ULXV-yaIYIWl",
        "WKnv8gLTcFkX",
        "ewIjghW8hdfF",
        "YoGZCZJLtYEz",
        "Y0azMy_UweYM",
        "EfYVilooLFKV",
        "NiravP_Vsn7d",
        "_Zhl6VFCwZb1",
        "SsBJX0Ho3hfl",
        "sNn8KnPxxvMh",
        "r_pJsvi-oTKt",
        "iVJhrM6pnCxI",
        "GkVSUAsRxUo-",
        "W7izZznkWyWt",
        "kccdUkTICpRe",
        "pvcj_X8snEt-",
        "HQJcZJNPx4-n",
        "HLxuJ-QiyOEu",
        "b6XPXLa9FJc5",
        "HffH_qpjHqZe",
        "Ofi95mM-yWmq",
        "HLvQBcMgU8VT",
        "Zq6PDAulALNy",
        "R8PfFHHWMaY8"
      ],
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    },
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "042b7cd400c241f794f901805aaa1cd3": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_f866b3e489ed4bf49c8a75b321612d0d",
            "max": 498818054,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_1fc331390e7748438981f98c4a09f88b",
            "value": 498818054
          }
        },
        "0854e273d3154a78a00a685854e6e7a4": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_a3f67146220f400c82ff0236d07162f4",
              "IPY_MODEL_6cf7f72efe974f33b8849f3ab049daec",
              "IPY_MODEL_cf856c7982c94828af0e1e7d8e1e5285"
            ],
            "layout": "IPY_MODEL_e56544d71940410fa293f3888654c634"
          }
        },
        "09013d348dcc44669b6c4da46fc4e6bf": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "09af5f6f152b4b908810b4284bd937a6": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "0dcee213b9c944ddad9459cbc08d2084": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "0e1e9f21aca74539b8b610e6b4d5c4b0": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "0eb6efd617a04be782ecffda4c386669": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_4ed1d2955e8b4512aacdb235bd70ef07",
            "placeholder": "​",
            "style": "IPY_MODEL_9d1e2268fe7148df924023500d4d22f2",
            "value": "model.safetensors: 100%"
          }
        },
        "171f40bc89d6437eaa8534a743038349": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_4c6d013dd817420780e7cf4c379c667d",
            "max": 48,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_7ebc447c3dd44b9b97aee5a6e969c199",
            "value": 48
          }
        },
        "1b68b31a65a349ae843369b8a0b03323": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_9d9b592726bf40638ac4515ee8b04ec9",
            "placeholder": "​",
            "style": "IPY_MODEL_b0579c20e1dd45baa4eeb0af17dd920f",
            "value": "tokenizer_config.json: 100%"
          }
        },
        "1b7b474aca0b484baa8d31c29b33a927": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "1fc331390e7748438981f98c4a09f88b": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "210f48c9e624498d8bdbaafa19510a2c": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_cf4cdf07224147b59ea813fc1ce479af",
              "IPY_MODEL_b2b9eea953a541f3a0c7719d9d4bee54",
              "IPY_MODEL_fe9a348c3f0b4226901813b32905f517"
            ],
            "layout": "IPY_MODEL_e434200356ce43a7bca5b5cf1fbc53b6"
          }
        },
        "21540d448e6c460cbaad965a3659dc99": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "221ac8bc9583414584c39796d4ee7ab8": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "24eb9a1543994ebabc78fc21b38395f1": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "27418a91eee94ac193258ee194f4412d": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "29d4741a47a146d4be5a98914c3c3ef4": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_fe03e0bba3ba4413879e0aab1acfa73b",
            "max": 456318,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_328b0579a6e740cdb7f4c36625c9a063",
            "value": 456318
          }
        },
        "2b944f0e05ee40f8b1985bc57dfc351e": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_1b7b474aca0b484baa8d31c29b33a927",
            "placeholder": "​",
            "style": "IPY_MODEL_24eb9a1543994ebabc78fc21b38395f1",
            "value": " 483/483 [00:00&lt;00:00, 9.69kB/s]"
          }
        },
        "2bc953c0e4324f9b99ca60e4896a0b17": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "2d9b38d1a35149bfadc1a304015cba92": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_0e1e9f21aca74539b8b610e6b4d5c4b0",
            "placeholder": "​",
            "style": "IPY_MODEL_e4a4eb51118848649ba38369178f9905",
            "value": "tokenizer_config.json: 100%"
          }
        },
        "328b0579a6e740cdb7f4c36625c9a063": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "32a3e0e82d6b4e4d89622f2b1d825487": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_751b9fc0ab0043ac9971411e7302ad89",
            "placeholder": "​",
            "style": "IPY_MODEL_725b9f3a838945df92f7b316b67f3927",
            "value": " 268M/268M [00:01&lt;00:00, 252MB/s]"
          }
        },
        "32c49056d3d5465f8bdfe1b8fd5e22a2": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_6e32e3bd91c049b88833ad41ff4d0ab3",
            "placeholder": "​",
            "style": "IPY_MODEL_09013d348dcc44669b6c4da46fc4e6bf",
            "value": "tokenizer.json: 100%"
          }
        },
        "32fdaba87bd44637901f1d8c0c73e513": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "36986e412bea480b943b0f8c38ddb4a3": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_1b68b31a65a349ae843369b8a0b03323",
              "IPY_MODEL_d007c33743d6486aa84b0a62fa4e89a9",
              "IPY_MODEL_37a216a4c7e745f095b788eb4753cebd"
            ],
            "layout": "IPY_MODEL_fc5c525a02c247e6b438d5cd00d7efa7"
          }
        },
        "37a216a4c7e745f095b788eb4753cebd": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_27418a91eee94ac193258ee194f4412d",
            "placeholder": "​",
            "style": "IPY_MODEL_21540d448e6c460cbaad965a3659dc99",
            "value": " 25.0/25.0 [00:00&lt;00:00, 2.23kB/s]"
          }
        },
        "4079af81b66744b991c307ca0850f3f8": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "4292de59507a45f1a1948e1dbc48f096": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "44da2b9729b84af5b22533c92867f824": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_baf77f9862bc43728f1a0e1f42fdf2fd",
            "placeholder": "​",
            "style": "IPY_MODEL_55331618b0ef41ff8476ea8f8f035cce",
            "value": " 456k/456k [00:00&lt;00:00, 35.4MB/s]"
          }
        },
        "4a0fb05c031e42c4b9b1194a30c5da9f": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_832d655d892a4b6b9f0f07655bd281fd",
            "placeholder": "​",
            "style": "IPY_MODEL_53ef66d4635a42c28b6d52ff76f5e099",
            "value": "config.json: 100%"
          }
        },
        "4b6bef99bd0a463cacb1988e6a0f7223": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "4b973a3d4cef46efbc3b472462f8f2b0": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "4c6d013dd817420780e7cf4c379c667d": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "4ed1d2955e8b4512aacdb235bd70ef07": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "530e04ed3ab24eb8bb6f852d2fafffae": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "538d773cee224c3cb16064fcc851ac59": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_0dcee213b9c944ddad9459cbc08d2084",
            "placeholder": "​",
            "style": "IPY_MODEL_af259a9b6d064bada3350a45da3961d6",
            "value": "merges.txt: 100%"
          }
        },
        "53ef66d4635a42c28b6d52ff76f5e099": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "55331618b0ef41ff8476ea8f8f035cce": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "56a31ac940be454da77c66bd2d9e2e2e": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "5b3d5600f74449a2ae434e9c8551ee41": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_4079af81b66744b991c307ca0850f3f8",
            "max": 483,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_c44a9de67b66454f8e6027f933e9c055",
            "value": 483
          }
        },
        "5bbbe2ebd32c4d82909ef543d5008caa": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "5ef2208a369245579dd9c64f3904003c": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_959373a8b1f14477b18fa48d23ee10b2",
            "max": 267954768,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_7656f56c662746f7b9d03813e37e3656",
            "value": 267954768
          }
        },
        "5f3b86b081f54f889a7e50ad4223a734": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "5fdb52f0866c4fd2a1aef95b773beace": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "645f4bb400e34efab18ffe63946f5329": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "6669401af59949558d684420fbe80715": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_09af5f6f152b4b908810b4284bd937a6",
            "placeholder": "​",
            "style": "IPY_MODEL_93232b4e658e4555a984a512ac929de1",
            "value": " 481/481 [00:00&lt;00:00, 36.5kB/s]"
          }
        },
        "67708facdfe64ce587170066efc51f8a": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_4a0fb05c031e42c4b9b1194a30c5da9f",
              "IPY_MODEL_5b3d5600f74449a2ae434e9c8551ee41",
              "IPY_MODEL_2b944f0e05ee40f8b1985bc57dfc351e"
            ],
            "layout": "IPY_MODEL_5fdb52f0866c4fd2a1aef95b773beace"
          }
        },
        "6cf7f72efe974f33b8849f3ab049daec": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_9e188051f4c44e3f81b473843f762610",
            "max": 898823,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_5f3b86b081f54f889a7e50ad4223a734",
            "value": 898823
          }
        },
        "6e32e3bd91c049b88833ad41ff4d0ab3": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "716e17e0b8604bf68c43764fcbb8633a": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "725b9f3a838945df92f7b316b67f3927": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "72a7784ab9b644b6901a421afb1a19e5": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "751b9fc0ab0043ac9971411e7302ad89": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "7656f56c662746f7b9d03813e37e3656": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "7bd2fc7d573a4bbaad2699066f9e4b1d": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_538d773cee224c3cb16064fcc851ac59",
              "IPY_MODEL_29d4741a47a146d4be5a98914c3c3ef4",
              "IPY_MODEL_44da2b9729b84af5b22533c92867f824"
            ],
            "layout": "IPY_MODEL_955d93cae7b04cb38b75e937e1c6b025"
          }
        },
        "7ebc447c3dd44b9b97aee5a6e969c199": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "82eb5391297a4c1993258c7e816642d9": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_d7ef0074ac4a40db880ff1446ee08087",
            "max": 481,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_4b6bef99bd0a463cacb1988e6a0f7223",
            "value": 481
          }
        },
        "832d655d892a4b6b9f0f07655bd281fd": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "83a0cbc2f5304587a4877ef4e4c202d9": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "83ce9dac48454bdebf6a9f336e7eca68": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "85bc01eff9e045b7a5dda7055e17f0de": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "8765f04caa284ec2bca71aa96a57adac": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "89bcbe0eb4084684a5f4c498e0ff26c7": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "8c77c28c6e5f470eb0fe51875a70a9fe": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "8d4f25d91cd54cf39d6ed414431e7255": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_530e04ed3ab24eb8bb6f852d2fafffae",
            "placeholder": "​",
            "style": "IPY_MODEL_89bcbe0eb4084684a5f4c498e0ff26c7",
            "value": " 232k/232k [00:00&lt;00:00, 510kB/s]"
          }
        },
        "8fa0e4639c524185ad701c68d7c9ef4c": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "909fa9050b804169a443223310fd6dc9": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "93232b4e658e4555a984a512ac929de1": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "94fcde6fbb4748569ae0b65fa32dd45a": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_2d9b38d1a35149bfadc1a304015cba92",
              "IPY_MODEL_171f40bc89d6437eaa8534a743038349",
              "IPY_MODEL_b5d78abe651d4873b4293edb1c910675"
            ],
            "layout": "IPY_MODEL_e952f73bb71b42e2ae358eef9a519db0"
          }
        },
        "955d93cae7b04cb38b75e937e1c6b025": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "959373a8b1f14477b18fa48d23ee10b2": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "969eadd452e045da9ea282ed2f968d3b": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "98562d71d83141599a7b481afdc05e88": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_bf96e6e5ca6e439fa12b22913d7c0ff3",
            "placeholder": "​",
            "style": "IPY_MODEL_969eadd452e045da9ea282ed2f968d3b",
            "value": "config.json: 100%"
          }
        },
        "9c5173c7f40d4d2d9e9667d58009920c": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_4292de59507a45f1a1948e1dbc48f096",
            "placeholder": "​",
            "style": "IPY_MODEL_8c77c28c6e5f470eb0fe51875a70a9fe",
            "value": "model.safetensors: 100%"
          }
        },
        "9d1e2268fe7148df924023500d4d22f2": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "9d219091b8cc4be990cffa479edbaf64": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_f7ef09bb64fe4a0d9be14cd2eece8f65",
            "max": 231508,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_bbc946d030cc4bdeb828edca824356b5",
            "value": 231508
          }
        },
        "9d9b592726bf40638ac4515ee8b04ec9": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "9e188051f4c44e3f81b473843f762610": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "9f597c67fcda441c9037cf69b25ee4ac": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "a3377ac61e3b42ccb632dee2091795b0": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_dc95fea471be407b9972db4702b1a089",
            "placeholder": "​",
            "style": "IPY_MODEL_8765f04caa284ec2bca71aa96a57adac",
            "value": " 1.36M/1.36M [00:01&lt;00:00, 1.30MB/s]"
          }
        },
        "a3f67146220f400c82ff0236d07162f4": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_cc4ec11b351c48d78c8b573fd40301df",
            "placeholder": "​",
            "style": "IPY_MODEL_645f4bb400e34efab18ffe63946f5329",
            "value": "vocab.json: 100%"
          }
        },
        "a44c195268324772ab260d14f1d38dd9": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_deffb02b2090469e88a0fd3330a04557",
              "IPY_MODEL_9d219091b8cc4be990cffa479edbaf64",
              "IPY_MODEL_8d4f25d91cd54cf39d6ed414431e7255"
            ],
            "layout": "IPY_MODEL_32fdaba87bd44637901f1d8c0c73e513"
          }
        },
        "ac33694f66994c3e9b6a2e5e078fbfe5": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "af259a9b6d064bada3350a45da3961d6": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "b0579c20e1dd45baa4eeb0af17dd920f": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "b2b9eea953a541f3a0c7719d9d4bee54": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_83a0cbc2f5304587a4877ef4e4c202d9",
            "max": 466062,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_8fa0e4639c524185ad701c68d7c9ef4c",
            "value": 466062
          }
        },
        "b5d78abe651d4873b4293edb1c910675": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_cf801b8a1fcc455da2306242d9f54136",
            "placeholder": "​",
            "style": "IPY_MODEL_85bc01eff9e045b7a5dda7055e17f0de",
            "value": " 48.0/48.0 [00:00&lt;00:00, 1.14kB/s]"
          }
        },
        "baf77f9862bc43728f1a0e1f42fdf2fd": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "bbc946d030cc4bdeb828edca824356b5": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "bf6c4f5b56594db7bdf19b488d75660f": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_9c5173c7f40d4d2d9e9667d58009920c",
              "IPY_MODEL_042b7cd400c241f794f901805aaa1cd3",
              "IPY_MODEL_f2b199aadfb94784bca9834ed79f5af9"
            ],
            "layout": "IPY_MODEL_4b973a3d4cef46efbc3b472462f8f2b0"
          }
        },
        "bf96e6e5ca6e439fa12b22913d7c0ff3": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "c44a9de67b66454f8e6027f933e9c055": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "c547a419af514f2ab7034a127be355d4": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "c7ce951e4e1f4e93b6e27b5d40376060": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "c7de73c1144343269cfd7f0786267981": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_716e17e0b8604bf68c43764fcbb8633a",
            "max": 1355863,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_ff86746a739a47c98af4e72810c20177",
            "value": 1355863
          }
        },
        "cc4ec11b351c48d78c8b573fd40301df": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "cd7c8ce702ab4cc2bcea50e61eb7cf89": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_0eb6efd617a04be782ecffda4c386669",
              "IPY_MODEL_5ef2208a369245579dd9c64f3904003c",
              "IPY_MODEL_32a3e0e82d6b4e4d89622f2b1d825487"
            ],
            "layout": "IPY_MODEL_e8b4e83df5e74fbba199a65bcabda497"
          }
        },
        "cf4cdf07224147b59ea813fc1ce479af": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_c547a419af514f2ab7034a127be355d4",
            "placeholder": "​",
            "style": "IPY_MODEL_83ce9dac48454bdebf6a9f336e7eca68",
            "value": "tokenizer.json: 100%"
          }
        },
        "cf801b8a1fcc455da2306242d9f54136": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "cf856c7982c94828af0e1e7d8e1e5285": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_909fa9050b804169a443223310fd6dc9",
            "placeholder": "​",
            "style": "IPY_MODEL_56a31ac940be454da77c66bd2d9e2e2e",
            "value": " 899k/899k [00:00&lt;00:00, 1.96MB/s]"
          }
        },
        "d007c33743d6486aa84b0a62fa4e89a9": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_5bbbe2ebd32c4d82909ef543d5008caa",
            "max": 25,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_221ac8bc9583414584c39796d4ee7ab8",
            "value": 25
          }
        },
        "d58dba0481b147e1a55b243c0b130253": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "d7ef0074ac4a40db880ff1446ee08087": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "dc95fea471be407b9972db4702b1a089": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "deffb02b2090469e88a0fd3330a04557": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_ac33694f66994c3e9b6a2e5e078fbfe5",
            "placeholder": "​",
            "style": "IPY_MODEL_9f597c67fcda441c9037cf69b25ee4ac",
            "value": "vocab.txt: 100%"
          }
        },
        "e434200356ce43a7bca5b5cf1fbc53b6": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "e4a4eb51118848649ba38369178f9905": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "e56544d71940410fa293f3888654c634": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "e822f7a637d449ff96e29c4349454baa": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_98562d71d83141599a7b481afdc05e88",
              "IPY_MODEL_82eb5391297a4c1993258c7e816642d9",
              "IPY_MODEL_6669401af59949558d684420fbe80715"
            ],
            "layout": "IPY_MODEL_d58dba0481b147e1a55b243c0b130253"
          }
        },
        "e8b4e83df5e74fbba199a65bcabda497": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "e952f73bb71b42e2ae358eef9a519db0": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "f2b199aadfb94784bca9834ed79f5af9": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_f926396796ca4406a2230aa4632c02a8",
            "placeholder": "​",
            "style": "IPY_MODEL_2bc953c0e4324f9b99ca60e4896a0b17",
            "value": " 499M/499M [00:02&lt;00:00, 248MB/s]"
          }
        },
        "f7ef09bb64fe4a0d9be14cd2eece8f65": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "f866b3e489ed4bf49c8a75b321612d0d": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "f926396796ca4406a2230aa4632c02a8": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "f9b8cbed4a664e9aad77deea6f2b8c39": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "fc5c525a02c247e6b438d5cd00d7efa7": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "fcb33bb07eae44189c50b95c90988fa8": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_32c49056d3d5465f8bdfe1b8fd5e22a2",
              "IPY_MODEL_c7de73c1144343269cfd7f0786267981",
              "IPY_MODEL_a3377ac61e3b42ccb632dee2091795b0"
            ],
            "layout": "IPY_MODEL_c7ce951e4e1f4e93b6e27b5d40376060"
          }
        },
        "fe03e0bba3ba4413879e0aab1acfa73b": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "fe9a348c3f0b4226901813b32905f517": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_f9b8cbed4a664e9aad77deea6f2b8c39",
            "placeholder": "​",
            "style": "IPY_MODEL_72a7784ab9b644b6901a421afb1a19e5",
            "value": " 466k/466k [00:00&lt;00:00, 1.12MB/s]"
          }
        },
        "ff86746a739a47c98af4e72810c20177": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        }
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}