09_SMOTE.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "sound-marketplace",
   "metadata": {},
   "source": [
    "# Flight Delay Prediction\n",
    "## SMOTE to Handle Imbalance in ArrDel15\n",
    "- <strong>Environment Setup:</strong> import libraries and load the dataset\n",
    "- <strong>Imbalanced Data Analysis:</strong> understanding how the dataset balance is restored using random under sampling\n",
    "- <strong>SMOTE:</strong> Creating synthetic data for the minority class using SMOTE(Synthetic Minority Over-sampling Technique), takes random samples from the minority class, finds its nearest k neighbors, and then selects a point between the randomly selected data point and its nearest k neighbors to generate synthetic data\n",
    "- <strong>Inference:</strong> understand how the model produces results and choosing the best classifier"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "genetic-crash",
   "metadata": {},
   "source": [
    "## Environment Setup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "bridal-workshop",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/nive/anaconda3/lib/python3.8/site-packages/numpy/lib/arraysetops.py:583: FutureWarning: elementwise comparison failed; returning scalar instead, but in the future will perform elementwise comparison\n",
      "  mask |= (ar1 == a)\n"
     ]
    }
   ],
   "source": [
    "# Pre-requisites\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "# Store the classifier models to save time\n",
    "import joblib\n",
    "\n",
    "# Preprocessing\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "# Classifiers from scikit-learn\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.ensemble import GradientBoostingClassifier\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.ensemble import ExtraTreesClassifier\n",
    "\n",
    "# Performance metrics\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import f1_score\n",
    "from sklearn.metrics import precision_score\n",
    "from sklearn.metrics import recall_score\n",
    "from sklearn.metrics import classification_report\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from sklearn.metrics import plot_confusion_matrix\n",
    "\n",
    "from sklearn.metrics import roc_curve\n",
    "from sklearn.metrics import roc_auc_score\n",
    "\n",
    "# Imbalanced dataset helpers\n",
    "from imblearn.over_sampling import SMOTE\n",
    "\n",
    "df = pd.read_csv(\"Data/flight_and_weather_encoded.csv\", index_col=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "sought-firewall",
   "metadata": {},
   "source": [
    "## Dataset Without Handling Imbalance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "entire-shaft",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.0    1463378\n",
      "1.0     388058\n",
      "Name: ArrDel15, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# Plotting the percentage of observations that fall under each class\n",
    "ax = df[\"ArrDel15\"].value_counts().sort_values().plot(kind=\"barh\", color=[\"r\", \"g\"])\n",
    "ax.set_axisbelow(True)\n",
    "ax.grid()\n",
    "totals= []\n",
    "for i in ax.patches:\n",
    "    totals.append(i.get_width())\n",
    "total = sum(totals)\n",
    "for i in ax.patches:\n",
    "     ax.text(i.get_width()+.3, i.get_y()+.20, \n",
    "     str(round((i.get_width()/total)*100, 2))+'%', \n",
    "     fontsize=10, color='black')\n",
    "plt.title(\"ArrDel15\", fontsize=20)\n",
    "plt.xlabel(\"Count\", fontsize=14)\n",
    "plt.ylabel(\"Class\", fontsize=14)\n",
    "plt.show()\n",
    "print(df[\"ArrDel15\"].value_counts())\n",
    "fig = ax.get_figure()\n",
    "fig.savefig(\"Img/09_01_Dataset_ArrDel15_Class_Imbalance.png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "laden-hawaii",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Dataset shape: (1851436, 24)\n",
      "features_train shape: (1481148, 23) | features_test shape: (370288, 23)\n",
      "labels_train shape: (1481148,) | labels_test shape: (370288,)\n"
     ]
    }
   ],
   "source": [
    "features = df.loc[:, df.columns != \"ArrDel15\"]\n",
    "labels = np.asarray(df[\"ArrDel15\"])\n",
    "# Number of samples/observations/rows is greater than 100,000\n",
    "print(f\"\\nDataset shape: {df.shape}\")\n",
    "features_train, features_test, labels_train, labels_test = train_test_split(features, labels, test_size=0.20, random_state=42)\n",
    "print(f\"features_train shape: {features_train.shape} | features_test shape: {features_test.shape}\")\n",
    "print(f\"labels_train shape: {labels_train.shape} | labels_test shape: {labels_test.shape}\")\n",
    "# print(f\"{features_train.shape[1]} Features: {features_train.columns.to_list()}\")\n",
    "del features\n",
    "del labels"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "thirty-onion",
   "metadata": {},
   "source": [
    "## Training Data Without Handling Imbalance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "chronic-spectacular",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.0    1170643\n",
      "1.0     310505\n",
      "dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# Plotting the percentage of observations that fall under each class\n",
    "ax = pd.DataFrame(labels_train).value_counts().sort_values().plot(kind=\"barh\", color=[\"r\", \"g\"])\n",
    "ax.set_axisbelow(True)\n",
    "ax.grid()\n",
    "totals= []\n",
    "for i in ax.patches:\n",
    "    totals.append(i.get_width())\n",
    "total = sum(totals)\n",
    "for i in ax.patches:\n",
    "     ax.text(i.get_width()+.3, i.get_y()+.20, \n",
    "     str(round((i.get_width()/total)*100, 2))+'%', \n",
    "     fontsize=10, color='black')\n",
    "plt.title(\"ArrDel15\", fontsize=20)\n",
    "plt.xlabel(\"Count\", fontsize=14)\n",
    "plt.ylabel(\"Class\", fontsize=14)\n",
    "plt.show()\n",
    "print(pd.DataFrame(labels_train).value_counts())\n",
    "fig = ax.get_figure()\n",
    "fig.savefig(\"Img/09_02_Training_ArrDel15_Class_Imbalance.png\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "horizontal-explosion",
   "metadata": {},
   "source": [
    "## SMOTE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "numerous-dining",
   "metadata": {},
   "outputs": [],
   "source": [
    "# smote = SMOTE(random_state=42)\n",
    "# features_train, labels_train = smote.fit_resample(features_train, labels_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "charitable-cross",
   "metadata": {},
   "outputs": [],
   "source": [
    "# SMOTE takes too long - approx 47 minutes\n",
    "# features_train.to_csv(\"./Data/smote_features_train.csv\")\n",
    "# pd.DataFrame(labels_train).to_csv(\"./Data/smote_labels_train.csv\")\n",
    "features_train = pd.read_csv(\"./Data/smote_features_train.csv\", index_col=0)\n",
    "labels_train = pd.read_csv(\"./Data/smote_labels_train.csv\", index_col=0)\n",
    "# labels_train.shape\n",
    "labels_train = labels_train[\"0\"].values"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "anonymous-daniel",
   "metadata": {},
   "source": [
    "## Balancing the Training Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "electronic-venture",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.0    1170643\n",
      "1.0    1170643\n",
      "dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# Plotting the percentage of observations that fall under each class\n",
    "ax = pd.DataFrame(labels_train).value_counts().sort_values().plot(kind=\"barh\", color=[\"r\", \"g\"])\n",
    "ax.set_axisbelow(True)\n",
    "ax.grid()\n",
    "totals= []\n",
    "for i in ax.patches:\n",
    "    totals.append(i.get_width())\n",
    "total = sum(totals)\n",
    "for i in ax.patches:\n",
    "     ax.text(i.get_width()+.3, i.get_y()+.20, \n",
    "     str(round((i.get_width()/total)*100, 2))+'%', \n",
    "     fontsize=10, color='black')\n",
    "plt.title(\"ArrDel15\", fontsize=20)\n",
    "plt.xlabel(\"Count\", fontsize=14)\n",
    "plt.ylabel(\"Class\", fontsize=14)\n",
    "plt.show()\n",
    "print(pd.DataFrame(labels_train).value_counts())\n",
    "fig = ax.get_figure()\n",
    "fig.savefig(\"Img/09_03_SMOTE_Training_ArrDel15_Class.png\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "decreased-preview",
   "metadata": {},
   "source": [
    "### Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "blessed-artist",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.94      0.93      0.93    292735\n",
      "         1.0       0.74      0.78      0.76     77553\n",
      "\n",
      "    accuracy                           0.90    370288\n",
      "   macro avg       0.84      0.85      0.85    370288\n",
      "weighted avg       0.90      0.90      0.90    370288\n",
      "\n",
      "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay object at 0x7f7b66aba2e0>\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# model = LogisticRegression(n_jobs=-1)\n",
    "# model.fit(features_train, labels_train)\n",
    "# joblib.dump(model, \"./SMOTE/LogisticRegression.joblib\")\n",
    "model = joblib.load(\"./SMOTE/LogisticRegression.joblib\")\n",
    "model_pred = model.predict(features_test)\n",
    "# print(confusion_matrix(labels_test, model_pred))\n",
    "print(classification_report(labels_test, model_pred))\n",
    "conf_mat_plot = plot_confusion_matrix(model, features_test, labels_test)\n",
    "plt.title(\"Logistic Regression\")\n",
    "print(conf_mat_plot)\n",
    "plt.savefig(\"Img/09_04_SMOTE_LR_Confusion_Matrix\")\n",
    "del model\n",
    "del model_pred"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "quick-pioneer",
   "metadata": {},
   "source": [
    "### Decision Tree Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "compressed-hollow",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.92      0.90      0.91    292735\n",
      "         1.0       0.66      0.71      0.68     77553\n",
      "\n",
      "    accuracy                           0.86    370288\n",
      "   macro avg       0.79      0.80      0.80    370288\n",
      "weighted avg       0.87      0.86      0.86    370288\n",
      "\n",
      "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay object at 0x7f7b678778e0>\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# model = DecisionTreeClassifier()\n",
    "# model.fit(features_train, labels_train)\n",
    "# joblib.dump(model, \"./SMOTE/DecisionTreeClassifier.joblib\")\n",
    "model = joblib.load(\"./SMOTE/DecisionTreeClassifier.joblib\")\n",
    "model_pred = model.predict(features_test)\n",
    "# print(confusion_matrix(labels_test, model_pred))\n",
    "print(classification_report(labels_test, model_pred))\n",
    "conf_mat_plot = plot_confusion_matrix(model, features_test, labels_test)\n",
    "plt.title(\"Decision Tree Classifier\")\n",
    "print(conf_mat_plot)\n",
    "plt.savefig(\"Img/09_05_SMOTE_DTC_Confusion_Matrix\")\n",
    "del model\n",
    "del model_pred"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ignored-daily",
   "metadata": {},
   "source": [
    "### Gradient Boosting Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "devoted-aspect",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.93      0.95      0.94    292735\n",
      "         1.0       0.80      0.75      0.77     77553\n",
      "\n",
      "    accuracy                           0.91    370288\n",
      "   macro avg       0.87      0.85      0.86    370288\n",
      "weighted avg       0.91      0.91      0.91    370288\n",
      "\n",
      "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay object at 0x7f7b6579aeb0>\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# model = GradientBoostingClassifier()\n",
    "# model.fit(features_train, labels_train)\n",
    "# joblib.dump(model, \"./SMOTE/GradientBoostingClassifier.joblib\")\n",
    "model = joblib.load(\"./SMOTE/GradientBoostingClassifier.joblib\")\n",
    "model_pred = model.predict(features_test)\n",
    "# print(confusion_matrix(labels_test, model_pred))\n",
    "print(classification_report(labels_test, model_pred))\n",
    "conf_mat_plot = plot_confusion_matrix(model, features_test, labels_test)\n",
    "plt.title(\"Gradient Boosting Classifier (XGBoost)\")\n",
    "print(conf_mat_plot)\n",
    "plt.savefig(\"Img/09_06_SMOTE_GBC_Confusion_Matrix\")\n",
    "del model\n",
    "del model_pred"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "coastal-accountability",
   "metadata": {},
   "source": [
    "### Random Forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "cutting-employment",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.93      0.95      0.94    292735\n",
      "         1.0       0.81      0.74      0.78     77553\n",
      "\n",
      "    accuracy                           0.91    370288\n",
      "   macro avg       0.87      0.85      0.86    370288\n",
      "weighted avg       0.91      0.91      0.91    370288\n",
      "\n",
      "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay object at 0x7f7b66ae60a0>\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# model = RandomForestClassifier(n_jobs=-1)\n",
    "# model.fit(features_train, labels_train)\n",
    "# joblib.dump(model, \"./SMOTE/RandomForestClassifier.joblib\")\n",
    "model = joblib.load(\"./SMOTE/RandomForestClassifier.joblib\")\n",
    "model_pred = model.predict(features_test)\n",
    "# print(confusion_matrix(labels_test, model_pred))\n",
    "print(classification_report(labels_test, model_pred))\n",
    "conf_mat_plot = plot_confusion_matrix(model, features_test, labels_test)\n",
    "plt.title(\"Random Forest Classifier\")\n",
    "print(conf_mat_plot)\n",
    "plt.savefig(\"Img/09_07_SMOTE_RFC_Confusion_Matrix\")\n",
    "del model\n",
    "del model_pred"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "considerable-macro",
   "metadata": {},
   "source": [
    "### Extra Trees Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "fundamental-regular",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.94      0.94      0.94    292735\n",
      "         1.0       0.77      0.76      0.76     77553\n",
      "\n",
      "    accuracy                           0.90    370288\n",
      "   macro avg       0.85      0.85      0.85    370288\n",
      "weighted avg       0.90      0.90      0.90    370288\n",
      "\n",
      "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay object at 0x7f7b65715850>\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "model = ExtraTreesClassifier(n_jobs=-1)\n",
    "model.fit(features_train, labels_train)\n",
    "# joblib.dump(model, \"./SMOTE/ExtraTreesClassifier.joblib\")\n",
    "# model = joblib.load(\"./SMOTE/ExtraTreesClassifier.joblib\")\n",
    "model_pred = model.predict(features_test)\n",
    "# print(confusion_matrix(labels_test, model_pred))\n",
    "print(classification_report(labels_test, model_pred))\n",
    "conf_mat_plot = plot_confusion_matrix(model, features_test, labels_test)\n",
    "plt.title(\"Extra Trees Classifier\")\n",
    "print(conf_mat_plot)\n",
    "plt.savefig(\"Img/09_08_SMOTE_ETC_Confusion_Matrix\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cfe4c8d9",
   "metadata": {},
   "source": [
    "## Performance Evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "2c94435b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Setup\n",
    "perf_df = pd.DataFrame(columns=[\"classifiers\", \"fpr\", \"tpr\", \"auc\"])\n",
    "\n",
    "# Handling Extra Trees Classifier separately due to memory constraints\n",
    "fpr, tpr, _ = roc_curve(labels_test,  model_pred)\n",
    "auc = roc_auc_score(labels_test, model_pred)\n",
    "# Record the results\n",
    "perf_df = perf_df.append({\"classifiers\": \"ExtraTreesClassifier\",\n",
    "                                        \"fpr\": fpr, \n",
    "                                        \"tpr\": tpr, \n",
    "                                        \"auc\": auc}, ignore_index=True)\n",
    "del model\n",
    "del model_pred\n",
    "\n",
    "classifiers = [\"LogisticRegression\", \"DecisionTreeClassifier\", \"GradientBoostingClassifier\", \"RandomForestClassifier\"]\n",
    "for clf in classifiers:\n",
    "    # Load the trained classifiers\n",
    "    model = joblib.load(f\"./SMOTE/{clf}.joblib\")\n",
    "    model_pred = model.predict(features_test)\n",
    "    fpr, tpr, _ = roc_curve(labels_test,  model_pred)\n",
    "    auc = roc_auc_score(labels_test, model_pred)\n",
    "    # Record the results\n",
    "    perf_df = perf_df.append({\"classifiers\": clf,\n",
    "                                        \"fpr\": fpr, \n",
    "                                        \"tpr\": tpr, \n",
    "                                        \"auc\": auc}, ignore_index=True)\n",
    "    del model\n",
    "    del model_pred\n",
    "\n",
    "# Set name of the classifiers as index labels\n",
    "perf_df.set_index(\"classifiers\", inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "8d4f68fa",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(8,6), dpi=80)\n",
    "for clf_name in perf_df.index:\n",
    "    plt.plot(perf_df.loc[clf_name][\"fpr\"], \n",
    "             perf_df.loc[clf_name][\"tpr\"], \n",
    "             label=\"{}, AUC={:.3f}\".format(clf_name, perf_df.loc[clf_name][\"auc\"]))\n",
    "    \n",
    "plt.plot([0,1], [0,1], color='orange', linestyle='--')\n",
    "\n",
    "plt.xticks(np.arange(0.0, 1.1, step=0.1))\n",
    "plt.xlabel(\"Flase Positive Rate\", fontsize=15)\n",
    "\n",
    "plt.yticks(np.arange(0.0, 1.1, step=0.1))\n",
    "plt.ylabel(\"True Positive Rate\", fontsize=15)\n",
    "\n",
    "plt.title(\"ROC AUC Analysis\", fontweight=\"bold\", fontsize=15)\n",
    "plt.legend(prop={\"size\":13}, loc=\"lower right\")\n",
    "\n",
    "plt.show()\n",
    "fig.savefig(\"Img/09_09_SMOTE_Classifiers_AUC_ROC\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "grateful-brief",
   "metadata": {},
   "source": [
    "## Inference\n",
    "<table>\n",
    "    <thead>\n",
    "        <tr>\n",
    "            <th rowspan=\"3\">Classification Model</th>\n",
    "            <th colspan=\"8\">Performance Metric</th>\n",
    "        </tr>\n",
    "        <tr>\n",
    "            <th colspan=\"2\">Precision</th>\n",
    "            <th colspan=\"2\">Recall</th>\n",
    "            <th colspan=\"2\">F1 Score</th>\n",
    "            <th rowspan=\"2\" colspan=\"2\">Accuracy</th>\n",
    "        </tr>\n",
    "        <tr>\n",
    "            <th>0.0</th>\n",
    "            <th>1.0</th>\n",
    "            <th>0.0</th>\n",
    "            <th>1.0</th>\n",
    "            <th>0.0</th>\n",
    "            <th>1.0</th>\n",
    "        </tr>\n",
    "    </thead>\n",
    "    <tbody>\n",
    "        <tr>\n",
    "            <th>Logistic Regression</th>\n",
    "            <td>0.94</td>\n",
    "            <td>0.74</td>\n",
    "            <td>0.93</td>\n",
    "            <td>0.78</td>\n",
    "            <td>0.93</td>\n",
    "            <td>0.76</td>\n",
    "            <td>0.90</td>\n",
    "        </tr>\n",
    "        <tr>\n",
    "            <th>Decision Tree Classifier</th>\n",
    "            <td>0.92</td>\n",
    "            <td>0.66</td>\n",
    "            <td>0.90</td>\n",
    "            <td>0.71</td>\n",
    "            <td>0.91</td>\n",
    "            <td>0.68</td>\n",
    "            <td>0.86</td>\n",
    "        </tr>\n",
    "        <tr>\n",
    "            <th>Extra Trees Classifier</th>\n",
    "            <td>0.94</td>\n",
    "            <td>0.77</td>\n",
    "            <td>0.94</td>\n",
    "            <td>0.76</td>\n",
    "            <td>0.94</td>\n",
    "            <td>0.76</td>\n",
    "            <td>0.90</td>\n",
    "        </tr>\n",
    "        <tr>\n",
    "            <th>Gradient Boosting Classifier</th>\n",
    "            <td>0.93</td>\n",
    "            <td>0.80</td>\n",
    "            <td>0.95</td>\n",
    "            <td>0.75</td>\n",
    "            <td>0.94</td>\n",
    "            <td>0.77</td>\n",
    "            <td>0.91</td>\n",
    "        </tr>\n",
    "        <tr>\n",
    "            <th>Random Forest Classifier</th>\n",
    "            <td>0.93</td>\n",
    "            <td>0.81</td>\n",
    "            <td>0.95</td>\n",
    "            <td>0.74</td>\n",
    "            <td>0.94</td>\n",
    "            <td>0.78</td>\n",
    "            <td>0.91</td>\n",
    "        </tr>\n",
    "    </tbody>\n",
    "</table>"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}