mars_spectrometry.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pxUpv5xxLM26"
      },
      "source": [
        "# 1. Imports"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "p_LBeGJJ_3M8"
      },
      "outputs": [],
      "source": [
        "!pip install pandas-path, catboost, xgboost\n",
        "!pip install --upgrade matplotlib\n",
        "\n",
        "!git clone --recursive https://github.com/Microsoft/LightGBM\n",
        "!cd LightGBM && rm -rf build && mkdir build && cd build && cmake -DUSE_GPU=1 ../../LightGBM && make -j4 && cd ../python-package && python3 setup.py install --precompile --gpu;"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 44,
      "metadata": {
        "id": "G8OIWKKF4f5i"
      },
      "outputs": [],
      "source": [
        "import itertools\n",
        "from pathlib import Path\n",
        "from pandas_path import path\n",
        "from pprint import pprint\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "\n",
        "# visualization\n",
        "from matplotlib import pyplot as plt, cm\n",
        "import seaborn as sns\n",
        "from tqdm.notebook import tqdm\n",
        "\n",
        "# modeling\n",
        "from sklearn.dummy import DummyClassifier\n",
        "from sklearn.preprocessing import minmax_scale\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.metrics import make_scorer, log_loss\n",
        "from sklearn.model_selection import StratifiedKFold, cross_val_score\n",
        "\n",
        "# models\n",
        "from xgboost import XGBClassifier\n",
        "from catboost import CatBoostClassifier\n",
        "from lightgbm import LGBMClassifier\n",
        "\n",
        "pd.set_option(\"max_colwidth\", 80)\n",
        "RANDOM_SEED = 42  # For reproducibility"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "y4NhhHisLP_L"
      },
      "source": [
        "# 2. Data"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Competition data isn´t publicly available. Following data structure is expected:\n",
        "\n",
        "```\n",
        "../mars_spectrometry/\n",
        "├── train_features\n",
        "│   ├── sample_id_1.csv\n",
        "│   ├── sample_id_2.csv\n",
        "│   ├── sample_id_3.csv\n",
        "│   └── ...\n",
        "├── val_features\n",
        "│   ├── ...\n",
        "│   ...\n",
        "├── test_features\n",
        "│   ├── ...\n",
        "│   ...\n",
        "├── metadata.csv\n",
        "├── train_labels.csv\n",
        "├── val_labels.csv\n",
        "└── submission_format.csv\n",
        "```"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "02KAwxWdSNUW"
      },
      "source": [
        "# 3. EDA"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 238
        },
        "id": "2zy84_AVANcG",
        "outputId": "56efeb8c-1afd-45d9-baf3-edeea5bc0327"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-1ce3de15-4a9c-40ba-9612-4a91e42ec390\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>split</th>\n",
              "      <th>instrument_type</th>\n",
              "      <th>features_path</th>\n",
              "      <th>features_md5_hash</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>sample_id</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>S0000</th>\n",
              "      <td>train</td>\n",
              "      <td>commercial</td>\n",
              "      <td>train_features/S0000.csv</td>\n",
              "      <td>017b9a71a702e81a828e6242aa15f049</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>S0001</th>\n",
              "      <td>train</td>\n",
              "      <td>commercial</td>\n",
              "      <td>train_features/S0001.csv</td>\n",
              "      <td>0d09840214054d254bd49436c6a6f315</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>S0002</th>\n",
              "      <td>train</td>\n",
              "      <td>commercial</td>\n",
              "      <td>train_features/S0002.csv</td>\n",
              "      <td>3f58b3c9b001bfed6ed4e4f757083e09</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>S0003</th>\n",
              "      <td>train</td>\n",
              "      <td>commercial</td>\n",
              "      <td>train_features/S0003.csv</td>\n",
              "      <td>e9a12f96114a2fda60b36f4c0f513fb1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>S0004</th>\n",
              "      <td>train</td>\n",
              "      <td>commercial</td>\n",
              "      <td>train_features/S0004.csv</td>\n",
              "      <td>b67603d3931897bfa796ac42cc16de78</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-1ce3de15-4a9c-40ba-9612-4a91e42ec390')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-1ce3de15-4a9c-40ba-9612-4a91e42ec390 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-1ce3de15-4a9c-40ba-9612-4a91e42ec390');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "           split instrument_type             features_path  \\\n",
              "sample_id                                                    \n",
              "S0000      train      commercial  train_features/S0000.csv   \n",
              "S0001      train      commercial  train_features/S0001.csv   \n",
              "S0002      train      commercial  train_features/S0002.csv   \n",
              "S0003      train      commercial  train_features/S0003.csv   \n",
              "S0004      train      commercial  train_features/S0004.csv   \n",
              "\n",
              "                          features_md5_hash  \n",
              "sample_id                                    \n",
              "S0000      017b9a71a702e81a828e6242aa15f049  \n",
              "S0001      0d09840214054d254bd49436c6a6f315  \n",
              "S0002      3f58b3c9b001bfed6ed4e4f757083e09  \n",
              "S0003      e9a12f96114a2fda60b36f4c0f513fb1  \n",
              "S0004      b67603d3931897bfa796ac42cc16de78  "
            ]
          },
          "execution_count": 4,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "DATA_PATH = Path.cwd().parent / \"/content/mars_spectrometry/\"\n",
        "metadata = pd.read_csv(DATA_PATH / \"metadata.csv\", index_col=\"sample_id\")\n",
        "metadata.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bpH5jkALANuY",
        "outputId": "2250ebe8-a8fa-4d6d-e3ce-a7b400d6c88d"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Number of training samples:  766\n",
            "Number of validation samples:  293\n",
            "Number of testing samples:  511\n"
          ]
        }
      ],
      "source": [
        "train_files = metadata[metadata[\"split\"] == \"train\"][\"features_path\"].to_dict()\n",
        "val_files = metadata[metadata[\"split\"] == \"val\"][\"features_path\"].to_dict()\n",
        "test_files = metadata[metadata[\"split\"] == \"test\"][\"features_path\"].to_dict()\n",
        "\n",
        "print(\"Number of training samples: \", len(train_files))\n",
        "print(\"Number of validation samples: \", len(val_files))\n",
        "print(\"Number of testing samples: \", len(test_files))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 298
        },
        "id": "MRoeI77hSO78",
        "outputId": "512d2493-0c65-4679-af73-929d31750448"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "[Text(0, 0, '64'), Text(0, 0, '12'), Text(0, 0, '0')]"
            ]
          },
          "execution_count": 6,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Share of samples from commercial instruments vs. SAM testbed\n",
        "meta_instrument = (\n",
        "    metadata.reset_index()\n",
        "    .groupby([\"split\", \"instrument_type\"])[\"sample_id\"]\n",
        "    .aggregate(\"count\")\n",
        "    .reset_index()\n",
        ")\n",
        "meta_instrument = meta_instrument.pivot(\n",
        "    index=\"split\", columns=\"instrument_type\", values=\"sample_id\"\n",
        ").reset_index()\n",
        "meta_instrument.head()\n",
        "\n",
        "ax = meta_instrument.plot(\n",
        "    x=\"split\",\n",
        "    kind=\"barh\",\n",
        "    stacked=True,\n",
        "    title=\"Instrument type by dataset split\",\n",
        "    mark_right=True,\n",
        ")\n",
        "ax.bar_label(ax.containers[0], label_type=\"center\")\n",
        "ax.bar_label(ax.containers[1], label_type=\"center\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 238
        },
        "id": "jJGZUJoXSUig",
        "outputId": "a13113ba-6d97-4d3e-f1ce-bf976a20a142"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-58c5c567-bbf3-4bb8-b601-617f1cb38b26\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>basalt</th>\n",
              "      <th>carbonate</th>\n",
              "      <th>chloride</th>\n",
              "      <th>iron_oxide</th>\n",
              "      <th>oxalate</th>\n",
              "      <th>oxychlorine</th>\n",
              "      <th>phyllosilicate</th>\n",
              "      <th>silicate</th>\n",
              "      <th>sulfate</th>\n",
              "      <th>sulfide</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>sample_id</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>S0000</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>S0001</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>S0002</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>S0003</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>S0004</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-58c5c567-bbf3-4bb8-b601-617f1cb38b26')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-58c5c567-bbf3-4bb8-b601-617f1cb38b26 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-58c5c567-bbf3-4bb8-b601-617f1cb38b26');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "           basalt  carbonate  chloride  iron_oxide  oxalate  oxychlorine  \\\n",
              "sample_id                                                                  \n",
              "S0000           0          0         0           0        0            0   \n",
              "S0001           0          1         0           0        0            0   \n",
              "S0002           0          0         0           0        0            1   \n",
              "S0003           0          1         0           1        0            0   \n",
              "S0004           0          0         0           1        0            1   \n",
              "\n",
              "           phyllosilicate  silicate  sulfate  sulfide  \n",
              "sample_id                                              \n",
              "S0000                   0         0        1        0  \n",
              "S0001                   0         0        0        0  \n",
              "S0002                   0         0        0        0  \n",
              "S0003                   0         0        1        0  \n",
              "S0004                   1         0        0        0  "
            ]
          },
          "execution_count": 7,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "train_labels = pd.read_csv(DATA_PATH / \"train_labels.csv\", index_col=\"sample_id\")\n",
        "train_labels.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 238
        },
        "id": "izSX_7uCQ7XF",
        "outputId": "7e598965-04a4-48c2-eb45-519d10aec36c"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-b097a5f4-74ba-447f-8d18-b77b4b28365b\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>basalt</th>\n",
              "      <th>carbonate</th>\n",
              "      <th>chloride</th>\n",
              "      <th>iron_oxide</th>\n",
              "      <th>oxalate</th>\n",
              "      <th>oxychlorine</th>\n",
              "      <th>phyllosilicate</th>\n",
              "      <th>silicate</th>\n",
              "      <th>sulfate</th>\n",
              "      <th>sulfide</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>sample_id</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>S0766</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>S0767</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>S0768</th>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>S0769</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>S0770</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-b097a5f4-74ba-447f-8d18-b77b4b28365b')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-b097a5f4-74ba-447f-8d18-b77b4b28365b button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-b097a5f4-74ba-447f-8d18-b77b4b28365b');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "           basalt  carbonate  chloride  iron_oxide  oxalate  oxychlorine  \\\n",
              "sample_id                                                                  \n",
              "S0766           0          0         0           0        0            0   \n",
              "S0767           0          0         0           0        0            0   \n",
              "S0768           1          0         0           0        0            0   \n",
              "S0769           0          0         0           0        0            1   \n",
              "S0770           0          0         0           1        0            1   \n",
              "\n",
              "           phyllosilicate  silicate  sulfate  sulfide  \n",
              "sample_id                                              \n",
              "S0766                   0         0        0        0  \n",
              "S0767                   0         0        0        0  \n",
              "S0768                   1         1        0        0  \n",
              "S0769                   0         0        1        0  \n",
              "S0770                   1         0        0        0  "
            ]
          },
          "execution_count": 8,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "val_labels = pd.read_csv(DATA_PATH / \"val_labels.csv\", index_col=\"sample_id\")\n",
        "val_labels.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 323
        },
        "id": "JHTUuy2QSVnB",
        "outputId": "69fdbf70-3b78-43b7-d5cb-567b35dd9751"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "Text(0.5, 1.0, 'Compounds represented in training set')"
            ]
          },
          "execution_count": 9,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "sumlabs = train_labels.aggregate(\"sum\").sort_values()\n",
        "plt.barh(sumlabs.index, sumlabs, align=\"center\")\n",
        "plt.ylabel(\"Compounds\")\n",
        "plt.xticks(rotation=45)\n",
        "plt.xlabel(\"Count in training set\")\n",
        "plt.title(\"Compounds represented in training set\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "id": "565DiIgVSXAO"
      },
      "outputs": [],
      "source": [
        "# Select sample IDs for five commercial samples and five testbed samples\n",
        "sample_id_commercial = (\n",
        "    metadata[metadata[\"instrument_type\"] == \"commercial\"]\n",
        "    .index\n",
        "    .values[0:5]\n",
        ")\n",
        "sample_id_testbed = (\n",
        "    metadata[metadata[\"instrument_type\"] == \"sam_testbed\"]\n",
        "    .index\n",
        "    .values[0:5]\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "id": "W8VfItogSYae"
      },
      "outputs": [],
      "source": [
        "# Import sample files for EDA\n",
        "sample_commercial_dict = {}\n",
        "sample_testbed_dict = {}\n",
        "\n",
        "for i in range(0, 5):\n",
        "    comm_lab = sample_id_commercial[i]\n",
        "    sample_commercial_dict[comm_lab] = pd.read_csv(DATA_PATH / train_files[comm_lab])\n",
        "\n",
        "    test_lab = sample_id_testbed[i]\n",
        "    sample_testbed_dict[test_lab] = pd.read_csv(DATA_PATH / train_files[test_lab])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 465
        },
        "id": "KVaWomCMSZw5",
        "outputId": "96e9b5cd-20d5-4d0a-8b9e-f24b8951eaf1"
      },
      "outputs": [
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABEAAAADgCAYAAAADrBOyAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAAp8ElEQVR4nO3debxkdX3n/9ebRYyDbIIEgbZxyWSij4lLR83EqKOGEUxCZmIUMQpEQxZNTBxjMDqRIDo4vyiQuGLgJ2h+Im4Rf5pRo3FiJuoIuEs0aMDuFm0QBFxQwM/8cc6Vounue+reWk6dej0fj3rcqnNOVX9Odb3rfs/nniVVhSRJkiRJ0pDtNu8CJEmSJEmSps0GiCRJkiRJGjwbIJIkSZIkafBsgEiSJEmSpMGzASJJkiRJkgbPBogkSZIkSRo8GyCSJGmqkmxI8u0ku3dY9lFJtsyirrVK8uEkz5h3HZIkaTw2QCRJmrMkxyW5uG0SXJXkb5M8fN51TUpVfbWq9q6qW+ddiyRJWl42QCRJmqMkzwHOBF4KHAxsAF4NHDPHssaWZI951yBJkrQrNkAkSZqTJPsCpwLPrKp3VNV3qurmqnp3Vf1Ru8xeSc5M8rX2dmaSvdp5j0qyJcnzkmxr9x75lSRHJ/lSkmuT/MnIv3dKkrcmeVOSG5N8NslPJHl++/zNSY4crS/JOe3rbk1y2sphLElOSPK/k5yR5JvAKUl+LMnLk1yZ5Pok/9hO25ikVpokSU5Mcllbw1eS/FbH9yvtv7ctyQ1t/fdv5z0+ySfb6ZuTnDLyvJV//8R23nVJfjvJzyT5TJJvJXnlyPIr6/bKdj3+OcljdlHXb7Trc12S9yW552r1SpKk2bMBIknS/PwscGfgnbtY5gXAw4AHAD8NPAR44cj8H29f41DgT4HXA78OPBj4eeC/JTliZPlfAt4I7A98EngfzXjgUJpmzOtGln0DcAtwH+CBwJHA6LkvHgp8hWbPlZcAf97+u/8BOAB4HvDDHazTNuAXgX2AE4EzkjxoF+/BiiOBRwA/AewLPBH4ZjvvO8DTgP2AxwO/k+RXtnv+Q4H7Ak+i2evmBcBjgfsBT0zyyO2W/TJwIPAi4B1JDti+oCTHAH8C/BfgIOAjwJs71CtJkmbMBogkSfNzN+CaqrplF8s8BTi1qrZV1dXAnwFPHZl/M/CSqroZuIBmg/2sqrqxqj4PfIGmcbLiI1X1vvbffCvNRvvpI8/fmGS/JAcDRwN/0O6Zsg04Azh25LW+VlV/2b7W94HfAJ5dVVur6taq+qeq+v72K1RV76mqL1fjfwHvp2nWrOZm4K7ATwKpqsuq6qr2NT9cVZ+tqh9W1WdomhCP3O75L66qm6rq/TQNkze37+tWmsbFA0eW3Qac2e6R8xbgizSNle39NvDf21puoTmU6QHtXiA7rVeSJM2eDRBJkubnm8CBq5w/4x7AlSOPr2yn/eg1Rk4u+r325zdG5n8P2Hvk8fbzrtnB8/cG7gnsCVzVHiLyLZq9Q+4+8vzNI/cPpNkT5cu7WBcAkhyV5GPtITrfomm0HLja86rqQ8ArgVcB25KcnWSf9jUfmuTvk1yd5HqaxsT2r7n9uu/qfdpaVTXyePv3fcU9gbNG3qNrgQCH7qpeSZI0ezZAJEman4/S7DnxK7tY5ms0G9krNrTTpm0zTW0HVtV+7W2fqrrfyDKjDYJrgJuAe+/qRdvzl7yd5nCZg6tqP+C9NE2DVVXVX1TVg4Gfojm05I/aWf8fcBFweFXtC7y262vuxKFJRp+/s/d9M/BbI+/RflX1Y1X1T6vUK0mSZswGiCRJc1JV19Oct+NV7clL75Jkz3YPif/RLvZm4IVJDkpyYLv8m2ZQ21U0h6a8PMk+SXZLcu/tzpMxuvwPgXOBVyS5R5Ldk/xs2/AYdSdgL+Bq4JYkR9GcK2NV7UlLH5pkT5pDWG7itnOM3BW4tqpuSvIQ4LgxV3l7dwd+v/3/+DXg39E0arb3WuD5Se7X1rhvu/xq9UqSpBmzASJJ0hxV1cuB59Cc2PRqmj0KngX8TbvIacDFwGeAzwKXttNm4Wk0DYsvANcBbwMO2cXyz6Wp8RM0h4K8jO3GGlV1I/D7wIXtax5Hs+dGF/vQnOT1OppDUr4J/D/tvN8FTk1yI02T6MKOr7kzH6c5Yeo1NCd4fUJV3eEEplX1Tpr1vCDJDcDngKM61CtJkmYstz+8VZIkabklOQF4RlU9fN61SJKkyXEPEEmSJEmSNHg2QCRJkiRJ0uB5CIwkSZIkSRo89wCRJEmSJEmDZwNEkiRJkiQNng0QSZIkSZI0eDZAJEmSJEnS4NkAkSRJkiRJg2cDRJIkSZIkDZ4NEHWW5OFJ/inJ9UmuTfK/k/xMO++4JFcm+U6Sv0lywMjzDkjyznbelUmO2+511/xcaVHNKU/PSnJxku8necPMVlaagFlnJsleSc5p592Y5FNJjprtWktrN6ffM29KclWSG5J8KckzZrfG0vrMIzMjy9w3yU1J3jT9NV1uNkDUSZJ9gP8f+EvgAOBQ4M+A7ye5H/A64KnAwcB3gVePPP1VwA/aeU8BXtM+h/U8V1pUc8zT14DTgHOntW7SNMwpM3sAm4FHAvsCLwQuTLJxWuspTcocf8/8d2BjVe0D/DJwWpIHT2k1pYmZY2ZGX+MTE18x3UGqat41aAEk2QT8XVXtt4N5L6X5ZXdc+/jewGXA3YAfAtcB96+qL7Xz3whsraqT1/Pcaa6vNE3zyFNV3Tjyb5wGHFZVJ0xvLaXJmXdmRv6tzwB/VlVvn8JqShPTh8wk+bfAh4FnV9WFU1lRaULmmZkkxwL/BfgCcJ+q+vWpruyScw8QdfUl4NYk5yU5Ksn+I/PuB3x65UFVfZmmC/oT7e2WlS+E1qfb56z3udKimkeepEU298wkObid/vmJrJE0XXPLTJJXJ/ku8M/AVcB7J7li0pTMJTPtnienAs+Z+Bpph2yAqJOqugF4OFDA64Grk1zUDgj3Bq7f7inXA3dt592wk3ms87nSQppTnqSFNe/MJNkT+GvgvKr65/WtjTR988xMVf1u+/jngXcA31/v+kjTNsfMvBg4p6q2TGI9tDobIOqsqi6rqhOq6jDg/sA9gDOBbwP7bLf4PsCNq8xjnc+VFtYc8iQttHllJsluwBtp/mL3rHWviDQj8/w9U1W3VtU/AocBv7O+NZFmY9aZSfIA4LHAGRNaBXVgA0Rr0v4F7A00Xw6fB356ZV6SewF70exK9iVgjyT3HXn6T3PbLsTrea40CDPKkzQYs8pMkgDn0Jy47ler6ubprJE0XXP8PbMHcO+JrIQ0QzPKzKOAjcBXk3wdeC7wq0kuncY6qeFJUNVJkp8EHg+8paq2JDkcuIDmZD1nAh9t519Kc6bjParq2Pa5F9DsTvYM4AE0x4L+h6r6fHtm5DU9d/prLU3HHPO0B81g9EU0f5X7TZrjVm+ZwWpLazbHzLy2fc5jq+rbM1lZaQLmkZkkdwceTXMlje/R/GX7HcCTq+qiWay3tFZzysxduP3eIc+laYj8TlVdPdUVXmZV5c3bqjeaS0FdCGwFvtP+fB2wTzv/OOCr7bx3AQeMPPcA4G/aeV8Fjtvutdf8XG/eFvE2xzydQvMLevR2yrzfD2/eVrvNIzPAPduM3ESzC/PK7Snzfj+8eVvtNqfMHAT8L+BbNOdE+Czwm/N+L7x563Kb19hsu+VOAd407/di6Df3AJEkSZIkSYPnOUAkSZIkSdLg2QCRJEmSJEmDZwNEkiRJkiQNng0QSZIkSZI0eDZAJEmSJEnS4O0x7wJm5cADD6yNGzfOuwypFy655JJrquqgtT7fPEm36ZInMyPdxsxI4zEz0nh2lZmlaYBs3LiRiy++eN5lSL2Q5Mr1PN88SbfpkiczI93GzEjjMTPSeHaVGQ+BkSRJkiRJgzfTPUCSnAv8IrCtqu7fTjsAeAuwEbgCeGJVXZckwFnA0cB3gROq6tL2OccDL2xf9rSqOm+W6yFJkiRJkmZj48nvucO0K05//NivM+tDYN4AvBI4f2TaycAHq+r0JCe3j/8YOAq4b3t7KPAa4KFtw+RFwCaggEuSXFRV181sLSRJkiRpDXa0IQdr25iTlsHOMrPx5PeMnZuZNkCq6h+SbNxu8jHAo9r75wEfpmmAHAOcX1UFfCzJfkkOaZf9QFVdC5DkA8DjgDdPu35JkiTdnhtzUjc7y8rofHMj3Wa1zKxFH84BcnBVXdXe/zpwcHv/UGDzyHJb2mk7my5JkqQZ2tXgdBoDV2lRmQepu40nv2dqmenVVWCqqpLUpF4vyUnASQAbNmyY1MtKS8k8SeMxMxqyaQxMzYyGaJqNDzOjIZp2s7APe4B8oz20hfbntnb6VuDwkeUOa6ftbPodVNXZVbWpqjYddNAuL50taRXmSRqPmdEQTfOvcmZGQzPtDTkzoyGZ5u+XUX1ogFwEHN/ePx5418j0p6XxMOD69lCZ9wFHJtk/yf7Ake00SZIkTYm78EvdzGpDThqKteal91eBSfJmmpOYHphkC83VXE4HLkzydOBK4Int4u+luQTu5TSXwT0RoKquTfJi4BPtcqeunBBVkiRJk+WGnNTNerPiCVC1bNaTmbXmZdZXgXnyTmY9ZgfLFvDMnbzOucC5EyxNkiRJ25nH4FRaRGZF6m6ezcJenQRVkiRJ8+dfsqXubH5I3c07LzZAJEmS9CPzHpxKi8KsSOPpQ2ZsgEiSJKkXA1NpEbiHlDSePv1+sQEiSZK05Po0OJX6zKxI3fWxWWgDRJIkaUn1cXAq9ZFZkcbT12ahDRBJkqQl1NfBqdQ3ZkXqru/NQhsgkiRJS8SNOak78yJ1twh5sQEiSZK0BPr+VzmpTxZhQ07qk0XJjA0QSZKkgVuUgak0b2ZFGs+iZcYGiCRJ0oAt2uBUmhezInW3qHmxASJJkjRAizo4lebBvEjdLXJebIBIkiQNyCIPTKVZMy/SeBY9MzZAJEmSBmLRB6bSrJgVaTxDyYwNEEmSpAEYyuBUmiavhiSNZ2i/W2yASJIkLbChDU6laTErUndDbRb2pgGS5A+BZwAFfBY4ETgEuAC4G3AJ8NSq+kGSvYDzgQcD3wSeVFVXzKNuSZKkeXBjTurOvEjdDTkvvWiAJDkU+H3gp6rqe0kuBI4FjgbOqKoLkrwWeDrwmvbndVV1nyTHAi8DnjSn8iVJkmZqyINTaZLMijSeoWemFw2Q1h7AjyW5GbgLcBXwaOC4dv55wCk0DZBj2vsAbwNemSRVVbMsWJIkaZaGPjCVJsm8SN0tS1560QCpqq1J/hz4KvA94P00h7x8q6puaRfbAhza3j8U2Nw+95Yk19McJnPN6OsmOQk4CWDDhg3TXg1p0MyTNB4zo0kb+uDUzGhShp6VFWZGkzDUc33sTC8aIEn2p9mr4wjgW8Bbgcet93Wr6mzgbIBNmza5d4i0DuZJGo+Z0aQsy8acmdF6LduGnJnRei3L75dRvWiAAI8F/rWqrgZI8g7g54D9kuzR7gVyGLC1XX4rcDiwJckewL40J0OVJEkahGXbmJPWYxk35KT1WNbMdGqAJAnNFVqeDBxYVf8+ySOAH6+qCydQx1eBhyW5C80hMI8BLgb+HngCzZVgjgfe1S5/Ufv4o+38D3n+D0mSNBTLOjCV1sK8SN0te1667gFyKvALwJnAa9tpW4AzgHU3QKrq40neBlwK3AJ8kmZ3rvcAFyQ5rZ12TvuUc4A3JrkcuJbmijGSJEkLb9kHp1JXZkXqzr0KG10bICcAD6yqa5K8pp32r8C9JlVIVb0IeNF2k78CPGQHy94E/Nqk/m1JkqR5c2NO6sasSOMxM7fp2gDZHfh2e3/lUJO9R6ZJkiRpDRyYSt2ZF6k783JHu3Vc7m+BVyTZC350TpAXA++eVmGSJElD5+BU6s68SN2Zlx3rugfIHwJvAK4H9qTZ8+P9wNOmU5YkSdJwOTCVujMv0njMzM6t2gBJsjvNlVaOA/YB7glsrqqvT7k2SZKkwXFgKnVjVqTxmJnVrdoAqapbk7yiqs4FbgK2Tb8sSZKkYXFgKnVnXqTuzEt3XQ+BeXeSX6oqz/khSZI0Bi89KHXnhpw0HjMznq4NkDsDb0vyUWAzt10JhqryPCCSJEk74MBU6s68SN3ZXF+brg2Qz7U3SZIkdeDGnNSNG3LSePz9snadGiBV9WfTLkSSJGkIHJhK3ZkXqTvzsn6dGiBJHr2zeVX1ocmVI0mStJj8K7bUnRty0njMzGR0PQTmnO0eHwTcCdgC3GuiFUmSJC0YB6ZSd+ZF6s68TFbXQ2COGH2cZHfghcCN0yhKkiRpEbjXh9SdG3LSeMzM5HXdA+R2qurWJC+h2QPkFZMtSZIkqf8cmErdmRepO/MyPWtqgLR+AfjhpAqRJElaBO71IXXnhpw0HjMzXV1PgroZqJFJdwHuDDxzGkVJkiT1kQNTqbu15sWsaBn5+2U2uu4B8uvbPf4O8KWqumFShSTZD/gr4P40zZbfAL4IvAXYCFwBPLGqrksS4CzgaOC7wAlVdemkapEkSRrlwFTqzrxI3blX4Wx1bYD8TFX9+fYTkzynqiZ1DpCzgP9ZVU9IcieavUz+BPhgVZ2e5GTgZOCPgaOA+7a3hwKvaX9KkiRNlBtzUnfmRerOvMxe1wbInwJ3aIDQXAlm3Q2QJPsCjwBOAKiqHwA/SHIM8Kh2sfOAD9M0QI4Bzq+qAj6WZL8kh1TVVeutRZIkCfyrnDQON+Sk8ZiZ+dhlAyTJo9u7uyf5j0BGZt+LyV0G9wjgauD/TfLTwCXAs4GDR5oaXwcObu8fCmweef6WdtrtGiBJTgJOAtiwYcOESpWWk3mSxmNmFpsD09kzM4vLvMyHmVlM5mW+VtsD5Jz2552Bc0emF01D4vcmWMeDgN+rqo8nOYvmcJfb/sGqSlI7fPZOVNXZwNkAmzZtGuu5km7PPEnjMTOLyYHp/JiZxWNe5svMLBbz0g+7bIBU1REASc6vqqdNsY4twJaq+nj7+G00DZBvrBzakuQQYFs7fytw+MjzD2unSZIkjc3DXaTu3JCTxmNm+qPTOUCm3Pygqr6eZHOSf1tVXwQeA3yhvR0PnN7+fFf7lIuAZyW5gObkp9d7/g9JkrQWDkyl7syL1J3N9f7p1ABJsg9wCvBI4EBGzgVSVZM64Oz3gL9urwDzFeBEYDfgwiRPB64Entgu+16aS+BeTnMZ3BMnVIMkSVoibsxJ3ZkXqTvz0k9drwLzaprDTE4F3gT8OvBHwNsnVUhVfQrYtINZj9nBsgU8c1L/tiRJWi4OTKXuzIs0HjPTX10bIEcC/66qvpnk1qp6V5KLgXcDZ0yvPEmSpMla68DUQamWjRtx0njMTP91bYDsBlzf3v92kn1pLjl7n6lUJUmSNGEOTKXuzIvUnXlZHF0bIJ+mOf/HB4GP0BwS823gS1OqS5IkaSI8CZ3UnXmRxmPzY7F0bYD8Jred+PTZwEuB/YCpXh1GkqR52NFgxkHKYnJgKnVnXqTxmJnFs2oDJMnuwAnASwCqahvwjOmWJUnSbK02iNl48nscrCwYB6ZSN2ZFGo+ZWVyrNkCq6tYkv0tzGVxJkgZlvbt7q38cmErdmRdpPJ5Ie7F1PQTmfOC3ac79IUnSINj8GBY35KTuzIs0HjMzDF0bIA8Bfi/J84DNQK3MqKpHTKMwSZKmxcbH8DgwlbozL1J35mVYujZAXt/eJElaWDY+hsnBqdSdeZG6My/D06kBUlXnTbsQSZKmZRKNDwcy/ePAVOrOvEjjMTPD1KkBkiQ0V355MnBgVf37JI8AfryqLpxmgZIkrdWk9vhwINMvDkql7syLNB4zM2xdD4E5FfgF4Ezgte20LcAZgA0QSVKvuMfHcDkwlbozL9J4zMzwdW2AnAA8sKquSfKadtq/AveaSlWSJK2BjY/hclAqdWdepPGYmeXRtQGyO/Dt9v7KFWD2HpkmSdLceKjLsDkwlbozL1J35mX5dG2AvBd4RZI/hB+dE+TFwLsnWUyS3YGLga1V9YtJjgAuAO4GXAI8tap+kGQv4HzgwcA3gSdV1RWTrEWS1G+TvKKLg5j+cnAqdbfWvJgVLSN/vyynrg2Q5wDnAdcDe9Ls+fF+4GkTrufZwGXAPu3jlwFnVNUFSV4LPB14Tfvzuqq6T5Jj2+WeNOFaJEk95R4fw+fAVOrOvEjjMTPLq+tlcG8A/nOSuwP3BDZX1dcnWUiSw4DHAy8BntPuZfJo4Lh2kfOAU2gaIMe09wHeBrwySaqqkCQNlo2P4XNQKnVnXqTxmBl13QOEJPvRXAnmHsDXkry3qq6bYC1nAs8D7to+vhvwraq6pX28BTi0vX8osBmgqm5Jcn27/DUTrEeS1BM2PpaDA1OpO/MijcdDxAQdGyBJHg28A/gicCWwAXhVkl+tqg+ut4gkvwhsq6pLkjxqva838ronAScBbNiwYVIvK/XSjr7UJ/mFbZ40D4vc+DAz3bkhJzAzXZkXrTAz3ZgZjeq6B8grgZOq6sKVCUl+DXgV8JMTqOPngF9OcjRwZ5pzgJwF7Jdkj3YvkMOAre3yW4HDgS1J9gD2pTkZ6u1U1dnA2QCbNm3y8BgNSpcv840nv2diX9zmSbO0yI2PFWamGwemWmFmVmdeNMrM7Np6xxJmZpi6NkDuAbx9u2nvBF4/iSKq6vnA8wHaPUCeW1VPSfJW4Ak0V4I5HnhX+5SL2scfbed/yPN/aMgmebULqc+G0PhQN27ISeNx932pO3/HaGe6NkDeCDwT+IuRab9Dcynaafpj4IIkpwGfBM5pp58DvDHJ5cC1wLFTrkOaKRseWjaT+Mw7YFkcDkyl7syLNB4zo13p2gB5IPDbSZ5Hc/jJocDdgY8n+YeVharqEestqKo+DHy4vf8V4CE7WOYm4NfW+29JfWHDQ8tokp97ByyLwUGp1J15kcZjZtRF1wbI65nQ4S7SsrPZIXmoyzJyYCp1Z16k8XiImLrq1ACpqvOmXYg0RPNudvilrr6x8bF83JCTuvOkjdJ4/B2jcXXdA4QkP09zKMzeo9Or6qWTLkpaZPNueqzwS119YuNjOTkwlbozL1J3Ngu1Vp0aIEn+Engi8BHgeyOzvPKKllpfmh3gF7n6ycbH8nJ3ZKk78yJ1Z7NQ69F1D5CnAPevqq9Nsxipz2x2SN3Z+FhebshJ3bkhJ43HzGi9ujZANgPfn2YhUt/Y8JDGZ+NjeTkolcZjs1Dqzt8xmpSuDZCnA69P8mbgG6MzquofdvwUabHY8JDWbhL58XO/uByYSt2ZF2k8Ngs1SV0bIA8GjgIewR3PAbJh0kVJ02azQ5oM9/hYbm7ISeMxM1J35kXT0LUB8lLgl6rq76ZZjDQtfWl4+GWsIZhknszE4nJgKnVnXqTxmBlNS9cGyHcAD3XRwuhLwwP8EtawuMeHwN2RpXG4ISd1Z140bV0bIH8KnJnkVGDb6Iyq+uHEq5LGYLNDmj4bHwIHptI4zIs0HjOjWejaADm3/flbI9NCcw6Q3SdakbSKvjQ8/KLVMrDxoRXu9SF1Z16k8ZgZzUrXBsgRU61C2gUbHtLs2fjQCv8iJ3VnXqTx2PjQrHVqgFTVlQBJdgMOrqqrplqVllZfmh3gF6uWk40PjXJjTurOvEjjsfmheejUAEmyH/Bq4AnAzcC/SfLLwEOq6oXTK09D15eGh1+kWnaTyKI5Gg435KTuzIs0HjOjeep6CMxrgeuAewJfaKd9FHg5sO4GSJLDgfOBg2nOK3J2VZ2V5ADgLcBG4ArgiVV1XZIAZwFHA98FTqiqS9dbh6arL80O8MtTWmHjQ9vzL3JSd27ISd2td8xhZjQJXRsgjwHuUVU3JymAqro6yd0nVMctwH+tqkuT3BW4JMkHgBOAD1bV6UlOBk4G/hg4Crhve3so8Jr2p3qiT80O8AtT2p6HumhHbH5I3dj4kMZjZtQXXRsg1wMHAj8690eSDaOP16M9p8hV7f0bk1wGHAocAzyqXew84MM0DZBjgPOrqoCPJdkvySGem2R++tTw8EtS2jX3+tD2bHxI3bkhJ43HzKhPdtkASfLkqnoz8FfA25O8ANgtyc8CL6U5NGaikmwEHgh8nNufcPXrNIfIQNMc2TzytC3tNBsgM2LDQ1pM7n6qUQ5Kpe7MizQeM6M+Wm0PkNcBbwZeBnwPeBWwJ3BuO++sSRaTZG/g7cAfVNUNzak+GlVVK4ffjPF6JwEnAWzYsGGSpS6VPjU7wC/EeTFPi8lDXean75lxYKq+6XNmzIv6aIiZMS+attUaIIGm+UDT7Jhow+N2/1CyJ03z46+r6h3t5G+sHNqS5BBgWzt9K3D4yNMPa6fdTlWdDZwNsGnTprGaJ8usTw0PvwT7wzwtFhsf89fnzLgxpz7qa2bckFNf9TEz/n5R363WANk9yX+kbYTsSFV9aL1FtFd1OQe4rKpeMTLrIuB44PT257tGpj8ryQU0Jz+93vN/rJ0ND2k4bHxoVxyYSt2ZF2k8ZkaLYLUGyF40jYmdNUAKuNcE6vg54KnAZ5N8qp32JzSNjwuTPB24EnhiO++9NJfAvZzmMrgnTqCGpdGXhodfdNLk2PjQavwrttSdeZG6s/GhRbJaA+Q7VTWJBscuVdU/svMmy2N2sHwBz5xqUQPTh6aHX3DS5Nn4UBduzEndmRepO5sfWjRdL4OrBTPvhodfaNJ0eSlbdeGGnNSdeZHGY2a0iDqdBFX9Nu9mB/hFJs3CJLNuZofNv8hJ3ZkXaTw2PrTIdtkAqaq7zqoQra4PjY4VfoFJs+WhLupivZ8TPx9aNjY/pO7Mi4bAQ2B6ymaHJLDxoe4cmErdmRdpPGZGQ2EDpEf60vTwS0qaPxsfGocDU6k78yJ1556FGhobIHPSl2bHCr+cpH6w8aFxuCEndWdepPGYGQ2RDZAZ6FuzA/xSkvrIK7uoK/8iJ3VnXqTx2PjQkNkAmZK+NT38MpL6y8G5uvKzIo3HDTmpO3/HaBnYAJkAmx2S1sKBhrpy7yBpPH6/SuOxWahlYQNkTH1qdvhlIy0mN2Y1DjfkpO7MizQeM6NlYwNkFX1pePjlIi02T26qcTkolcZjZqTuzIuWlQ0Q+tPk2J5fLNLis/Ghcfh5kcbjHnXS6ia9rWNmtMiWvgHSl+aHXyTSsLghq64m+XvIz4uWgZmRujMv0u0tfQNkHvzykIbLgYbGYaNMGo+ZkebDzGgobIDMiF8a0vA5MNc8+HmRujMv0njMjIZmoRsgSR4HnAXsDvxVVZ0+55IAvyikZWPjQ/Pg50Uaj5mRujMvGqqFbYAk2R14FfALwBbgE0kuqqovzLIOvxyk5WXjQ/Pg50Uaj5mRxmNmNGQL2wABHgJcXlVfAUhyAXAMMFYD5IrTHz/WRoxfCJK86oDmwc+MNB4zI423rWNmtAwWuQFyKLB55PEW4KGjCyQ5CTgJYMOGDTt9IcMura5rnobMxofG0SUzXQamfma0LCaRGfOiZeK2jjS+RW6ArKqqzgbOBti0aVPNuRxpoS1znmx8aC26ZsbPhtQwM9J4lnlsJq3VIjdAtgKHjzw+rJ0mSevm+T0kSZKkYdlt3gWswyeA+yY5IsmdgGOBi+Zck6QBsPkhSZIkDc/C7gFSVbckeRbwPprL4J5bVZ+fc1nS0ptU82CR2fiQJEmS+mdhGyAAVfVe4L3zrkNSY9mbHzY+JEmSpP5a6AaIJPWBjQ9JkiSp/2yASNIa2fiQJEmSFocNEEkag00PSZIkaTHZAJGkDmx8SJIkSYttkS+DK6lnhtokGOp6SZIkScvEPUAkTZTNAkmSJEl95B4gkiRJkiRp8GyASJIkSZKkwUtVzbuGmUhyNXDlKosdCFwzg3ImyZpnZxHr3lnN96yqg9b6ogPIk7WtXZ/rm1dtq+ZpAJlZjyGu1xDXCWa3XmZm14a4XkNcJzAzfeF6LZZZrNdOM7M0DZAuklxcVZvmXcc4rHl2FrHuedbc5/fL2tauz/X1ubYuFr3+nRnieg1xnWDx1mvR6u1qiOs1xHWCxVuvRau3K9drscx7vTwERpIkSZIkDZ4NEEmSJEmSNHg2QG7v7HkXsAbWPDuLWPc8a+7z+2Vta9fn+vpcWxeLXv/ODHG9hrhOsHjrtWj1djXE9RriOsHirdei1duV67VY5rpengNEkiRJkiQNnnuASJIkSZKkwbMB0kryuCRfTHJ5kpPnXc+oJFck+WySTyW5uJ12QJIPJPmX9uf+7fQk+Yt2PT6T5EEzqvHcJNuSfG5k2tg1Jjm+Xf5fkhw/h5pPSbK1fa8/leTokXnPb2v+YpL/NDJ9Zp+dJIcn+fskX0jy+STPbqf35r3uQ5b6lpk+56PPOViEz/sk9CEz69G3vK1Vn3O6Vn3O93r1saauzEx/M9PWM8jc9K2ecZgZMzNRVbX0N2B34MvAvYA7AZ8GfmredY3UdwVw4HbT/gdwcnv/ZOBl7f2jgb8FAjwM+PiManwE8CDgc2utETgA+Er7c//2/v4zrvkU4Lk7WPan2s/FXsAR7edl91l/doBDgAe19+8KfKmtrRfvdV+y1LfM9Dkffc5B3z/vE/ps9CIz61yHXuVtHevR25xOeJ16ke91rlfvahqzfjPT08zsYr0WOjd9q2cN9ZsZMzOxm3uANB4CXF5VX6mqHwAXAMfMuabVHAOc194/D/iVkennV+NjwH5JDpl2MVX1D8C166zxPwEfqKprq+o64APA42Zc884cA1xQVd+vqn8FLqf53Mz0s1NVV1XVpe39G4HLgEPpz3vd5yzNLTN9zkefc7AAn/dJ6HNm1qNXv6O66HNO16rP+V6nPta0XmamB5mBweamb/VMgpkxM2tiA6RxKLB55PGWdlpfFPD+JJckOamddnBVXdXe/zpwcHu/T+sybo19qf1Z7a5m567shkYPa06yEXgg8HH681735f9wETLTl/+znelVDnr6eZ+EPtfW1SLkba2G9Fkb1at8r0EfaxqHmen352tnFjk3fatnXGamv5+tXellZmyALIaHV9WDgKOAZyZ5xOjMqiqaL4beWoQaW68B7g08ALgKePlcq9mJJHsDbwf+oKpuGJ23QO/1NC1UZvpWDz3LgZ/33luovK3VUNaDnuV7SZmZxWNu5svMLJ7eZsYGSGMrcPjI48Paab1QVVvbn9uAd9LsIvSNld252p/b2sX7tC7j1jj32qvqG1V1a1X9EHg9zXvNLmqbec1J9qTZGPzrqnpHO7kv7/Xc/w9hYTLTl/+zO+hTDnr+eZ+EPtfWyYLkba2G9FkD+pXvdehjTZ2ZGaDfn687GEBu+lbPWMwM0N/P1g71OTM2QBqfAO6b5IgkdwKOBS6ac00AJPk3Se66ch84EvgcTX3Ht4sdD7yrvX8R8LT2zMEPA64f2a1q1sat8X3AkUn2b3eTOrKdNjPbHSP4n2ne65Waj02yV5IjgPsC/4cZf3aSBDgHuKyqXjEyqy/v9dyztECZ6cv/2R30JQcL8HmfhLlnZj0WKG9rNaTPGtCffK9TH2vqxMwsXmZgELnpWz2dmRkzM3HVgzPi9uFGc6bdL9GcffYF865npK570ZwF99PA51dqA+4GfBD4F+DvgAPa6QFe1a7HZ4FNM6rzzTS7N91Mc8zW09dSI/AbNCfDuRw4cQ41v7Gt6TM0oTtkZPkXtDV/EThqHp8d4OE0u8Z9BvhUezu6T+/1vLPUx8z0OR99zsEifN4ntJ69/P3Tsfbe5W0d69LbnE54nXqR7wmsW+9q6li3melxZnaxXgufm77VM0bdZsbMTPSW9h+TJEmSJEkaLA+BkSRJkiRJg2cDRJIkSZIkDZ4NEEmSJEmSNHg2QCRJkiRJ0uDZAJEkSZIkSYNnA0SSJEmSJA2eDRBJkiRJkjR4NkAkSdJUJPl8kkfNuw5JkiSAPeZdgCRJWkxJvj3y8C7A94Fb28e/VVX3m31VkiRJO5aqmncNkiRpwSW5AnhGVf3dvGuRJEnaEQ+BkSRJU5HkiiSPbe+fkuStSd6U5MYkn03yE0men2Rbks1Jjhx57r5JzklyVZKtSU5Lsvv81kaSJC06GyCSJGlWfgl4I7A/8EngfTRjkUOBU4HXjSz7BuAW4D7AA4EjgWfMsFZJkjQwNkAkSdKsfKSq3ldVtwBvBQ4CTq+qm4ELgI1J9ktyMHA08AdV9Z2q2gacARw7t8olSdLC8ySokiRpVr4xcv97wDVVdevIY4C9gXsAewJXJVlZfjdg8yyKlCRJw2QDRJIk9c1mmivKHNjuLSJJkrRuHgIjSZJ6paquAt4PvDzJPkl2S3LvJI+cd22SJGlx2QCRJEl99DTgTsAXgOuAtwGHzLUiSZK00FJV865BkiRJkiRpqtwDRJIkSZIkDZ4NEEmSJEmSNHg2QCRJkiRJ0uDZAJEkSZIkSYNnA0SSJEmSJA2eDRBJkiRJkjR4NkAkSZIkSdLg2QCRJEmSJEmDZwNEkiRJkiQN3v8FMKJYrBLTwysAAAAASUVORK5CYII=",
            "text/plain": [
              "<Figure size 1080x216 with 5 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABEAAAADgCAYAAAADrBOyAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAAu6UlEQVR4nO3dd5glZZX48e9hBFHJgiwMDAOKAVgXFUFXRJRFBQOmRUyACQMGVncVlVUXxXWDiNmFFQVRggnRxYABdVdBSZJMKIPDCAwKkn6KMJ7fH1Xt3Gk6VHdX3apb9/t5nn763roVzr1d5623Tr9VNzITSZIkSZKkPlur7QAkSZIkSZKaZgFEkiRJkiT1ngUQSZIkSZLUexZAJEmSJElS71kAkSRJkiRJvWcBRJIkSZIk9Z4FEEmSVIuIeHtEnFTTuvaMiKvrWFdTIuLsiHhJ23FIkqRqLIBIktSCiNg9Ir4fETdFxA0R8X8R8fBJ86wXEbdGxFemWH5ZRPwpIjadNP3CiMiIWDrNdms5aR+FAoUkSdIgCyCSJA1ZRGwAfBn4ALAJsBj4F+D2SbM+s5y2d0T81RSruhJ4zsB6/xq4ZxMxS5IkjToLIJIkDd/9ATLz5MxclZl/yMyvZ+bFk+Y7CPgocDHw/CnW80ngwEnznzjdRiPiKODRwAfLkSUfLKc/MCLOKkei/Cwi9h9YZt+IuDwibomIFRHxjxFxL+ArwJblem6NiC3LRdaNiFPL+S+IiL8ZWNeWEfG5iLg+Iq6MiNcMvHaPiPhERNwYEZcDa4yGmfQ+IiLeGxErI+LmiLgkInYqX3tSOQrm5ohYHhFvH1huaTk65oXlazdGxMsj4uERcXFE/H7iMynnP7gcmfPBcqTOTyNirxnielFE/KRc79ciYpvZ4pUkScNjAUSSpOH7ObAqIk6IiH0iYuPJM5Qnz3sCnyp/Dpw8D3AOsEFEPCgiFgEHANPegyMz3wJ8D3hVZq6Xma8qixlnAZ8G7lOu48MRsUO52MeAl2Xm+sBOwLcy8zZgH+A35XrWy8zflPPvB3yGYmTLp4HTI2LtiFgL+BLwY4oRL3sBh0XEE8rl3gbct/x5AkUxZzqPB/agKCRtCOwP/K587bbys9oIeBLwioh42qTldwO2B54NHAO8Bfg7YEdg/4h4zKR5fwlsWsb4+YjYZHJAEbEf8GbgGcBmFJ/zyRXilSRJQ2IBRJKkIcvMm4HdgQSOA66PiDMiYvOB2V4AXJyZlwOnADtGxEOmWN3EKJC9gZ8AK+YYzpOBZZn58cy8MzMvBD4H/H35+h3ADhGxQWbemJkXzLK+8zPzs5l5B3A0sC7wCIoRHZtl5pGZ+afM/FX53g8ol9sfOCozb8jM5cD7Z9jGHcD6wAOByMyfZOY1AJl5dmZekpl/LkfUnAw8ZtLy78jMP2bm1ykKJidn5srMXEFRuBj8nFcCx2TmHZl5KvAzisLKZC8H/rWM5U7gXcDOZSFr2nglSdLwWACRJKkF5UnwwZm5FcXIii0pRiNMOJBi5Aflifl3mHpUxCeB5wIHM8PlLzPYBtitvPzj9xHxe+B5wMQ9R54J7AtcFRHfiYhHzrK+5RMPMvPPwNUU720biktmBrfzZmCi6LPl4LLAVdNtIDO/BXwQ+BCwMiKOLe+rQkTsFhHfLi+zuYmiMLHppFVcN/D4D1M8X2/g+YrMzElxbcldbQO8b+C93QAEsHimeCVJ0vBYAJEkqWWZ+VPgExSFECLibyku0XhTRFwbEddSXIrx3Ii426Rlr6K4Geq+wOerbG7S8+XAdzJzo4Gf9TLzFeX6f5SZ+1FcHnM6cNo065mw9cSD8rKXrYDflNu5ctJ21s/MfcvZrxlcFlgy45vIfH9mPgzYgeLSkn8qX/o0cAawdWZuSHEPlZhpXbNYHBGDyy8p389kyykuFRp8f/fIzO/PEq8kSRoSCyCSJA1ZedPR10fEVuXzrSm+zeWccpaDKO7LsQOwc/mzE3APintvTPZi4HHlvTlmcx2w3cDzLwP3j4gXlPfqWLu8KeiDImKdiHheRGxYXtJyM/DngfXcOyI2nLT+h0XEM8pCzWEU32JzDvBD4JaIeGN5w9NFEbFTrP7q39MoCj4bl5/Lq6d7A2V8u0XE2hSXsPxxIK71gRsy848RsSvF6JiFuA/wmvJz+XvgQcCZU8z30TL+HcsYNyznny1eSZI0JBZAJEkavlsoRnScGxG3URQILgVeHxHrUtwP4wOZee3Az5UUl7vc5TKYzPxlZp5XcdvvA55VflPJ+zPzFoqbdB5AMbLhWuDfgLuX878AWBYRN1NcTvK8cps/pbi/xq/Kyz4mLgv5IsXNRW8sl31Gef+MVRT3G9mZYsTKb4H/prgpKBRfAzwxmuXr5XudzgYU9w+5sVzmd8B/lK+9EjgyIm4B3srqESvzdS7FaJzfAkcBz8rMu9zANDO/QPG5nVJ+Vpeyulg1U7ySJGlIYs3LWiVJkgTF1+ACL8nM3duORZIkLZwjQCRJkiRJUu9ZAJEkSZIkSb3nJTCSJEmSJKn3HAEiSZIkSZJ6zwKIJEmSJEnqPQsgkiRJkiSp9yyASJIkSZKk3rMAIkmSJEmSes8CiCRJkiRJ6j0LIKosInaPiO9HxE0RcUNE/F9EPLx87bkRcVVE3BYRp0fEJgPL3TrpZ1VEfKB8bWlE5KTX/3mKbW8SEddHxP8O7x1LC9dG3kTEJyLiT5NeXzT8dy/NXVvHmoj4u4i4oFz31RGx/3DfuTR/LR1rLpv02p0R8aXhv3upHi3l0SYRcWpE/C4ifhsRn4qIDYb/7sdHZGbbMWgElIn4a+AVwGnAOsCjgWuBVcA5wJOAC4BjgbUy84Ap1rNeucy+mfndiFgKXAmsnZl3zrD944AHlOvdvca3JjWmrbyJiE8AV2fmEQ28LakxLebMDsDZwEHAWcCGwEaZ+cua36JUu7b7aOWyAfwKeFtmnljTW5OGpsXjz4eB+wHPAgL4HHBxZr6u7veowt3aDkAj4/4AmXly+fwPwNcBIuJdwJcy87vl838GfhIR62fmLZPW80xgJfC9qhuOiL8FdqJobF68kDchDVlreSONqLZy5gjgvzLzK+Xz35U/0ijowrFmD2BTipM3aRS1lUfbAqdn5s3lur8APHUhb0Qz8xIYVfVzYFVEnBAR+0TExgOv7Qj8eOJJ+R+zP1E2JJMcBJyYdx16dFU55PjjEbHpxMQohu1/EHgV4HAljZpW8qb0ynL45vkR8cwa3os0DG3lzCMAIuKSiLgmIk4aHN4sdVybx5rBZT+XmbfN/21IrWorjz4EPDkiNi63+UzgK6gxFkBUSVmV3J2iCHEccH1EnBERmwPrATdNWuQmYP3BCRGxDfAY4ISByb8FHg5sAzysXOZTA6+/Bjg3M8+v791Iw9Fi3rwf2B64D/DPwCci4lE1vS2pMS3mzFbACyg6ntsD9wA+UM+7kprVYt5MLHtPiuH7n6jh7UitaDGPLqC43GZi5OEq4MP1vCtNxQKIKsvMn2TmwZm5FcUlKVsCxwC3ApNv1rMBMHlI2AuA/83MKwfWeWtmnpeZd2bmdRQjPR4fEetHxJYUBZC3NPOOpOYNO2/K1y/IzN+Vr59JcaB9RhPvT6pbGzlDMdT545n588y8FXgXsG/d701qSkt5M+EZwA3Ad2p7Q1ILWsqj0yhGn6xfrvOXwEn1vjMNsgCiecnMn1JU+ncCLgP+ZuK1iNgOuDtFMg86kDUrolOuuvy9FrArsAVweURcC7wP2DUirg2/0UIjaEh5M93rMcdwpdYNMWcuHpjGpMfSSGnhWDPdkH9pZA0xj3amuAfVbWUB/qNYgG+UBRBVEhEPjIjXR8RW5fOtgedQ3BH5U8BTIuLREXEv4Ejg84M3BSpvZLoY+Myk9e4WEQ+IiLUi4t4UQ/fPzsybKK5/W0rRMOwMvBW4ENg5M1c1+X6lOrSUN0TEsyJivfL1xwPPB85o/h1LC9NWzgAfB14YEduVw/kPB77c7LuV6tFi3lBu87HMftIndVqLefQj4CURcY+IuAdwCEVRXg2xAKKqbgF2A86NiNsoGoNLgddn5mXAyykah5UUQ7heOWn5g5jUUJS2A75arv9S4HaKxobMvD0zr534objW7o7ysTQKhp43pdcCK4DfA/8BvDQzz67tXUnNaSVnMvN44ETgXOCq8vXX1PrOpOa0dayBYsj/D9KvjNboayuPXkTxD9+rKfpu25XrUkPC0WqSJEmSJKnvHAEiSZIkSZJ6zwKIJEmSJEnqPQsgkiRJkiSp9yyASJIkSZKk3rMAIkmSJEmSeu9ubQcwLJtuumkuXbq07TCkTjj//PN/m5mbzTafeSOtViVvzBlpNY810txVzZvJzCNptZnyaGwKIEuXLuW8885rOwypEyLiqirzmTfSalXyxpyRVvNYI81d1byZzDySVpspj7wERpIkSZIk9d7YjACRJEmSpPk64vRLOOmcX9eyrmXvflIt65FGydLD/6fW9c0njyyASJIkSeq1uk+8Fmrp4f9jEUQjpWs5BPPLIwsgkiRJkkZSF0/KpFE1DvlkAUSSJElSp43DiZk0LOOcTxZAJEmSJHXKOJ+gSXUzn1azACJJkiSpVZ6gSfUyp6ZmAUSSJElSK8b1JM0boKoJ45ZPfguM1JBRbUw8uEprqvIVhuaNBPd70/9wZ1ab15zRXI1qv6oK80Ft6ENODSt3LIBorPWhsZiJX7GmcbTQvDZvNG5Ov3AFh5160byXN2dUVRf7Xe67GmVdyKlRyyELIBorXWgkJNXHnJbmx9zRMLWxv43aSZk0F8PMqb7lkgUQ9Z6dPKk/zGdp/swftaHJ/a5vJ2ZSFebUwgy1ABIRxwNPBlZm5k7ltE2AU4GlwDJg/8y8MSICeB+wL/D/gIMz84JymYOAI8rVvjMzTxjm+9BosKMnjT7zWFo480htqXvfG4eTM2kmdebUuObTsEeAfAL4IHDiwLTDgW9m5rsj4vDy+RuBfYDty5/dgI8Au5UFk7cBuwAJnB8RZ2TmjUN7F+o0O3rS6DOPpYUzj9SmOva/cT1Bk6ZiTtVjqAWQzPxuRCydNHk/YM/y8QnA2RQFkP2AEzMzgXMiYqOI2KKc96zMvAEgIs4Cngic3HT86jY7endlI6dR04U83nz9ddoOQVowrw9Xmxa6/7lPSWtaSE6ZT2vqwj1ANs/Ma8rH1wKbl48XA8sH5ru6nDbddI0xb64ljbYuFD6gKH6c+5a92w5Dmrdh5JLHP81kvvug+5U0NXOqXl0ogPxFZmZEVPzW+dlFxCHAIQBLliypa7XqGK8vrZd5o2Eb9f9UmzPqiiZyqaljonnTT/PZB8e937UQ5lG/WfhoRhcKINdFxBaZeU15icvKcvoKYOuB+bYqp61g9SUzE9PPnmrFmXkscCzALrvsUlthRd1RV2fPhmI180bDcvqFKzjs1IsaWfcwc9qcUReM2vHQvOmfue6D9r0WzjzqL4uJzelCAeQM4CDg3eXvLw5Mf1VEnEJxE9SbyiLJ14B3RcTG5XyPB9405JjVAV4LJ40uR25J9fF+C2qbxQ+pPhY/mjXsr8E9mWL0xqYRcTXFt7m8GzgtIl4MXAXsX85+JsVX4F5B8TW4LwTIzBsi4h3Aj8r5jpy4IarGh0PCpNHlXcyl+vjPALXJEzWpXuZU84b9LTDPmealvaaYN4FDp1nP8cDxNYamEWLDII0uT9ak+vjPALXJ/phUL3NqOLpwCYxUmUMspdH1vON+MK/lzGPpruwoq03uf1K9zKnhWavtAKSqLH5Io+3/fjn3qxXNY+mu7CirTe5/Ur3MqeFyBIg678Fv+yo3375qTsvYKEjdYgFTqoe5pDZ5oibVyzZ9+CyAqNM80Eqjz4O71A5zSXWyTybVy/5RO7wERp3lgVYafR7cpfrMJZ/MJdXJPplUL/tH7XEEiDrJA600+jy4S/U5/cIVlec1l1Qn23KpXuZUuxwBos6x+CGNH3NYmtlhp15UaT5zSXXyRE2qlznVPgsg6hSLH1I/OFRfqs8Rp19Sab7nP2JJw5FonHiiJtXLnOoGCyDqDIsf0vgxh6XZnXTOryvN986n/XXDkWhceKIm1cuc6g4LIOoEix9Sf1TNZ3NYmp35pGHzRE2qlznVLZVughoRAbwEeA6waWY+OCL2AP4qM09rMkD123wKH2DDIHVV1Zw+5tk7NxuIJGnOPFGT6mVOdU/Vb4E5EtgbOAb4aDntauC9gAUQzYujPqTx9bSHLG47BKnzHP2hYbFPJtXP4kc3Vb0E5mDgyZl5CpDltCuB7ZoISv3ngVbqH0/WJGn02CeT6mfxo7uqFkAWAbeWjycKIOsNTJMq80Ar9c/pF65oOwSpVywoahjsk0n1s/jRbVULIF8Bjo6Iu8Nf7gnyDuBLTQWmfvJAK/XTYadeVGk+81mSusE+mVQ/ix/dV7UA8g/AXwE3ARtSjPzYBnhjQ3GphzzQSv3kf6qleplTapp9Mql+Fj9Gw6w3QY2IRcCzgOcCG1AUPpZn5rUNx6Ye8UArSZLUPvtkUr3MqdEy6wiQzFwFHJ2Zf8zMlZn5I4sfmgsbBam//E+1VC9zSk2yTybVy5waPVW/BvdLEfGUzPSeH5oTh4JJ/fXgt3217RCksfSo+27SdggaQfbJpHpZ/BhNVQsg6wKfjYgfAMtZ/U0wZOaBTQSm0eeBVuq3m29fVWk+c1uqpupx81MvfWTDkahv7JNJ9bL4MbqqFkAuLX8aEREPAE4dmLQd8FZgI+ClwPXl9Ddn5pnlMm8CXgysAl6TmV9rKj7NnQdaqd8cpi+1w5zSXNknk+pl8WO0VSqAZOa/NBlEZv4M2Bn+ctPVFcAXgBcC783M/xycPyJ2AA4AdgS2BL4REfcv71eilnmglSRpbubToZZmY59MqpfFj9FXqQASEY+b7rXM/FZ94QCwF/DLzLwqIqabZz/glMy8HbgyIq4AdgV+UHMsmiMPtFL/OfpDaoc5pbmwTybVy+JHP1S9BOZjk55vBqwDXE1xuUqdDgBOHnj+qog4EDgPeH1m3ggsBs4ZmOfqcppa5IFW6j//Sy3Vz7xS3eyTSfWy+NEfVS+B2XbweXmZyhHALXUGExHrAE8F3lRO+gjwDoqbrr4DeA/wojms7xDgEIAlS5bUGaom8UDbH+aN6jBOOW7OaBj6llPmTXPsk40P82g4zKl+WWs+C5X32jgKeEO94bAPcEFmXldu57rMXJWZfwaOo7jMBYp7hGw9sNxW5bTJcR6bmbtk5i6bbbZZzaFqgo1Cv5g3mo6XvkzNnNFCjOvoD/OmGfbJxot51Dxzqn/mVQAp7Q38ua5ASs9h4PKXiNhi4LWns/qbaM4ADoiIu0fEtsD2wA9rjkUV2ChIktQsj52qwj6ZVC9zqp+q3gR1OcVlKBPuCawLHFpXIBFxL4qiyssGJv97ROxcbnvZxGuZeVlEnAZcDtwJHOo3wAyfjYI0Phz9IdVvXEd/qF5HnH4JJ53z6zktY1stzczznP6qehPU5096fhvw88y8ua5AMvM24N6Tpr1ghvmPorgMRy2wUZDGR9V8f/4jvP5YqpvHT83EGzNK9fM8p9+qFkAenpn/OXliRLwuM4+uOSZ1nI2CpKm882l/3XYI0shw9IcWyuKHVD/Pc/qv6j1A3jrN9CPqCkSjwUZBGi9e+iK1x7zSdCx+SPXzPGc8zDgCJCIeVz5cFBGPBWLg5e2o+Wtw1W02CpIkLZyjP7QQFj+k+nmeMz5muwTmY+XvdYHjB6YncC3w6iaCUvfYKEjjx9EfUnvMK03F4odUP89zxsuMBZDM3BYgIk7MzAOHE5K6xkZBGj9V837dRTH7TJL+wtEfmi+LH1L9PM8ZP5VugmrxY3zZKEiayU+P2rftEKTe8ViqyeyPSfUzr8ZTpQJIRGwAvB14DLApA/cCyUy/97CnbBSk8eSlL1IzHP2h+bA/JtXPvBpfVb8F5sPAQ4EjgU0o7v3xa+C9DcWlltkoSJI0fB5PNcj+mFQ/82q8VRoBAjweeFBm/i4iVmXmFyPiPOBLWATpHRsFaXw5+kNqhqM/NFf2x6T6mVeqOgJkLeCm8vGtEbEhcA1wv0ai0siwUZD6o2qn4Jhn79xsINKY8piqCZ6kSfUzrwTVR4D8mOL+H98EvkdxScytwM8bikstmUvDYKMgjaenPWRx2yFII8XRH5oLT9Kk+plXmlB1BMhLgWXl49cCfwA2Avx2mB6x+CGNLy99kdplbgk8SZOaYF5p0KwjQCJiEXAwcBRAZq4EXtJsWBo2ix+SJNXP0R+qypM0qX7mlSabdQRIZq4CXgnc0Xw4aoPFD2m8OfpDape5JU/SpPqZV5pK1UtgTgRe3mQgaof/mZLGm8UPqTkeY1WFJ2lS/cwrTafqTVB3BV4dEW8AlgM58UJm7tFEYGqeDYMkSe3y2Dq+djvqLK675U9zWsb9RZqd5ziaSdUCyHHlj3rChkGSoz+k5jj6QzOZz/5hWyzNznMczaZSASQzT2g6EA2PDYMkSe3z+DqeLH5IzfAcR1VUugdIFF4aEd+KiIvLaXtExP7Nhqe62TBIAkd/SE1y9IemY/FDaobnOKqq6k1QjwReDBwLLCmnXQ28sYmg1AwbBklg8UPqAvNr/Fj8kJrhOY7momoB5GDgyZl5CqtvgHolsF1dgUTEsoi4JCIuiojzymmbRMRZEfGL8vfG5fSIiPdHxBURcXFEPLSuOFSwYZAkae4c/aGpzHW/2Hz9deyLSRVY/NBcVb0J6iLg1vLxRAFkvYFpdXlsZv524PnhwDcz890RcXj5/I3APsD25c9uwEfK35qCDYOkCY7+kNpnfo2X+RQ/zn3L3g1FI/WH5ziaj6oFkDOBoyPiH6AYgQG8A/hSU4GV9gP2LB+fAJxNUQDZDzgxMxM4JyI2iogtMvOahuMZOTYMkiacfuGKtkOQes3RH5rMfpjUDHNL81X1EpjXAVsANwEbUoz82IZ67wGSwNcj4vyIOKSctvlAUeNaYPPy8WJg+cCyV5fT1hARh0TEeRFx3vXXX19jqKPBhkHzMe5502eHnXpRpflsC+bGnNFcmF+Fccgb+2Fq2jjk0VTMLS1EpQJIZt6cmU+nuAHqI4D7ZubTM/OWGmPZPTMfSnF5y6ERscekGJLVl99UkpnHZuYumbnLZpttVmOo/WPDoAnmTT956UtzzBmBoz/mqu954wmahqHveTQVc0sLVXUECBGxEbA3xSUpe03ckLQumbmi/L0S+AKwK3BdRGxRbn8LYGU5+wpg64HFtyqnqWRHTJKkbrEjPh48QZOaYW6pDpUKIBHxOGAZ8Brg4cCrgSsjYq86goiIe0XE+hOPgccDlwJnAAeVsx0EfLF8fAZwYPltMI8AbvL+H6vZOEga5OgPqVn+00ET7INJzTC3VJeqN0H9IHBIZp42MSEi/h74EPDAGuLYHPhCcW9V7gZ8OjO/GhE/Ak6LiBcDVwH7l/OfCewLXAH8P+CFNcTQCzYOkiR1j8fb/rMPJjXD3FKdqhZAtgQ+N2naF4Dj6ggiM38F/M0U038H3GWUSXk/kEPr2HafzKVxsGGQxoOjP6RmOfpD4Ama1BRzS3Wreg+QT3LXgsMrgBPrDUfzZQdM0mQWP6RuMMf6zRM0qRnmlppQdQTIQ4CXR8QbKG42uhi4D3BuRHx3YqbM3GOa5dWQ+RQ+bBwkSVo4//kgT9CkZphbakrVAshx1HS5i+pj8UPSdBz9IXWDOdZfnqBJzTC31KRKBZDMPKHpQDQ3Fj8kTcf/SkvNM8/GmydoUjPMLTWt6ggQIuLRFJfCrDc4PTPfVXdQmt58O1w2DpIms12QmmWO9ZMnaFIzzC0NQ6UCSER8gOIraL8H/GHgpWwiKE3N4oek2Xjpi9Q8R3+ML0/QpGaYWxqWqiNAngfslJm/aTIY1c/GQZKk4fP42z+eoEnNMLc0TFW/Bnc5cHuTgah+Ng7SeHH0h9Q8R3+MJ0/QpGaYWxq2qiNAXgwcFxEnA9cNvpCZ3516EdXJxkHSTDwpk7rDY3C/2AeTmmFuqQ1VCyAPA/YB9uCu9wBZUndQWpMnNpLqYudBmj+Px+PHEzSpGeaW2lK1APIu4CmZ+Y0mg9FdHXH6JXNexgZCGi9e+iJ1h3nWH56gSc0wt9SmqgWQ2wAvdWnBSef8uvK8Ng6SJDXD0R/jxRM0qRnmltpWtQDyVuCYiDgSWDn4Qmb+ufaoBMytgbBxkMaToz+k7jDPRt+D3/ZVbr591ZyW8e8uVWPxQ11QtQByfPn7ZQPTguIeIItqjUiAxQ9Js7P4IQ2Hoz/Gw3z+zravUjUWP9QVVQsg2zYahdZg8UOSpNHi8Xi0WfyQmmPxQ11SqQCSmVcBRMRawOaZeU2jUY0xix+SqnD0hzQcjv7oP4sfUnMsfqhr1qoyU0RsFBGfBv4IXFFOe2pEvLPJ4MaNnSxJkkaPHfbRZfFDao7FD3VRpQII8FHgJmAb4E/ltB8Az24iqHFkAyGpKkd/SMPhPyb6zeKH1BzPbdRVVe8BshewZWbeEREJkJnXR8R9mgttfNhASKrK4ofULebaaLL4ITXHcxt1WdURIDcBmw5OiIglQC33AomIrSPi2xFxeURcFhGvLae/PSJWRMRF5c++A8u8KSKuiIifRcQT6oijDTYQkiR1j6M/+ms+fS/7X1I1ntuo62YsgETEc8qH/w18LiIeC6wVEY8ETqC4NKYOdwKvz8wdgEcAh0bEDuVr783MncufM8u4dgAOAHYEngh8OCJG7ut4bSAkzYWjP6RuMddGj30vqTnml0bBbCNA/qv8/W/AqcCHgLWB44EvAu+rI4jMvCYzLygf3wL8BFg8wyL7Aadk5u2ZeSXFjVl3rSOWYbGBkCSpmxz90U/2vaTmmF8aFbMVQAIgC+/LzB0y816Z+aDMPCYzs+6AImIp8BDg3HLSqyLi4og4PiI2LqctBpYPLHY1MxdMOsUGQtJcOfpD6hZzbbTY95KaY35plMx2E9RF5WUvMd0MmfmtuoKJiPWAzwGHZebNEfER4B1Alr/fA7xoDus7BDgEYMmSJXWFuSA2EOq6LubNuLP40W3mTL84+mM4hpk39r3UV104/phfGjWzFUDuDnyM6QsgCWxXRyARsTZF8eNTmfl5gMy8buD144Avl09XAFsPLL5VOW3N4DKPBY4F2GWXXWofrTJXNhAaBV3LG6nrzJnx4/F54YaVN/a91GdtH3/ML42i2Qogt2VmLQWOmUREUBRafpKZRw9M3yIzJ75p5unApeXjM4BPR8TRwJbA9sAPm45zIWwgJM2Hoz+k4XH0R7/Y95KaY35pVM1WABmWRwEvAC6JiIvKaW8GnhMRO1OMNFkGvAwgMy+LiNOAyym+QebQzFw15Jgrs4GQJKkfPEaPBvteUnPML42y2Qog0977o06Z+b/TbOvMGZY5CjiqsaBaYgMhaYKjP6ThcfRHf3hyJjXH/NKom7EAkpnrDyuQvppLI2EDIWnC3kefXWk+2w1peMy3bptPEcu/qVSdxQ/1wWxfg6sFsPghab5+sfK2tkOQxoajP0afxQ+pWRY/1BcWQBpi8UPSfHnpi9Q95lt3WfyQmmXxQ31iAaQBFj8kzdfpF97lG70lNcjRH6PN4ofULIsf6puufAtMb9iRkrQQh516UaX57GBIC1e14Gi+dc98+1v+LaVqjjj9Ek4659dzWsb80iiwAFIjK6SSFsICqjRcVQuO6haLH1KzHFmlPrMAUhOLH1L37H302b28majth8bJfP4LWSfzrVvmc2J2t4Ar/tW/o1SFxQ/1nQWQGlj8kKpr+2Rm1Nl+qOscyaSmeGImNcsc0ziwALJAFj+k6XkiJPVfn/PcY3Y3eMmL1DyLHxoXFkAWwOKHNLU+nxC1yTZEXWKeaxgsfkjNs/ihcWIBZJ4sfkhT86SoGbYh6opxyXFzrn0WP6TmWfzQuLEAMg8WP6SpjcuJ0bDZhqgrxiXHzbnR5N9Nqs6vudW4sgAyRxY/pKmNy4nRsNmGSMNlzo0m/25SdY760DizADIHFj+kqY1r8SOABBZvdA/+6QkP4GkPWdx2SFJj+prn29/nXpz1uj3bDkMLYH9Lqs7ih8adBZCKLH5IU6v7pMjckbqnrjw/5tk7WyhUrTxmSNVZ/JAsgMzKhkKa3ukXrpjXcuaI1G/muIbB/UyqxhsKS6tZAJmBxQ9peo6KkuamztFSy979pErrmy7vZlp28jJzids8l6RusfghrWmttgPoKosf0vQsfkhzU/elYlXXN9V8sy07+LrFD0kaP7bn6jNHgExilVSql7kh9Z95Lkmjz7Zc48ACSGkh/52zsZCmZm5I/WeeS9Losy3XuBjpS2Ai4okR8bOIuCIiDp/veix+SJIkSRpHns9onIxsASQiFgEfAvYBdgCeExE7zHU9Fj+kZpgfUv+Z55IkaZSMbAEE2BW4IjN/lZl/Ak4B9hvGhpe9+0l2+jTWZtv/zQ9pTXXnRNX1TTXfXPJ3pnnNc0mSNGpG+R4gi4HlA8+vBnYbnCEiDgEOAViyZEktG7XDp76rmjfmglRoK2cWsr65LGuuqwlN9NGkcWMeSXM3yiNAZpWZx2bmLpm5y2abbbbg9dkJ1DioO2+kvjNnpLkzb6SFM4+kuRvlESArgK0Hnm9VTqudhQ9JkiRJkkbbKI8A+RGwfURsGxHrAAcAZ8x1JbNd32zxQ5Ikafi8B420cN63TVrTyI4Aycw7I+JVwNeARcDxmXnZfNZl4kuSJHWPfTRp4cwjabWRLYAAZOaZwJltxyFJkiRJkrptlC+BkSRJkiRJqiQys+0YhiIirgeummW2TYHfDiGc2RjHmoyj/hi2ycxZbxc+YnkzE2OszyjE2VSMs+bNCOZMV2LpShzQnVi6EgfMP5Y+H2u6FE+XYgHjmc1s8VTKm8lGMI+6EktX4gBjmUrtx5+xKYBUERHnZeYuxmEcXY2jCzFM1sWYJjPG+oxCnF2PsUvxdSWWrsQB3YmlK3FAN2LpQgyDuhRPl2IB45lNm/F06bPoSixdiQOMZVhxeAmMJEmSJEnqPQsgkiRJkiSp9yyArOnYtgMoGceajGO1LsQwWRdjmswY6zMKcXY9xi7F15VYuhIHdCeWrsQB3YilCzEM6lI8XYoFjGc2bcbTpc+iK7F0JQ4wlqnUHof3AJEkSZIkSb3nCBBJkiRJktR7FkBKEfHEiPhZRFwREYc3vK1lEXFJRFwUEeeV0zaJiLMi4hfl743L6RER7y/jujgiHrqA7R4fESsj4tKBaXPebkQcVM7/i4g4qKY43h4RK8rP5KKI2HfgtTeVcfwsIp4wMH1Bf7OI2Doivh0Rl0fEZRHx2mF/JjPEMPTPY66Gvb0ptt+J/XmWGFvfxyrEuG5E/DAiflzG+C/l9G0j4twyllMjYp1y+t3L51eUry8dWNeU+2aNsS6KiAsj4stdjbHCe2g0b7qSF13a97u2j3dlP46a+iIL/ftUiLOVY01dn88Ctt+JXJ4lnlb6KtGh9mWWeDrVlxt2LrWZQ13Kn67kTlfyphP5kplj/wMsAn4JbAesA/wY2KHB7S0DNp007d+Bw8vHhwP/Vj7eF/gKEMAjgHMXsN09gIcCl853u8AmwK/K3xuXjzeuIY63A/84xbw7lH+PuwPbln+nRXX8zYAtgIeWj9cHfl5ub2ifyQwxDP3z6HLOdHl/7vo+ViHGANYrH68NnFtu+zTggHL6R4FXlI9fCXy0fHwAcOpM+2bNf/PXAZ8Gvlw+71yMbedNV/KiS/t+1/bxruzH1NAXqePv03bONPn5LHD7ncjlWeJ5Oy30VehQ+zJLPK18Pl3JpTZzqEv505Xc6UredCFfHAFS2BW4IjN/lZl/Ak4B9htyDPsBJ5SPTwCeNjD9xCycA2wUEVvMZwOZ+V3ghgVu9wnAWZl5Q2beCJwFPLGGOKazH3BKZt6emVcCV1D8vRb8N8vMazLzgvLxLcBPgMUM8TOZIYbpNPZ5zFHrOdOV/XmWGFvfxyrEmJl5a/l07fIngccBn50mxonYPwvsFRHB9PtmLSJiK+BJwH+Xz6NrMVbQeN50JS+6tO93aR8fgf24M21TqfVjzSSN99UmdCWXZ4lnOo32VbrUvswSz3Ta6Mt1JZeGkkNdyp+u5E5X8qYL+WIBpLAYWD7w/Gpm/kMsVAJfj4jzI+KQctrmmXlN+fhaYPMhxTbX7TYZz6vKIVbHTwy/GlYcUQwtfgjFfwZb+UwmxQAtfh4VDHt7VXVpf15DF/axGWJbFBEXASspDmS/BH6fmXdOsb2/xFK+fhNw76ZjBI4B3gD8uXx+7w7GOJu2tt/q/taFfb9D+/gxdGc/rqMv0vQ+3WbOdqmvNqFrfx9oua/ShfZlhnigO325NnKpaznU+v4xyVif90wRBwzpM7EA0o7dM/OhwD7AoRGxx+CLmZkUjcZQtbXd0keA+wI7A9cA7xnWhiNiPeBzwGGZefPga8P6TKaIobXPoy9a3p/X0IV9bCaZuSozdwa2oqioP7DNeCaLiCcDKzPz/LZjGXXD3t+6su93YR/v4H7cyb5Ih3T682l7+6VW+ypdaV9miGfc+3KdzaEO5M9Yn/dME8fQPhMLIIUVwNYDz7cqpzUiM1eUv1cCX6DojF03MdSr/L1ySLHNdbuNxJOZ15Ud1D8Dx7F6OG+jcUTE2hTJ96nM/Hw5eaifyVQxtPV5zMGwt1dVJ/bnQV3Yx6rKzN8D3wYeSTHU8W5TbO8vsZSvbwj8ruEYHwU8NSKWUQxxfBzwvo7FWEVb229lf+vivt/yPt6p/bimvkjT+3RrOduxvtqETv192uyrdK19GYG+3NBzqYM51Jn8GefznuniGOpnkg3e/GZUfoC7UdzAZVtW30Rlx4a2dS9g/YHH36e4buo/WPMGNP9ePn4Sa96A5ocL3P5S1rwJz5y2S3HDmyspbnqzcfl4kxri2GLg8T9QXOsFsCNr3vjmVxQ3vVnw36x8bycCx0yaPrTPZIYYhv55dDVnRmF/7vI+ViHGzYCNysf3AL4HPBn4DGvemPGV5eNDWfPGjKfNtG828Dffk9U3j+xkjDPEPpS86UJedGnf7+I+3vZ+TE19kTr+Pl3ImaY+nxriWEqHjnFTxNNKX4UOtS+zxNOZvtywc6kLOdSl/OlC7nQlb7qQL43s9KP4Q3Gn259TXBf8lga3s135B/oxcNnEtiiu6/0m8AvgGxM7UrmTfKiM6xJglwVs+2SKIUV3UFwn9eL5bBd4EcUNaK4AXlhTHJ8st3MxcMakJHhLGcfPgH3q+psBu1MM87oYuKj82XeYn8kMMQz98+hqznR9f+76PlYhxgcDF5YxXgq8tZy+HfDDcnufAe5eTl+3fH5F+fp2s+2bNX+me7L6xLGTMbaZN13Jiy7t+13cx9vej6mxL7LQv0/bOdP057OAGDqRy7PE00pfhQ61L7PE06m+3DBzqe0c6lL+dCV3upI3XciXKBeWJEmSJEnqLe8BIkmSJEmSes8CiCRJkiRJ6j0LIJIkSZIkqfcsgEiSJEmSpN6zACJJkiRJknrPAogkSZIkSeo9CyCSJEmSJKn3LIBIkqRGRMRlEbFn23FIkiQB3K3tACRJ0miKiFsHnt4TuB1YVT5/WWbuOPyoJEmSphaZ2XYMkiRpxEXEMuAlmfmNtmORJEmaipfASJKkRkTEsoj4u/Lx2yPiMxFxUkTcEhGXRMT9I+JNEbEyIpZHxOMHlt0wIj4WEddExIqIeGdELGrv3UiSpFFnAUSSJA3LU4BPAhsDFwJfo+iLLAaOBP5rYN5PAHcC9wMeAjweeMkQY5UkST1jAUSSJA3L9zLza5l5J/AZYDPg3Zl5B3AKsDQiNoqIzYF9gcMy87bMXAm8FzigtcglSdLI8yaokiRpWK4bePwH4LeZuWrgOcB6wJbA2sA1ETEx/1rA8mEEKUmS+skCiCRJ6prlFN8os2k5WkSSJGnBvARGkiR1SmZeA3wdeE9EbBARa0XEfSPiMW3HJkmSRpcFEEmS1EUHAusAlwM3Ap8Ftmg1IkmSNNIiM9uOQZIkSZIkqVGOAJEkSZIkSb1nAUSSJEmSJPWeBRBJkiRJktR7FkAkSZIkSVLvWQCRJEmSJEm9ZwFEkiRJkiT1ngUQSZIkSZLUexZAJEmSJElS71kAkSRJkiRJvff/AWYxZq03nVzsAAAAAElFTkSuQmCC",
            "text/plain": [
              "<Figure size 1080x216 with 5 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Plot time vs. temperature for both sample types\n",
        "\n",
        "# For commercial\n",
        "fig, ax = plt.subplots(1, 5, figsize=(15, 3), constrained_layout=True, sharey=\"all\")\n",
        "fig.supylabel(\"Temperature\")\n",
        "fig.supxlabel(\"Time\")\n",
        "fig.suptitle(\"Commercial samples\")\n",
        "\n",
        "for i in range(0, 5):\n",
        "    sample_lab = sample_id_commercial[i]\n",
        "    plt.subplot(1, 5, i + 1)\n",
        "    plt.scatter(\n",
        "        sample_commercial_dict[sample_lab][\"time\"],\n",
        "        sample_commercial_dict[sample_lab][\"temp\"],\n",
        "    )\n",
        "    plt.title(sample_lab)\n",
        "\n",
        "# For SAM testbed\n",
        "fig, ax = plt.subplots(1, 5, figsize=(15, 3), constrained_layout=True, sharey=\"all\")\n",
        "fig.supylabel(\"Temperature\")\n",
        "fig.supxlabel(\"Time\")\n",
        "fig.suptitle(\"SAM testbed samples\")\n",
        "\n",
        "for i in range(0, 5):\n",
        "    sample_lab = sample_id_testbed[i]\n",
        "    plt.subplot(1, 5, i + 1)\n",
        "    plt.scatter(\n",
        "        sample_testbed_dict[sample_lab][\"time\"],\n",
        "        sample_testbed_dict[sample_lab][\"temp\"],\n",
        "    )\n",
        "    plt.title(sample_lab)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 465
        },
        "id": "MV3OcSo6Sbtc",
        "outputId": "9b90ba37-f76e-4f15-e64d-077a42190560"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1080x216 with 5 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1080x216 with 5 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "# For commercial\n",
        "fig, ax = plt.subplots(1, 5, figsize=(15, 3), constrained_layout=True)\n",
        "fig.suptitle(\"Commercial samples\")\n",
        "fig.supxlabel(\"Temperature\")\n",
        "fig.supylabel(\"Abundance\")\n",
        "\n",
        "for i in range(0, 5):\n",
        "    sample_lab = sample_id_commercial[i]\n",
        "    sample_df = sample_commercial_dict[sample_lab]\n",
        "\n",
        "    plt.subplot(1, 5, i + 1)\n",
        "    for m in sample_df[\"m/z\"].unique():\n",
        "        plt.plot(\n",
        "            sample_df[sample_df[\"m/z\"] == m][\"temp\"],\n",
        "            sample_df[sample_df[\"m/z\"] == m][\"abundance\"],\n",
        "        )\n",
        "\n",
        "    plt.title(sample_lab)\n",
        "\n",
        "# For testbed\n",
        "fig, ax = plt.subplots(1, 5, figsize=(15, 3), constrained_layout=True)\n",
        "fig.suptitle(\"SAM testbed samples\")\n",
        "fig.supxlabel(\"Temperature\")\n",
        "fig.supylabel(\"Abundance\")\n",
        "\n",
        "for i in range(0, 5):\n",
        "    sample_lab = sample_id_testbed[i]\n",
        "    sample_df = sample_testbed_dict[sample_lab]\n",
        "\n",
        "    plt.subplot(1, 5, i + 1)\n",
        "    for m in sample_df[\"m/z\"].unique():\n",
        "        plt.plot(\n",
        "            sample_df[sample_df[\"m/z\"] == m][\"temp\"],\n",
        "            sample_df[sample_df[\"m/z\"] == m][\"abundance\"],\n",
        "        )\n",
        "    plt.title(sample_lab)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3CKqjQr6SdRt"
      },
      "source": [
        "# 4. Preprocessing"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "id": "vaVMzUl3SfcY"
      },
      "outputs": [],
      "source": [
        "# Selecting a testbed sample to demonstrate preprocessing steps\n",
        "sample_lab = sample_id_testbed[1]\n",
        "sample_df = sample_testbed_dict[sample_lab]"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cxvUeMBKSkjY"
      },
      "source": [
        "## 1. Standardizing which m/z values to include"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "id": "4syOHgMZSr6E"
      },
      "outputs": [],
      "source": [
        "def drop_frac_and_He(df):\n",
        "    \"\"\"\n",
        "    Drops fractional m/z values, m/z values > 100, and carrier gas m/z\n",
        "\n",
        "    Args:\n",
        "        df: a dataframe representing a single sample, containing m/z values\n",
        "\n",
        "    Returns:\n",
        "        The dataframe without fractional an carrier gas m/z\n",
        "    \"\"\"\n",
        "\n",
        "    # drop fractional m/z values\n",
        "    df = df[df[\"m/z\"].transform(round) == df[\"m/z\"]]\n",
        "    assert df[\"m/z\"].apply(float.is_integer).all(), \"not all m/z are integers\"\n",
        "\n",
        "    # drop m/z values greater than 99\n",
        "    df = df[df[\"m/z\"] < 100]\n",
        "\n",
        "    # drop carrier gas\n",
        "    df = df[df[\"m/z\"] != 4]\n",
        "\n",
        "    return df"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "id": "ow37k0F1StPG",
        "outputId": "13d0cb75-f5f2-4c7d-b38d-4b154a23a537"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "Text(0.5, 1.0, 'After dropping selected m/z values')"
            ]
          },
          "execution_count": 16,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAogAAAFrCAYAAABScXgLAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAABED0lEQVR4nO3debwcVZn/8c+XhB2EABEhCwkQ0YAjYAQcGEVQCIsTdBCCCgGj0Z8gqDgCbiCQGZhRUVxQlEiCQmBQJGJYIou4sQQNS4jINWwJCQlkYV8Cz++PcxqqO9339iW3b3ff+32/Xv26VadOVT1V3V336ao6pxQRmJmZmZmVrNXsAMzMzMystThBNDMzM7MyThDNzMzMrIwTRDMzMzMr4wTRzMzMzMo4QTQzMzOzMk4QzczMzKyME0Qz61WSfiTpaz20rOGSnpY0II/fJOkTPbHsvLyrJU3oqeV1Y71nSnpc0uLeXndP6On3wcx6nxNEM+sxkh6U9JykpyStkPRnSZ+W9OqxJiI+HRFn1Lms93VWJyIejoiNIuLlHoj9NEk/r1j+ARExdU2X3c04hgMnAqMj4k09uNxrJe3XU8szs77NCaKZ9bQPRMTGwDbAWcBJwAU9vRJJA3t6mS1iOPBERCzpqQVK2hAYA/y+p5ZpZn2bE0Qza4iIWBkRM4DDgQmSdgKQdKGkM/PwFpKuymcbl0n6g6S1JF1ESpR+ky8hf0nSCEkhaaKkh4EbCmXFZHE7SbdJelLSlZI2y+vaW9KCYoyls5SSxgJfBg7P67szT3/1UmmO66uSHpK0RNI0SZvkaaU4Jkh6OF8e/kqtfSNpkzz/0ry8r+blvw+YBWyd47iwyrx7S1qQ98kSSYskHSLpQEn/yPvxyxWz7Qv8KSJeyPv66fx6Jsc9omId6+Z6OxXKBuezw2+UNCi/b0slLc/DQ2tsa9mZ2cr3LO+LC/J2LMyX10u3DGwv6feSVuZ9emmtfWpmPcsJopk1VETcBiwA/q3K5BPztMHAlqQkLSLiSOBh0tnIjSLifwrzvAd4K7B/jVUeBXwc2ApYBZxbR4zXAP8FXJrX9/Yq1Y7Or/cC2wIbAd+vqLMXsAMpIfu6pLfWWOX3gE3yct6TYz4mIn4HHAA8muM4usb8bwLWA4YAXwd+AnwMeAdpP39N0shC/QOB3+Zt3TQveyPgu8AfgIUV++MF4FfAEYXiw4Df5zObawE/I50lHg48V2Vf1OtC0vu0PbALsB9Qun/xDOA6YBAwlLTfzKwXOEE0s97wKLBZlfKXSIncNhHxUkT8ISKii2WdFhHPRMRzNaZfFBH3RMQzwNeAw0pnpNbQR4FvR8T8iHgaOAUYX3H28hsR8VxE3AncCayWaOZYxgOnRMRTEfEg8C3gyG7E8hIwOSJeAqYDWwDfzcubC9xbse4DgZkVcRwOfAT4j7ycShfnOEs+ksuIiCci4pcR8WxEPAVMJiW63SJpyxzb5/J7ugQ4p7Del0hJ6NYR8XxE/LG76zCz18cJopn1hiHAsirl/wt0ANdJmi/p5DqW9Ug3pj8ErE1KoNbU1nl5xWUPJJ35LCm2On6WdJax0hY5psplDelGLE8UGuaUEuXHCtOfK61b0tuAlRHx6n6RtAvpjN8HI2JpjXXcCGwgafd8CXpn4Io8/waSfpwvjz8J3Axs+joS8W1I+2JRvqS9Avgx8MY8/UuAgNskzZX08W4u38xep756k7eZtQhJ7yQlP6ud/clnn04ETsz3u90g6faIuB6odSaxqzOMwwrDw0lnoR4HngE2KMQ1gHRpu97lPkpKaIrLXkVKzKref1fD47x2ZuzewrIW1pxjzZSdPZT0RuDXwLER8bdaM0XEy5IuI11mfgy4Kr9fkN6zHYDdI2KxpJ2Bv5GSuUpl+510ebzkEeAFYIuIWFUlhsXAJ3PcewG/k3RzRHR0usVmtsZ8BtHMGkLSGyQdTLoE+vOIuLtKnYNzQwQBK4GXgVfy5MdI9+h118ckjZa0AXA6cHk+2/YPYD1JB0laG/gqsG5hvseAESp0yVPhEuDzkkZK2ojX7llcLbHpTI7lMmCypI0lbQN8Afh553O+bq/ef5gvh19Oej8uq2Pei0mNjD6ah0s2Jp2lXKHUCOjUTpYxB3i3Up+Vm5AuzQMQEYtI9xh+K39e1pK0naT35Hg/XGj8spyUxL+CmTWcE0Qz62m/kfQU6ezQV4BvA8fUqDsK+B3wNPAX4IcRcWOe9t/AV/Olxy92Y/0XkRo+LCY15DgeUqtq4DPAT0ln654hNZAp+b/89wlJf62y3Cl52TcDDwDPA5/tRlxFn83rn086s3pxXn6PkrQpMBr4cy4aSmrE8rlCS+anlfpeXE1E3Jrj3Bq4ujDpO8D6pLOhtwDX1IohImYBlwJ3AXcAV1VUOQpYh3Q2dTkpgd0qT3sncKukp4EZwAkRMb/LDTezNaau7wc3M7N2JOkw4NCIOKzZsZhZe/EZRDOzvmsFqVWwmVm3+AyimZmZmZXxGUQzMzMzK+ME0czMzMzKOEE0MzMzszJOEM3MzMysjBNEMzMzMyvjBNHMzMzMyjhB7GMk/T9Jj+WnI2zey+s+TVKjHhfW1brnStq7GeuuRtKFks5sgThC0vbNjuP1kLS3pAVd17R2J2lPSffn49YhvbTOmyR9ojfWVbHe4Xk7B/T2umuR9KCk9zU5hrb+vjfz/1+jOEFsMfmL+lw+gCyX9FtJw+qcd23SY832i4iNIuKJxkbbOiJix4i4qdlx9CRJR0v6Y7PjKJL0Lkl/7rqm2epyUrZc0roVk04Hvp+PW79u5x82XYmIh/N2vtzsWHpSq/woLpJ0iqT/anYc7coJYmv6QERsRHoe6WPA9+qcb0vSs2fndneFShr2eZA0sFHLtl51EDCz2UFY+5E0gvQc6AD+vWLyNryO41aN9azxscbHqz7Dx6s14ASxhUXE86QH148ulUlaV9I3JT2cLyX/SNL6kt4M3JerrZB0Q67/r5Jul7Qy//3XwrJukjRZ0p+AZ4FtJb1F0ixJyyTdl5/lWpWkkZJ+L+kpSbOALQrTRuSzABMlPQzcIGktSV+V9JCkJZKmSdqkov4kSY9KWiTpi4XlnSbpckmX5vX9VdLbC9NfvUSS616Wl/9Uvvw8plB3V0l/y9P+Ly+z6i9fSdvnbVwp6XFJlxamdWdfHSxpjqQVkv4s6V8K04ZJ+pWkpZKekPR9SW8FfgS8K59NXtHZ+19Y1n/mffeopI/XiifXvUnSmTmepyX9RtLmkn4h6cn8eRlRMduBwExJX8rzlF4vSbqwyjpOknR5Rdl3JZ2bh4+RNC+/F/MlfaqTeMvOKqnijEUX+/gkSQvzeu6TtG9n+8Ya4ijgFuBCYEKpUNI/gW2B3+TP0l/ypDvz+OG5Xmfv74P5Pb4LeEZVEjxJ75f09/xd/j6gwrSjJf1J0jmSngBOk7RJPoYsVTpmfVX5R3Sh/vfz8v5e/Ezl79Z/S7otf5eulLRZnlY61g0s1D0jL+8pSddJKh5Lj8rrf0LS19TJ5WBJB0q6Ny9nocqPoTX3X8Uy1pJ0sqR/5nVeVoo9T98rz79C0iN5X0wCPgqUjgu/yXW3lvTLvA8fkHR8YTnr5+/wckn3Au+sFk+hfkj6jNKtCE/lfbZdjuXJHOc6hfqDgDcDf1E6thWPV69IOrrKOq6WdFxF2Z2SPpSHv5u3+UlJd0j6txqxrna5XOX/o2ruY0nrSfp5Ll+hdBzesrN90zAR4VcLvYAHgffl4Q2AqcC0wvRzgBnAZsDGwG+A/87TRpB+nQ/M45sBy4EjgYHAEXl88zz9JuBhYMc8fRPgEeCYPL4L8DgwukasfyFd0l4XeDfwFPDzilimARsC6wMfBzpI/ww2An4FXFRR/5Jc/23A0sK+OA14CTgUWBv4IvAAsHaV/XYa8DwpmRkA/DdwS562DvAQcEJezoeAF4Eza2zjJcBXSD+m1gP2yuUbdravSP8Ez8zDuwBLgN1zPBNyvOvm8Tvz+7phxTqOBv5YEU9n7/9Y0hnnnfKyLs77dPsa23ZTfj+2y+/9vcA/gPflbZoG/KxQfytgIfkRnYXyYcCjwAFV1rEN6cfHxnl8ALAI2COPH5TXL+A9ue6uedrewILCssq2pRv7eIf8Xm1d+Kxt1+zven975c/aZ4B3kL7LWxamPUj+/tZ4r2u+v4X55+TP4vpV1r0F6fhUOn58HlgFfCJPPzqPfzZ/9tfPn/8r8/dsRP5uTKyo//m8vMOBlcBmhe/WQl77Lv6S1Y+NAwt1/0lKZtbP42flaaOBp4G9SMeub+Z9974a+3gR8G95eFDhu1TP/isdP08gJfJD8/fnx8Alhe/zU6T/JWsDmwM7V34f8/hawB3A13Ps2wLzgf3z9LOAP5COZcOAeyh836tsW+T34w2k/1kvANfn5ZaOXxMK9ceX4q5YzgGk49WwKtOOAv5UGB9Nep55aT99LG/zQOBEYDGwXp52WuE93rtyW7qxjz9FOq5vkN+rdwBvaMp3ttkHDb8q3pD0IXo6fyhfyh/kt+VpAp6h8M8NeBfwQB4eQfmB50jgtorl/wU4Og/fBJxemHY48IeK+j8GTq0S53DSAXLDQtnFrH4Q3LYw/XrgM4XxHfI2DizUf0th+v8AF+Th08hJXh5fi/KDYfHLdxrwu0Ld0cBzefjdVCQ5wB+pnSBOA84HhlaUd7qvKE9ezgPOqKh7HykhehcpER5YZd1HU0gQ63j/p5D/seTxN9N1gviVwvi3gKsL4x8A5hTGJ5bej0LZ+qR/Aid18pn+I3BUHn4/8M9O6v4aOCEP7039CWJn+3h70j/H95F/UPjVuy9SgvMSsEUe/zvw+cL0V7+/Nd7rmu9vYf6Pd7L+oyg/fghYQHmC+HBh+gDSD8fRhbJPATcV6j9K+XHkNuDIPHxTxXdxdF7eAKoniF8t1P0McE0e/jqFJIeUNLxI7QTx4RznGyrK69l/pePnPGDfQr2teO04fQpwRY11v/p9zOO7F/dpLjuF/KOTlCyOLUybRNcJ4p6F8bLjDun49Z3C+EWl96NQ9mbSsWCvGuvYmHSM3SaPTwamdBLTcuDtefg06k8QO9vHHwf+DPxLI7+T9bx8ibk1HRIRm5LOJh0H/F7Sm4DBpAPEHfnU8wrgmlxezdaks2VFDwFDCuOPFIa3AXYvLTsv/6PAm2ose3lEPFOx7ErF5VfG8xDpC7FljfoP5XlWmxYRr5AO8MXpRYsLw88C6+VLOlsDCyN/K6uss9KXSP9MblO6VF26bNudfbUNcGJF3WE5lmHAQxGxqpMYSrp6/7dm9f3XlccKw89VGd+oMH4gq9/PcwFwX0Sc3ck6LiadcQD4SB4HQNIBkm5Ruky/Iq9ji9UX0aWa+zgiOoDPkQ7gSyRNl1Trc2ONMQG4LiIez+MXU7jMXIfOvkMlnX2Py74b+ftfWb84vgXpDFnl8ap47Kw8jtQ8XuVpa1P7s115vCp97yrjfhborPHhf5C+Qw8p3Rrzrlxez/6jUPeKQr15wMuk4/Qw0tnOemwDbF2xzi/z2vG+YccrpVsB3k86PpLLNiGdgfxqRFRt/BcRTwG/JZ19hHTc+kVhGV9UuiVmZd6eTXj9x6ta+/gi4FpgutKtQv+j1AC11zlBbGER8XJE/Ir0wdmLdAnzOWDHiNg0vzaJ1KClmkdJH8Si4aQzaK+upjD8CPD7wrI3jdTa7v9VWfYiYJCkDSuWvdpmdBJP6Sxk8Us+rGL6o9Wm5QPA0Irp9VgEDJGkQlnNVuIRsTgiPhkRW5N+mf9Q6T647uyrR4DJFXU3iIhL8rThqn5TfFSMd/X+L2L1/dcj8gHqPcCsQtnJpF/kE7uY/f+AvSUNBT5IThCVWrL+knTZbMv8o2gmhXvDKjxLSpBLisl4Z/uYiLg4IvYiff4C6CyhtR6kdI/sYcB7JC2WtJh0afbtKtxH3IVO39+s8vtSVPbdyN//yu99cf7HSWd0Ko9XxWNn5XGk5vEqT3spL7c7FpGOc6W41ydd4qwqIm6PiHHAG0ln4y/Lk+rZfxTqHlBRd72IWJinbVdr9VWW80DFcjaOiAML29aQ4xXpfsaHImIpvPr/4mLgxog4v4t5LwGOyMn1esCNeRn/RjphcBgwKB+vVlL9ePUMhWOVUpdGxRM5NfdxRLwUEd+IiNHAvwIHk86A9zoniC1MyTjSvSTz8lmznwDnSHpjrjNE0v41FjETeLOkj0gaqHSz92jgqhr1r8r1j5S0dn69U6nBRJmIeAiYDXxD0jqS9iJdkuzMJcDnlRq3bAT8F3Bpxdmzr0naQNKOpPv7Li1Me4ekD+Vk6nOke1Bu6WKdlf5CSriPy/tkHLBbrcqSPpwTG0iXEwJ4hW7sK9J79mlJu+f3dENJB0namHRZahFwVi5fT9Keeb7HgKHKN17X8f5fBhwtabSkDYBTu7lvOrMXcFdEPJnXewBwPPDBiHiusxnzQfom4Gekfxjz8qR1SPffLAVW5WXu18mi5gAfkTRA0lhSwlpScx9L2kHSPjkhfZ6UZL/Sra23NXEI6Ts3Gtg5v95Kuv+s1j++x0j3lpV09h2qx2+BHQvHj+OpfrYfSD/OSd+nyfkztA3wBaDYz90bgePzd//DeZuKZ9g/Vvgung5cHt3v2uZy4ANKjQ3XIZ0Fr/oDKh+HPyppk4h4CXiS1z7n3dl/P8rbvU1e7uB8nIR0Nu19kg7Lx8/NJe2cp1W+Z7cBTyk1Hlo/f293klRqjHIZcIqkQfkY+9lu7pvOHEh6z0smk+4FPaGOeWeSfhicTvr/VNqHG5NOaCwFBkr6Oul+yGr+QbpqdZDSj+uvko51JTX3saT3SnpbTiqfJP2waMrxyglia/qNpKdJH47JpBtvS11AnES62fsWSU8CvyPdy7eaSP0gHky6mfYJ0q+fgwuXeSrrP0X6Bz2e9Et4MelMS2WfZSUfId1nsoyUjEzrYrumkE6f30xqYPI8qx8Ufp+373rgmxFxXWHalaR7/5aT7q/8UD4Q1i0iXiQ1TJlIus/zY6Rk74Uas7wTuDW/HzNI98fN786+iojZwCeB7+fYO0j3MJX+EX2AdJ/cw6TL5ofnWW8gdf2xWFLpPav5/kfE1cB38nwd+W9Pqewu4nDSL+J5eq1l4I86mf9i0j2Ar15ezvvweNI/iuWkz9OMTpZxAmlfrSBdzv91YVk19zHpPTmLdPZmMekf+ymdrMd61gTSfWcP5zPyiyNiMem9+qiqnz0/DZiqdAnusC7e3y7lY96HSZ+DJ4BRwJ+6mO2zpDNB80n30V5MOoaV3JqX8zjpOH1olPc9exHpvrzFpDNRx9NN+bj/WWA66Yfk06R76Godr44EHszHhk+TviddfT8qfZf0PbxO0lOkH+G75+U8TEq+TiQd9+cApbPAFwCj83v263xsO5j0g+AB0n76KemyLMA3SJeVHwCuI+2vnlJ5vDoC2ANYXjhefbTajBHxAqkBZdnxinTZ9xpS8vcQ6f9X1dsaImIl6V7Sn5LOOj9DOraX1NzHpB8ul5P+/88j/U/syX1TN5XfQmHWHErdqTxAakSw2v14kk4j3bT+sQas+1bgRxHxs55edl+h1A3FoRFxb7NjMWs2pS5SPpFvW6g2/SZSg4Wf9vB6NyL9QBoVEQ/05LL7CqUuYf4GDAknOGvEZxCt35H0HklvypdIJgD/QuFmZiuXL21Nc3Jo1vskfUDptpsNSffr3k1qEWvVbQKc6ORwzbm3eOuPdiBd1tyQdPno0IhY1NyQWle+LH9Ws+Mw66fGkS4xinTf93gnP7VFxD9Il4FtDfkSs5mZmZmV8SVmMzMzMyvjBNHMzMzMyrT1PYi5n6DZpB7tD5Y0ktQdwOakx/AcGREv5v7PppGeafgEcHhEPJiXcQqpy5OXgeMj4tpcPpbUFH0A8NOI6PIerC222CJGjBjRsxtpZi3rjjvueDwiaj3JqK34+GXW/3R2DGvrBJHUL9o8Xuus8mzgnIiYnvtkm0h6BuVE0mPhtpc0Ptc7XNJoUj92O5Ie+/M7SW/Oy/oB6VE9C4DbJc3oqhXniBEjmD17ds9uoZm1LEn1PB6sLfj4Zdb/dHYMa9tLzLnn9YNIHVGWHp20D6mDSYCppB78IbUCm5qHLwf2zfXHAdMj4oXcp1QH6akauwEduUPkF0lnJUs9yZuZmZn1aW2bIJKeGPElXnsEzebAikInywt47cHqQ8g9nufpK3P9V8sr5qlVbmZmZtbntWWCKOlgYElE3NECsUySNFvS7KVLlzY7HDMzM7M11pYJIrAn8O+SHiRd/t2H1KBk08JzPYeSnoFI/jsMIE/fhNRY5dXyinlqla8mIs6PiDERMWbw4D5xr7qZmZn1c22ZIEbEKRExNCJGkBqZ3BARHwVuBA7N1SYAV+bhGXmcPP2G3BP9DGC8pHVzC+hRwG3A7cAoSSPzY8bG57pmZmtM0hRJSyTdU2XaiZJC0hZ5XJLOldQh6S5JuxbqTpB0f35NKJS/Q9LdeZ5z8z3XZmZ1a8sEsRMnAV+Q1EG6x/CCXH4BsHku/wJwMkBEzCU9cu1e0rN4j42Il/N9iscB15JaSV+W65qZ9YQLgbGVhZKGAfsBDxeKDyD9eB0FTCL1zICkzYBTgd1JDetOlTQoz3Me8MnCfKuty8ysM+3ezQ0RcRNwUx6eTzpQVtZ5HvhwjfknA5OrlM8EZvZgqGZmAETEzZJGVJl0Dqnx3ZWFsnHAtHzV4xZJm0raCtgbmBURywAkzQLGSroJeENE3JLLp5F6dLi6MVtjZn1RXzuDaGbWliSNI3X6f2fFpO72tjAkD1eWV1unG9mZWVVOEM3MmkzSBsCXga/35nrdyM7ManGCaGbWfNsBI4E7c+8MQ4G/SnoT3e9tYWEeriw3M6ubE0QzsyaLiLsj4o0RMSL3zrAA2DUiFpN6UDgqt2beA1gZEYtIjej2kzQoN07ZD7g2T3tS0h659fJRlN/TaGbWpbZvpGK9a8TJv212CC3pwbMOanYI1kYkXUJqZLKFpAXAqRFxQY3qM4EDSY8CfRY4BiAilkk6g9QtF8DppQYrwGdILaXXJzVOcQOVrJHHsHY9DrTrPvH/o9X15P52gmhm1ssi4ogupo8oDAdwbI16U4ApVcpnAzutWZRm1p/5ErOZmZmZlXGCaGZmZmZlnCCamZmZWRkniGZmZmZWxgmimZmZmZVxgmhmZmZmZZwgmpmZmVkZ94PYB7nzUDMzM1sTPoNoZmZmZmWcIJqZmZlZGSeIZmZmZlbGCaKZmZmZlXGCaGZmZmZlnCCamZmZWRkniGZmZmZWxgmimZmZmZVxgmhmZmZmZZwgmpmZmVkZJ4hmZmZmVqYtE0RJ60m6TdKdkuZK+kYuv1DSA5Lm5NfOuVySzpXUIekuSbsWljVB0v35NaFQ/g5Jd+d5zpWkXt9QMzMzsyYY2OwAXqcXgH0i4mlJawN/lHR1nvafEXF5Rf0DgFH5tTtwHrC7pM2AU4ExQAB3SJoREctznU8CtwIzgbHA1ZiZmZn1cW15BjGSp/Po2vkVncwyDpiW57sF2FTSVsD+wKyIWJaTwlnA2DztDRFxS0QEMA04pFHbY2ZmZtZK2jJBBJA0QNIcYAkpybs1T5qcLyOfI2ndXDYEeKQw+4Jc1ln5girl1eKYJGm2pNlLly5d080yMzMza7p2vcRMRLwM7CxpU+AKSTsBpwCLgXWA84GTgNMbHMf5eV2MGTOms7OYZmZm/cqIk3/b7BDsdWrbBLEkIlZIuhEYGxHfzMUvSPoZ8MU8vhAYVphtaC5bCOxdUX5TLh9apb6ZmTVYuyYV7Rq3WTVteYlZ0uB85hBJ6wPvB/6e7x0ktzg+BLgnzzIDOCq3Zt4DWBkRi4Brgf0kDZI0CNgPuDZPe1LSHnlZRwFX9t4WmpmZmTVPu55B3AqYKmkAKcm9LCKuknSDpMGAgDnAp3P9mcCBQAfwLHAMQEQsk3QGcHuud3pELMvDnwEuBNYntV52C2YzMzPrF9oyQYyIu4BdqpTvU6N+AMfWmDYFmFKlfDaw05pFamZmZtZ+2vISs5lZO5M0RdISSfcUyv5X0t9zLwxXlG6jydNOyZ323ydp/0L52FzWIenkQvlISbfm8kslrdNrG2dmfYITRDOz3nchqfP9olnAThHxL8A/SL0yIGk0MB7YMc/zw9zN1wDgB6QHAYwGjsh1Ac4GzomI7YHlwMTGbo6Z9TVOEM3MellE3Awsqyi7LiJW5dFbeK0nhXHA9Ih4ISIeIN1LvVt+dUTE/Ih4EZgOjMsN6/YBSk+Umoo7+jezbnKCaGbWej7Oaw3jutvR/+bAikKy6Y7+zazbnCCambUQSV8BVgG/aPS6IuL8iBgTEWMGDx7c6NWZWRtpy1bMZmZ9kaSjgYOBfXPvC1C7o39qlD9Bet78wHwW0R39m1m3+QyimVkLkDQW+BLw7xHxbGHSDGC8pHUljQRGAbeR+m8dlVssr0NqyDIjJ5Y3Aofm+Sfgjv7NrJucIJqZ9TJJlwB/AXaQtEDSROD7wMbALElzJP0IICLmApcB9wLXAMdGxMv57OBxpCdCzSM9MGBuXsVJwBckdZDuSbygFzfPzPoAX2I2M+tlEXFEleKaSVxETAYmVymfSXpSVGX5fFIrZzOz18VnEM3MzMysjBNEMzMzMyvjBNHMzMzMyjhBNDMzM7MyThDNzMzMrIwTRDMzMzMr4wTRzMzMzMo4QTQzMzOzMk4QzczMzKyME0QzMzMzK+ME0czMzMzKOEE0MzMzszJOEM3MzMysjBNEMzMzMyvjBNHMzMzMyrRlgihpPUm3SbpT0lxJ38jlIyXdKqlD0qWS1snl6+bxjjx9RGFZp+Ty+yTtXygfm8s6JJ3c6xtpZmZm1iRtmSACLwD7RMTbgZ2BsZL2AM4GzomI7YHlwMRcfyKwPJefk+shaTQwHtgRGAv8UNIASQOAHwAHAKOBI3JdMzMzsz6vLRPESJ7Oo2vnVwD7AJfn8qnAIXl4XB4nT99XknL59Ih4ISIeADqA3fKrIyLmR8SLwPRc18zMzKzPa8sEESCf6ZsDLAFmAf8EVkTEqlxlATAkDw8BHgHI01cCmxfLK+apVV4tjkmSZkuavXTp0h7YMjMzM7PmatsEMSJejoidgaGkM35vaVIc50fEmIgYM3jw4GaEYGZmZtaj6koQJR0h6a15eAdJN0u6UVJTkrKiiFgB3Ai8C9hU0sA8aSiwMA8vBIYB5OmbAE8UyyvmqVVuZmZm1ufVewbxTGBZHv4mcBvwe+CHjQiqK5IGS9o0D68PvB+YR0oUD83VJgBX5uEZeZw8/YaIiFw+PrdyHgmMIm3b7cCo3Cp6HVJDlhkN3zAzMzOzFjCw6yoADI6IxyStB+xFSrJeAh5vWGSd2wqYmlsbrwVcFhFXSboXmC7pTOBvwAW5/gXARZI6SInueICImCvpMuBeYBVwbES8DCDpOOBaYAAwJSLm9t7mmZmZmTVPvQniUknbA28Dbo+IFyRtAKhxodUWEXcBu1Qpn0+6H7Gy/HngwzWWNRmYXKV8JjBzjYM1MzMzazP1JohnAHcALwOH57L3AXc2IigzMzMza566EsSIuDBfiiUins3Ft5Av1ZqZmZlZ31HvGUQi4llJb5T0pkYGZGZmZmbNVVeCKGksqaHHVhWTgtSIw8zMzMz6iHq7ufkB6T7EDSNircLLyaGZmZlZH1NvgjgI+HFEPNfIYMzM+gNJUyQtkXRPoWwzSbMk3Z//DsrlknSupA5Jd0natTDPhFz/fkkTCuXvkHR3nufc/Ox5M7O61ZsgXgAc08hAzMz6kQuBsRVlJwPXR8Qo4Po8DnAAqRP/UcAk4DxICSVwKrA7qXuvU0tJZa7zycJ8lesyM+tUvQniHsB5kv6RH7P36quRwZmZ9UURcTOvPZ2qZBwwNQ9PBQ4plE+L5BbSI0W3AvYHZkXEsohYDswCxuZpb4iIW/ITo6YVlmVmVpd6WzH/NL/MzKwxtoyIRXl4MbBlHh4CPFKotyCXdVa+oEr5aiRNIp2VZPjw4WsYvpn1JfX2gzi161pmZtYTIiIkRS+s53zgfIAxY8Y0fH1m1j7qvcSMpGMk3SDpvvzX9ySamfWcx/LlYfLfJbl8ITCsUG9oLuusfGiVcjOzutXbD+JXgKOAbwEPAdsAX5K0dX6WsXXTiJN/2+wQzKy1zAAmAGflv1cWyo+TNJ3UIGVlRCySdC3wX4WGKfsBp0TEMklPStoDuJV07P5eb26ImbW/eu9B/ASwd0Q8VCrIB6ebASeIZmbdIOkSYG9gC0kLSK2RzwIukzSR9EP8sFx9JnAg0AE8S+5RIieCZwC353qnR0Sp4ctnSC2l1weuzi8zs7rVmyBuCCytKHuCdPAxM7NuiIgjakzat0rdAI6tsZwpwJQq5bOBndYkRjPr3+q9B/Ea4BeSdpC0vqS3kLphuLZxoZmZmZlZM9SbIB4HPAXcBTwNzAGeAT7bmLDMzMzMrFnq7ebmSeAoSUcDWwCPR8QrjQzMzMzMzJqjZoIoaUREPJiHt62YvFHp0Z4RMb9h0ZmZmZlZr+vsDOLdwMZ5uAMIoPKB7wEMaEBcZmZmZtYkNRPEiNi4MFx3h9pmZmZm1t7qSvwknVuj/Ds9Go2ZmZmZNV29ZwaPrlF+ZA/FYWZmZmYtotNWzJI+XqpXGC7ZFni8IVGZmZmZWdN01c1N6QzhOpSfLQzgMdLzQs3MzMysD+k0QYyI9wJIOjMivto7IZmZmZlZM9V1D2IxOVSyVunVuNBqkzRM0o2S7pU0V9IJufw0SQslzcmvAwvznCKpQ9J9kvYvlI/NZR2STi6Uj5R0ay6/VNI6vbuVZmZmZs1RbyvmrSVdIekJYBXwUuHVDKuAEyNiNLAHcKyk0XnaORGxc37NBMjTxgM7AmOBH0oaIGkA8APgAGA0cERhOWfnZW0PLAcm9tbGmZmZmTVTvWcAfwy8COxLehbzrsAM4NMNiqtTEbEoIv6ah58C5gFDOpllHDA9Il6IiAdIHX/vll8dETE/Il4EpgPjlB4Tsw9weZ5/KnBIQzbGzMzMrMXUmyD+K/DxiJgDRETcSTqjdmKjAquXpBHALsCtueg4SXdJmiJpUC4bAjxSmG1BLqtVvjmwIiJWVZRXW/8kSbMlzV66dGlPbJKZmZlZU9WbIL5MuqwLsELSYOAZOj9r13CSNgJ+CXwuIp4EzgO2A3YGFgHfanQMEXF+RIyJiDGDBw9u9OrMzMzMGq7eBPFWoNTg41rgUuBXwOxGBFUPSWuTksNfRMSvACLisYh4OSJeAX5CuoQMsBAYVph9aC6rVf4EsKmkgRXlZmZmZn1evQnikcDv8/DngBuAe4CPNCCmLuV7BC8A5kXEtwvlWxWqfZAUI6T7JcdLWlfSSGAUcBtwOzAqt1heh9SQZUZEBHAjcGiefwJwZSO3yczMzKxVdNVR9qv1IuJxgIh4DjizcSHVZU9S0nq3pDm57MukVsg7kzryfhD4FEBEzJV0GXAv6VL5sRHxMoCk40hnRQcAUyJibl7eScB0SWcCfyMlpGZmZmZ9Xr0J4sOSbgIuBq6IiGcaF1LXIuKPgKpMmtnJPJOByVXKZ1abLyLm89olajMzM7N+o95LzMOBq0jd2iyWdImkDxTu0TMzMzOzPqLeJ6k8HhE/jIi9gJ2AO0ln4xY1MjgzMzMz632v51F5bwS2BLYAVvRoNGZmZmbWdPU+am+0pDMkdQC/zsWHRMSohkVmZmZmZk1R7z2EfyL1Ofgp4Mbcz6CZmZmZ9UFdJoiSBpCeUHJ6RDzf+JDMzMzMrJm6vMSc+wv8JPBi48MxMzMzs2art5HKRaQubszMzMysj6s3QdwN+K6kByX9QdLNpVcjgzMz628kfV7SXEn35D5n18uPA71VUoekS/OjQcmPD700l98qaURhOafk8vsk7d+0DTKztlRvI5Wf5JeZmTWIpCHA8cDoiHguPyJ0PHAgcE5ETJf0I2Ai6d7wicDyiNhe0njgbOBwSaPzfDsCWwO/k/Tm0iNGzcy6UleCGBFTGx2ImZkB6bi8vqSXgA1IDyTYB/hInj4VOI2UII7LwwCXA9+XpFw+PSJeAB7IXZTtBvyll7bBzNpcvf0gStInJd0g6a5c9m5JhzU2PDOz/iMiFgLfBB4mJYYrgTuAFRGxKldbAAzJw0OAR/K8q3L9zYvlVeZ5laRJkmZLmr106dKe3yAza1v13oN4OulSxvmk5zJDOuCc1IigzMz6I0mDSGf/RpIuDW8IjG3U+iLi/IgYExFjBg8e3KjVmFkbqjdBPBo4OCKmA5HLHgC2bURQZmb91PuAByJiaUS8BPwK2BPYVFLplqChwMI8vBAYBpCnbwI8USyvMo+ZWZfqTRAHAE/n4VKCuFGhzMzM1tzDwB6SNsj3Eu4L3AvcCBya60wArszDM/I4efoNERG5fHxu5TwSGAXc1kvbYGZ9QL0J4kzg25LWhXRPInAG8JtGBWZm1t9ExK2kxiZ/Be4mHaPPJ93O84Xc2GRz4II8ywXA5rn8C8DJeTlzgctIyeU1wLFuwWxm3VFvNzdfILWcWwmsTTpzeB1wVIPiMjPrlyLiVODUiuL5pFbIlXWfBz5cYzmTgck9HqCZ9Qv1dnPzJPBBSW8EtgEeiYjFDY3MzMzMzJqirgRR0mDguYhYIukJ4ChJLwM/j4hXGhqhmZmZmfWqeu9BvIp0kzOkSxZfJF12/lYjgjIzMzOz5qn3HsQ3A3Py8MeAfyXdhzgX+HzPh2VmZmZmzVJvgvgysI6kNwMrI+JhSWuRuroxMzMzsz6k3gTxalKXCZsD03PZaNzxqpmZmVmfU2+C+AlSZ6wvARflsi147SHxZv3aiJN/27BlP3jWQQ1btpmZWTV1NVKJiBci4nzgQmCQJEXETfnRe71O0jBJN0q6V9JcSSfk8s0kzZJ0f/47KJdL0rmSOiTdJWnXwrIm5Pr3S5pQKH+HpLvzPOfmzsHNzMzM+ry6EkRJm0qaBjwHPAY8J+kiSZs1NLraVgEnRsRoYA/gWEmjSU8RuD4iRgHX53GAA0itsEcBk4DzICWUpA5pdyd1QntqKanMdT5ZmG9sL2yXmZmZWdPV283Nz4ANgF1IDVN2AdYFpjQork5FxKKI+GsefgqYBwwBxpGe+EL+e0geHgdMi+QW0oPvtwL2B2ZFxLKIWA7MAsbmaW+IiFvyc02nFZZlZmZm1qfVew/iPsCbIuK5PD5P0tHAow2JqhskjSAlrLcCW0bEojxpMbBlHh4CPFKYbUEu66x8QZXyauufRDoryfDhw9dgS8zMzMxaQ71nEP8OjKgoGw7c16PRdJOkjYBfAp/LjwN8VT7zF42OISLOj4gxETFm8ODBjV6dmZmZWcPVPIMo6eOF0euB6yRdRDrjNozUYfZF1ebtDZLWJiWHv4iIX+XixyRtFRGL8mXiJbl8ISnmkqG5bCGwd0X5Tbl8aJX6ZmZmZn1eZ2cQjyy83gV05L+H5b//zH97XW5RfAEwLyK+XZg0g9QdD/nvlYXyo3Jr5j1InX0vAq4F9pM0KDdO2Q+4Nk97UtIeeV1HFZZlZmZm1qfVPIMYEe/tzUC6aU9S4nq3pDm57MvAWcBlkiYCD5GSWYCZwIGkJPdZ4BiAiFgm6Qzg9lzv9IhYloc/Q+rWZ31SR+FXN3B7zMzMzFpGvY1UyGfYPkBqrLEQuKqQTPWqiPgjUKtfwn2r1A/g2BrLmkKV1tgRMRvYaQ3CNDMzM2tL9faDWLqk/GngX4BPAR253MzMzMz6kHrPIH4H+EzxySmSDgfOBd7ZgLjMzMzMrEnq7ebmzcBlFWWXA9v3bDhmZmZm1mz1Joj3A+Mryj5MuuxsZmZmZn1IvZeYPwdcJel4UuvgEaTnEx/cmLDMzMzMrFnqShAj4s+StgMOArYGfgPMbFYrZjMzMzNrnLq7uYmI5cDPGxiLmZmZmbWAeu9BNDMzM7N+wgmimZmZmZWpmSBKentvBmJmZmZmraGzM4h/KA1Iur8XYjEzMzOzFtBZI5UVkg4G7gW2kjSSKs8/joj5jQrOzMzMzHpfZwniCaRH7G1DOtNYrVPsAAb0fFhmZmZm1iw1LzFHxBURsX1ErA08GxFrVXk5OTQz60GSNpV0uaS/S5on6V2SNpM0S9L9+e+gXFeSzpXUIekuSbsWljMh179f0oTmbZGZtaN6WzFvDiBpLUlbSXLrZzOzxvgucE1EvAV4OzAPOBm4PiJGAdfncYADSE+1GgVMAs4DkLQZcCqwO7AbcGopqTQzq0e9id66kqYBzwMLgeckTZW0SeNCMzPrX/Ix9d3ABQAR8WJErADGAVNztanAIXl4HDAtkluATSVtBewPzIqIZfkhB7OAsb22IWbW9upNEL8HbAjsBKwPvA3YADi3QXGZmfVHI4GlwM8k/U3STyVtCGwZEYtyncXAlnl4CPBIYf4FuaxWeRlJkyTNljR76dKlPbwpZtbO6k0QxwJHRsQ/IuKFiPgHcAz+RWpm1pMGArsC50XELsAzvHY5GYCICFIDwTUWEedHxJiIGDN48OCeWKSZ9RH1JojPA5VHjy2AF3o2HDOzfm0BsCAibs3jl5MSxsfypWPy3yV5+kJgWGH+obmsVrmZWV3qTRB/CsyS9GlJB0j6NHAtcH7jQjMz618iYjHwiKQdctG+pL5oZwCllsgTgCvz8AzgqNyaeQ9gZb4UfS2wn6RBuXHKfrnMzKwunfWDWDQZeBT4CLB1Hv4fYEqD4jIz668+C/xC0jrAfNLtPGsBl0maCDwEHJbrzgQOBDqAZ3NdImKZpDOA23O90yNiWe9tgpm1u7oSxHzPyxScEJqZNVREzAHGVJm0b5W6ARxbYzk+ZpvZ6+b+DM3MzMysjBNEMzMzMyvjBNHMzMzMytSVIEr6cI3yQ3s2nPpImiJpiaR7CmWnSVooaU5+HViYdkp+Vul9kvYvlI/NZR2STi6Uj5R0ay6/NN8sbmZmZtYv1HsG8YIa5c3q5uZCqnfSfU5E7JxfMwEkjQbGAzvmeX4oaYCkAcAPSM8yHQ0ckesCnJ2XtT2wHJjY0K0xMzMzayGdtmKWtG0eXEvSSECFyduSOtDudRFxs6QRdVYfB0yPiBeAByR1kB5eD9AREfMBJE0HxkmaB+xD6tIH0nNPTwPO66HwzczMzFpaV93cdJAe6STgnxXTFpMSp1ZynKSjgNnAifkh9UOAWwp1is8krXxW6e7A5sCKiFhVpf5qJE0CJgEMHz68J7bBzMzMrKk6vcQcEWtFxADgD3m4+No6IlrpSSrnAdsBOwOLgG/1xkr9LFMzMzPra+rtKPs9jQ5kTUXEY6VhST8BrsqjnT2TtFr5E8Cmkgbms4h+hqmZmZn1K/W2Yh4p6WJJ90p6uPhqdID1Kj3IPvsgUGrhPAMYL2ndfB/lKOA20iOoRuVtW4fUkGVGfjLBjUCphXbxuadmZmZmfV69z2K+mHQP4omk5302laRLgL2BLSQtAE4F9pa0M+meyQeBTwFExFxJl5EeeL8KODYiXs7LOY70APsBwJSImJtXcRIwXdKZwN+o3YrbzMzMrM+pN0HcEdgzIl5pZDD1iogjqhTXTOIiYjIwuUr5TNLD7ivL5/NaS2czMzOzfqXefhBvBnZpZCBmZmZm1hrqPYP4IHCNpCtI3du8KiK+3tNBmZmZmVnz1JsgbkhqFbw25S1/zczMzKyPqbebm2MaHYiZmZmZtYa6EsTCI/dWU3pUnZmZmZn1DfVeYi4+cq8k8t8BPRqRmZmZmTVVvZeYy1o7S3oTqe/BPzQiKDMzMzNrnnq7uSkTEYuBzwH/3aPRmJmZmVnTva4EMdsB2KCnAjEzMzOz1lBvI5U/8No9h5ASwx2B0xsRlJmZmZk1T72NVH5aMf4McGdE3N/D8ZiZmZlZk9XbSGVqowMxMzMzs9ZQ1z2IktaW9A1J8yU9n/9+Q9I6jQ7QzMzMzHpXvZeY/wfYDfg08BCwDfA14A3A5xsTmpmZmZk1Q70J4oeBt0fEE3n8Pkl/Be7ECaKZmZlZn1JvNzfqZrmZmZmZtal6E8T/A34jaX9Jb5U0Fvg1cFnDIjMz66ckDZD0N0lX5fGRkm6V1CHp0tL935LWzeMdefqIwjJOyeX3Sdq/SZtiZm2q3gTxS8DvgB8AdwDfA24ETmpQXGZm/dkJwLzC+NnAORGxPbAcmJjLJwLLc/k5uR6SRgPjSf3VjgV+KGlAL8VuZn1AXQliRLwYEV+PiO0jYoOIGBURX4uIFxodoJlZfyJpKHAQuf9ZSQL2AS7PVaYCh+ThcXmcPH3fXH8cMD0iXoiIB4AOUkNDM7O6dJogStpT0tk1pp0laY/GhGVm1m99h3TV5pU8vjmwIiJW5fEFwJA8PAR4BCBPX5nrv1peZZ5XSZokabak2UuXLu3hzTCzdtbVGcQvAzfXmPZ74Cs9G46ZWf8l6WBgSUTc0Rvri4jzI2JMRIwZPHhwb6zSzNpEV93c7AxcU2PaLOCCHo3GzKx/2xP4d0kHAuuR+pr9LrCppIH5LOFQYGGuvxAYBiyQNBDYBHiiUF5SnMfMrEtdnUF8A1DraSlrAxv3bDhmZv1XRJwSEUMjYgSpkckNEfFRUqPAQ3O1CcCVeXhGHidPvyEiIpePz62cRwKjgNt6aTPMrA/oKkH8O7BfjWn75elmZtZYJwFfkNRBusewdPXmAmDzXP4F4GSAiJhL6obsXtJVoGMj4uVej9rM2lZXCeI5wI8lfUjSWgCS1pL0IeBHwLcbHWAtkqZIWiLpnkLZZpJmSbo//x2UyyXp3Nwn2F2Sdi3MMyHXv1/ShEL5OyTdnec5N7cMNDPrFRFxU0QcnIfnR8RuuSeJD5d6kIiI5/P49nn6/ML8kyNiu4jYISKubtZ2mFl76jRBjIiLSc9hngo8L+lR4Pk8/r8RcUnjQ6zpQlL/XkUnA9dHxCjg+jwOcADpEssoYBJwHqSEEjgV2J3UBcSppaQy1/lkYb7KdZmZmZn1SV32gxgR3yZ1j/AB4Iv575Bc3jQRcTOwrKK42CdYZV9h0yK5hXTD91bA/sCsiFgWEctJDW/G5mlviIhb8v080wrLMjMzM+vTumrFDEBEPAlc2+BYesKWEbEoDy8GtszDtfoE66x8QZXy1UiaRDoryfDhw9cwfDMzM7Pmq/dRe20nn/mLXliP+xEzMzOzPqWvJYiP5cvD5L9LcnmtPsE6Kx9apdzMzMysz+trCWKxT7DKvsKOyq2Z9wBW5kvR1wL7SRqUG6fsB1ybpz0paY/cevmowrLMzMzM+rS67kFsRZIuAfYGtpC0gNQa+SzgMkkTgYeAw3L1mcCBpAfWPwscAxARyySdAdye650eEaWGL58htZReH7g6v8zMzMz6vLZNECPiiBqT9q1SN4BjayxnCjClSvlsYKc1idHMzMysHfW1S8xmZmZmtoacIJqZmZlZGSeIZmZmZlbGCaKZmZmZlXGCaGZmZmZlnCCamZmZWRkniGZmZmZWxgmimZmZmZVxgmhmZmZmZZwgmpmZmVmZtn3UnpmZNc+Ik3/b7BDMrIF8BtHMzMzMyjhBNDMzM7MyThDNzMzMrIwTRDMzMzMr4wTRzMzMzMo4QTQzMzOzMk4QzczMzKyME0QzMzMzK+ME0czMzMzKOEE0MzMzszJOEM3MWoSkYZJulHSvpLmSTsjlm0maJen+/HdQLpekcyV1SLpL0q6FZU3I9e+XNKFZ22Rm7ckJoplZ61gFnBgRo4E9gGMljQZOBq6PiFHA9Xkc4ABgVH5NAs6DlFACpwK7A7sBp5aSSjOzejhBNDNrERGxKCL+moefAuYBQ4BxwNRcbSpwSB4eB0yL5BZgU0lbAfsDsyJiWUQsB2YBY3tvS8ys3TlBNDNrQZJGALsAtwJbRsSiPGkxsGUeHgI8UphtQS6rVV65jkmSZkuavXTp0p7dADNra30uQZT0oKS7Jc2RNDuX+f4dM2sbkjYCfgl8LiKeLE6LiACiJ9YTEedHxJiIGDN48OCeWKSZ9RF9LkHM3hsRO0fEmDzu+3fMrC1IWpuUHP4iIn6Vix/Ll47Jf5fk8oXAsMLsQ3NZrXIzs7r01QSxku/fMbOWJ0nABcC8iPh2YdIMoHQlYwJwZaH8qHw1ZA9gZb4UfS2wn6RB+cftfrnMzKwuA5sdQAMEcJ2kAH4cEefToPt3IN3DQzr7yPDhw3tqG8ysf9oTOBK4W9KcXPZl4CzgMkkTgYeAw/K0mcCBQAfwLHAMQEQsk3QGcHuud3pELOuVLTCzPqEvJoh7RcRCSW8EZkn6e3FiREROHntETkDPBxgzZkyPLdfM+p+I+COgGpP3rVI/gGNrLGsKMKXnojOz/qTPXWKOiIX57xLgCtI9hL5/x8zMzKxOfSpBlLShpI1Lw6T7bu7B9++YmZmZ1a2vXWLeErgi3efNQODiiLhG0u34/h0zMzOzuvSpBDEi5gNvr1L+BL5/x8zMzKwufSpBNOuLRpz824Yt+8GzDmrYss3MrH31qXsQzczMzGzNOUE0MzMzszJOEM3MzMysjBNEMzMzMyvjBNHMzMzMyjhBNDMzM7MyThDNzMzMrIwTRDMzMzMr446yzfoxd8JtZmbV+AyimZmZmZVxgmhmZmZmZZwgmpmZmVkZJ4hmZmZmVsYJopmZmZmVcYJoZmZmZmWcIJqZmZlZGSeIZmZmZlbGHWWbWUM0shPuRnIH32ZmPoNoZmZmZhWcIJqZmZlZGSeIZmZmZlbGCaKZmZmZlXGCaGZmZmZlnCB2QtJYSfdJ6pB0crPjMTPrDh/DzOz1coJYg6QBwA+AA4DRwBGSRjc3KjOz+vgYZmZrwglibbsBHRExPyJeBKYD45ock5lZvXwMM7PXzR1l1zYEeKQwvgDYvbKSpEnApDz6tKT76lj2FsDjaxxhYznGntMOcTrGTGd3q/o2DQqjJ3R5DOvG8asdPh+1tHPs4Pibqe1irzh+1RN/zWOYE8Q1FBHnA+d3Zx5JsyNiTINC6hGOsee0Q5yOsX+q9/jVzvu+nWMHx99M7Rw7rHn8vsRc20JgWGF8aC4zM2sHPoaZ2evmBLG224FRkkZKWgcYD8xockxmZvXyMczMXjdfYq4hIlZJOg64FhgATImIuT20+G5dkm4Sx9hz2iFOx9jH9PAxrJ33fTvHDo6/mdo5dljD+BURPRWImZmZmfUBvsRsZmZmZmWcIJqZmZlZGSeIvaiVH3sl6UFJd0uaI2l2LttM0ixJ9+e/g3o5pimSlki6p1BWNSYl5+Z9e5ekXZsY42mSFuZ9OUfSgYVpp+QY75O0fy/FOEzSjZLulTRX0gm5vNX2Za04W2p/9ietfMyqpruf9VYkaYCkv0m6Ko+PlHRrfg8uzQ2OWpKkTSVdLunvkuZJele77HtJn8+fmXskXSJpvVbe973x/9EJYi9Rezz26r0RsXOh36STgesjYhRwfR7vTRcCYyvKasV0ADAqvyYB5zUxRoBz8r7cOSJmAuT3ezywY57nh/lz0WirgBMjYjSwB3BsjqXV9mWtOKG19me/0CbHrErd/ay3ohOAeYXxs0mf/+2B5cDEpkRVn+8C10TEW4C3k7aj5fe9pCHA8cCYiNiJ1KhrPK297y+kwf8fnSD2nnZ87NU4YGoengoc0psrj4ibgWV1xjQOmBbJLcCmkrZqUoy1jAOmR8QLEfEA0EH6XDRURCyKiL/m4adIB+0htN6+rBVnLU3Zn/1I2x2zXsdnvaVIGgocBPw0jwvYB7g8V2nl2DcB3g1cABARL0bECtpk35N6dVlf0kBgA2ARLbzve+P/oxPE3lPtsVed/fPrbQFcJ+kOpcdvAWwZEYvy8GJgy+aEVqZWTK22f4/Lp/KnFC6pND1GSSOAXYBbaeF9WREntOj+7OPaev/W+VlvNd8BvgS8ksc3B1ZExKo83srvwUhgKfCzfIn8p5I2pA32fUQsBL4JPExKDFcCd9A++76kR4/pThCtZK+I2JV0KvpYSe8uTozUH1JL9YnUijFl5wHbATuTDjbfamo0maSNgF8Cn4uIJ4vTWmlfVomzJfenta52+awXSToYWBIRdzQ7ltdpILArcF5E7AI8Q8Xl5Bbe94NIZ9lGAlsDG1L91qG20RP72gli72npx17lX1BExBLgCtLlpcdKp6Hz3yXNi/BVtWJqmf0bEY9FxMsR8QrwE1677Nm0GCWtTfqH+YuI+FUubrl9WS3OVtyf/URb7t9uftZbyZ7Av0t6kHQ5fx/SPX2b5sue0NrvwQJgQUSUzvpfTkoY22Hfvw94ICKWRsRLwK9I70e77PuSHj2mO0HsPS372CtJG0rauDQM7AfcQ4pvQq42AbiyORGWqRXTDOCo3FprD2Bl4VR7r6q4t+ODpH0JKcbxktaVNJJ0w/BtvRCPSPcFzYuIbxcmtdS+rBVnq+3PfqRlj1m1vI7PesuIiFMiYmhEjCDt6xsi4qPAjcChuVpLxg4QEYuBRyTtkIv2Be6lDfY96dLyHpI2yJ+hUuxtse8LevaYHhF+9dILOBD4B/BP4CvNjqcQ17bAnfk1txQb6f6X64H7gd8Bm/VyXJeQLim+RPp1OrFWTIBILS7/CdxNao3WrBgvyjHclb+YWxXqfyXHeB9wQC/FuBfpUsNdwJz8OrAF92WtOFtqf/anV6ses17HZ6ipx7LXsR17A1fl4W1JP3w6gP8D1m12fJ3EvTMwO+//XwOD2mXfA98A/k76AXoRsG4r7/ve+P/oR+2ZmZmZWRlfYjYzMzOzMk4QzczMzKyME0QzMzMzK+ME0czMzMzKOEE0MzMzszJOEM3MzMysjBNEMzMzMyvjBNHMzMzMyjhBNDMzM7MyThDNzMzMrIwTRDMzMzMr4wTRzMzMzMo4QTQzMzOzMk4QzczMzKyME0QzsxYg6QhJFzc7DjMzcIJoZtYqDgJmNjsIMzNwgmhm1hCSHpT0n5LukvSMpAskbSnpaklPSfqdpEG57lrA+4FrJH1f0tOF1ypJpzV1Y8ys33GCaGbWOP9BSvzeDHwAuBr4MjCYdPw9PtfbDZgfEY9HxHERsVFEbATsBSwHruz1yM2sXxvY7ADMzPqw70XEYwCS/gAsiYi/5fErgH1zvdUuL0saDPwa+GxpHjOz3uIziGZmjfNYYfi5KuMb5eEDKSSIktYGLgcujojpjQ7SzKySE0QzsyaS9CZgK+CvheLvAU8CX21KUGbW7/kSs5lZcx0AXBMRASDpU8B7gN0j4pWmRmZm/ZbPIJqZNVfl/YdHANsCjxZaMn+5OaGZWX+l/KPVzMx6maSBwGJg24h4stnxmJmV+AyimVnzbAZ8zcmhmbUan0E0MzMzszI+g2hmZmZmZZwgmpmZmVkZJ4hmZmZmVsYJopmZmZmVcYJoZmZmZmWcIJqZmZlZmf8P6aivvttH5FUAAAAASUVORK5CYII=",
            "text/plain": [
              "<Figure size 720x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "# m/z values before limiting\n",
        "fig, ax = plt.subplots(1, 2, figsize=(10, 5))\n",
        "fig.suptitle(\"Distribution of m/z values\")\n",
        "fig.supxlabel(\"m/z\")\n",
        "fig.supylabel(\"Count of observations\")\n",
        "\n",
        "plt.subplot(1, 2, 1)\n",
        "plt.hist(sample_df[\"m/z\"])\n",
        "plt.title(\"Before dropping selected m/z values\")\n",
        "\n",
        "# m/z values after limiting\n",
        "sample_df = drop_frac_and_He(sample_df)\n",
        "plt.subplot(1, 2, 2)\n",
        "plt.hist(sample_df[\"m/z\"])\n",
        "plt.title(\"After dropping selected m/z values\")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4CmfDC2YSvi4"
      },
      "source": [
        "## 2. Removing background ion presences"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "metadata": {
        "id": "Qj9Sck3iSwmn"
      },
      "outputs": [],
      "source": [
        "def remove_background_abundance(df):\n",
        "    \"\"\"\n",
        "    Subtracts minimum abundance value\n",
        "\n",
        "    Args:\n",
        "        df: dataframe with 'm/z' and 'abundance' columns\n",
        "\n",
        "    Returns:\n",
        "        dataframe with minimum abundance subtracted for all observations\n",
        "    \"\"\"\n",
        "\n",
        "    df[\"abundance_minsub\"] = df.groupby([\"m/z\"])[\"abundance\"].transform(\n",
        "        lambda x: (x - x.min())\n",
        "    )\n",
        "\n",
        "    return df"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "id": "cNMwbDw9SzhH",
        "outputId": "b1d20e77-a23c-4bf0-ed25-87da27d5031e"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "Text(0.5, 1.0, 'After subtracting minimum abundance')"
            ]
          },
          "execution_count": 18,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1080x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Abundance values before subtracting minimum\n",
        "fig, ax = plt.subplots(1, 2, figsize=(15, 5))\n",
        "fig.suptitle(\"Abundance values across temperature by m/z\")\n",
        "fig.supxlabel(\"Temperature\")\n",
        "fig.supylabel(\"Abundance levels\")\n",
        "\n",
        "plt.subplot(1, 2, 1)\n",
        "for m in sample_df[\"m/z\"].unique():\n",
        "    plt.plot(\n",
        "        sample_df[sample_df[\"m/z\"] == m][\"temp\"],\n",
        "        sample_df[sample_df[\"m/z\"] == m][\"abundance\"],\n",
        "    )\n",
        "plt.title(\"Before subtracting minimum abundance\")\n",
        "\n",
        "# After subtracting minimum abundance value\n",
        "sample_df = remove_background_abundance(sample_df)\n",
        "\n",
        "plt.subplot(1, 2, 2)\n",
        "for m in sample_df[\"m/z\"].unique():\n",
        "    plt.plot(\n",
        "        sample_df[sample_df[\"m/z\"] == m][\"temp\"],\n",
        "        sample_df[sample_df[\"m/z\"] == m][\"abundance_minsub\"],\n",
        "    )\n",
        "plt.title(\"After subtracting minimum abundance\")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XST12ElGS1iI"
      },
      "source": [
        "## 3. Convert abundances to relative abundances"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "metadata": {
        "id": "cFMdr1E8S2a3"
      },
      "outputs": [],
      "source": [
        "def scale_abun(df):\n",
        "    \"\"\"\n",
        "    Scale abundance from 0-100 according to the min and max values across entire sample\n",
        "\n",
        "    Args:\n",
        "        df: dataframe containing abundances and m/z\n",
        "\n",
        "    Returns:\n",
        "        dataframe with additional column of scaled abundances\n",
        "    \"\"\"\n",
        "\n",
        "    df[\"abun_minsub_scaled\"] = minmax_scale(df[\"abundance_minsub\"].astype(float))\n",
        "\n",
        "    return df"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "N37hKjiZS3sX"
      },
      "source": [
        "## 4. Putting it all together"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "metadata": {
        "id": "9plpq9zOTBtB"
      },
      "outputs": [],
      "source": [
        "# Preprocess function\n",
        "def preprocess_sample(df):\n",
        "    df = drop_frac_and_He(df)\n",
        "    df = remove_background_abundance(df)\n",
        "    df = scale_abun(df)\n",
        "    return df"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "id": "-d1Ygj5-TDEU",
        "outputId": "ff1de97a-9b38-432a-8b5c-f4f569ac91bb"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1080x216 with 5 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1080x216 with 5 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "# For commercial\n",
        "fig, ax = plt.subplots(1, 5, figsize=(15, 3), constrained_layout=True)\n",
        "fig.suptitle(\"Commercial samples\")\n",
        "fig.supxlabel(\"Temperature\")\n",
        "fig.supylabel(\"Abundance\")\n",
        "\n",
        "for i in range(0, 5):\n",
        "    sample_lab = sample_id_commercial[i]\n",
        "    sample_df = sample_commercial_dict[sample_lab]\n",
        "    sample_df = preprocess_sample(sample_df)\n",
        "\n",
        "    plt.subplot(1, 5, i + 1)\n",
        "    for m in sample_df[\"m/z\"].unique():\n",
        "        plt.plot(\n",
        "            sample_df[sample_df[\"m/z\"] == m][\"temp\"],\n",
        "            sample_df[sample_df[\"m/z\"] == m][\"abun_minsub_scaled\"],\n",
        "        )\n",
        "\n",
        "    plt.title(sample_lab)\n",
        "\n",
        "# For testbed\n",
        "fig, ax = plt.subplots(1, 5, figsize=(15, 3), constrained_layout=True)\n",
        "fig.suptitle(\"SAM testbed samples\")\n",
        "fig.supxlabel(\"Temperature\")\n",
        "fig.supylabel(\"Abundance\")\n",
        "\n",
        "for i in range(0, 5):\n",
        "    sample_lab = sample_id_testbed[i]\n",
        "    sample_df = sample_testbed_dict[sample_lab]\n",
        "    sample_df = preprocess_sample(sample_df)\n",
        "\n",
        "    plt.subplot(1, 5, i + 1)\n",
        "    for m in sample_df[\"m/z\"].unique():\n",
        "        plt.plot(\n",
        "            sample_df[sample_df[\"m/z\"] == m][\"temp\"],\n",
        "            sample_df[sample_df[\"m/z\"] == m][\"abun_minsub_scaled\"],\n",
        "        )\n",
        "    plt.title(sample_lab)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "L5D9FdwgTL5_"
      },
      "source": [
        "# 5. Feature Engineering"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 22,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "9s_DTZX4TOG2",
        "outputId": "83b62ed3-870d-4c59-b91e-5d7142b6c294"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-bde8334b-9fda-448f-a393-fec1772a1117\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>temp_bin</th>\n",
              "      <th>m/z</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>(0, 100]</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>(0, 100]</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>(0, 100]</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>(0, 100]</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>(0, 100]</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-bde8334b-9fda-448f-a393-fec1772a1117')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-bde8334b-9fda-448f-a393-fec1772a1117 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-bde8334b-9fda-448f-a393-fec1772a1117');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "   temp_bin  m/z\n",
              "0  (0, 100]    0\n",
              "1  (0, 100]    1\n",
              "2  (0, 100]    2\n",
              "3  (0, 100]    3\n",
              "4  (0, 100]    4"
            ]
          },
          "execution_count": 22,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Create a series of temperature bins\n",
        "temprange = pd.interval_range(start=0, end=1500, freq=100)\n",
        "temprange\n",
        "\n",
        "# Make dataframe with rows that are combinations of all temperature bins\n",
        "# and all m/z values\n",
        "allcombs = list(itertools.product(temprange, [*range(0, 100)]))\n",
        "\n",
        "allcombs_df = pd.DataFrame(allcombs, columns=[\"temp_bin\", \"m/z\"])\n",
        "allcombs_df.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "metadata": {
        "id": "Uu3vG1rTTQk1"
      },
      "outputs": [],
      "source": [
        "def abun_per_tempbin(df):\n",
        "\n",
        "    \"\"\"\n",
        "    Transforms dataset to take the preprocessed max abundance for each\n",
        "    temperature range for each m/z value\n",
        "\n",
        "    Args:\n",
        "        df: dataframe to transform\n",
        "\n",
        "    Returns:\n",
        "        transformed dataframe\n",
        "    \"\"\"\n",
        "\n",
        "    # Bin temperatures\n",
        "    df[\"temp_bin\"] = pd.cut(df[\"temp\"], bins=temprange)\n",
        "\n",
        "    # Combine with a list of all temp bin-m/z value combinations\n",
        "    df = pd.merge(allcombs_df, df, on=[\"temp_bin\", \"m/z\"], how=\"left\")\n",
        "\n",
        "    # Aggregate to temperature bin level to find max\n",
        "    df = df.groupby([\"temp_bin\", \"m/z\"]).max(\"abun_minsub_scaled\").reset_index()\n",
        "\n",
        "    # Fill in 0 for abundance values without information\n",
        "    df = df.replace(np.nan, 0)\n",
        "\n",
        "    # Reshape so each row is a single sample\n",
        "    df = df.pivot_table(columns=[\"m/z\", \"temp_bin\"], values=[\"abun_minsub_scaled\"])\n",
        "\n",
        "    return df"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 24,
      "metadata": {
        "id": "b1vtCAz2TR1J"
      },
      "outputs": [],
      "source": [
        "# Assembling preprocessed and transformed training set\n",
        "def prepare_features(train_files):\n",
        "    \"\"\"\n",
        "    Preprocesses and feature engineers the features for training.\n",
        "\n",
        "    Args:\n",
        "        train_files: a dictionary with sample_ids and the corresponding csv paths\n",
        "\n",
        "    Returns:\n",
        "        The dataframe with the preprocessed and scaled features\n",
        "    \"\"\"\n",
        "\n",
        "    train_features_dict = {}\n",
        "    print(\"Total number of train files: \", len(train_files))\n",
        "\n",
        "    for i, (sample_id, filepath) in enumerate(tqdm(train_files.items())):\n",
        "\n",
        "        # Load training sample\n",
        "        temp = pd.read_csv(DATA_PATH / filepath)\n",
        "\n",
        "        # Preprocessing training sample\n",
        "        train_sample_pp = preprocess_sample(temp)\n",
        "\n",
        "        # Feature engineering\n",
        "        train_sample_fe = abun_per_tempbin(train_sample_pp).reset_index(drop=True)\n",
        "        train_features_dict[sample_id] = train_sample_fe\n",
        "\n",
        "    train_features = pd.concat(\n",
        "        train_features_dict, names=[\"sample_id\", \"dummy_index\"]\n",
        "    ).reset_index(level=\"dummy_index\", drop=True)\n",
        "\n",
        "    return train_features"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 25,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0,
          "referenced_widgets": [
            "dc7505a241264fa2be96c28436207db0",
            "f11b0cb6112b4dbe95d708986bf97a68",
            "b4613e7f5da44afab9ddddef6dcacf51",
            "398efc4a630f47cc9c27707d177d7cc0",
            "87c0eb619b614d6aa14a10bdbef59e34",
            "b48c0f8e99274ee391f10f7b6f82ef65",
            "22d2d0a3ee044435964c21ec184c31b8",
            "521d30efe29e4049be8dd0c6c12a0d2f",
            "6cb51ddcdfb44e1f8e562d97020ec548",
            "269a9b6202b5475793236b270473e29b",
            "3e9e7c013ef54979ac103b138f84726c",
            "d895b8b2fd694dd5988f31c7f9430fec",
            "4cc88f64551a45c2a2aa3c1efc7137b6",
            "6cec1c4d6b8948d98c9c9bd9a52fa5e4",
            "a41dc73f5caf4f59a8c491368b6ebf9f",
            "793435bbd40a41b4a7ae799ef72ca996",
            "2c59011ecc00415594670ea73666b29f",
            "a494aa9596dc4d49964656d1e5a0b2c6",
            "50af3fc5de1646f2b94022d9dc48c83b",
            "92481ce80e7d43d7923e9bb238ee12cf",
            "af90bbbbd0004d0f904f8a9daec10a48",
            "74c08b5283674dc7a251cd2236e8cf79"
          ]
        },
        "id": "B3PYYQEYEJna",
        "outputId": "07d5dfe8-0aad-4cb8-a4ac-a62bcff6e34d"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Total number of train files:  766\n"
          ]
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "dc7505a241264fa2be96c28436207db0",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "  0%|          | 0/766 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Total number of train files:  293\n"
          ]
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "d895b8b2fd694dd5988f31c7f9430fec",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "  0%|          | 0/293 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "train_features = prepare_features(train_files)\n",
        "val_features = prepare_features(val_files)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 26,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "id": "karNpmqY5Ubb",
        "outputId": "f988f148-e3d8-4f8c-8f62-d37e5f3a0cb4"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-edb6feb0-6c68-40e6-8fdb-b0747e3b2529\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead tr th {\n",
              "        text-align: left;\n",
              "    }\n",
              "\n",
              "    .dataframe thead tr:last-of-type th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr>\n",
              "      <th>m/z</th>\n",
              "      <th colspan=\"10\" halign=\"left\">0</th>\n",
              "      <th>...</th>\n",
              "      <th colspan=\"10\" halign=\"left\">99</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>temp_bin</th>\n",
              "      <th>(0, 100]</th>\n",
              "      <th>(100, 200]</th>\n",
              "      <th>(200, 300]</th>\n",
              "      <th>(300, 400]</th>\n",
              "      <th>(400, 500]</th>\n",
              "      <th>(500, 600]</th>\n",
              "      <th>(600, 700]</th>\n",
              "      <th>(700, 800]</th>\n",
              "      <th>(800, 900]</th>\n",
              "      <th>(900, 1000]</th>\n",
              "      <th>...</th>\n",
              "      <th>(500, 600]</th>\n",
              "      <th>(600, 700]</th>\n",
              "      <th>(700, 800]</th>\n",
              "      <th>(800, 900]</th>\n",
              "      <th>(900, 1000]</th>\n",
              "      <th>(1000, 1100]</th>\n",
              "      <th>(1100, 1200]</th>\n",
              "      <th>(1200, 1300]</th>\n",
              "      <th>(1300, 1400]</th>\n",
              "      <th>(1400, 1500]</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>sample_id</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>S0000</th>\n",
              "      <td>0.004085</td>\n",
              "      <td>0.004641</td>\n",
              "      <td>0.001394</td>\n",
              "      <td>0.000188</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000e+00</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>S0001</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000e+00</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>S0002</th>\n",
              "      <td>0.002270</td>\n",
              "      <td>0.002495</td>\n",
              "      <td>0.001688</td>\n",
              "      <td>0.000636</td>\n",
              "      <td>0.000597</td>\n",
              "      <td>0.000819</td>\n",
              "      <td>0.000155</td>\n",
              "      <td>0.000235</td>\n",
              "      <td>0.000227</td>\n",
              "      <td>0.000086</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000001</td>\n",
              "      <td>9.397170e-07</td>\n",
              "      <td>0.000001</td>\n",
              "      <td>0.000001</td>\n",
              "      <td>0.000001</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>S0003</th>\n",
              "      <td>0.001336</td>\n",
              "      <td>0.002464</td>\n",
              "      <td>0.001099</td>\n",
              "      <td>0.000992</td>\n",
              "      <td>0.000676</td>\n",
              "      <td>0.000883</td>\n",
              "      <td>0.000892</td>\n",
              "      <td>0.000631</td>\n",
              "      <td>0.000361</td>\n",
              "      <td>0.000299</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000005</td>\n",
              "      <td>4.693477e-06</td>\n",
              "      <td>0.000004</td>\n",
              "      <td>0.000004</td>\n",
              "      <td>0.000004</td>\n",
              "      <td>0.000003</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>S0004</th>\n",
              "      <td>0.005993</td>\n",
              "      <td>0.012429</td>\n",
              "      <td>0.009380</td>\n",
              "      <td>0.005099</td>\n",
              "      <td>0.006921</td>\n",
              "      <td>0.001966</td>\n",
              "      <td>0.000499</td>\n",
              "      <td>0.000880</td>\n",
              "      <td>0.000767</td>\n",
              "      <td>0.000903</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000005</td>\n",
              "      <td>3.174421e-06</td>\n",
              "      <td>0.000004</td>\n",
              "      <td>0.000004</td>\n",
              "      <td>0.000003</td>\n",
              "      <td>0.000004</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 1500 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-edb6feb0-6c68-40e6-8fdb-b0747e3b2529')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-edb6feb0-6c68-40e6-8fdb-b0747e3b2529 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-edb6feb0-6c68-40e6-8fdb-b0747e3b2529');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "m/z              0                                                          \\\n",
              "temp_bin   (0, 100] (100, 200] (200, 300] (300, 400] (400, 500] (500, 600]   \n",
              "sample_id                                                                    \n",
              "S0000      0.004085   0.004641   0.001394   0.000188   0.000000   0.000000   \n",
              "S0001      0.000000   0.000000   0.000000   0.000000   0.000000   0.000000   \n",
              "S0002      0.002270   0.002495   0.001688   0.000636   0.000597   0.000819   \n",
              "S0003      0.001336   0.002464   0.001099   0.000992   0.000676   0.000883   \n",
              "S0004      0.005993   0.012429   0.009380   0.005099   0.006921   0.001966   \n",
              "\n",
              "m/z                                                     ...         99  \\\n",
              "temp_bin  (600, 700] (700, 800] (800, 900] (900, 1000]  ... (500, 600]   \n",
              "sample_id                                               ...              \n",
              "S0000       0.000000   0.000000   0.000000    0.000000  ...   0.000000   \n",
              "S0001       0.000000   0.000000   0.000000    0.000000  ...   0.000000   \n",
              "S0002       0.000155   0.000235   0.000227    0.000086  ...   0.000001   \n",
              "S0003       0.000892   0.000631   0.000361    0.000299  ...   0.000005   \n",
              "S0004       0.000499   0.000880   0.000767    0.000903  ...   0.000005   \n",
              "\n",
              "m/z                                                                     \\\n",
              "temp_bin     (600, 700] (700, 800] (800, 900] (900, 1000] (1000, 1100]   \n",
              "sample_id                                                                \n",
              "S0000      0.000000e+00   0.000000   0.000000    0.000000     0.000000   \n",
              "S0001      0.000000e+00   0.000000   0.000000    0.000000     0.000000   \n",
              "S0002      9.397170e-07   0.000001   0.000001    0.000001     0.000000   \n",
              "S0003      4.693477e-06   0.000004   0.000004    0.000004     0.000003   \n",
              "S0004      3.174421e-06   0.000004   0.000004    0.000003     0.000004   \n",
              "\n",
              "m/z                                                            \n",
              "temp_bin  (1100, 1200] (1200, 1300] (1300, 1400] (1400, 1500]  \n",
              "sample_id                                                      \n",
              "S0000              0.0          0.0          0.0          0.0  \n",
              "S0001              0.0          0.0          0.0          0.0  \n",
              "S0002              0.0          0.0          0.0          0.0  \n",
              "S0003              0.0          0.0          0.0          0.0  \n",
              "S0004              0.0          0.0          0.0          0.0  \n",
              "\n",
              "[5 rows x 1500 columns]"
            ]
          },
          "execution_count": 26,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "train_features.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 27,
      "metadata": {
        "id": "asulKK8STVhW"
      },
      "outputs": [],
      "source": [
        "# Make sure that all sample IDs in features and labels are identical\n",
        "assert train_features.index.equals(train_labels.index)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0nOAwuXgTaz8"
      },
      "source": [
        "# 6. Perform Modeling"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 28,
      "metadata": {
        "id": "Az9e5fImTdha"
      },
      "outputs": [],
      "source": [
        "# Define stratified k-fold validation\n",
        "skf = StratifiedKFold(n_splits=10, random_state=RANDOM_SEED, shuffle=True)\n",
        "\n",
        "# Define log loss\n",
        "log_loss_scorer = make_scorer(log_loss, needs_proba=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 38,
      "metadata": {
        "id": "Q9ehrKEPTiMv"
      },
      "outputs": [],
      "source": [
        "# Check log loss score for baseline dummy model\n",
        "def logloss_cross_val(clf, X, y):\n",
        "\n",
        "    # Generate a score for each label class\n",
        "    log_loss_cv = {}\n",
        "    for col in y.columns:\n",
        "\n",
        "        y_col = y[col]  # take one label at a time\n",
        "        log_loss_cv[col] = np.mean(\n",
        "            cross_val_score(clf, X.values, y_col, cv=skf, scoring=log_loss_scorer)\n",
        "        )\n",
        "\n",
        "    avg_log_loss = np.mean(list(log_loss_cv.values()))\n",
        "\n",
        "    return log_loss_cv, avg_log_loss"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "l4OCJmtKTe9C"
      },
      "source": [
        "## Baseline dummy classifier"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 39,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kaQucTkJTjIA",
        "outputId": "923ceb52-7bab-4448-e064-5ecbbd08f6bc"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Dummy model log-loss:\n",
            "{'basalt': 0.3645372208379112,\n",
            " 'carbonate': 0.3824414251351717,\n",
            " 'chloride': 0.3697285633598953,\n",
            " 'iron_oxide': 0.5277071928108552,\n",
            " 'oxalate': 0.16528503978963754,\n",
            " 'oxychlorine': 0.5452451136713539,\n",
            " 'phyllosilicate': 0.6247744558229936,\n",
            " 'silicate': 0.39720419245594735,\n",
            " 'sulfate': 0.49271320494614584,\n",
            " 'sulfide': 0.18955707573177416}\n",
            "\n",
            "Average log-loss\n"
          ]
        },
        {
          "data": {
            "text/plain": [
              "0.40591934845616856"
            ]
          },
          "execution_count": 39,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Dummy classifier\n",
        "dummy_clf = DummyClassifier(strategy=\"prior\")\n",
        "\n",
        "# Cross Validation\n",
        "print(\"Dummy model log-loss:\")\n",
        "dummy_labelwise_logloss, dummy_avg_logloss = logloss_cross_val(dummy_clf, train_features, train_labels)\n",
        "pprint(dummy_labelwise_logloss)\n",
        "print(\"\\nAverage log-loss\")\n",
        "dummy_avg_logloss"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hOWm0SLdTlQj"
      },
      "source": [
        "## Logistic regression model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 47,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YzUIK-XaTpg2",
        "outputId": "8ae09221-d107-406e-d911-a623d693c0ec"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Logistic regression model log-loss:\n",
            "\n",
            "{'basalt': 0.27672359546453806,\n",
            " 'carbonate': 0.2614589394727089,\n",
            " 'chloride': 0.2841182874837477,\n",
            " 'iron_oxide': 0.38376399163199165,\n",
            " 'oxalate': 0.004094102362585186,\n",
            " 'oxychlorine': 0.29910400940977827,\n",
            " 'phyllosilicate': 0.4110026855029848,\n",
            " 'silicate': 0.35111797415663965,\n",
            " 'sulfate': 0.3446689121220305,\n",
            " 'sulfide': 0.13631160119625768}\n",
            "Average log-loss\n"
          ]
        },
        {
          "data": {
            "text/plain": [
              "0.2752364098803262"
            ]
          },
          "execution_count": 47,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Define logistic regression model\n",
        "logreg_clf = LogisticRegression(\n",
        "    penalty=\"l1\", solver=\"liblinear\", C=10, random_state=RANDOM_SEED\n",
        ")\n",
        "print(\"Logistic regression model log-loss:\\n\")\n",
        "logreg_labelwise_logloss, logreg_avg_logloss = logloss_cross_val(logreg_clf, train_features, train_labels)\n",
        "pprint(logreg_labelwise_logloss)\n",
        "print(\"Average log-loss\")\n",
        "logreg_avg_logloss"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FGvr8yQ83jQL"
      },
      "source": [
        "## XGBoost model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 33,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Sok7impCigMq",
        "outputId": "98c4f5f4-a35c-4809-9fc7-3004d31539fe"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "XGBoost model log-loss:\n",
            "\n",
            "{'basalt': 0.2048042224619102, 'carbonate': 0.13357722719604742, 'chloride': 0.19081779801383258, 'iron_oxide': 0.2282876959209538, 'oxalate': 0.02619778705282484, 'oxychlorine': 0.19990084894222576, 'phyllosilicate': 0.2718878906244667, 'silicate': 0.22779872838696663, 'sulfate': 0.20202114710406177, 'sulfide': 0.07551884889244612}\n",
            "Average log-loss\n"
          ]
        },
        {
          "data": {
            "text/plain": [
              "0.1760812194595736"
            ]
          },
          "execution_count": 33,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Define the model\n",
        "xgboost_model = XGBClassifier(\n",
        "    objective='binary:logistic',\n",
        "    tree_method='gpu_hist',\n",
        "    gpu_id=0,\n",
        "    seed=RANDOM_SEED\n",
        "    )\n",
        "\n",
        "print(\"XGBoost model log-loss:\\n\")\n",
        "xgboost_labelwise_logloss, xgboost_avg_logloss = logloss_cross_val(xgboost_model, train_features, train_labels)\n",
        "print(xgboost_labelwise_logloss)\n",
        "print(\"Average log-loss\")\n",
        "xgboost_avg_logloss"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Cf-trxsfv6v0"
      },
      "source": [
        "## CatBoost model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 42,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1z-955vXwXLs",
        "outputId": "8e7de76b-1c22-4829-afdc-a52e7867eb2a"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "CatBoost model log-loss:\n",
            "\n",
            "{'basalt': 0.21887766313698456, 'carbonate': 0.14202721119814415, 'chloride': 0.2017391712910893, 'iron_oxide': 0.23234814410597476, 'oxalate': 0.023221540586441016, 'oxychlorine': 0.2039719944368204, 'phyllosilicate': 0.26827676210542806, 'silicate': 0.24054264940687817, 'sulfate': 0.2207616145389581, 'sulfide': 0.09127113676510631}\n",
            "Average log-loss\n"
          ]
        },
        {
          "data": {
            "text/plain": [
              "0.1843037887571825"
            ]
          },
          "execution_count": 42,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Define the model. Adding iterations parameters to speed up pred.\n",
        "catboost_model = CatBoostClassifier(\n",
        "    task_type='GPU',\n",
        "    iterations=250,\n",
        "    random_seed=RANDOM_SEED,\n",
        "    silent=True\n",
        "    )\n",
        "\n",
        "print(\"CatBoost model log-loss:\\n\")\n",
        "catboost_labelwise_logloss, catboost_avg_logloss = logloss_cross_val(catboost_model, train_features, train_labels)\n",
        "print(catboost_labelwise_logloss)\n",
        "print(\"Average log-loss\")\n",
        "catboost_avg_logloss"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "R-34X3VP8K0X"
      },
      "source": [
        "## LightGBM model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 45,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "koKl-kuN8Pjy",
        "outputId": "7024926c-a875-46a2-94e8-8358d4c57d8e"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "LightGBM model log-loss:\n",
            "\n",
            "{'basalt': 0.2724098706088447, 'carbonate': 0.17389896569613236, 'chloride': 0.29201949205004085, 'iron_oxide': 0.2696584949125955, 'oxalate': 0.05909615464942266, 'oxychlorine': 0.2608955521550483, 'phyllosilicate': 0.31594809483366404, 'silicate': 0.3536488986996479, 'sulfate': 0.2795565018890081, 'sulfide': 0.1172351142285039}\n",
            "Average log-loss\n"
          ]
        },
        {
          "data": {
            "text/plain": [
              "0.23943671397229083"
            ]
          },
          "execution_count": 45,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Define the model\n",
        "lightgbm_model = LGBMClassifier(\n",
        "    objective='binary',\n",
        "    device_type='gpu',\n",
        "    random_seed=RANDOM_SEED,\n",
        "    verbose=-1\n",
        ")\n",
        "\n",
        "print(\"LightGBM model log-loss:\\n\")\n",
        "lightgbm_labelwise_logloss, lightgbm_avg_logloss = logloss_cross_val(lightgbm_model, train_features, train_labels)\n",
        "print(lightgbm_labelwise_logloss)\n",
        "print(\"Average log-loss\")\n",
        "lightgbm_avg_logloss"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6WZcINZYQbx2"
      },
      "source": [
        "# 7. Compare performance before optimization"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 48,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 449
        },
        "id": "SjuzR9qbsA14",
        "outputId": "059ddca9-68df-4794-87a8-f7ff5ce9efbb"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "(array([0, 1, 2, 3, 4]),\n",
              " [Text(0, 0, 'dummy_clf'),\n",
              "  Text(1, 0, 'log_reg_model'),\n",
              "  Text(2, 0, 'xgboost'),\n",
              "  Text(3, 0, 'catboost'),\n",
              "  Text(4, 0, 'lightgbm')])"
            ]
          },
          "execution_count": 48,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Create dictionary with model_names and their corresponding performance\n",
        "model_names = {'dummy_clf':dummy_avg_logloss,\n",
        "               'log_reg_model':logreg_avg_logloss,\n",
        "               'xgboost':xgboost_avg_logloss,\n",
        "               'catboost':catboost_avg_logloss,\n",
        "               'lightgbm':lightgbm_avg_logloss}\n",
        "\n",
        "# Plot the results (lower is better)\n",
        "ax = sns.barplot(x=list(model_names.keys()), y=list(model_names.values()))\n",
        "plt.title(\"Model performances\")\n",
        "plt.xlabel('model_name')\n",
        "plt.ylabel('log_loss')\n",
        "plt.xticks(rotation=45)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Sf6QDN9B_4vy"
      },
      "source": [
        "# 8. Optuna optimization"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RasZhlmYAPJZ"
      },
      "source": [
        "## Install/Import"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "k9fx3MSdBy-K"
      },
      "outputs": [],
      "source": [
        "!pip install optuna\n",
        "\n",
        "import optuna"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "utjES9mPS0Ge"
      },
      "source": [
        "## XGBoost optimization"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 50,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HPYI2uHUPhOm",
        "outputId": "5060d555-1418-4e47-8201-24bc2e50cb67"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\u001b[32m[I 2022-04-21 20:04:24,540]\u001b[0m A new study created in memory with name: XGBoost optimization\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:04:40,337]\u001b[0m Trial 0 finished with value: 0.16658454721823482 and parameters: {'depth': 5, 'n_estimators': 400, 'iterations': 80, 'learning_rate': 0.13, 'lambda': 0.00288632979835963, 'alpha': 0.042126769442889164}. Best is trial 0 with value: 0.16658454721823482.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:04:49,350]\u001b[0m Trial 1 finished with value: 0.1706214995843769 and parameters: {'depth': 9, 'n_estimators': 130, 'iterations': 80, 'learning_rate': 0.32, 'lambda': 0.32371728751351364, 'alpha': 0.012520451149282406}. Best is trial 0 with value: 0.16658454721823482.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:05:02,289]\u001b[0m Trial 2 finished with value: 0.17912825143550926 and parameters: {'depth': 6, 'n_estimators': 340, 'iterations': 390, 'learning_rate': 0.44, 'lambda': 3.701481341922972, 'alpha': 0.3287101719128261}. Best is trial 0 with value: 0.16658454721823482.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:05:18,062]\u001b[0m Trial 3 finished with value: 0.1839623767779263 and parameters: {'depth': 3, 'n_estimators': 470, 'iterations': 270, 'learning_rate': 0.35000000000000003, 'lambda': 1.351223434779919, 'alpha': 0.0011096501534686137}. Best is trial 0 with value: 0.16658454721823482.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:05:27,761]\u001b[0m Trial 4 finished with value: 0.18139689980341395 and parameters: {'depth': 11, 'n_estimators': 170, 'iterations': 200, 'learning_rate': 0.47000000000000003, 'lambda': 0.001223371305510308, 'alpha': 0.003685728270190711}. Best is trial 0 with value: 0.16658454721823482.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:05:35,592]\u001b[0m Trial 5 finished with value: 0.20502551061752236 and parameters: {'depth': 12, 'n_estimators': 80, 'iterations': 380, 'learning_rate': 0.85, 'lambda': 0.0038822474494699596, 'alpha': 0.08383058899551178}. Best is trial 0 with value: 0.16658454721823482.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:05:50,397]\u001b[0m Trial 6 finished with value: 0.19063392276246074 and parameters: {'depth': 5, 'n_estimators': 440, 'iterations': 240, 'learning_rate': 0.52, 'lambda': 0.14783220064050706, 'alpha': 0.5882782285871188}. Best is trial 0 with value: 0.16658454721823482.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:06:02,559]\u001b[0m Trial 7 finished with value: 0.18076220343400123 and parameters: {'depth': 4, 'n_estimators': 300, 'iterations': 120, 'learning_rate': 0.45, 'lambda': 0.017827894090961156, 'alpha': 0.0018484286363770071}. Best is trial 0 with value: 0.16658454721823482.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:06:10,908]\u001b[0m Trial 8 finished with value: 0.18077919322953417 and parameters: {'depth': 11, 'n_estimators': 110, 'iterations': 210, 'learning_rate': 0.5800000000000001, 'lambda': 8.467298187187106, 'alpha': 0.13073004583906667}. Best is trial 0 with value: 0.16658454721823482.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:06:23,135]\u001b[0m Trial 9 finished with value: 0.1931348015377495 and parameters: {'depth': 10, 'n_estimators': 310, 'iterations': 340, 'learning_rate': 0.66, 'lambda': 0.17618729116428436, 'alpha': 0.001606696750575315}. Best is trial 0 with value: 0.16658454721823482.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:06:40,158]\u001b[0m Trial 10 finished with value: 0.19929920196749346 and parameters: {'depth': 7, 'n_estimators': 400, 'iterations': 500, 'learning_rate': 0.01, 'lambda': 0.01708107744563329, 'alpha': 9.939305874678189}. Best is trial 0 with value: 0.16658454721823482.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:06:51,970]\u001b[0m Trial 11 finished with value: 0.1621500327485607 and parameters: {'depth': 9, 'n_estimators': 240, 'iterations': 90, 'learning_rate': 0.13, 'lambda': 0.45398146869266764, 'alpha': 0.016646599814269226}. Best is trial 11 with value: 0.1621500327485607.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:07:03,249]\u001b[0m Trial 12 finished with value: 0.15675631456119363 and parameters: {'depth': 8, 'n_estimators': 200, 'iterations': 140, 'learning_rate': 0.08, 'lambda': 0.037603456185284705, 'alpha': 0.01733322612877168}. Best is trial 12 with value: 0.15675631456119363.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:07:14,464]\u001b[0m Trial 13 finished with value: 0.16302262645850926 and parameters: {'depth': 8, 'n_estimators': 220, 'iterations': 160, 'learning_rate': 0.15000000000000002, 'lambda': 0.022856594063094822, 'alpha': 0.011229119836074549}. Best is trial 12 with value: 0.15675631456119363.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:07:26,241]\u001b[0m Trial 14 finished with value: 0.16848549329928225 and parameters: {'depth': 8, 'n_estimators': 230, 'iterations': 160, 'learning_rate': 0.18000000000000002, 'lambda': 0.49551394163755097, 'alpha': 0.01220201452410933}. Best is trial 12 with value: 0.15675631456119363.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:07:38,304]\u001b[0m Trial 15 finished with value: 0.16437871218718583 and parameters: {'depth': 9, 'n_estimators': 230, 'iterations': 140, 'learning_rate': 0.04, 'lambda': 0.05908820235028741, 'alpha': 0.02471548082168537}. Best is trial 12 with value: 0.15675631456119363.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:07:48,396]\u001b[0m Trial 16 finished with value: 0.17293225037285584 and parameters: {'depth': 9, 'n_estimators': 180, 'iterations': 310, 'learning_rate': 0.27, 'lambda': 0.7440422551697732, 'alpha': 1.9276569987610184}. Best is trial 12 with value: 0.15675631456119363.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:07:59,460]\u001b[0m Trial 17 finished with value: 0.21632384087496853 and parameters: {'depth': 7, 'n_estimators': 260, 'iterations': 110, 'learning_rate': 1.0, 'lambda': 0.06971095993441395, 'alpha': 0.006155383492079995}. Best is trial 12 with value: 0.15675631456119363.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:08:09,467]\u001b[0m Trial 18 finished with value: 0.1714226953128693 and parameters: {'depth': 10, 'n_estimators': 170, 'iterations': 190, 'learning_rate': 0.23, 'lambda': 1.2411598404371786, 'alpha': 0.05430032265144513}. Best is trial 12 with value: 0.15675631456119363.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:08:24,234]\u001b[0m Trial 19 finished with value: 0.16313437586777613 and parameters: {'depth': 6, 'n_estimators': 350, 'iterations': 480, 'learning_rate': 0.09, 'lambda': 0.03644274304246995, 'alpha': 0.025749917394087395}. Best is trial 12 with value: 0.15675631456119363.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:08:35,523]\u001b[0m Trial 20 finished with value: 0.19496840874520938 and parameters: {'depth': 10, 'n_estimators': 260, 'iterations': 250, 'learning_rate': 0.7000000000000001, 'lambda': 0.008961515240307397, 'alpha': 0.18443043620713787}. Best is trial 12 with value: 0.15675631456119363.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:08:46,685]\u001b[0m Trial 21 finished with value: 0.16238184288381693 and parameters: {'depth': 8, 'n_estimators': 210, 'iterations': 160, 'learning_rate': 0.15000000000000002, 'lambda': 0.02791243237009956, 'alpha': 0.005542390895601359}. Best is trial 12 with value: 0.15675631456119363.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:08:57,204]\u001b[0m Trial 22 finished with value: 0.16560639397505145 and parameters: {'depth': 8, 'n_estimators': 190, 'iterations': 110, 'learning_rate': 0.22, 'lambda': 0.16437582742905307, 'alpha': 0.004128650026162799}. Best is trial 12 with value: 0.15675631456119363.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:09:06,200]\u001b[0m Trial 23 finished with value: 0.17484750050729628 and parameters: {'depth': 7, 'n_estimators': 130, 'iterations': 170, 'learning_rate': 0.35000000000000003, 'lambda': 0.008138589729380667, 'alpha': 0.006119288031255175}. Best is trial 12 with value: 0.15675631456119363.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:09:19,089]\u001b[0m Trial 24 finished with value: 0.15769047346609888 and parameters: {'depth': 9, 'n_estimators': 270, 'iterations': 80, 'learning_rate': 0.06999999999999999, 'lambda': 0.05764403558752663, 'alpha': 0.025742014877676533}. Best is trial 12 with value: 0.15675631456119363.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:09:32,219]\u001b[0m Trial 25 finished with value: 0.15618879734667374 and parameters: {'depth': 9, 'n_estimators': 280, 'iterations': 80, 'learning_rate': 0.060000000000000005, 'lambda': 0.0808408722711427, 'alpha': 0.022674221126685937}. Best is trial 25 with value: 0.15618879734667374.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:09:45,474]\u001b[0m Trial 26 finished with value: 0.1561850297605533 and parameters: {'depth': 11, 'n_estimators': 270, 'iterations': 130, 'learning_rate': 0.060000000000000005, 'lambda': 0.070494218855612, 'alpha': 0.05277449197608768}. Best is trial 26 with value: 0.1561850297605533.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:10:00,553]\u001b[0m Trial 27 finished with value: 0.21194020965972551 and parameters: {'depth': 12, 'n_estimators': 330, 'iterations': 130, 'learning_rate': 0.01, 'lambda': 0.09999798546614737, 'alpha': 0.07258433505206575}. Best is trial 26 with value: 0.1561850297605533.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:10:14,783]\u001b[0m Trial 28 finished with value: 0.1812714476183822 and parameters: {'depth': 11, 'n_estimators': 380, 'iterations': 140, 'learning_rate': 0.3, 'lambda': 0.00950584861861678, 'alpha': 0.23205694369348803}. Best is trial 26 with value: 0.1561850297605533.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:10:27,834]\u001b[0m Trial 29 finished with value: 0.1643497689073275 and parameters: {'depth': 10, 'n_estimators': 290, 'iterations': 230, 'learning_rate': 0.09999999999999999, 'lambda': 0.24307777275562215, 'alpha': 0.04120556352396425}. Best is trial 26 with value: 0.1561850297605533.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:10:42,004]\u001b[0m Trial 30 finished with value: 0.17209422408572225 and parameters: {'depth': 11, 'n_estimators': 360, 'iterations': 190, 'learning_rate': 0.21000000000000002, 'lambda': 0.003443597134403389, 'alpha': 0.9818032743183654}. Best is trial 26 with value: 0.1561850297605533.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:10:54,909]\u001b[0m Trial 31 finished with value: 0.15769047346609888 and parameters: {'depth': 9, 'n_estimators': 270, 'iterations': 80, 'learning_rate': 0.06999999999999999, 'lambda': 0.04943035072963475, 'alpha': 0.03347396790131185}. Best is trial 26 with value: 0.1561850297605533.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:11:08,490]\u001b[0m Trial 32 finished with value: 0.15801984995710888 and parameters: {'depth': 10, 'n_estimators': 290, 'iterations': 80, 'learning_rate': 0.06999999999999999, 'lambda': 0.09228394925229952, 'alpha': 0.048546042913984795}. Best is trial 26 with value: 0.1561850297605533.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:11:21,767]\u001b[0m Trial 33 finished with value: 0.22841281404419972 and parameters: {'depth': 6, 'n_estimators': 270, 'iterations': 100, 'learning_rate': 0.01, 'lambda': 0.05324713903017978, 'alpha': 0.01744880479893584}. Best is trial 26 with value: 0.1561850297605533.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:11:35,764]\u001b[0m Trial 34 finished with value: 0.164711204193082 and parameters: {'depth': 9, 'n_estimators': 330, 'iterations': 120, 'learning_rate': 0.13, 'lambda': 0.03715105929510017, 'alpha': 0.12913450508157756}. Best is trial 26 with value: 0.1561850297605533.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:11:46,051]\u001b[0m Trial 35 finished with value: 0.18123774599651427 and parameters: {'depth': 12, 'n_estimators': 190, 'iterations': 80, 'learning_rate': 0.38, 'lambda': 0.01548955935754217, 'alpha': 0.03269492662055817}. Best is trial 26 with value: 0.1561850297605533.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:11:57,768]\u001b[0m Trial 36 finished with value: 0.1713386501634683 and parameters: {'depth': 8, 'n_estimators': 250, 'iterations': 140, 'learning_rate': 0.26, 'lambda': 0.22430406170411504, 'alpha': 0.008753910246130172}. Best is trial 26 with value: 0.1561850297605533.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:12:08,925]\u001b[0m Trial 37 finished with value: 0.1697916224514635 and parameters: {'depth': 11, 'n_estimators': 210, 'iterations': 280, 'learning_rate': 0.19, 'lambda': 0.1031686310814099, 'alpha': 0.017937415479616955}. Best is trial 26 with value: 0.1561850297605533.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:12:19,026]\u001b[0m Trial 38 finished with value: 0.1566457797965039 and parameters: {'depth': 6, 'n_estimators': 160, 'iterations': 100, 'learning_rate': 0.09, 'lambda': 0.0049311261038336935, 'alpha': 0.048563052722402876}. Best is trial 26 with value: 0.1561850297605533.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:12:28,819]\u001b[0m Trial 39 finished with value: 0.15915037793650144 and parameters: {'depth': 5, 'n_estimators': 150, 'iterations': 390, 'learning_rate': 0.12, 'lambda': 0.0012547671216406675, 'alpha': 0.09275257027117793}. Best is trial 26 with value: 0.1561850297605533.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:12:38,105]\u001b[0m Trial 40 finished with value: 0.17529729510406447 and parameters: {'depth': 3, 'n_estimators': 150, 'iterations': 210, 'learning_rate': 0.39, 'lambda': 0.005302293132064082, 'alpha': 0.002586060651448032}. Best is trial 26 with value: 0.1561850297605533.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:12:46,369]\u001b[0m Trial 41 finished with value: 0.17532736995978976 and parameters: {'depth': 7, 'n_estimators': 90, 'iterations': 100, 'learning_rate': 0.06999999999999999, 'lambda': 0.001808722887208765, 'alpha': 0.05689696751931455}. Best is trial 26 with value: 0.1561850297605533.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:13:00,401]\u001b[0m Trial 42 finished with value: 0.15665812353035025 and parameters: {'depth': 6, 'n_estimators': 310, 'iterations': 120, 'learning_rate': 0.060000000000000005, 'lambda': 0.03750466929438572, 'alpha': 0.3742174827125889}. Best is trial 26 with value: 0.1561850297605533.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:13:14,601]\u001b[0m Trial 43 finished with value: 0.1579466135623414 and parameters: {'depth': 6, 'n_estimators': 310, 'iterations': 170, 'learning_rate': 0.05, 'lambda': 0.014381992134950707, 'alpha': 0.9514277187928534}. Best is trial 26 with value: 0.1561850297605533.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:13:31,758]\u001b[0m Trial 44 finished with value: 0.17296580137728845 and parameters: {'depth': 4, 'n_estimators': 500, 'iterations': 130, 'learning_rate': 0.17, 'lambda': 0.0021505902720330857, 'alpha': 0.423480952300329}. Best is trial 26 with value: 0.1561850297605533.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:13:41,308]\u001b[0m Trial 45 finished with value: 0.15935901572154093 and parameters: {'depth': 5, 'n_estimators': 140, 'iterations': 420, 'learning_rate': 0.12, 'lambda': 0.03228887726681336, 'alpha': 0.2506194086448684}. Best is trial 26 with value: 0.1561850297605533.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:13:50,095]\u001b[0m Trial 46 finished with value: 0.16311048730636815 and parameters: {'depth': 6, 'n_estimators': 110, 'iterations': 110, 'learning_rate': 0.26, 'lambda': 0.12665103645488435, 'alpha': 0.13699370405768466}. Best is trial 26 with value: 0.1561850297605533.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:14:07,161]\u001b[0m Trial 47 finished with value: 0.159192807333915 and parameters: {'depth': 7, 'n_estimators': 420, 'iterations': 150, 'learning_rate': 0.03, 'lambda': 0.3258702778976667, 'alpha': 5.520869123412087}. Best is trial 26 with value: 0.1561850297605533.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:14:19,567]\u001b[0m Trial 48 finished with value: 0.19485524751622435 and parameters: {'depth': 6, 'n_estimators': 310, 'iterations': 180, 'learning_rate': 0.53, 'lambda': 2.9169748112300278, 'alpha': 0.5890098940418405}. Best is trial 26 with value: 0.1561850297605533.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:14:30,822]\u001b[0m Trial 49 finished with value: 0.16492002309326698 and parameters: {'depth': 4, 'n_estimators': 200, 'iterations': 310, 'learning_rate': 0.16, 'lambda': 0.02015634090620232, 'alpha': 0.07250820891999482}. Best is trial 26 with value: 0.1561850297605533.\u001b[0m\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Number of finished trials:  50\n",
            "Best trial:\n",
            "  Value: 0.1561850297605533\n",
            "  Params: \n",
            "    depth: 11\n",
            "    n_estimators: 270\n",
            "    iterations: 130\n",
            "    learning_rate: 0.060000000000000005\n",
            "    lambda: 0.070494218855612\n",
            "    alpha: 0.05277449197608768\n"
          ]
        }
      ],
      "source": [
        "# define hyperparameters and scopes\n",
        "def objective(trial):\n",
        "\n",
        "    # Define model parameters\n",
        "    params = {\n",
        "        \"depth\": trial.suggest_int(\"depth\", 3, 12, step=1),\n",
        "        \"n_estimators\": trial.suggest_int(\"n_estimators\", 80, 500, step=10),\n",
        "        \"iterations\": trial.suggest_int(\"iterations\", 80, 500, step=10),\n",
        "        \"learning_rate\": trial.suggest_float(\"learning_rate\", 0.01, 1.0, step=0.01),\n",
        "        'lambda': trial.suggest_loguniform('lambda', 1e-3, 10.0),\n",
        "        'alpha': trial.suggest_loguniform('alpha', 1e-3, 10.0),\n",
        "    }\n",
        "\n",
        "    # Define model\n",
        "    model = XGBClassifier(**params,\n",
        "                        objective='binary:logistic',\n",
        "                        eval_metric= \"logloss\",\n",
        "                        tree_method='gpu_hist',\n",
        "                        gpu_id=0,\n",
        "                        seed=RANDOM_SEED\n",
        "                        )\n",
        "    \n",
        "    \n",
        "    # Evaluate log_loss for each label\n",
        "    aggr_log_loss = {}\n",
        "\n",
        "    for col in val_labels.columns:\n",
        "\n",
        "        train_y_col = train_labels[col]\n",
        "        val_y_col = val_labels[col]\n",
        "\n",
        "        model.fit(\n",
        "            train_features.values,\n",
        "            train_y_col\n",
        "            )\n",
        "\n",
        "        pred_labels = model.predict_proba(val_features.values)\n",
        "        aggr_log_loss[col] = log_loss(y_true=val_y_col.values, y_pred=pred_labels, labels=[0,1])\n",
        "\n",
        "    # Calculate aggregated log_loss\n",
        "    avg_log_loss = np.mean(list(aggr_log_loss.values()))\n",
        "\n",
        "    return avg_log_loss\n",
        "\n",
        "xgboost_study = optuna.create_study(direction=\"minimize\", study_name='XGBoost optimization')\n",
        "xgboost_study.optimize(objective, n_trials=100, timeout=600)\n",
        "\n",
        "print(\"Number of finished trials: \", len(xgboost_study.trials))\n",
        "print(\"Best trial:\")\n",
        "xgboost_trial = xgboost_study.best_trial\n",
        "\n",
        "print(\"  Value: {}\".format(xgboost_trial.value))\n",
        "print(\"  Params: \")\n",
        "for key, value in xgboost_trial.params.items():\n",
        "\n",
        "    print(\"    {}: {}\".format(key, value))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 51,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 542
        },
        "id": "jpKC8r07Lcfi",
        "outputId": "467547b3-b26c-493e-bfba-23a687c77949"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<html>\n",
              "<head><meta charset=\"utf-8\" /></head>\n",
              "<body>\n",
              "    <div>            <script src=\"https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/MathJax.js?config=TeX-AMS-MML_SVG\"></script><script type=\"text/javascript\">if (window.MathJax) {MathJax.Hub.Config({SVG: {font: \"STIX-Web\"}});}</script>                <script type=\"text/javascript\">window.PlotlyConfig = {MathJaxConfig: 'local'};</script>\n",
              "        <script src=\"https://cdn.plot.ly/plotly-2.8.3.min.js\"></script>                <div id=\"35f9a321-78a6-4853-a0c9-2b1bdad1a65f\" class=\"plotly-graph-div\" style=\"height:525px; width:100%;\"></div>            <script type=\"text/javascript\">                                    window.PLOTLYENV=window.PLOTLYENV || {};                                    if (document.getElementById(\"35f9a321-78a6-4853-a0c9-2b1bdad1a65f\")) {                    Plotly.newPlot(                        \"35f9a321-78a6-4853-a0c9-2b1bdad1a65f\",                        [{\"mode\":\"markers\",\"name\":\"Objective Value\",\"x\":[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49],\"y\":[0.16658454721823482,0.1706214995843769,0.17912825143550926,0.1839623767779263,0.18139689980341395,0.20502551061752236,0.19063392276246074,0.18076220343400123,0.18077919322953417,0.1931348015377495,0.19929920196749346,0.1621500327485607,0.15675631456119363,0.16302262645850926,0.16848549329928225,0.16437871218718583,0.17293225037285584,0.21632384087496853,0.1714226953128693,0.16313437586777613,0.19496840874520938,0.16238184288381693,0.16560639397505145,0.17484750050729628,0.15769047346609888,0.15618879734667374,0.1561850297605533,0.21194020965972551,0.1812714476183822,0.1643497689073275,0.17209422408572225,0.15769047346609888,0.15801984995710888,0.22841281404419972,0.164711204193082,0.18123774599651427,0.1713386501634683,0.1697916224514635,0.1566457797965039,0.15915037793650144,0.17529729510406447,0.17532736995978976,0.15665812353035025,0.1579466135623414,0.17296580137728845,0.15935901572154093,0.16311048730636815,0.159192807333915,0.19485524751622435,0.16492002309326698],\"type\":\"scatter\"},{\"name\":\"Best Value\",\"x\":[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49],\"y\":[0.16658454721823482,0.16658454721823482,0.16658454721823482,0.16658454721823482,0.16658454721823482,0.16658454721823482,0.16658454721823482,0.16658454721823482,0.16658454721823482,0.16658454721823482,0.16658454721823482,0.1621500327485607,0.15675631456119363,0.15675631456119363,0.15675631456119363,0.15675631456119363,0.15675631456119363,0.15675631456119363,0.15675631456119363,0.15675631456119363,0.15675631456119363,0.15675631456119363,0.15675631456119363,0.15675631456119363,0.15675631456119363,0.15618879734667374,0.1561850297605533,0.1561850297605533,0.1561850297605533,0.1561850297605533,0.1561850297605533,0.1561850297605533,0.1561850297605533,0.1561850297605533,0.1561850297605533,0.1561850297605533,0.1561850297605533,0.1561850297605533,0.1561850297605533,0.1561850297605533,0.1561850297605533,0.1561850297605533,0.1561850297605533,0.1561850297605533,0.1561850297605533,0.1561850297605533,0.1561850297605533,0.1561850297605533,0.1561850297605533,0.1561850297605533],\"type\":\"scatter\"}],                        {\"title\":{\"text\":\"Optimization History Plot\"},\"xaxis\":{\"title\":{\"text\":\"#Trials\"}},\"yaxis\":{\"title\":{\"text\":\"Objective Value\"}},\"template\":{\"data\":{\"bar\":[{\"error_x\":{\"color\":\"#2a3f5f\"},\"error_y\":{\"color\":\"#2a3f5f\"},\"marker\":{\"line\":{\"color\":\"#E5ECF6\",\"width\":0.5},\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"bar\"}],\"barpolar\":[{\"marker\":{\"line\":{\"color\":\"#E5ECF6\",\"width\":0.5},\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"barpolar\"}],\"carpet\":[{\"aaxis\":{\"endlinecolor\":\"#2a3f5f\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"minorgridcolor\":\"white\",\"startlinecolor\":\"#2a3f5f\"},\"baxis\":{\"endlinecolor\":\"#2a3f5f\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"minorgridcolor\":\"white\",\"startlinecolor\":\"#2a3f5f\"},\"type\":\"carpet\"}],\"choropleth\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"type\":\"choropleth\"}],\"contour\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"contour\"}],\"contourcarpet\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"type\":\"contourcarpet\"}],\"heatmap\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"heatmap\"}],\"heatmapgl\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"heatmapgl\"}],\"histogram\":[{\"marker\":{\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"histogram\"}],\"histogram2d\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"histogram2d\"}],\"histogram2dcontour\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"histogram2dcontour\"}],\"mesh3d\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"type\":\"mesh3d\"}],\"parcoords\":[{\"line\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"parcoords\"}],\"pie\":[{\"automargin\":true,\"type\":\"pie\"}],\"scatter\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatter\"}],\"scatter3d\":[{\"line\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatter3d\"}],\"scattercarpet\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattercarpet\"}],\"scattergeo\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattergeo\"}],\"scattergl\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattergl\"}],\"scattermapbox\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattermapbox\"}],\"scatterpolar\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatterpolar\"}],\"scatterpolargl\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatterpolargl\"}],\"scatterternary\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatterternary\"}],\"surface\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"surface\"}],\"table\":[{\"cells\":{\"fill\":{\"color\":\"#EBF0F8\"},\"line\":{\"color\":\"white\"}},\"header\":{\"fill\":{\"color\":\"#C8D4E3\"},\"line\":{\"color\":\"white\"}},\"type\":\"table\"}]},\"layout\":{\"annotationdefaults\":{\"arrowcolor\":\"#2a3f5f\",\"arrowhead\":0,\"arrowwidth\":1},\"autotypenumbers\":\"strict\",\"coloraxis\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"colorscale\":{\"diverging\":[[0,\"#8e0152\"],[0.1,\"#c51b7d\"],[0.2,\"#de77ae\"],[0.3,\"#f1b6da\"],[0.4,\"#fde0ef\"],[0.5,\"#f7f7f7\"],[0.6,\"#e6f5d0\"],[0.7,\"#b8e186\"],[0.8,\"#7fbc41\"],[0.9,\"#4d9221\"],[1,\"#276419\"]],\"sequential\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"sequentialminus\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]},\"colorway\":[\"#636efa\",\"#EF553B\",\"#00cc96\",\"#ab63fa\",\"#FFA15A\",\"#19d3f3\",\"#FF6692\",\"#B6E880\",\"#FF97FF\",\"#FECB52\"],\"font\":{\"color\":\"#2a3f5f\"},\"geo\":{\"bgcolor\":\"white\",\"lakecolor\":\"white\",\"landcolor\":\"#E5ECF6\",\"showlakes\":true,\"showland\":true,\"subunitcolor\":\"white\"},\"hoverlabel\":{\"align\":\"left\"},\"hovermode\":\"closest\",\"mapbox\":{\"style\":\"light\"},\"paper_bgcolor\":\"white\",\"plot_bgcolor\":\"#E5ECF6\",\"polar\":{\"angularaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"bgcolor\":\"#E5ECF6\",\"radialaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"}},\"scene\":{\"xaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"gridwidth\":2,\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\"},\"yaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"gridwidth\":2,\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\"},\"zaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"gridwidth\":2,\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\"}},\"shapedefaults\":{\"line\":{\"color\":\"#2a3f5f\"}},\"ternary\":{\"aaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"baxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"bgcolor\":\"#E5ECF6\",\"caxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"}},\"title\":{\"x\":0.05},\"xaxis\":{\"automargin\":true,\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\",\"title\":{\"standoff\":15},\"zerolinecolor\":\"white\",\"zerolinewidth\":2},\"yaxis\":{\"automargin\":true,\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\",\"title\":{\"standoff\":15},\"zerolinecolor\":\"white\",\"zerolinewidth\":2}}}},                        {\"responsive\": true}                    ).then(function(){\n",
              "                            \n",
              "var gd = document.getElementById('35f9a321-78a6-4853-a0c9-2b1bdad1a65f');\n",
              "var x = new MutationObserver(function (mutations, observer) {{\n",
              "        var display = window.getComputedStyle(gd).display;\n",
              "        if (!display || display === 'none') {{\n",
              "            console.log([gd, 'removed!']);\n",
              "            Plotly.purge(gd);\n",
              "            observer.disconnect();\n",
              "        }}\n",
              "}});\n",
              "\n",
              "// Listen for the removal of the full notebook cells\n",
              "var notebookContainer = gd.closest('#notebook-container');\n",
              "if (notebookContainer) {{\n",
              "    x.observe(notebookContainer, {childList: true});\n",
              "}}\n",
              "\n",
              "// Listen for the clearing of the current output cell\n",
              "var outputEl = gd.closest('.output');\n",
              "if (outputEl) {{\n",
              "    x.observe(outputEl, {childList: true});\n",
              "}}\n",
              "\n",
              "                        })                };                            </script>        </div>\n",
              "</body>\n",
              "</html>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Plot_optimization_histor: shows the scores from all trials as well as the best score so far at each point.\n",
        "optuna.visualization.plot_optimization_history(xgboost_study)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 52,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 542
        },
        "id": "5iy2B0Q17VP1",
        "outputId": "1071924d-3cdf-45e7-d448-78131304eb1c"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<html>\n",
              "<head><meta charset=\"utf-8\" /></head>\n",
              "<body>\n",
              "    <div>            <script src=\"https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/MathJax.js?config=TeX-AMS-MML_SVG\"></script><script type=\"text/javascript\">if (window.MathJax) {MathJax.Hub.Config({SVG: {font: \"STIX-Web\"}});}</script>                <script type=\"text/javascript\">window.PlotlyConfig = {MathJaxConfig: 'local'};</script>\n",
              "        <script src=\"https://cdn.plot.ly/plotly-2.8.3.min.js\"></script>                <div id=\"fc406907-905c-4f73-8b7d-edade7da76f7\" class=\"plotly-graph-div\" style=\"height:525px; width:100%;\"></div>            <script type=\"text/javascript\">                                    window.PLOTLYENV=window.PLOTLYENV || {};                                    if (document.getElementById(\"fc406907-905c-4f73-8b7d-edade7da76f7\")) {                    Plotly.newPlot(                        \"fc406907-905c-4f73-8b7d-edade7da76f7\",                        [{\"cliponaxis\":false,\"hovertemplate\":[\"alpha (LogUniformDistribution): 0.0016443164602132048<extra></extra>\",\"iterations (IntUniformDistribution): 0.00357050915856684<extra></extra>\",\"depth (IntUniformDistribution): 0.006351153553098985<extra></extra>\",\"lambda (LogUniformDistribution): 0.006806638944702465<extra></extra>\",\"n_estimators (IntUniformDistribution): 0.006818791624193394<extra></extra>\",\"learning_rate (DiscreteUniformDistribution): 0.9748085902592251<extra></extra>\"],\"marker\":{\"color\":\"rgb(66,146,198)\"},\"orientation\":\"h\",\"text\":[\"0.0016443164602132048\",\"0.00357050915856684\",\"0.006351153553098985\",\"0.006806638944702465\",\"0.006818791624193394\",\"0.9748085902592251\"],\"textposition\":\"outside\",\"texttemplate\":\"%{text:.2f}\",\"x\":[0.0016443164602132048,0.00357050915856684,0.006351153553098985,0.006806638944702465,0.006818791624193394,0.9748085902592251],\"y\":[\"alpha\",\"iterations\",\"depth\",\"lambda\",\"n_estimators\",\"learning_rate\"],\"type\":\"bar\"}],                        {\"showlegend\":false,\"title\":{\"text\":\"Hyperparameter Importances\"},\"xaxis\":{\"title\":{\"text\":\"Importance for Objective Value\"}},\"yaxis\":{\"title\":{\"text\":\"Hyperparameter\"}},\"template\":{\"data\":{\"bar\":[{\"error_x\":{\"color\":\"#2a3f5f\"},\"error_y\":{\"color\":\"#2a3f5f\"},\"marker\":{\"line\":{\"color\":\"#E5ECF6\",\"width\":0.5},\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"bar\"}],\"barpolar\":[{\"marker\":{\"line\":{\"color\":\"#E5ECF6\",\"width\":0.5},\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"barpolar\"}],\"carpet\":[{\"aaxis\":{\"endlinecolor\":\"#2a3f5f\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"minorgridcolor\":\"white\",\"startlinecolor\":\"#2a3f5f\"},\"baxis\":{\"endlinecolor\":\"#2a3f5f\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"minorgridcolor\":\"white\",\"startlinecolor\":\"#2a3f5f\"},\"type\":\"carpet\"}],\"choropleth\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"type\":\"choropleth\"}],\"contour\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"contour\"}],\"contourcarpet\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"type\":\"contourcarpet\"}],\"heatmap\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"heatmap\"}],\"heatmapgl\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"heatmapgl\"}],\"histogram\":[{\"marker\":{\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"histogram\"}],\"histogram2d\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"histogram2d\"}],\"histogram2dcontour\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"histogram2dcontour\"}],\"mesh3d\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"type\":\"mesh3d\"}],\"parcoords\":[{\"line\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"parcoords\"}],\"pie\":[{\"automargin\":true,\"type\":\"pie\"}],\"scatter\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatter\"}],\"scatter3d\":[{\"line\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatter3d\"}],\"scattercarpet\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattercarpet\"}],\"scattergeo\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattergeo\"}],\"scattergl\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattergl\"}],\"scattermapbox\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattermapbox\"}],\"scatterpolar\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatterpolar\"}],\"scatterpolargl\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatterpolargl\"}],\"scatterternary\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatterternary\"}],\"surface\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"surface\"}],\"table\":[{\"cells\":{\"fill\":{\"color\":\"#EBF0F8\"},\"line\":{\"color\":\"white\"}},\"header\":{\"fill\":{\"color\":\"#C8D4E3\"},\"line\":{\"color\":\"white\"}},\"type\":\"table\"}]},\"layout\":{\"annotationdefaults\":{\"arrowcolor\":\"#2a3f5f\",\"arrowhead\":0,\"arrowwidth\":1},\"autotypenumbers\":\"strict\",\"coloraxis\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"colorscale\":{\"diverging\":[[0,\"#8e0152\"],[0.1,\"#c51b7d\"],[0.2,\"#de77ae\"],[0.3,\"#f1b6da\"],[0.4,\"#fde0ef\"],[0.5,\"#f7f7f7\"],[0.6,\"#e6f5d0\"],[0.7,\"#b8e186\"],[0.8,\"#7fbc41\"],[0.9,\"#4d9221\"],[1,\"#276419\"]],\"sequential\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"sequentialminus\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]},\"colorway\":[\"#636efa\",\"#EF553B\",\"#00cc96\",\"#ab63fa\",\"#FFA15A\",\"#19d3f3\",\"#FF6692\",\"#B6E880\",\"#FF97FF\",\"#FECB52\"],\"font\":{\"color\":\"#2a3f5f\"},\"geo\":{\"bgcolor\":\"white\",\"lakecolor\":\"white\",\"landcolor\":\"#E5ECF6\",\"showlakes\":true,\"showland\":true,\"subunitcolor\":\"white\"},\"hoverlabel\":{\"align\":\"left\"},\"hovermode\":\"closest\",\"mapbox\":{\"style\":\"light\"},\"paper_bgcolor\":\"white\",\"plot_bgcolor\":\"#E5ECF6\",\"polar\":{\"angularaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"bgcolor\":\"#E5ECF6\",\"radialaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"}},\"scene\":{\"xaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"gridwidth\":2,\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\"},\"yaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"gridwidth\":2,\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\"},\"zaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"gridwidth\":2,\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\"}},\"shapedefaults\":{\"line\":{\"color\":\"#2a3f5f\"}},\"ternary\":{\"aaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"baxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"bgcolor\":\"#E5ECF6\",\"caxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"}},\"title\":{\"x\":0.05},\"xaxis\":{\"automargin\":true,\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\",\"title\":{\"standoff\":15},\"zerolinecolor\":\"white\",\"zerolinewidth\":2},\"yaxis\":{\"automargin\":true,\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\",\"title\":{\"standoff\":15},\"zerolinecolor\":\"white\",\"zerolinewidth\":2}}}},                        {\"responsive\": true}                    ).then(function(){\n",
              "                            \n",
              "var gd = document.getElementById('fc406907-905c-4f73-8b7d-edade7da76f7');\n",
              "var x = new MutationObserver(function (mutations, observer) {{\n",
              "        var display = window.getComputedStyle(gd).display;\n",
              "        if (!display || display === 'none') {{\n",
              "            console.log([gd, 'removed!']);\n",
              "            Plotly.purge(gd);\n",
              "            observer.disconnect();\n",
              "        }}\n",
              "}});\n",
              "\n",
              "// Listen for the removal of the full notebook cells\n",
              "var notebookContainer = gd.closest('#notebook-container');\n",
              "if (notebookContainer) {{\n",
              "    x.observe(notebookContainer, {childList: true});\n",
              "}}\n",
              "\n",
              "// Listen for the clearing of the current output cell\n",
              "var outputEl = gd.closest('.output');\n",
              "if (outputEl) {{\n",
              "    x.observe(outputEl, {childList: true});\n",
              "}}\n",
              "\n",
              "                        })                };                            </script>        </div>\n",
              "</body>\n",
              "</html>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Visualize parameter importances.\n",
        "optuna.visualization.plot_param_importances(xgboost_study)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Y4rrW6JhS6gf"
      },
      "source": [
        "## Catboost optimization"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 53,
      "metadata": {
        "id": "beG8dT5dlFDA"
      },
      "outputs": [],
      "source": [
        "from catboost import Pool"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 63,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DcTo6x6Uy0O5",
        "outputId": "1e737040-eaef-4c0d-d847-df7a1272edfa"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\u001b[32m[I 2022-04-21 21:23:55,925]\u001b[0m A new study created in memory with name: CatBoost optimization\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 21:24:31,652]\u001b[0m Trial 0 finished with value: 0.2011284955822706 and parameters: {'iterations': 170, 'learning_rate': 0.03, 'depth': 4, 'l2_leaf_reg': 2, 'border_count': 213}. Best is trial 0 with value: 0.2011284955822706.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 21:25:05,213]\u001b[0m Trial 1 finished with value: 0.23205987725946833 and parameters: {'iterations': 170, 'learning_rate': 0.8400000000000001, 'depth': 7, 'l2_leaf_reg': 1, 'border_count': 22}. Best is trial 0 with value: 0.2011284955822706.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 21:25:17,082]\u001b[0m Trial 2 finished with value: 0.20905687438786952 and parameters: {'iterations': 10, 'learning_rate': 0.31, 'depth': 9, 'l2_leaf_reg': 2, 'border_count': 28}. Best is trial 0 with value: 0.2011284955822706.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 21:25:45,993]\u001b[0m Trial 3 finished with value: 0.18760980856657072 and parameters: {'iterations': 80, 'learning_rate': 0.43, 'depth': 9, 'l2_leaf_reg': 3, 'border_count': 19}. Best is trial 3 with value: 0.18760980856657072.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 21:26:06,845]\u001b[0m Trial 4 finished with value: 0.22105361250020933 and parameters: {'iterations': 90, 'learning_rate': 0.86, 'depth': 4, 'l2_leaf_reg': 3, 'border_count': 109}. Best is trial 3 with value: 0.18760980856657072.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 21:27:14,553]\u001b[0m Trial 5 finished with value: 0.20302186583516346 and parameters: {'iterations': 180, 'learning_rate': 0.51, 'depth': 8, 'l2_leaf_reg': 2, 'border_count': 134}. Best is trial 3 with value: 0.18760980856657072.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 21:27:56,113]\u001b[0m Trial 6 finished with value: 0.22429679276054895 and parameters: {'iterations': 170, 'learning_rate': 0.78, 'depth': 7, 'l2_leaf_reg': 2, 'border_count': 73}. Best is trial 3 with value: 0.18760980856657072.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 21:28:12,688]\u001b[0m Trial 7 finished with value: 0.2077911973701443 and parameters: {'iterations': 10, 'learning_rate': 0.49, 'depth': 10, 'l2_leaf_reg': 3, 'border_count': 108}. Best is trial 3 with value: 0.18760980856657072.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 21:28:24,072]\u001b[0m Trial 8 finished with value: 0.43339964428250066 and parameters: {'iterations': 10, 'learning_rate': 0.04, 'depth': 4, 'l2_leaf_reg': 2, 'border_count': 119}. Best is trial 3 with value: 0.18760980856657072.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 21:29:43,127]\u001b[0m Trial 9 finished with value: 0.22402196239813174 and parameters: {'iterations': 180, 'learning_rate': 0.91, 'depth': 10, 'l2_leaf_reg': 2, 'border_count': 24}. Best is trial 3 with value: 0.18760980856657072.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 21:30:13,214]\u001b[0m Trial 10 finished with value: 0.1832027623679508 and parameters: {'iterations': 80, 'learning_rate': 0.32, 'depth': 6, 'l2_leaf_reg': 4, 'border_count': 249}. Best is trial 10 with value: 0.1832027623679508.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 21:30:43,362]\u001b[0m Trial 11 finished with value: 0.1656859419312326 and parameters: {'iterations': 80, 'learning_rate': 0.27, 'depth': 6, 'l2_leaf_reg': 4, 'border_count': 248}. Best is trial 11 with value: 0.1656859419312326.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 21:31:08,743]\u001b[0m Trial 12 finished with value: 0.16790647081832402 and parameters: {'iterations': 60, 'learning_rate': 0.27, 'depth': 6, 'l2_leaf_reg': 4, 'border_count': 248}. Best is trial 11 with value: 0.1656859419312326.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 21:31:29,754]\u001b[0m Trial 13 finished with value: 0.17603582243341565 and parameters: {'iterations': 50, 'learning_rate': 0.2, 'depth': 6, 'l2_leaf_reg': 4, 'border_count': 191}. Best is trial 11 with value: 0.1656859419312326.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 21:32:11,740]\u001b[0m Trial 14 finished with value: 0.20703039267274442 and parameters: {'iterations': 130, 'learning_rate': 0.68, 'depth': 6, 'l2_leaf_reg': 4, 'border_count': 245}. Best is trial 11 with value: 0.1656859419312326.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 21:32:30,560]\u001b[0m Trial 15 finished with value: 0.18738740693913522 and parameters: {'iterations': 50, 'learning_rate': 0.17, 'depth': 5, 'l2_leaf_reg': 4, 'border_count': 178}. Best is trial 11 with value: 0.1656859419312326.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 21:33:03,675]\u001b[0m Trial 16 finished with value: 0.19790252546066647 and parameters: {'iterations': 120, 'learning_rate': 0.62, 'depth': 5, 'l2_leaf_reg': 4, 'border_count': 227}. Best is trial 11 with value: 0.1656859419312326.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 21:33:22,370]\u001b[0m Trial 17 finished with value: 0.183132828831548 and parameters: {'iterations': 50, 'learning_rate': 0.19, 'depth': 5, 'l2_leaf_reg': 3, 'border_count': 174}. Best is trial 11 with value: 0.1656859419312326.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 21:34:24,113]\u001b[0m Trial 18 finished with value: 0.1948045334186567 and parameters: {'iterations': 130, 'learning_rate': 0.31, 'depth': 8, 'l2_leaf_reg': 1, 'border_count': 205}. Best is trial 11 with value: 0.1656859419312326.\u001b[0m\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Number of finished trials:  19\n",
            "Best trial:\n",
            "  Value: 0.1656859419312326\n",
            "  Params: \n",
            "    iterations: 80\n",
            "    learning_rate: 0.27\n",
            "    depth: 6\n",
            "    l2_leaf_reg: 4\n",
            "    border_count: 248\n"
          ]
        }
      ],
      "source": [
        "def objective(trial):\n",
        "\n",
        "    # Define model parameter scopes\n",
        "    catboost_param = {\n",
        "        \"iterations\": trial.suggest_int(\"iterations\", 10, 200, step=10),\n",
        "        \"learning_rate\": trial.suggest_float(\"learning_rate\", 0.01, 1.0, step=0.01),\n",
        "        \"depth\": trial.suggest_int(\"depth\", 4, 10, step=1),\n",
        "        \"l2_leaf_reg\": trial.suggest_int(\"l2_leaf_reg\", 1, 4, step=1),\n",
        "        \"border_count\": trial.suggest_int(\"border_count\", 1,255 , step=1)\n",
        "    }\n",
        "\n",
        "    # Define model\n",
        "    model = CatBoostClassifier(**catboost_param,\n",
        "                               objective='Logloss',\n",
        "                               loss_function='Logloss',\n",
        "                               task_type='GPU',\n",
        "                               random_seed=RANDOM_SEED,\n",
        "                               silent=True\n",
        "                               )\n",
        "    \n",
        "    # Evaluate log_loss for each label\n",
        "    aggr_log_loss = {}\n",
        "\n",
        "    for col in val_labels.columns:\n",
        "\n",
        "        train_y_col = train_labels[col]\n",
        "        val_y_col = val_labels[col]\n",
        "\n",
        "        pool = Pool(data=train_features.values, label=train_y_col)  \n",
        "        model.fit(pool)\n",
        "\n",
        "        pred_labels = model.predict_proba(val_features.values)\n",
        "        aggr_log_loss[col] = log_loss(y_true=val_y_col.values, y_pred=pred_labels, labels=[0,1])\n",
        "\n",
        "    # Calculate aggregated log_loss\n",
        "    avg_log_loss = np.mean(list(aggr_log_loss.values()))\n",
        "\n",
        "    return avg_log_loss\n",
        "\n",
        "\n",
        "catboost_study = optuna.create_study(direction=\"minimize\", study_name='CatBoost optimization')\n",
        "catboost_study.optimize(objective, n_trials=100, timeout=600)\n",
        "\n",
        "print(\"Number of finished trials: \", len(catboost_study.trials))\n",
        "print(\"Best trial:\")\n",
        "catboost_trial = catboost_study.best_trial\n",
        "\n",
        "print(\"  Value: {}\".format(catboost_trial.value))\n",
        "print(\"  Params: \")\n",
        "for key, value in catboost_trial.params.items():\n",
        "\n",
        "    print(\"    {}: {}\".format(key, value))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 64,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 542
        },
        "id": "3WpdV3N2A6kj",
        "outputId": "b6600565-442c-4c07-ae63-f1a1d6b6b7d7"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<html>\n",
              "<head><meta charset=\"utf-8\" /></head>\n",
              "<body>\n",
              "    <div>            <script src=\"https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/MathJax.js?config=TeX-AMS-MML_SVG\"></script><script type=\"text/javascript\">if (window.MathJax) {MathJax.Hub.Config({SVG: {font: \"STIX-Web\"}});}</script>                <script type=\"text/javascript\">window.PlotlyConfig = {MathJaxConfig: 'local'};</script>\n",
              "        <script src=\"https://cdn.plot.ly/plotly-2.8.3.min.js\"></script>                <div id=\"7357d0a7-21c8-449b-ba12-2118192ec0c6\" class=\"plotly-graph-div\" style=\"height:525px; width:100%;\"></div>            <script type=\"text/javascript\">                                    window.PLOTLYENV=window.PLOTLYENV || {};                                    if (document.getElementById(\"7357d0a7-21c8-449b-ba12-2118192ec0c6\")) {                    Plotly.newPlot(                        \"7357d0a7-21c8-449b-ba12-2118192ec0c6\",                        [{\"mode\":\"markers\",\"name\":\"Objective Value\",\"x\":[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18],\"y\":[0.2011284955822706,0.23205987725946833,0.20905687438786952,0.18760980856657072,0.22105361250020933,0.20302186583516346,0.22429679276054895,0.2077911973701443,0.43339964428250066,0.22402196239813174,0.1832027623679508,0.1656859419312326,0.16790647081832402,0.17603582243341565,0.20703039267274442,0.18738740693913522,0.19790252546066647,0.183132828831548,0.1948045334186567],\"type\":\"scatter\"},{\"name\":\"Best Value\",\"x\":[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18],\"y\":[0.2011284955822706,0.2011284955822706,0.2011284955822706,0.18760980856657072,0.18760980856657072,0.18760980856657072,0.18760980856657072,0.18760980856657072,0.18760980856657072,0.18760980856657072,0.1832027623679508,0.1656859419312326,0.1656859419312326,0.1656859419312326,0.1656859419312326,0.1656859419312326,0.1656859419312326,0.1656859419312326,0.1656859419312326],\"type\":\"scatter\"}],                        {\"title\":{\"text\":\"Optimization History Plot\"},\"xaxis\":{\"title\":{\"text\":\"#Trials\"}},\"yaxis\":{\"title\":{\"text\":\"Objective Value\"}},\"template\":{\"data\":{\"bar\":[{\"error_x\":{\"color\":\"#2a3f5f\"},\"error_y\":{\"color\":\"#2a3f5f\"},\"marker\":{\"line\":{\"color\":\"#E5ECF6\",\"width\":0.5},\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"bar\"}],\"barpolar\":[{\"marker\":{\"line\":{\"color\":\"#E5ECF6\",\"width\":0.5},\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"barpolar\"}],\"carpet\":[{\"aaxis\":{\"endlinecolor\":\"#2a3f5f\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"minorgridcolor\":\"white\",\"startlinecolor\":\"#2a3f5f\"},\"baxis\":{\"endlinecolor\":\"#2a3f5f\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"minorgridcolor\":\"white\",\"startlinecolor\":\"#2a3f5f\"},\"type\":\"carpet\"}],\"choropleth\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"type\":\"choropleth\"}],\"contour\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"contour\"}],\"contourcarpet\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"type\":\"contourcarpet\"}],\"heatmap\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"heatmap\"}],\"heatmapgl\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"heatmapgl\"}],\"histogram\":[{\"marker\":{\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"histogram\"}],\"histogram2d\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"histogram2d\"}],\"histogram2dcontour\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"histogram2dcontour\"}],\"mesh3d\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"type\":\"mesh3d\"}],\"parcoords\":[{\"line\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"parcoords\"}],\"pie\":[{\"automargin\":true,\"type\":\"pie\"}],\"scatter\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatter\"}],\"scatter3d\":[{\"line\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatter3d\"}],\"scattercarpet\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattercarpet\"}],\"scattergeo\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattergeo\"}],\"scattergl\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattergl\"}],\"scattermapbox\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattermapbox\"}],\"scatterpolar\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatterpolar\"}],\"scatterpolargl\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatterpolargl\"}],\"scatterternary\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatterternary\"}],\"surface\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"surface\"}],\"table\":[{\"cells\":{\"fill\":{\"color\":\"#EBF0F8\"},\"line\":{\"color\":\"white\"}},\"header\":{\"fill\":{\"color\":\"#C8D4E3\"},\"line\":{\"color\":\"white\"}},\"type\":\"table\"}]},\"layout\":{\"annotationdefaults\":{\"arrowcolor\":\"#2a3f5f\",\"arrowhead\":0,\"arrowwidth\":1},\"autotypenumbers\":\"strict\",\"coloraxis\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"colorscale\":{\"diverging\":[[0,\"#8e0152\"],[0.1,\"#c51b7d\"],[0.2,\"#de77ae\"],[0.3,\"#f1b6da\"],[0.4,\"#fde0ef\"],[0.5,\"#f7f7f7\"],[0.6,\"#e6f5d0\"],[0.7,\"#b8e186\"],[0.8,\"#7fbc41\"],[0.9,\"#4d9221\"],[1,\"#276419\"]],\"sequential\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"sequentialminus\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]},\"colorway\":[\"#636efa\",\"#EF553B\",\"#00cc96\",\"#ab63fa\",\"#FFA15A\",\"#19d3f3\",\"#FF6692\",\"#B6E880\",\"#FF97FF\",\"#FECB52\"],\"font\":{\"color\":\"#2a3f5f\"},\"geo\":{\"bgcolor\":\"white\",\"lakecolor\":\"white\",\"landcolor\":\"#E5ECF6\",\"showlakes\":true,\"showland\":true,\"subunitcolor\":\"white\"},\"hoverlabel\":{\"align\":\"left\"},\"hovermode\":\"closest\",\"mapbox\":{\"style\":\"light\"},\"paper_bgcolor\":\"white\",\"plot_bgcolor\":\"#E5ECF6\",\"polar\":{\"angularaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"bgcolor\":\"#E5ECF6\",\"radialaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"}},\"scene\":{\"xaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"gridwidth\":2,\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\"},\"yaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"gridwidth\":2,\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\"},\"zaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"gridwidth\":2,\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\"}},\"shapedefaults\":{\"line\":{\"color\":\"#2a3f5f\"}},\"ternary\":{\"aaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"baxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"bgcolor\":\"#E5ECF6\",\"caxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"}},\"title\":{\"x\":0.05},\"xaxis\":{\"automargin\":true,\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\",\"title\":{\"standoff\":15},\"zerolinecolor\":\"white\",\"zerolinewidth\":2},\"yaxis\":{\"automargin\":true,\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\",\"title\":{\"standoff\":15},\"zerolinecolor\":\"white\",\"zerolinewidth\":2}}}},                        {\"responsive\": true}                    ).then(function(){\n",
              "                            \n",
              "var gd = document.getElementById('7357d0a7-21c8-449b-ba12-2118192ec0c6');\n",
              "var x = new MutationObserver(function (mutations, observer) {{\n",
              "        var display = window.getComputedStyle(gd).display;\n",
              "        if (!display || display === 'none') {{\n",
              "            console.log([gd, 'removed!']);\n",
              "            Plotly.purge(gd);\n",
              "            observer.disconnect();\n",
              "        }}\n",
              "}});\n",
              "\n",
              "// Listen for the removal of the full notebook cells\n",
              "var notebookContainer = gd.closest('#notebook-container');\n",
              "if (notebookContainer) {{\n",
              "    x.observe(notebookContainer, {childList: true});\n",
              "}}\n",
              "\n",
              "// Listen for the clearing of the current output cell\n",
              "var outputEl = gd.closest('.output');\n",
              "if (outputEl) {{\n",
              "    x.observe(outputEl, {childList: true});\n",
              "}}\n",
              "\n",
              "                        })                };                            </script>        </div>\n",
              "</body>\n",
              "</html>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Plot_optimization_histor: shows the scores from all trials as well as the best score so far at each point.\n",
        "optuna.visualization.plot_optimization_history(catboost_study)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 56,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 542
        },
        "id": "rx4TVfUFA_2Z",
        "outputId": "e7479e20-12ec-4813-c864-7f8cd773e698"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<html>\n",
              "<head><meta charset=\"utf-8\" /></head>\n",
              "<body>\n",
              "    <div>            <script src=\"https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/MathJax.js?config=TeX-AMS-MML_SVG\"></script><script type=\"text/javascript\">if (window.MathJax) {MathJax.Hub.Config({SVG: {font: \"STIX-Web\"}});}</script>                <script type=\"text/javascript\">window.PlotlyConfig = {MathJaxConfig: 'local'};</script>\n",
              "        <script src=\"https://cdn.plot.ly/plotly-2.8.3.min.js\"></script>                <div id=\"5ea4c818-360f-4ba4-9d91-28328c94144e\" class=\"plotly-graph-div\" style=\"height:525px; width:100%;\"></div>            <script type=\"text/javascript\">                                    window.PLOTLYENV=window.PLOTLYENV || {};                                    if (document.getElementById(\"5ea4c818-360f-4ba4-9d91-28328c94144e\")) {                    Plotly.newPlot(                        \"5ea4c818-360f-4ba4-9d91-28328c94144e\",                        [{\"cliponaxis\":false,\"hovertemplate\":[\"depth (IntUniformDistribution): 0.3098591549295773<extra></extra>\",\"iterations (IntUniformDistribution): 0.3380281690140846<extra></extra>\",\"learning_rate (DiscreteUniformDistribution): 0.35211267605633817<extra></extra>\"],\"marker\":{\"color\":\"rgb(66,146,198)\"},\"orientation\":\"h\",\"text\":[\"0.3098591549295773\",\"0.3380281690140846\",\"0.35211267605633817\"],\"textposition\":\"outside\",\"texttemplate\":\"%{text:.2f}\",\"x\":[0.3098591549295773,0.3380281690140846,0.35211267605633817],\"y\":[\"depth\",\"iterations\",\"learning_rate\"],\"type\":\"bar\"}],                        {\"showlegend\":false,\"title\":{\"text\":\"Hyperparameter Importances\"},\"xaxis\":{\"title\":{\"text\":\"Importance for Objective Value\"}},\"yaxis\":{\"title\":{\"text\":\"Hyperparameter\"}},\"template\":{\"data\":{\"bar\":[{\"error_x\":{\"color\":\"#2a3f5f\"},\"error_y\":{\"color\":\"#2a3f5f\"},\"marker\":{\"line\":{\"color\":\"#E5ECF6\",\"width\":0.5},\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"bar\"}],\"barpolar\":[{\"marker\":{\"line\":{\"color\":\"#E5ECF6\",\"width\":0.5},\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"barpolar\"}],\"carpet\":[{\"aaxis\":{\"endlinecolor\":\"#2a3f5f\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"minorgridcolor\":\"white\",\"startlinecolor\":\"#2a3f5f\"},\"baxis\":{\"endlinecolor\":\"#2a3f5f\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"minorgridcolor\":\"white\",\"startlinecolor\":\"#2a3f5f\"},\"type\":\"carpet\"}],\"choropleth\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"type\":\"choropleth\"}],\"contour\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"contour\"}],\"contourcarpet\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"type\":\"contourcarpet\"}],\"heatmap\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"heatmap\"}],\"heatmapgl\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"heatmapgl\"}],\"histogram\":[{\"marker\":{\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"histogram\"}],\"histogram2d\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"histogram2d\"}],\"histogram2dcontour\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"histogram2dcontour\"}],\"mesh3d\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"type\":\"mesh3d\"}],\"parcoords\":[{\"line\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"parcoords\"}],\"pie\":[{\"automargin\":true,\"type\":\"pie\"}],\"scatter\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatter\"}],\"scatter3d\":[{\"line\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatter3d\"}],\"scattercarpet\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattercarpet\"}],\"scattergeo\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattergeo\"}],\"scattergl\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattergl\"}],\"scattermapbox\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattermapbox\"}],\"scatterpolar\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatterpolar\"}],\"scatterpolargl\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatterpolargl\"}],\"scatterternary\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatterternary\"}],\"surface\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"surface\"}],\"table\":[{\"cells\":{\"fill\":{\"color\":\"#EBF0F8\"},\"line\":{\"color\":\"white\"}},\"header\":{\"fill\":{\"color\":\"#C8D4E3\"},\"line\":{\"color\":\"white\"}},\"type\":\"table\"}]},\"layout\":{\"annotationdefaults\":{\"arrowcolor\":\"#2a3f5f\",\"arrowhead\":0,\"arrowwidth\":1},\"autotypenumbers\":\"strict\",\"coloraxis\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"colorscale\":{\"diverging\":[[0,\"#8e0152\"],[0.1,\"#c51b7d\"],[0.2,\"#de77ae\"],[0.3,\"#f1b6da\"],[0.4,\"#fde0ef\"],[0.5,\"#f7f7f7\"],[0.6,\"#e6f5d0\"],[0.7,\"#b8e186\"],[0.8,\"#7fbc41\"],[0.9,\"#4d9221\"],[1,\"#276419\"]],\"sequential\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"sequentialminus\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]},\"colorway\":[\"#636efa\",\"#EF553B\",\"#00cc96\",\"#ab63fa\",\"#FFA15A\",\"#19d3f3\",\"#FF6692\",\"#B6E880\",\"#FF97FF\",\"#FECB52\"],\"font\":{\"color\":\"#2a3f5f\"},\"geo\":{\"bgcolor\":\"white\",\"lakecolor\":\"white\",\"landcolor\":\"#E5ECF6\",\"showlakes\":true,\"showland\":true,\"subunitcolor\":\"white\"},\"hoverlabel\":{\"align\":\"left\"},\"hovermode\":\"closest\",\"mapbox\":{\"style\":\"light\"},\"paper_bgcolor\":\"white\",\"plot_bgcolor\":\"#E5ECF6\",\"polar\":{\"angularaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"bgcolor\":\"#E5ECF6\",\"radialaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"}},\"scene\":{\"xaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"gridwidth\":2,\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\"},\"yaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"gridwidth\":2,\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\"},\"zaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"gridwidth\":2,\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\"}},\"shapedefaults\":{\"line\":{\"color\":\"#2a3f5f\"}},\"ternary\":{\"aaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"baxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"bgcolor\":\"#E5ECF6\",\"caxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"}},\"title\":{\"x\":0.05},\"xaxis\":{\"automargin\":true,\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\",\"title\":{\"standoff\":15},\"zerolinecolor\":\"white\",\"zerolinewidth\":2},\"yaxis\":{\"automargin\":true,\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\",\"title\":{\"standoff\":15},\"zerolinecolor\":\"white\",\"zerolinewidth\":2}}}},                        {\"responsive\": true}                    ).then(function(){\n",
              "                            \n",
              "var gd = document.getElementById('5ea4c818-360f-4ba4-9d91-28328c94144e');\n",
              "var x = new MutationObserver(function (mutations, observer) {{\n",
              "        var display = window.getComputedStyle(gd).display;\n",
              "        if (!display || display === 'none') {{\n",
              "            console.log([gd, 'removed!']);\n",
              "            Plotly.purge(gd);\n",
              "            observer.disconnect();\n",
              "        }}\n",
              "}});\n",
              "\n",
              "// Listen for the removal of the full notebook cells\n",
              "var notebookContainer = gd.closest('#notebook-container');\n",
              "if (notebookContainer) {{\n",
              "    x.observe(notebookContainer, {childList: true});\n",
              "}}\n",
              "\n",
              "// Listen for the clearing of the current output cell\n",
              "var outputEl = gd.closest('.output');\n",
              "if (outputEl) {{\n",
              "    x.observe(outputEl, {childList: true});\n",
              "}}\n",
              "\n",
              "                        })                };                            </script>        </div>\n",
              "</body>\n",
              "</html>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Visualize parameter importances.\n",
        "optuna.visualization.plot_param_importances(catboost_study)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JtuujCc0S-8N"
      },
      "source": [
        "## LightGBM optimization"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 57,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0ThIVw28hzLr",
        "outputId": "09ec4ccd-393f-4ad4-a8bf-973591d28a4d"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\u001b[32m[I 2022-04-21 20:36:16,984]\u001b[0m A new study created in memory with name: LightGBM optimization\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:36:31,001]\u001b[0m Trial 0 finished with value: 0.32447133707751674 and parameters: {'n_estimators': 530, 'learning_rate': 0.41000000000000003, 'max_depth': 7, 'num_leaves': 6, 'max_bin': 35}. Best is trial 0 with value: 0.32447133707751674.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:36:50,984]\u001b[0m Trial 1 finished with value: 0.9213429188623252 and parameters: {'n_estimators': 600, 'learning_rate': 0.8200000000000001, 'max_depth': 5, 'num_leaves': 7, 'max_bin': 170}. Best is trial 0 with value: 0.32447133707751674.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:37:14,679]\u001b[0m Trial 2 finished with value: 0.8782220570917417 and parameters: {'n_estimators': 580, 'learning_rate': 0.59, 'max_depth': 17, 'num_leaves': 7, 'max_bin': 117}. Best is trial 0 with value: 0.32447133707751674.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:37:56,421]\u001b[0m Trial 3 finished with value: 0.3113087476668822 and parameters: {'n_estimators': 360, 'learning_rate': 0.27, 'max_depth': 14, 'num_leaves': 10, 'max_bin': 252}. Best is trial 3 with value: 0.3113087476668822.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:38:21,963]\u001b[0m Trial 4 finished with value: 0.3103635469365468 and parameters: {'n_estimators': 380, 'learning_rate': 0.4, 'max_depth': 13, 'num_leaves': 11, 'max_bin': 106}. Best is trial 4 with value: 0.3103635469365468.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:38:49,956]\u001b[0m Trial 5 finished with value: 1.6546637518238896 and parameters: {'n_estimators': 990, 'learning_rate': 0.88, 'max_depth': 8, 'num_leaves': 12, 'max_bin': 241}. Best is trial 4 with value: 0.3103635469365468.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:39:02,557]\u001b[0m Trial 6 finished with value: 0.22660635917828248 and parameters: {'n_estimators': 150, 'learning_rate': 0.2, 'max_depth': 9, 'num_leaves': 7, 'max_bin': 42}. Best is trial 6 with value: 0.22660635917828248.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:39:23,131]\u001b[0m Trial 7 finished with value: 0.30587880271168394 and parameters: {'n_estimators': 650, 'learning_rate': 0.2, 'max_depth': 8, 'num_leaves': 4, 'max_bin': 47}. Best is trial 6 with value: 0.22660635917828248.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:40:00,627]\u001b[0m Trial 8 finished with value: 0.31449688781481766 and parameters: {'n_estimators': 720, 'learning_rate': 0.33, 'max_depth': 9, 'num_leaves': 10, 'max_bin': 186}. Best is trial 6 with value: 0.22660635917828248.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:41:01,230]\u001b[0m Trial 9 finished with value: 0.22370778606755523 and parameters: {'n_estimators': 320, 'learning_rate': 0.09, 'max_depth': 18, 'num_leaves': 7, 'max_bin': 235}. Best is trial 9 with value: 0.22370778606755523.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:41:09,788]\u001b[0m Trial 10 finished with value: 0.25534910434462754 and parameters: {'n_estimators': 140, 'learning_rate': 0.05, 'max_depth': 18, 'num_leaves': 2, 'max_bin': 207}. Best is trial 9 with value: 0.22370778606755523.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:41:24,492]\u001b[0m Trial 11 finished with value: 0.26191851045117226 and parameters: {'n_estimators': 120, 'learning_rate': 0.01, 'max_depth': 12, 'num_leaves': 5, 'max_bin': 86}. Best is trial 9 with value: 0.22370778606755523.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:42:01,019]\u001b[0m Trial 12 finished with value: 0.27395997856563564 and parameters: {'n_estimators': 280, 'learning_rate': 0.14, 'max_depth': 4, 'num_leaves': 9, 'max_bin': 149}. Best is trial 9 with value: 0.22370778606755523.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:42:17,758]\u001b[0m Trial 13 finished with value: 0.30453902727329263 and parameters: {'n_estimators': 260, 'learning_rate': 0.56, 'max_depth': 15, 'num_leaves': 8, 'max_bin': 74}. Best is trial 9 with value: 0.22370778606755523.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:42:30,953]\u001b[0m Trial 14 finished with value: 0.17236905070633085 and parameters: {'n_estimators': 90, 'learning_rate': 0.15000000000000002, 'max_depth': 11, 'num_leaves': 4, 'max_bin': 203}. Best is trial 14 with value: 0.17236905070633085.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:43:01,593]\u001b[0m Trial 15 finished with value: 0.17498547112354712 and parameters: {'n_estimators': 450, 'learning_rate': 0.09999999999999999, 'max_depth': 16, 'num_leaves': 3, 'max_bin': 218}. Best is trial 14 with value: 0.17236905070633085.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:43:25,278]\u001b[0m Trial 16 finished with value: 0.2733415601174604 and parameters: {'n_estimators': 860, 'learning_rate': 0.47000000000000003, 'max_depth': 11, 'num_leaves': 2, 'max_bin': 212}. Best is trial 14 with value: 0.17236905070633085.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:43:46,702]\u001b[0m Trial 17 finished with value: 0.5707544506277301 and parameters: {'n_estimators': 480, 'learning_rate': 0.7100000000000001, 'max_depth': 15, 'num_leaves': 4, 'max_bin': 201}. Best is trial 14 with value: 0.17236905070633085.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:44:11,155]\u001b[0m Trial 18 finished with value: 0.2692996681002731 and parameters: {'n_estimators': 440, 'learning_rate': 0.28, 'max_depth': 16, 'num_leaves': 3, 'max_bin': 158}. Best is trial 14 with value: 0.17236905070633085.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:44:34,101]\u001b[0m Trial 19 finished with value: 3.2646089767671342 and parameters: {'n_estimators': 780, 'learning_rate': 0.97, 'max_depth': 11, 'num_leaves': 4, 'max_bin': 223}. Best is trial 14 with value: 0.17236905070633085.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:45:03,949]\u001b[0m Trial 20 finished with value: 0.22419798132341362 and parameters: {'n_estimators': 250, 'learning_rate': 0.18000000000000002, 'max_depth': 13, 'num_leaves': 5, 'max_bin': 182}. Best is trial 14 with value: 0.17236905070633085.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:45:28,350]\u001b[0m Trial 21 finished with value: 0.1691697655455637 and parameters: {'n_estimators': 320, 'learning_rate': 0.09, 'max_depth': 18, 'num_leaves': 3, 'max_bin': 234}. Best is trial 21 with value: 0.1691697655455637.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:45:39,172]\u001b[0m Trial 22 finished with value: 0.19418531963765612 and parameters: {'n_estimators': 90, 'learning_rate': 0.09999999999999999, 'max_depth': 16, 'num_leaves': 3, 'max_bin': 225}. Best is trial 21 with value: 0.1691697655455637.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:45:56,375]\u001b[0m Trial 23 finished with value: 0.1911536097744522 and parameters: {'n_estimators': 200, 'learning_rate': 0.28, 'max_depth': 18, 'num_leaves': 3, 'max_bin': 252}. Best is trial 21 with value: 0.1691697655455637.\u001b[0m\n",
            "\u001b[32m[I 2022-04-21 20:46:51,262]\u001b[0m Trial 24 finished with value: 0.16829793995794928 and parameters: {'n_estimators': 430, 'learning_rate': 0.02, 'max_depth': 16, 'num_leaves': 5, 'max_bin': 193}. Best is trial 24 with value: 0.16829793995794928.\u001b[0m\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Number of finished trials:  25\n",
            "Best trial:\n",
            "  Value: 0.16829793995794928\n",
            "  Params: \n",
            "    n_estimators: 430\n",
            "    learning_rate: 0.02\n",
            "    max_depth: 16\n",
            "    num_leaves: 5\n",
            "    max_bin: 193\n"
          ]
        }
      ],
      "source": [
        "def objective(trial):\n",
        "\n",
        "    # Define model parameter scopes\n",
        "    lightgbm_param = {\n",
        "        \"n_estimators\": trial.suggest_int(\"n_estimators\", 80, 1000, step=10),\n",
        "        \"learning_rate\": trial.suggest_float(\"learning_rate\", 0.01, 1.0, step=0.01),\n",
        "        \"max_depth\": trial.suggest_int(\"max_depth\", 3, 18, step=1),\n",
        "        \"num_leaves\": trial.suggest_int('num_leaves', 2, 12, step=1),\n",
        "        \"max_bin\": trial.suggest_int('max_bin', 25, 255, step=1)\n",
        "    }\n",
        "\n",
        "    # Define model\n",
        "    model = LGBMClassifier(**lightgbm_param,\n",
        "        objective='binary',\n",
        "        device_type='gpu',\n",
        "        random_seed=RANDOM_SEED,\n",
        "        verbose=-1\n",
        "    )\n",
        "\n",
        "    # Evaluate log_loss for each label\n",
        "    aggr_log_loss = {}\n",
        "\n",
        "    for col in val_labels.columns:\n",
        "\n",
        "        train_y_col = train_labels[col]\n",
        "        val_y_col = val_labels[col]\n",
        "\n",
        "        model.fit(\n",
        "            train_features.values,\n",
        "            train_y_col\n",
        "            )\n",
        "\n",
        "        pred_labels = model.predict_proba(val_features.values)\n",
        "        aggr_log_loss[col] = log_loss(y_true=val_y_col.values, y_pred=pred_labels, labels=[0,1])\n",
        "\n",
        "    # Calculate aggregated log_loss\n",
        "    avg_log_loss = np.mean(list(aggr_log_loss.values()))\n",
        "\n",
        "    return avg_log_loss\n",
        "\n",
        "\n",
        "lightgbm_study = optuna.create_study(direction=\"minimize\", study_name='LightGBM optimization')\n",
        "lightgbm_study.optimize(objective, n_trials=100, timeout=600)\n",
        "\n",
        "print(\"Number of finished trials: \", len(lightgbm_study.trials))\n",
        "print(\"Best trial:\")\n",
        "lightgbm_trial = lightgbm_study.best_trial\n",
        "\n",
        "print(\"  Value: {}\".format(lightgbm_trial.value))\n",
        "print(\"  Params: \")\n",
        "for key, value in lightgbm_trial.params.items():\n",
        "\n",
        "    print(\"    {}: {}\".format(key, value))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 58,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 542
        },
        "id": "jpNdWCSVhVrs",
        "outputId": "38e4b27b-22cd-4591-d504-b5980e1e8293"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<html>\n",
              "<head><meta charset=\"utf-8\" /></head>\n",
              "<body>\n",
              "    <div>            <script src=\"https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/MathJax.js?config=TeX-AMS-MML_SVG\"></script><script type=\"text/javascript\">if (window.MathJax) {MathJax.Hub.Config({SVG: {font: \"STIX-Web\"}});}</script>                <script type=\"text/javascript\">window.PlotlyConfig = {MathJaxConfig: 'local'};</script>\n",
              "        <script src=\"https://cdn.plot.ly/plotly-2.8.3.min.js\"></script>                <div id=\"1a5358e4-be51-48b3-8616-799dcdea7832\" class=\"plotly-graph-div\" style=\"height:525px; width:100%;\"></div>            <script type=\"text/javascript\">                                    window.PLOTLYENV=window.PLOTLYENV || {};                                    if (document.getElementById(\"1a5358e4-be51-48b3-8616-799dcdea7832\")) {                    Plotly.newPlot(                        \"1a5358e4-be51-48b3-8616-799dcdea7832\",                        [{\"mode\":\"markers\",\"name\":\"Objective Value\",\"x\":[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24],\"y\":[0.32447133707751674,0.9213429188623252,0.8782220570917417,0.3113087476668822,0.3103635469365468,1.6546637518238896,0.22660635917828248,0.30587880271168394,0.31449688781481766,0.22370778606755523,0.25534910434462754,0.26191851045117226,0.27395997856563564,0.30453902727329263,0.17236905070633085,0.17498547112354712,0.2733415601174604,0.5707544506277301,0.2692996681002731,3.2646089767671342,0.22419798132341362,0.1691697655455637,0.19418531963765612,0.1911536097744522,0.16829793995794928],\"type\":\"scatter\"},{\"name\":\"Best Value\",\"x\":[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24],\"y\":[0.32447133707751674,0.32447133707751674,0.32447133707751674,0.3113087476668822,0.3103635469365468,0.3103635469365468,0.22660635917828248,0.22660635917828248,0.22660635917828248,0.22370778606755523,0.22370778606755523,0.22370778606755523,0.22370778606755523,0.22370778606755523,0.17236905070633085,0.17236905070633085,0.17236905070633085,0.17236905070633085,0.17236905070633085,0.17236905070633085,0.17236905070633085,0.1691697655455637,0.1691697655455637,0.1691697655455637,0.16829793995794928],\"type\":\"scatter\"}],                        {\"title\":{\"text\":\"Optimization History Plot\"},\"xaxis\":{\"title\":{\"text\":\"#Trials\"}},\"yaxis\":{\"title\":{\"text\":\"Objective Value\"}},\"template\":{\"data\":{\"bar\":[{\"error_x\":{\"color\":\"#2a3f5f\"},\"error_y\":{\"color\":\"#2a3f5f\"},\"marker\":{\"line\":{\"color\":\"#E5ECF6\",\"width\":0.5},\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"bar\"}],\"barpolar\":[{\"marker\":{\"line\":{\"color\":\"#E5ECF6\",\"width\":0.5},\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"barpolar\"}],\"carpet\":[{\"aaxis\":{\"endlinecolor\":\"#2a3f5f\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"minorgridcolor\":\"white\",\"startlinecolor\":\"#2a3f5f\"},\"baxis\":{\"endlinecolor\":\"#2a3f5f\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"minorgridcolor\":\"white\",\"startlinecolor\":\"#2a3f5f\"},\"type\":\"carpet\"}],\"choropleth\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"type\":\"choropleth\"}],\"contour\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"contour\"}],\"contourcarpet\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"type\":\"contourcarpet\"}],\"heatmap\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"heatmap\"}],\"heatmapgl\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"heatmapgl\"}],\"histogram\":[{\"marker\":{\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"histogram\"}],\"histogram2d\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"histogram2d\"}],\"histogram2dcontour\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"histogram2dcontour\"}],\"mesh3d\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"type\":\"mesh3d\"}],\"parcoords\":[{\"line\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"parcoords\"}],\"pie\":[{\"automargin\":true,\"type\":\"pie\"}],\"scatter\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatter\"}],\"scatter3d\":[{\"line\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatter3d\"}],\"scattercarpet\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattercarpet\"}],\"scattergeo\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattergeo\"}],\"scattergl\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattergl\"}],\"scattermapbox\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattermapbox\"}],\"scatterpolar\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatterpolar\"}],\"scatterpolargl\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatterpolargl\"}],\"scatterternary\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatterternary\"}],\"surface\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"surface\"}],\"table\":[{\"cells\":{\"fill\":{\"color\":\"#EBF0F8\"},\"line\":{\"color\":\"white\"}},\"header\":{\"fill\":{\"color\":\"#C8D4E3\"},\"line\":{\"color\":\"white\"}},\"type\":\"table\"}]},\"layout\":{\"annotationdefaults\":{\"arrowcolor\":\"#2a3f5f\",\"arrowhead\":0,\"arrowwidth\":1},\"autotypenumbers\":\"strict\",\"coloraxis\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"colorscale\":{\"diverging\":[[0,\"#8e0152\"],[0.1,\"#c51b7d\"],[0.2,\"#de77ae\"],[0.3,\"#f1b6da\"],[0.4,\"#fde0ef\"],[0.5,\"#f7f7f7\"],[0.6,\"#e6f5d0\"],[0.7,\"#b8e186\"],[0.8,\"#7fbc41\"],[0.9,\"#4d9221\"],[1,\"#276419\"]],\"sequential\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"sequentialminus\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]},\"colorway\":[\"#636efa\",\"#EF553B\",\"#00cc96\",\"#ab63fa\",\"#FFA15A\",\"#19d3f3\",\"#FF6692\",\"#B6E880\",\"#FF97FF\",\"#FECB52\"],\"font\":{\"color\":\"#2a3f5f\"},\"geo\":{\"bgcolor\":\"white\",\"lakecolor\":\"white\",\"landcolor\":\"#E5ECF6\",\"showlakes\":true,\"showland\":true,\"subunitcolor\":\"white\"},\"hoverlabel\":{\"align\":\"left\"},\"hovermode\":\"closest\",\"mapbox\":{\"style\":\"light\"},\"paper_bgcolor\":\"white\",\"plot_bgcolor\":\"#E5ECF6\",\"polar\":{\"angularaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"bgcolor\":\"#E5ECF6\",\"radialaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"}},\"scene\":{\"xaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"gridwidth\":2,\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\"},\"yaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"gridwidth\":2,\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\"},\"zaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"gridwidth\":2,\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\"}},\"shapedefaults\":{\"line\":{\"color\":\"#2a3f5f\"}},\"ternary\":{\"aaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"baxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"bgcolor\":\"#E5ECF6\",\"caxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"}},\"title\":{\"x\":0.05},\"xaxis\":{\"automargin\":true,\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\",\"title\":{\"standoff\":15},\"zerolinecolor\":\"white\",\"zerolinewidth\":2},\"yaxis\":{\"automargin\":true,\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\",\"title\":{\"standoff\":15},\"zerolinecolor\":\"white\",\"zerolinewidth\":2}}}},                        {\"responsive\": true}                    ).then(function(){\n",
              "                            \n",
              "var gd = document.getElementById('1a5358e4-be51-48b3-8616-799dcdea7832');\n",
              "var x = new MutationObserver(function (mutations, observer) {{\n",
              "        var display = window.getComputedStyle(gd).display;\n",
              "        if (!display || display === 'none') {{\n",
              "            console.log([gd, 'removed!']);\n",
              "            Plotly.purge(gd);\n",
              "            observer.disconnect();\n",
              "        }}\n",
              "}});\n",
              "\n",
              "// Listen for the removal of the full notebook cells\n",
              "var notebookContainer = gd.closest('#notebook-container');\n",
              "if (notebookContainer) {{\n",
              "    x.observe(notebookContainer, {childList: true});\n",
              "}}\n",
              "\n",
              "// Listen for the clearing of the current output cell\n",
              "var outputEl = gd.closest('.output');\n",
              "if (outputEl) {{\n",
              "    x.observe(outputEl, {childList: true});\n",
              "}}\n",
              "\n",
              "                        })                };                            </script>        </div>\n",
              "</body>\n",
              "</html>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Plot_optimization_histor: shows the scores from all trials as well as the best score so far at each point.\n",
        "optuna.visualization.plot_optimization_history(lightgbm_study)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 59,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 542
        },
        "id": "h7PfmtLphaQd",
        "outputId": "26001bfc-a649-4d0a-a3f1-3fc33ec69631"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<html>\n",
              "<head><meta charset=\"utf-8\" /></head>\n",
              "<body>\n",
              "    <div>            <script src=\"https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/MathJax.js?config=TeX-AMS-MML_SVG\"></script><script type=\"text/javascript\">if (window.MathJax) {MathJax.Hub.Config({SVG: {font: \"STIX-Web\"}});}</script>                <script type=\"text/javascript\">window.PlotlyConfig = {MathJaxConfig: 'local'};</script>\n",
              "        <script src=\"https://cdn.plot.ly/plotly-2.8.3.min.js\"></script>                <div id=\"0cd9e4ee-5929-491c-ac46-86d521b8cc15\" class=\"plotly-graph-div\" style=\"height:525px; width:100%;\"></div>            <script type=\"text/javascript\">                                    window.PLOTLYENV=window.PLOTLYENV || {};                                    if (document.getElementById(\"0cd9e4ee-5929-491c-ac46-86d521b8cc15\")) {                    Plotly.newPlot(                        \"0cd9e4ee-5929-491c-ac46-86d521b8cc15\",                        [{\"cliponaxis\":false,\"hovertemplate\":[\"max_depth (IntUniformDistribution): 0.0025269937329817304<extra></extra>\",\"num_leaves (IntUniformDistribution): 0.0150375080663586<extra></extra>\",\"max_bin (IntUniformDistribution): 0.023830487752902835<extra></extra>\",\"n_estimators (IntUniformDistribution): 0.08255551522031257<extra></extra>\",\"learning_rate (DiscreteUniformDistribution): 0.8760494952274442<extra></extra>\"],\"marker\":{\"color\":\"rgb(66,146,198)\"},\"orientation\":\"h\",\"text\":[\"0.0025269937329817304\",\"0.0150375080663586\",\"0.023830487752902835\",\"0.08255551522031257\",\"0.8760494952274442\"],\"textposition\":\"outside\",\"texttemplate\":\"%{text:.2f}\",\"x\":[0.0025269937329817304,0.0150375080663586,0.023830487752902835,0.08255551522031257,0.8760494952274442],\"y\":[\"max_depth\",\"num_leaves\",\"max_bin\",\"n_estimators\",\"learning_rate\"],\"type\":\"bar\"}],                        {\"showlegend\":false,\"title\":{\"text\":\"Hyperparameter Importances\"},\"xaxis\":{\"title\":{\"text\":\"Importance for Objective Value\"}},\"yaxis\":{\"title\":{\"text\":\"Hyperparameter\"}},\"template\":{\"data\":{\"bar\":[{\"error_x\":{\"color\":\"#2a3f5f\"},\"error_y\":{\"color\":\"#2a3f5f\"},\"marker\":{\"line\":{\"color\":\"#E5ECF6\",\"width\":0.5},\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"bar\"}],\"barpolar\":[{\"marker\":{\"line\":{\"color\":\"#E5ECF6\",\"width\":0.5},\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"barpolar\"}],\"carpet\":[{\"aaxis\":{\"endlinecolor\":\"#2a3f5f\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"minorgridcolor\":\"white\",\"startlinecolor\":\"#2a3f5f\"},\"baxis\":{\"endlinecolor\":\"#2a3f5f\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"minorgridcolor\":\"white\",\"startlinecolor\":\"#2a3f5f\"},\"type\":\"carpet\"}],\"choropleth\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"type\":\"choropleth\"}],\"contour\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"contour\"}],\"contourcarpet\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"type\":\"contourcarpet\"}],\"heatmap\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"heatmap\"}],\"heatmapgl\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"heatmapgl\"}],\"histogram\":[{\"marker\":{\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"histogram\"}],\"histogram2d\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"histogram2d\"}],\"histogram2dcontour\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"histogram2dcontour\"}],\"mesh3d\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"type\":\"mesh3d\"}],\"parcoords\":[{\"line\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"parcoords\"}],\"pie\":[{\"automargin\":true,\"type\":\"pie\"}],\"scatter\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatter\"}],\"scatter3d\":[{\"line\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatter3d\"}],\"scattercarpet\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattercarpet\"}],\"scattergeo\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattergeo\"}],\"scattergl\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattergl\"}],\"scattermapbox\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattermapbox\"}],\"scatterpolar\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatterpolar\"}],\"scatterpolargl\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatterpolargl\"}],\"scatterternary\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatterternary\"}],\"surface\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"surface\"}],\"table\":[{\"cells\":{\"fill\":{\"color\":\"#EBF0F8\"},\"line\":{\"color\":\"white\"}},\"header\":{\"fill\":{\"color\":\"#C8D4E3\"},\"line\":{\"color\":\"white\"}},\"type\":\"table\"}]},\"layout\":{\"annotationdefaults\":{\"arrowcolor\":\"#2a3f5f\",\"arrowhead\":0,\"arrowwidth\":1},\"autotypenumbers\":\"strict\",\"coloraxis\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"colorscale\":{\"diverging\":[[0,\"#8e0152\"],[0.1,\"#c51b7d\"],[0.2,\"#de77ae\"],[0.3,\"#f1b6da\"],[0.4,\"#fde0ef\"],[0.5,\"#f7f7f7\"],[0.6,\"#e6f5d0\"],[0.7,\"#b8e186\"],[0.8,\"#7fbc41\"],[0.9,\"#4d9221\"],[1,\"#276419\"]],\"sequential\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"sequentialminus\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]},\"colorway\":[\"#636efa\",\"#EF553B\",\"#00cc96\",\"#ab63fa\",\"#FFA15A\",\"#19d3f3\",\"#FF6692\",\"#B6E880\",\"#FF97FF\",\"#FECB52\"],\"font\":{\"color\":\"#2a3f5f\"},\"geo\":{\"bgcolor\":\"white\",\"lakecolor\":\"white\",\"landcolor\":\"#E5ECF6\",\"showlakes\":true,\"showland\":true,\"subunitcolor\":\"white\"},\"hoverlabel\":{\"align\":\"left\"},\"hovermode\":\"closest\",\"mapbox\":{\"style\":\"light\"},\"paper_bgcolor\":\"white\",\"plot_bgcolor\":\"#E5ECF6\",\"polar\":{\"angularaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"bgcolor\":\"#E5ECF6\",\"radialaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"}},\"scene\":{\"xaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"gridwidth\":2,\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\"},\"yaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"gridwidth\":2,\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\"},\"zaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"gridwidth\":2,\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\"}},\"shapedefaults\":{\"line\":{\"color\":\"#2a3f5f\"}},\"ternary\":{\"aaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"baxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"bgcolor\":\"#E5ECF6\",\"caxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"}},\"title\":{\"x\":0.05},\"xaxis\":{\"automargin\":true,\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\",\"title\":{\"standoff\":15},\"zerolinecolor\":\"white\",\"zerolinewidth\":2},\"yaxis\":{\"automargin\":true,\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\",\"title\":{\"standoff\":15},\"zerolinecolor\":\"white\",\"zerolinewidth\":2}}}},                        {\"responsive\": true}                    ).then(function(){\n",
              "                            \n",
              "var gd = document.getElementById('0cd9e4ee-5929-491c-ac46-86d521b8cc15');\n",
              "var x = new MutationObserver(function (mutations, observer) {{\n",
              "        var display = window.getComputedStyle(gd).display;\n",
              "        if (!display || display === 'none') {{\n",
              "            console.log([gd, 'removed!']);\n",
              "            Plotly.purge(gd);\n",
              "            observer.disconnect();\n",
              "        }}\n",
              "}});\n",
              "\n",
              "// Listen for the removal of the full notebook cells\n",
              "var notebookContainer = gd.closest('#notebook-container');\n",
              "if (notebookContainer) {{\n",
              "    x.observe(notebookContainer, {childList: true});\n",
              "}}\n",
              "\n",
              "// Listen for the clearing of the current output cell\n",
              "var outputEl = gd.closest('.output');\n",
              "if (outputEl) {{\n",
              "    x.observe(outputEl, {childList: true});\n",
              "}}\n",
              "\n",
              "                        })                };                            </script>        </div>\n",
              "</body>\n",
              "</html>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Visualize parameter importances.\n",
        "optuna.visualization.plot_param_importances(lightgbm_study)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hf_P9PORSizW"
      },
      "source": [
        "# 8. Compare models after optimization"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 75,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 362
        },
        "id": "ptaVGP6zDY8p",
        "outputId": "5c17ef2a-a706-47d0-e052-fd99395a3829"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Group before and after optimization model performances\n",
        "model_groups = [xgboost_avg_logloss, xgboost_trial.value, # xgboost\n",
        "                catboost_avg_logloss, catboost_trial.value, #catboost\n",
        "                lightgbm_avg_logloss, lightgbm_trial.value # lightgbm\n",
        "                ]\n",
        "\n",
        "# Corresponding labels\n",
        "group_labels = ['xgboost baseline', 'xgboost optimized', 'catboost baseline', 'catboost optimized', 'lightgbm baseline', 'lightgbm optimized']\n",
        "\n",
        "# Set positions for each bar on x-axis (grouping)\n",
        "x_pos = [0,1,4,5,8,9]\n",
        "\n",
        "# Create bars with fitting colors\n",
        "plt.bar(x_pos, model_groups, color=['blue', 'blue', 'green', 'green', 'black', 'black'])\n",
        " \n",
        "# Create names on the x-axis\n",
        "plt.xticks(x_pos, group_labels)\n",
        "plt.xticks(rotation=45, ha=\"right\")\n",
        "\n",
        "# Labeling\n",
        "plt.title(\"Model performances\")\n",
        "plt.xlabel('model_name')\n",
        "plt.ylabel('log_loss')\n",
        " \n",
        "# Show graphic\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gcTYIxebTtiK"
      },
      "source": [
        "# 9. Training the models on all of the data"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 76,
      "metadata": {
        "id": "eUkkIvfiEFc9"
      },
      "outputs": [],
      "source": [
        "# Merge both preprocessed feature dataframes\n",
        "all_features = pd.concat([train_features, val_features])\n",
        "\n",
        "# Merge both label dataframes\n",
        "all_labels = pd.concat([train_labels, val_labels])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 77,
      "metadata": {
        "id": "qR8S8WmATvZl"
      },
      "outputs": [],
      "source": [
        "# Initialize dict to hold fitted models\n",
        "def create_fitted_model_dict(model_name, model_params, labels, features):\n",
        "    fitted_logreg_dict = {}\n",
        "\n",
        "    # Split into binary classifier for each class\n",
        "    for col in labels.columns:\n",
        "        y_train_col = labels[col]  # Train on one class at a time\n",
        "\n",
        "        if model_name == 'xgboost':\n",
        "            model = XGBClassifier(**model_params)\n",
        "        elif model_name == 'catboost':\n",
        "            model = CatBoostClassifier(**model_params)\n",
        "        elif model_name == 'lightgbm':\n",
        "            model = LGBMClassifier(**model_params)\n",
        "        else:\n",
        "            raise Exception('Unknown model name!')\n",
        "        \n",
        "        fitted_logreg_dict[col] = model.fit(features.values, y_train_col)  # Train\n",
        "\n",
        "    return fitted_logreg_dict"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 78,
      "metadata": {
        "id": "0VSVr7PcVFNh"
      },
      "outputs": [],
      "source": [
        "xgboost_model_params = {\n",
        "    'tree_method':'gpu_hist',\n",
        "    'objective':'binary:logistic',\n",
        "    'gpu_id':0,\n",
        "    'seed':RANDOM_SEED,\n",
        "    }\n",
        "    \n",
        "# Add params acquired by optuna optimization\n",
        "xgboost_model_params.update(xgboost_trial.params)\n",
        "\n",
        "fitted_xgboost_dict = create_fitted_model_dict('xgboost', xgboost_model_params, all_labels, all_features)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 79,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PiwzBBOHVcjz",
        "outputId": "d6b0bce5-c32c-4672-d2f4-b6d5b1b6d859"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "Warning: less than 75% gpu memory available for training. Free: 10075.75 Total: 16280.875\n",
            "Warning: less than 75% gpu memory available for training. Free: 10075.75 Total: 16280.875\n",
            "Warning: less than 75% gpu memory available for training. Free: 10075.75 Total: 16280.875\n",
            "Warning: less than 75% gpu memory available for training. Free: 10075.75 Total: 16280.875\n",
            "Warning: less than 75% gpu memory available for training. Free: 10075.75 Total: 16280.875\n",
            "Warning: less than 75% gpu memory available for training. Free: 10075.75 Total: 16280.875\n",
            "Warning: less than 75% gpu memory available for training. Free: 10075.75 Total: 16280.875\n",
            "Warning: less than 75% gpu memory available for training. Free: 10075.75 Total: 16280.875\n",
            "Warning: less than 75% gpu memory available for training. Free: 10075.75 Total: 16280.875\n",
            "Warning: less than 75% gpu memory available for training. Free: 10075.75 Total: 16280.875\n"
          ]
        }
      ],
      "source": [
        "catboost_model_params = {\n",
        "    'task_type':'GPU',\n",
        "    'random_seed':RANDOM_SEED,\n",
        "    'silent':True,\n",
        "    }\n",
        "\n",
        "# Add params acquired by optuna optimization\n",
        "catboost_model_params.update(catboost_trial.params)\n",
        "\n",
        "fitted_catboost_dict = create_fitted_model_dict('catboost', catboost_model_params, all_labels, all_features)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 80,
      "metadata": {
        "id": "n-h0cQR4EyWJ"
      },
      "outputs": [],
      "source": [
        "lightgbm_model_params = {\n",
        "    'objective':'binary',\n",
        "    'device_type':'gpu',\n",
        "    'random_seed':RANDOM_SEED,\n",
        "    'verbose':-1\n",
        "    }\n",
        "\n",
        "# Add params acquired by optuna optimization\n",
        "lightgbm_model_params.update(lightgbm_trial.params)\n",
        "\n",
        "fitted_lightgbm_dict = create_fitted_model_dict('lightgbm', lightgbm_model_params, all_labels, all_features)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0Ct9mSGDEByu"
      },
      "source": [
        "# 10. Preparing a submission"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 81,
      "metadata": {
        "id": "VmptQFkZEIqS"
      },
      "outputs": [],
      "source": [
        "def predict_for_sample(sample_id, fitted_model_dict, files, compounds_order):\n",
        "\n",
        "    # Import sample\n",
        "    temp_sample = pd.read_csv(DATA_PATH / files[sample_id])\n",
        "\n",
        "    # Preprocess sample\n",
        "    temp_sample = preprocess_sample(temp_sample)\n",
        "\n",
        "    # Feature engineering on sample\n",
        "    temp_sample = abun_per_tempbin(temp_sample)\n",
        "\n",
        "    # Generate predictions for each class\n",
        "    temp_sample_preds_dict = {}\n",
        "\n",
        "    for compound in compounds_order:\n",
        "        clf = fitted_model_dict[compound]\n",
        "        temp_sample_preds_dict[compound] = clf.predict_proba(temp_sample.values)[:, 1][0]\n",
        "\n",
        "    return temp_sample_preds_dict"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 82,
      "metadata": {
        "id": "89TFVgaBEGPV"
      },
      "outputs": [],
      "source": [
        "# Import submission format\n",
        "submission_template_df = pd.read_csv(\n",
        "    DATA_PATH / \"submission_format.csv\", index_col=\"sample_id\"\n",
        ")\n",
        "compounds_order = submission_template_df.columns\n",
        "sample_order = submission_template_df.index\n",
        "\n",
        "# Create dict with both validation and test sample IDs and paths\n",
        "all_test_files = test_files.copy()\n",
        "all_test_files.update(val_files)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hKHw-AXgF52Q"
      },
      "source": [
        "## Inference on each model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 83,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 49,
          "referenced_widgets": [
            "f69ee1ff7c8a4ccf9a4b828b06cddefc",
            "52b2c31c39024479b61891f810c63953",
            "bace33a2e1e94e46b147561e51440899",
            "096a8a7e3e894fdd93a545ed60201e3e",
            "83e7a7f74a644d0a81a0e4ae61c95475",
            "4fc5af050f2245e0a0a57a4f0c6bbd7f",
            "0afae99fafcc49c8a318aeccd453570d",
            "ca15d12cde804976a76f2013468740a8",
            "0835a02905014bd0b1bf4975ef85ecae",
            "59d7e5e6471a426eaa0b7a8824127e0e",
            "4c4eeb3d09a549e7a83ff09d39925614"
          ]
        },
        "id": "ZfFg5yGMVMa7",
        "outputId": "5f3f2b62-8c95-4cc8-ad44-28f805476b07"
      },
      "outputs": [
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "f69ee1ff7c8a4ccf9a4b828b06cddefc",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "  0%|          | 0/804 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# XGBoost predictions\n",
        "xgboost_submission_df = pd.DataFrame(\n",
        "    [\n",
        "        predict_for_sample(sample_id, fitted_xgboost_dict, all_test_files, compounds_order)\n",
        "        for sample_id in tqdm(sample_order)\n",
        "    ],\n",
        "    index=sample_order,\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 84,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 49,
          "referenced_widgets": [
            "e896b60e93b84045b9f968b599ea2b9c",
            "25e12f9c0d4b4fa7bc0582b157c0534f",
            "85877d0b57654cc997bf3b0de157ce2c",
            "33472ae7e48e42e1a94db93c497ffe97",
            "c8458d926eb8432fa4cdd3a1566d7518",
            "f38a7f893f6e4db68348a7a9334bd872",
            "93455f6549a4402fb440713750932fef",
            "a470af39f60843dab4e57d2953db899d",
            "ce617f64add64ade913747a77f264f6a",
            "fd7620b77b9d4227a21fa5d47d327e4d",
            "10597454bd3a4c58860fdcaa4fb618e7"
          ]
        },
        "id": "BS9TiR0WEKAx",
        "outputId": "4aac7a08-3bf1-4d32-9c28-bdbd7b5fe8db"
      },
      "outputs": [
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "e896b60e93b84045b9f968b599ea2b9c",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "  0%|          | 0/804 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# CatBoost predictions\n",
        "catboost_submission_df = pd.DataFrame(\n",
        "    [\n",
        "        predict_for_sample(sample_id, fitted_catboost_dict, all_test_files, compounds_order)\n",
        "        for sample_id in tqdm(sample_order)\n",
        "    ],\n",
        "    index=sample_order,\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 85,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 49,
          "referenced_widgets": [
            "8e5a3e9fd287438da3af7ed8f47a4269",
            "6c44298cb38f44ad91e356af47225b88",
            "e005b353ed6b4845ab7644998f7f6809",
            "77bef106f21246498041b61ba72b8a14",
            "d7bbc5578d2049c5a967f40f847e0b5d",
            "e431bca0354d4c0b948d17628e09c95d",
            "cc70f9aca42f4f81b8abf9a8bc15d164",
            "1bf3e1cf6bc342c5bfc9c45ee8fe96cd",
            "abd14469c5d44750a0366a6ee64f2193",
            "670ea2fc34274f74a83e28e1015609b3",
            "a3d57991c59843dcbbfcdeabde832b62"
          ]
        },
        "id": "efr1wlYo4xqd",
        "outputId": "4e6a8b87-95a5-4fdf-822f-0542931805ce"
      },
      "outputs": [
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "8e5a3e9fd287438da3af7ed8f47a4269",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "  0%|          | 0/804 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# LightGBM predictions\n",
        "lightgbm_submission_df = pd.DataFrame(\n",
        "    [\n",
        "        predict_for_sample(sample_id, fitted_lightgbm_dict, all_test_files, compounds_order)\n",
        "        for sample_id in tqdm(sample_order)\n",
        "    ],\n",
        "    index=sample_order,\n",
        ")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "aktH_v7IF_12"
      },
      "source": [
        "## Stack predictions & take average"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 86,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 238
        },
        "id": "eTBk9Q-MVTNw",
        "outputId": "bc6019e5-a6c8-4102-fbd9-d82ac5527eec"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-ce65c2f1-ce8c-4a7a-90c8-f81db68e5b0b\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>basalt</th>\n",
              "      <th>carbonate</th>\n",
              "      <th>chloride</th>\n",
              "      <th>iron_oxide</th>\n",
              "      <th>oxalate</th>\n",
              "      <th>oxychlorine</th>\n",
              "      <th>phyllosilicate</th>\n",
              "      <th>silicate</th>\n",
              "      <th>sulfate</th>\n",
              "      <th>sulfide</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>sample_id</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>S0766</th>\n",
              "      <td>0.002441</td>\n",
              "      <td>0.001332</td>\n",
              "      <td>0.010178</td>\n",
              "      <td>0.010396</td>\n",
              "      <td>0.000211</td>\n",
              "      <td>0.007367</td>\n",
              "      <td>0.015185</td>\n",
              "      <td>0.103990</td>\n",
              "      <td>0.002720</td>\n",
              "      <td>0.000571</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>S0767</th>\n",
              "      <td>0.021792</td>\n",
              "      <td>0.017664</td>\n",
              "      <td>0.001783</td>\n",
              "      <td>0.022095</td>\n",
              "      <td>0.000134</td>\n",
              "      <td>0.005550</td>\n",
              "      <td>0.032132</td>\n",
              "      <td>0.075339</td>\n",
              "      <td>0.011882</td>\n",
              "      <td>0.000371</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>S0768</th>\n",
              "      <td>0.890998</td>\n",
              "      <td>0.072644</td>\n",
              "      <td>0.002076</td>\n",
              "      <td>0.047650</td>\n",
              "      <td>0.000311</td>\n",
              "      <td>0.018337</td>\n",
              "      <td>0.722315</td>\n",
              "      <td>0.895352</td>\n",
              "      <td>0.055063</td>\n",
              "      <td>0.000242</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>S0769</th>\n",
              "      <td>0.008833</td>\n",
              "      <td>0.001750</td>\n",
              "      <td>0.047314</td>\n",
              "      <td>0.037409</td>\n",
              "      <td>0.000156</td>\n",
              "      <td>0.945314</td>\n",
              "      <td>0.016984</td>\n",
              "      <td>0.003225</td>\n",
              "      <td>0.912814</td>\n",
              "      <td>0.000631</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>S0770</th>\n",
              "      <td>0.003607</td>\n",
              "      <td>0.001922</td>\n",
              "      <td>0.008824</td>\n",
              "      <td>0.745228</td>\n",
              "      <td>0.000339</td>\n",
              "      <td>0.978265</td>\n",
              "      <td>0.938980</td>\n",
              "      <td>0.001836</td>\n",
              "      <td>0.003916</td>\n",
              "      <td>0.000135</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-ce65c2f1-ce8c-4a7a-90c8-f81db68e5b0b')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-ce65c2f1-ce8c-4a7a-90c8-f81db68e5b0b button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-ce65c2f1-ce8c-4a7a-90c8-f81db68e5b0b');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "             basalt  carbonate  chloride  iron_oxide   oxalate  oxychlorine  \\\n",
              "sample_id                                                                     \n",
              "S0766      0.002441   0.001332  0.010178    0.010396  0.000211     0.007367   \n",
              "S0767      0.021792   0.017664  0.001783    0.022095  0.000134     0.005550   \n",
              "S0768      0.890998   0.072644  0.002076    0.047650  0.000311     0.018337   \n",
              "S0769      0.008833   0.001750  0.047314    0.037409  0.000156     0.945314   \n",
              "S0770      0.003607   0.001922  0.008824    0.745228  0.000339     0.978265   \n",
              "\n",
              "           phyllosilicate  silicate   sulfate   sulfide  \n",
              "sample_id                                                \n",
              "S0766            0.015185  0.103990  0.002720  0.000571  \n",
              "S0767            0.032132  0.075339  0.011882  0.000371  \n",
              "S0768            0.722315  0.895352  0.055063  0.000242  \n",
              "S0769            0.016984  0.003225  0.912814  0.000631  \n",
              "S0770            0.938980  0.001836  0.003916  0.000135  "
            ]
          },
          "execution_count": 86,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Stacked prediction and take average of all 3 model preditictions\n",
        "final_submission_df = (xgboost_submission_df+catboost_submission_df+lightgbm_submission_df)/3\n",
        "final_submission_df.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 87,
      "metadata": {
        "id": "acxvTJz-01Xw"
      },
      "outputs": [],
      "source": [
        "final_submission_df.to_csv('submission_.csv')"
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "collapsed_sections": [
        "pxUpv5xxLM26",
        "y4NhhHisLP_L",
        "02KAwxWdSNUW",
        "3CKqjQr6SdRt",
        "cxvUeMBKSkjY",
        "4CmfDC2YSvi4",
        "XST12ElGS1iI",
        "L5D9FdwgTL5_",
        "0nOAwuXgTaz8",
        "hOWm0SLdTlQj",
        "6WZcINZYQbx2",
        "utjES9mPS0Ge"
      ],
      "name": "mars_spectrometry.ipynb",
      "provenance": [],
      "toc_visible": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    },
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "0835a02905014bd0b1bf4975ef85ecae": {
          "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": ""
          }
        },
        "096a8a7e3e894fdd93a545ed60201e3e": {
          "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_59d7e5e6471a426eaa0b7a8824127e0e",
            "placeholder": "​",
            "style": "IPY_MODEL_4c4eeb3d09a549e7a83ff09d39925614",
            "value": " 804/804 [04:40&lt;00:00,  1.21it/s]"
          }
        },
        "0afae99fafcc49c8a318aeccd453570d": {
          "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": ""
          }
        },
        "10597454bd3a4c58860fdcaa4fb618e7": {
          "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": ""
          }
        },
        "1bf3e1cf6bc342c5bfc9c45ee8fe96cd": {
          "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
          }
        },
        "22d2d0a3ee044435964c21ec184c31b8": {
          "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": ""
          }
        },
        "25e12f9c0d4b4fa7bc0582b157c0534f": {
          "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_f38a7f893f6e4db68348a7a9334bd872",
            "placeholder": "​",
            "style": "IPY_MODEL_93455f6549a4402fb440713750932fef",
            "value": "100%"
          }
        },
        "269a9b6202b5475793236b270473e29b": {
          "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
          }
        },
        "2c59011ecc00415594670ea73666b29f": {
          "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
          }
        },
        "33472ae7e48e42e1a94db93c497ffe97": {
          "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_fd7620b77b9d4227a21fa5d47d327e4d",
            "placeholder": "​",
            "style": "IPY_MODEL_10597454bd3a4c58860fdcaa4fb618e7",
            "value": " 804/804 [04:58&lt;00:00,  1.20it/s]"
          }
        },
        "398efc4a630f47cc9c27707d177d7cc0": {
          "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_269a9b6202b5475793236b270473e29b",
            "placeholder": "​",
            "style": "IPY_MODEL_3e9e7c013ef54979ac103b138f84726c",
            "value": " 766/766 [03:12&lt;00:00,  1.50it/s]"
          }
        },
        "3e9e7c013ef54979ac103b138f84726c": {
          "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": ""
          }
        },
        "4c4eeb3d09a549e7a83ff09d39925614": {
          "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": ""
          }
        },
        "4cc88f64551a45c2a2aa3c1efc7137b6": {
          "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_2c59011ecc00415594670ea73666b29f",
            "placeholder": "​",
            "style": "IPY_MODEL_a494aa9596dc4d49964656d1e5a0b2c6",
            "value": "100%"
          }
        },
        "4fc5af050f2245e0a0a57a4f0c6bbd7f": {
          "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
          }
        },
        "50af3fc5de1646f2b94022d9dc48c83b": {
          "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
          }
        },
        "521d30efe29e4049be8dd0c6c12a0d2f": {
          "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
          }
        },
        "52b2c31c39024479b61891f810c63953": {
          "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_4fc5af050f2245e0a0a57a4f0c6bbd7f",
            "placeholder": "​",
            "style": "IPY_MODEL_0afae99fafcc49c8a318aeccd453570d",
            "value": "100%"
          }
        },
        "59d7e5e6471a426eaa0b7a8824127e0e": {
          "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
          }
        },
        "670ea2fc34274f74a83e28e1015609b3": {
          "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
          }
        },
        "6c44298cb38f44ad91e356af47225b88": {
          "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_e431bca0354d4c0b948d17628e09c95d",
            "placeholder": "​",
            "style": "IPY_MODEL_cc70f9aca42f4f81b8abf9a8bc15d164",
            "value": "100%"
          }
        },
        "6cb51ddcdfb44e1f8e562d97020ec548": {
          "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": ""
          }
        },
        "6cec1c4d6b8948d98c9c9bd9a52fa5e4": {
          "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_50af3fc5de1646f2b94022d9dc48c83b",
            "max": 293,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_92481ce80e7d43d7923e9bb238ee12cf",
            "value": 293
          }
        },
        "74c08b5283674dc7a251cd2236e8cf79": {
          "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": ""
          }
        },
        "77bef106f21246498041b61ba72b8a14": {
          "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_670ea2fc34274f74a83e28e1015609b3",
            "placeholder": "​",
            "style": "IPY_MODEL_a3d57991c59843dcbbfcdeabde832b62",
            "value": " 804/804 [03:44&lt;00:00,  1.29it/s]"
          }
        },
        "793435bbd40a41b4a7ae799ef72ca996": {
          "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
          }
        },
        "83e7a7f74a644d0a81a0e4ae61c95475": {
          "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
          }
        },
        "85877d0b57654cc997bf3b0de157ce2c": {
          "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_a470af39f60843dab4e57d2953db899d",
            "max": 804,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_ce617f64add64ade913747a77f264f6a",
            "value": 804
          }
        },
        "87c0eb619b614d6aa14a10bdbef59e34": {
          "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
          }
        },
        "8e5a3e9fd287438da3af7ed8f47a4269": {
          "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_6c44298cb38f44ad91e356af47225b88",
              "IPY_MODEL_e005b353ed6b4845ab7644998f7f6809",
              "IPY_MODEL_77bef106f21246498041b61ba72b8a14"
            ],
            "layout": "IPY_MODEL_d7bbc5578d2049c5a967f40f847e0b5d"
          }
        },
        "92481ce80e7d43d7923e9bb238ee12cf": {
          "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": ""
          }
        },
        "93455f6549a4402fb440713750932fef": {
          "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": ""
          }
        },
        "a3d57991c59843dcbbfcdeabde832b62": {
          "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": ""
          }
        },
        "a41dc73f5caf4f59a8c491368b6ebf9f": {
          "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_af90bbbbd0004d0f904f8a9daec10a48",
            "placeholder": "​",
            "style": "IPY_MODEL_74c08b5283674dc7a251cd2236e8cf79",
            "value": " 293/293 [01:10&lt;00:00,  3.93it/s]"
          }
        },
        "a470af39f60843dab4e57d2953db899d": {
          "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
          }
        },
        "a494aa9596dc4d49964656d1e5a0b2c6": {
          "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": ""
          }
        },
        "abd14469c5d44750a0366a6ee64f2193": {
          "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": ""
          }
        },
        "af90bbbbd0004d0f904f8a9daec10a48": {
          "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
          }
        },
        "b4613e7f5da44afab9ddddef6dcacf51": {
          "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_521d30efe29e4049be8dd0c6c12a0d2f",
            "max": 766,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_6cb51ddcdfb44e1f8e562d97020ec548",
            "value": 766
          }
        },
        "b48c0f8e99274ee391f10f7b6f82ef65": {
          "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
          }
        },
        "bace33a2e1e94e46b147561e51440899": {
          "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_ca15d12cde804976a76f2013468740a8",
            "max": 804,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_0835a02905014bd0b1bf4975ef85ecae",
            "value": 804
          }
        },
        "c8458d926eb8432fa4cdd3a1566d7518": {
          "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
          }
        },
        "ca15d12cde804976a76f2013468740a8": {
          "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
          }
        },
        "cc70f9aca42f4f81b8abf9a8bc15d164": {
          "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": ""
          }
        },
        "ce617f64add64ade913747a77f264f6a": {
          "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": ""
          }
        },
        "d7bbc5578d2049c5a967f40f847e0b5d": {
          "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
          }
        },
        "d895b8b2fd694dd5988f31c7f9430fec": {
          "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_4cc88f64551a45c2a2aa3c1efc7137b6",
              "IPY_MODEL_6cec1c4d6b8948d98c9c9bd9a52fa5e4",
              "IPY_MODEL_a41dc73f5caf4f59a8c491368b6ebf9f"
            ],
            "layout": "IPY_MODEL_793435bbd40a41b4a7ae799ef72ca996"
          }
        },
        "dc7505a241264fa2be96c28436207db0": {
          "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_f11b0cb6112b4dbe95d708986bf97a68",
              "IPY_MODEL_b4613e7f5da44afab9ddddef6dcacf51",
              "IPY_MODEL_398efc4a630f47cc9c27707d177d7cc0"
            ],
            "layout": "IPY_MODEL_87c0eb619b614d6aa14a10bdbef59e34"
          }
        },
        "e005b353ed6b4845ab7644998f7f6809": {
          "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_1bf3e1cf6bc342c5bfc9c45ee8fe96cd",
            "max": 804,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_abd14469c5d44750a0366a6ee64f2193",
            "value": 804
          }
        },
        "e431bca0354d4c0b948d17628e09c95d": {
          "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
          }
        },
        "e896b60e93b84045b9f968b599ea2b9c": {
          "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_25e12f9c0d4b4fa7bc0582b157c0534f",
              "IPY_MODEL_85877d0b57654cc997bf3b0de157ce2c",
              "IPY_MODEL_33472ae7e48e42e1a94db93c497ffe97"
            ],
            "layout": "IPY_MODEL_c8458d926eb8432fa4cdd3a1566d7518"
          }
        },
        "f11b0cb6112b4dbe95d708986bf97a68": {
          "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_b48c0f8e99274ee391f10f7b6f82ef65",
            "placeholder": "​",
            "style": "IPY_MODEL_22d2d0a3ee044435964c21ec184c31b8",
            "value": "100%"
          }
        },
        "f38a7f893f6e4db68348a7a9334bd872": {
          "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
          }
        },
        "f69ee1ff7c8a4ccf9a4b828b06cddefc": {
          "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_52b2c31c39024479b61891f810c63953",
              "IPY_MODEL_bace33a2e1e94e46b147561e51440899",
              "IPY_MODEL_096a8a7e3e894fdd93a545ed60201e3e"
            ],
            "layout": "IPY_MODEL_83e7a7f74a644d0a81a0e4ae61c95475"
          }
        },
        "fd7620b77b9d4227a21fa5d47d327e4d": {
          "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
          }
        }
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}