LGBM_Regression_Model.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "machine_shape": "hm"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "gpuClass": "standard"
  },
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 118,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TbP1SYI7lt5O",
        "outputId": "8271817f-a943-4980-b589-e4496d2dece8"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Requirement already satisfied: ipython-autotime in /usr/local/lib/python3.7/dist-packages (0.3.1)\n",
            "Requirement already satisfied: ipython in /usr/local/lib/python3.7/dist-packages (from ipython-autotime) (7.9.0)\n",
            "Requirement already satisfied: setuptools>=18.5 in /usr/local/lib/python3.7/dist-packages (from ipython->ipython-autotime) (57.4.0)\n",
            "Requirement already satisfied: jedi>=0.10 in /usr/local/lib/python3.7/dist-packages (from ipython->ipython-autotime) (0.18.2)\n",
            "Requirement already satisfied: pexpect in /usr/local/lib/python3.7/dist-packages (from ipython->ipython-autotime) (4.8.0)\n",
            "Requirement already satisfied: pygments in /usr/local/lib/python3.7/dist-packages (from ipython->ipython-autotime) (2.6.1)\n",
            "Requirement already satisfied: pickleshare in /usr/local/lib/python3.7/dist-packages (from ipython->ipython-autotime) (0.7.5)\n",
            "Requirement already satisfied: traitlets>=4.2 in /usr/local/lib/python3.7/dist-packages (from ipython->ipython-autotime) (5.1.1)\n",
            "Requirement already satisfied: backcall in /usr/local/lib/python3.7/dist-packages (from ipython->ipython-autotime) (0.2.0)\n",
            "Requirement already satisfied: prompt-toolkit<2.1.0,>=2.0.0 in /usr/local/lib/python3.7/dist-packages (from ipython->ipython-autotime) (2.0.10)\n",
            "Requirement already satisfied: decorator in /usr/local/lib/python3.7/dist-packages (from ipython->ipython-autotime) (4.4.2)\n",
            "Requirement already satisfied: parso<0.9.0,>=0.8.0 in /usr/local/lib/python3.7/dist-packages (from jedi>=0.10->ipython->ipython-autotime) (0.8.3)\n",
            "Requirement already satisfied: six>=1.9.0 in /usr/local/lib/python3.7/dist-packages (from prompt-toolkit<2.1.0,>=2.0.0->ipython->ipython-autotime) (1.15.0)\n",
            "Requirement already satisfied: wcwidth in /usr/local/lib/python3.7/dist-packages (from prompt-toolkit<2.1.0,>=2.0.0->ipython->ipython-autotime) (0.2.5)\n",
            "Requirement already satisfied: ptyprocess>=0.5 in /usr/local/lib/python3.7/dist-packages (from pexpect->ipython->ipython-autotime) (0.7.0)\n",
            "The autotime extension is already loaded. To reload it, use:\n",
            "  %reload_ext autotime\n",
            "Mounted at /content/drive\n",
            "time: 8.41 s (started: 2022-11-23 04:44:09 +00:00)\n"
          ]
        }
      ],
      "source": [
        "!pip install ipython-autotime\n",
        "%load_ext autotime\n",
        "import pandas as pd\n",
        "from google.colab import drive\n",
        "import pathlib\n",
        "import numpy as np\n",
        "import gc\n",
        "import matplotlib.pyplot as plt\n",
        "from xgboost import XGBRegressor\n",
        "import pyarrow.parquet as pq\n",
        "import lightgbm as lgb\n",
        "from sklearn.model_selection import RandomizedSearchCV\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import mean_squared_error\n",
        "from sklearn.metrics import mean_absolute_error\n",
        "from sklearn.preprocessing import MinMaxScaler\n",
        "\n",
        "\n",
        "drive.mount('/content/drive',force_remount=True)"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "random_state=17"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1HoVPF1Pl19I",
        "outputId": "e30753a9-caaf-4eb1-94dd-24ac9219a22d"
      },
      "execution_count": 92,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 691 µs (started: 2022-11-23 03:13:14 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "taxiData = pd.DataFrame()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "iMG7Aehrl3vk",
        "outputId": "7c30fde9-fbc8-4109-f843-8b7182a3b03f"
      },
      "execution_count": 93,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 1.47 ms (started: 2022-11-23 03:13:14 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "data_dir = \"/content/drive/MyDrive/yellow_tripdata_2020_data.parquet\"\n",
        "data_dir = pathlib.Path(data_dir)\n",
        "yellowTaxiData2020 = pq.ParquetFile(data_dir)\n",
        "del data_dir"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Wzd85aEEl6LO",
        "outputId": "d3c44b4d-0ac5-457b-c5d6-3a79d3b0a887"
      },
      "execution_count": 94,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 68.5 ms (started: 2022-11-23 03:13:14 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "for taxiDataBatch in yellowTaxiData2020.iter_batches(batch_size=2000000):\n",
        "    taxiDataBatchDF = taxiDataBatch.to_pandas()\n",
        "    taxiDataBatchDF.drop(columns=['tpep_pickup_datetime', 'tpep_dropoff_datetime', 'pickup_date', 'pickup_time', 'dropoff_date', 'dropoff_time'], inplace=True)\n",
        "    taxiData = pd.concat([taxiData, taxiDataBatchDF])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "WyicXH5frT0L",
        "outputId": "409cfced-94a2-4cbb-ffc4-cc4e6e2c9f44"
      },
      "execution_count": 95,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 7.02 s (started: 2022-11-23 03:13:17 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "taxiData.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "60B0tMrml6s0",
        "outputId": "aab3cf9e-8b09-4989-e9e4-d94accc3a459"
      },
      "execution_count": 96,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(4069192, 18)"
            ]
          },
          "metadata": {},
          "execution_count": 96
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 31.2 ms (started: 2022-11-23 03:13:25 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "del yellowTaxiData2020"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "O_KbmYEgl9dW",
        "outputId": "7f837bf8-d6a1-4498-b208-0607b31830a8"
      },
      "execution_count": 97,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 704 µs (started: 2022-11-23 03:13:28 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "gc.collect()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "jclHmuszl_AY",
        "outputId": "d327c6f3-40f5-4bfb-d439-af2fea8d7c35"
      },
      "execution_count": 98,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "172"
            ]
          },
          "metadata": {},
          "execution_count": 98
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 739 ms (started: 2022-11-23 03:13:30 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "data_dir = \"/content/drive/MyDrive/yellow_tripdata_2021_data.parquet\"\n",
        "data_dir = pathlib.Path(data_dir)\n",
        "yellowTaxiData2021 = pq.ParquetFile(data_dir)\n",
        "del data_dir"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9S0PTF_ppBAk",
        "outputId": "31d5cef6-7dd6-49f2-d942-8d57bbe67cc2"
      },
      "execution_count": 99,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 8.96 ms (started: 2022-11-23 03:13:31 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "gc.collect()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fCJFC5jGpIFz",
        "outputId": "7df3fc70-e81e-4def-ce0d-c9140209ddbe"
      },
      "execution_count": 100,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "88"
            ]
          },
          "metadata": {},
          "execution_count": 100
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 199 ms (started: 2022-11-23 03:13:34 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "for taxiDataBatch in yellowTaxiData2021.iter_batches(batch_size=2000000):\n",
        "    taxiDataBatchDF = taxiDataBatch.to_pandas()\n",
        "    taxiDataBatchDF.drop(columns=['tpep_pickup_datetime', 'tpep_dropoff_datetime', 'pickup_date', 'pickup_time', 'dropoff_date', 'dropoff_time'], inplace=True)\n",
        "    taxiData = pd.concat([taxiData, taxiDataBatchDF])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Fc4O6wHppUXy",
        "outputId": "9581e0c4-a9b0-4f21-b471-c0f8abbf154c"
      },
      "execution_count": 101,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 30.4 s (started: 2022-11-23 03:13:37 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "scaler = MinMaxScaler()\n",
        "taxiData = pd.DataFrame(scaler.fit_transform(taxiData.values), columns=taxiData.columns, index=taxiData.index)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TPErxJ91eWSo",
        "outputId": "ab6c6429-c3ee-46e4-b76c-050d4598a84f"
      },
      "execution_count": 102,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 17.3 s (started: 2022-11-23 03:14:13 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X_train, X_test, y_train, y_test = train_test_split(taxiData.loc[:, (taxiData.columns!='ride_duration')], \n",
        "                                                    taxiData.loc[:, 'ride_duration'], \n",
        "                                                    test_size=0.2, \n",
        "                                                    random_state=random_state)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9vk_Bwj-pZY_",
        "outputId": "cddb1d08-2737-45ff-fb17-c587b2fccead"
      },
      "execution_count": 104,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 21.3 s (started: 2022-11-23 03:14:56 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X_train, X_val, y_train, y_val = train_test_split(taxiData.loc[:, (taxiData.columns!='ride_duration')], \n",
        "                                                    taxiData.loc[:, 'ride_duration'], \n",
        "                                                    test_size=0.2, \n",
        "                                                    random_state=random_state)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "W-LxkWKmpcsz",
        "outputId": "611b8480-66f4-44ff-a7ba-f9b17a3da859"
      },
      "execution_count": 105,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 23.8 s (started: 2022-11-23 03:15:17 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "del yellowTaxiData2021\n",
        "del taxiData\n",
        "gc.collect()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GiAS8tAspjuE",
        "outputId": "0be41245-3ee8-499c-b12d-0ea280c84207"
      },
      "execution_count": 106,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "153"
            ]
          },
          "metadata": {},
          "execution_count": 106
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 95.9 ms (started: 2022-11-23 03:16:02 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "lgbmRegressor = lgb.LGBMRegressor()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5ofnN__F5fzQ",
        "outputId": "b480c6d8-fb75-42bd-a808-01b7609947d8"
      },
      "execution_count": 107,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 875 µs (started: 2022-11-23 03:16:07 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "lgbmParamsForRandomCV = {'n_estimators': [5, 6, 7, 8, 9, 10], 'max_features': ['auto', 'sqrt'], 'max_depth': [10, 15, 20, 25, 30, 35, 40], 'min_samples_split': [2, 5, 10], 'min_samples_leaf': [1, 2, 4]}\n",
        "\n",
        "randomSearchCV = RandomizedSearchCV(estimator = lgbmRegressor, param_distributions = lgbmParamsForRandomCV, n_iter = 10, verbose=2, random_state=random_state, n_jobs = -1)\n",
        "randomSearchCV.fit(X_train[:5000000], y_train[:5000000])\n",
        "print(randomSearchCV.best_estimator_)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8MUyKE7vqN3l",
        "outputId": "be3344e2-5b5c-41eb-c6ed-0d00f92229d3"
      },
      "execution_count": 108,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Fitting 5 folds for each of 10 candidates, totalling 50 fits\n",
            "LGBMRegressor(max_depth=40, max_features='sqrt', min_samples_leaf=2,\n",
            "              min_samples_split=5, n_estimators=10)\n",
            "time: 4min 7s (started: 2022-11-23 03:16:09 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "lgbmRegressor = lgb.LGBMRegressor(max_depth=40, max_features='sqrt', min_samples_leaf=2,\n",
        "              min_samples_split=5, n_estimators=10, objective='regression', feature_fraction=0.9, bagging_fraction=0.5, max_bin=1000, learning_rate=0.1, random_state=random_state)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-DQXmocA9tr4",
        "outputId": "b68ec26c-f96e-421f-d67a-7cec1a77da26"
      },
      "execution_count": 109,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 1.1 ms (started: 2022-11-23 03:20:27 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "lgbmRegressor.fit(X_train, y_train, eval_set=[(X_val, y_val)], eval_metric='l1',callbacks=[lgb.early_stopping(20)])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SgynYnt1A3sL",
        "outputId": "bea51e16-7711-4421-a9d2-2b1dd75c377d"
      },
      "execution_count": 110,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[1]\tvalid_0's l1: 0.163832\tvalid_0's l2: 0.040337\n",
            "Training until validation scores don't improve for 20 rounds.\n",
            "[2]\tvalid_0's l1: 0.153269\tvalid_0's l2: 0.035597\n",
            "[3]\tvalid_0's l1: 0.144177\tvalid_0's l2: 0.031751\n",
            "[4]\tvalid_0's l1: 0.136353\tvalid_0's l2: 0.0286089\n",
            "[5]\tvalid_0's l1: 0.129572\tvalid_0's l2: 0.026024\n",
            "[6]\tvalid_0's l1: 0.123772\tvalid_0's l2: 0.0239192\n",
            "[7]\tvalid_0's l1: 0.11879\tvalid_0's l2: 0.0221901\n",
            "[8]\tvalid_0's l1: 0.114496\tvalid_0's l2: 0.020764\n",
            "[9]\tvalid_0's l1: 0.11382\tvalid_0's l2: 0.0205308\n",
            "[10]\tvalid_0's l1: 0.113275\tvalid_0's l2: 0.0203421\n",
            "Did not meet early stopping. Best iteration is:\n",
            "[10]\tvalid_0's l1: 0.113275\tvalid_0's l2: 0.0203421\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "LGBMRegressor(bagging_fraction=0.5, feature_fraction=0.9, max_bin=1000,\n",
              "              max_depth=40, max_features='sqrt', min_samples_leaf=2,\n",
              "              min_samples_split=5, n_estimators=10, objective='regression',\n",
              "              random_state=17)"
            ]
          },
          "metadata": {},
          "execution_count": 110
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 37.1 s (started: 2022-11-23 03:20:43 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "predictions_train = lgbmRegressor.predict(X_train, num_iteration=lgbmRegressor.best_iteration_)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DOOQNArRpeOb",
        "outputId": "29bc928f-dbac-42b5-d375-594747854e6d"
      },
      "execution_count": 116,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 18.5 s (started: 2022-11-23 04:02:04 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "predictions = lgbmRegressor.predict(X_test, num_iteration=lgbmRegressor.best_iteration_)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UjCLMs-HDF1L",
        "outputId": "e70413ef-1cda-4ae1-b474-6fdc80607c87"
      },
      "execution_count": 111,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 2.28 s (started: 2022-11-23 03:21:58 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "rmse_train = mean_squared_error(y_train, predictions_train)**0.5\n",
        "print(f'The RMSE of training prediction is: {rmse_train}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pCJ8PnqspnoU",
        "outputId": "a0ff96ab-5b67-4153-8b62-ab4c2fbeb141"
      },
      "execution_count": 117,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The RMSE of training prediction is: 0.14264482481396076\n",
            "time: 282 ms (started: 2022-11-23 04:02:55 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "mse_train = mean_squared_error(y_train, predictions_train)\n",
        "print(f'The MSE of training prediction is: {mse_train}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HN73x_piysoU",
        "outputId": "809a8ca9-861e-4637-bd0f-04dd72fd38de"
      },
      "execution_count": 120,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The MSE of training prediction is: 0.020347546046205554\n",
            "time: 139 ms (started: 2022-11-23 04:46:19 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "mae_train = mean_absolute_error(y_train, predictions_train)\n",
        "print(f'The MAE of training prediction is: {mae_train}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XXNQ9uHMzVpb",
        "outputId": "488bf363-e160-49a8-c37c-cdb020c19e25"
      },
      "execution_count": 121,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The MAE of training prediction is: 0.11329334110638502\n",
            "time: 394 ms (started: 2022-11-23 04:46:29 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "rmse_test = mean_squared_error(y_test, predictions)**0.5\n",
        "print(f'The RMSE of prediction is: {rmse_test}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DqVqmfE7DUgx",
        "outputId": "4a3a5611-8c3a-485b-dcc8-c2079314efe6"
      },
      "execution_count": 112,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The RMSE of prediction is: 0.14262587545327904\n",
            "time: 40.5 ms (started: 2022-11-23 03:22:05 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "mse_test = mean_squared_error(y_test, predictions)\n",
        "print(f'The MSE of prediction is: {mse_test}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "N1JdF82ty13x",
        "outputId": "08b78c59-864e-466c-e82b-72dc3ea19f0b"
      },
      "execution_count": 122,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The MSE of prediction is: 0.02034214034881426\n",
            "time: 143 ms (started: 2022-11-23 04:46:37 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "mae_test = mean_absolute_error(y_test, predictions)\n",
        "print(f'The MAE of training prediction is: {mae_test}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vnS3FTb1zRBs",
        "outputId": "bdad23cc-3161-48fa-b95e-f1aefb9bd798"
      },
      "execution_count": 123,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The MAE of training prediction is: 0.11327522834307302\n",
            "time: 61.1 ms (started: 2022-11-23 04:46:48 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "lgb.plot_importance(lgbmRegressor, figsize=(10, 9))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 603
        },
        "id": "zgmA1gCqLO2x",
        "outputId": "7161df11-d1dc-4ea5-ec24-6b991e1579d1"
      },
      "execution_count": 113,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f1c7b9dec10>"
            ]
          },
          "metadata": {},
          "execution_count": 113
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 720x648 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAArAAAAImCAYAAACre7noAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzde5hVZd3/8fdXPJF4yAf0V5khkqICkZhmoQ36oB2wtB8dPDwKooZlPpVoPHkiO2lqeXrSMEvylKWlJpVaMklmKYpIWVi/oEhNFNMcQpjB7++PvUa34xw2bGbGNbxf17Uv1l73Wuv+rnu68jP3vteeyEwkSZKkstigtwuQJEmS1oQBVpIkSaVigJUkSVKpGGAlSZJUKgZYSZIklYoBVpIkSaVigJUkdSgiPhcR3+rtOiSpWvg9sJLUPSJiMbAtsLpq906Z+Vid1zwmM39eX3XlExHTgaGZeURv1yKpdzkDK0nd66DMHFD1Wuvwui5ExIa92f/aKmvdkrqHAVaSelhEbBkRV0TE4xHxaER8MSL6FW07RsSdEbEsIp6KiGsiYqui7Spge+DHEdEUEadERENE/L3N9RdHxH8W29Mj4oaIuDoi/gVM7Kz/dmqdHhFXF9uDIyIjYlJELImIf0bElIh4W0Q8FBHPRMQlVedOjIi7I+KSiHg2Iv4YEftXtb8+Im6JiKcj4s8RcWybfqvrngJ8DvhIce/zi+MmRcQfIuK5iPhLRHys6hoNEfH3iDgpIpYW9zupqr1/RJwfEX8t6vtVRPQv2t4eEb8u7ml+RDSs1Q9bUrcwwEpSz7sSaAGGAm8FDgCOKdoC+ArwemAX4I3AdIDM/C/gb7w0q/vVGvv7AHADsBVwTRf912Iv4M3AR4ALgFOB/wR2Az4cEe9qc+z/AwYCZwI/jIiti7bvAX8v7nUC8OWI2K+Duq8AvgxcX9z7W4pjlgLjgS2AScDXI2L3qmv8H2BL4A3AZOB/I+K1Rdt5wGjgHcDWwCnACxHxBmAW8MVi/1TgxogYtAZjJKkbGWAlqXvdVMziPRMRN0XEtsB7gU9l5vLMXAp8HfgoQGb+OTPvyMyVmfkk8DXgXR1fvib3ZOZNmfkClaDXYf81+kJmPp+ZtwPLgesyc2lmPgrMoRKKWy0FLsjM5sy8HlgIvC8i3gi8E/hsca0HgW8BR7ZXd2auaK+QzJyVmf8vK34J3A7sU3VIM3BW0f9PgCZg54jYADga+O/MfDQzV2fmrzNzJXAE8JPM/EnR9x3A3GLcJL0KuKZIkrrXwdUPXEXEnsBGwOMR0bp7A2BJ0b4tcCGVELZ50fbPOmtYUrX9ps76r9ETVdsr2nk/oOr9o/nyp4X/SmXG9fXA05n5XJu2PTqou10R8R4qM7s7UbmP1wALqg5ZlpktVe//XdQ3ENiUyuxwW28CPhQRB1Xt2wiY3VU9knqGAVaSetYSYCUwsE2wavVlIIERmfl0RBwMXFLV3varY5ZTCW0AFGtZ237UXX1OV/2va2+IiKgKsdsDtwCPAVtHxOZVIXZ74NGqc9ve68veR8QmwI1UZm1vzszmiLiJyjKMrjwFPA/sCMxv07YEuCozj33FWZJeFVxCIEk9KDMfp/Ix9/kRsUVEbFA8uNW6TGBzKh9zP1usxTy5zSWeAIZUvX8E2DQi3hcRGwGnAZvU0f+6tg1wYkRsFBEforKu9yeZuQT4NfCViNg0IkZSWaN6dSfXegIYXHz8D7AxlXt9EmgpZmMPqKWoYjnFt4GvFQ+T9YuIvYtQfDVwUEQcWOzftHggbLs1v31J3cEAK0k970gq4ethKssDbgBeV7R9HtgdeJbKg0Q/bHPuV4DTijW1UzPzWeDjVNaPPkplRvbvdK6z/te131J54Osp4EvAhMxcVrQdCgymMhv7I+DMLr7f9gfFv8si4oFi5vZE4PtU7uMwKrO7tZpKZbnBfcDTwDnABkW4/gCVbz14ksqM7Mn430zpVcM/ZCBJ6hYRMZHKH10Y09u1SOpb/G1SkiRJpWKAlSRJUqm4hECSJEml4gysJEmSSsUAK0mSpFLxDxmsR7baaqscOnRob5dRasuXL2ezzTbr7TJKy/Grn2NYP8ewPo5f/RzD2tx///1PZWbbP8wCGGDXK9tuuy1z587t7TJKrbGxkYaGht4uo7Qcv/o5hvVzDOvj+NXPMaxNRPy1ozaXEEiSJKlUDLCSJEkqFQOsJEmSSsUAK0mSpFIxwEqSJKlUDLCSJEkqFQOsJEmSSsUAK0mSpFIxwEqSJKlUDLCSJEkqFQOsJEmSSsUAK0mSpFIxwEqSJKlUDLCSJEkqFQOsJEmSSsUAK0mSpFIxwEqSJKlUDLCSJEkqFQOsJEmSSsUAK0mSpFIxwEqSJKlUDLCSJEkqFQOsJEmSSsUAK0mSpFIxwEqSJKlUDLCSJEkqFQOsJEmSSsUAK0mSpFIxwEqSJK2HnnnmGSZMmMCwYcPYZZdduOeeezj99NMZOXIko0aN4oADDuCxxx7r7TLbFZnZ2zWoh2w/ZGhu8OELe7uMUjtpRAvnL9iwt8soLcevfo5h/RzD+jh+9evuMVx89vtqOu6oo45in3324ZhjjmHVqlX8+9//ZoMNNmCLLbYA4KKLLuLhhx/msssu67ZaOxMR92fmHu219bkZ2IjYKiI+3kn7r9dBHxMj4pJie0pEHNnJsQ0R8Y56+5QkSVpXnn32We666y4mT54MwMYbb8xWW231YngFWL58ORHRWyV2qi/+CrUV8HHgG9U7I2LDzGzJzHUaJjOzq19LGoAmoO7gLEmStC4sWrSIQYMGMWnSJObPn8/o0aO58MIL2WyzzTj11FP57ne/y5Zbbsns2bN7u9R29bkZWOBsYMeIeDAi7ouIORFxC/AwQEQ0Ff82RMRdETErIhZGxGUR0eF4RMSkiHgkIu4F3lm1f3pETC22T4yIhyPioYj4XkQMBqYAny7q2SciDoqI30bEvIj4eURsW3Wdb0dEY0T8JSJOrOrjyOKa8yPiqmLfoIi4sbjH+yLinUiSJNWgpaWFBx54gOOPP5558+ax2WabcfbZZwPwpS99iSVLlnD44YdzySWX9HKl7euLM7DTgOGZOSoiGoBZxftF7Ry7J7Ar8FfgZ8AHgRvaHhQRrwM+D4wGngVmA/M66HuHzFwZEVtl5jMRcRnQlJnnFdd6LfD2zMyIOAY4BTipOH8YMBbYHFgYEZcCOwGnAe/IzKciYuvi2AuBr2fmryJie+A2YJd2aj8OOA5g4MBBnDGipcOBU9e27V9Zu6S14/jVzzGsn2NYH8evft09ho2NjV0e8/TTTzNw4EBWrFhBY2MjO+64I9deey3777//i8cMGTKEadOmMXbs2G6rdW31xQDb1r0dhNfWtr8ARMR1wBjaCbDAXkBjZj5ZHHs9lWDZ1kPANRFxE3BTB31uB1xfhOKNgeraZmXmSmBlRCwFtgX2A36QmU8BZObTxbH/CexatTZli4gYkJlN1Z1l5gxgBlQe4nLhfX18eKE+jl/9HMP6OYb1cfzq1+0PcR3eUNNxX//613nd617HzjvvTGNjI/vssw9veMMbePOb3wzAxRdfzOjRo2loqO16PWl9+F/g8k7a2n4FQ71fyfA+YF/gIODUiBjRzjEXA1/LzFuKGeLpVW0rq7ZX0/nPZwMqM7nP11WxJElaL1188cUcfvjhrFq1iiFDhvCd73yHY445hoULF7LBBhvwpje9qde+gaArfTHAPkflI/ha7BkRO1BZQvARipnKdvwWuDAi/gP4F/AhYH71AcX62Tdm5uyI+BXwUWBAUc8WVYduCTxabB9VQ413Aj+KiK9l5rKI2LqYhb0d+CRwbtH/qMx8sLML9d+oHwtr/GoNta+xsbHm32z1So5f/RzD+jmG9XH86vdqGcNRo0Yxd+7cl+278cYbe6maNdPnHuLKzGXA3RHxO4pw14n7gEuAP1D5KP9HHVzzcSozpfcAdxfHt9UPuDoiFlBZH3tRZj4D/Bg4pPUhruI6P4iI+4Gnarif3wNfAn4ZEfOBrxVNJwJ7FA93PUzlYTFJkqQ+ry/OwJKZh3XSNqDq7b8yc3yN1/wO8J129k+vejumnfZHgJFtdt/cxXXIzOFV2zOBmW3an6IyayxJkrRe6XMzsJIkSerb+uQMbC0ysxFobLs/In4LbNJm939l5oIeKEuSJEldWG8DbEcyc6/erkGSJEkdcwmBJEmSSsUAK0mSpFIxwEqSJKlUDLCSJEkqFQOsJEmSSsUAK0mSpFIxwEqSJKlUDLCSJEkqFQOsJEmSSsUAK0mSpFIxwEqSJKlUDLCSJEkqFQOsJEmSSsUAK0mSpFIxwEqSJKlUDLCSJEkqFQOsJEmSSsUAK0mSpFIxwEqSJKlUDLCSJEkqFQOsJEmSSsUAK0mS1MbgwYMZMWIEo0aNYo899nhx/8UXX8ywYcPYbbfdOOWUU3qxwvXbhr1dgHrOiubVDJ42q7fLKLWTRrQw0TFca45f/RzD+jmG9Sn7+C0++301Hzt79mwGDhz4svc333wz8+fPZ5NNNmHp0qXdUaJqsF7MwEbE9IiY2o3XHxQRv42IeRGxT0R8KCL+EBGzOzh+YkRc0l31SJKkde/SSy9l2rRpbLLJJgBss802vVzR+mu9CLDtiYh1Ofu8P7AgM9+amXOAycCxmTl2HfbRpXV8T5IkrbciggMOOIDRo0czY8YMAB555BHmzJnDXnvtxbve9S7uu+++Xq5y/dVnA09EnAocBSwFlgD3R0Qj8CAwBrguIh4EzqMyDvcBx2fmyohYDHwfeA+wAjgsM/8cEYOBbwMDgSeBScDWwFeB/hGxB/Cj4vpXRMQtmXlyByW+PiJ+BuwI/CgzTynqPhT4HBDArMz8bLG/KTMHFNsTgPGZOTEirgSeB94K3A18ps04HAccBzBw4CDOGNGy5oOpF23bv/LxmdaO41c/x7B+jmF9yj5+jY2NNR331a9+lUGDBvHPf/6TqVOnsmLFCp599lkWLFjA2WefzR//+Efe//73c+211xIRa1RDU1NTzXWofX0ywEbEaOCjwCgq9/gAcH/RvHFm7hERmwJ/AvbPzEci4rvA8cAFxXHPZuaIiDiy2DceuBiYmZkzI+Jo4KLMPDgizgD2yMwTiv7HAlMzc24nZY6iEjpXAgsj4mJgNXAOMBr4J3B7RBycmTd1ccvbAe/IzNVtGzJzBjADYPshQ/P8BX3yR95jThrRgmO49hy/+jmG9XMM61P28Vt8eMManzN//nyam5vZeeed+eQnP8nYsWMZO3Ys5513HsOHD2fQoEFrdL3GxkYaGta8Dr2kry4h2IfKrOa/M/NfwC1VbdcX/+4MLMrMR4r3M4F9q467rurfvYvtvYFri+2rqMy0rq1fZOazmfk88DDwJuBtQGNmPpmZLcA1bWrqyA/aC6+SJGnNLV++nOeee+7F7dtvv53hw4dz8MEHM3t25fGWRx55hFWrVr3sIS/1nPL+CrX2ltd4XHawva6srNpeTdc/i+oaNm3TVus9SZKkLjzxxBMccsghALS0tHDYYYfx7ne/m1WrVnH00UczfPhwNt54Y2bOnLnGywe0bvTVAHsXcGVEfIXKPR4EfLPNMQuBwRExNDP/DPwX8Muq9o8AZxf/3lPs+zWVpQlXAYcDc9Zx3fcCF0XEQCpLCA6lsmwB4ImI2KWo+xDguTW9eP+N+rFwDb4+RK/U2Ni4Vh8/qcLxq59jWD/HsD7rw/gNGTKE+fPnv2L/xhtvzNVXX90LFamtPhlgM/OBiLgemE/lIa5XPCaYmc9HxCTgB8XT+/cBl1Ud8tqIeIjKTOmhxb5PAt+JiJN56SGudVn34xExDZjNSw9x3Vw0TwNuLfqdCwxYl31LkiSVRZ8MsACZ+SXgS212n9fmmF9QeZCqPee2fgNA1fF/BfZrp68rgSur3jd0UVvb48dXbV/HS+tvq8+5Abihnf0TO+tLkiSpr+mrD3FJkiSpj+qzM7D1yMzB6+I6EXEgla/FqrYoMw9ZF9eXJElaHxlgu1Fm3gbc1tt1SJIk9SUuIZAkSVKpGGAlSZJUKgZYSZIklYoBVpIkSaVigJUkSVKpGGAlSZJUKgZYSZIklYoBVpIkSaVigJUkSVKpGGAlSZJUKgZYSZIklYoBVpIkSaVigJUkSVKpGGAlSZJUKgZYSZIklYoBVpIkSaVigJUkSVKpGGAlSZJUKgZYSZIklYoBVpIkSaVigJUkSVKpGGAlSWrH6tWreetb38r48eMBWLRoEXvttRdDhw7lIx/5CKtWrerlCqX114a9XYB6zorm1QyeNqu3yyi1k0a0MNExXGuOX/0cw/pd+e7NajruwgsvZJddduFf//oXAJ/97Gf59Kc/zUc/+lGmTJnCFVdcwfHHH9+dpUrqgDOwayAivhURu3bSPj0ipvZQLY0RsUex/bme6FOS1hd///vfmTVrFscccwwAmcmdd97JhAkTADjqqKO46aaberNEab1mgF0DmXlMZj7c23W0wwArSevQpz71Kb761a+ywQaV/0wuW7aMrbbaig03rHxwud122/Hoo4/2ZonSes0lBO2IiMHAz4D7gd2B3wNHAj8Bpmbm3Ih4N/BloB/wVGbu3+YaxwIfLF5PZuaAYv8EYHxmToyIK4HngT2ALYDPZOatHdTUH/gO8Bbgj0D/Yv/ZQP+IeBD4fWYe3ua844DjAAYOHMQZI1rWfmDEtv0rH+Fq7Th+9XMM69fU1ERjY2OH7ffccw/Nzc0899xzPPjggyxbtoy7776bFStWvHje0qVLWb58eafX6au6Gj91zTGsnwG2YzsDkzPz7oj4NvDx1oaIGARcDuybmYsiYuvqEyPiBGAccHBmroyIzvoZDOwJ7AjMjoihmfl8O8cdD/w7M3eJiJHAAwCZOS0iTsjMUe1dPDNnADMAth8yNM9f4I+8HieNaMExXHuOX/0cw/pd+e7NaGho6LD9tttu4/7772fixIk8//zz/Otf/+L73/8+K1euZMyYMWy44Ybcc8897LTTTp1ep69qbGxcL+97XXIM6+cSgo4tycy7i+2rgTFVbW8H7srMRQCZ+XRV25HAe4AJmbmyhn6+n5kvZOafgL8Awzo4bt+iDjLzIeChmu9EklSzr3zlK/z9739n8eLFfO9732O//fbjmmuuYezYsdxwww0AzJw5kw984AO9XKm0/jLAdiy7eN+RBVRmVbfr4NxN11E/kqQedM455/C1r32NoUOHsmzZMiZPntzbJUnrLT+H6tj2EbF3Zt4DHAb8CjioaPsN8I2I2KF1CUHVLOw84FLglog4MDMfA56IiF2AhcAhwHNV/XwoImYCOwBDimPac1dRx50RMRwYWdXWHBEbZWZzZzfUf6N+LDz7fTXevtrT2NjI4sMberuM0nL86ucY1m9N1h42NDS8+FHvkCFDuPfee7unKElrxBnYji0EPhERfwBeSyWUApCZT1J5MOqHETEfuL76xMz8FTAVmBURA4FpwK3Ar4HH2/TzN+Be4KfAlA7Wv1L0P6Co5ywqD5i1mgE8FBHXrM2NSpIklYkzsB1rycwj2uxraN3IzJ9SCZ1U7ZtetX0bcFvx9obi1Z6fZ+aUrorJzBXARzto+yzw2a6uIUmS1Bc4AytJkqRScQa2HZm5GBjeA/1MbLsvIg4Ezmmze1FmHtLd9UiSJJWBAfZVps3SA0mSJLXhEgJJkiSVigFWkiRJpWKAlSRJUqkYYCVJklQqBlhJkiSVigFWkiRJpWKAlSRJUqkYYCVJklQqBlhJkiSVigFWkiRJpWKAlSRJUqkYYCVJklQqBlhJkiSVigFWkiRJpWKAlSRJUqkYYCVJklQqBlhJkiSVigFWkiRJpWKAlSRJUqkYYCVJklQqBlhJkiSVigFWkrReWbVqFXvuuSdvectb2G233TjzzDMBuPPOO9l9990ZPnw4Rx11FC0tLb1cqaSObNjbBajnrGhezeBps3q7jFI7aUQLEx3Dteb41c8x7Nzis9/X5TEbbbQRd955JwMGDKC5uZkxY8Zw4IEHctRRR/GLX/yCnXbaiTPOOIOZM2cyefLkHqha0ppar2dgI+JbEbFrJ+3TI2JqN/U9MSIu6Y5rS5I6FhEMGDAAgObmZpqbm+nXrx8bb7wxO+20EwDjxo3jxhtv7M0yJXVivQ6wmXlMZj7c23WsSxHhrLokdWH16tWMGjWKbbbZhnHjxrHnnnvS0tLC3LlzAbjhhhtYsmRJL1cpqSORmb1dQ7eLiMHAz4D7gd2B3wNHAj8Bpmbm3Ih4N/BloB/wVGbuHxHTgabMPC8ijgU+WLyezMwBxbUnAOMzc2JEXAk8D+wBbAF8JjNv7aCmicD7gdcAOwI/ysxTirZDgc8BAczKzM8W+5u66PetwN2Z+Zmqfo4DjgMYOHDQ6DMuuLyOkdS2/eGJFb1dRXk5fvVzDDs34g1bdnlMU1PTizOwTU1NnH766Zx44on8+9//5pvf/CbNzc3sscce3HPPPXzrW9/q7pJLp3r8tHYcw9qMHTv2/szco7229Wm2bmdgcmbeHRHfBj7e2hARg4DLgX0zc1FEbF19YkScAIwDDs7MlRHRWT+DgT2phNLZETE0M5/v4NhRVELnSmBhRFwMrAbOAUYD/wRuj4iDM/OmLu5vO+Admbm6emdmzgBmAGw/ZGiev2B9+pGveyeNaMExXHuOX/0cw84tPryhy2MaGxtpaHjpuAceeIBly5YxdepUPvGJTwBw++23s3Llypcdp4q246c15xjWb31aQrAkM+8utq8GxlS1vR24KzMXAWTm01VtRwLvASZk5soa+vl+Zr6QmX8C/gIM6+TYX2Tms0XAfRh4E/A2oDEzn8zMFuAaYN8a+v1B2/AqSXqlZ555hmeeeQaAFStWcMcddzBs2DCWLl0KwMqVKznnnHOYMmVKb5YpqRPr06/xbddK1Lp2YgGVmdLtgEXtnLtpHf1UB+LVdP3z6Kzf5V2cK0kCli1bxtixY1m9ejUvvPACH/7whxk/fjwnn3wyt956Ky+88ALHH388++23X2+XKqkD61OA3T4i9s7Me4DDgF8BBxVtvwG+ERE7tC4hqJqFnQdcCtwSEQdm5mPAExGxC7AQOAR4rqqfD0XETGAHYEhxzJq4F7goIgZSWUJwKHBx0dZZv13qv1E/FtbwFTPqWGNjY00fUap9jl/9HMP67bjjjsybN+8V+88991zOPffcXqhI0ppan5YQLAQ+ERF/AF5LJZQCkJlPUnnQ6YcRMR+4vvrEzPwVMBWYVQTLacCtwK+Bx9v08zcqIfSnwJRO1r+2KzMfL64/G5gP3J+ZNxfNnfUrSZK0XlifZmBbMvOINvsaWjcy86dUQidV+6ZXbd8G3Fa8vaF4tefnmdnlwqnMvBK4sur9+Krt64Dr2jmn3X4zc2JX/UmSJPUV69MMrCRJkvqA9WIGNjMXA8N7oJ+JbfdFxIFUvhar2qLMPKS765EkSeqL1osA25vaLD2QJElSnVxCIEmSpFIxwEqSJKlUDLCSJEkqFQOsJEmSSsUAK0mSpFIxwEqSJKlUDLCSJEkqFQOsJEmSSsUAK0mSpFIxwEqSJKlUDLCSJEkqFQOsJEmSSsUAK0mSpFIxwEqSJKlUDLCSJEkqFQOsJEmSSsUAK0mSpFIxwEqSJKlUDLCSJEkqFQOsJEmSSsUAK0mSpFIxwEqSetTzzz/PnnvuyVve8hZ22203zjzzTAAmTpzIDjvswKhRoxg1ahQPPvhgL1cq6dVqw94uYF2KiNXAAir39QfgKGAb4NbMHF513HSgKTPPi4gri/YbuqGeicDtmflY8f5bwNcy8+G1uNZ0Xl7zu4B/Af2B3wCfy8y/d3aNFc2rGTxt1pp2rSonjWhhomO41hy/+r3ax3Dx2e/r8phNNtmEO++8kwEDBtDc3MyYMWN4z3veA8C5557LhAkTurtMSSXX12ZgV2TmqCKsrgKm9HI9E4HXt77JzGPWJrx24OTMfAuwMzAPuDMiNl5H15akbhMRDBgwAIDm5maam5uJiF6uSlKZ9LUAW20OMHRtToyITSPiOxGxICLmRcTYYn+/iDgvIn4XEQ9FxCeL/WdExH3F/hlRMQHYA7gmIh6MiP4R0RgRexTnHFpc/3cRcU5V300R8aWImB8Rv4mIbTurNSu+DvwDeM/a3K8k9bTVq1czatQottlmG8aNG8dee+0FwKmnnsrIkSP59Kc/zcqVK3u5SkmvVn1qCUGriNiQSpj72Vpe4hNUsuGIiBgG3B4ROwGTgMHAqMxsiYiti+Mvycyzir6vAsZn5g0RcQIwNTPnFm2t9b0eOAcYDfyzuP7BmXkTsBnwm8w8NSK+ChwLfLGGmh8AhgE3txmL44DjAAYOHMQZI1rWbkQEwLb9Kx/hau04fvV7tY9hY2NjzcdecMEFNDU1cfrppzNs2DAOOuggjjrqKJqbmzn//POZMmUKRx111DqvsampaY3q1Ms5fvVzDOvX1wJs/4hoXfU/B7gCeF0Hx2Yn1xkDXAyQmX+MiL8COwH/CVyWmS1F29PF8WMj4hTgNcDWwO+BH3dy/bcBjZn5JEBEXAPsC9xEZenDrcVx9wPjOrlOtXY/f8vMGcAMgO2HDM3zF/S1H3nPOmlEC47h2nP86vdqH8PFhzes8TkPPPAAy5YtY9KkSS/u23jjjTnvvPNoaFjz63WlsbGxW667vnD86ucY1q+vLSFoXQM7KjM/mZmrgGXAa9sctzXw1LroMCI2Bb4BTMjMEcDlwKZ1XLI5M1vD9Wpq/yXjrVQeXJOkV7Unn3ySZ555BoAVK1Zwxx13MGzYMB5//HEAMpObbrqJ4cOHd3YZSeuxvhZgXyEzm4DHI2I/gOJj/3cDv+rktDnA4cXxOwHbAwuBO4CPFUsUWq/VGlafiogBQPXjs88Bm7dz/XuBd0XEwIjoBxwK/HJt7q9Yb3silZnmtV0yIUk95vHHH2fs2LGMHDmSt73tbYwbN47x48dz+OGHM2LECEaMGMFTTz3Faaed1tulSnqVevV+DrVuHQn8b0R8rXj/+cz8f1Xt34yIC4rtJcBY4NKIWAC0ABMzc2XxNVg7AQ9FRDNweWZeEhGXA7+j8rZ+usYAACAASURBVCDVfVXXvRK4LCJWAHu37szMxyNiGjCbykf/szLzZWtXa3BuRJxOZdnCb4CxxYxzh/pv1I+FNXzFjTrW2Ni4Vh+RqsLxq19fGMORI0cyb968V+y/8847e6EaSWXUpwJsZg7oYP/DVEJpe20TO7jcpLY7irWvnyle1ftPA14xVZCZNwI3Vu1qqGq7DriunXMGVG3fANxQbE+voWZJkqQ+r88vIZAkSVLfYoCVJElSqRhgJUmSVCoGWEmSJJWKAVaSJEmlYoCVJElSqRhgJUmSVCoGWEmSJJWKAVaSJEmlYoCVJElSqRhgJUmSVCoGWEmSJJWKAVaSJEmlYoCVJElSqRhgJUmSVCoGWEmSJJWKAVaSJEmlYoCVJElSqRhgJUmSVCoGWEmSJJWKAVaSJEmlYoCVJElSqRhgJUmSVCoGWEmSJJWKAVaSJEmlYoCVtF5ZsmQJY8eOZdddd2W33XbjwgsvBOD0009n5MiRjBo1igMOOIDHHnuslyuVJHVkw94uQD1nRfNqBk+b1dtllNpJI1qY6BiutZ4Yv8Vnv6/T9g033JDzzz+f3Xffneeee47Ro0czbtw4Tj75ZL7whS8AcNFFF3HWWWdx2WWXdWutkqS10+szsBHxrYjYtZP26RExtSdrqkdENNV4XENE3Nrd9Uh6ude97nXsvvvuAGy++ebssssuPProo2yxxRYvHrN8+XIiordKlCR1oddnYDPzmN6uoS+JiA0zs6W365DKYPHixcybN4+99toLgFNPPZXvfve7bLnllsyePbuXq5MkdSQys2c6ihgM/Ay4H9gd+D1wJPATYGpmzo2IdwNfBvoBT2Xm/hExHWjKzPMi4ljgg8XrycwcUFx7AjA+MydGxJXA88AewBbAZzKz3ZnOiJgF/E9mPhQR84AfZeZZEXEWsCQzL4+Ik4EPA5sU7WcW5x4BnAhsDPwW+Hhmro6IpswcEBEDgR8DX8zMV3xmGhENwHTgKWB4MS5HZGZGxP7AeVR+wbgPOD4zV0bEYmCPzHwqIvYAzsvMhmKMdgSGAH/LzEOr+jkOOA5g4MBBo8+44PKuflTqxLb94YkVvV1FefXE+I14w5Y1HbdixQr++7//myOOOIJ99933ZW3XXHMNq1atYtKkSd1RYl2ampoYMGBAb5dRao5hfRy/+jmGtRk7duz9mblHe209PQO7MzA5M++OiG8DH29tiIhBwOXAvpm5KCK2rj4xIk4AxgEHF2Gus34GA3tSCXWzI2JoZj7fznFzgH0i4q9AC/DOYv8+wJSIOAB4c3GtAG6JiH2BJ4GPAO/MzOaI+AZwOPDdotZtgVuA0zLzjk7qfCuwG/AYcDfwzoiYC1wJ7J+Zj0TEd4HjgQs6u2FgV2BMZr4sHmTmDGAGwPZDhub5C3p90r3UThrRgmO49npi/BYf3tDlMc3NzYwfP54pU6bwmc985hXtQ4YM4b3vfS8zZ87shgrr09jYSENDQ2+XUWqOYX0cv/o5hvXr6TWwSzLz7mL7amBMVdvbgbsycxFAZj5d1XYk8B5gQmaurKGf72fmC5n5J+AvwLAOjpsD7EsluM4CBkTEa4AdMnMhcEDxmgc8UFznzcD+wGjgvoh4sHg/pLjmRsAvgFO6CK8A92bm3zPzBeBBKsF7Z2BRZj5SHDOzqLErt7QNr5JeKTOZPHkyu+yyy8vC65/+9KcXt2+++WaGDevo/zYkSb2tp6eS2q5XqHX9wgJgFLAdsKidczddy37uo7LU4C/AHcBA4FgqH+dDZdb1K5n5zeqTIuKTwMzM/J92rtlSnH8g8MsO+m1VHcZX0/XPo4WXfuloe8/LuzhXEnD33Xdz1VVXMWLECEaNGgXAl7/8Za644goWLlzIBhtswJve9Ca/gUCSXsV6OsBuHxF7Z+Y9wGHAr4CDirbfAN+IiB1alxBUzcLOAy6l8hH+gZn5GPBEROwCLAQOAZ6r6udDETET2IHKzOjC9orJzFURsQT4EHAWMIjK2tPzikNuA74QEddkZlNEvAFopjLDenNEfD0zlxbLHTbPzL9SCctHAz+IiM9m5jlrOEYLgcHFsoc/A//FS0F4MZWZ358C/3cNr0v/jfqxsIuvGFLnGhsba/qIWu17NYzfmDFjaG/t/3vf+95eqEaStDZ6egnBQuATEfEH4LVUQikAmfkklYeNfhgR84Hrq0/MzF8BU4FZxQNS04BbgV8Dj7fp52/AvVSC3pQO1r+2mgMsLT5+n0NllndO0eftwLXAPRGxALiBSlB9GDgNuD0iHqIye/u6qlpXA4cC+0XEx1kDRa2TqATgBcALQOtU0OeBC4t1sqvX5LqSJEl9RU/PwLZk5hFt9jW0bmTmT6mETqr2Ta/avo3KrChUwuQNHfTz88ycUktBmXk6cHqx/RiVZQPV7RcCF7Zz3vW0CdnF/gHFvyupLCPoqN9GoLHq/QlV27+g8oBX23PmADu1s396R/1IkiT1Nb3+hwwkSZKkNdFjM7CZuZjK9512dz8T2+6LiAOBtmtRF2XmId1dT0SMAK5qs3tlZu7V3X1LkiT1RevFF1q2WXrQ0323foOCJEmS1gGXEEiSJKlUDLCSJEkqFQOsJEmSSsUAK0mSpFIxwEqSJKlUDLCSJEkqlZoCbETsGBGbFNsNEXFiRGzVvaVJkiRJr1TrDOyNwOqIGArMAN4IXNttVUmSJEkdqDXAvpCZLcAhwMWZeTLwuu4rS5IkSWpfrQG2OSIOBY4Cbi32bdQ9JUmSJEkdqzXATgL2Br6UmYsiYgfgqu4rS5IkSWrfhrUclJkPR8Rnge2L94uAc7qzMEmSJKk9tX4LwUHAg8DPivejIuKW7ixMkiRJak+tSwimA3sCzwBk5oPAkG6qSZIkSepQzQ9xZeazbfa9sK6LkSRJkrpS0xpY4PcRcRjQLyLeDJwI/Lr7ypIkSZLaV+sM7CeB3YCVVP6AwbPAp7qrKEmSJKkjXc7ARkQ/YFZmjgVO7f6SJEmSpI51OQObmauBFyJiyx6oR5IkSepUrWtgm4AFEXEHsLx1Z2ae2C1VSZIkSR2oNcD+sHhJkiRJvaqmh7gyc2Z7r+4uTlLPO/roo9lmm20YPnz4y/ZffPHFDBs2jN12241TTjmll6qTJKnGGdiIWARk2/2Z6R8zKJEVzasZPG1Wb5dRaieNaGFiicdw8dnv6/KYiRMncsIJJ3DkkUe+uG/27NncfPPNzJ8/n0022YSlS5d2Z5mSJHWq1q/R2gN4W/HaB7gIuLq7imorIr4VEbt20j49Iqb2VD31ioimGo65MiIm9EQ9UrV9992Xrbfe+mX7Lr30UqZNm8Ymm2wCwDbbbNMbpUmSBNS+hGBZ1evRzLwA6HoqZx3JzGMy8+Ge6k/Syz3yyCPMmTOHvfbai3e9613cd999vV2SJGk9VusSgt2r3m5AZUa21gfAahYRg4GfAfcDuwO/B44EfgJMzcy5EfFu4MtAP+CpzNy/zTWOBT5YvJ7MzAHF/gnA+MycGBFXAs8X97EF8JnMvLWDmmYB/5OZD0XEPOBHmXlWRJwFLMnMyyPiZODDwCZF+5nFuUdQ+atlGwO/BT5efC1Z67UHAj8Gvljc48XAOGAJsKrquDOAg4D+VP4C2seAIcAPMnP34pg3A9e3vq869zjgOICBAwdxxoiWjn8A6tK2/SvLCMqqsbGxpuP+8Y9/sHz58hePf/bZZ1mwYAFnn302f/zjH3n/+9/PtddeS0SsUf9NTU0116D2OYb1cwzr4/jVzzGsX60h9Pyq7RZgEZXA1h12BiZn5t0R8W3g460NETEIuBzYNzMXRcTLPueMiBOoBMCDM3NlF/9xHQzsCewIzI6IoZn5fDvHzQH2iYi/Urn3dxb79wGmRMQBwJuLawVwS0TsCzwJfAR4Z2Y2R8Q3gMOB7xa1bgvcApyWmXdExAeLe98V2BZ4GPh20dclmXlWcd5VVIL4jyPi2YgYlZkPApOA77QtPjNnADMAth8yNM9fsM5/71ivnDSihTKP4eLDG2o7bvFiNttsMxoaKsfvvPPOfPKTn2Ts2LGMHTuW8847j+HDhzNo0KA16r+xsfHFa2rtOIb1cwzr4/jVzzGsX61rYCdn5tjiNS4zj6NqhnAdW5KZdxfbVwNjqtreDtyVmYsAMvPpqrYjgfcAEzJzZQ39fD8zX8jMPwF/AYZ1cNwcYF8qwXUWMCAiXgPskJkLgQOK1zzggeI6bwb2B0YD90XEg8X71ofeNgJ+AZySmXcU+/YFrsvM1Zn5GHBnVQ1jI+K3EbEA2I/Kn/UF+BYwqfhraR+h8md+pXXu4IMPZvbs2UBlOcGqVasYOHBgL1clSVpf1Rpgb6hx37rQ9tsOXvHtBx1YQGVWdbsOzt10Lfu5j8pSg32Au6gE1WOpLHOAyqzrVzJzVPEamplXFPtnVu3fOTOnF+e0FOcf2NVNRcSmwDeoBPMRVGagW+/lRiqhfTxwf2Yu6+p6UlcOPfRQ9t57bxYuXMh2223HFVdcwdFHH81f/vIXhg8fzkc/+lFmzpy5xssHJElaVzr9LDQihlGZ7duy+Ii71Ra8MhCuK9tHxN6ZeQ9wGPArKus/AX4DfCMidmhdQlA1CzsPuJTKR/gHFrOYT0TELsBC4BDguap+PhQRM4EdqMyMLmyvmMxcFRFLgA8BZwGDgPOKF8BtwBci4prMbIqINwDNVGZYb46Ir2fm0mK5w+aZ+VcqYflo4AcR8dnMPIdKOP5YUdM2wFgqM6qt4/xURAwAJlD88pCZz0fEbcV9T+5qYPtv1I+FNXyNkjrW2NhY88fwZXXddde1u//qq3vsi0ckSepUV4v5dqYyu7cVL4VIqATBY7uppoXAJ4r1rw9TCWcHAWTmk8VDST+MiA2ApVTWvFK0/6r4Oq1ZETEOmAbcSmU96lxgQFU/fwPupRLGp3Sw/rXVHGD/zFwREXOozPLOKfq8vQjJ9xQzUk3AEZn5cEScBtxe1NoMfAL4a3He6og4lErgfq64z/2Ke/4bcE9x3DMRcTnwO+AfVGaEq11DJZzf3sW4SpIk9QmdBtjMvJnKLGLrjGhPaMnMI9rsa6iq6afAT6sbqz6aJzNvozIrCpWZyo6WOvw8M6fUUlBmng6cXmw/RmV5QHX7hcCF7Zx3PXB9O/sHFP+u5OXLCE7ooP/TgNM6KG8M8J3qbzeQJEnqy2p9nHpeRHyCynKCF5cOZObR3VKVahIRP6LyLQr79XYtkiRJPaXWAHsV8Ecqs4VnUfk6qD+s62IyczEwvKvj1kE/E9vui4gDgXPa7F6UmYd0dz1r69VcmyRJUnepNcAOzcwPRcQHMnNmRFxLsQa0r2iz9ECSJEmvUrV+jVZz8e8zETEc2JLKk/KSJElSj6p1BnZGRLyWyoNMt1B5mv+MbqtKkiRJ6kBNATYzv1Vs/pKX/pqUJEmS1ONqWkIQEdtGxBUR8dPi/a4R0eUX50uSJEnrWq1rYK+k8oDT64v3jwCf6o6CJEmSpM7UGmAHZub3gRcAMrMF8IvzJUmS1ONqDbDLI+I/gASIiLcDz3ZbVZIkSVIHav0Wgs9Q+faBHSPibmAQMKHbqpIkSZI60GmAjYjtM/NvmflARLwL2BkIYGFmNnd2riRJktQdulpCcFPV9vWZ+fvM/J3hVZIkSb2lqwAbVdt+/6skSZJ6XVcBNjvYliRJknpFVw9xvSUi/kVlJrZ/sU3xPjNzi26tTpIkSWqj0wCbmf16qhBJkiSpFrV+D6wkSZL0qmCAlSRJUqkYYCVJklQqBlhJkiSVigFWkiRJpWKAlSRJUqkYYCVJklQqBlipZI4++mi22WYbhg8f/uK+008/nZEjRzJq1CgOOOAAHnvssV6sUJKk7tXVX+IqhYhYDSwANgJagO8CX8/MF4r2McDXgNa/HPa1zJxRtE0HmjLzvG6o62Dgkcx8uHh/FnBXZv58La41EdgjM08oaj4WeBLYjMq9n9baT0dWNK9m8LRZa9q1qpw0ooWJ3TiGi89+X5fHTJw4kRNOOIEjjzzyxX0nn3wyX/jCFwC46KKLOOuss7jsssu6rU5JknpTX5mBXZGZozJzN2Ac8B7gTICI+D/AtcCUzBwGjAE+FhFdJ4X6HQzs2vomM89Ym/Daga8X9/xm4HrgzogYtI6urVexfffdl6233vpl+7bY4qW/6rx8+XIioqfLkiSpx/SVAPuizFwKHAecEJX/in8CuDIzHyjanwJOAaZ1dI2oODcifhcRCyLiI1Vtny32zY+Is4t9x0bEfcW+GyPiNRHxDuD9wLkR8WBE7BgRV0bEhOKc/SNiXnGtb0fEJsX+xRHx+Yh4oGgbVsM9Xw/cDhy2dqOmvuDUU0/ljW98I9dccw1nnXVWb5cjSVK36RNLCNrKzL9ERD9gG2A3YGabQ+YW+zvyQWAU8BZgIHBfRNxV7PsAsFdm/jsiWqfBfpiZlwNExBeByZl5cUTcAtyamTcUbRT/bgpcCeyfmY9ExHeB44ELius9lZm7R8THganAMTXc9gPAK8JuRBxHJdAzcOAgzhjRUsOl1JFt+1eWEXSXxsbGmo77xz/+wfLly192/Lhx4xg3bhzXXHMNU6dOZdKkSd1TZB2amppqvke1zzGsn2NYH8evfo5h/fpkgF0HxgDXZeZq4ImI+CXwNuBdwHcy898Amfl0cfzwIrhuBQwAbuvi+jsDizLzkeL9TCozxa0B9ofFv/dTCdO1aPcz42Kt7wyA7YcMzfMX+COvx0kjWujOMVx8eENtxy1ezGabbUZDwyuPHzJkCO9973uZObPt7229r7Gxsd2aVTvHsH6OYX0cv/o5hvXrc0sIACJiCLAaWAo8DIxuc8ho4PfrsMsrgRMycwTweWDTOq+3svh3NbX/kvFW4A919quS+tOf/vTi9s0338ywYV2uPJEkqbT6XIAtHmS6DLgkMxP4X2BiRIwq2v8DOAf4aieXmQN8JCL6FdfbF7gXuAOYFBGvKa7VuoRgc+DxiNgIOLzqOs8VbW0tBAZHxNDi/X8Bv1zjmy1ExP8FDgCuW9trqDwOPfRQ9t57bxYuXMh2223HFVdcwbRp0xg+fDgjR47k9ttv58ILL+ztMiVJ6jZ95fPk/hHxIC99jdZVVL42i8x8PCKOAC6PiM2pfNR+QWb+uOr80yLiU1Xv3wjsDcwHEjglM/8B/KwIwnMjYhXwE+BzwOnAb6l8rdVveSm0fq/o90RgQuvFM/P5iJgE/CAiNgTuoxK618Sni/vaDPgdsF9mPtnZCf036sfCGr6mSR1rbGys+WP+7nLdda/8PWXy5Mm9UIkkSb2jTwTYzOzXRftdVNawttc2HZjeTtPJxavt8WcDZ7fZdylwaTvH3k3V12gBE6vafkHlY/+25wyu2p4LNBTbV1JZqtBZzZIkSX1en1tCIEmSpL7NACtJkqRSMcBKkiSpVAywkiRJKhUDrCRJkkrFACtJkqRSMcBKkiSpVAywkiRJKhUDrCRJkkrFACtJkqRSMcBKkiSpVAywkiRJKhUDrCRJkkrFACtJkqRSMcBKkiSpVAywkiRJKhUDrCRJkkrFACtJkqRSMcBKkiSpVAywkiRJKhUDrCRJkkrFACtJkqRSMcBKkiSpVAywkiRJKhUDrCRJkkplw94uQFobgwcPZvPNN6dfv35suOGGzJ07t7dLkiRJPcQAux5Z0byawdNm9XYZnVp89vtqPnb27NkMHDiwG6uRJEmvRq+6JQQRMT0ipnbj9QdFxG8jYl5E7BMRH4qIP0TE7O7qs50ammo45sqImNAT9UiSJJXJqy7Atici1uVM8f7Agsx8a2bOASYDx2bm2HXYh7pZRHDAAQcwevRoZsyY0dvlSJKkHhSZ2ds1EBGnAkcBS4ElwP3AeOBBYAxwXbF9HpVlD/cBx2fmyohYDHwfeA+wAjgsM/8cEYOBbwMDgSeBScDWwC1Af+BR4EfAycX2LZl5cju1zQL+JzMfioh5wI8y86yIOAtYkpmXR8TJwIeBTYr2M4tzjwBOBDYGfgt8PDNXR0RTZg6IiIHAj4EvAj8BLgbGFWOwCvh2Zt4QEWcABxV1/xr4GDAE+EFm7l709Wbg+tb3VfX///buPczqst77+PsjUKJ4iECvtsYG1BRCBCWVLfIMmKfEgvKQh014IkpFrzbu2M9Wt9r2nDu1nRaeICS10IxkGxAyiYSpnETBSZ8YwzwhnhgDGobv88fvnlzMnhMsZtb8hs/ruubit+7f6bu+OPrxXvdaaywwFqBbt+6HXXnrXc3/iymBg/fZo1nHrVmzhu7du/Pee+8xYcIExo8fzyGHHNLC1UFVVRVdunRp8fu0V+5f8dzD4rmHxXH/iuceNs+wYcMWRcSg+vaVfA2spMOArwMDyOpZTBZgAT4REYMk7Qy8DBwTEX+U9FPgW8Ct6bgPIuJgSaPT2AiyMDglIqZIOhe4PSJGpjA4KCIuSvcfBkyIiIbeBTQfOFrSq8Am4Kg0fjQwTtJxwAHA4YCAGZKGkoXm04GjIqJa0h3AWcBP0333JgvTl0fEHElfBQ4E+gJ7AyvIAjjAf0fENem8qcCIiPi1pA8kDYiIpWQB/b66xUfEJGASQI/e+8cty0v+V96oyrPKtvqcZcuWUV1dTVnZ1p+7tcrLy1vlPu2V+1c897B47mFx3L/iuYfFawtLCI4mm7X8a0R8SBbqaj2U/jwQWBURf0yPpwBDC457oODPwWl7MPCztD2VbCZ3W8xP9zoKmAl0kbQL0CsiKoDj0s8SsvB9EFmgPQY4DHhW0tL0uHe6ZidgLvCvETEnjQ0FHoiImoh4HXiioIZhad3ucmA48Pk0fjdwjqQOZGH5Z+wAPvroI9atW/f37dmzZ9OvX78SV2VmZmatpW1Px8FHzTwuGtjeHp4FBgF/AuaQLUm4gI9niQVcHxE/KTxJ0sVkM8D/Vs81N6Xzjwd+19jN0+zzHWSzxqslXQXsnHY/DPwHWdhdFBFrt/rZ5dBbb73FqFGjANi0aRNnnnkmJ5xwQomrMjMzs9bSFgLsk8BkSdeT1XMy8JM6x1QAPSXtHxGvAP/MlsHvdOCG9OfCNPZ7sqUJU8leup+/LcVFxN8krQZOBa4BupOtxf1+OmQW8D1J0yKiStI+QDXZDOuvJP0gIt6W1BXYLSJeJQvZ5wK/kPTdiLgx9eGbkqYAewHDyGZUa8PqO5K6AKcA01NtGyTNAu4kezNaozp36kDFVnxMVVvVu3dvli1bVuoyzMzMrERKHmAjYrGkh4BlZG/ieraeYzZIOocs8NW+ievHBYd8StLzwEbgjDR2MXBfeoNV7Zu4ttV8svW36yXNB/ZNY0TEbEl9gIWSAKqAsyNihaTLgdmSdiILtRcCr6bzaiSdQbZmdh1ZCB1Otvb1z6QgHhHvS7oLeAF4s57+TANGAbOLeH5mZmZmuVHyAAsQEdcC19YZ/n6dY+YCAxu4xM0R8d06x79KFgjr3msyMLngcVkz6rsCuCJtv062bKBw/23AbfWc9xAfr+MtHO+S/txItoyg1kUN3P9y4PIGyhsC3BcRNU09DzMzM7P2oE0EWNs2kn4J7Ec9Qd3MzMysvcp9gI2IntvjOpKOB26sM7wqIkZtj+u3hLZcm5mZmVlLyX2A3V4iYhbZG7LMzMzMrA1rC58Da2ZmZmbWbA6wZmZmZpYrDrBmZmZmlisOsGZmZmaWKw6wZmZmZpYrDrBmZmZmlisOsGZmZmaWKw6wZmZmZpYrDrBmZmZmlisOsGZmZmaWKw6wZmZmZpYrDrBmZmZmlisOsGZmZmaWKw6wZmZmZpYrDrBmZmZmlisOsGZmZmaWKw6wZmZmZpYrDrBmZmZmlisOsGZmZmaWKw6wZmZmZpYrDrBmZmZmlisOsO2ApEsl7VLqOranmpoaBg4cyIgRI0pdipmZmbUxHUtdQHshqWNEbCrR7S8F7gf+2thB66tr6DlxZutU1IDKG05q1nG33XYbffr04cMPP2zhiszMzCxvSj4DK6mnpJckTZO0UtJ0SbtIulLSs5JekDRJktLx4yWtkPS8pAfT2P+RtDT9LJG0Wxq/LF3jeUlXF9xvpaS7JL0oabakzmnfF9KxSyXdLOmFNN4hPa691jfTeJmk+ZJmACsaeY6j03nLJE0tqOOJND5XUo80PlnSKQXnVhXcqzz1p7ZfkjQe+AdgnqR52/mvpyRee+01Zs6cyfnnn1/qUszMzKwNKnmATQ4E7oiIPsCHwLeB/46IL0REP6AzUPta8kRgYET0B8alsQnAhRExADgaWC/pOOAA4HBgAHCYpKHp+AOAH0XE54H3ga+l8fuAb6br1BTUdx7wQUR8AfgCcIGkXmnfocAlEfG5+p6YpM8DlwPDI+IQ4JK064fAlPQ8pgG3N6NPA8lmW/sCvYGjIuJ24HVgWEQMa8Y12rxLL72Um266iZ12aiv/eJqZmVlb0laWEKyOiAVp+35gPLBK0r8CuwBdgReBXwPPA9MkPQo8ms5ZAPyXpGnAIxHxWgqwxwFL0jFdyILrn4FVEbE0jS8CekraE9gtIham8Z/xcWg+DuhfMDO6R7rW34BnImJVI89tOPCLiHgHICLeTeODga+m7anATU01Kd3rNQBJS4GewFONnSBpLDAWoFu37lx5cKlWOWTKy8sb3b9w4UKqq6tZt24dS5cuZe3atU2e05qqqqraVD154/4Vzz0snntYHPeveO5h8dpKgI16Ht8BDIqI1ZKuAnZO+04ChgInA/8u6eCIuEHSTOBLwAJJxwMCro+InxReWFJPYGPBUA3ZDG9jBFwcEbPqXKsM+Kg5T3ArbCLNjEvaCfhEwb66dTf59xcRk4BJAD167x+3LC/tX3nlWWWN7p81axaLFi1izJgxbNiwgQ8//JC7776b+++/v3UKbEJ5eTllZWWlLiO33L/iuYfFcw+L4/4Vzz0sXlt5jbaHpMFp+0w+nlV8R1IXvzsNrwAAG8JJREFU4BT4e6D7bETMA75LNhPaRdJ+EbE8Im4EngUOAmYB56bzkbSPpL0aKiAi3gfWSToiDX29YPcs4FuSOqVrfU7Srs18bk8Ap0r6dDq3axr/fcE9zgLmp+1K4LC0/WWgUzPusQ7YrZn1tGnXX389r732GpWVlTz44IMMHz68zYRXMzMzaxvaygxsBXChpHvJ3gx1J/Ap4AXgTbJQCtABuF/SHmSzordHxPuSvidpGLCZbKnB4xGxUVIfYGF6/1cVcDZbrm2t6zzgLkmbgd8BH6Txu8lerl+c3ky2BhjZnCcWES9Kuhb4naQasiUNY4CLgfskXZaud0465S7gV5KWAb+heTO8k4DfSHq9vayDNTMzM2uIIuq+et/KBWQv6T+W3qxVUpK6RETtu/4nAp+JiEuaOC03DjzwwKioqCh1Gbnml32K4/4Vzz0snntYHPeveO5h80haFBGD6tvXVmZg24qTJP0bWV9eJZspNTMzM7M2pOQBNiIqgZLPvgJExEPAQ9tyblrjOreeXcdExNqiCjMzMzOzvyt5gG0vUkgdUOo6zMzMzNq7tvIpBGZmZmZmzeIAa2ZmZma54gBrZmZmZrniAGtmZmZmueIAa2ZmZma54gBrZmZmZrniAGtmZmZmueIAa2ZmZma54gBrZmZmZrniAGtmZmZmueIAa2ZmZma54gBrZmZmZrniAGtmZmZmueIAa2ZmZma54gBrZmZmZrniAGtmZmZmueIAa2ZmZma54gBrZmZmZrniAGtmZmZmueIAa2ZmZma54gBrZmZmZrniAGtmZmZmueIAa2ZmZma50rHUBVjrWV9dQ8+JM1v0HpU3nNTo/g0bNjB06FA2btzIpk2bOOWUU7j66qtbtCYzMzNrXxxgrVV98pOf5IknnqBLly5UV1czZMgQTjzxRI488shSl2ZmZmY50W6WEEi6W1LfRvZfJWlCa9ZUTw3XSPriNp67p6Rvb++aWpskunTpAkB1dTXV1dVIKnFVZmZmliftJsBGxPkRsaLUdTQmIq6MiN9u4+l7ArkPsAA1NTUMGDCAvfbai2OPPZYjjjii1CWZmZlZjigiSl3DVpHUE/gNsAg4FHgRGA38DzAhIp6TdAJwHdABeCcijpF0FVAVEd+XdAHw1fSzJiK6pGufAoyIiDGSJgMbgEHA7sB3IuKxBmoaA4wEdgUOAL4PfAL4Z2Aj8KWIeDdd87GImC6pEpgCnAx0Ak6NiJcK60zXfgEYAdwAfAWoAOZExGWSLgNOAz4J/DIi/qOe2sYCYwG6det+2JW33tX8Zm+Dg/fZo9nHVlVVccUVVzB+/Hh69erVglVtP1VVVX+fQbat5/4Vzz0snntYHPeveO5h8wwbNmxRRAyqb19e18AeCJwXEQsk3UvBzKSk7sBdwNCIWCWpa+GJki4CjgVGRsTGJl6+7gkcDuwHzJO0f0RsaODYfsBAYGfgFeC7ETFQ0g/IAvat9ZzzTkQcmpYGTADOb6SWiUC/iBiQnsdxZGH5cEDADElDI+LJwpMiYhIwCaBH7/3jluUt+1deeVbZVh2/ePFi1q5dyznnnNMyBW1n5eXllJWVlbqM3HL/iuceFs89LI77Vzz3sHh5XUKwOiIWpO37gSEF+44EnoyIVQAR8W7BvtHAicApEbGxGff5eURsjoiXgT8BBzVy7LyIWBcRa4APgF+n8eVkQbg+j6Q/FzVyTEOOSz9LgMWptgO28hqtbs2aNbz//vsArF+/njlz5nDQQY211czMzGxLeZ2BrbvuobnrIJYDA4B9gVX1nLtzEfcpDMSbCx5vpuE+1x5TU3DMJrb8H4u6NdUScH1E/KSRmrbQuVMHKpr4mKuW9sYbb/CNb3yDmpoaNm/ezGmnncaIESNKWpOZmZnlS14DbA9JgyNiIXAm8BTZWlKAp4E7JPWqXUJQMAu7BLiT7OX24yPideAtSX3I1paOAtYV3OdUSVOAXkDvdExLqyRb84qkQ9O9SXXtVnDcLOB7kqZFRJWkfYDqiHi7FWrcZv3792fJkiWlLsPMzMxyLK9LCCqACyWtBD5FFkoBSC/hjwUekbQMeKjwxIh4imy96UxJ3cjWlj4G/B54o859/gw8AzwOjGtk/ev29DDQVdKLwEXAH1Pda4EFkl6QdHNEzAZ+BiyUtByYzpYB18zMzKxdyusM7KaIOLvOWFntRkQ8ThY6KRi7qmB7FtkMJmTBb3oD9/ltRIxrqpiImAxMLnjcs759ETGmgWOeq60/ItaTrW2t7z5n1nl8G3BbU/WZmZmZtSd5nYE1MzMzsx1U7mZgI6KS7COrWvo+Y+qOSToeuLHO8KqIGNXS9ZiZmZlZJncBtpTqLD0wMzMzsxLwEgIzMzMzyxUHWDMzMzPLFQdYMzMzM8sVB1gzMzMzyxUHWDMzMzPLFQdYMzMzM8sVB1gzMzMzyxUHWDMzMzPLFQdYMzMzM8sVB1gzMzMzyxUHWDMzMzPLFQdYMzMzM8sVB1gzMzMzyxUHWDMzMzPLFQdYMzMzM8sVB1gzMzMzyxUHWDMzMzPLFQdYMzMzM8sVB1gzMzMzyxUHWDMzMzPLFQdYMzMzM8uVjqUuwFrP+uoaek6cWdQ1Km84qdH9q1evZvTo0bz11ltIYuzYsVxyySVF3dPMzMysUJuagZV0t6S+jey/StKEVqqlXNKg7XzNUyWtlDRve163LenYsSO33HILK1as4Omnn+ZHP/oRK1asKHVZZmZm1o60qQAbEedHRHtOO+cBF0TEsJa4uKSSz6h/5jOf4dBDDwVgt912o0+fPvzlL38pcVVmZmbWnpQkwErqKeklSdPSjOR0SbsUznpKOkHSYknLJM2t5xoXSHpcUmdJVQXjp0ianLYnS/qxpOck/VHSiEZq6izpwVTPL4HOBfvuTNd4UdLVaWy4pEcLjjk2nYekMyQtl/SCpBvT2JXAEOAeSTdLellS97RvJ0mvSOqefh6W9Gz6OSodc7ikhZKWSPq9pAPT+BhJMyQ9AfyvPpVSZWUlS5Ys4Ygjjih1KWZmZtaOlHLG7kDgvIhYIOle4Nu1O1KwuwsYGhGrJHUtPFHSRcCxwMiI2Cipsfv0BA4H9gPmSdo/IjbUc9y3gL9GRB9J/YHFBfv+PSLeldQBmJv2zwPukNQ9ItYA5wD3SvoH4EbgMOA9YLakkRFxjaThwISIeC6F7rOAW4EvAssiYo2knwE/iIinJPUAZgF9gJeAoyNik6QvAtcBX0v1HQr0j4h36z4pSWOBsQDdunXnyoM3NdarJpWXlzfruPXr13PJJZdw/vnns3jx4qZPyImqqqpm98D+N/eveO5h8dzD4rh/xXMPi1fKALs6Ihak7fuB8QX7jgSejIhVAHWC2WhgNVl4rW7GfX4eEZuBlyX9CTgIWFrPcUOB29P9npf0fMG+01IQ7Ah8BuibjpkKnC3pPmBwqu0koDyFWiRNS9d+lC3dC/yKLMCeC9yXxr8I9C0I5btL6gLsAUyRdAAQQKeCa82pL7ym5zIJmATQo/f+ccvy4v7KK88qa/KY6upqRowYwbhx4/jOd75T1P3amvLycsrKykpdRm65f8VzD4vnHhbH/Suee1i8UgbYaOJxQ5YDA4B9gVX1nLvzdroPAJJ6AROAL0TEe2l5Qu097gN+DWwAfpFmR5t13YhYLemtNCt7ONlsLGTLOo6sO0ss6b+BeRExSlJPoLxg90db85xaUkRw3nnn0adPn3YXXs3MzKxtKGWA7SFpcEQsBM4EngJOTvueJnt5vlftEoKCGcYlwJ3ADEnHR8TrwFuS+gAVwChgXcF9TpU0BegF9E7H1OfJVMcTkvoB/dP47mQB8QNJewMnksJjRLwu6XXgcrKZU4BngNsldSNbQnAG8MMG7nk32ezz1IioSWOzgYuBmwEkDYiIpWQzsLXvhhrTwPUa1blTByqa+BisYi1YsICpU6dy8MEHM2DAAACuu+46vvSlL7Xofc3MzGzHUcoAWwFcmNa/riALpScDpLWgY4FHJO0EvE225pW0/6n0cVozJR0LTAQeA9YAzwFdCu7zZ7JQuTswroH1r6T73ydpJbASWJTutUzSErI1qKuBBXXOmwZ0j4iV6fg3JE0kWyMrYGZE/KqBe84gm8W9r2BsPPCjtIShI1mwHgfcRLaE4HKguA9zbUFDhgwhYqsmuc3MzMy2SikD7KaIOLvOWFntRkQ8DjxeuDMirirYnkX2BieA6emnPr+NiHFNFRMR64GvN7BvTCOnDiF7w1nh8Q8AD9RznbI6Q4eQvXnrpYJj3gFOr+fchcDnCoYuT+OTgcmN1GdmZmbWrpT8c0PzTNIisuUF/7IN504k++SDs5o61szMzMw+VpIAGxGVQL9WuM+YumOSjif7mKtCqyJi1DZc/7BtLI2IuAG4YVvPNzMzM9tR7XAzsHWWHpiZmZlZzrSpr5I1MzMzM2uKA6yZmZmZ5YoDrJmZmZnligOsmZmZmeWKA6yZmZmZ5YoDrJmZmZnligOsmZmZmeWKA6yZmZmZ5YoDrJmZmZnligOsmZmZmeWKA6yZmZmZ5YoDrJmZmZnligOsmZmZmeWKA6yZmZmZ5YoDrJmZmZnligOsmZmZmeWKA6yZmZmZ5YoDrJmZmZnligOsmZmZmeWKA6yZmZmZ5YoDrJmZmZnligOsbbVzzz2Xvfbai379+pW6FDMzM9sBdSx1AdZ61lfX0HPizEaPqbzhpCavM2bMGC666CJGjx69vUozMzMza7ZWn4GVdLekvo3sv0rShNasqRiSqrbi2O6S/iBpiaSjW7KuljR06FC6du1a6jLMzMxsB9XqM7ARcX5r37MNOQZY3pI9kNQhImpa6vpmZmZmpdZiM7CSekp6SdI0SSslTZe0i6RySYPSMSdIWixpmaS59VzjAkmPS+pcONMp6RRJk9P2ZEk/lvScpD9KGtFITTMl9U/bSyRdmbavkXRB2r5M0rOSnpd0dcG5Z0t6RtJSST+R1KHOtbtJWijppPTcn0jXmCuph6QBwE3AV9I1bpZ0acH510q6pIkaHpW0SNKLksYWjFdJukXSMmBw8/6GzMzMzPKppWdgDwTOi4gFku4Fvl27Q1J34C5gaESskrTFa9KSLgKOBUZGxEZJjd2nJ3A4sB8wT9L+EbGhnuPmA0dLehXYBByVxo8Gxkk6DjggXUvADElDgTXA6cBREVEt6Q7gLOCnqda9gRnA5RExR9KvgSkRMUXSucDtETEyBeZBEXGRpJ7AI8CtknYCvg4c3lANEfEkcG5EvCupM/CspIcjYi2wK/CHiPiXuk84Bd2xAN26defKgzc11kfKy8sb3V/rzTff5KOPPmr28e1FVVXVDvectyf3r3juYfHcw+K4f8VzD4vX0gF2dUQsSNv3A+ML9h0JPBkRqwAi4t2CfaOB1WThtboZ9/l5RGwGXpb0J+AgYGk9x81PNawCZgLHStoF6BURFWkW9jhgSTq+C1mY7A8cRhYaAToDb6djOgFzgQsj4ndpbDDw1bQ9lWzmdQsRUSlpraSBwN7AkohYmwJsfTU8CYyXNCqNfzaNrwVqgIfra0xETAImAfTovX/csrzxv/LKs8oa3f/34yor2XXXXSkra97x7UV5efkO95y3J/eveO5h8dzD4rh/xXMPi9fSb+KKJh43ZDnZrOq+DZy78zbe51lgENmM65NkIfECYFHaL+D6iBiQfvaPiHvS+JSC8QMj4qp0zqZ0/vHNfG6F7gbGAOcA9zZWg6Qy4IvA4Ig4JNVe24cNrbnu9YwzzmDw4MFUVFSw7777cs8997TWrc3MzMxafAa2h6TBEbEQOBN4Cjg57XsauENSr9olBAWzsEuAO8lePj8+Il4H3pLUB6gARgHrCu5zqqQpQC+gdzrmf4mIv0laDZwKXAN0B76ffgBmAd+TNC0iqiTtA1STzbD+StIPIuLttNxht4h4lSwsnwv8QtJ3I+JG4PdkSwKmki01mN9Af36Z6uiU+tNYDXsA70XEXyUdRDaDvVU6d+pARTM+JqspDzzwQNHXMDMzM9tWLR1gK4AL0/rXFWSh9GSAiFiT1mc+ktaAvk225pW0/6n0cVozJR0LTAQeI1uP+hzZS+u1/gw8A+wOjGtg/Wut+cAxEbFe0nyyWd756Z6zU0hemJYKVAFnR8QKSZcDs1Ot1cCFwKvpvBpJZ5AF7nXAxcB9ki5L9Z5TXyEpUM8D3q+dQW2oBuA3ZOt0V6a+Pt3IczQzMzNrt1o6wG6KiLPrjJXVbkTE48DjhTsLXponImaRzUgCTE8/9fltRIxrTkERcQVwRdp+newl+8L9twG31XPeQ8BD9Yx3SX9uZMtlBMPrOXYyMLn2cQrDR5LNCDdZA3BiA8+pS33jZmZmZu2Rv0q2RJR9mcMrwNyIeLnU9ZiZmZnlRYvNwEZEJdCvpa5fcJ8xdcckHQ/cWGd4VUSMqntsqUTECrL1umZmZma2FVr9m7haQ52lB2ZmZmbWjngJgZmZmZnligOsmZmZmeWKA6yZmZmZ5YoDrJmZmZnligOsmZmZmeWKA6yZmZmZ5YoDrJmZmZnligOsmZmZmeWKA6yZmZmZ5YoDrJmZmZnligOsmZmZmeWKA6yZmZmZ5YoDrJmZmZnligOsmZmZmeWKA6yZmZmZ5YoDrJmZmZnligOsmZmZmeWKA6yZmZmZ5YoDrJmZmZnligOsmZmZmeWKA6yZmZmZ5YoDrJmZmZnligOsmZmZmeWKA6yZmZmZ5YoDrJmZmZnligOsmZmZmeWKIqLUNVgrkbQOqCh1HTnXDXin1EXkmPtXPPeweO5hcdy/4rmHzfOPEdG9vh0dW7sSK6mKiBhU6iLyTNJz7uG2c/+K5x4Wzz0sjvtXPPeweF5CYGZmZma54gBrZmZmZrniALtjmVTqAtoB97A47l/x3MPiuYfFcf+K5x4WyW/iMjMzM7Nc8QysmZmZmeWKA+wOQtIJkiokvSJpYqnryQNJ90p6W9ILBWNdJc2R9HL681OlrLEtk/RZSfMkrZD0oqRL0rh72AySdpb0jKRlqX9Xp/Fekv6QfpcfkvSJUtfa1knqIGmJpMfSY/dwK0iqlLRc0lJJz6Ux/x43k6Q9JU2X9JKklZIGu3/Fc4DdAUjqAPwIOBHoC5whqW9pq8qFycAJdcYmAnMj4gBgbnps9dsE/EtE9AWOBC5M/9y5h82zERgeEYcAA4ATJB0J3Aj8ICL2B94DzithjXlxCbCy4LF7uPWGRcSAgo9+8u9x890G/CYiDgIOIftn0f0rkgPsjuFw4JWI+FNE/A14EPhKiWtq8yLiSeDdOsNfAaak7SnAyFYtKkci4o2IWJy215H9S3sf3MNmiUxVetgp/QQwHJiext2/JkjaFzgJuDs9Fu7h9uDf42aQtAcwFLgHICL+FhHv4/4VzQF2x7APsLrg8WtpzLbe3hHxRtp+E9i7lMXkhaSewEDgD7iHzZZe+l4KvA3MAf4f8H5EbEqH+He5abcC/wpsTo8/jXu4tQKYLWmRpLFpzL/HzdMLWAPcl5ax3C1pV9y/ojnAmm2jyD7Cwx/j0QRJXYCHgUsj4sPCfe5h4yKiJiIGAPuSvZJyUIlLyhVJI4C3I2JRqWvJuSERcSjZMrQLJQ0t3Onf40Z1BA4F7oyIgcBH1Fku4P5tGwfYHcNfgM8WPN43jdnWe0vSZwDSn2+XuJ42TVInsvA6LSIeScPu4VZKLznOAwYDe0qq/Rpw/y437ijgy5IqyZZODSdbj+geboWI+Ev6823gl2T/M+Xf4+Z5DXgtIv6QHk8nC7TuX5EcYHcMzwIHpHfefgL4OjCjxDXl1QzgG2n7G8CvSlhLm5bWGt4DrIyI/yrY5R42g6TukvZM252BY8nWEc8DTkmHuX+NiIh/i4h9I6In2b/3noiIs3APm03SrpJ2q90GjgNewL/HzRIRbwKrJR2Yho4BVuD+Fc1fZLCDkPQlsrVgHYB7I+LaEpfU5kl6ACgDugFvAf8BPAr8HOgBvAqcFhF13+hlgKQhwHxgOR+vP/y/ZOtg3cMmSOpP9uaODmSTDT+PiGsk9SabTewKLAHOjoiNpas0HySVARMiYoR72HypV79MDzsCP4uIayV9Gv8eN4ukAWRvIvwE8CfgHNLvNO7fNnOANTMzM7Nc8RICMzMzM8sVB1gzMzMzyxUHWDMzMzPLFQdYMzMzM8sVB1gzMzMzyxUHWDOzNkBSjaSlBT89t+EaIyX13f7VgaR/kDS9Ja7dyD0HpI8ANDPbQsemDzEzs1awPn1tbDFGAo+RfVB6s0jqGBGbmjouIl7n4w//b3Hpm7IGAIOA/2mt+5pZPngG1sysjZJ0mKTfSVokaVbBV09eIOlZScskPSxpF0n/BHwZuDnN4O4nqVzSoHROt/SVqkgaI2mGpCeAuenblu6V9IykJZK+Uk8tPSW9UHD+o5LmSKqUdJGk76Rzn5bUNR1XLum2VM8Lkg5P413T+c+n4/un8askTZW0AJgKXAOcns4/XdLhkham+/y+9tuNUj2PSPqNpJcl3VRQ9wmSFqdezU1jTT5fM2vbPANrZtY2dJa0NG2vAk4Dfgh8JSLWSDoduBY4F3gkIu4CkPSfwHkR8UNJM4DHImJ62tfY/Q4F+kfEu5KuI/ua1XPT19c+I+m3EfFRI+f3AwYCOwOvAN+NiIGSfgCMJvvmP4BdImKApKHAvem8q4ElETFS0nDgp2SzrQB9gSERsV7SGGBQRFyUns/uwNERsUnSF4HrgK+l8wakejYCFZJ+CGwA7gKGRsSq2mAN/Ps2PF8za0McYM3M2oYtlhBI6kcW9uakINoBeCPt7peC655AF2DWNtxvTsFXVx4HfFnShPR4Z7KvuFzZyPnzImIdsE7SB8Cv0/hyoH/BcQ8ARMSTknZPgXEIKXhGxBOSPp3CKcCMiFjfwD33AKZIOgAIoFPBvrkR8QGApBXAPwKfAp6MiFXpXsU8XzNrQxxgzczaJgEvRsTgevZNBkZGxLI0S1nWwDU28fFSsZ3r7CucbRTwtYio2Ir6NhZsby54vJkt/9tS9/vKm/r+8sZmQb9HFpxHpTe5lTdQTw2N//dtW56vmbUhXgNrZtY2VQDdJQ0GkNRJ0ufTvt2ANyR1As4qOGdd2lerEjgsbTf2BqxZwMVKU72SBhZf/t+dnq45BPggzZLOJ9UtqQx4JyI+rOfcus9nD+AvaXtMM+79NDBUUq90r9olBC35fM2sFTjAmpm1QRHxN7LQeaOkZcBS4J/S7iuAPwALgJcKTnsQuCy9MWk/4PvAtyQtAbo1crvvkb0c/7ykF9Pj7WVDuv+PgfPS2FXAYZKeB24AvtHAufOAvrVv4gJuAq5P12vyFcSIWAOMBR5JPXwo7WrJ52tmrUARTb2aY2ZmtvUklQMTIuK5UtdiZu2LZ2DNzMzMLFc8A2tmZmZmueIZWDMzMzPLFQdYMzMzM8sVB1gzMzMzyxUHWDMzMzPLFQdYMzMzM8sVB1gzMzMzy5X/D4Cjr2ki2YEaAAAAAElFTkSuQmCC\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 322 ms (started: 2022-11-23 03:22:09 +00:00)\n"
          ]
        }
      ]
    }
  ]
}