Enhanced Changeover Detection in Industry 4.0 environments with Machine Learning.ipynb

{
    "cells": [
        {
            "cell_type": "markdown",
            "source": [
                "# Enhanced Changeover Detection in Industry 4.0 environments with Machine Learning\r\n",
                "\r\n",
                "This notebook contains all the preliminary models used in the OBerA project.\r\n",
                "It belongs to the article \"Enhanced Changeover Detection in Industry 4.0 Environments with Machine Learning\" which can be found in Open Access by the following link: <https://www.mdpi.com/1424-8220/21/17/5896/htm>\r\n"
            ],
            "metadata": {}
        },
        {
            "cell_type": "markdown",
            "source": [
                "## ⚠️ <span style=\"color:red\">**Before you hit *\"Run All\"* be advised that it will also run SVC which will take more than 15 minutes to run. If this is something you want to avoid, we have some static variables as outputs of SVC prepared.**</span> ⚠️\r\n",
                "\r\n",
                "## <span style=\"color:yellow\">**You might encounter a *\"ValueError: object of too small depth for desired array\"* when running Balanced Random Forest if your Python version is *lower* than 3.8**</span>"
            ],
            "metadata": {}
        },
        {
            "cell_type": "markdown",
            "source": [
                "## Packages used\r\n",
                "\r\n",
                "Except for imblearn other packages used here are rather common for Anaconda distribution. Most of them are of the newest (or rather new) versions as of August 2021. You shall be able replicate the same results with the newest versions of these packages. However, below the version list of the packages used in the imports will be presented.\r\n",
                "\r\n",
                "*For those interested, a 'requirements.txt' (generated by conda) file with all the packages used within the environment was also added to the project*\r\n",
                "\r\n",
                "\r\n",
                "conda version : 4.10.3\r\n",
                "\r\n",
                "python version : 3.9.6\r\n",
                "\r\n",
                "\r\n",
                "pandas==1.3.2\r\n",
                "\r\n",
                "numpy==1.21.2\r\n",
                "\r\n",
                "matplotlib==3.4.3\r\n",
                "\r\n",
                "seaborn==0.11.2\r\n",
                "\r\n",
                "tensorflow==2.6.0\r\n",
                "\r\n",
                "scikit-learn==0.24.2\r\n",
                "\r\n",
                "imbalanced-learn==0.8.0\r\n"
            ],
            "metadata": {}
        },
        {
            "cell_type": "markdown",
            "source": [
                "# Imports\r\n"
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 1,
            "source": [
                "# Importing libraries\r\n",
                "import pandas as pd #for working with dataframes\r\n",
                "import numpy as np #for numerical and vector operations\r\n",
                "import matplotlib as mpl #for visualization\r\n",
                "import matplotlib.pyplot as plt #for visualization\r\n",
                "import seaborn as sns #for visualization\r\n",
                "import tensorflow as tf #for neural networks\r\n",
                "from tensorflow import keras #for neural networks\r\n",
                "\r\n",
                "# Metrics\r\n",
                "from sklearn import metrics #metrics\r\n",
                "from sklearn.metrics import classification_report #for F1 score and other metrics\r\n",
                "from sklearn.metrics import f1_score #for F1 score specifically\r\n",
                "from sklearn.metrics import matthews_corrcoef #for Matthews correlation coefficient (MCC)\r\n",
                "from sklearn.metrics import confusion_matrix #for confusion matrix\r\n",
                "from sklearn.metrics import roc_curve, roc_auc_score #ROC and AUC scores\r\n",
                "\r\n",
                "# Importing machine learning tools for preprocessing\r\n",
                "from sklearn.model_selection import train_test_split #for splitting the data into training and test sets\r\n",
                "from sklearn.preprocessing import StandardScaler #for feature scaling\r\n",
                "\r\n",
                "# Models\r\n",
                "from sklearn.tree import DecisionTreeClassifier #basic decision tree\r\n",
                "from tensorflow.keras.optimizers import Adam #optimizer for neural networks\r\n",
                "from imblearn.ensemble import BalancedRandomForestClassifier #balanced random forest classifier\r\n",
                "from sklearn.ensemble import RandomForestClassifier #random forest classifier\r\n",
                "from sklearn.ensemble import ExtraTreesClassifier #extra trees classifier\r\n",
                "from sklearn.svm import SVC #support vector machine linear classifier"
            ],
            "outputs": [],
            "metadata": {}
        },
        {
            "cell_type": "markdown",
            "source": [
                "# Function definition\r\n",
                "\r\n",
                "Here we define a few functions that will be used to evaluate the models.\r\n"
            ],
            "metadata": {}
        },
        {
            "cell_type": "markdown",
            "source": [
                "## Confusion matrix function\r\n"
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 2,
            "source": [
                "# Define the confusion matrix function with graphical output\r\n",
                "\r\n",
                "def cf_matrix(test_labels, y_pred):\r\n",
                "\r\n",
                "      # Create a basic confusion matrix\r\n",
                "      cf = confusion_matrix(test_labels, y_pred)\r\n",
                "\r\n",
                "      # Configuration of a confusion matrix\r\n",
                "      group_counts = [\"{0:0.0f}\".format(value) for value in cf.flatten()]\r\n",
                "      group_percentages = [\"{0:.2%}\".format(value) for value in cf.flatten()/np.sum(cf)]\r\n",
                "\r\n",
                "      # Labeling\r\n",
                "      labels = [f\"{v1}\\n{v2}\" for v1, v2 in zip(group_counts,group_percentages)]\r\n",
                "      labels = [f\"{v2}\\n{v3}\" for v2, v3 in zip(group_counts,group_percentages)]\r\n",
                "      labels = np.asarray(labels).reshape(2,2)\r\n",
                "\r\n",
                "      # Plotting \r\n",
                "      sns.heatmap(cf, annot=labels, fmt='', cmap='Blues',)\r\n",
                "      plt.title('Confusion-Matrix', fontsize = 20) #title with fontsize 20\r\n",
                "      plt.xlabel('Predicted Label', fontsize = 15) #x-axis label with fontsize 15\r\n",
                "      plt.ylabel('True Label', fontsize = 15) #y-axis label with fontsize 15\r\n",
                "\r\n",
                "      # Assigning correct labels and their position\r\n",
                "      ax = plt.gca()\r\n",
                "      ax.set_xticklabels(['Changeover', 'Production'])\r\n",
                "      ax.set_yticklabels(['Changeover', 'Production'], va='center')\r\n",
                "\r\n",
                "      # Displaying\r\n",
                "      print(cf_matrix)"
            ],
            "outputs": [],
            "metadata": {}
        },
        {
            "cell_type": "markdown",
            "source": [
                "## ROC curve function\r\n",
                " "
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 32,
            "source": [
                "# Define the ROC curve function with graphical output\r\n",
                "\r\n",
                "def plot_roc_curve(model):\r\n",
                "\r\n",
                "    #setting up the parameters\r\n",
                "    try:\r\n",
                "        probs = model.predict(test_features)\r\n",
                "        fpr, tpr, thresholds = roc_curve(test_labels, probs)\r\n",
                "    except ValueError: #this is needed for Neural Network\r\n",
                "        probs = np.argmax(model.predict(test_features, batch_size=1, verbose=0), axis=-1)\r\n",
                "        fpr, tpr, thresholds = roc_curve(test_labels, probs)\r\n",
                "        \r\n",
                "    #plotting the ROC curve\r\n",
                "    plt.plot(fpr, tpr, color='orange', label='ROC')\r\n",
                "    plt.plot([0, 1], [0, 1], color='darkblue', linestyle='--')\r\n",
                "    plt.xlabel('False Positive Rate')\r\n",
                "    plt.ylabel('True Positive Rate')\r\n",
                "    plt.title('Receiver Operating Characteristic (ROC) Curve')\r\n",
                "    plt.legend()\r\n",
                "    plt.show()\r\n",
                "\r\n",
                "    #print AUC score\r\n",
                "    print(f\"AUC score: {roc_auc_score(test_labels, probs)}\")"
            ],
            "outputs": [],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 4,
            "source": [
                "# Import data from csv\r\n",
                "df = pd.read_csv('Data from 31.03.2021 with Labels Outer Join.csv', delimiter=',', index_col=0,) #Data from 31.03.2021 with Labels.csv\r\n",
                "df.head() #show first 5 entries"
            ],
            "outputs": [
                {
                    "output_type": "execute_result",
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>DateTime</th>\n",
                            "      <th>Distance</th>\n",
                            "      <th>Power</th>\n",
                            "      <th>Liter</th>\n",
                            "      <th>V1</th>\n",
                            "      <th>V2</th>\n",
                            "      <th>X</th>\n",
                            "      <th>Y</th>\n",
                            "      <th>Phase</th>\n",
                            "      <th>Phase_compressed</th>\n",
                            "      <th>Production</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>28</th>\n",
                            "      <td>2021-03-31 06:58:28</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>3278.437</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1.206000</td>\n",
                            "      <td>1.142373</td>\n",
                            "      <td>1</td>\n",
                            "      <td>1</td>\n",
                            "      <td>0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>29</th>\n",
                            "      <td>2021-03-31 06:58:29</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>3292.477</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1.346650</td>\n",
                            "      <td>1.052089</td>\n",
                            "      <td>1</td>\n",
                            "      <td>1</td>\n",
                            "      <td>0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>30</th>\n",
                            "      <td>2021-03-31 06:58:30</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>3289.743</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1.406439</td>\n",
                            "      <td>0.724651</td>\n",
                            "      <td>1</td>\n",
                            "      <td>1</td>\n",
                            "      <td>0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>31</th>\n",
                            "      <td>2021-03-31 06:58:30</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>3289.743</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1.406439</td>\n",
                            "      <td>0.551185</td>\n",
                            "      <td>1</td>\n",
                            "      <td>1</td>\n",
                            "      <td>0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>32</th>\n",
                            "      <td>2021-03-31 06:58:30</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>3289.743</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1.433358</td>\n",
                            "      <td>0.724651</td>\n",
                            "      <td>1</td>\n",
                            "      <td>1</td>\n",
                            "      <td>0</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "               DateTime  Distance     Power  Liter   V1   V2         X  \\\n",
                            "28  2021-03-31 06:58:28       1.0  3278.437    0.0  0.0  0.0  1.206000   \n",
                            "29  2021-03-31 06:58:29       1.0  3292.477    0.0  0.0  0.0  1.346650   \n",
                            "30  2021-03-31 06:58:30       1.0  3289.743    0.0  0.0  0.0  1.406439   \n",
                            "31  2021-03-31 06:58:30       1.0  3289.743    0.0  0.0  0.0  1.406439   \n",
                            "32  2021-03-31 06:58:30       1.0  3289.743    0.0  0.0  0.0  1.433358   \n",
                            "\n",
                            "           Y  Phase  Phase_compressed  Production  \n",
                            "28  1.142373      1                 1           0  \n",
                            "29  1.052089      1                 1           0  \n",
                            "30  0.724651      1                 1           0  \n",
                            "31  0.551185      1                 1           0  \n",
                            "32  0.724651      1                 1           0  "
                        ]
                    },
                    "metadata": {},
                    "execution_count": 4
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "markdown",
            "source": [
                "# Removing unnecessary columns\r\n",
                "\r\n",
                "We need to ensure that our data contains entries useful for binary classification only. Everything else has to be discarded.\r\n"
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 5,
            "source": [
                "# Dropping the columns we do not need\r\n",
                "df.drop(['DateTime', 'Phase', 'Phase_compressed'], axis=1, inplace=True)\r\n",
                "df.head() #show first 5 entries"
            ],
            "outputs": [
                {
                    "output_type": "execute_result",
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>Distance</th>\n",
                            "      <th>Power</th>\n",
                            "      <th>Liter</th>\n",
                            "      <th>V1</th>\n",
                            "      <th>V2</th>\n",
                            "      <th>X</th>\n",
                            "      <th>Y</th>\n",
                            "      <th>Production</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>28</th>\n",
                            "      <td>1.0</td>\n",
                            "      <td>3278.437</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1.206000</td>\n",
                            "      <td>1.142373</td>\n",
                            "      <td>0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>29</th>\n",
                            "      <td>1.0</td>\n",
                            "      <td>3292.477</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1.346650</td>\n",
                            "      <td>1.052089</td>\n",
                            "      <td>0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>30</th>\n",
                            "      <td>1.0</td>\n",
                            "      <td>3289.743</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1.406439</td>\n",
                            "      <td>0.724651</td>\n",
                            "      <td>0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>31</th>\n",
                            "      <td>1.0</td>\n",
                            "      <td>3289.743</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1.406439</td>\n",
                            "      <td>0.551185</td>\n",
                            "      <td>0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>32</th>\n",
                            "      <td>1.0</td>\n",
                            "      <td>3289.743</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1.433358</td>\n",
                            "      <td>0.724651</td>\n",
                            "      <td>0</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "    Distance     Power  Liter   V1   V2         X         Y  Production\n",
                            "28       1.0  3278.437    0.0  0.0  0.0  1.206000  1.142373           0\n",
                            "29       1.0  3292.477    0.0  0.0  0.0  1.346650  1.052089           0\n",
                            "30       1.0  3289.743    0.0  0.0  0.0  1.406439  0.724651           0\n",
                            "31       1.0  3289.743    0.0  0.0  0.0  1.406439  0.551185           0\n",
                            "32       1.0  3289.743    0.0  0.0  0.0  1.433358  0.724651           0"
                        ]
                    },
                    "metadata": {},
                    "execution_count": 5
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "markdown",
            "source": [
                "# Data description\r\n"
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 6,
            "source": [
                "# Using the describe function to get 'raw' insights\r\n",
                "df.describe()"
            ],
            "outputs": [
                {
                    "output_type": "execute_result",
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>Distance</th>\n",
                            "      <th>Power</th>\n",
                            "      <th>Liter</th>\n",
                            "      <th>V1</th>\n",
                            "      <th>V2</th>\n",
                            "      <th>X</th>\n",
                            "      <th>Y</th>\n",
                            "      <th>Production</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>count</th>\n",
                            "      <td>39591.000000</td>\n",
                            "      <td>39591.000000</td>\n",
                            "      <td>39591.000000</td>\n",
                            "      <td>39591.000000</td>\n",
                            "      <td>39591.000000</td>\n",
                            "      <td>39591.000000</td>\n",
                            "      <td>39591.000000</td>\n",
                            "      <td>39591.000000</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>mean</th>\n",
                            "      <td>0.987144</td>\n",
                            "      <td>941.545698</td>\n",
                            "      <td>9.170771</td>\n",
                            "      <td>0.528782</td>\n",
                            "      <td>0.793968</td>\n",
                            "      <td>2.993836</td>\n",
                            "      <td>1.131005</td>\n",
                            "      <td>0.349019</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>std</th>\n",
                            "      <td>0.112657</td>\n",
                            "      <td>2426.354646</td>\n",
                            "      <td>22.060219</td>\n",
                            "      <td>0.499177</td>\n",
                            "      <td>0.404459</td>\n",
                            "      <td>1.347984</td>\n",
                            "      <td>0.669302</td>\n",
                            "      <td>0.476666</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>min</th>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>-31337.590000</td>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>-0.097020</td>\n",
                            "      <td>-0.035480</td>\n",
                            "      <td>0.000000</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>25%</th>\n",
                            "      <td>1.000000</td>\n",
                            "      <td>-328.850000</td>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>1.000000</td>\n",
                            "      <td>2.312167</td>\n",
                            "      <td>0.777261</td>\n",
                            "      <td>0.000000</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>50%</th>\n",
                            "      <td>1.000000</td>\n",
                            "      <td>1558.843000</td>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>1.000000</td>\n",
                            "      <td>1.000000</td>\n",
                            "      <td>3.523018</td>\n",
                            "      <td>1.024344</td>\n",
                            "      <td>0.000000</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>75%</th>\n",
                            "      <td>1.000000</td>\n",
                            "      <td>2948.010000</td>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>1.000000</td>\n",
                            "      <td>1.000000</td>\n",
                            "      <td>3.901250</td>\n",
                            "      <td>1.554425</td>\n",
                            "      <td>1.000000</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>max</th>\n",
                            "      <td>1.000000</td>\n",
                            "      <td>19332.470000</td>\n",
                            "      <td>74.000000</td>\n",
                            "      <td>1.000000</td>\n",
                            "      <td>1.000000</td>\n",
                            "      <td>15.642741</td>\n",
                            "      <td>13.604223</td>\n",
                            "      <td>1.000000</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "           Distance         Power         Liter            V1            V2  \\\n",
                            "count  39591.000000  39591.000000  39591.000000  39591.000000  39591.000000   \n",
                            "mean       0.987144    941.545698      9.170771      0.528782      0.793968   \n",
                            "std        0.112657   2426.354646     22.060219      0.499177      0.404459   \n",
                            "min        0.000000 -31337.590000      0.000000      0.000000      0.000000   \n",
                            "25%        1.000000   -328.850000      0.000000      0.000000      1.000000   \n",
                            "50%        1.000000   1558.843000      0.000000      1.000000      1.000000   \n",
                            "75%        1.000000   2948.010000      0.000000      1.000000      1.000000   \n",
                            "max        1.000000  19332.470000     74.000000      1.000000      1.000000   \n",
                            "\n",
                            "                  X             Y    Production  \n",
                            "count  39591.000000  39591.000000  39591.000000  \n",
                            "mean       2.993836      1.131005      0.349019  \n",
                            "std        1.347984      0.669302      0.476666  \n",
                            "min       -0.097020     -0.035480      0.000000  \n",
                            "25%        2.312167      0.777261      0.000000  \n",
                            "50%        3.523018      1.024344      0.000000  \n",
                            "75%        3.901250      1.554425      1.000000  \n",
                            "max       15.642741     13.604223      1.000000  "
                        ]
                    },
                    "metadata": {},
                    "execution_count": 6
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 7,
            "source": [
                "# Correlation heatmap\r\n",
                "plt.figure(figsize=(16, 6))\r\n",
                "heatmap = sns.heatmap(df.corr(), vmin=-1, vmax=1, annot=True, fmt='.2%')\r\n",
                "heatmap.set_title('Correlation Heatmap', fontdict={'fontsize':12}, pad=12)\r\n"
            ],
            "outputs": [
                {
                    "output_type": "execute_result",
                    "data": {
                        "text/plain": [
                            "Text(0.5, 1.0, 'Correlation Heatmap')"
                        ]
                    },
                    "metadata": {},
                    "execution_count": 7
                },
                {
                    "output_type": "display_data",
                    "data": {
                        "image/png": "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",
                        "text/plain": [
                            "<Figure size 1152x432 with 2 Axes>"
                        ]
                    },
                    "metadata": {
                        "needs_background": "light"
                    }
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "markdown",
            "source": [
                "# Data preparation\r\n"
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 8,
            "source": [
                "# Splitting the data into training and test sets\r\n",
                "train_df, test_df = train_test_split(df, test_size=0.2, random_state=0)\r\n",
                "\r\n",
                "# Using numpy to create arrays of lables and features\r\n",
                "train_labels = np.array(train_df.pop('Production'))\r\n",
                "test_labels = np.array(test_df.pop('Production'))\r\n",
                "train_features = np.array(train_df)\r\n",
                "test_features = np.array(test_df)\r\n",
                "\r\n",
                "# Scaling the features using Standard Scaler\r\n",
                "scaler = StandardScaler()\r\n",
                "train_features = scaler.fit_transform(train_features)\r\n",
                "test_features = scaler.transform(test_features)\r\n",
                "\r\n",
                "# Having a look at the results\r\n",
                "print('Training labels shape:', train_labels.shape)\r\n",
                "print('Test labels shape:', test_labels.shape)\r\n",
                "print('Training features shape:', train_features.shape)\r\n",
                "print('Test features shape:', test_features.shape)"
            ],
            "outputs": [
                {
                    "output_type": "stream",
                    "name": "stdout",
                    "text": [
                        "Training labels shape: (31672,)\n",
                        "Test labels shape: (7919,)\n",
                        "Training features shape: (31672, 7)\n",
                        "Test features shape: (7919, 7)\n"
                    ]
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "markdown",
            "source": [
                "# Decision Tree\r\n"
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 15,
            "source": [
                "# Create a Decision Tree classifier object\r\n",
                "clf_DT = DecisionTreeClassifier()\r\n",
                "\r\n",
                "# Train Decision Tree Classifier\r\n",
                "clf_DT = clf_DT.fit(train_features,train_labels)\r\n",
                "\r\n",
                "# Predict the response for the test dataset\r\n",
                "y_pred_DT = clf_DT.predict(test_features)"
            ],
            "outputs": [],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 16,
            "source": [
                "# Get the metrics for the Decision Tree\r\n",
                "print(classification_report(test_labels,y_pred_DT))"
            ],
            "outputs": [
                {
                    "output_type": "stream",
                    "name": "stdout",
                    "text": [
                        "              precision    recall  f1-score   support\n",
                        "\n",
                        "           0       0.97      0.97      0.97      5148\n",
                        "           1       0.94      0.94      0.94      2771\n",
                        "\n",
                        "    accuracy                           0.96      7919\n",
                        "   macro avg       0.95      0.95      0.95      7919\n",
                        "weighted avg       0.96      0.96      0.96      7919\n",
                        "\n"
                    ]
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 17,
            "source": [
                "# Get macro average F1 score\r\n",
                "f1_DT = f1_score(test_labels, y_pred_DT, average='macro')\r\n",
                "f1_DT"
            ],
            "outputs": [
                {
                    "output_type": "execute_result",
                    "data": {
                        "text/plain": [
                            "0.9523297792664269"
                        ]
                    },
                    "metadata": {},
                    "execution_count": 17
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 18,
            "source": [
                "# Calculate Matthews Correlation coefficient (for future use)\r\n",
                "MCC_DT = matthews_corrcoef(test_labels, y_pred_DT)\r\n",
                "MCC_DT"
            ],
            "outputs": [
                {
                    "output_type": "execute_result",
                    "data": {
                        "text/plain": [
                            "0.9046701421177795"
                        ]
                    },
                    "metadata": {},
                    "execution_count": 18
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 19,
            "source": [
                "# Plot the confusion matrix for the Decision Tree\r\n",
                "cf_matrix(test_labels,y_pred_DT)"
            ],
            "outputs": [
                {
                    "output_type": "stream",
                    "name": "stdout",
                    "text": [
                        "<function cf_matrix at 0x000001BDDDED1280>\n"
                    ]
                },
                {
                    "output_type": "display_data",
                    "data": {
                        "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAEgCAYAAAC6tbgBAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAA8l0lEQVR4nO3dd3wUVdfA8d9JI6EXaQLSBBEQsXcEBQUb2LA+ig0LtseCIIpYsOtrRUVF7IgKiv1RioiKFEU6ihKkhiIlQAgp5/3jTsKS7G42IexuhvPlM5/duXNn5s5uOHP3zp07oqoYY4zxr4RYF8AYY8yeZYHeGGN8zgK9Mcb4nAV6Y4zxOQv0xhjjcxbojTHG5yzQm2JE5GYRmS8iWSKiInJrFPaZLiLpe3o/ewMRaeZ9byNjXRYTHyzQx5iItBGR50VkrohsEpEdIrJSRL4QkatEJDXK5bkQeBbYDjwD3A9MjWYZ4okXMFVE8kWkZZh8EwPy9tnNffYpj+0YUyAp1gXYm4nIYOA+3Al3KvAmsAWoD3QGXgOuBw6PYrHOKHhV1ZVR3O/JUdxXaeXi/q9cBdxddKGItAJODMgXayuAA4FNsS6IiQ/x8Ee5VxKRu3G15WXA+ar6S5A8ZwC3R7lo+wJEOcijqn9Fc3+llAGsAq4QkcGqmltk+dWAAJ8DvaJctmJUNQdYGOtymPhhTTcxICLNgCFADnBasCAPoKqfA92LrNtbRCZ7zTxZIjJHRAaKSKUg+0n3psoi8oSI/CMi2SKyWETuEhEJyDtERBTo4s0XNENoQZnDtfuKyKSCvAFpIiKXi8hPIrJWRLaLyDIR+UZELghW1iDbrSQiA0RktohsE5HNIvKDiPQO9rkWlNF7P0pE1nn7neGdOMvqVaABO3/xFOwzGbgc+AmYF2xFETlMRJ4Vkd9F5F+vPH+KyFMiUqtI3knAG97sG4Hfg/d3U/hdiUhnEblYRH4RkS0Fn1+w70pEjvKaBf8WkRpF9tlQRDK8bbQp8ydk4pbV6GPjCiAZGKWqc8NlVNXsgvci8jAwEFgHvIdr5ukBPAycKiLdvNpcoGTgf7ia+le45oVewKNAKu5XBcAk77UP0DQgfXcM9cq7BBiNa0poCBwBnA98EG5lEUkBvsE1iywEXgQqA+cBH4hIR1Ut1pTilX8a8DfwNlAbuAD4VES6qurEMhzL+8DTuNr7JwHpZ+Ga2gYA+4dY9xrgbOB74DsgETgUuA3oISJHqWqml3cksBHoCXwKzArYzsYi270d6AZ8BkwEahCCqv7i/Yp8AnfS6g0gIgnAO0A9oI+q2i8BP1JVm6I8AeMBBa4uxTrHeOv8AzQISE/C/UdX4O4i66R76V8CaQHp9XBBYyOQXGSdSe7Potj+m3nbGhmifMXWA9YDy4HKQfLvE6Ss6UXSBgaUP6lI+QuO7dggZVTgviLbOrVgW6X8rhRY7r1/DXeibByw/GvcCawy8JCXv0+RbTQFEoNs+yov/11F0vsE207A8iHe8q3AIZF+V+xsXlLgWi/tPm/+rVj/v7Bpz03WdBMbDb3X5aVY50rv9SFVXV2QqK69+HYgH1fbDOZmVc0KWGcNrrZYAzigFGUoixwgr2iiqq6LYN0rcUHoNg1oF/fK/6A3G+yYl+KCbuD+vsGdJI+MrNhBvYqrjV8JICJNcTXqd1V1W6iVVHWpqhb7DIARwGbcSagshqvqb5FmVlXFnURWAM+IyA3AvcAi3EV/41PWdBMbBW3jpRkj+lDvdULRBar6h4gsB5qLSE1V3RiweJOqLg6yvWXea60gy8rLu8BNwDwR+RDXdPGzqpbYG0REquGaQlZo8OaEgs/hkCDLZoUIrMtwv4wK9tEMF/h2oapDgpVJXfPHHOBKEXkId5JJwJ0AQvLa8a8FLgTa4k6wgZWsRuHWD2NaaVdQ1XUicjHu83sR1432AlXdWsYymArAAn1srATaAI1LsU5B++uqEMtXAft5+TYGpG8MlhnXBAGuhrqn/Bf4C1cDHuBNuSLyJXB7iBNQgUiOF6BmkGUbQ6yTy64Bthmu6aKoIWHK9SrwHO4i+RXAzAhq1R/g2uj/xv2SWg0UXHu5FSh2IT1Cq0vOEtQ03K+b5sBEVf29jNsxFYQ13cTGFO+1NH3HC2rBDUIsb1gkX3nL915DVQ5qFk1Q1TxVfVZVD8ZdsDwXGIu7gPl1sJ5CAfb48arqJFWVolMJq70NZAGv4Griw8NlFpHDcUH+O6CNql6hqgO9Xw0PACllLT+l+0UY6FlckF+Huxh8yW6UwVQAFuhj4w1c2/W5ItI2XMaAYFhQa+wcJM/+uF8HS4o025SnDd5rkyD7rw60Dreyqq5R1TGq2hvXbNASaB8mfybu10Aj74akorp4r79GUPZy432+H+E+76243jjhFPTEGafFe0QdCaQFWaeg2ancf22JyPlAX2AyrjlwLfByiM/Y+IQF+hhQ1XRc80AK8IVX6ytGRLrjukSCu3AHcI+I1A3Ikwg8ifsuX99DRS4IvAuB4wJPTt7+n6ZIwPL6v58c2FffS0/GdXcECHkB0zMCdz3jCW8/BdvYB3cRsSBPtN2Dq6Wfqju7RYaS7r12DkwUkXq4NvJg1nuv+5WxfEGJSAtc09N64BJVXQZcBlTBdVctaxOSiXPWRh8jqvqwiCTh2oini8hPwAx2DoHQCWjlpaGqP4nI40B/YK6IfISrUfbA1Yyn4PpI70lP4E4mP3oXV7fjatbJwO/AwQF503DNFeki8guuJ0wqrpfKgbga7oIS9vck7vh6Ar97bfuVcX3w6wGPq+qUMOvvEar6D66NOxLTgR+Bc7zveAru++2B6+0S7A7kn3EnwVtFpDbuzlyA5yO5kB2Md4Idhbv20VNVl3vH8rWIPAXcgfu8byrL9k2ci3X/zr19wgW954G5uK52O3AXGr/C9bOuVCT/hbhgkYkLtPOAQUBqkG2nU6RvesCyIbg23s5F0icRpB99wPKrvH1m4y4GvgLUKboeLvj3947jH6+sa3Fj+lwHpERSVtzJ4W7v88nyjnsKcFGQvM0oZV//CL6fwn70EeQN1Y++NjDMO8btuCaph3EnrVDH3R0X8Lew896AZuG+u3CfA/CUl/ZskPzJwC/e8rNj/X/CpvKfxPuijTHG+JS10RtjjM9ZoDfGGJ+zQG+MMT5ngd4YY3yuQnWvTDvkRrtybIrZMP2FWBfBxKHUJEq6y7lEpYk5Wb+9sNv721MqVKA3xpioEn80eligN8aYUCRuK+mlYoHeGGNCsRq9Mcb4nNXojTHG5xL25OMaoscCvTHGhGJNN8YY43PWdGOMMT5nNXpjjPE5q9EbY4zPWY3eGGN8znrdGGOMz1mN3hhjfC7B2uiNMcbfrEZvjDE+Z71ujDHG5+xirDHG+Jw13RhjjM9Z040xxvic1eiNMcbnrEZvjDE+ZzV6Y4zxOet1Y4wxPmc1emOM8TlrozfGGJ+zGr0xxvicT2r0UT1diUiCiPSO5j6NMabMJCHyKY5FtXSqmg/cGM19GmNMWUlCQsRTPItF0823InIH8AGwtSBRVf+NQVmMMSYk8UnTTSwC/ZXea7+ANAVaxKAsxhgTmj/ifPQDvao2j/Y+jTGmLPxSo496w5KIVBaRe0RkuDffSkTOiHY5jDGmJCIS8RTPYnEF4Q1gB3CsN78ceCgG5TDGmLASEhIinuJZLErXUlUfB3IAVDUL37SEGWN8RUoxxbFYXIzdISJpuAuwiEhLIDsG5TDGmLDivUkmUrEI9EOAr4EmIvIucBzQJwblMMaYsCzQl5Gq/k9EZgJH437w3KKq66JdDmOMKYkF+jISkXHA+8A4Vd1aUn5jjIkVvwT6WFyMfQo4AZgvIh+KyHkikhqDchhjTFiSIBFP8SwWTTffA9+LSCJwEnANMAKoHu2yGGNMOH6p0cdkmGKv182ZwAXAocCbsSiHMcaE45dAH4s7Yz8AFuBq8y/i+tXfFO1yGGNMicq5H72IJIrIbyLyuTdfW0S+FZE/vddaAXkHishiEVkkIqcGpB8mInO8Zc9JBGejWN0Z21JVr1PVCd7QxcYYE3f2wBAIt+AqugUGAONVtRUw3ptHRNoCFwLtgO7AMK+5G+AloC/Qypu6l7TTWAT68UA/EfnIm24SkeQYlMMYY8Iqz0AvIo2B04HXApJ7srPp+k2gV0D6KFXNVtUlwGLgSBFpCFRX1Z9VVYG3AtYJKRZt9C8BycAwb/4/XtrVMShL3ElIEH58tz8r12zi3Fte5qDWjXh+0IVUSavE0pXruWLQm2Ru3U5SUgIvDb6Ejm2akJSYwLtfTOPJEf8D4JtXb6HBPtXJys4B4MzrX2Dthi2xPCxTTgbfM5DJ30+idu06jPn088L09959m1HvvUNiYhKdOp3If+/oz5zZs3lwyL0AqCrX9buJk7t2i1XRK6TSjGEjIn1xNe0Cw1V1eMD8M0B/oFpAWn1VXQWgqqtEpJ6X3giYGpBvuZeW470vmh5WLAL9Eap6cMD8BBH5PQbliEs3XtyFRUsyqFbF9Th9afDFDPi/sUyZuZjLeh7Nfy8/mQeGfcG5XQ+lUkoSR/R+mLTUZH77+B5GfzWDf1a557dcMehNfp3/TywPxewBPXudw0UXX8qggXcVpk37ZSqTJozno7GfkZKSwvr16wHYv1Ur3hv9MUlJSaxdu4bzz+nJiZ27kJRkj4qOWCmuxXpBfXiwZd4IvWtUdaaIdC7jnjVMelixaLrJ88a3AUBEWgB5MShH3GlUrybdj2/HG2N/Kkxr1bQeU2YuBmDC1IX0OrkjAIpSOTWFxMQE0iqlsCMnj8yt22NRbBNFhx1+BNVr1Ngl7cMP3ufKq/uSkpICQJ06dQBIS0srDOrZ2dm+6UESTeXYdHMccJaIpAOjgJNE5B0gw2uOwXtd4+VfDjQJWL8xsNJLbxwkPaxYBPo7gYkiMklEvgcmALfHoBxx54k7z2XQs5+Qn7/zBD3/r1Wc0fkgAM7pdiiN67uL8mO++41t23ew5Nuh/PHVAzzz1ng2bN5WuN4rQy5l6qgBDLimxOs0poJbmp7OrzNncMmF53Pl5Zcyd87swmWzZ//O2Wedznm9zuKewfdbbb6UyivQq+pAVW2sqs1wF1knqOqlwDjgci/b5cCn3vtxwIUiUklEmuMuuk7zmnkyReRor7fNZQHrhBT1QK+q43GFvtmbDlDVidEuR7zpcUJ71vybyW8Llu2Sfu2Qd7m2dyd+fLc/VStXYkeO+/FzRLtm5OXl0+KUQRx4+n3c8p+TaNbI1eSuuHskR/R+mK5X/h/HHdKSi884MurHY6InNy+PzZs38877o/nv7f258/ZbcdfpoEOHgxk77gve++AjXn/1FbKzbaDY0ojCg0ceBbqJyJ9AN28eVZ0HjAbm4waB7KeqBS0f1+Mu6C4G/gK+KmknsRjrpjJwG9BUVa8R94SpA1T185LW9bNjOrbgjBMPovvx7aiUkkz1KqmMeOgyrrznLc684UUA9t+vHj1OaAdA7x6H87+f5pObm8/aDVv4edbfHNZ2P9JXrGfl2k0AbNmWzQdfzeCIdk157/NpMTs2s2fVr1+fk7t2Q0Q4qEMHEhIS2LBhA7Vr1y7M06JlS9LS0lj85x+0a39QDEtbseyJoQ1UdRIwyXu/Hjg5RL6hwNAg6TOA9qXZZyyfMHWMN29PmAIGPz+O/bvfS5vT7+OyAW8wafofXHnPW9StVRVwNYsB15zKqx9NAWD56n/pfMQBAFROTeHIDs1YlJ5BYmICdWpWASApKYHTOrVn3l+rYnNQJiq6nNyVab+4Dhrp6UvIycmhVq1aLF++jNzcXABWrlzB0vQl7NuoxA4aJkAUavRREYsGu5aqeoGIXATuCVOR3Nm1t+rd/XCuvaATAJ9OmMVbn7r/0C9/MJnh91/KzI8GIQJvfzqVuX+upHJqCuNe7EdyUiKJiQlM/GUhI8b8GMtDMOXorjtuY8b0aWzcuIFuJ3Xi+n43cfbZ5zL43rs5p+cZJCcn8+DQRxERfvt1JiNee5XkpCQkIYG77x1CrVq1S96JKeSX0CQFbXlR26HIT7ifKj+q6qFeD5z3VbXEhuS0Q26MbmFNhbBh+guxLoKJQ6lJu/+Av+a3fhFxzFnyzOlxe1aIRY3+PuwJU8aYiiBuQ3fpxGKY4m9F5FcifMJU4N1mSY07k7RPu+gU1Biz1/NL000set0c6r0tuEK4n4jUAJaqam7R/IF3m/ml6aZG1TReuu9i2rZsiCpcd/+79DzpYE7r1J4dOXksWb6Ovve9w6YtWcXW7XdRZ64451hEhDfG/MgL700CYNC1p3HlOccWDnVw3wvj+GbKfI45uAXP3n0BO3JyuWzgG/y9bB01qqbx9mNXcla/F6N52CZCq1etYtDA/qxfvw6RBM47vzeX/OfyXfKoKo89MpQpk78nNS2VB4c+yoFtXSVo8+bN3D/4HhYv/gMR4f4HH+bgjofwf089wY9TJnNAmwMZ+sjjAHw27hM2b9pUbPvGSYjzB4pEKhZNN8NwY9DPxtXo23vv64jIdar6vxiUKaqe7H8e//tpPhff+TrJSYlUTk2hauVK3Pv8OPLy8nno5p7ceeUp3PPcrvdBtG3ZkCvOOZYT/vMEO3LyGPfiDXw1ZR5//bMWgOffmcgzb4/fZZ1b/nMSF935Gk0b1qHv+Scw4OmxDOzbncdHfBO14zWlk5iUyB39B3Bg23Zs3bqFC88/l6OPOY6W++9fmGfKD5P5Z2k6n331P+bM/p2HHhjCu6M+BODxR4Zy3PEn8NQzz5GzYwdZ27eTmZnJ77N+46OxnzGw/+38+ccimuzXlHGfjGXYK6+FKInxS40+Ft0r04FDVPVwVT0MOASYC3QFHo9BeaKqWpVUjj+0JSPH/gxATm4em7ZkMX7qQvLy3IjN0+YsoVH9msXWbdO8AdPmpJO1PYe8vHx+mLmYnl0OLpYvUE5uHmmVkqmclkxObh7NG+/DvvVqFg6rYOJP3br1CmvnVapUpUWLFqxZk7FLnokTxnPmWb0QEToc3JHMzM2sXbuGLVu2MHPmdM4+9zwAklNSqF69OgkJQk5ODqrK9uxskpKSGDniNS6+9D8kJ9vgsaGIRD7Fs1gE+jbeXV8AqOp8XOD/OwZlibrmjeqwbsMWht9/KT+/fxfDBl9M5dSUXfJc1vMYvvlxfrF15/21kuMP3Z/aNaqQlppM9+Pb0bhB4XMKuO7CTkz7YCAv33cJNaulAfDEiP/x4j0XcePFXXh51GTuv/FM7h+2V9+bVqGsWLGchQsWcFCHXU/oa9ZkUL9Bg8L5+vUbsCYjg+XLllGrVm0GDxpI73N7MWTwILZt20aVKlXp2u0ULji3F40aNaZqtWrMmzuXLid1jfYhVSh+6Ucfi0C/SEReEpETvWkY8IeIVMINwelrSUmJdGzThFc//IFjLnqMbVnZ3HHlzqFj+191Knl5+Yz6cnqxdRctyeCpkd/y+Us3Mu7Ffsz+YwW5ue6u6Fc//IG2Zw7hqAsfZfW6zTx62zkAzP5jBSde/hTd+z5Hs8Z1WLV2E4Lw9qNXMOKhy6hXu1qx/Zj4sG3rVm6/9WbuHHA3VatW3XVhkG7RIkJeXi4LF8zn/AsvYvTHn5CWlsaI19yAildcdQ2jx3zKHf0H8OLzz3LDTTcz5qMPufO2Wxj+8rBi2zNWo98dfXBjNNwK/Bf420vLAbrEoDxRtSJjAyvWbGT63KUAjP1uFh3buEHqLjnzKE7r1J4+g0aGXP/NT37m2Isfo9tVz7Bh01YWe+3za/7NJD9fUVVGjPmRw9s3LbbugKu788jwrxh0bQ8efPlL3v9yOjdc1Lncj9HsvpycHG679WZOO/1MunY7pdjyevUbkLF6deF8RsZq6tarR/36DahfvwEdvF8A3U7pzsIFu/46XODNN23ajM/GfcITTz/L4sV/snRp+p47oAoqIUEinuJZLAY1y1LVp1T1bFXtpapPquo2Vc1XVd8/HSNjfSbLV2+gVVP3fIHORx7Awr9X0+3YA7m9T1fOu/UVsraH/mFTMCRCkwa16HnSwYz+egYADfapXpin50kHM7/IsAeXnnkUX/8wj42ZWVROTSE/X8nPVyqnWvtsvFFVhgweRIsWLbiszxVB83TuchKfjfsEVWX277OoWrUadevWY5+6danfoAHpS1xL6C9Tf6ZFy5a7rPvi889yw403k5ubS36e+0WYIAlsz7JhrovyS6CPRffK44AhQNPA/atqi2iXJVZue+xD3ni4DylJiaSvcF0pp7zTn0opSXz+0o0ATJuTzs1DR9Gwbg2GDb6Ys296CYD3n7ya2jWrkJObx62PjmZjpuuCOfSWXnQ4oDGqytJV/3LTQ+8X7i8tNZlLzzyKM25wd5A+984E3n/yanbk5HL5wJHRPXhTot9+ncnn4z6lVevW9D6nJwA33Xobq1a5Ycd7X3ARJ3Q6kSmTv+eMHt1ITU3jgYceLlx/wN33MvCuO8jJyaFx4yY88NAjhcsmjP+O9u0Pol69+gB06HgI5/Y6k9atW3NAmzZRPMqKId6bZCIViyEQFuKabGYS8MARbxS3sPzSj96ULxsCwQRTHkMgdBj8XcQxZ/YDXeP2tBCLfvSbVLXE8ZONMSbW4r03TaRiEegnisgTwBig8CkIqvprDMpijDEh+STOxyTQH+W9Hh6QpsBJMSiLMcaEFO8XWSMVi0HNfN+F0hjjD9Z0sxtE5HSgHZBakKaqD8SiLMYYE4pP4nxMule+DFTG3Rz1GnAeYA80NcbEHb/U6GNxZ+yxqnoZsEFV78c9O7ZJDMphjDFh+WUIhFg03RQMsr5NRPYF1gPNY1AOY4wJyy81+pCBXkQql2ZDqrotwqyfi0hN4AngV1yPGxsQ2xgTd/aGXjdbcEE4UomRZFLVB723H4vI50Cqqm4qxX6MMSYqfFKhDxvor6R0gT5iInIs0Kxg/yKCqr61J/ZljDFl5fumG1UduSd2KCJvAy2BWewc60YBC/TGmLjikzhfuouxItIWOAzXS2aEqq4Wkf2BDFXNjHAzhwNtNdqjqRljTCn5vkYfSESqAiNwfd5zvPW+BlYDDwP/AHdEuM+5QANgVUkZjTEmlvaqQA88DRwLnAz8CAQ+oeBLXJAPG+hF5DNcE001YL6ITGPXQc3OirzYxhiz5+0NvW4CnQPcoqoTRaRo75qluIeIlGQcUB/4oUj6icCKCMthjDFR45MKfcSBPg13Y1Mw1Qh4gEgYPYG7VXV2YKKIbAXuA16PsCzGGBMVfmm6iXQIhOnAZSGWnQf8FME2mhUN8gCqOgPX1dIYY+LK3jYEwj3AdyLyHfAhrq39NBH5Ly7Qd4pgG6lhlqVFWA5jjImahHiP4BGKqEavqlNwF2IrAS8AAtwPtAC6qur0CDYzXUSuKZooIlfhnh9rjDFxJSFBIp7iWcT96FX1R+AEEUkDagEbSzG+DcCtwFgRuYSdgf1wIAU4uxTbMcaYqIjz+B2xsoxeuR3Xlz6rpIyBVDUDOFZEugDtveQvVHVCGcpgjDF73N52MRYROU1EfsIF+tXAdhH5yXtaVMRUdaKqPu9NFuSNMXGrvC7GikiqiEwTkd9FZJ6I3O+l1xaRb0XkT++1VsA6A0VksYgsEpFTA9IPE5E53rLnJIKzUUSBXkSuBT7DjWh5C3C+97oFGOctN8YYX5FS/CtBNnCSqh4MdAS6i8jRwABgvKq2AsZ78wXDzVyIe+Rqd2BYwD1MLwF9gVbe1L2knUfadHM3MFxVry+S/rL3aMBBwCsRbssYYyqE8mqj98b22uLNJnuT4u4v6uylvwlMAu7y0kepajawREQWA0eKSDpQXVV/BhCRt4BewFdhjyPCctYBxoRY9jFQO8LtGGNMhVGaXjci0ldEZgRMfQO3JSKJIjILWAN8q6q/APVVdRWA91rPy94IWBaw+nIvrZH3vmh6WJHW6Cfihir4NsiyE4HJEW7HGGMqjNL0o1fV4cDwMMvzgI7eE/bGikj7UHkhaFuQhkkPK9yjBNsGzD4HvCYidYBPcGekerhukT2Aq0vakTHGVDR7otONqm4UkUm4tvUMEWmoqqtEpCEutoKrqTcJWK0xsNJLbxwkPaxwNfq57HqmEOBabyp6ZvmaCB8laIwxFUV5da8UkbpAjhfk04CuwGO4wR4vBx71Xj/1VhkHvCciTwP74i66TlPVPBHJ9C7k/oIbmub5kvYfLtB3KeMxGWOML5Rjjb4h8KbXcyYBGK2qn4vIz8Bob4SAf3A9GlHVeSIyGpgP5AL9vKYfgOuBkbihY76ihAuxEP5Rgt+X+ZCMMcYHEssp0nsDOh4SJH09bniZYOsMBYYGSZ/BzptOI1LqO2NFJIEgA5SVcjgEY4yJe365MzbSRwkK0B+4BmgeIpu10RtjfMUvY91E2o/+ZtwdW6/jLsIOBR4A/gDScXdpGWOMr4hIxFM8izTQX4N7CtTj3vwnqno/7vbchbgrwsYY4yt+efBIpIG+OTDLu+qbA9QEUNV8YBiuW5AxxvjK3lajXw9U9d7/w65Xj2thT4gyxvhQYoJEPMWzSHvd/AgcAXwJvAcMEZHawA6gH27UNWOM8ZX4Dt+RizTQD2HnwDkP45pu+uBq8t8CN5VzuYwxJub88szYiAK9qi4CFnnvs3Fj0d+yB8tljDEx55M4H/kTpkIRkXNFJK/knMYYU7H45WJsWZ4Za4wxe4U4j98Rs0BvjDEhxHtvmkhZoDfGmBDivUkmUhUq0G+Y/kKsi2Di0O2fLYh1EUwcevHsA3d7G7t9ETNOhHvC1OgIt9G45CzGGFPx7A01+roRbiMbe2asMcaHfNJEH/bBI/aEKWPMXs0uxhpjjM/5JM5boDfGmFB80kRvgd4YY0LZq8a6McaYvZHvu1caY8zezicV+tIFeu8h4Y2BJsDvqrp1j5TKGGPigF963UT8y0REbgBWAEuBH4ADvPQxInLrHimdMcbEUIJEPsWziAK9iNwJPA28CpzErg9emQRcUO4lM8aYGEsQiXiKZ5E23fQDBqvq4yKSWGTZIqB1+RbLGGNiL87jd8QiDfQNgJkhluUDqeVTHGOMiR/x3iQTqUjb6BcDJ4ZY1gmYXz7FMcaY+CGl+BfPIq3RPwMME5EdwEdeWj0RuQq4DbhmD5TNGGNiKsknHekjfTj4ayJSCxgM3O8lfwlsA4ao6nt7qHzGGBMze8MwxbtQ1SdE5GXgWKAO8C/ws6pu2lOFM8aYWPJLG32pbphS1Uzgmz1UFmOMiSs+qdBHFui9m6XCUtVhu18cY4yJH/HePz5Skdbowz2sVb1XC/TGGF9J9MnF2IgOQ1UTik5AbeAi4Heg7Z4spDHGxEICEvEUz8o8eqWqbgQ+EJEawCtA53IqkzHGxAWftNyUy3DLS4DDy2E7xhgTV8prUDMRaSIiE0VkgYjME5FbvPTaIvKtiPzpvdYKWGegiCwWkUUicmpA+mEiMsdb9pxE0Ad0twK9iDQEbscFe2OM8ZVyHNQsF7hdVQ8Ejgb6iUhbYAAwXlVbAeO9ebxlFwLtgO64G1YLxhl7CegLtPKm7iXtPNJeN2vZedG1QApQDdgOnBPJdowxpiIpr6YbVV0FrPLeZ4rIAqAR0JOdzd5v4kYDvstLH6Wq2cASEVkMHCki6UB1Vf3ZlU/eAnoBX4Xb/+70utkOLAe+VtX1EW7HGGMqjNI8eERE+uJq2gWGq+rwIPmaAYcAvwD1vZMAqrpKROp52RoBUwNWW+6l5Xjvi6aHVWKgF5Fk4DtgiaquLCm/Mcb4RWnatr2gXiywBxKRqsDHwK2qujlM83qwBRomPaxIjiMPmAAcGEFeY4zxDRGJeIpgW8m4IP+uqo7xkjO8a50F1zzXeOnLcY9sLdAYWOmlNw6SHlaJgV5V84E/gfol5TXGGD+RUkxht+POBK8DC1T16YBF44DLvfeXA58GpF8oIpVEpDnuous0r5knU0SO9rZ5WcA6IUXaRj8IeExE5qjqnAjXMcaYCq0ch0A4DvgPMEdEZnlpdwOPAqO9Id//Ac4HUNV5IjIa96yPXKCfquZ5610PjATScBdhw16IhTCBXkQ6Ab+q6hbgHtyIlbNEZAWQQZF2IVU9MoKDLdh2I6Bp4P5VdXKk6xtjTDSUV5hX1SlhNndyiHWGAkODpM8A2pdm/+Fq9BOBY4BpwFxv2m0i8hjuYeLzce3/4E4aFuiNMXElwSfjFIcL9IVHqKpXlOM+ewEHeP1DjTEmbvlkTLOyj3WzG/4GkgEL9MaYuLa3PGHqNBFpE8mGVPWtCPe5DdfWP56AYK+qN0e4vjHGRIU/wnzJgX5whNtRINJAP86bjDEmru0tNfouwIzy3KGqvikiKUBrL2mRquaU5z6MMaY8JO4lgT5LVbeW5w5FpDNu8J503C+jJiJyuXWvNMbEG3+E+dhcjH0KOEVVFwGISGvgfeCwGJTFGGNC8kmFPiaBPrkgyAOo6h/eGBDGGBNX4v0RgZEKGei958LuCTNE5HXgbW/+EmDmHtqXMcaUmdXoy+56oB9wM64JbDIwLAblMMaYsMTvNfo9xbsj9mlvMsaYuLW39LopNyIyWlV7i8gcggyUr6odolWWimDwPQOZ/P0kateuw5hPPwdg4YIFPPTAfezIziYxKZG77xnCQR3cx/b6q68w9uOPSEhM4K6B93Dc8SfEsvimnNRMS+Lyw/alemoSqjAlfQOT/trAaW324bhmNdmS7YaLGjd/DfMytpIocNEhDdmvZioKfDQ7gz/XbQPgzLZ1OapJDSqnJHLbZ4vC7NUU8Emcj2qN/hbv9Ywo7rPC6tnrHC66+FIGDbyrMO3/nn6C627ox/EnnMgPk7/nmaef4PWRb/PX4sV8/eUXjBn3BWvWZHDt1Vcw7otvSExMDLMHUxHk58OYOWtYtmk7lZISuKtLMxaucT2eJyz+l/GL/90l/3HNagHw8IQlVE1JpN+xTXh8UjoKzFmVyfd/bWDIKS2jfRgVll8CfdTG7Cl4LiJwg6ouDZyAG6JVjorisMOPoHqNGrukCcKWLe4/+ZbMTOrWdY+XnDRxPN1PO52UlBQaN25CkyZNmTtndtTLbMrf5uxclm3aDkB2bj4ZmTuomRq6k1qD6iksWuv9jezIIysnn/1qpQKQvmE7m7Nz93yhfURK8S+exWJwtm5B0npEvRQVUP8Bd/N/Tz7OKSefyFNPPsbN/70NgIyMDOo3aFCYr36D+qzJyIhVMc0eUrtyMo1rpJK+IQuAE1vU4u6TmnPpoQ1JS3b/lVdsyqZDw2okCNSpnEyTmqnUSrPey2WVIJFP8SyabfTX42ruLUUksLpZDfgpWuWoyEZ/8D533jWQrqecyjdff8mQewcx/PWRoMWfDeyXMTqMUylRuObIRnw0J4Ptufn8sGQDXy1cB8AZbety7kH1eefXVfy8dCMNqqVwV+fm/JuVw5J/s8gP8vdhIlOOT5iKqWjW6N8DzsQ93/DMgOkwVb0kiuWosD77dCwndzsFgFNO7VHYPFO/QQMyVq8uzJexOoO69erFpIym/CUIXH1UY6Yv38zvKzMByMzOQ3G9Gn5M30hTr3kmX+HjOWt4ZOISXpm6nLTkBNZs2RG7wldw1nRTSqq6SVXTgWeBfwPa53NE5KholaMiq1uvHjOmTwNg2i9T2a9pMwBO7HISX3/5BTt27GD58mX880867Q+yTkx+cemhDVmduYMJARdeq1fa+WP84IbVWLnZjfidnCikJLqg06ZuFfIVVmdaoC8ra7opu5eAQwPmtwZJ2+vddcdtzJg+jY0bN9DtpE5c3+8mBg95kMcffZi83FxSKlVi8JAHANh//1ac0r0HZ591GomJidx9z2DrceMTLeukcdR+NVmxaTsDuzQHXFfKwxvXoFGNSgCs35bD+7+5X3TVKiVx47FNUGBjVi5vzlhRuK1e7epxeJPqJCcKD3Xfn5/SN/Kl1/xjgov3mnqkRKPcficis1S1Y5G02ZH0o9+eW7z/vTG3f7Yg1kUwcejFsw/c7Sg95c8NEcec41vVituzQix63fwtIjeLSLI33YJ7vKAxxsQVKcUUz2IR6K8DjgVWAMuBo4C+oTKLSF8RmSEiM15/dXiUimiMMW4IhEineBaLsW7WABeWIv9wYDhY040xJsriO35HLOqBXkTeIPhYN1dGuyyxtnrVKgYN7M/69esQSeC883tzyX8u3yXPF5+P443XXwWgcuUqDLp3CAe0aUP6kr/pf/t/C/MtX76MG268mUsv68P/PfUEP06ZzAFtDmToI48D8Nm4T9i8aVOx7Zv4kJQg/PeEpiQlutrhbys288XCdZzdvh7tG1QlL19ZuzWHd35dSVZO/i7r1quawlVHNCqcr1MlmS8WrGXiXxtCrt+idhoXdmxAbr7yxvQVrN2aQ1pyAlce0YgXf1oW7cOPW365GBuLXjefB7xPBc4GVsagHDGXmJTIHf0HcGDbdmzduoULzz+Xo485jpb771+Yp1GjxowY+Q7Va9Rgyg/f88CQe3l31Ic0a96C0WM+BSAvL49uXTpxUtduZGZm8vus3/ho7GcM7H87f/6xiCb7NWXcJ2MZ9sprsTpUU4LcfOW5KUvJzlMSBG7v1Ix5GVtYsGYrn85bQ75Cz3Z1OaV1HT6dt3aXddds2cEjE5cArgL6cI9Whf3tQ61/cqvavPrLcupUTuGE5rUYM3cNPQ7Yh28WrY/2oce1OG+RiVgsmm4+DpwXkfeB76JdjnhQt269wvFqqlSpSosWLVizJmOXQN/xkJ29Tjt06EhGxupi2/ll6s80adKEffdtxNatW8jJyUFV2Z6dTVJSEiNHvMbFl/6H5GS7FT6eZee5H7qJCUKC1zG7YAAzgPR/t3NIo2pht3FAvSqs3bqDf7Nyw66flw/JiQmkJAl5quxTJZkaaUksXr+tXI+povNJnI/JxdiiWgH7xboQsbZixXIWLljAQR0ODpln7JiPOP6ETsXSv/7qC7qf5gYFrVKlKl27ncIF5/aiUaPGVK1WjXlz59LlpK57rOymfAgwsEtzHjutNQvXbCF9w/Zdlh/TtAbzMraE3cbhjaszc/nmoMsC1//fH+u4+JCGdGlZm+//2sCZbevy+fy1Qdfbq/mk200s2ugzcW304r2uBu4Ku5LPbdu6ldtvvZk7B9xN1apVg+aZ9stUxo75iJFvv7dLes6OHXw/cQK33Hp7YdoVV13DFVddA8CQwYO44aabGfPRh/z80xRatT6AvtfZYKHxSIFHJi4hLTmBvkc1pmG1SqzKdHe8ntq6DnkK05cFD+IAiQIHNajKp/PWFFtWdP3lm7J58vt0APavk8am7bmICFce0Yi8fGXM3AwyvbHu92Y21k0ZqWo1Va0e8Nq6aHPO3iQnJ4fbbr2Z004/k67eODZF/bFoIfffdw/PPD+MmjVr7bJsypTJtGnbjjr77FNsvQUL5gPQtGkzPhv3CU88/SyLF//J0qXp5X4cpvxk5eTz57pttK1fBYCj9qtB+4ZVGRlwl2sw7RpUZdnG7cUCdEnrdz9gH75auI7T2uzDFwvWMn3ZJjq3rF0+B1PB+aRCH9XRK8MOcaCqv0arLPFCVRkyeBAtWrTgsj5XBM2zauVKbrvlJoY+8jjNmjUvtvyrL7+gx2mnB133xeefZfCQB8jNzSU/z/3nT5AEtmdtD5rfxE7VlETyVMnKySc5QTigbhW+/XM9betVoVurOjzzw1Jy8sL3Lj6scXVmFGm2KWn9o/erwdyMLW6/iQkokA+F4+Xs9XzyMUSz6eYp7zUVOBz4HfcxdgB+AY6PYlniwm+/zuTzcZ/SqnVrep/TE4Cbbr2NVatcJ6TeF1zEKy+/yMZNG3n4wfsB11Pn/dFjAMjKymLqTz9x730PFNv2hPHf0b79QdSrVx+ADh0P4dxeZ9K6dWsOaNMmGodnSqF6ahKXHbYvCeJ6evy6PJO5q7cwpFtLkhKEm45zl7GWbMhi1KzV1EhN4pJDGjLsZ9cVMjlRaFOvSuGYNwV6H9wg6PoF6xy1Xw2e//EfACYsXs/VR7qmmzdm7JUd4YrxS/fKWIx1MwoYqqpzvPn2wB2q2qekde2GKROMjXVjgimPsW5m/ZMZcczpuF+1uD0rxKIffZuCIA+gqnNFpGMMymGMMWH55FpsTAL9AhF5DXgH19HgUsCqZMaYuOOXpptYBPorgOuBW7z5ybjx6I0xJq74pUYfi+6V24EXgcHAvcALXpoxxsSV8uxeKSIjRGSNiMwNSKstIt+KyJ/ea62AZQNFZLGILBKRUwPSDxOROd6y5ySCB0RHPdCLSGfgT+AFYBjwh4gUv93TGGNirXw70o8EuhdJGwCMV9VWwHhvHhFpixvlt523zjARKXhs3Eu4od1beVPRbRYTiyEQngJOUdUTVbUTcCrwfzEohzHGhFWeDwdX1cnAv0WSewJveu/fBHoFpI9S1WxVXQIsBo4UkYZAdVX9WV2XybcC1gkpFoE+WVUXFcyo6h+AjbZljIk7pXk4eOBDkrwp5AOVAtRX1VUA3ms9L70REDhe9HIvrZH3vmh6WLG4GDtTRF4H3vbmLwFmxqAcxhgTXikuxgY+JGkP7VnDpIcVi0B/HdAPuBlX6Mm4tnpjjIkrUehemSEiDVV1ldcsUzAi3XKgSUC+xrjndiz33hdNDyuqTTcikgDMVNWnVfUcVT1bVf9PVbOjWQ5jjImESORTGY0DCh77djnwaUD6hSJSSUSa4y66TvOadzJF5Givt81lAeuEFNUavarmi8jvIrKfqv4TzX0bY0xplWd93nvIUmdgHxFZDtwHPAqMFpGrgH+A8wFUdZ6IjAbmA7lAP1UtGJb0elwPnjTgK28KKxZNNw2BeSIyDSh8/I2qnhWDshhjTGjlGOlV9aIQi04OkX8oMDRI+gygfWn2HYtAf38M9mmMMaXmlwePRHM8+lTchdj9gTnA66qaG639G2NMafkjzEe3Rv8mkAP8APQA2rJzvBtjjIk/Pon00Qz0bVX1IACvH/20KO7bGGNKzUavLL2cgjeqmhvBODzGGBNTfglT0Qz0B4tIwQMtBUjz5gVQVa0exbIYY0yJLNCXkqomlpzLGGPihzXdGGOMz1mN3hhjfM4ncd4CvTHGhGI1emOM8T1/RHoL9MYYE0KCP+K8BXpjjAnFmm6MMcbnrHulMcb4nT/ivAV6Y4wJxSdx3gK9McaEYm30xhjjc34ZfNECvTHGhOCPMG+B3hhjQvJJhd4CvTHGhGLdK40xxuesRm+MMT5ngd4YY3zOmm6MMcbnrEZvjDE+55M4b4HeGGNC8kmkt0BvjDEhWBu9Mcb4nD14xBhj/M4CvTHG+Js13RhjjM/5pXulqGqsy2DKQET6qurwWJfDxBf7uzDBJMS6AKbM+sa6ACYu2d+FKcYCvTHG+JwFemOM8TkL9BWXtcOaYOzvwhRjF2ONMcbnrEZvjDE+Z4HeGGN8zgJ9ORCRBiIySkT+EpH5IvKliPQVkc9jXTYTP0QkT0RmichcEflQRCrvxrYmicjhZVivpojcEDC/r4h8VNZymIrBAv1uEhEBxgKTVLWlqrYF7gbqx7Zk5U9EEmNdhgouS1U7qmp7YAdwXeDCKH2+NYHCQK+qK1X1vCjs18SQBfrd1wXIUdWXCxJUdRbwA1BVRD4SkYUi8q53UkBEBovIdK9mNzwgfZKIPCYi00TkDxE5wUuvLCKjRWS2iHwgIr8U1OZE5BQR+VlEfvVqiVW99JNF5DcRmSMiI0Skkoj0EJHRBeUUkc4i8lkJ20n3yjsFOD8Kn+fe4gdgf+87mCgi7wFzRCRVRN7wvrffRKQLgIikeb8aZ4vIB0BawYZEZEvA+/NEZKT3vr6IjBWR373pWOBRoKX3y+IJEWkmInO9/KH23UdExojI1yLyp4g8Hq0PyZQPC/S7rz0wM8SyQ4BbgbZAC+A4L/0FVT3Cq9mlAWcErJOkqkd6693npd0AbFDVDsCDwGEAIrIPcA/QVVUPBWYAt4lIKjASuEBVD8KNaXQ98C1wtIhU8bZ7AfBBqO0ElGm7qh6vqqNK88GY4EQkCegBzPGSjgQGeb8G+wF439tFwJve93k9sM37GxiK9zdQgueA71X1YOBQYB4wAPjL+2VxZ5H8ofYN0BH393IQcIGINCn1gZuYsUC/Z01T1eWqmg/MApp56V28Wvkc4CSgXcA6Y7zXmQH5jwdGAajqXGC2l3407iTyo4jMAi4HmgIHAEtU9Q8v35tAJ1XNBb4GzvSCzenAp2G2U+CDMn8CJlCa9/nOAP4BXvfSp6nqEu/98cDbAKq6EFgKtAY6Ae946bPZ+TcQzknAS946eaq6qYT8ofYNMF5VN6nqdmA+u/59mDhno1fuvnlAqDbO7ID3eUCSV0MaBhyuqstEZAiQGmSdPHZ+P6HG0BPgW1W9aJdEkY5hyvsBrub2LzBdVTO9pqNi2wmwNcz2TOSyVLVjYILXahf4+YYbLzHUTS+B6akh8kQi3L6L/S3vxn5MlFmNfvdNACqJyDUFCSJyBHBiiPwF/xHXee3gkVwImwL09rbdFvfzGWAqcJyI7O8tqywirYGFQLOCdOA/wPfe+0m4n/HXsLOmHmo7JvomA5cAeN/BfsCiIuntgQ4B62SIyIEikgCcHZA+Htfkg4gkikh1IBOoVsp9mwrOAv1uUndr8dlAN3HdK+cBQ4CVIfJvBF7Ftc9+AkyPYDfDgLoiMhu4C/ezfZOqrgX6AO97y6YCbbyf11cAH3rNQ/nAy97+84DPcW3En3tpQbdTio/BlJ9hQKL3vX0A9FHVbFwTTFXv++kPTAtYZwDuu5wArApIvwXXTDgH1xTYTlXX45ro5orIExHu21RwNgRCBSCu212yqm4XkZa4mlprVd0R46IZYyoAa2erGCoDE0UkGdeOer0FeWNMpKxGb4wxPmdt9MYY43MW6I0xxucs0BtjjM9ZoPc5ERkiIhowrRSRj73eO3tqn2d4+2rmzTfz5s8oYdXAbfQWkT7lWKaqXhnCbtPLc+Nu7muIiKzbnW0EbGukiMwoj22ZvZf1utk7bAK6e+9b4MbLGS8i7VQ1Gne9rgKOwd3IFanewD64MXuMMbvBAv3eIVdVp3rvp4rIP7jRE08DPiyaWUTSVDWrvHbu3XQztcSMxpg9wppu9k4Fo202g8KhiJ8SkXtFZDmw2UtPEJEBIrJYRLLFDZ18eeCGxBkiImtEJFNE3gKqF8kTtOlGRK7xhsTdLiIZ4oZ0riFumN1zgRMDmpyGBKzXU0RmeOutFpHHvXsMArd9rlfeLBGZTDnd6Ssip4vIt97xbhaRqSJySoi8x4kb9nm7uGGBjw+S52oRmed9vktFpH95lNOYQBbo907NvNfVAWkX48bnuQE3HC3A87jhi4fjRrocC4woErBvBgZ7ec4DsoASxysXkXuAV3Bj8PTCjcmyCaiKa1qaCPyGa/I5BnjNW683boTPacBZwP1AX+CRgG0firuF/3fgHGAcUDgO/25qDnyGGz/oXOAn4CsROa5Ivsq40SZfxo3jv9HL1yCgnHfihjb4BDdU9UvAg7t7jcCYYlTVJh9PuHF31uGa6ZJww85OxNXaG3p50nHt6KkB6+2PGyPn8iLbews36iVAIm5Mn5eK5PkWN6JiM2++mTd/hjdfE9gGPB2m3B/hntoVmCa4oXPfKJJ+Je4EU8ebH40bSlcC8gzyytCnhM9LgRsj/GwTvM/0G2BEkc9cgYsD0qriRgx91JuvDmwB7iuyzQdwJ+BEb34kMCPWf0c2VezJavR7hzpAjjctwl2QvUBVAwfAGq9uMLQCJ+MC/VgRSSqYcOPsdPTG32kCNMSNaR9oDOEdg3vgyhulPI6CERVHFynTBNyooO29fEcC41Q18LbvksoUERFpLCJvisgKIBf3mZ7CznHbA40teKOqW3AnwCO9pGOAKriB54oeS32gcXmU1xiwi7F7i01AV1wtczWwskgQBMgoMr8PrsYe6mEVDYGCZog1RZYVnS+qjve6Kmyu4vbxXr8MsbzgqUcNylCmEnnDAI/DDfM7GFiMG0v+AaBekexbtPgF7TXsHF644FjmhdhdE9yvF2N2mwX6vUOuqpbUF7to4P8XV2M9DlezL2oNO/9+iga5ovNFrfdeG+KalSL1r/faF9d+X1TBU5pWl6FMkdgf93jIHqr6dUGiiKQFyVs1SO+leuw8uRUcyxkUP8mCjQNvypEFehPKBFyNvoaqfhssg4gswwXVnrhHFBY4p4Rt/4xrU78cuCNEnh0Uf1rSImAFru3/1TDbnw6cJSIDA365lFSmSBQE9MIx2kWkKe5kGOzRfmcD73n5qgLdcBetYednsK+qflEOZTMmJAv0JihVXSQiLwOjRORx3HNOU3HPt22tqlerap637EnvTtAfcD1RDixh2xtF5EFgqIik4JpiKuF69tyvqitwN1f1FJFewHJcc9NKEbkdeFvc05K+wp0QWuB67pynqtuAx4BfcG35r+Pa7q8qxeF3FJGiT/5ai7sXYDnwlIjci2vCuR938ikqyzu+qrgL1ncAKcCzAZ/BEOBZ72QxGXdxtzXQRVXPDrJNY8om1leDbdqzE16vmxLypANPBkkX4FZcO3I2Lth9D1xWJM+D3rJM4F1cV82QvW4C1r0W1zsmG/fLYDRQ3Vu2D+5i5r/eukMC1uuBO6lsxfUemgU8BCQF5Dkf14a+HfcoxiOIvNdNsGmSt/wIXNfOLOBP3JO5RhLQM4adPZ1O8MqWjevq2SnI/i7F3deQBWzAnaBuC1i+y7Ztsqksk41Hb4wxPmfdK40xxucs0BtjjM9ZoDfGGJ+zQG+MMT5ngd4YY3zOAr0xxvicBXpjjPE5C/TGGONz/w8eCOydZ1dDKwAAAABJRU5ErkJggg==",
                        "text/plain": [
                            "<Figure size 432x288 with 2 Axes>"
                        ]
                    },
                    "metadata": {
                        "needs_background": "light"
                    }
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 33,
            "source": [
                "# Plot the ROC curve and get the AUC score for the Decision Tree\r\n",
                "plot_roc_curve(clf_DT)"
            ],
            "outputs": [
                {
                    "output_type": "display_data",
                    "data": {
                        "image/png": "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",
                        "text/plain": [
                            "<Figure size 432x288 with 1 Axes>"
                        ]
                    },
                    "metadata": {
                        "needs_background": "light"
                    }
                },
                {
                    "output_type": "stream",
                    "name": "stdout",
                    "text": [
                        "AUC score: 0.9516893598001501\n"
                    ]
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "markdown",
            "source": [
                "# Neural Network\r\n"
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 21,
            "source": [
                "# Create neural network object\r\n",
                "clf_NN = tf.keras.models.Sequential([\r\n",
                "    tf.keras.layers.Dense(100,activation='relu',input_shape=(7,)),\r\n",
                "    tf.keras.layers.Dropout(0.1), #prevention of overfitting\r\n",
                "    tf.keras.layers.Dense(30,activation='relu'),\r\n",
                "    tf.keras.layers.Dense(2,activation='softmax')\r\n",
                "    ])\r\n",
                "\r\n",
                "# Set up and train neural network\r\n",
                "opt = keras.optimizers.Adam(learning_rate=0.01)\r\n",
                "clf_NN.compile(loss='sparse_categorical_crossentropy', optimizer=opt, metrics=['accuracy'])\r\n",
                "history = clf_NN.fit(train_features,train_labels, epochs=50, batch_size=50,validation_data=(test_features, test_labels))\r\n",
                "clf_NN.evaluate(test_features, test_labels)\r\n",
                "\r\n",
                "# Predict the response for the test dataset\r\n",
                "y_pred_NN = clf_NN.predict(test_features,batch_size=1,verbose=0)\r\n",
                "\r\n",
                "weights = clf_NN.get_weights() #returns a numpy list of weights"
            ],
            "outputs": [
                {
                    "output_type": "stream",
                    "name": "stdout",
                    "text": [
                        "Epoch 1/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.4710 - accuracy: 0.7745 - val_loss: 0.4315 - val_accuracy: 0.7901\n",
                        "Epoch 2/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.4161 - accuracy: 0.7962 - val_loss: 0.3881 - val_accuracy: 0.8001\n",
                        "Epoch 3/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3876 - accuracy: 0.8117 - val_loss: 0.3945 - val_accuracy: 0.8247\n",
                        "Epoch 4/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3760 - accuracy: 0.8233 - val_loss: 0.3539 - val_accuracy: 0.8355\n",
                        "Epoch 5/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3600 - accuracy: 0.8302 - val_loss: 0.3381 - val_accuracy: 0.8478\n",
                        "Epoch 6/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3523 - accuracy: 0.8357 - val_loss: 0.3359 - val_accuracy: 0.8423\n",
                        "Epoch 7/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3528 - accuracy: 0.8352 - val_loss: 0.3169 - val_accuracy: 0.8538\n",
                        "Epoch 8/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3458 - accuracy: 0.8406 - val_loss: 0.3338 - val_accuracy: 0.8521\n",
                        "Epoch 9/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3421 - accuracy: 0.8414 - val_loss: 0.3504 - val_accuracy: 0.8382\n",
                        "Epoch 10/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3332 - accuracy: 0.8451 - val_loss: 0.3129 - val_accuracy: 0.8574\n",
                        "Epoch 11/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3318 - accuracy: 0.8464 - val_loss: 0.3047 - val_accuracy: 0.8640\n",
                        "Epoch 12/50\n",
                        "634/634 [==============================] - 1s 968us/step - loss: 0.3278 - accuracy: 0.8488 - val_loss: 0.3063 - val_accuracy: 0.8659\n",
                        "Epoch 13/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3337 - accuracy: 0.8485 - val_loss: 0.3155 - val_accuracy: 0.8562\n",
                        "Epoch 14/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3200 - accuracy: 0.8532 - val_loss: 0.3252 - val_accuracy: 0.8463\n",
                        "Epoch 15/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3237 - accuracy: 0.8534 - val_loss: 0.3217 - val_accuracy: 0.8549\n",
                        "Epoch 16/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3225 - accuracy: 0.8546 - val_loss: 0.3205 - val_accuracy: 0.8559\n",
                        "Epoch 17/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3173 - accuracy: 0.8585 - val_loss: 0.2950 - val_accuracy: 0.8669\n",
                        "Epoch 18/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3184 - accuracy: 0.8565 - val_loss: 0.3234 - val_accuracy: 0.8632\n",
                        "Epoch 19/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3155 - accuracy: 0.8568 - val_loss: 0.3178 - val_accuracy: 0.8538\n",
                        "Epoch 20/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3113 - accuracy: 0.8608 - val_loss: 0.2880 - val_accuracy: 0.8712\n",
                        "Epoch 21/50\n",
                        "634/634 [==============================] - 1s 998us/step - loss: 0.3089 - accuracy: 0.8619 - val_loss: 0.2939 - val_accuracy: 0.8678\n",
                        "Epoch 22/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3165 - accuracy: 0.8586 - val_loss: 0.2873 - val_accuracy: 0.8696\n",
                        "Epoch 23/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3072 - accuracy: 0.8622 - val_loss: 0.2799 - val_accuracy: 0.8726\n",
                        "Epoch 24/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3042 - accuracy: 0.8623 - val_loss: 0.2838 - val_accuracy: 0.8728\n",
                        "Epoch 25/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3113 - accuracy: 0.8595 - val_loss: 0.2930 - val_accuracy: 0.8713\n",
                        "Epoch 26/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3027 - accuracy: 0.8640 - val_loss: 0.2841 - val_accuracy: 0.8764\n",
                        "Epoch 27/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3086 - accuracy: 0.8621 - val_loss: 0.2884 - val_accuracy: 0.8725\n",
                        "Epoch 28/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3019 - accuracy: 0.8649 - val_loss: 0.2792 - val_accuracy: 0.8761\n",
                        "Epoch 29/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3021 - accuracy: 0.8646 - val_loss: 0.2888 - val_accuracy: 0.8746\n",
                        "Epoch 30/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3032 - accuracy: 0.8637 - val_loss: 0.2883 - val_accuracy: 0.8741\n",
                        "Epoch 31/50\n",
                        "634/634 [==============================] - 1s 993us/step - loss: 0.3032 - accuracy: 0.8645 - val_loss: 0.2986 - val_accuracy: 0.8651\n",
                        "Epoch 32/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3036 - accuracy: 0.8671 - val_loss: 0.2873 - val_accuracy: 0.8732\n",
                        "Epoch 33/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.3021 - accuracy: 0.8657 - val_loss: 0.2802 - val_accuracy: 0.8735\n",
                        "Epoch 34/50\n",
                        "634/634 [==============================] - 1s 964us/step - loss: 0.2981 - accuracy: 0.8687 - val_loss: 0.2929 - val_accuracy: 0.8720\n",
                        "Epoch 35/50\n",
                        "634/634 [==============================] - 1s 985us/step - loss: 0.3010 - accuracy: 0.8681 - val_loss: 0.2843 - val_accuracy: 0.8754\n",
                        "Epoch 36/50\n",
                        "634/634 [==============================] - 1s 987us/step - loss: 0.3014 - accuracy: 0.8675 - val_loss: 0.2727 - val_accuracy: 0.8829\n",
                        "Epoch 37/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.2986 - accuracy: 0.8695 - val_loss: 0.2771 - val_accuracy: 0.8775\n",
                        "Epoch 38/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.2993 - accuracy: 0.8667 - val_loss: 0.2971 - val_accuracy: 0.8684\n",
                        "Epoch 39/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.2968 - accuracy: 0.8689 - val_loss: 0.2902 - val_accuracy: 0.8740\n",
                        "Epoch 40/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.2986 - accuracy: 0.8674 - val_loss: 0.2886 - val_accuracy: 0.8699\n",
                        "Epoch 41/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.2957 - accuracy: 0.8683 - val_loss: 0.2995 - val_accuracy: 0.8670\n",
                        "Epoch 42/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.2937 - accuracy: 0.8688 - val_loss: 0.2713 - val_accuracy: 0.8833\n",
                        "Epoch 43/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.2914 - accuracy: 0.8729 - val_loss: 0.2742 - val_accuracy: 0.8828\n",
                        "Epoch 44/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.2959 - accuracy: 0.8702 - val_loss: 0.2761 - val_accuracy: 0.8799\n",
                        "Epoch 45/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.2973 - accuracy: 0.8686 - val_loss: 0.2842 - val_accuracy: 0.8783\n",
                        "Epoch 46/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.2932 - accuracy: 0.8717 - val_loss: 0.2806 - val_accuracy: 0.8759\n",
                        "Epoch 47/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.2973 - accuracy: 0.8694 - val_loss: 0.2810 - val_accuracy: 0.8792\n",
                        "Epoch 48/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.2933 - accuracy: 0.8732 - val_loss: 0.2831 - val_accuracy: 0.8751\n",
                        "Epoch 49/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.2910 - accuracy: 0.8734 - val_loss: 0.2867 - val_accuracy: 0.8749\n",
                        "Epoch 50/50\n",
                        "634/634 [==============================] - 1s 1ms/step - loss: 0.2906 - accuracy: 0.8711 - val_loss: 0.2727 - val_accuracy: 0.8838\n",
                        "248/248 [==============================] - 0s 658us/step - loss: 0.2727 - accuracy: 0.8838\n"
                    ]
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 22,
            "source": [
                "# Predicted values as integers\r\n",
                "y_pred_NN = np.argmax(clf_NN.predict(test_features, batch_size=1, verbose=0), axis=-1)\r\n",
                "print(y_pred_NN)"
            ],
            "outputs": [
                {
                    "output_type": "stream",
                    "name": "stdout",
                    "text": [
                        "[0 0 0 ... 0 1 0]\n"
                    ]
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 23,
            "source": [
                "# Get the metrics for the Neural Network\r\n",
                "print(classification_report(test_labels,y_pred_NN))"
            ],
            "outputs": [
                {
                    "output_type": "stream",
                    "name": "stdout",
                    "text": [
                        "              precision    recall  f1-score   support\n",
                        "\n",
                        "           0       0.90      0.93      0.91      5148\n",
                        "           1       0.86      0.80      0.83      2771\n",
                        "\n",
                        "    accuracy                           0.88      7919\n",
                        "   macro avg       0.88      0.86      0.87      7919\n",
                        "weighted avg       0.88      0.88      0.88      7919\n",
                        "\n"
                    ]
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 24,
            "source": [
                "# Get macro average F1 score\r\n",
                "f1_NN = f1_score(test_labels, y_pred_NN, average='macro')\r\n",
                "f1_NN"
            ],
            "outputs": [
                {
                    "output_type": "execute_result",
                    "data": {
                        "text/plain": [
                            "0.870230256384607"
                        ]
                    },
                    "metadata": {},
                    "execution_count": 24
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 35,
            "source": [
                "# Calculate Matthews Correlation coefficient (for future use)\r\n",
                "MCC_NN = matthews_corrcoef(test_labels, y_pred_NN)\r\n",
                "MCC_NN"
            ],
            "outputs": [
                {
                    "output_type": "execute_result",
                    "data": {
                        "text/plain": [
                            "0.7416407827369598"
                        ]
                    },
                    "metadata": {},
                    "execution_count": 35
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 26,
            "source": [
                "# Plot the confusion matrix for the Neural Network\r\n",
                "cf_matrix(test_labels,y_pred_NN)"
            ],
            "outputs": [
                {
                    "output_type": "stream",
                    "name": "stdout",
                    "text": [
                        "<function cf_matrix at 0x000001BDDDED1280>\n"
                    ]
                },
                {
                    "output_type": "display_data",
                    "data": {
                        "image/png": "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",
                        "text/plain": [
                            "<Figure size 432x288 with 2 Axes>"
                        ]
                    },
                    "metadata": {
                        "needs_background": "light"
                    }
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 34,
            "source": [
                "# Plot the ROC curve and get the AUC score for the Neural Network\r\n",
                "plot_roc_curve(clf_NN)"
            ],
            "outputs": [
                {
                    "output_type": "display_data",
                    "data": {
                        "image/png": "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",
                        "text/plain": [
                            "<Figure size 432x288 with 1 Axes>"
                        ]
                    },
                    "metadata": {
                        "needs_background": "light"
                    }
                },
                {
                    "output_type": "stream",
                    "name": "stdout",
                    "text": [
                        "AUC score: 0.8645716176842123\n"
                    ]
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "markdown",
            "source": [
                "# Balanced Random Forest\r\n"
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 36,
            "source": [
                "# Create a Balanced Random Forest classifier object\r\n",
                "clf_BRF = BalancedRandomForestClassifier(n_estimators=100, random_state=42)\r\n",
                "\r\n",
                "# Train the Balanced Random Forest Classifier\r\n",
                "clf_BRF = clf_BRF.fit(train_features,train_labels)\r\n",
                "\r\n",
                "# Predict the response for the test dataset\r\n",
                "y_pred_BRF = clf_BRF.predict(test_features)"
            ],
            "outputs": [],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 37,
            "source": [
                "# Get the metrics for the Balanced Random Forest\r\n",
                "print(classification_report(test_labels,y_pred_BRF))"
            ],
            "outputs": [
                {
                    "output_type": "stream",
                    "name": "stdout",
                    "text": [
                        "              precision    recall  f1-score   support\n",
                        "\n",
                        "           0       0.98      0.97      0.98      5148\n",
                        "           1       0.95      0.96      0.96      2771\n",
                        "\n",
                        "    accuracy                           0.97      7919\n",
                        "   macro avg       0.96      0.97      0.97      7919\n",
                        "weighted avg       0.97      0.97      0.97      7919\n",
                        "\n"
                    ]
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 38,
            "source": [
                "# Get macro average F1 score\r\n",
                "f1_BRF = f1_score(test_labels, y_pred_BRF, average='macro')\r\n",
                "f1_BRF"
            ],
            "outputs": [
                {
                    "output_type": "execute_result",
                    "data": {
                        "text/plain": [
                            "0.9654635653916577"
                        ]
                    },
                    "metadata": {},
                    "execution_count": 38
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 39,
            "source": [
                "# Calculate Matthews Correlation coefficient (for future use)\r\n",
                "MCC_BRF = matthews_corrcoef(test_labels, y_pred_BRF)\r\n",
                "MCC_BRF"
            ],
            "outputs": [
                {
                    "output_type": "execute_result",
                    "data": {
                        "text/plain": [
                            "0.9310423462283628"
                        ]
                    },
                    "metadata": {},
                    "execution_count": 39
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 40,
            "source": [
                "# Plot the confusion matrix for the Balanced Random Forest\r\n",
                "cf_matrix(test_labels,y_pred_BRF)"
            ],
            "outputs": [
                {
                    "output_type": "stream",
                    "name": "stdout",
                    "text": [
                        "<function cf_matrix at 0x000001BDDDED1280>\n"
                    ]
                },
                {
                    "output_type": "display_data",
                    "data": {
                        "image/png": "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",
                        "text/plain": [
                            "<Figure size 432x288 with 2 Axes>"
                        ]
                    },
                    "metadata": {
                        "needs_background": "light"
                    }
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 41,
            "source": [
                "# Plot the ROC curve and get the AUC score for the Balanced Random Forest\r\n",
                "plot_roc_curve(clf_BRF)"
            ],
            "outputs": [
                {
                    "output_type": "display_data",
                    "data": {
                        "image/png": "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",
                        "text/plain": [
                            "<Figure size 432x288 with 1 Axes>"
                        ]
                    },
                    "metadata": {
                        "needs_background": "light"
                    }
                },
                {
                    "output_type": "stream",
                    "name": "stdout",
                    "text": [
                        "AUC score: 0.9676371535357462\n"
                    ]
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "markdown",
            "source": [
                "# Random Forest\r\n"
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 42,
            "source": [
                "# Create a Random Forest classifier object with best parameters found via grid search\r\n",
                "clf_RF = RandomForestClassifier(n_estimators=48, bootstrap=False, max_depth=30, min_samples_leaf=1, min_samples_split=2, random_state=0)\r\n",
                "\r\n",
                "# Train the Random Forest Classifier\r\n",
                "clf_RF = clf_RF.fit(train_features, train_labels)\r\n",
                "\r\n",
                "# Predict the response for the test dataset\r\n",
                "y_pred_RF = clf_RF.predict(test_features)"
            ],
            "outputs": [],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 43,
            "source": [
                "# Get the metrics for the Random Tree\r\n",
                "print(classification_report(test_labels,y_pred_RF))"
            ],
            "outputs": [
                {
                    "output_type": "stream",
                    "name": "stdout",
                    "text": [
                        "              precision    recall  f1-score   support\n",
                        "\n",
                        "           0       0.98      0.99      0.98      5148\n",
                        "           1       0.97      0.95      0.96      2771\n",
                        "\n",
                        "    accuracy                           0.98      7919\n",
                        "   macro avg       0.97      0.97      0.97      7919\n",
                        "weighted avg       0.98      0.98      0.98      7919\n",
                        "\n"
                    ]
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 44,
            "source": [
                "# Get macro average F1 score\r\n",
                "f1_RF = f1_score(test_labels, y_pred_RF, average='macro')\r\n",
                "f1_RF"
            ],
            "outputs": [
                {
                    "output_type": "execute_result",
                    "data": {
                        "text/plain": [
                            "0.9725230691977741"
                        ]
                    },
                    "metadata": {},
                    "execution_count": 44
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 45,
            "source": [
                "# Calculate Matthews Correlation coefficient (for future use)\r\n",
                "MCC_RF = matthews_corrcoef(test_labels, y_pred_RF)\r\n",
                "MCC_RF"
            ],
            "outputs": [
                {
                    "output_type": "execute_result",
                    "data": {
                        "text/plain": [
                            "0.9451775126239347"
                        ]
                    },
                    "metadata": {},
                    "execution_count": 45
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 46,
            "source": [
                "# Plot the confusion matrix for the Random Forest\r\n",
                "cf_matrix(test_labels,y_pred_RF)"
            ],
            "outputs": [
                {
                    "output_type": "stream",
                    "name": "stdout",
                    "text": [
                        "<function cf_matrix at 0x000001BDDDED1280>\n"
                    ]
                },
                {
                    "output_type": "display_data",
                    "data": {
                        "image/png": "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",
                        "text/plain": [
                            "<Figure size 432x288 with 2 Axes>"
                        ]
                    },
                    "metadata": {
                        "needs_background": "light"
                    }
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 47,
            "source": [
                "# Plot the ROC curve and get the AUC score for the Random Forest\r\n",
                "plot_roc_curve(clf_RF)"
            ],
            "outputs": [
                {
                    "output_type": "display_data",
                    "data": {
                        "image/png": "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",
                        "text/plain": [
                            "<Figure size 432x288 with 1 Axes>"
                        ]
                    },
                    "metadata": {
                        "needs_background": "light"
                    }
                },
                {
                    "output_type": "stream",
                    "name": "stdout",
                    "text": [
                        "AUC score: 0.9702020131919086\n"
                    ]
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "markdown",
            "source": [
                "# Extra trees\r\n",
                "\r\n",
                "*Additional model that is not featured in the article. However, it looks very promising and is likely to be considered in the future research.*\r\n"
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 48,
            "source": [
                "# Create an Extra Trees classifier object\r\n",
                "clf_ET = ExtraTreesClassifier(n_estimators=250, n_jobs=-1, max_features=1)\r\n",
                "\r\n",
                "# Train Extra Trees Classifier\r\n",
                "clf_ET = clf_ET.fit(train_features,train_labels)\r\n",
                "\r\n",
                "# Predict the response for the test dataset\r\n",
                "y_pred_ET = clf_ET.predict(test_features)"
            ],
            "outputs": [],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 49,
            "source": [
                "# Get the metrics for the Extra Trees\r\n",
                "print(classification_report(test_labels,y_pred_ET))"
            ],
            "outputs": [
                {
                    "output_type": "stream",
                    "name": "stdout",
                    "text": [
                        "              precision    recall  f1-score   support\n",
                        "\n",
                        "           0       0.97      0.99      0.98      5148\n",
                        "           1       0.98      0.95      0.96      2771\n",
                        "\n",
                        "    accuracy                           0.98      7919\n",
                        "   macro avg       0.98      0.97      0.97      7919\n",
                        "weighted avg       0.98      0.98      0.98      7919\n",
                        "\n"
                    ]
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 50,
            "source": [
                "# Get macro average F1 score\r\n",
                "f1_ET = f1_score(test_labels, y_pred_ET, average='macro')\r\n",
                "f1_ET"
            ],
            "outputs": [
                {
                    "output_type": "execute_result",
                    "data": {
                        "text/plain": [
                            "0.9733000427879573"
                        ]
                    },
                    "metadata": {},
                    "execution_count": 50
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 51,
            "source": [
                "# Calculate Matthews Correlation coefficient (for future use)\r\n",
                "MCC_ET = matthews_corrcoef(test_labels, y_pred_ET)\r\n",
                "MCC_ET"
            ],
            "outputs": [
                {
                    "output_type": "execute_result",
                    "data": {
                        "text/plain": [
                            "0.9468742360079958"
                        ]
                    },
                    "metadata": {},
                    "execution_count": 51
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 52,
            "source": [
                "# Plot the confusion matrix for the Extra Trees\r\n",
                "cf_matrix(test_labels,y_pred_ET)"
            ],
            "outputs": [
                {
                    "output_type": "stream",
                    "name": "stdout",
                    "text": [
                        "<function cf_matrix at 0x000001BDDDED1280>\n"
                    ]
                },
                {
                    "output_type": "display_data",
                    "data": {
                        "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAEgCAYAAAC6tbgBAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAA8vUlEQVR4nO3dd5wURfrH8c93A0lAkiAIkkQ5zBnDKWbMOQeMnFlPPfMpnqdn/p2eomLELGbMhwgiBpKJoCiniEgUBAkLbHh+f1TvMiwzs7OwOzM7PG9f/ZqZ6uru6ll8uqa6ukpmhnPOudyVl+kCOOecq10e6J1zLsd5oHfOuRzngd4553KcB3rnnMtxHuidcy7HeaB3q5F0saRJkookmaRL03DMqZKm1vZx1gWSOkV/tyczXRaXHTzQZ5ik7pL+I2mCpIWSVkiaIeltSWdJapDm8pwA3AssA/4N3AR8ns4yZJMoYJqkMkldk+QbFpP39LU85uk1sR/nyhVkugDrMkk3ADcSLrifAwOBxUAboBfwKHAesEMai3VI+auZzUjjcfdJ47Gqq4Tw/8pZwLWVV0rqBuwZky/TfgX+BCzMdEFcdsiGf5TrJEnXEmrLvwDHmtmoOHkOAS5Pc9HaAaQ5yGNm/0vn8appNjATOEPSDWZWUmn92YCAt4Aj0ly21ZhZMfBdpsvhsoc33WSApE5AP6AYOChekAcws7eA3pW2PU7SiKiZp0jSeEnXSKof5zhTo6WRpDslTZO0XNIUSVdJUkzefpIM2Cv6XN4MYeVlTtbuK2l4ed6YNEnqI+lTSXMlLZP0i6T3JR0fr6xx9ltf0tWSvpG0VNIfkj6WdFy877W8jNH7FyT9Fh13bHThXFOPABuy8hdP+TELgT7Ap8DEeBtK2l7SvZK+ljQ/Ks8Pku6W1LxS3uHAE9HHJ2L/DtG/m4q/laRekk6SNErS4vLvL97fStLOUbPgj5LWr3TMtpJmR/vovsbfkMtaXqPPjDOAQuAFM5uQLKOZLS9/L+lW4BrgN+A5QjPPgcCtwAGS9otqc7EKgf8SaurvEpoXjgBuAxoQflUADI9eTwc6xqSvjVui8v4EDCI0JbQFdgSOBV5MtrGkesD7hGaR74AHgEbAMcCLkrYxs9WaUqLyjwZ+BJ4GWgDHA29I2tfMhq3BuTwP3EOovb8ek34YoantamCTBNueAxwJfAR8AOQD2wGXAQdK2tnMFkV5nwQWAIcDbwBfxexnQaX9Xg7sB7wJDAPWJwEzGxX9iryTcNE6DkBSHvAM0Bo43cz8l0AuMjNf0rwAQwEDzq7GNrtE20wDNoxJLyD8j27AtZW2mRqlvwM0jElvTQgaC4DCStsMD/8sVjt+p2hfTyYo32rbAfOA6UCjOPlbxSnr1Epp18SUv6BS+cvPbdc4ZTTgxkr7OqB8X9X8WxkwPXr/KOFC2T5m/XuEC1gj4J9R/tMr7aMjkB9n32dF+a+qlH56vP3ErO8XrV8CbJvq34qVzUsG/CVKuzH6/FSm/7/wpfYWb7rJjLbR6/RqbHNm9PpPM5tVnmihvfhyoIxQ24znYjMritlmDqG2uD6wWTXKsCaKgdLKiWb2WwrbnkkIQpdZTLt4VP6bo4/xzvlnQtCNPd77hIvkTqkVO65HCLXxMwEkdSTUqJ81s6WJNjKzn81ste8AeBz4g3ARWhMDzOzLVDObmREuIr8C/5Z0PvB3YDLhpr/LUd50kxnlbePVGSN6u+j1w8orzOx7SdOBzpKamdmCmNULzWxKnP39Er02j7OupjwLXARMlPQSoeniMzOrsjeIpCaEppBfLX5zQvn3sG2cdV8lCKy/EH4ZlR+jEyHwrcLM+sUrk4Xmj/HAmZL+SbjI5BEuAAlF7fh/AU4AehAusLGVrI2SbZ/E6OpuYGa/STqJ8P09QOhGe7yZLVnDMrg6wAN9ZswAugPtq7FNefvrzATrZwIbR/kWxKQviJeZ0AQBoYZaW/4K/I9QA746WkokvQNcnuACVC6V8wVoFmfdggTblLBqgO1EaLqorF+Scj0C3Ee4SX4GMC6FWvWLhDb6Hwm/pGYB5fdeLgVWu5GeollVZ4lrNOHXTWdgmJl9vYb7cXWEN91kxsjotTp9x8trwRsmWN+2Ur6aVha9JqocNKucYGalZnavmW1NuGF5NPAa4Qbme/F6CsWo9fM1s+FmpspLFZs9DRQBDxNq4gOSZZa0AyHIfwB0N7MzzOya6FfDP4B6a1p+qveLMNa9hCD/G+Fm8MlrUQZXB3igz4wnCG3XR0vqkSxjTDAsrzX2ipNnE8Kvg58qNdvUpN+j1w5xjt8U2DTZxmY2x8xeNbPjCM0GXYEtkuRfRPg1sFH0QFJle0WvX6RQ9hoTfb8vE77vJYTeOMmU98QZbKv3iNoJaBhnm/Jmpxr/tSXpWKAvMILQHDgXeCjBd+xyhAf6DDCzqYTmgXrA21GtbzWSehO6REK4cQdwvaQNYvLkA3cR/paP1VKRywPvd8BusRen6Pj3UClgRf3f94ntqx+lFxK6OwIkvIEZeZxwP+PO6Djl+2hFuIlYnifdrifU0g+wld0iE5kavfaKTZTUmtBGHs+86HXjNSxfXJK6EJqe5gEnm9kvwGnAeoTuqmvahOSynLfRZ4iZ3SqpgNBGPEbSp8BYVg6BsAfQLUrDzD6VdAdwJTBB0suEGuWBhJrxSEIf6dp0J+Fi8kl0c3UZoWZdCHwNbB2TtyGhuWKqpFGEnjANCL1U/kSo4X5bxfHuIpzf4cDXUdt+I0If/NbAHWY2Msn2tcLMphHauFMxBvgEOCr6G48k/H0PJPR2ifcE8meEi+ClkloQnswF+E8qN7LjiS6wLxDufRxuZtOjc3lP0t3AFYTv+6I12b/Lcpnu37muL4Sg9x9gAqGr3QrCjcZ3Cf2s61fKfwIhWCwiBNqJwHVAgzj7nkqlvukx6/oR2nh7VUofTpx+9DHrz4qOuZxwM/BhoGXl7QjB/8roPKZFZZ1LGNPnXKBeKmUlXByujb6foui8RwInxsnbiWr29U/h71PRjz6FvIn60bcA+kfnuIzQJHUr4aKV6Lx7EwL+YlY+G9Ap2d8u2fcA3B2l3RsnfyEwKlp/ZKb/n/Cl5hdFf2jnnHM5ytvonXMux3mgd865HOeB3jnncpwHeuecy3F1qntlw20v9DvHbjW/j7k/00VwWahBAVU95Vyl6sScoi/vX+vj1Rav0TvnXCLKS32paldhcp3xkr6SNDZKayFpSDQRzZDYiWgUJhSaImmypANi0reP9jNF0n2VH0qMxwO9c84lIqW+pGYvM9vGzMqfhr8aGGpm3QjzVFwdDqsehGdmNic8U9E/5unwBwnDWHSLllVmoYvHA71zziVSgzX6BA4HBkbvB7JyzuHDCTPQLTezn4ApwE6S2gJNzewzCw9BPUUK8xR7oHfOuUSqUaOX1Deam7h86Vtpbwb8V9K4mHVtzGwmQPTaOkrfiJVzRkCYpGijaJkeJz2pOnUz1jnn0iov9QFEzWwAyYet3s3MZkQD2g2RlGx+3nhtQZYkPSmv0TvnXCI12HRjZjOi1zmEeRl2AmZHzTFEr3Oi7NNZdUjw9oQB8Kaz6oRF5elJeaB3zrlEauhmrKT1oukxkbQesD9hoL7BQJ8oWx/CDGRE6SdEw313Jtx0HR017yyS1DPqbXNazDYJedONc84lsuY3WStrA7wW9YQsAJ6zMET0GGCQpLMIo7weC2BmEyUNAiYRpsC8wFbOg3we8CRhKPB3WTlnRUIe6J1zLpHUu00mZWY/sup8DeXp80gwpaiZ3QLcEid9LElmZ4vHA71zziVSczX6jPJA75xziVSj100280DvnHOJeI3eOedyXF7WjlNWLR7onXMuEa/RO+dcjquhXjeZ5oHeOecS8ZuxzjmX47zpxjnncpw33TjnXI7zGr1zzuU4r9E751yO8xq9c87lOO9145xzOc5r9M45l+O8jd4553Kc1+idcy7H5UiNPq2XK0l5ko5L5zGdc26N1eDk4JmU1tKZWRlwYTqP6Zxza0p5eSkv2SwTTTdDJF0BvAgsKU80s/kZKItzziWkHGm6yUSgPzN6vSAmzYAuGSiLc84llhtxPv2B3sw6p/uYzjm3JnKlRp/2hiVJjSRdL2lA9LmbpEPSXQ7nnKuKpJSXbJaJOwhPACuAXaPP04F/ZqAczjmXVF5eXspLNstE6bqa2R1AMYCZFZEzLWHOuZyiaixZLBM3Y1dIaki4AYukrsDyDJTDOeeSyvYmmVRlItD3A94DOkh6FtgNOD0D5XDOuaQ80K8hM/uvpHFAT8IPnkvM7Ld0l8M556rigX4NSRoMPA8MNrMlVeV3zrlMyZVAn4mbsXcDfwYmSXpJ0jGSGmSgHM45l5TylPKSzTLRdPMR8JGkfGBv4BzgcaBpusvinHPJ5EqNPiPDFEe9bg4Fjge2AwZmohzOOZeMB/o1JOlFYGdCz5sHgOHRqJbOOZddciPOZ6RG/wRwkpmVZuDYzjmXslyp0WfiZuxQ4AJJL0fLRZIKM1AO55xLqqbHupGUL+lLSW9Fn1tIGiLph+i1eUzeayRNkTRZ0gEx6dtLGh+tu08pHDwTNfoHgUKgf/T51Cjt7AyUJat89/ZNLFqynNKyMkpKy9j95Dto3rQRT99+Jh3bteDnGfM55crHWLCoiMKCfO6//kS267ExZVbGFXe8wsfjfgDg/UcuYcNWTSlaXgzAoefdz9zfF2fy1FwtOXC/vWm03nrk5+WRX5DP84Ne5f77/s3wYUPJUx7NW7bk5lv+RevWbTJd1DqpFsawuQT4lpWdT64GhprZbZKujj5fJakHcAKwOdAO+EDSplFLyINAX+Bz4B2gN/BusoNmItDvaGZbx3z+UNLXGShHVurd917mLVj5eMEVZ+zH8NGTueuJIVxxxn5cccb+XH/fG5x51G4A7HjcrWzQvDGv338+u59yJ2YGwBnXDeSLSdMycg4uvR59YiDNm7eo+Hz6mWdz4cWXAvDsM0/x8IMP8Pcb/5Gh0tVxNdhyI6k9cDBwC3BZlHw40Ct6PxAYDlwVpb9gZsuBnyRNAXaSNBVoamafRft8CjiCKgJ9JppuSqPxbQCQ1AXw9voEDum1Fc+8OQqAZ94cxaF7bQVA9y4bMmz0ZADm/r6YhYuK2L7Hxhkrp8sejRs3rni/rKgoZ9qZM6GGm27+DVwJxHY+aWNmMwGi19ZR+kbALzH5pkdpG0XvK6cnlYka/d+AYZJ+JFwvOwJnZKAcWcfMeLP/hZgZj73yCY+/+gmtWzZh1m9/ADDrtz/YoEUTAMZ//yuH9tqSl94fR/s2zdm2Rwfab9icsRN/BuDhfqdQWlbG60O/4rZH3svYOblaJjj3nLOQxDHHHs8xxx0PwH/u/T/eHPw6jRs34dEnnspwIeuu6lwkJfUlNKmUG2Bm5fNuHALMMbNxknqlsrs4aZYkPalMPDA1VFI3YDNCob+Lfp6s8/Y+4/+YOXchGzRvzFsPXcjkqbMS5h34xmd079yGT569kmkz5/P51z9RUhp+GJ1x7ZPMmLuQxo3q8/xdZ3PSITvx3Fuj03UaLo0GPvM8rVu3Yd68eZx79hl07tKF7XfYkYsu+SsXXfJXHnvkYV547hnOv/DiTBe1TqpOoI+C+oAEq3cDDpN0ENAAaCrpGWC2pLZmNlNSW2BOlH860CFm+/bAjCi9fZz0pDIywxShVn+RmX0NbOwzTAUz5y4EQlPM4A+/YcfNOzFn3iI2bBXu22zYqilz5y8CoLS0jCvvfpWeJ9zGcX8dQLMmDZkybS4AM6L9LF66nBffHcuOm3fMwNm4dCi/ydqyZUv23nc/Joz/ZpX1Bx58CB8M+W8mipYTamoIBDO7xszam1knwk3WD83sFGAw0CfK1gd4I3o/GDhBUn1JnYFuwOioeWeRpJ5Rb5vTYrZJKJMzTO0SffYZpoBGDerRuFH9ivf77tKdif+bwdsfjeeUQ3cG4JRDd+at4eF/5IYNCmnUoB4Ae+/cnZLSMr77cRb5+Xm0bLYeAAUFeRy0xxZM/N/MDJyRq21Lly5lyZLFFe8/+/QTNtmkGz//PLUiz/BhH9K5c5cMlbDuS8NUgrcB+0n6Adgv+oyZTQQGAZMID5deEPPs0XnAo8AU4H9UcSMWMtNG39XMjpd0IoQZplLpB5rrWrdswov3nANAQX4+L747liGffsu4idN45vYz6XPELvwy83dOvvIxADZo3oQ3+19AWZkxY+4Czro+jCJRv7CAwQ9cQGFBPvn5eQwb9R2Pv/pJxs7L1Z758+bx14svAKCktJSDDj6E3f68B5ddchFTp/5EXp5o23Yjrr/xpgyXtO6qjdBkZsMJvWsws3nAPgny3ULooVM5fSywRXWOqfLueOki6VPCiX1iZttFPXCeN7Odqtq24bYXprewrk74fcz9mS6Cy0INCta+c2TnS99OOeb89O+Ds7bCmoka/Y34DFPOuboga0N39WSi180QSV+Q4gxTsV2WCtr3oqDV5ukpqHNunZcrrcqZ6HWzHaHv/ExCt6CNJXWVFPeiY2YDzGwHM9shV4L8+o0b8tydZ/HVq9fz5SvXs/NWnSvWXXrqPhR9eX/FDdVY9esV8PHTVzDqxasZ9/J1XH/uQavlqbz9Llt3YfSL1zDymb/RpUOriuMPfuCCWjo7VxM++XgEhx18AIf03o/HHonfY2/M6FEcd9ThHHnYwZzZ5xQApv70I8cddXjFsutO2/HMU08C8H9338kxRx7KdddcWbGPNwe/zrNP+yjhieTlKeUlm2Wi6aY/YQz6bwg1+i2i9y0lnWtmOd8X7K4rj+G/n07ipL89RmFBfkXvmfZtmrF3z+5Mmzk/7nbLV5TQu+99LClaQUFBHh8+fhn//WQSo8dPTbj9JafuzYl/e5SObVvS99g/c/U9r3FN397c8fj7tX6ebs2UlpZy6y3/4OFHnqBNmzacdPwx9Nprb7pusklFnj/++INbb76J/g8/Stt27Zg3bx4AnTp3YdCrb1TsZ7+99mDvffdj0aJFfP3Vl7z82ptcc+Xl/PD9ZDps3JHBr79G/4cfzch51gVeo19zU4Fto1r69sC2wARgX+CODJQnrZqs14Ddt+vKk699BkBxSSkLFxcBcMcVR3Pdva+T7Ab5kqIVABQW5FNQkL9K3njbF5eU0rB+IY0aFlJcUkrn9q1o17oZI8dNqY3TczVgwvhv6NChI+07dKCwXj16H3Qww4cNXSXPu2+/yT777kfbdu2A0I++slGff0aHDh1o124j8vJEcXExZsay5cspKCjgyccf5aRTTqWw0AePTURKfclmmajRd4/6iAJgZpMkbWtmP+bK1TOZzhu15LffFzPgplPYctON+PLbX7jijpfZa+fNmDFnAeO//zXp9nl54tPnrqJrhw14+MURjJkQhjw4eM8t425/5+P/5YHrT6RoeTFnXf8U/7rsSG7q/1atnZ9be3Nmz2bDthtWfG7dpg3jv1n1Qaifp06lpKSEs04/lSVLlnDyKadx6OFHrJLnvXffpvdB4VnE9dZrzL777c/xRx/BTj13oXGTJkycMIFzz7+w1s+nLsuVmJSJQD9Z0oPAC9Hn44HvJdUHijNQnrQqKMhnm+4duOz2lxgz4Wfu+tvRXH/uQey+3SYccn7V3QTLyoyeJ9zG+o0b8uI959Cja1t++vU3rjrrgLjbf/P9r+zZ524AdtuuKzPnLkSIp287g+KSUq6+5zXmRE/buuxgcYYuqRxwSkpLmTRpIgMee5Lly5dx2kknsOXWW9OpU7jfU7xiBR8N+5BLLr28YpszzjqHM84Kz2r0u+E6zr/oYl59+SU++3Qk3TbdjL7nnl+LZ1U35Uicz0jTzemEJ7ouBf4K/BilFQN7ZaA8afXr7N/5dc6Cipr4ax98xTbdO9Bxo5aMfvEavnv7JjZq3YzPnruKNi2bJNzPwsVFjBj7A/vv2oMu7TdIafurz+7Nvwa8y3V/OZCbH3qH598Zw/kn9qrN03VroE2bDZk1c+U4R3Nmz6Z169ar5dlt9z/TqFEjmjdvwXY77MD3k7+rWD9y5Ai699iclq1arbb/b7+dBEDHjp14c/Dr3HnPvUyZ8sMqT9S6IFduxqY90JtZkZndbWZHmtkRZnaXmS01szIzy/nZMWbPW8T0Wb/TrWP4H7fXTpvx1Xe/0HGfa+h+8I10P/hGfp2zgF1Oup3Z81atabdq3pj1GzcEoEH9QvbeeTMmT53NxCkzqtz+lEN35r2PJ7JgURGNGtSjrMwoKzMaNfD22Wyz+RZbMm3aVKZP/4XiFSt475232XOvvVfJs9fe+/DFuLGUlJRQVFTE+G++oXOXitG/efedtznwoIPj7v+B/9zL+RdeTElJCWXRQHh5ymNZ0bLaO6k6KlcCfSYmB98N6EfoYllxfDNbZwbkuOz2l3ji1tOpV5DP1F9/o++NzyTM23aD9el/w0kcedGDbNiqKY/841Ty8/LIyxOvDPmCdz+eUOXxGjYo5JRDd65o2rnvmQ95/q6zWVFcQp9rnqyp03I1pKCggGuuu4Hz+p5NWVkpRxx5NJts0o1BLz4PwHHHn0iXrl3Zbfc/c+yRh6G8PI46+hi6ddsUgKKiIj7/9NO4k418OPQDtthiy4rB0LbaZluOPuJQNt10Uzbr3j19J1lH5ErTTSaGQPiO0GQzjpgJR6IxH5LyIRBcPD4EgounJoZA2OqGD1KOOd/8Y9+svSxk4mbsQjOrcrQ155zLNO91s+aGSboTeBWomHDEzL7IQFmccy6hHInzGQn0O0evO8SkGbB3nLzOOZcx2X6TNVWZGNQs57tQOudygzfdrAVJBwObE+ZOBMDMVu8i4JxzGZQjcT4j3SsfAhoRHo56FDgG8JmrnXNZJ1dq9Jl4MnZXMzsN+N3MbiLMHduhim2ccy7tfFCzNVcUvS6V1A6YB3ROkt855zIiV2r0CQO9pEbV2ZGZLU0x61uSmgF3Al8Qetz4gNjOuayzLvS6WQxxhtFLLD+VTGZ2c/T2FUlvAQ3MbGE1juOcc2mRIxX6pIH+TKoX6FMmaVegU/nxJWFmT9XGsZxzbk3lfNONmT1ZGweU9DTQFfiKlWPdGOCB3jmXVXIkzlfvZqykHsD2hF4yj5vZLEmbALPNLNXZK3YAeli6R1NzzrlqyvkafSxJjYHHCX3ei6Pt3gNmAbcC04ArUjzmBGBDYGZ1C+ucc+m0TgV64B5gV2Af4BMgdoaCdwhBPmmgl/QmoYmmCTBJ0mhWHdTssNSL7ZxztW9d6HUT6yjgEjMbJqly75qfCZOIVGUw0Ab4uFL6nkDyGbGdcy4DcqRCn3Kgb0h4sCmeJsRMIJLE4cC1ZrbKdPaSlgA3Ao+lWBbnnEuLXGm6SXUIhDHAaQnWHQN8msI+OlUO8gBmNpbQ1dI557LKujYEwvXAB5I+AF4itLUfJOmvhEC/Rwr7aJBkXcMUy+Gcc2mTl+0RPEUp1ejNbCThRmx94H5AwE1AF2BfMxuTwm7GSDqncqKkswjzxzrnXFbJy1PKSzZLuR+9mX0C/FlSQ6A5sKAa49sAXAq8JulkVgb2HYB6wJHV2I9zzqVFlsfvlK3J6JXLCH3pi6rKGMvMZgO7StoL2CJKftvMPlyDMjjnXK3LlZuxKQd6SQcR2uq3j7YrkTQOuMXM3k51P2Y2DBhW3YI651y65UicT62NXtJfgDcJI1peAhwbvS4GBkfrnXMup6ga/yXdj9RA0mhJX0uaKOmmKL2FpCGSfohem8dsc42kKZImSzogJn17SeOjdfcphZ8dqXavvBYYYGb7m9lDZvZq9Lo/8AhwXYr7cc65OiNPqS9VWA7sbWZbA9sAvSX1BK4GhppZN2Bo9Ll8XLETCHNr9wb6xzys+iDQF+gWLb2rPI8Uz7cl8GqCda8ALVLcj3PO1Rk11evGgsXRx8JoMcKDpAOj9IHAEdH7w4EXzGy5mf0ETAF2ktQWaGpmn0UDQz4Vs03i80jxfIcRhiqIZ09gRIr7cc65OiNPSnmR1FfS2Jilb+y+JOVL+gqYAwwxs1FAGzObCRC9to6ybwT8ErP59Chto+h95fSkkk0l2CPm433Ao5JaAq9HBW1N6BZ5IHB2VQdyzrm6pjo3Y81sADAgyfpSYJtoKtXXJG2RKC/EbfS3JOlJJet1M6HSDgT8JVoqH/A9UpxK0Dnn6ora6F5pZgskDSe0rc+W1NbMZkbNMnOibNMJ836Uaw/MiNLbx0lPKlmg36saZXfOuZxTU3Fe0gZAcRTkGwL7ArcTRvXtA9wWvb4RbTIYeE7SPUA7wk3X0WZWKmlRdCN3FGEMsv9UdfxkUwl+tOan5ZxzdV9+zdXo2wIDo54zecAgM3tL0mfAoGgomGmEruuY2URJg4BJQAlwQdT0A3Ae8CRhjLB3oyWpaj8ZKymPOAOUVXM4BOecy3o11XQTjdy7bZz0eYRxxOJtcwtwS5z0sawcXSAlqU4lKOBK4Bygc4Js3kbvnMspuTLWTardKy8mdOR/jHAT9hbgH8D3wFRC533nnMspCt0mU1qyWaqB/hzCLFB3RJ9fN7ObCE9tfUe4UeCcczklVyYeSTXQdwa+im4GFAPNAMysDOhPuFvsnHM5ZV2r0c8DGkfvp7HqTYXm+AxRzrkclJ+nlJdslmqvm0+AHYF3gOeAfpJaACuACwiD8TjnXE7J7vCdulQDfT9WjqdwK6Hp5nRCTX4IcFENl8s55zIuV+aMTSnQm9lkYHL0fjlhLPpLarFczjmXcTkS51Nuo09I0tGSSqvO6ZxzdUuu3IxdkzljnXNunZDl8TtlHuidcy6BbO9NkyoP9M45l0C2N8mkqk4F+t9GVTkap1sH/fWNSZkugstCDx7do+pMVVjrm5hZItkMU4NS3Ef7qrM451zdsy7U6DdIcR/L8TljnXM5KEea6JNOPOIzTDnn1ml+M9Y553JcjsR5D/TOOZdIjjTRe6B3zrlE1qmxbpxzbl2U890rnXNuXZcjFfrqBfpokvD2QAfgazNbUiulcs65LJArvW5S/mUi6XzgV+Bn4GNgsyj9VUmX1krpnHMug/KU+pLNUgr0kv4G3AM8AuzNqhOvDAeOr/GSOedchuVJKS/ZLNWmmwuAG8zsDkn5ldZNBjat2WI551zmZXn8TlmqgX5DYFyCdWVAg5opjnPOZY9sb5JJVapt9FOAPROs2wPw4QOdczlH1fgvm6Vao/830F/SCuDlKK21pLOAy4BzaqFszjmXUQU50pE+1cnBH5XUHLgBuClKfgdYCvQzs+dqqXzOOZcx68IwxaswszslPQTsCrQE5gOfmdnC2iqcc85lUq600VfrgSkzWwS8X0tlcc65rJIjFfrUAn30sFRSZtZ/7YvjnHPZI9v7x6cq1Rr9/UnWWfTqgd45l1Pyc+RmbEqnYWZ5lRegBXAi8DWw9rPwOudclslDKS/ZbI2vV2a2wMxeBB4CHq65IjnnXHaQUl+S70cdJA2T9K2kiZIuidJbSBoi6YfotXnMNtdImiJpsqQDYtK3lzQ+WnefUugaVBM/TH4CdqiB/TjnXFapwUHNSoDLzexPQE/gAkk9gKuBoWbWDRgafSZadwKwOdCb8BxT+fAzDwJ9gW7R0rvK86jmea9CUlvgckKwd865nFJTg5qZ2Uwz+yJ6vwj4FtgIOBwYGGUbCBwRvT8ceMHMlpvZT4TRCXaKYm5TM/vMzAx4KmabhFLtdTOXlTddy9UDmgDLgKNS2Y9zztUl1el0I6kvoaZdboCZDYiTrxOwLTAKaGNmMyFcDCS1jrJtBHwes9n0KK04el85Pam16XWzLDrIe2Y2L8X9OOdcnVGdiUeioL5aYI8lqTHwCnCpmf2RpHk93gpLkp5UlYFeUiHwAfCTmc2oKr9zzuWKmuxdGcXSV4BnzezVKHm2pLZRbb4tMCdKn06Yya9ce2BGlN4+TnpSqZxHKfAh8KcU8jrnXM6QlPJSxX4EPAZ8a2b3xKwaDPSJ3vcB3ohJP0FSfUmdCTddR0fNPIsk9Yz2eVrMNglVWaM3szJJPwBtqsrrnHO5pAZ7x+8GnAqMl/RVlHYtcBswKBoJeBpwLICZTZQ0iDAEfAlwgZmVRtudBzwJNATejZakUm2jvw64XdJ4Mxuf4jbOOVen1dQQCGY2ksTXjX0SbHMLcEuc9LHAFtU5fsJAL2kP4AszWwxcTxix8itJvwKzqXQDwMx2SvWgkjYCOsYe38xGVKfgzjlX27L7edfUJavRDwN2AUYDE6JlrUm6nTCZ+CRC+z+Ei4YHeudcVsnLkXGKkwX6ijM0szNq8JhHAJuZ2fIa3KdzztW4HBnTrHrj0deQH4FCwAO9cy6rrSszTB0kqXsqOzKzp1I85lJCW/9QYoK9mV2c4vbOOZcWuRHmqw70N6S4n/IxF1IxOFqccy6rrSs1+r2AsTV5QDMbKKkesGmUNNnMimvyGM45VxPy15FAX2RmS2rygJJ6EUZpm0r4ZdRBUh/vXumcyza5EeYzczP2bmB/M5sMIGlT4Hlg+wyUxTnnEsqRCn1GAn1heZAHMLPvo8F+nHMuq2T7FIGpShjoo3lha8NYSY8BT0efTwbG1dKxnHNujXmNfs2dB1wAXExoAhsB9M9AOZxzLinleo2+tkRPxN4TLc45l7XWlV43NUbSIDM7TtJ44syIYmZbpassdUG/v1/LxyOG06JFS1567U0A+v/nXoYPG0peXh4tWrTgpn/+iw1at6G4uJib+13Pd5MmUVJayiGHHc6ZZ/8lw2fgakLzhgX02WEjmjYowDBG/rSAYVPmA9Cra3N6dW1BqRkTZi7mtQlz6Ni8ASdv1w4IP5ff+nYuX89YBMCFu23M+g0KyMuDKb8t5YUvZ1U9NdE6LkfifFpr9JdEr4ek8Zh11qGHH8nxJ57MDdddXZF22hlncf5F4Wt8/tmnGPBQf6674SY++O97rFhRzKDX3qSoqIhjjjiY3gceTLuN2ifavasjSg1eGT+bXxYso35BHtfs3ZlvZy+maYMCtm7XhH9+8CMlZUaT+vkAzPhjObd9+CNlBk0bFHD9Pl0YP3MRZQaPjprOspIyAPr2bM/27ZsydvofmTy9rJcrgT5tY/aUT4ALnG9mP8cuwPnpKkddsf0OO7L++uuvkta4ceOK90VFRRVP7UmiqGgpJSUlLF++jMLCQtaLyevqrj+WlfDLgmUALC8pY9aiFTRrWMgeXZrz/uR5lJSFOvmi5WEg2OJSI0qiME+r1NjLg3yewlyoXpuvmqrxXzbLxM3Y/YCrKqUdGCfNxXH/ff/H24PfoHGTJgx4bCAA++x3AMOHfcj+e/+ZZcuWcfnfrmb99ZtltqCuxrVoVEiHZg2YOr+Io7ZswyYtG3HY5q0pLi3j1fGz+fn3cEHo1Lwhp+7QlhaN6vHkmF8rAj/ARbtvTKfmDZk4azFfeG2+SjkySnH6avSSzova57tL+iZm+QnwWatSdOHFf+XdD4Zz4MGH8MLzzwAwccJ48vPyeH/oCN569wOeeeoJpv/yS4ZL6mpS/Xzxl57teenrWSwrKSNf0KheHncM+4lXx8/m7J1XNtNN/b2Im4f8yO0f/kjvzVpREBOt/jNyGle9/T0F+WKz1utl4lTqlDwp5SWbpXO45eeAQwkT2R4as2xvZiensRw5ofdBh/DhB0MAePftt9hl9z9TWFhIi5Yt2Xqb7Zg0sUbmiXFZIE/Qd5cOjP5lIV9FN1Z/Lyrhy1/D+59/X4YZNK6Xv8p2sxatYHlJGe2a1l8lvaTM+GbGIrZu1yQ9J1CH5UrTTTrb6Bea2VTgXmB+TPt8saSd01WOumzaz1Mr3o8Y9iGdOncGoG3btowZ9TlmRtHSpYz/5ms6de6SoVK6mnbq9u2Y9cdyhv4wvyLt6xmLKmrkrRvXIz9PLF5RSstGhRXNDS0aFdKmST3mLS2mfr5o2iC01OYJNt+wMbMW+ZQQVclT6ks2y0Qb/YPAdjGfl8RJW+ddc+VljBszhgULfqf3Pnty7gUXMfLjj/h56lQk0bZdO677+00AHHfiSfS7/lqOPfJQzIzDjjiKTTfbLMNn4GpC15YN6dmxGdMXLuPafcLF+42Jc/h06u+cukM7/r5vF0rKjKfG/hryt2rEAZu1pLQMzIwXvprFkhWlNKmfz3m7dKAgX+QJJs9Zysc//p7JU6sTsr2mniqZpffeu6SvzGybSmnfpNKPfsmKNBfW1QlXvPltpovgstCDR/dY6yg98offU445u3drnrVXhUxMifijpIslFUbLJYTpBZ1zLquoGks2y0SgPxfYFfgVmA7sDPRNlFlSX0ljJY19/NEBaSqic86FIRBSXbJZJsa6mQOcUI38A4AB4E03zrk0y+74nbK0B3pJTxB/rJsz012WTIs3nk2sgU88xrtvh/TS0lJ++vF/DB3xacXDUKWlpZxywjFs0Lo19z3wMAD33nMXn4wcwWbd/8TNt94OwFtvvsEfCxdy0imnpefEXLUV5InL9+xEQZ7Iy4Mvpy/irW/ncmiPDdiqXRPMYNHyEp4aO4OFy0pW2TbZeDgH/2kDdu/crOLJ2TcmzmHirMV0admQE7dtS0mp8fjo6cxdUkzDwjzO3rk9/xk5Le3nn61y5WZsJnrdvBXzvgFwJDAjA+XIuHjj2cTqc8ZZ9DnjLAA+Gv4hzz49cJUnXp9/5ik6d+7C4iWLAVi0aBFff/0lg14dzHVXXcEP30+mw8YdefON17j/wUdq/XzcmispM/49YirLS408wRW9OjNx9mKGfD+PNyfNBWCvri046E+teP7LWatsm2g8nFmLVgAw9If5fPDDvFW22bdbSwZ8Pp2WjQrZo0sLXhk/m4O6b8B73/2WnhOuI7K8RSZlaW+jN7NXYpZngeOALdJdjmwQbzybRN5/5216H3hwxefZs2bx8ccfccTRx1ak5eWJkuJizIxly5dTUFDIU088xoknnUphoU/ile2Wl4Yfuvl5Il9gtnJ8GoB6BfGjTqLxcJIpLTPq5Yl6+aLUjFbrFdKsYQE//La0hs4mN+TKzdhM1Ogr6wZsnOlCZLOioiI+/WQkV13394q0u+64lUv+egVLl66cu3299Rqz9777c+KxR7LTzj1p3KQxEyeOp+95F2Si2K6aBFyzTxc2aFyPj/43n6m/FwFw2OYbsPPGzVhWXMr/jfg56T5ix8Mp16trc3buuD7Tfi/ilW9ms7S4jPcn/8bJ27VlRanx5NhfOXrLNgyOfjm4GNkewVOUiTb6RYQ2ekWvs/ABzZIa8dEwtt5224pmmxEfDaNFi5b02HwLxo4ZtUre0888m9PPPBuAf9x4PeddcDGvvfISn3/6Cd023Yyz/3JeuovvUmTArUN/pGFhHn/p2YF2Tesz44/lDJ44l8ET53LAZi3p1bUFb30bPyBXHg8HYMSP83knyn/o5htw9FZteHrcTKYvXM4dw6cCsEmrRixYVoKAs3baiFIzXvlmdkW7/ros28ewSVUmmm6amFnTmNdNzeyVdJejLvnvu++s0mzz9Zdf8NGwDzn4gL255m+XM3b0KK67+m+rbPPdt5MA6NixE28NfoPb7/43U6b8sMowCi47FRWX8cNvS+jRZtWhpsf88gfbbhR/fJp44+FAGL7YCBeRkT8toFPzhqtte2D3Vrz77VwO/tMGvPXtXEZPW8hem7SoyVOqs7zpppokJR3iwMy+SFdZ6pJFixYxbuwY/vmvOyrSLrr0ci669HIAxo4ZxVNPPs4tt925ynb977+Xv9/4D0pKSigrCzWzPIlly5alr/AuZY3r5VNqRlFxGYV5onvrxrw/+Tc2aFyPuYvDTdWt2jauuMFaWbzxcCBMPvJH1Etnm3ZNmPHHquPb9Oy4PhNmLWZpcRn18oVZuDdQLz8Tj9hkoWyP4ClKZ9PN3dFrA2AH4GvC17gVMArYPY1lyQrxxrMpKQn/Ux5zXHjUYNjQIfTcdTcaNmqU8n6HDf2AzbfYkg1atwFgq6234bgjD6Xbppux6Wbda/5E3Fpbv0EBfXZshyTygHHT/2DCrMX07dmeNo3rUQbMX1rMc1/MrMh/yvZteeCTXxKOhzNx1mKO2qI17Zs1wID5S4p59suZFccszBc9N27GfSNDu//QH+bTt2d7SsuMx0b/muZvIDvlSvfKTIx18wJwi5mNjz5vAVxhZqdXta0/MOXi8bFuXDw1MdbNV9MWpRxzttm4SdZeFTLx+6x7eZAHMLMJwDYZKIdzziUlpb5UvS89LmmOpAkxaS0kDZH0Q/TaPGbdNZKmSJos6YCY9O0ljY/W3SdVffRMBPpvJT0qqZekPSU9AniVzDmXdWp44pEngd6V0q4GhppZN2Bo9BlJPQhDxWwebdNfUvnMMg8SxgfrFi2V97maTAT6M4CJwCXApcCkKM0557JKTdbozWwEML9S8uHAwOj9QOCImPQXzGy5mf0ETAF2ktQWaGpmn1lod38qZpuEMjGo2TJJDwAfEHp9TTaz4nSXwznnqlKdRndJfVl1JN4B0aCMybQxs5kAZjZTUusofSPg85h806O04uh95fSkMvHAVC/ClWsq4XvsIKlPdLVzzrnsUY1IHzvSbi0d2ZKkJ5WJIRDuBvY3s8kAkjYFnge2z0BZnHMuoTR0r5wtqW1Um28LzInSpwMdYvK1Jwz+OD16Xzk9qUy00ReWB3kAM/se8BG3nHNZJw2Tgw8G+kTv+wBvxKSfIKm+pM6Em66jo2aeRZJ6Rr1tTovZJqFM1OjHSXoMeDr6fDIwLgPlcM655GqwQi/peaAX0ErSdOBG4DZgkKSzgGnAsQBmNlHSIEJnlRLgAjMrH3zoPEIPnobAu9GS/NgZeGCqPnAB4UlYASOA/ma2POmG+ANTLj5/YMrFUxMPTH03c2nKMad720ZZ+8BUWmv0kvKAcWa2BXBPOo/tnHPVlSODV6a3jd7MyoCvJfn48865rOejV665tsBESaOBilkzzOywDJTFOecSy/YInqJMBPqbMnBM55yrtlyZeCSd49E3AM4FNgHGA4+ZWUnyrZxzLnNyI8ynt0Y/kPD47sfAgUAPwng3zjmXnXIk0qcz0Pcwsy0Bon70o9N4bOecq7ZcmXgknYG+YuAyMytJYQhl55zLqFwJU+kM9FtL+iN6L6Bh9FmAmVnTNJbFOeeq5IG+mswsv+pczjmXPbzpxjnncpzX6J1zLsflSJz3QO+cc4l4jd4553JebkR6D/TOOZfAWkwoklU80DvnXALedOOccznOu1c651yuy40474HeOecSyZE474HeOecS8TZ655zLcbky+KIHeuecSyA3wrwHeuecSyhHKvQe6J1zLhHvXumccznOa/TOOZfjPNA751yO86Yb55zLcV6jd865HJcjcd4DvXPOJZQjkd4DvXPOJeBt9M45l+N84hHnnMt1Huidcy63edONc87luFzpXikzy3QZ3BqQ1NfMBmS6HC67+L8LF09epgvg1ljfTBfAZSX/d+FW44HeOedynAd655zLcR7o6y5vh3Xx+L8Ltxq/GeuccznOa/TOOZfjPNA751yO80BfAyRtKOkFSf+TNEnSO5L6Snor02Vz2UNSqaSvJE2Q9JKkRmuxr+GSdliD7ZpJOj/mcztJL69pOVzd4IF+LUkS8Bow3My6mlkP4FqgTWZLVvMk5We6DHVckZltY2ZbACuAc2NXpun7bQZUBHozm2Fmx6ThuC6DPNCvvb2AYjN7qDzBzL4CPgYaS3pZ0neSno0uCki6QdKYqGY3ICZ9uKTbJY2W9L2kP0fpjSQNkvSNpBcljSqvzUnaX9Jnkr6IaomNo/R9JH0pabykxyXVl3SgpEHl5ZTUS9KbVexnalTekcCxafg+1xUfA5tEf4Nhkp4DxktqIOmJ6O/2paS9ACQ1jH41fiPpRaBh+Y4kLY55f4ykJ6P3bSS9JunraNkVuA3oGv2yuFNSJ0kTovyJjn26pFclvSfpB0l3pOtLcjXDA/3a2wIYl2DdtsClQA+gC7BblH6/me0Y1ewaAofEbFNgZjtF290YpZ0P/G5mWwE3A9sDSGoFXA/sa2bbAWOByyQ1AJ4EjjezLQljGp0HDAF6Slov2u/xwIuJ9hNTpmVmtruZvVCdL8bFJ6kAOBAYHyXtBFwX/Rq8ACD6u50IDIz+nucBS6N/A7cQ/Ruown3AR2a2NbAdMBG4Gvhf9Mvib5XyJzo2wDaEfy9bAsdL6lDtE3cZ44G+do02s+lmVgZ8BXSK0veKauXjgb2BzWO2eTV6HReTf3fgBQAzmwB8E6X3JFxEPpH0FdAH6AhsBvxkZt9H+QYCe5hZCfAecGgUbA4G3kiyn3IvrvE34GI1jL7fscA04LEofbSZ/RS93x14GsDMvgN+BjYF9gCeidK/YeW/gWT2Bh6Mtik1s4VV5E90bIChZrbQzJYBk1j134fLcj565dqbCCRq41we874UKIhqSP2BHczsF0n9gAZxtill5d8n0Rh6AoaY2YmrJErbJCnvi4Sa23xgjJktipqOVttPjCVJ9udSV2Rm28QmRK12sd9vsvESEz30EpveIEGeVCQ79mr/ltfiOC7NvEa/9j4E6ks6pzxB0o7Angnyl/+P+FvUDp7KjbCRwHHRvnsQfj4DfA7sJmmTaF0jSZsC3wGdytOBU4GPovfDCT/jz2FlTT3Rflz6jQBOBoj+BhsDkyulbwFsFbPNbEl/kpQHHBmTPpTQ5IOkfElNgUVAk2oe29VxHujXkoVHi48E9lPoXjkR6AfMSJB/AfAIoX32dWBMCofpD2wg6RvgKsLP9oVmNhc4HXg+Wvc50D36eX0G8FLUPFQGPBQdvxR4i9BG/FaUFnc/1fgaXM3pD+RHf7cXgdPNbDmhCaZx9Pe5Ehgds83VhL/lh8DMmPRLCM2E4wlNgZub2TxCE90ESXemeGxXx/kQCHWAQre7QjNbJqkroaa2qZmtyHDRnHN1gLez1Q2NgGGSCgntqOd5kHfOpcpr9M45l+O8jd4553KcB3rnnMtxHuidcy7HeaDPcZL6SbKYZYakV6LeO7V1zEOiY3WKPneKPh9Sxaax+zhO0uk1WKbGURmS7jPKc+FaHqufpN/WZh8x+3pS0tia2Jdbd3mvm3XDQqB39L4LYbycoZI2N7N0PPU6E9iF8CBXqo4DWhHG7HHOrQUP9OuGEjP7PHr/uaRphNETDwJeqpxZUkMzK6qpg0cP3XxeZUbnXK3wppt1U/lom52gYijiuyX9XdJ04I8oPU/S1ZKmSFquMHRyn9gdKegnaY6kRZKeAppWyhO36UbSOdGQuMskzVYY0nl9hWF2jwb2jGly6hez3eGSxkbbzZJ0R/SMQey+j47KWyRpBDX0pK+kgyUNic73D0mfS9o/Qd7dFIZ9XqYwLPDucfKcLWli9P3+LOnKmiinc7E80K+bOkWvs2LSTiKMz3M+YThagP8Qhi8eQBjp8jXg8UoB+2LghijPMUARUOV45ZKuBx4mjMFzBGFMloVAY0LT0jDgS0KTzy7Ao9F2xxFG+BwNHAbcBPQF/hWz7+0Ij/B/DRwFDAYqxuFfS52BNwnjBx0NfAq8K2m3SvkaEUabfIgwjv+CKN+GMeX8G2Fog9cJQ1U/CNy8tvcInFuNmfmSwwth3J3fCM10BYRhZ4cRau1tozxTCe3oDWK224QwRk6fSvt7ijDqJUA+YUyfByvlGUIYUbFT9LlT9PmQ6HMzYClwT5Jyv0yYtSs2TYShc5+olH4m4QLTMvo8iDCUrmLyXBeV4fQqvi8DLkzxu82LvtP3gccrfecGnBST1pgwYuht0eemwGLgxkr7/AfhApwffX4SGJvpf0e+1O3Fa/TrhpZAcbRMJtyQPd7MYgfAGmphMLRy+xAC/WuSCsoXwjg720Tj73QA2hLGtI/1KsntQphw5Ylqnkf5iIqDKpXpQ8KooFtE+XYCBptZ7GPfVZUpJZLaSxoo6VeghPCd7s/KcdtjvVb+xswWEy6AO0VJuwDrEQaeq3wubYD2NVFe58Bvxq4rFgL7EmqZs4AZlYIgwOxKn1sRauyJJqtoC5Q3Q8yptK7y58paRq8zk+ZaXavo9Z0E68tnPdpwDcpUpWgY4MGEYX5vAKYQxpL/B9C6UvbFtvoN7TmsHF64/FwmJjhcB8KvF+fWmgf6dUOJmVXVF7ty4J9PqLHuRqjZVzaHlf9+Kge5yp8rmxe9tiU0K6VqfvTal9B+X1n5LE2z1qBMqdiEMD3kgWb2XnmipIZx8jaO03upNSsvbuXncgirX2TBx4F3NcgDvUvkQ0KNfn0zGxIvg6RfCEH1cMIUheWOqmLfnxHa1PsAVyTIs4LVZ0uaDPxKaPt/JMn+xwCHSbom5pdLVWVKRXlArxijXVJHwsUw3tR+RwLPRfkaA/sRblrDyu+gnZm9XQNlcy4hD/QuLjObLOkh4AVJdxDmOW1AmN92UzM728xKo3V3RU+CfkzoifKnKva9QNLNwC2S6hGaYuoTevbcZGa/Eh6uOlzSEcB0QnPTDEmXA08rzJb0LuGC0IXQc+cYM1sK3A6MIrTlP0Zouz+rGqe/jaTKM3/NJTwLMB24W9LfCU04NxEuPpUVRefXmHDD+gqgHnBvzHfQD7g3uliMINzc3RTYy8yOjLNP59ZMpu8G+1K7C1GvmyryTAXuipMu4FJCO/JyQrD7CDitUp6bo3WLgGcJXTUT9rqJ2fYvhN4xywm/DAYBTaN1rQg3M+dH2/aL2e5AwkVlCaH30FfAP4GCmDzHEtrQlxGmYtyR1HvdxFuGR+t3JHTtLAJ+IMzM9SQxPWNY2dPpz1HZlhO6eu4R53inEJ5rKAJ+J1ygLotZv8q+ffFlTRYfj94553Kcd690zrkc54HeOedynAd655zLcR7onXMux3mgd865HOeB3jnncpwHeuecy3Ee6J1zLsf9P6a7dNyMNkf7AAAAAElFTkSuQmCC",
                        "text/plain": [
                            "<Figure size 432x288 with 2 Axes>"
                        ]
                    },
                    "metadata": {
                        "needs_background": "light"
                    }
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 53,
            "source": [
                "# Plot the ROC curve and get the AUC score for the Decision Tree\r\n",
                "plot_roc_curve(clf_ET)"
            ],
            "outputs": [
                {
                    "output_type": "display_data",
                    "data": {
                        "image/png": "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",
                        "text/plain": [
                            "<Figure size 432x288 with 1 Axes>"
                        ]
                    },
                    "metadata": {
                        "needs_background": "light"
                    }
                },
                {
                    "output_type": "stream",
                    "name": "stdout",
                    "text": [
                        "AUC score: 0.969951612003218\n"
                    ]
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "markdown",
            "source": [
                "# SVC\r\n"
            ],
            "metadata": {}
        },
        {
            "cell_type": "markdown",
            "source": [
                "## ⚠️ <span style=\"color:red\">**SVC takes a lot of time to run!!! More than 15 minutes!**</span> ⚠️"
            ],
            "metadata": {}
        },
        {
            "cell_type": "markdown",
            "source": [
                "<span style=\"color:blue\">**You can use static metrics in place of calculated ones if you do not want to wait >20 minutes for the SVC to run.**</span>\r\n",
                "\r\n",
                "We obtained those during our runs. Static metrics can be found in the end of SVC section."
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 54,
            "source": [
                "# Create an SVC classifier object\r\n",
                "clf_SVC = SVC(kernel='rbf', C=100000)\r\n",
                "\r\n",
                "# Train the SVC Classifier\r\n",
                "clf_SVC = clf_SVC.fit(train_features, train_labels)\r\n",
                "\r\n",
                "# Predict the response for the test dataset\r\n",
                "y_pred_SVC = clf_SVC.predict(test_features)"
            ],
            "outputs": [],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 55,
            "source": [
                "# Get the metrics for the SVC\r\n",
                "print(classification_report(test_labels,y_pred_SVC))"
            ],
            "outputs": [
                {
                    "output_type": "stream",
                    "name": "stdout",
                    "text": [
                        "              precision    recall  f1-score   support\n",
                        "\n",
                        "           0       0.87      0.96      0.91      5148\n",
                        "           1       0.91      0.73      0.81      2771\n",
                        "\n",
                        "    accuracy                           0.88      7919\n",
                        "   macro avg       0.89      0.84      0.86      7919\n",
                        "weighted avg       0.88      0.88      0.88      7919\n",
                        "\n"
                    ]
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 56,
            "source": [
                "# Get macro average F1 score\r\n",
                "f1_SVC = f1_score(test_labels, y_pred_SVC, average='macro')\r\n",
                "f1_SVC"
            ],
            "outputs": [
                {
                    "output_type": "execute_result",
                    "data": {
                        "text/plain": [
                            "0.8602850449576763"
                        ]
                    },
                    "metadata": {},
                    "execution_count": 56
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 57,
            "source": [
                "# Calculate Matthews Correlation coefficient (for future use)\r\n",
                "MCC_SVC = matthews_corrcoef(test_labels, y_pred_SVC)\r\n",
                "MCC_SVC"
            ],
            "outputs": [
                {
                    "output_type": "execute_result",
                    "data": {
                        "text/plain": [
                            "0.7308375730727619"
                        ]
                    },
                    "metadata": {},
                    "execution_count": 57
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 58,
            "source": [
                "# Plot the confusion matrix for the SVC\r\n",
                "cf_matrix(test_labels,y_pred_SVC)"
            ],
            "outputs": [
                {
                    "output_type": "stream",
                    "name": "stdout",
                    "text": [
                        "<function cf_matrix at 0x000001BDDDED1280>\n"
                    ]
                },
                {
                    "output_type": "display_data",
                    "data": {
                        "image/png": "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",
                        "text/plain": [
                            "<Figure size 432x288 with 2 Axes>"
                        ]
                    },
                    "metadata": {
                        "needs_background": "light"
                    }
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 59,
            "source": [
                "# Plot the ROC curve and get the AUC score for the SVC\r\n",
                "plot_roc_curve(clf_SVC)"
            ],
            "outputs": [
                {
                    "output_type": "display_data",
                    "data": {
                        "image/png": "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",
                        "text/plain": [
                            "<Figure size 432x288 with 1 Axes>"
                        ]
                    },
                    "metadata": {
                        "needs_background": "light"
                    }
                },
                {
                    "output_type": "stream",
                    "name": "stdout",
                    "text": [
                        "AUC score: 0.8449814750789129\n"
                    ]
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "markdown",
            "source": [
                "### Static SVC metrics"
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 60,
            "source": [
                "f1_SVC_static = 0.8602850449576763"
            ],
            "outputs": [],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 61,
            "source": [
                "MCC_SVC_static = 0.7308375730727619"
            ],
            "outputs": [],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 62,
            "source": [
                "AUC_SVC_static = 0.8449814750789129"
            ],
            "outputs": [],
            "metadata": {}
        },
        {
            "cell_type": "markdown",
            "source": [
                "# Comparison tables\r\n",
                "\r\n",
                "Cells below contain code to create a nice-looking table which compares F1, AUC and MCC of all algorithms used in OBerA.\r\n",
                "\r\n",
                "<span style=\"color:yellow\">**Please note that the table uses static values for SVC to prevent users from getting errors in case they did not run their own SVC. If you have generated your own values, please change the corresponding inputs in the following cells. If you did not modify our script, you shall simply remove \"_static\" from each value and it should work.**</span>"
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 63,
            "source": [
                "# Create a skeleton of the table\r\n",
                "comp_table = pd.DataFrame(index=['Neural Network', 'SVC', 'Decision Tree', 'Balanced Random Forest', 'Random Forest', 'Extra Trees'], columns=['F1 score', 'AUC score', 'MCC'])"
            ],
            "outputs": [],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 64,
            "source": [
                "# Add values to the table\r\n",
                "comp_table.iloc[0] = [f1_NN, roc_auc_score(test_labels, y_pred_NN), MCC_NN]\r\n",
                "comp_table.iloc[1] = [f1_SVC_static, AUC_SVC_static, MCC_SVC_static] #if needed, put your SVC values here or simply remove \"_static\" from each parameter\r\n",
                "comp_table.iloc[2] = [f1_DT, roc_auc_score(test_labels, y_pred_DT), MCC_DT]\r\n",
                "comp_table.iloc[3] = [f1_BRF, roc_auc_score(test_labels, y_pred_BRF), MCC_BRF]\r\n",
                "comp_table.iloc[4] = [f1_RF, roc_auc_score(test_labels, y_pred_RF), MCC_RF]\r\n",
                "comp_table.iloc[5] = [f1_ET, roc_auc_score(test_labels, y_pred_ET), MCC_ET]"
            ],
            "outputs": [],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 65,
            "source": [
                "# Round all the values down to two digits\r\n",
                "for i in range(len(comp_table)):\r\n",
                "    comp_table.iloc[i] = comp_table.iloc[i].apply(lambda x: round(x, 2))"
            ],
            "outputs": [],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 66,
            "source": [
                "# Show the table\r\n",
                "comp_table"
            ],
            "outputs": [
                {
                    "output_type": "execute_result",
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>F1 score</th>\n",
                            "      <th>AUC score</th>\n",
                            "      <th>MCC</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>Neural Network</th>\n",
                            "      <td>0.87</td>\n",
                            "      <td>0.86</td>\n",
                            "      <td>0.74</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>SVC</th>\n",
                            "      <td>0.86</td>\n",
                            "      <td>0.84</td>\n",
                            "      <td>0.73</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>Decision Tree</th>\n",
                            "      <td>0.95</td>\n",
                            "      <td>0.95</td>\n",
                            "      <td>0.9</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>Balanced Random Forest</th>\n",
                            "      <td>0.97</td>\n",
                            "      <td>0.97</td>\n",
                            "      <td>0.93</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>Random Forest</th>\n",
                            "      <td>0.97</td>\n",
                            "      <td>0.97</td>\n",
                            "      <td>0.95</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>Extra Trees</th>\n",
                            "      <td>0.97</td>\n",
                            "      <td>0.97</td>\n",
                            "      <td>0.95</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "                       F1 score AUC score   MCC\n",
                            "Neural Network             0.87      0.86  0.74\n",
                            "SVC                        0.86      0.84  0.73\n",
                            "Decision Tree              0.95      0.95   0.9\n",
                            "Balanced Random Forest     0.97      0.97  0.93\n",
                            "Random Forest              0.97      0.97  0.95\n",
                            "Extra Trees                0.97      0.97  0.95"
                        ]
                    },
                    "metadata": {},
                    "execution_count": 66
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "markdown",
            "source": [
                "# Graphical comparison tables\r\n",
                "\r\n",
                "Below are the graphs that compare all the models in a better, human readable way.\r\n",
                "\r\n",
                "<span style=\"color:yellow\">Please note that you need to run your SVC model for these graphs to work, Otherwise you could modify the code to avoid plotting SVC at all.</span>"
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 70,
            "source": [
                "%%capture --no-display\r\n",
                "\r\n",
                "fig, axs = plt.subplots(nrows=2, ncols=3, figsize=(25, 15)) #creating a skeleton of the graph\r\n",
                "\r\n",
                "labels = ['Changeover', 'Production'] #adding labels\r\n",
                "\r\n",
                "sns.heatmap(confusion_matrix(test_labels, y_pred_DT), cbar=False, annot=True, fmt='g', ax=axs[0,0]) #adding Decision Tree to subplot 0,0\r\n",
                "axs[0,0].set_title('Confusion Matrix Decision Tree')\r\n",
                "\r\n",
                "sns.heatmap(confusion_matrix(test_labels, y_pred_NN), cbar=False, annot=True, fmt='g', ax=axs[0,1]) #adding Neural Network to subplot 0,1\r\n",
                "axs[0,1].set_title('Confusion Matrix Neural Network')\r\n",
                "\r\n",
                "sns.heatmap(confusion_matrix(test_labels, y_pred_BRF), cbar=False, annot=True, fmt='g', ax=axs[0,2])#adding Balanced Random forest to subplot 0,2\r\n",
                "axs[0,2].set_title('Confusion Matrix Balanced RF')\r\n",
                "\r\n",
                "sns.heatmap(confusion_matrix(test_labels, y_pred_RF), cbar=False, annot=True, fmt='g', ax=axs[1,0])#adding Random Forest to subplot 1,0\r\n",
                "axs[1,0].set_title('Confusion Matrix Random Forest')\r\n",
                "\r\n",
                "sns.heatmap(confusion_matrix(test_labels, y_pred_SVC), cbar=False, annot=True, fmt='g', ax=axs[1,1])#adding SVC to subplot 1,1\r\n",
                "axs[1,1].set_title('Confusion Matrix SVC')\r\n",
                "\r\n",
                "for i in range(0,3): #iterating through our graphs to give a label to each\r\n",
                "    axs[0,i].set(xlabel='Predicted labels')\r\n",
                "    axs[0,i].set(ylabel='True labels')\r\n",
                "    axs[0,i].xaxis.set_ticklabels(['Changeover', 'Production'])\r\n",
                "    axs[0,i].yaxis.set_ticklabels(['Changeover', 'Production'], va='center')\r\n",
                "    axs[1,i].set(xlabel='Predicted labels')\r\n",
                "    axs[1,i].set(ylabel='True labels')\r\n",
                "    axs[1,i].xaxis.set_ticklabels(['Changeover', 'Production'])\r\n",
                "    axs[1,i].yaxis.set_ticklabels(['Changeover', 'Production'], va='center')\r\n",
                "\r\n",
                "#fig.delaxes(axs[1,2]) #this line can be used to delete the unnecessary subplot from the entire plot\r\n",
                "\r\n",
                "# Add our table with metrics from before to the graph\r\n",
                "table = axs[1,2].table(cellText=comp_table.values,\r\n",
                "          colLabels=comp_table.columns,\r\n",
                "          rowLabels=comp_table.index,\r\n",
                "          loc=\"lower right\",\r\n",
                "          cellLoc='center',\r\n",
                "          bbox=[0.4, 0, 0.75, 1],\r\n",
                "          fontsize=64).set_fontsize(14)\r\n",
                "\r\n",
                "plt.axis('off')"
            ],
            "outputs": [
                {
                    "output_type": "execute_result",
                    "data": {
                        "text/plain": [
                            "(0.0, 1.0, 0.0, 1.0)"
                        ]
                    },
                    "metadata": {},
                    "execution_count": 70
                },
                {
                    "output_type": "display_data",
                    "data": {
                        "image/png": "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",
                        "text/plain": [
                            "<Figure size 1800x1080 with 6 Axes>"
                        ]
                    },
                    "metadata": {
                        "needs_background": "light"
                    }
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "markdown",
            "source": [
                "Next graph is the same but includes the confusion matrix for Extra Trees model in place of the table."
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": 71,
            "source": [
                "fig, axs = plt.subplots(nrows=2, ncols=3, figsize=(25, 15))\r\n",
                "\r\n",
                "labels = ['Changeover', 'Production']\r\n",
                "\r\n",
                "sns.heatmap(confusion_matrix(test_labels, y_pred_DT), cbar=False, annot=True, fmt='g', ax=axs[0,0])\r\n",
                "axs[0,0].set_title('Confusion Matrix Decision Tree')\r\n",
                "\r\n",
                "sns.heatmap(confusion_matrix(test_labels, y_pred_NN), cbar=False, annot=True, fmt='g', ax=axs[0,1])\r\n",
                "axs[0,1].set_title('Confusion Matrix Neural Network')\r\n",
                "\r\n",
                "sns.heatmap(confusion_matrix(test_labels, y_pred_BRF), cbar=False, annot=True, fmt='g', ax=axs[0,2])\r\n",
                "axs[0,2].set_title('Confusion Matrix Balanced RF')\r\n",
                "\r\n",
                "sns.heatmap(confusion_matrix(test_labels, y_pred_RF), cbar=False, annot=True, fmt='g', ax=axs[1,0])\r\n",
                "axs[1,0].set_title('Confusion Matrix Random Forest')\r\n",
                "\r\n",
                "sns.heatmap(confusion_matrix(test_labels, y_pred_SVC), cbar=False, annot=True, fmt='g', ax=axs[1,1])\r\n",
                "axs[1,1].set_title('Confusion Matrix SVC')\r\n",
                "\r\n",
                "sns.heatmap(confusion_matrix(test_labels, y_pred_ET), cbar=False, annot=True, fmt='g', ax=axs[1,2])\r\n",
                "axs[1,2].set_title('Confusion Matrix Extra Trees')\r\n",
                "\r\n",
                "for i in range(0,3): #iterating through our graphs to give a label to each\r\n",
                "    axs[0,i].set(xlabel='Predicted labels')\r\n",
                "    axs[0,i].set(ylabel='True labels')\r\n",
                "    axs[0,i].xaxis.set_ticklabels(['Changeover', 'Production'])\r\n",
                "    axs[0,i].yaxis.set_ticklabels(['Changeover', 'Production'], va='center')\r\n",
                "    axs[1,i].set(xlabel='Predicted labels')\r\n",
                "    axs[1,i].set(ylabel='True labels')\r\n",
                "    axs[1,i].xaxis.set_ticklabels(['Changeover', 'Production'])\r\n",
                "    axs[1,i].yaxis.set_ticklabels(['Changeover', 'Production'], va='center')"
            ],
            "outputs": [
                {
                    "output_type": "display_data",
                    "data": {
                        "image/png": "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",
                        "text/plain": [
                            "<Figure size 1800x1080 with 6 Axes>"
                        ]
                    },
                    "metadata": {
                        "needs_background": "light"
                    }
                }
            ],
            "metadata": {}
        },
        {
            "cell_type": "code",
            "execution_count": null,
            "source": [],
            "outputs": [],
            "metadata": {}
        }
    ],
    "metadata": {
        "orig_nbformat": 4,
        "language_info": {
            "name": "python",
            "version": "3.9.6",
            "mimetype": "text/x-python",
            "codemirror_mode": {
                "name": "ipython",
                "version": 3
            },
            "pygments_lexer": "ipython3",
            "nbconvert_exporter": "python",
            "file_extension": ".py"
        },
        "kernelspec": {
            "name": "python3",
            "display_name": "Python 3.9.6 64-bit ('base': conda)"
        },
        "interpreter": {
            "hash": "ffe16161eec1dcdad78b77b0d7a1d179ba96f37233aa4c571b1f2b600a0e4afb"
        }
    },
    "nbformat": 4,
    "nbformat_minor": 2
}