BallByBallEventPredictor.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "title: \"Project Step II\"\n",
    "subtitle: \"Group 11 - Ball by Ball Event Predictor\"\n",
    "date: 10/17/2023\n",
    "date-modified: last-modified\n",
    "date-format: long\n",
    "format:\n",
    "  html:\n",
    "    theme: [cosmo, theme.scss]\n",
    "    toc: true\n",
    "    embed-resources: true\n",
    "    number-sections: true\n",
    "author:\n",
    "  - name: Tegveer Ghura\n",
    "    affiliations:\n",
    "      - id: gu\n",
    "        name: Georgetown University\n",
    "        city: Washington\n",
    "        state: DC\n",
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Preparation "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import the necessary libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import random\n",
    "\n",
    "random.seed(1310)\n",
    "from seaborn.palettes import color_palette\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "%matplotlib inline\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "import re\n",
    "import plotly.express as px\n",
    "import plotly.io as pio\n",
    "\n",
    "# pio.renderers.default = \"notebook\"\n",
    "pio.renderers.default = \"plotly_mimetype+notebook_connected\"\n",
    "from sklearn.metrics import (\n",
    "    classification_report,\n",
    "    confusion_matrix,\n",
    "    ConfusionMatrixDisplay,\n",
    ")\n",
    "from sklearn.exceptions import ConvergenceWarning\n",
    "from sklearn.exceptions import DataConversionWarning\n",
    "\n",
    "ConvergenceWarning(\"ignore\")\n",
    "DataConversionWarning(\"ignore\")\n",
    "from sklearn.model_selection import (\n",
    "    cross_val_score,\n",
    "    RepeatedKFold,\n",
    "    GridSearchCV,\n",
    "    RandomizedSearchCV,\n",
    ")\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.ensemble import (\n",
    "    GradientBoostingClassifier,\n",
    "    BaggingClassifier,\n",
    "    RandomForestClassifier,\n",
    "    StackingClassifier,\n",
    "    AdaBoostClassifier,\n",
    ")\n",
    "from xgboost import XGBClassifier\n",
    "#from catboost import CatBoostClassifier\n",
    "import pickle\n",
    "import joblib\n",
    "import json\n",
    "import sys\n",
    "import os\n",
    "import shutil\n",
    "import warnings"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Load the dataset and display the dataframe "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv(\"data/raw/deliveries.csv\", encoding=\"ISO-8859-1\", na_filter=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "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>season</th>\n",
       "      <th>match_id</th>\n",
       "      <th>venue</th>\n",
       "      <th>inning</th>\n",
       "      <th>batting_team</th>\n",
       "      <th>bowling_team</th>\n",
       "      <th>batsman</th>\n",
       "      <th>non_striker</th>\n",
       "      <th>bowler</th>\n",
       "      <th>ball</th>\n",
       "      <th>...</th>\n",
       "      <th>noball_runs</th>\n",
       "      <th>penalty_runs</th>\n",
       "      <th>batsman_runs</th>\n",
       "      <th>extra_runs</th>\n",
       "      <th>total_runs</th>\n",
       "      <th>player_dismissed</th>\n",
       "      <th>dismissal_kind</th>\n",
       "      <th>fielder</th>\n",
       "      <th>wickets</th>\n",
       "      <th>commentary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2008</td>\n",
       "      <td>1</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>SC Ganguly</td>\n",
       "      <td>BB McCullum</td>\n",
       "      <td>P Kumar</td>\n",
       "      <td>0.1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>Kumar to Ganguly, 1 leg bye, and thats runs im...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2008</td>\n",
       "      <td>1</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>BB McCullum</td>\n",
       "      <td>SC Ganguly</td>\n",
       "      <td>P Kumar</td>\n",
       "      <td>0.2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>Kumar to McCullum, no run, this one nips back ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2008</td>\n",
       "      <td>1</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>BB McCullum</td>\n",
       "      <td>SC Ganguly</td>\n",
       "      <td>P Kumar</td>\n",
       "      <td>0.3</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>Kumar to McCullum, 1 wide, thats wide, as Prav...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2008</td>\n",
       "      <td>1</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>BB McCullum</td>\n",
       "      <td>SC Ganguly</td>\n",
       "      <td>P Kumar</td>\n",
       "      <td>0.3</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>Kumar to McCullum, no run, too straight, McCul...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2008</td>\n",
       "      <td>1</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>BB McCullum</td>\n",
       "      <td>SC Ganguly</td>\n",
       "      <td>P Kumar</td>\n",
       "      <td>0.4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>Kumar to McCullum, no run, steps down the trac...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   season match_id      venue  inning           batting_team  \\\n",
       "0    2008        1  Bangalore       1  Kolkata Knight Riders   \n",
       "1    2008        1  Bangalore       1  Kolkata Knight Riders   \n",
       "2    2008        1  Bangalore       1  Kolkata Knight Riders   \n",
       "3    2008        1  Bangalore       1  Kolkata Knight Riders   \n",
       "4    2008        1  Bangalore       1  Kolkata Knight Riders   \n",
       "\n",
       "                  bowling_team      batsman  non_striker   bowler  ball  ...  \\\n",
       "0  Royal Challengers Bangalore   SC Ganguly  BB McCullum  P Kumar   0.1  ...   \n",
       "1  Royal Challengers Bangalore  BB McCullum   SC Ganguly  P Kumar   0.2  ...   \n",
       "2  Royal Challengers Bangalore  BB McCullum   SC Ganguly  P Kumar   0.3  ...   \n",
       "3  Royal Challengers Bangalore  BB McCullum   SC Ganguly  P Kumar   0.3  ...   \n",
       "4  Royal Challengers Bangalore  BB McCullum   SC Ganguly  P Kumar   0.4  ...   \n",
       "\n",
       "   noball_runs  penalty_runs  batsman_runs  extra_runs  total_runs  \\\n",
       "0            0             0             0           1           1   \n",
       "1            0             0             0           0           0   \n",
       "2            0             0             0           1           1   \n",
       "3            0             0             0           0           0   \n",
       "4            0             0             0           0           0   \n",
       "\n",
       "   player_dismissed  dismissal_kind  fielder wickets  \\\n",
       "0                                                  0   \n",
       "1                                                  0   \n",
       "2                                                  0   \n",
       "3                                                  0   \n",
       "4                                                  0   \n",
       "\n",
       "                                          commentary  \n",
       "0  Kumar to Ganguly, 1 leg bye, and thats runs im...  \n",
       "1  Kumar to McCullum, no run, this one nips back ...  \n",
       "2  Kumar to McCullum, 1 wide, thats wide, as Prav...  \n",
       "3  Kumar to McCullum, no run, too straight, McCul...  \n",
       "4  Kumar to McCullum, no run, steps down the trac...  \n",
       "\n",
       "[5 rows x 23 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "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>season</th>\n",
       "      <th>match_id</th>\n",
       "      <th>venue</th>\n",
       "      <th>inning</th>\n",
       "      <th>batting_team</th>\n",
       "      <th>bowling_team</th>\n",
       "      <th>batsman</th>\n",
       "      <th>non_striker</th>\n",
       "      <th>bowler</th>\n",
       "      <th>ball</th>\n",
       "      <th>...</th>\n",
       "      <th>noball_runs</th>\n",
       "      <th>penalty_runs</th>\n",
       "      <th>batsman_runs</th>\n",
       "      <th>extra_runs</th>\n",
       "      <th>total_runs</th>\n",
       "      <th>player_dismissed</th>\n",
       "      <th>dismissal_kind</th>\n",
       "      <th>fielder</th>\n",
       "      <th>wickets</th>\n",
       "      <th>commentary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>131996</th>\n",
       "      <td>2019</td>\n",
       "      <td>final</td>\n",
       "      <td>Hyderabad</td>\n",
       "      <td>2</td>\n",
       "      <td>Chennai Super Kings</td>\n",
       "      <td>Mumbai Indians</td>\n",
       "      <td>RA Jadeja</td>\n",
       "      <td>SR Watson</td>\n",
       "      <td>SL Malinga</td>\n",
       "      <td>19.2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>5</td>\n",
       "      <td>Malinga to Jadeja, 1 run, low full toss on mid...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131997</th>\n",
       "      <td>2019</td>\n",
       "      <td>final</td>\n",
       "      <td>Hyderabad</td>\n",
       "      <td>2</td>\n",
       "      <td>Chennai Super Kings</td>\n",
       "      <td>Mumbai Indians</td>\n",
       "      <td>SR Watson</td>\n",
       "      <td>RA Jadeja</td>\n",
       "      <td>SL Malinga</td>\n",
       "      <td>19.3</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>5</td>\n",
       "      <td>Malinga to Watson, 2 runs, yorker, just outsid...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131998</th>\n",
       "      <td>2019</td>\n",
       "      <td>final</td>\n",
       "      <td>Hyderabad</td>\n",
       "      <td>2</td>\n",
       "      <td>Chennai Super Kings</td>\n",
       "      <td>Mumbai Indians</td>\n",
       "      <td>SR Watson</td>\n",
       "      <td>RA Jadeja</td>\n",
       "      <td>SL Malinga</td>\n",
       "      <td>19.4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>SR Watson</td>\n",
       "      <td>run out</td>\n",
       "      <td>KH Pandya</td>\n",
       "      <td>6</td>\n",
       "      <td>Malinga to Watson, 1 run, OUT, yorker, just ou...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131999</th>\n",
       "      <td>2019</td>\n",
       "      <td>final</td>\n",
       "      <td>Hyderabad</td>\n",
       "      <td>2</td>\n",
       "      <td>Chennai Super Kings</td>\n",
       "      <td>Mumbai Indians</td>\n",
       "      <td>SN Thakur</td>\n",
       "      <td>RA Jadeja</td>\n",
       "      <td>SL Malinga</td>\n",
       "      <td>19.5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>6</td>\n",
       "      <td>Malinga to Thakur, 2 runs, full toss on leg, a...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>132000</th>\n",
       "      <td>2019</td>\n",
       "      <td>final</td>\n",
       "      <td>Hyderabad</td>\n",
       "      <td>2</td>\n",
       "      <td>Chennai Super Kings</td>\n",
       "      <td>Mumbai Indians</td>\n",
       "      <td>SN Thakur</td>\n",
       "      <td>RA Jadeja</td>\n",
       "      <td>SL Malinga</td>\n",
       "      <td>19.6</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>SN Thakur</td>\n",
       "      <td>lbw</td>\n",
       "      <td></td>\n",
       "      <td>7</td>\n",
       "      <td>Malinga to Thakur, OUT, Mumbai Indians win IPL...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        season match_id      venue  inning         batting_team  \\\n",
       "131996    2019    final  Hyderabad       2  Chennai Super Kings   \n",
       "131997    2019    final  Hyderabad       2  Chennai Super Kings   \n",
       "131998    2019    final  Hyderabad       2  Chennai Super Kings   \n",
       "131999    2019    final  Hyderabad       2  Chennai Super Kings   \n",
       "132000    2019    final  Hyderabad       2  Chennai Super Kings   \n",
       "\n",
       "          bowling_team    batsman non_striker      bowler  ball  ...  \\\n",
       "131996  Mumbai Indians  RA Jadeja   SR Watson  SL Malinga  19.2  ...   \n",
       "131997  Mumbai Indians  SR Watson   RA Jadeja  SL Malinga  19.3  ...   \n",
       "131998  Mumbai Indians  SR Watson   RA Jadeja  SL Malinga  19.4  ...   \n",
       "131999  Mumbai Indians  SN Thakur   RA Jadeja  SL Malinga  19.5  ...   \n",
       "132000  Mumbai Indians  SN Thakur   RA Jadeja  SL Malinga  19.6  ...   \n",
       "\n",
       "        noball_runs  penalty_runs  batsman_runs  extra_runs  total_runs  \\\n",
       "131996            0             0             1           0           1   \n",
       "131997            0             0             2           0           2   \n",
       "131998            0             0             1           0           1   \n",
       "131999            0             0             2           0           2   \n",
       "132000            0             0             0           0           0   \n",
       "\n",
       "        player_dismissed  dismissal_kind    fielder wickets  \\\n",
       "131996                                                    5   \n",
       "131997                                                    5   \n",
       "131998         SR Watson         run out  KH Pandya       6   \n",
       "131999                                                    6   \n",
       "132000         SN Thakur             lbw                  7   \n",
       "\n",
       "                                               commentary  \n",
       "131996  Malinga to Jadeja, 1 run, low full toss on mid...  \n",
       "131997  Malinga to Watson, 2 runs, yorker, just outsid...  \n",
       "131998  Malinga to Watson, 1 run, OUT, yorker, just ou...  \n",
       "131999  Malinga to Thakur, 2 runs, full toss on leg, a...  \n",
       "132000  Malinga to Thakur, OUT, Mumbai Indians win IPL...  \n",
       "\n",
       "[5 rows x 23 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(132001, 23)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['season', 'match_id', 'venue', 'inning', 'batting_team', 'bowling_team',\n",
       "       'batsman', 'non_striker', 'bowler', 'ball', 'wide_runs', 'bye_runs',\n",
       "       'legbye_runs', 'noball_runs', 'penalty_runs', 'batsman_runs',\n",
       "       'extra_runs', 'total_runs', 'player_dismissed', 'dismissal_kind',\n",
       "       'fielder', 'wickets', 'commentary'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Check for Missing Values/NAs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "season              0\n",
       "match_id            0\n",
       "venue               0\n",
       "inning              0\n",
       "batting_team        0\n",
       "bowling_team        0\n",
       "batsman             0\n",
       "non_striker         0\n",
       "bowler              0\n",
       "ball                0\n",
       "wide_runs           0\n",
       "bye_runs            0\n",
       "legbye_runs         0\n",
       "noball_runs         0\n",
       "penalty_runs        0\n",
       "batsman_runs        0\n",
       "extra_runs          0\n",
       "total_runs          0\n",
       "player_dismissed    0\n",
       "dismissal_kind      0\n",
       "fielder             0\n",
       "wickets             0\n",
       "commentary          0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isna().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Remove Unnecessary Columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "columns_to_remove = [\"player_dismissed\", \"fielder\", \"penalty_runs\"]\n",
    "df.drop(labels=columns_to_remove, axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "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>season</th>\n",
       "      <th>match_id</th>\n",
       "      <th>venue</th>\n",
       "      <th>inning</th>\n",
       "      <th>batting_team</th>\n",
       "      <th>bowling_team</th>\n",
       "      <th>batsman</th>\n",
       "      <th>non_striker</th>\n",
       "      <th>bowler</th>\n",
       "      <th>ball</th>\n",
       "      <th>wide_runs</th>\n",
       "      <th>bye_runs</th>\n",
       "      <th>legbye_runs</th>\n",
       "      <th>noball_runs</th>\n",
       "      <th>batsman_runs</th>\n",
       "      <th>extra_runs</th>\n",
       "      <th>total_runs</th>\n",
       "      <th>dismissal_kind</th>\n",
       "      <th>wickets</th>\n",
       "      <th>commentary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2008</td>\n",
       "      <td>1</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>SC Ganguly</td>\n",
       "      <td>BB McCullum</td>\n",
       "      <td>P Kumar</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>Kumar to Ganguly, 1 leg bye, and thats runs im...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2008</td>\n",
       "      <td>1</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>BB McCullum</td>\n",
       "      <td>SC Ganguly</td>\n",
       "      <td>P Kumar</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>Kumar to McCullum, no run, this one nips back ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2008</td>\n",
       "      <td>1</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>BB McCullum</td>\n",
       "      <td>SC Ganguly</td>\n",
       "      <td>P Kumar</td>\n",
       "      <td>0.3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>Kumar to McCullum, 1 wide, thats wide, as Prav...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2008</td>\n",
       "      <td>1</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>BB McCullum</td>\n",
       "      <td>SC Ganguly</td>\n",
       "      <td>P Kumar</td>\n",
       "      <td>0.3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>Kumar to McCullum, no run, too straight, McCul...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2008</td>\n",
       "      <td>1</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>BB McCullum</td>\n",
       "      <td>SC Ganguly</td>\n",
       "      <td>P Kumar</td>\n",
       "      <td>0.4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>Kumar to McCullum, no run, steps down the trac...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   season match_id      venue  inning           batting_team  \\\n",
       "0    2008        1  Bangalore       1  Kolkata Knight Riders   \n",
       "1    2008        1  Bangalore       1  Kolkata Knight Riders   \n",
       "2    2008        1  Bangalore       1  Kolkata Knight Riders   \n",
       "3    2008        1  Bangalore       1  Kolkata Knight Riders   \n",
       "4    2008        1  Bangalore       1  Kolkata Knight Riders   \n",
       "\n",
       "                  bowling_team      batsman  non_striker   bowler  ball  \\\n",
       "0  Royal Challengers Bangalore   SC Ganguly  BB McCullum  P Kumar   0.1   \n",
       "1  Royal Challengers Bangalore  BB McCullum   SC Ganguly  P Kumar   0.2   \n",
       "2  Royal Challengers Bangalore  BB McCullum   SC Ganguly  P Kumar   0.3   \n",
       "3  Royal Challengers Bangalore  BB McCullum   SC Ganguly  P Kumar   0.3   \n",
       "4  Royal Challengers Bangalore  BB McCullum   SC Ganguly  P Kumar   0.4   \n",
       "\n",
       "   wide_runs  bye_runs  legbye_runs  noball_runs  batsman_runs  extra_runs  \\\n",
       "0          0         0            1            0             0           1   \n",
       "1          0         0            0            0             0           0   \n",
       "2          1         0            0            0             0           1   \n",
       "3          0         0            0            0             0           0   \n",
       "4          0         0            0            0             0           0   \n",
       "\n",
       "   total_runs dismissal_kind  wickets  \\\n",
       "0           1                       0   \n",
       "1           0                       0   \n",
       "2           1                       0   \n",
       "3           0                       0   \n",
       "4           0                       0   \n",
       "\n",
       "                                          commentary  \n",
       "0  Kumar to Ganguly, 1 leg bye, and thats runs im...  \n",
       "1  Kumar to McCullum, no run, this one nips back ...  \n",
       "2  Kumar to McCullum, 1 wide, thats wide, as Prav...  \n",
       "3  Kumar to McCullum, no run, too straight, McCul...  \n",
       "4  Kumar to McCullum, no run, steps down the trac...  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature Engineering `bowler_type`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the code cell below, we have created a custom list of all spin bowlers for each team. We have then modified the `bowler` column in the dataframe which will be used to identify the type of bowler, either \"spinner\" or \"pacer\". Therefore, instead of dropping the `bowler` column, we have created a new column `bowler_type` which will be used for our analysis and enhance the predictive power of our models."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Replacing all spinner names with 'spinner' and all pacer names with 'pacer' under `bowler` column\n",
    "\n",
    "rcb_spin = [\n",
    "    \"SB Joshi\",\n",
    "    \"KP Appanna\",\n",
    "    \"A Kumble\",\n",
    "    \"J Arunkumar\",\n",
    "    \"Abdur Razzak\",\n",
    "    \"CL White\",\n",
    "    \"TM Dilshan\",\n",
    "    \"KP Pietersen\",\n",
    "    \"RE van der Merwe\",\n",
    "    \"S Sriram\",\n",
    "    \"DL Vettori\",\n",
    "    \"KB Arun Karthik\",\n",
    "    \"CH Gayle\",\n",
    "    \"S Sriram\",\n",
    "    \"AA Kazi\",\n",
    "    \"R Ninan\",\n",
    "    \"J Syed Mohammad\",\n",
    "    \"M Muralitharan\",\n",
    "    \"M Kartik\",\n",
    "    \"P R Barman\",\n",
    "    \"Sachin Baby\",\n",
    "    \"YS Chahal\",\n",
    "    \"Yuvraj Singh\",\n",
    "    \"SB Jakati\",\n",
    "    \"Iqbal Abdulla\",\n",
    "    \"T Shamsi\",\n",
    "    \"Parvez Rasool\",\n",
    "    \"S Baby\",\n",
    "    \"TM Head\",\n",
    "    \"KM Jadhav\",\n",
    "    \"P Negi\",\n",
    "    \"S Badree\",\n",
    "    \"Washington Sundar\",\n",
    "    \"M Ali\",\n",
    "    \"M Ashwin\",\n",
    "]\n",
    "\n",
    "mi_spin = [\n",
    "    \"A Dananjaya\",\n",
    "    \"AG Murtaza\",\n",
    "    \"A Roy\",\n",
    "    \"GJ Maxwell\",\n",
    "    \"Harbhajan Singh\",\n",
    "    \"J Suchith\",\n",
    "    \"J Yadav\",\n",
    "    \"JP Duminy\",\n",
    "    \"KH Pandya\",\n",
    "    \"KV Sharma\",\n",
    "    \"M Markande\",\n",
    "    \"N Rana\",\n",
    "    \"PP Ojha\",\n",
    "    \"RD Chahar\",\n",
    "    \"RG Sharma\",\n",
    "    \"RJ Peterson\",\n",
    "    \"RW Price\",\n",
    "    \"S Gopal\",\n",
    "    \"SD Chitnis\",\n",
    "    \"ST Jayasuriya\",\n",
    "    \"TL Suman\",\n",
    "    \"VS Yeligati\",\n",
    "]\n",
    "\n",
    "kkr_spin = [\n",
    "    \"BAW Mendis\",\n",
    "    \"BJ Hodge\",\n",
    "    \"DJ Hussey\",\n",
    "    \"GB Hogg\",\n",
    "    \"J Botha\",\n",
    "    \"KC Cariappa\",\n",
    "    \"M Kartik\",\n",
    "    \"CH Gayle\",\n",
    "    \"Iqbal Abdulla\",\n",
    "    \"Kuldeep Yadav\",\n",
    "    \"MB Parmar\",\n",
    "    \"MK Tiwary\",\n",
    "    \"Mohammad Hafeez\",\n",
    "    \"N Rana\",\n",
    "    \"PP Chawla\",\n",
    "    \"RS Gavaskar\",\n",
    "    \"S Ladda\",\n",
    "    \"Shakib Al Hasan\",\n",
    "    \"SMSM Senanayake\",\n",
    "    \"SP Narine\",\n",
    "    \"YK Pathan\",\n",
    "]\n",
    "\n",
    "csk_spin = [\n",
    "    \"DJ Hussey\",\n",
    "    \"F du Plessis\",\n",
    "    \"Harbhajan Singh\",\n",
    "    \"Imran Tahir\",\n",
    "    \"RA Jadeja\",\n",
    "    \"KV Sharma\",\n",
    "    \"M Santner\",\n",
    "    \"M Muralitharan\",\n",
    "    \"P Negi\",\n",
    "    \"R Ashwin\",\n",
    "    \"S Badree\",\n",
    "    \"S Randiv\",\n",
    "    \"S Vidyut\",\n",
    "    \"SB Jakati\",\n",
    "    \"SK Raina\",\n",
    "    \"SMSM Senanayake\",\n",
    "]\n",
    "\n",
    "dc_spin = [\n",
    "    \"A Mishra\",\n",
    "    \"AA Jhunjhunwala\",\n",
    "    \"Ankit Sharma\",\n",
    "    \"CL White\",\n",
    "    \"PP Ojha\",\n",
    "    \"DB Ravi Teja\",\n",
    "    \"JP Duminy\",\n",
    "    \"LPC Silva\",\n",
    "    \"R Sharma\",\n",
    "    \"RG Sharma\",\n",
    "    \"S Dhawan\",\n",
    "    \"Shahid Afridi\",\n",
    "    \"TL Suman\",\n",
    "    \"Y Venugopal Rao\",\n",
    "]\n",
    "\n",
    "dd_spin = [\n",
    "    \"A Mishra\",\n",
    "    \"AJ Finch\",\n",
    "    \"AR Patel\",\n",
    "    \"BMAJ Mendis\",\n",
    "    \"DL Vettori\",\n",
    "    \"GJ Maxwell\",\n",
    "    \"H Vihari\",\n",
    "    \"Imran Tahir\",\n",
    "    \"J Suchith\",\n",
    "    \"J Botha\",\n",
    "    \"J Yadav\",\n",
    "    \"JP Duminy\",\n",
    "    \"KP Pietersen\",\n",
    "    \"MK Tiwary\",\n",
    "    \"P Negi\",\n",
    "    \"R Sharma\",\n",
    "    \"R Tewatia\",\n",
    "    \"RE van der Merwe\",\n",
    "    \"S Ladda\",\n",
    "    \"S Lamichhane\",\n",
    "    \"S Nadeem\",\n",
    "    \"S Sriram\",\n",
    "    \"Shoaib Malik\",\n",
    "    \"TM Dilshan\",\n",
    "    \"Sunny Gupta\",\n",
    "    \"V Sehwag\",\n",
    "    \"Y Venugopal Rao\",\n",
    "    \"Yuvraj Singh\",\n",
    "    \"Y Nagar\",\n",
    "]\n",
    "\n",
    "srh_spin = [\n",
    "    \"A Mishra\",\n",
    "    \"Ankit Sharma\",\n",
    "    \"Bipul Sharma\",\n",
    "    \"CL White\",\n",
    "    \"DJ Hooda\",\n",
    "    \"KS Williamson\",\n",
    "    \"KV Sharma\",\n",
    "    \"Mohammad Nabi\",\n",
    "    \"Parvez Rasool\",\n",
    "    \"Rashid Khan\",\n",
    "    \"S Nadeem\",\n",
    "    \"Shakib Al Hasan\",\n",
    "    \"Y Venugopal Rao\",\n",
    "    \"YK Pathan\",\n",
    "    \"Yuvraj Singh\",\n",
    "]\n",
    "\n",
    "rr_spin = [\n",
    "    \"A Chandila\",\n",
    "    \"AA Chavan\",\n",
    "    \"AA Jhunjhunwala\",\n",
    "    \"AC Voges\",\n",
    "    \"AJ Finch\",\n",
    "    \"AL Menaria\",\n",
    "    \"Ankit Sharma\",\n",
    "    \"AS Raut\",\n",
    "    \"BJ Hodge\",\n",
    "    \"D Short\",\n",
    "    \"DJ Hooda\",\n",
    "    \"D Salunkhe\",\n",
    "    \"GB Hogg\",\n",
    "    \"I Sodhi\",\n",
    "    \"J Botha\",\n",
    "    \"K Gowtham\",\n",
    "    \"L Livingstone\",\n",
    "    \"LRPL Taylor\",\n",
    "    \"M Lomror\",\n",
    "    \"PV Tambe\",\n",
    "    \"ND Doshi\",\n",
    "    \"R Parag\",\n",
    "    \"R Tewatia\",\n",
    "    \"S Badree\",\n",
    "    \"S Gopal\",\n",
    "    \"RA Jadeja\",\n",
    "    \"S Midhun\",\n",
    "    \"SK Warne\",\n",
    "    \"YK Pathan\",\n",
    "]\n",
    "\n",
    "punj_spin = [\n",
    "    \"AC Gilchrist\",\n",
    "    \"AR Patel\",\n",
    "    \"BA Bhatt\",\n",
    "    \"Bipul Sharma\",\n",
    "    \"DJ Hussey\",\n",
    "    \"GJ Maxwell\",\n",
    "    \"Gurkeerat Singh\" \"H Brar\",\n",
    "    \"Karanveer Singh\",\n",
    "    \"KC Cariappa\",\n",
    "    \"M Ashwin\",\n",
    "    \"M Kartik\",\n",
    "    \"M Ur Rahman\",\n",
    "    \"M Vijay\",\n",
    "    \"MK Tiwary\",\n",
    "    \"P Sahu\",\n",
    "    \"PP Chawla\",\n",
    "    \"R Ashwin\",\n",
    "    \"R Tewatia\",\n",
    "    \"RR Powar\",\n",
    "    \"Shivam Sharma\",\n",
    "    \"SN Khan\",\n",
    "    \"Swapnil Singh\",\n",
    "    \"V Chakravarthy\",\n",
    "    \"Yuvraj Singh\",\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Replacing all spin bowlers' names with 'spinner' under `bowler` column\n",
    "\n",
    "df.loc[\n",
    "    (df.bowling_team == \"Royal Challengers Bangalore\") & (df.bowler.isin(rcb_spin)),\n",
    "    \"bowler\",\n",
    "] = \"spinner\"\n",
    "df.loc[\n",
    "    (df.bowling_team == \"Delhi Capitals\") & (df.bowler.isin(dd_spin)), \"bowler\"\n",
    "] = \"spinner\"\n",
    "df.loc[\n",
    "    (df.bowling_team == \"Delhi Daredevils\") & (df.bowler.isin(dd_spin)), \"bowler\"\n",
    "] = \"spinner\"\n",
    "df.loc[\n",
    "    (df.bowling_team == \"Mumbai Indians\") & (df.bowler.isin(mi_spin)), \"bowler\"\n",
    "] = \"spinner\"\n",
    "df.loc[\n",
    "    (df.bowling_team == \"Chennai Super Kings\") & (df.bowler.isin(csk_spin)), \"bowler\"\n",
    "] = \"spinner\"\n",
    "df.loc[\n",
    "    (df.bowling_team == \"Kings XI Punjab\") & (df.bowler.isin(punj_spin)), \"bowler\"\n",
    "] = \"spinner\"\n",
    "df.loc[\n",
    "    (df.bowling_team == \"Sunrisers Hyderabad\") & (df.bowler.isin(srh_spin)), \"bowler\"\n",
    "] = \"spinner\"\n",
    "df.loc[\n",
    "    (df.bowling_team == \"Kolkata Knight Riders\") & (df.bowler.isin(kkr_spin)), \"bowler\"\n",
    "] = \"spinner\"\n",
    "df.loc[\n",
    "    (df.bowling_team == \"Deccan Chargers\") & (df.bowler.isin(dc_spin)), \"bowler\"\n",
    "] = \"spinner\"\n",
    "df.loc[\n",
    "    (df.bowling_team == \"Rajasthan Royals\") & (df.bowler.isin(rr_spin)), \"bowler\"\n",
    "] = \"spinner\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Replacing all fast/pace bowlers' names with 'pacer' under `bowler` column\n",
    "\n",
    "l = [\"spinner\"]\n",
    "df.loc[~df.bowler.isin(l), \"bowler\"] = \"pacer\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "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>season</th>\n",
       "      <th>match_id</th>\n",
       "      <th>venue</th>\n",
       "      <th>inning</th>\n",
       "      <th>batting_team</th>\n",
       "      <th>bowling_team</th>\n",
       "      <th>batsman</th>\n",
       "      <th>non_striker</th>\n",
       "      <th>bowler</th>\n",
       "      <th>ball</th>\n",
       "      <th>wide_runs</th>\n",
       "      <th>bye_runs</th>\n",
       "      <th>legbye_runs</th>\n",
       "      <th>noball_runs</th>\n",
       "      <th>batsman_runs</th>\n",
       "      <th>extra_runs</th>\n",
       "      <th>total_runs</th>\n",
       "      <th>dismissal_kind</th>\n",
       "      <th>wickets</th>\n",
       "      <th>commentary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2008</td>\n",
       "      <td>1</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>SC Ganguly</td>\n",
       "      <td>BB McCullum</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>Kumar to Ganguly, 1 leg bye, and thats runs im...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2008</td>\n",
       "      <td>1</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>BB McCullum</td>\n",
       "      <td>SC Ganguly</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>Kumar to McCullum, no run, this one nips back ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2008</td>\n",
       "      <td>1</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>BB McCullum</td>\n",
       "      <td>SC Ganguly</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>Kumar to McCullum, 1 wide, thats wide, as Prav...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2008</td>\n",
       "      <td>1</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>BB McCullum</td>\n",
       "      <td>SC Ganguly</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>Kumar to McCullum, no run, too straight, McCul...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2008</td>\n",
       "      <td>1</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>BB McCullum</td>\n",
       "      <td>SC Ganguly</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>Kumar to McCullum, no run, steps down the trac...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131996</th>\n",
       "      <td>2019</td>\n",
       "      <td>final</td>\n",
       "      <td>Hyderabad</td>\n",
       "      <td>2</td>\n",
       "      <td>Chennai Super Kings</td>\n",
       "      <td>Mumbai Indians</td>\n",
       "      <td>RA Jadeja</td>\n",
       "      <td>SR Watson</td>\n",
       "      <td>pacer</td>\n",
       "      <td>19.2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td></td>\n",
       "      <td>5</td>\n",
       "      <td>Malinga to Jadeja, 1 run, low full toss on mid...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131997</th>\n",
       "      <td>2019</td>\n",
       "      <td>final</td>\n",
       "      <td>Hyderabad</td>\n",
       "      <td>2</td>\n",
       "      <td>Chennai Super Kings</td>\n",
       "      <td>Mumbai Indians</td>\n",
       "      <td>SR Watson</td>\n",
       "      <td>RA Jadeja</td>\n",
       "      <td>pacer</td>\n",
       "      <td>19.3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td></td>\n",
       "      <td>5</td>\n",
       "      <td>Malinga to Watson, 2 runs, yorker, just outsid...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131998</th>\n",
       "      <td>2019</td>\n",
       "      <td>final</td>\n",
       "      <td>Hyderabad</td>\n",
       "      <td>2</td>\n",
       "      <td>Chennai Super Kings</td>\n",
       "      <td>Mumbai Indians</td>\n",
       "      <td>SR Watson</td>\n",
       "      <td>RA Jadeja</td>\n",
       "      <td>pacer</td>\n",
       "      <td>19.4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>run out</td>\n",
       "      <td>6</td>\n",
       "      <td>Malinga to Watson, 1 run, OUT, yorker, just ou...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131999</th>\n",
       "      <td>2019</td>\n",
       "      <td>final</td>\n",
       "      <td>Hyderabad</td>\n",
       "      <td>2</td>\n",
       "      <td>Chennai Super Kings</td>\n",
       "      <td>Mumbai Indians</td>\n",
       "      <td>SN Thakur</td>\n",
       "      <td>RA Jadeja</td>\n",
       "      <td>pacer</td>\n",
       "      <td>19.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td></td>\n",
       "      <td>6</td>\n",
       "      <td>Malinga to Thakur, 2 runs, full toss on leg, a...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>132000</th>\n",
       "      <td>2019</td>\n",
       "      <td>final</td>\n",
       "      <td>Hyderabad</td>\n",
       "      <td>2</td>\n",
       "      <td>Chennai Super Kings</td>\n",
       "      <td>Mumbai Indians</td>\n",
       "      <td>SN Thakur</td>\n",
       "      <td>RA Jadeja</td>\n",
       "      <td>pacer</td>\n",
       "      <td>19.6</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>lbw</td>\n",
       "      <td>7</td>\n",
       "      <td>Malinga to Thakur, OUT, Mumbai Indians win IPL...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>86617 rows × 20 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        season match_id      venue  inning           batting_team  \\\n",
       "0         2008        1  Bangalore       1  Kolkata Knight Riders   \n",
       "1         2008        1  Bangalore       1  Kolkata Knight Riders   \n",
       "2         2008        1  Bangalore       1  Kolkata Knight Riders   \n",
       "3         2008        1  Bangalore       1  Kolkata Knight Riders   \n",
       "4         2008        1  Bangalore       1  Kolkata Knight Riders   \n",
       "...        ...      ...        ...     ...                    ...   \n",
       "131996    2019    final  Hyderabad       2    Chennai Super Kings   \n",
       "131997    2019    final  Hyderabad       2    Chennai Super Kings   \n",
       "131998    2019    final  Hyderabad       2    Chennai Super Kings   \n",
       "131999    2019    final  Hyderabad       2    Chennai Super Kings   \n",
       "132000    2019    final  Hyderabad       2    Chennai Super Kings   \n",
       "\n",
       "                       bowling_team      batsman  non_striker bowler  ball  \\\n",
       "0       Royal Challengers Bangalore   SC Ganguly  BB McCullum  pacer   0.1   \n",
       "1       Royal Challengers Bangalore  BB McCullum   SC Ganguly  pacer   0.2   \n",
       "2       Royal Challengers Bangalore  BB McCullum   SC Ganguly  pacer   0.3   \n",
       "3       Royal Challengers Bangalore  BB McCullum   SC Ganguly  pacer   0.3   \n",
       "4       Royal Challengers Bangalore  BB McCullum   SC Ganguly  pacer   0.4   \n",
       "...                             ...          ...          ...    ...   ...   \n",
       "131996               Mumbai Indians    RA Jadeja    SR Watson  pacer  19.2   \n",
       "131997               Mumbai Indians    SR Watson    RA Jadeja  pacer  19.3   \n",
       "131998               Mumbai Indians    SR Watson    RA Jadeja  pacer  19.4   \n",
       "131999               Mumbai Indians    SN Thakur    RA Jadeja  pacer  19.5   \n",
       "132000               Mumbai Indians    SN Thakur    RA Jadeja  pacer  19.6   \n",
       "\n",
       "        wide_runs  bye_runs  legbye_runs  noball_runs  batsman_runs  \\\n",
       "0               0         0            1            0             0   \n",
       "1               0         0            0            0             0   \n",
       "2               1         0            0            0             0   \n",
       "3               0         0            0            0             0   \n",
       "4               0         0            0            0             0   \n",
       "...           ...       ...          ...          ...           ...   \n",
       "131996          0         0            0            0             1   \n",
       "131997          0         0            0            0             2   \n",
       "131998          0         0            0            0             1   \n",
       "131999          0         0            0            0             2   \n",
       "132000          0         0            0            0             0   \n",
       "\n",
       "        extra_runs  total_runs dismissal_kind  wickets  \\\n",
       "0                1           1                       0   \n",
       "1                0           0                       0   \n",
       "2                1           1                       0   \n",
       "3                0           0                       0   \n",
       "4                0           0                       0   \n",
       "...            ...         ...            ...      ...   \n",
       "131996           0           1                       5   \n",
       "131997           0           2                       5   \n",
       "131998           0           1        run out        6   \n",
       "131999           0           2                       6   \n",
       "132000           0           0            lbw        7   \n",
       "\n",
       "                                               commentary  \n",
       "0       Kumar to Ganguly, 1 leg bye, and thats runs im...  \n",
       "1       Kumar to McCullum, no run, this one nips back ...  \n",
       "2       Kumar to McCullum, 1 wide, thats wide, as Prav...  \n",
       "3       Kumar to McCullum, no run, too straight, McCul...  \n",
       "4       Kumar to McCullum, no run, steps down the trac...  \n",
       "...                                                   ...  \n",
       "131996  Malinga to Jadeja, 1 run, low full toss on mid...  \n",
       "131997  Malinga to Watson, 2 runs, yorker, just outsid...  \n",
       "131998  Malinga to Watson, 1 run, OUT, yorker, just ou...  \n",
       "131999  Malinga to Thakur, 2 runs, full toss on leg, a...  \n",
       "132000  Malinga to Thakur, OUT, Mumbai Indians win IPL...  \n",
       "\n",
       "[86617 rows x 20 columns]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[df.bowler == \"pacer\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "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>season</th>\n",
       "      <th>match_id</th>\n",
       "      <th>venue</th>\n",
       "      <th>inning</th>\n",
       "      <th>batting_team</th>\n",
       "      <th>bowling_team</th>\n",
       "      <th>batsman</th>\n",
       "      <th>non_striker</th>\n",
       "      <th>bowler</th>\n",
       "      <th>ball</th>\n",
       "      <th>wide_runs</th>\n",
       "      <th>bye_runs</th>\n",
       "      <th>legbye_runs</th>\n",
       "      <th>noball_runs</th>\n",
       "      <th>batsman_runs</th>\n",
       "      <th>extra_runs</th>\n",
       "      <th>total_runs</th>\n",
       "      <th>dismissal_kind</th>\n",
       "      <th>wickets</th>\n",
       "      <th>commentary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>56</th>\n",
       "      <td>2008</td>\n",
       "      <td>1</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>RT Ponting</td>\n",
       "      <td>BB McCullum</td>\n",
       "      <td>spinner</td>\n",
       "      <td>9.1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td></td>\n",
       "      <td>1</td>\n",
       "      <td>Joshi to Ponting, 1 run, some good turn, spins...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57</th>\n",
       "      <td>2008</td>\n",
       "      <td>1</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>BB McCullum</td>\n",
       "      <td>RT Ponting</td>\n",
       "      <td>spinner</td>\n",
       "      <td>9.2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td></td>\n",
       "      <td>1</td>\n",
       "      <td>Joshi to McCullum, 1 run, too straight and fla...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>58</th>\n",
       "      <td>2008</td>\n",
       "      <td>1</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>RT Ponting</td>\n",
       "      <td>BB McCullum</td>\n",
       "      <td>spinner</td>\n",
       "      <td>9.3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td></td>\n",
       "      <td>1</td>\n",
       "      <td>Joshi to Ponting, 1 run, fired on leg again an...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59</th>\n",
       "      <td>2008</td>\n",
       "      <td>1</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>BB McCullum</td>\n",
       "      <td>RT Ponting</td>\n",
       "      <td>spinner</td>\n",
       "      <td>9.4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>1</td>\n",
       "      <td>Joshi to McCullum, no run, bit of flight there...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60</th>\n",
       "      <td>2008</td>\n",
       "      <td>1</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>BB McCullum</td>\n",
       "      <td>RT Ponting</td>\n",
       "      <td>spinner</td>\n",
       "      <td>9.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td></td>\n",
       "      <td>1</td>\n",
       "      <td>Joshi to McCullum, SIX runs, snuff is snuff, s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131984</th>\n",
       "      <td>2019</td>\n",
       "      <td>final</td>\n",
       "      <td>Hyderabad</td>\n",
       "      <td>2</td>\n",
       "      <td>Chennai Super Kings</td>\n",
       "      <td>Mumbai Indians</td>\n",
       "      <td>SR Watson</td>\n",
       "      <td>DJ Bravo</td>\n",
       "      <td>spinner</td>\n",
       "      <td>17.2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td></td>\n",
       "      <td>4</td>\n",
       "      <td>KH Pandya to Watson, SIX runs, hammered! Full,...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131985</th>\n",
       "      <td>2019</td>\n",
       "      <td>final</td>\n",
       "      <td>Hyderabad</td>\n",
       "      <td>2</td>\n",
       "      <td>Chennai Super Kings</td>\n",
       "      <td>Mumbai Indians</td>\n",
       "      <td>SR Watson</td>\n",
       "      <td>DJ Bravo</td>\n",
       "      <td>spinner</td>\n",
       "      <td>17.3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td></td>\n",
       "      <td>4</td>\n",
       "      <td>KH Pandya to Watson, SIX runs, over Krunal's h...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131986</th>\n",
       "      <td>2019</td>\n",
       "      <td>final</td>\n",
       "      <td>Hyderabad</td>\n",
       "      <td>2</td>\n",
       "      <td>Chennai Super Kings</td>\n",
       "      <td>Mumbai Indians</td>\n",
       "      <td>SR Watson</td>\n",
       "      <td>DJ Bravo</td>\n",
       "      <td>spinner</td>\n",
       "      <td>17.4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td></td>\n",
       "      <td>4</td>\n",
       "      <td>KH Pandya to Watson, SIX runs, into the second...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131987</th>\n",
       "      <td>2019</td>\n",
       "      <td>final</td>\n",
       "      <td>Hyderabad</td>\n",
       "      <td>2</td>\n",
       "      <td>Chennai Super Kings</td>\n",
       "      <td>Mumbai Indians</td>\n",
       "      <td>SR Watson</td>\n",
       "      <td>DJ Bravo</td>\n",
       "      <td>spinner</td>\n",
       "      <td>17.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>4</td>\n",
       "      <td>KH Pandya to Watson, no run, full, fast, and w...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131988</th>\n",
       "      <td>2019</td>\n",
       "      <td>final</td>\n",
       "      <td>Hyderabad</td>\n",
       "      <td>2</td>\n",
       "      <td>Chennai Super Kings</td>\n",
       "      <td>Mumbai Indians</td>\n",
       "      <td>SR Watson</td>\n",
       "      <td>DJ Bravo</td>\n",
       "      <td>spinner</td>\n",
       "      <td>17.6</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td></td>\n",
       "      <td>4</td>\n",
       "      <td>KH Pandya to Watson, 1 run, overpitched just o...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>45384 rows × 20 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        season match_id      venue  inning           batting_team  \\\n",
       "56        2008        1  Bangalore       1  Kolkata Knight Riders   \n",
       "57        2008        1  Bangalore       1  Kolkata Knight Riders   \n",
       "58        2008        1  Bangalore       1  Kolkata Knight Riders   \n",
       "59        2008        1  Bangalore       1  Kolkata Knight Riders   \n",
       "60        2008        1  Bangalore       1  Kolkata Knight Riders   \n",
       "...        ...      ...        ...     ...                    ...   \n",
       "131984    2019    final  Hyderabad       2    Chennai Super Kings   \n",
       "131985    2019    final  Hyderabad       2    Chennai Super Kings   \n",
       "131986    2019    final  Hyderabad       2    Chennai Super Kings   \n",
       "131987    2019    final  Hyderabad       2    Chennai Super Kings   \n",
       "131988    2019    final  Hyderabad       2    Chennai Super Kings   \n",
       "\n",
       "                       bowling_team      batsman  non_striker   bowler  ball  \\\n",
       "56      Royal Challengers Bangalore   RT Ponting  BB McCullum  spinner   9.1   \n",
       "57      Royal Challengers Bangalore  BB McCullum   RT Ponting  spinner   9.2   \n",
       "58      Royal Challengers Bangalore   RT Ponting  BB McCullum  spinner   9.3   \n",
       "59      Royal Challengers Bangalore  BB McCullum   RT Ponting  spinner   9.4   \n",
       "60      Royal Challengers Bangalore  BB McCullum   RT Ponting  spinner   9.5   \n",
       "...                             ...          ...          ...      ...   ...   \n",
       "131984               Mumbai Indians    SR Watson     DJ Bravo  spinner  17.2   \n",
       "131985               Mumbai Indians    SR Watson     DJ Bravo  spinner  17.3   \n",
       "131986               Mumbai Indians    SR Watson     DJ Bravo  spinner  17.4   \n",
       "131987               Mumbai Indians    SR Watson     DJ Bravo  spinner  17.5   \n",
       "131988               Mumbai Indians    SR Watson     DJ Bravo  spinner  17.6   \n",
       "\n",
       "        wide_runs  bye_runs  legbye_runs  noball_runs  batsman_runs  \\\n",
       "56              0         0            0            0             1   \n",
       "57              0         0            0            0             1   \n",
       "58              0         0            0            0             1   \n",
       "59              0         0            0            0             0   \n",
       "60              0         0            0            0             6   \n",
       "...           ...       ...          ...          ...           ...   \n",
       "131984          0         0            0            0             6   \n",
       "131985          0         0            0            0             6   \n",
       "131986          0         0            0            0             6   \n",
       "131987          0         0            0            0             0   \n",
       "131988          0         0            0            0             1   \n",
       "\n",
       "        extra_runs  total_runs dismissal_kind  wickets  \\\n",
       "56               0           1                       1   \n",
       "57               0           1                       1   \n",
       "58               0           1                       1   \n",
       "59               0           0                       1   \n",
       "60               0           6                       1   \n",
       "...            ...         ...            ...      ...   \n",
       "131984           0           6                       4   \n",
       "131985           0           6                       4   \n",
       "131986           0           6                       4   \n",
       "131987           0           0                       4   \n",
       "131988           0           1                       4   \n",
       "\n",
       "                                               commentary  \n",
       "56      Joshi to Ponting, 1 run, some good turn, spins...  \n",
       "57      Joshi to McCullum, 1 run, too straight and fla...  \n",
       "58      Joshi to Ponting, 1 run, fired on leg again an...  \n",
       "59      Joshi to McCullum, no run, bit of flight there...  \n",
       "60      Joshi to McCullum, SIX runs, snuff is snuff, s...  \n",
       "...                                                   ...  \n",
       "131984  KH Pandya to Watson, SIX runs, hammered! Full,...  \n",
       "131985  KH Pandya to Watson, SIX runs, over Krunal's h...  \n",
       "131986  KH Pandya to Watson, SIX runs, into the second...  \n",
       "131987  KH Pandya to Watson, no run, full, fast, and w...  \n",
       "131988  KH Pandya to Watson, 1 run, overpitched just o...  \n",
       "\n",
       "[45384 rows x 20 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[df.bowler == \"spinner\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# sanity check - check for NaN values in `bowler` column\n",
    "\n",
    "df.bowler.isna().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature Engineering `batsman` and `non_striker` Type"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In cricket, the batting order is the sequence in which batters play through their team's innings, there always being two batters taking part at any one time. All eleven players in a team are required to bat if the innings is completed (i.e., if the innings does not close early due to a declaration or other factor).\n",
    "\n",
    "The batting order is colloquially subdivided into:\n",
    "\n",
    "1. Top order (batters one to three)\n",
    "2. Middle order (batters four to eight)\n",
    "3. Tail enders (batters nine to eleven) \n",
    "\n",
    "In the code cell below, we have created a custom list of all top order batsman and tail enders for each team. We have then modified the `batsman` and `non_striker` columns in the dataframe which will be used to identify the type of batsman, \"top_order\", \"middle_order\" or \"tail_ender\". This column, we belive, should lend to the predictive power of our models. For example, a top order batsman is more likely to score a boundary than a tail ender, purely based on batting skill. Moreover, if given the scenario that a top order batsman bats till the last 5 overs of the game with a tail ender at the non-striker's end, the top order batsman is more likely to retain strike to maximize the number of runs scored in the last 5 overs.\n",
    "\n",
    "We deliberately chose not to take the easier route of engineering these columns column based on the `wickets` column. Doing so would induce multicollinearity in the data and, more importantly, alter the `batsman` and `non_striker` columns in a way that would compromise the integrity of our data.\n",
    "\n",
    "**Note:** It is not uncommon in cricket that an opening batsman may play in the middle order or vice-versa, given various factors such as recent form of the batsman or captain's/coach's discretion. Therefore, we have categorized individual batters with regards to their ideal batting position. For example, Robin Uthappa started off as a middle order batsman in 2008 with Mumbai Indians, but was promoted to the top order when he joined Royal Challengers Bangalore and Kolkata Knight Riders in 2010 and 2014 respectively."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Creating a list of all top order batsmen and tail enders for each team\n",
    "\n",
    "rcb_top_order = [\n",
    "    \"TM Dilshan\",\n",
    "    \"KP Pietersen\",\n",
    "    \"CH Gayle\",\n",
    "    \"PA Patel\",\n",
    "    \"V Kohli\",\n",
    "    \"AB de Villiers\",\n",
    "    \"R Dravid\",\n",
    "    \"JH Kallis\",\n",
    "    \"MK Pandey\",\n",
    "    \"W Jaffer\",\n",
    "    \"S Chanderpaul\",\n",
    "    \"LRPL Taylor\",\n",
    "    \"SR Watson\",\n",
    "    \"Mandeep Singh\",\n",
    "    \"A Mukund\",\n",
    "    \"BB McCullum\",\n",
    "    \"CA Pujara\",\n",
    "    \"MA Agarwal\",\n",
    "    \"KL Rahul\",\n",
    "    \"Q de Kock\",\n",
    "    \"J Arunkumar\",\n",
    "    \"S Sriram\",\n",
    "]\n",
    "\n",
    "rcb_tail_enders = [\n",
    "    \"SB Joshi\",\n",
    "    \"KP Appanna\",\n",
    "    \"A Kumble\",\n",
    "    \"Abdur Razzak\",\n",
    "    \"DL Vettori\",\n",
    "    \"KB Arun Karthik\",\n",
    "    \"S Sriram\",\n",
    "    \"AA Kazi\",\n",
    "    \"R Ninan\",\n",
    "    \"J Syed Mohammad\",\n",
    "    \"M Muralitharan\",\n",
    "    \"M Kartik\",\n",
    "    \"P R Barman\",\n",
    "    \"YS Chahal\",\n",
    "    \"SB Jakati\",\n",
    "    \"T Shamsi\",\n",
    "    \"Parvez Rasool\",\n",
    "    \"P Kumar\",\n",
    "    \"Z Khan\",\n",
    "    \"R Vinay Kumar\",\n",
    "    \"DW Steyn\",\n",
    "    \"ND Doshi\",\n",
    "    \"DP Nannes\",\n",
    "    \"JJ van der Wath\",\n",
    "    \"S Aravind\",\n",
    "    \"S Badree\",\n",
    "    \"CK Langeveldt\",\n",
    "    \"RR Bhatkal\",\n",
    "    \"P Parameswaran\",\n",
    "    \"JD Unadkat\",\n",
    "    \"RP Singh\",\n",
    "    \"R Rampaul\",\n",
    "    \"MA Starc\",\n",
    "    \"M Ashwin\" \"AB Dinda\",\n",
    "    \"VR Aaron\",\n",
    "    \"AN Ahmed\",\n",
    "    \"AF Milne\",\n",
    "    \"KW Richardson\",\n",
    "    \"CJ Jordan\",\n",
    "    \"TS Mills\",\n",
    "    \"A Choudhary\",\n",
    "    \"B Stanlake\",\n",
    "    \"Avesh Khan\",\n",
    "    \"UT Yadav\",\n",
    "    \"K Khejroliya\",\n",
    "    \"Mohammed Siraj\",\n",
    "    \"TG Southee\",\n",
    "    \"N Saini\",\n",
    "]\n",
    "\n",
    "mi_top_order = [\n",
    "    \"N Rana\",\n",
    "    \"RG Sharma\",\n",
    "    \"ST Jayasuriya\",\n",
    "    \"SR Tendulkar\",\n",
    "    \"YV Takawale\",\n",
    "    \"AM Rahane\",\n",
    "    \"DR Smith\",\n",
    "    \"L Ronchi\",\n",
    "    \"AC Blizzard\",\n",
    "    \"RV Uthappa\",\n",
    "    \"AT Rayudu\",\n",
    "    \"DJ Jacobs\",\n",
    "    \"S Dhawan\",\n",
    "    \"SS Tiwary\",\n",
    "    \"JP Duminy\",\n",
    "    \"AP Tare\",\n",
    "    \"AJ Finch\",\n",
    "    \"E Lewis\",\n",
    "    \"Ishan Kishan\",\n",
    "    \"Q de Kock\",\n",
    "    \"AM Rahane\",\n",
    "    \"LMP Simmons\",\n",
    "    \"CM Gautam\",\n",
    "    \"C Madan\",\n",
    "    \"PA Patel\",\n",
    "    \"BR Dunk\",\n",
    "    \"JC Buttler\",\n",
    "    \"SM Pollock\",\n",
    "    \"AS Yadav\",\n",
    "    \"HH Gibbs\",\n",
    "    \"UBT Chand\",\n",
    "    \"GJ Maxwell\",\n",
    "    \"MJ Guptill\",\n",
    "    \"RE Levi\",\n",
    "]\n",
    "\n",
    "mi_tail_enders = [\n",
    "    \"A Dananjaya\",\n",
    "    \"AG Murtaza\",\n",
    "    \"Harbhajan Singh\",\n",
    "    \"J Yadav\",\n",
    "    \"M Markande\",\n",
    "    \"PP Ojha\",\n",
    "    \"RD Chahar\",\n",
    "    \"RW Price\",\n",
    "    \"VS Yeligati\",\n",
    "    \"Z Khan\",\n",
    "    \"SL Malinga\",\n",
    "    \"JJ Bumrah\",\n",
    "    \"MG Johnson\",\n",
    "    \"P Suyal\",\n",
    "    \"A Nehra\",\n",
    "    \"DS Kulkarni\",\n",
    "    \"MA Khote\",\n",
    "    \"CRD Fernando\",\n",
    "    \"DJ Thornely\",\n",
    "    \"RR Raje\",\n",
    "    \"DR Smith\",\n",
    "    \"RP Singh\",\n",
    "    \"CJ McKay\",\n",
    "    \"NLTC Perera\",\n",
    "    \"R Shukla\",\n",
    "    \"A Nel\",\n",
    "    \"Z Khan\",\n",
    "    \"R McLaren\",\n",
    "    \"AN Ahmed\",\n",
    "    \"MM Patel\",\n",
    "    \"YS Chahal\",\n",
    "    \"P Kumar\",\n",
    "    \"K Santokie\",\n",
    "    \"M de Lange\",\n",
    "    \"R Vinay Kumar\",\n",
    "    \"MJ McClenaghan\",\n",
    "    \"TG Southee\",\n",
    "    \"Mustafizur Rahman\",\n",
    "    \"PJ Sangwan\",\n",
    "    \"R Salam\",\n",
    "    \"J Behrendorff\",\n",
    "    \"A Joseph\",\n",
    "    \"BB Sran\",\n",
    "]\n",
    "\n",
    "kkr_top_order = [\n",
    "    \"SC Ganguly\",\n",
    "    \"BB McCullum\",\n",
    "    \"RT Ponting\",\n",
    "    \"WP Saha\",\n",
    "    \"Salman Butt\",\n",
    "    \"AB Agarkar\",\n",
    "    \"Mohammad Hafeez\",\n",
    "    \"A Chopra\",\n",
    "    \"BJ Hodge\",\n",
    "    \"MK Tiwary\",\n",
    "    \"CA Pujara\",\n",
    "    \"CH Gayle\",\n",
    "    \"OA Shah\",\n",
    "    \"Mandeep Singh\",\n",
    "    \"AD Mathews\",\n",
    "    \"DJ Hussey\",\n",
    "    \"MS Bisla\",\n",
    "    \"JH Kallis\",\n",
    "    \"YK Pathan\",\n",
    "    \"G Gambhir\",\n",
    "    \"BJ Haddin\",\n",
    "    \"SP Goswami\",\n",
    "    \"EJG Morgan\",\n",
    "    \"RV Uthappa\",\n",
    "    \"MK Pandey\",\n",
    "    \"CA Lynn\",\n",
    "    \"C Munro\",\n",
    "    \"SP Narine\",\n",
    "    \"N Rana\",\n",
    "    \"S Gill\",\n",
    "    \"N Naik\",\n",
    "    \"J Denly\",\n",
    "]\n",
    "\n",
    "kkr_tail_enders = [\n",
    "    \"BAW Mendis\",\n",
    "    \"GB Hogg\",\n",
    "    \"KC Cariappa\",\n",
    "    \"M Kartik\",\n",
    "    \"Kuldeep Yadav\",\n",
    "    \"MB Parmar\",\n",
    "    \"R Vinay Kumar\",\n",
    "    \"PP Chawla\",\n",
    "    \"SMSM Senanayake\",\n",
    "    \"AB Dinda\",\n",
    "    \"I Sharma\",\n",
    "    \"AB Agarkar\",\n",
    "    \"Umar Gul\",\n",
    "    \"Shoaib Akhtar\",\n",
    "    \"CK Langeveldt\",\n",
    "    \"SE Bond\",\n",
    "    \"JD Unadkat\",\n",
    "    \"L Balaji\",\n",
    "    \"B Lee\",\n",
    "    \"M de Lange\",\n",
    "    \"PJ Sangwan\",\n",
    "    \"M Morkel\",\n",
    "    \"Mohammed Shami\",\n",
    "    \"R McLaren\",\n",
    "    \"Shami Ahmed\",\n",
    "    \"AS Rajpoot\",\n",
    "    \"TA Boult\",\n",
    "    \"MG Johnson\",\n",
    "    \"T Curran\",\n",
    "    \"S Mavi\",\n",
    "    \"P Krishna\",\n",
    "    \"J Searles\",\n",
    "    \"L Ferguson\",\n",
    "    \"H Gurney\",\n",
    "    \"P Raj\",\n",
    "    \"S Warrier\",\n",
    "    \"UT Yadav\",\n",
    "]\n",
    "\n",
    "csk_top_order = [\n",
    "    \"F du Plessis\",\n",
    "    \"SK Raina\",\n",
    "    \"M Vijay\",\n",
    "    \"ML Hayden\",\n",
    "    \"PA Patel\",\n",
    "    \"MEK Hussey\",\n",
    "    \"BB McCullum\",\n",
    "    \"S Vidyut\",\n",
    "    \"AT Rayudu\",\n",
    "    \"SP Fleming\",\n",
    "    \"DR Smith\",\n",
    "    \"S Anirudha\",\n",
    "    \"WP Saha\",\n",
    "    \"SR Watson\",\n",
    "    \"GJ Bailey\",\n",
    "]\n",
    "\n",
    "csk_tail_enders = [\n",
    "    \"Imran Tahir\",\n",
    "    \"M Muralitharan\",\n",
    "    \"R Ashwin\",\n",
    "    \"S Badree\",\n",
    "    \"S Randiv\",\n",
    "    \"S Vidyut\",\n",
    "    \"SB Jakati\",\n",
    "    \"P Amarnath\",\n",
    "    \"Joginder Sharma\",\n",
    "    \"M Ntini\",\n",
    "    \"L Balaji\",\n",
    "    \"S Tyagi\",\n",
    "    \"NLTC Perera\",\n",
    "    \"C Ganapathy\",\n",
    "    \"DE Bollinger\",\n",
    "    \"TG Southee\",\n",
    "    \"VY Mahesh\",\n",
    "    \"BW Hilfenhaus\",\n",
    "    \"DP Nannes\",\n",
    "    \"AS Rajpoot\",\n",
    "    \"B Laughlin\",\n",
    "    \"MM Sharma\",\n",
    "    \"IC Pandey\",\n",
    "    \"V Shankar\",\n",
    "    \"A Nehra\",\n",
    "    \"RG More\",\n",
    "    \"DL Chahar\",\n",
    "    \"M Wood\",\n",
    "    \"L Ngidi\",\n",
    "    \"KM Asif\",\n",
    "    \"S Kuggeleijn\",\n",
    "]\n",
    "\n",
    "dc_top_order = [\n",
    "    \"AC Gilchrist\",\n",
    "    \"VVS Laxman\",\n",
    "    \"HH Gibbs\",\n",
    "    \"LPC Silva\",\n",
    "    \"SB Styris\",\n",
    "    \"MD Mishra\",\n",
    "    \"TL Suman\",\n",
    "    \"Anirudh Singh\",\n",
    "    \"S Dhawan\",\n",
    "    \"IR Jaggi\",\n",
    "    \"KC Sangakkara\",\n",
    "    \"S Sohal\",\n",
    "    \"MJ Lumb\",\n",
    "    \"DJ Harris\",\n",
    "    \"PA Patel\",\n",
    "    \"PA Reddy\",\n",
    "]\n",
    "\n",
    "dc_tail_enders = [\n",
    "    \"A Mishra\",\n",
    "    \"Ankit Sharma\",\n",
    "    \"PP Ojha\",\n",
    "    \"WPUJC Vaas\",\n",
    "    \"RP Singh\",\n",
    "    \"D Kalyankrishna\",\n",
    "    \"DP Vijaykumar\",\n",
    "    \"PM Sarvesh Kumar\",\n",
    "    \"Jaskaran Singh\",\n",
    "    \"KAJ Roach\",\n",
    "    \"Harmeet Singh\",\n",
    "    \"RJ Harris\",\n",
    "    \"DW Steyn\",\n",
    "    \"I Sharma\",\n",
    "    \"Anand Rajan\",\n",
    "    \"DJ Harris\",\n",
    "    \"TP Sudhindra\",\n",
    "    \"V Pratap Singh\",\n",
    "    \"J Theron\",\n",
    "]\n",
    "\n",
    "dd_top_order = [\n",
    "    \"G Gambhir\",\n",
    "    \"V Sehwag\",\n",
    "    \"S Dhawan\",\n",
    "    \"AB de Villiers\",\n",
    "    \"TM Dilshan\",\n",
    "    \"DA Warner\",\n",
    "    \"KD Karthik\",\n",
    "    \"UBT Chand\",\n",
    "    \"AJ Finch\",\n",
    "    \"NV Ojha\",\n",
    "    \"CA Ingram\",\n",
    "    \"KP Pietersen\",\n",
    "    \"DPMD Jayawardene\",\n",
    "    \"MC Juneja\",\n",
    "    \"Q de Kock\",\n",
    "    \"M Vijay\",\n",
    "    \"MA Agarwal\",\n",
    "    \"SS Iyer\",\n",
    "    \"JP Duminy\",\n",
    "    \"MK Tiwary\",\n",
    "    \"SV Samson\",\n",
    "    \"KK Nair\",\n",
    "    \"AP Tare\",\n",
    "    \"C Munro\",\n",
    "    \"JJ Roy\",\n",
    "    \"P Shaw\",\n",
    "    \"S Dhawan\",\n",
    "]\n",
    "\n",
    "dd_tail_enders = [\n",
    "    \"A Mishra\",\n",
    "    \"Imran Tahir\",\n",
    "    \"J Yadav\",\n",
    "    \"R Sharma\",\n",
    "    \"S Lamichhane\",\n",
    "    \"S Nadeem\",\n",
    "    \"S Sriram\",\n",
    "    \"Sunny Gupta\",\n",
    "    \"GD McGrath\",\n",
    "    \"B Geeves\",\n",
    "    \"Mohammad Asif\",\n",
    "    \"VY Mahesh\",\n",
    "    \"PJ Sangwan\",\n",
    "    \"DP Nannes\",\n",
    "    \"UT Yadav\",\n",
    "    \"Y Nagar\",\n",
    "    \"A Nehra\",\n",
    "    \"AB Dinda\",\n",
    "    \"M Morkel\",\n",
    "    \"S Nadeem\",\n",
    "    \"VR Aaron\",\n",
    "    \"AM Salvi\",\n",
    "    \"DAJ Bracewell\",\n",
    "    \"Sunny Gupta\",\n",
    "    \"S Kaul\",\n",
    "    \"Mohammed Shami\",\n",
    "    \"WD Parnell\",\n",
    "    \"JD Unadkat\",\n",
    "    \"R Sharma\",\n",
    "    \"JA Morkel\",\n",
    "    \"DJ Muthuswami\",\n",
    "    \"Z Khan\",\n",
    "    \"GS Sandhu\",\n",
    "    \"J Yadav\",\n",
    "    \"K Rabada\",\n",
    "    \"TA Boult\",\n",
    "    \"Avesh Khan\",\n",
    "    \"L Plunkett\",\n",
    "    \"J Dala\",\n",
    "]\n",
    "\n",
    "srh_top_order = [\n",
    "    \"PA Reddy\",\n",
    "    \"PA Patel\",\n",
    "    \"CL White\",\n",
    "    \"KC Sangakkara\",\n",
    "    \"Q de Kock\",\n",
    "    \"GH Vihari\",\n",
    "    \"S Dhawan\",\n",
    "    \"A Ashish Reddy\",\n",
    "    \"AJ Finch\",\n",
    "    \"KL Rahul\",\n",
    "    \"NV Ojha\",\n",
    "    \"DA Warner\",\n",
    "    \"KS Williamson\",\n",
    "    \"RS Bopara\",\n",
    "    \"MC Henriques\",\n",
    "    \"DJ Hooda\",\n",
    "    \"WP Saha\",\n",
    "    \"R Bhui\",\n",
    "    \"MK Pandey\",\n",
    "    \"A Hales\",\n",
    "    \"SP Goswami\",\n",
    "    \"J Bairstow\",\n",
    "    \"V Shankar\",\n",
    "    \"MJ Guptill\",\n",
    "]\n",
    "\n",
    "srh_tail_enders = [\n",
    "    \"DW Steyn\",\n",
    "    \"I Sharma\",\n",
    "    \"A Ashish Reddy\",\n",
    "    \"Anand Rajan\",\n",
    "    \"B Kumar\",\n",
    "    \"TA Boult\",\n",
    "    \"Sandeep Sharma\",\n",
    "    \"P Kumar\",\n",
    "    \"A Nehra\",\n",
    "    \"Mustafizur Rahman\",\n",
    "    \"BB Sran\",\n",
    "    \"S Kaul\",\n",
    "    \"Mohammed Siraj\",\n",
    "    \"B Stanlake\",\n",
    "    \"Basil Thampi\",\n",
    "    \"K Ahmed\",\n",
    "    \"S Sharma\",\n",
    "    \"A Mishra\",\n",
    "    \"Ankit Sharma\",\n",
    "    \"Parvez Rasool\",\n",
    "    \"S Nadeem\",\n",
    "]\n",
    "\n",
    "rr_top_order = [\n",
    "    \"T Kohli\",\n",
    "    \"YK Pathan\",\n",
    "    \"SR Watson\",\n",
    "    \"M Kaif\",\n",
    "    \"Kamran Akmal\",\n",
    "    \"GC Smith\",\n",
    "    \"M Rawat\",\n",
    "    \"SA Asnodkar\",\n",
    "    \"NK Patel\",\n",
    "    \"Younis Khan\",\n",
    "    \"NV Ojha\",\n",
    "    \"MJ Lumb\",\n",
    "    \"DR Martyn\",\n",
    "    \"AA Jhunjhunwala\",\n",
    "    \"FY Fazal\",\n",
    "    \"AG Paunikar\",\n",
    "    \"R Dravid\",\n",
    "    \"AM Rahane\",\n",
    "    \"SP Goswami\",\n",
    "    \"OA Shah\",\n",
    "    \"BJ Hodge\",\n",
    "    \"MDKJ Perera\",\n",
    "    \"DH Yagnik\",\n",
    "    \"SV Samson\",\n",
    "    \"KK Nair\",\n",
    "    \"Ankit Sharma\",\n",
    "    \"KK Cooper\",\n",
    "    \"UBT Chand\",\n",
    "    \"SPD Smith\",\n",
    "    \"DJ Hooda\",\n",
    "    \"D Short\",\n",
    "    \"H Klaasen\",\n",
    "    \"RA Tripathi\",\n",
    "    \"JC Buttler\",\n",
    "]\n",
    "\n",
    "rr_tail_enders = [\n",
    "    \"A Chandila\",\n",
    "    \"AA Chavan\",\n",
    "    \"AS Raut\",\n",
    "    \"GB Hogg\",\n",
    "    \"I Sodhi\",\n",
    "    \"PV Tambe\",\n",
    "    \"ND Doshi\",\n",
    "    \"S Badree\",\n",
    "    \"S Gopal\",\n",
    "    \"S Midhun\",\n",
    "    \"SK Warne\",\n",
    "    \"MM Patel\",\n",
    "    \"SK Trivedi\",\n",
    "    \"Pankaj Singh\",\n",
    "    \"SW Tait\",\n",
    "    \"Kamran Khan\",\n",
    "    \"A Uniyal\",\n",
    "    \"M Morkel\",\n",
    "    \"S Narwal\",\n",
    "    \"SB Wagh\",\n",
    "    \"AP Dole\",\n",
    "    \"FY Fazal\",\n",
    "    \"A Singh\",\n",
    "    \"S Sreesanth\",\n",
    "    \"R Shukla\",\n",
    "    \"VS Malik\",\n",
    "    \"KW Richardson\",\n",
    "    \"DS Kulkarni\",\n",
    "    \"R Bhatia\",\n",
    "    \"TG Southee\",\n",
    "    \"J Theron\",\n",
    "    \"B Brainder Sran\",\n",
    "    \"JD Unadkat\",\n",
    "    \"B Laughlin\",\n",
    "    \"J Archer\",\n",
    "    \"Anureet Singh\",\n",
    "    \"VR Aaron\",\n",
    "    \"O Thomas\",\n",
    "]\n",
    "\n",
    "punj_top_order = [\n",
    "    \"K Goel\",\n",
    "    \"JR Hopes\",\n",
    "    \"KC Sangakkara\",\n",
    "    \"DPMD Jayawardene\",\n",
    "    \"Yuvraj Singh\",\n",
    "    \"SM Katich\",\n",
    "    \"SE Marsh\",\n",
    "    \"LA Pomersbach\",\n",
    "    \"RS Bopara\",\n",
    "    \"MS Bisla\",\n",
    "    \"PC Valthaty\",\n",
    "    \"AC Gilchrist\",\n",
    "    \"Mandeep Singh\",\n",
    "    \"N Saini\",\n",
    "    \"M Vohra\",\n",
    "    \"Gurkeerat Singh\",\n",
    "    \"CA Pujara\",\n",
    "    \"V Sehwag\",\n",
    "    \"WP Saha\",\n",
    "    \"KK Nair\" \"M Vijay\",\n",
    "    \"HM Amla\",\n",
    "    \"EJG Morgan\",\n",
    "    \"MJ Guptill\",\n",
    "    \"KL Rahul\",\n",
    "    \"MA Agarwal\",\n",
    "    \"CH Gayle\",\n",
    "    \"AJ Finch\",\n",
    "]\n",
    "\n",
    "punj_tail_enders = [\n",
    "    \"BA Bhatt\",\n",
    "    \"H Brar\",\n",
    "    \"Karanveer Singh\",\n",
    "    \"KC Cariappa\",\n",
    "    \"M Ashwin\",\n",
    "    \"M Kartik\",\n",
    "    \"M Ur Rahman\",\n",
    "    \"RJ Harris\",\n",
    "    \"P Sahu\",\n",
    "    \"RR Powar\",\n",
    "    \"Shivam Sharma\",\n",
    "    \"Swapnil Singh\",\n",
    "    \"J Theron\",\n",
    "    \"VS Malik\",\n",
    "    \"L Ablish\",\n",
    "    \"P Kumar\",\n",
    "    \"V Chakravarthy\",\n",
    "    \"B Lee\",\n",
    "    \"S Sreesanth\",\n",
    "    \"VRV Singh\",\n",
    "    \"Gagandeep Singh\",\n",
    "    \"YA Abdulla\",\n",
    "    \"R McLaren\",\n",
    "    \"Harmeet Singh\",\n",
    "    \"P Awana\",\n",
    "    \"MS Gony\",\n",
    "    \"Sandeep Sharma (1)\",\n",
    "    \"MG Neser\",\n",
    "    \"Sandeep Sharma\",\n",
    "    \"L Balaji\",\n",
    "    \"MG Johnson\",\n",
    "    \"BE Hendricks\",\n",
    "    \"Anureet Singh\",\n",
    "    \"MM Sharma\",\n",
    "    \"KJ Abbott\",\n",
    "    \"VR Aaron\",\n",
    "    \"T Natarajan\",\n",
    "    \"I Sharma\",\n",
    "    \"MJ Henry\",\n",
    "    \"BB Sran\",\n",
    "    \"AS Rajpoot\",\n",
    "    \"Mohammed Shami\",\n",
    "    \"H Viljoen\",\n",
    "    \"A Singh\",\n",
    "    \"AJ Tye\",\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Replacing all batsman and non_striker names with \"top\"\n",
    "\n",
    "df.loc[\n",
    "    (df.batting_team == \"Royal Challengers Bangalore\")\n",
    "    & (df.batsman.isin(rcb_top_order)),\n",
    "    \"batsman\",\n",
    "] = \"top\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Delhi Capitals\") & (df.batsman.isin(dd_top_order)), \"batsman\"\n",
    "] = \"top\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Delhi Daredevils\") & (df.batsman.isin(dd_top_order)), \"batsman\"\n",
    "] = \"top\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Chennai Super Kings\") & (df.batsman.isin(csk_top_order)),\n",
    "    \"batsman\",\n",
    "] = \"top\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Kings XI Punjab\") & (df.batsman.isin(punj_top_order)),\n",
    "    \"batsman\",\n",
    "] = \"top\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Sunrisers Hyderabad\") & (df.batsman.isin(srh_top_order)),\n",
    "    \"batsman\",\n",
    "] = \"top\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Kolkata Knight Riders\") & (df.batsman.isin(kkr_top_order)),\n",
    "    \"batsman\",\n",
    "] = \"top\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Deccan Chargers\") & (df.batsman.isin(dc_top_order)), \"batsman\"\n",
    "] = \"top\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Rajasthan Royals\") & (df.batsman.isin(rr_top_order)), \"batsman\"\n",
    "] = \"top\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Mumbai Indians\") & (df.batsman.isin(mi_top_order)), \"batsman\"\n",
    "] = \"top\"\n",
    "\n",
    "\n",
    "df.loc[\n",
    "    (df.batting_team == \"Royal Challengers Bangalore\")\n",
    "    & (df.non_striker.isin(rcb_top_order)),\n",
    "    \"non_striker\",\n",
    "] = \"top\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Delhi Capitals\") & (df.non_striker.isin(dd_top_order)),\n",
    "    \"non_striker\",\n",
    "] = \"top\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Delhi Daredevils\") & (df.non_striker.isin(dd_top_order)),\n",
    "    \"non_striker\",\n",
    "] = \"top\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Chennai Super Kings\") & (df.non_striker.isin(csk_top_order)),\n",
    "    \"non_striker\",\n",
    "] = \"top\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Kings XI Punjab\") & (df.non_striker.isin(punj_top_order)),\n",
    "    \"non_striker\",\n",
    "] = \"top\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Sunrisers Hyderabad\") & (df.non_striker.isin(srh_top_order)),\n",
    "    \"non_striker\",\n",
    "] = \"top\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Kolkata Knight Riders\") & (df.non_striker.isin(kkr_top_order)),\n",
    "    \"non_striker\",\n",
    "] = \"top\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Deccan Chargers\") & (df.non_striker.isin(dc_top_order)),\n",
    "    \"non_striker\",\n",
    "] = \"top\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Rajasthan Royals\") & (df.non_striker.isin(rr_top_order)),\n",
    "    \"non_striker\",\n",
    "] = \"top\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Mumbai Indians\") & (df.non_striker.isin(mi_top_order)),\n",
    "    \"non_striker\",\n",
    "] = \"top\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Replacing all batsman and non_striker names with \"tail\"\n",
    "\n",
    "df.loc[\n",
    "    (df.batting_team == \"Royal Challengers Bangalore\")\n",
    "    & (df.batsman.isin(rcb_tail_enders)),\n",
    "    \"batsman\",\n",
    "] = \"tail\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Delhi Capitals\") & (df.batsman.isin(dd_tail_enders)), \"batsman\"\n",
    "] = \"tail\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Delhi Daredevils\") & (df.batsman.isin(dd_tail_enders)),\n",
    "    \"batsman\",\n",
    "] = \"tail\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Chennai Super Kings\") & (df.batsman.isin(csk_tail_enders)),\n",
    "    \"batsman\",\n",
    "] = \"tail\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Kings XI Punjab\") & (df.batsman.isin(punj_tail_enders)),\n",
    "    \"batsman\",\n",
    "] = \"tail\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Sunrisers Hyderabad\") & (df.batsman.isin(srh_tail_enders)),\n",
    "    \"batsman\",\n",
    "] = \"tail\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Kolkata Knight Riders\") & (df.batsman.isin(kkr_tail_enders)),\n",
    "    \"batsman\",\n",
    "] = \"tail\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Deccan Chargers\") & (df.batsman.isin(dc_tail_enders)),\n",
    "    \"batsman\",\n",
    "] = \"tail\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Rajasthan Royals\") & (df.batsman.isin(rr_tail_enders)),\n",
    "    \"batsman\",\n",
    "] = \"tail\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Mumbai Indians\") & (df.batsman.isin(mi_tail_enders)), \"batsman\"\n",
    "] = \"tail\"\n",
    "\n",
    "df.loc[\n",
    "    (df.batting_team == \"Royal Challengers Bangalore\")\n",
    "    & (df.non_striker.isin(rcb_tail_enders)),\n",
    "    \"non_striker\",\n",
    "] = \"tail\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Delhi Capitals\") & (df.non_striker.isin(dd_tail_enders)),\n",
    "    \"non_striker\",\n",
    "] = \"tail\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Delhi Daredevils\") & (df.non_striker.isin(dd_tail_enders)),\n",
    "    \"non_striker\",\n",
    "] = \"tail\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Chennai Super Kings\") & (df.non_striker.isin(csk_tail_enders)),\n",
    "    \"non_striker\",\n",
    "] = \"tail\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Kings XI Punjab\") & (df.non_striker.isin(punj_tail_enders)),\n",
    "    \"non_striker\",\n",
    "] = \"tail\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Sunrisers Hyderabad\") & (df.non_striker.isin(srh_tail_enders)),\n",
    "    \"non_striker\",\n",
    "] = \"tail\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Kolkata Knight Riders\")\n",
    "    & (df.non_striker.isin(kkr_tail_enders)),\n",
    "    \"non_striker\",\n",
    "] = \"tail\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Deccan Chargers\") & (df.non_striker.isin(dc_tail_enders)),\n",
    "    \"non_striker\",\n",
    "] = \"tail\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Rajasthan Royals\") & (df.non_striker.isin(rr_tail_enders)),\n",
    "    \"non_striker\",\n",
    "] = \"tail\"\n",
    "df.loc[\n",
    "    (df.batting_team == \"Mumbai Indians\") & (df.non_striker.isin(mi_tail_enders)),\n",
    "    \"non_striker\",\n",
    "] = \"tail\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "# replace remaining names with \"middle\"\n",
    "\n",
    "l = [\"top\", \"tail\"]\n",
    "df.loc[~df.batsman.isin(l), \"batsman\"] = \"middle\"\n",
    "df.loc[~df.non_striker.isin(l), \"non_striker\"] = \"middle\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0\n",
      "0\n"
     ]
    }
   ],
   "source": [
    "# sanity check for nulls\n",
    "\n",
    "print(df.batsman.isna().sum())\n",
    "print(df.non_striker.isna().sum())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(89893, 20)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[df.batsman == \"top\"].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(36128, 20)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[df.batsman == \"middle\"].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5980, 20)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[df.batsman == \"tail\"].shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Renaming Old Team Names To Current Names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Empty DataFrame\n",
      "Columns: [season, match_id, venue, inning, batting_team, bowling_team, batsman, non_striker, bowler, ball, wide_runs, bye_runs, legbye_runs, noball_runs, batsman_runs, extra_runs, total_runs, dismissal_kind, wickets, commentary]\n",
      "Index: []\n",
      "Empty DataFrame\n",
      "Columns: [season, match_id, venue, inning, batting_team, bowling_team, batsman, non_striker, bowler, ball, wide_runs, bye_runs, legbye_runs, noball_runs, batsman_runs, extra_runs, total_runs, dismissal_kind, wickets, commentary]\n",
      "Index: []\n"
     ]
    }
   ],
   "source": [
    "# Renaming the teams with their latest names\n",
    "\n",
    "df.loc[(df.batting_team == \"Delhi Daredevils\"), \"batting_team\"] = \"Delhi Capitals\"\n",
    "df.loc[(df.bowling_team == \"Delhi Daredevils\"), \"bowling_team\"] = \"Delhi Capitals\"\n",
    "\n",
    "df.loc[(df.batting_team == \"Deccan Chargers\"), \"batting_team\"] = \"Sunrisers Hyderabad\"\n",
    "df.loc[(df.bowling_team == \"Deccan Chargers\"), \"bowling_team\"] = \"Sunrisers Hyderabad\"\n",
    "\n",
    "# sanity check\n",
    "print(\n",
    "    df.loc[\n",
    "        (df.batting_team == \"Delhi Daredevils\")\n",
    "        | (df.bowling_team == \"Delhi Daredevils\")\n",
    "    ]\n",
    ")\n",
    "print(\n",
    "    df.loc[\n",
    "        (df.batting_team == \"Deccan Chargers\") | (df.bowling_team == \"Deccan Chargers\")\n",
    "    ]\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature Engineering `ball_length` Column from `commentary` Column"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "def lower_comm(x):\n",
    "    return x.lower()\n",
    "\n",
    "\n",
    "df[\"commentary\"] = df[\"commentary\"].apply(lower_comm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "full_toss_length = [\"full toss\", \"beamer\", \"full-toss\", \"fulltoss\"]\n",
    "yorker_length = [\n",
    "    \"yorker\",\n",
    "    \"yorker length\",\n",
    "    \"yorker-ish\",\n",
    "    \"blockhole\",\n",
    "    \"block-hole\",\n",
    "    \"block hole\",\n",
    "    \"toe-crusher\",\n",
    "    \"toe crusher\",\n",
    "]\n",
    "full_length = [\n",
    "    \"full\",\n",
    "    \"full and straight\",\n",
    "    \"full length\",\n",
    "    \"fullish\",\n",
    "    \"full-ish\",\n",
    "    \"fuller\",\n",
    "    \"very full\",\n",
    "    \"full ball\",\n",
    "    \"full delivery\",\n",
    "    \"forward\",\n",
    "    \"front foot\",\n",
    "    \"flicked off the pads\",\n",
    "    \"glanced off the pads\",\n",
    "    \"flicks off the pad\",\n",
    "    \"off the pads\",\n",
    "    \"off the pad\",\n",
    "    \"clips\",\n",
    "    \"clipped\",\n",
    "    \"pads\",\n",
    "    \"on the pad\",\n",
    "    \"into the pads\",\n",
    "    \"cover drive\",\n",
    "    \"front\",\n",
    "    \"beaten\",\n",
    "    \"beaten on the drive\",\n",
    "    \"pitches it up\",\n",
    "    \"full on\",\n",
    "    \"slog\",\n",
    "    \"slogs\",\n",
    "    \"flat and quick\",\n",
    "    \"onto the pads\",\n",
    "    \"aimed at leg stump\",\n",
    "    \"aimed at leg-stump\",\n",
    "    \"pitched at leg stump\",\n",
    "    \"floated up\",\n",
    "    \"floats\",\n",
    "    \"floating\",\n",
    "    \"flights\",\n",
    "    \"flight\",\n",
    "    \"inside out\",\n",
    "    \"in the slot\",\n",
    "    \"slot delivery\",\n",
    "    \"long on\",\n",
    "    \"long-on\",\n",
    "    \"flighted\",\n",
    "    \"flighting\",\n",
    "    \"tossed up\",\n",
    "    \"overpitched\",\n",
    "    \"overpitching\",\n",
    "    \"overpitches\",\n",
    "    \"pitched up\",\n",
    "    \"half-volley\",\n",
    "    \"half volley\",\n",
    "    \"loopy\",\n",
    "    \"looping\",\n",
    "    \"reverse sweep\",\n",
    "    \"down the ground\",\n",
    "    \"reverse-sweep\",\n",
    "    \"sweep\",\n",
    "    \"sweeping\",\n",
    "    \"paddled\",\n",
    "    \"paddle\",\n",
    "    \"paddles\",\n",
    "    \"slog swept\",\n",
    "    \"slog-sweep\",\n",
    "    \"off-drive\",\n",
    "    \"straight-drive\",\n",
    "    \"worked\",\n",
    "    \"works\",\n",
    "    \"meeting the pitch\",\n",
    "    \"lunges forward\",\n",
    "    \"cover-drive\",\n",
    "    \"cover drive\",\n",
    "    \"off drive\",\n",
    "    \"slog-sweeps\",\n",
    "    \"straight drive\",\n",
    "    \"on-drive\",\n",
    "    \"on drive\",\n",
    "    \"helicopter shot\",\n",
    "    \"helicopter\",\n",
    "    \"over cover\",\n",
    "    \"over extra cover\",\n",
    "    \"pitch of the ball\",\n",
    "    \"to the pitch\",\n",
    "    \"slower full\",\n",
    "    \"slower, full\",\n",
    "]\n",
    "good_length = [\n",
    "    \"length ball\",\n",
    "    \"good length\",\n",
    "    \"length delivery\",\n",
    "    \"at a length\",\n",
    "    \", length,\",\n",
    "    \"length\",\n",
    "    \"peach\",\n",
    "    \"corridor of uncertainty\",\n",
    "    \"punched\",\n",
    "    \"googly\",\n",
    "    \"cramped\",\n",
    "    \"cramps\",\n",
    "    \"cramping\",\n",
    "    \"thigh pad\",\n",
    "    \"cramp\",\n",
    "    \"goodish\",\n",
    "    \"good-ish\",\n",
    "    \"nudges\",\n",
    "    \"driven on the up\",\n",
    "    \"slower length ball\",\n",
    "    \"slower, good length\",\n",
    "    \"on the up\",\n",
    "    \"length outside\",\n",
    "    \"length on\",\n",
    "    \"tight length\",\n",
    "]\n",
    "short_length = [\n",
    "    \"short\",\n",
    "    \"short ball\",\n",
    "    \"short delivery\",\n",
    "    \"bouncer\",\n",
    "    \"pulled\",\n",
    "    \"hooked\",\n",
    "    \"pulling\",\n",
    "    \"hooking\",\n",
    "    \"pull\",\n",
    "    \"hook\",\n",
    "    \"back of a length\",\n",
    "    \"back of length\",\n",
    "    \"back of good length\",\n",
    "    \"back foot\",\n",
    "    \"back-foot\",\n",
    "    \"heaved\",\n",
    "    \"hoicked\",\n",
    "    \"short of good length\",\n",
    "    \"short length\",\n",
    "    \"shortish\",\n",
    "    \"short-ish\",\n",
    "    \"cuts\",\n",
    "    \"cut\",\n",
    "    \"cutting it\",\n",
    "    \"cutting the\",\n",
    "    \"dragged back\",\n",
    "    \"drags his length back\",\n",
    "    \"back-of-a-length\",\n",
    "    \"holding back the length\",\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "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>season</th>\n",
       "      <th>match_id</th>\n",
       "      <th>venue</th>\n",
       "      <th>inning</th>\n",
       "      <th>batting_team</th>\n",
       "      <th>bowling_team</th>\n",
       "      <th>batsman</th>\n",
       "      <th>non_striker</th>\n",
       "      <th>bowler</th>\n",
       "      <th>ball</th>\n",
       "      <th>...</th>\n",
       "      <th>bye_runs</th>\n",
       "      <th>legbye_runs</th>\n",
       "      <th>noball_runs</th>\n",
       "      <th>batsman_runs</th>\n",
       "      <th>extra_runs</th>\n",
       "      <th>total_runs</th>\n",
       "      <th>dismissal_kind</th>\n",
       "      <th>wickets</th>\n",
       "      <th>commentary</th>\n",
       "      <th>ball_length</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2008</td>\n",
       "      <td>1</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>kumar to ganguly, 1 leg bye, and thats runs im...</td>\n",
       "      <td>full</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2008</td>\n",
       "      <td>1</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>kumar to mccullum, no run, this one nips back ...</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2008</td>\n",
       "      <td>1</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.3</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>kumar to mccullum, 1 wide, thats wide, as prav...</td>\n",
       "      <td>nan</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2008</td>\n",
       "      <td>1</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.3</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>kumar to mccullum, no run, too straight, mccul...</td>\n",
       "      <td>nan</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2008</td>\n",
       "      <td>1</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>kumar to mccullum, no run, steps down the trac...</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   season match_id      venue  inning           batting_team  \\\n",
       "0    2008        1  Bangalore       1  Kolkata Knight Riders   \n",
       "1    2008        1  Bangalore       1  Kolkata Knight Riders   \n",
       "2    2008        1  Bangalore       1  Kolkata Knight Riders   \n",
       "3    2008        1  Bangalore       1  Kolkata Knight Riders   \n",
       "4    2008        1  Bangalore       1  Kolkata Knight Riders   \n",
       "\n",
       "                  bowling_team batsman non_striker bowler  ball  ...  \\\n",
       "0  Royal Challengers Bangalore     top         top  pacer   0.1  ...   \n",
       "1  Royal Challengers Bangalore     top         top  pacer   0.2  ...   \n",
       "2  Royal Challengers Bangalore     top         top  pacer   0.3  ...   \n",
       "3  Royal Challengers Bangalore     top         top  pacer   0.3  ...   \n",
       "4  Royal Challengers Bangalore     top         top  pacer   0.4  ...   \n",
       "\n",
       "   bye_runs  legbye_runs  noball_runs  batsman_runs  extra_runs  total_runs  \\\n",
       "0         0            1            0             0           1           1   \n",
       "1         0            0            0             0           0           0   \n",
       "2         0            0            0             0           1           1   \n",
       "3         0            0            0             0           0           0   \n",
       "4         0            0            0             0           0           0   \n",
       "\n",
       "   dismissal_kind wickets                                         commentary  \\\n",
       "0                       0  kumar to ganguly, 1 leg bye, and thats runs im...   \n",
       "1                       0  kumar to mccullum, no run, this one nips back ...   \n",
       "2                       0  kumar to mccullum, 1 wide, thats wide, as prav...   \n",
       "3                       0  kumar to mccullum, no run, too straight, mccul...   \n",
       "4                       0  kumar to mccullum, no run, steps down the trac...   \n",
       "\n",
       "  ball_length  \n",
       "0        full  \n",
       "1        good  \n",
       "2         nan  \n",
       "3         nan  \n",
       "4        good  \n",
       "\n",
       "[5 rows x 21 columns]"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "conditions = [\n",
    "    df.commentary.str.contains(\"|\".join(full_toss_length)),\n",
    "    df.commentary.str.contains(\"|\".join(yorker_length)),\n",
    "    df.commentary.str.contains(\"|\".join(full_length)),\n",
    "    df.commentary.str.contains(\"|\".join(good_length)),\n",
    "    df.commentary.str.contains(\"|\".join(short_length)),\n",
    "]\n",
    "choices = [\"full toss\", \"yorker\", \"full\", \"good\", \"short\"]\n",
    "\n",
    "df[\"ball_length\"] = np.select(conditions, choices, default=np.nan)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.loc[df.ball_length == \"nan\"].shape[0]\n",
    "# df.loc[df.ball_length=='nan']['commentary'].to_csv('commentary_nan.csv')\n",
    "\n",
    "# replace all NaN values in `ball_length` column with 'Unknown'\n",
    "\n",
    "df.loc[df.ball_length == \"nan\", \"ball_length\"] = \"Unknown\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0, 21)"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[(df.ball_length == \"nan\")].shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature Engineering `cumulative_runs` from `total_runs` per ball "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "# sum total_runs until `inning` doesn't change\n",
    "\n",
    "df[\"cumulative_runs\"] = df.groupby(\n",
    "    [\"season\", \"match_id\", \"venue\", \"inning\", \"batting_team\", \"bowling_team\"]\n",
    ")[\"total_runs\"].cumsum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0           1\n",
       "1           1\n",
       "2           2\n",
       "3           2\n",
       "4           2\n",
       "         ... \n",
       "131996    143\n",
       "131997    145\n",
       "131998    146\n",
       "131999    148\n",
       "132000    148\n",
       "Name: cumulative_runs, Length: 132001, dtype: int64"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[\"cumulative_runs\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Cleaning `match_id` Column  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "playoff = [\n",
    "    \"3rd Place Playoff\",\n",
    "    \"eliminator\",\n",
    "    \"final\",\n",
    "    \"semi-final1\",\n",
    "    \"semi-final2\",\n",
    "    \"semi-final\",\n",
    "]\n",
    "df.loc[df.match_id.isin(playoff), \"match_id\"] = \"playoffs\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "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>season</th>\n",
       "      <th>match_id</th>\n",
       "      <th>venue</th>\n",
       "      <th>inning</th>\n",
       "      <th>batting_team</th>\n",
       "      <th>bowling_team</th>\n",
       "      <th>batsman</th>\n",
       "      <th>non_striker</th>\n",
       "      <th>bowler</th>\n",
       "      <th>ball</th>\n",
       "      <th>...</th>\n",
       "      <th>legbye_runs</th>\n",
       "      <th>noball_runs</th>\n",
       "      <th>batsman_runs</th>\n",
       "      <th>extra_runs</th>\n",
       "      <th>total_runs</th>\n",
       "      <th>dismissal_kind</th>\n",
       "      <th>wickets</th>\n",
       "      <th>commentary</th>\n",
       "      <th>ball_length</th>\n",
       "      <th>cumulative_runs</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>12801</th>\n",
       "      <td>2008</td>\n",
       "      <td>playoffs</td>\n",
       "      <td>Mumbai</td>\n",
       "      <td>1</td>\n",
       "      <td>Rajasthan Royals</td>\n",
       "      <td>Delhi Capitals</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>mcgrath to smith, no run, a peach from mcgrath...</td>\n",
       "      <td>full</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12802</th>\n",
       "      <td>2008</td>\n",
       "      <td>playoffs</td>\n",
       "      <td>Mumbai</td>\n",
       "      <td>1</td>\n",
       "      <td>Rajasthan Royals</td>\n",
       "      <td>Delhi Capitals</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>mcgrath to smith, 1 wide, and this one is wide...</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12803</th>\n",
       "      <td>2008</td>\n",
       "      <td>playoffs</td>\n",
       "      <td>Mumbai</td>\n",
       "      <td>1</td>\n",
       "      <td>Rajasthan Royals</td>\n",
       "      <td>Delhi Capitals</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>mcgrath to smith, no run, tight line and lengt...</td>\n",
       "      <td>full</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12804</th>\n",
       "      <td>2008</td>\n",
       "      <td>playoffs</td>\n",
       "      <td>Mumbai</td>\n",
       "      <td>1</td>\n",
       "      <td>Rajasthan Royals</td>\n",
       "      <td>Delhi Capitals</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.3</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>mcgrath to smith, no run, good length once aga...</td>\n",
       "      <td>full</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12805</th>\n",
       "      <td>2008</td>\n",
       "      <td>playoffs</td>\n",
       "      <td>Mumbai</td>\n",
       "      <td>1</td>\n",
       "      <td>Rajasthan Royals</td>\n",
       "      <td>Delhi Capitals</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>mcgrath to smith, no run, smith tries to flick...</td>\n",
       "      <td>full</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131996</th>\n",
       "      <td>2019</td>\n",
       "      <td>playoffs</td>\n",
       "      <td>Hyderabad</td>\n",
       "      <td>2</td>\n",
       "      <td>Chennai Super Kings</td>\n",
       "      <td>Mumbai Indians</td>\n",
       "      <td>middle</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>19.2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td></td>\n",
       "      <td>5</td>\n",
       "      <td>malinga to jadeja, 1 run, low full toss on mid...</td>\n",
       "      <td>full toss</td>\n",
       "      <td>143</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131997</th>\n",
       "      <td>2019</td>\n",
       "      <td>playoffs</td>\n",
       "      <td>Hyderabad</td>\n",
       "      <td>2</td>\n",
       "      <td>Chennai Super Kings</td>\n",
       "      <td>Mumbai Indians</td>\n",
       "      <td>top</td>\n",
       "      <td>middle</td>\n",
       "      <td>pacer</td>\n",
       "      <td>19.3</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td></td>\n",
       "      <td>5</td>\n",
       "      <td>malinga to watson, 2 runs, yorker, just outsid...</td>\n",
       "      <td>yorker</td>\n",
       "      <td>145</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131998</th>\n",
       "      <td>2019</td>\n",
       "      <td>playoffs</td>\n",
       "      <td>Hyderabad</td>\n",
       "      <td>2</td>\n",
       "      <td>Chennai Super Kings</td>\n",
       "      <td>Mumbai Indians</td>\n",
       "      <td>top</td>\n",
       "      <td>middle</td>\n",
       "      <td>pacer</td>\n",
       "      <td>19.4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>run out</td>\n",
       "      <td>6</td>\n",
       "      <td>malinga to watson, 1 run, out, yorker, just ou...</td>\n",
       "      <td>yorker</td>\n",
       "      <td>146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131999</th>\n",
       "      <td>2019</td>\n",
       "      <td>playoffs</td>\n",
       "      <td>Hyderabad</td>\n",
       "      <td>2</td>\n",
       "      <td>Chennai Super Kings</td>\n",
       "      <td>Mumbai Indians</td>\n",
       "      <td>middle</td>\n",
       "      <td>middle</td>\n",
       "      <td>pacer</td>\n",
       "      <td>19.5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td></td>\n",
       "      <td>6</td>\n",
       "      <td>malinga to thakur, 2 runs, full toss on leg, a...</td>\n",
       "      <td>full toss</td>\n",
       "      <td>148</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>132000</th>\n",
       "      <td>2019</td>\n",
       "      <td>playoffs</td>\n",
       "      <td>Hyderabad</td>\n",
       "      <td>2</td>\n",
       "      <td>Chennai Super Kings</td>\n",
       "      <td>Mumbai Indians</td>\n",
       "      <td>middle</td>\n",
       "      <td>middle</td>\n",
       "      <td>pacer</td>\n",
       "      <td>19.6</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>lbw</td>\n",
       "      <td>7</td>\n",
       "      <td>malinga to thakur, out, mumbai indians win ipl...</td>\n",
       "      <td>yorker</td>\n",
       "      <td>148</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>9186 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        season  match_id      venue  inning         batting_team  \\\n",
       "12801     2008  playoffs     Mumbai       1     Rajasthan Royals   \n",
       "12802     2008  playoffs     Mumbai       1     Rajasthan Royals   \n",
       "12803     2008  playoffs     Mumbai       1     Rajasthan Royals   \n",
       "12804     2008  playoffs     Mumbai       1     Rajasthan Royals   \n",
       "12805     2008  playoffs     Mumbai       1     Rajasthan Royals   \n",
       "...        ...       ...        ...     ...                  ...   \n",
       "131996    2019  playoffs  Hyderabad       2  Chennai Super Kings   \n",
       "131997    2019  playoffs  Hyderabad       2  Chennai Super Kings   \n",
       "131998    2019  playoffs  Hyderabad       2  Chennai Super Kings   \n",
       "131999    2019  playoffs  Hyderabad       2  Chennai Super Kings   \n",
       "132000    2019  playoffs  Hyderabad       2  Chennai Super Kings   \n",
       "\n",
       "          bowling_team batsman non_striker bowler  ball  ...  legbye_runs  \\\n",
       "12801   Delhi Capitals     top         top  pacer   0.1  ...            0   \n",
       "12802   Delhi Capitals     top         top  pacer   0.2  ...            0   \n",
       "12803   Delhi Capitals     top         top  pacer   0.2  ...            0   \n",
       "12804   Delhi Capitals     top         top  pacer   0.3  ...            0   \n",
       "12805   Delhi Capitals     top         top  pacer   0.4  ...            0   \n",
       "...                ...     ...         ...    ...   ...  ...          ...   \n",
       "131996  Mumbai Indians  middle         top  pacer  19.2  ...            0   \n",
       "131997  Mumbai Indians     top      middle  pacer  19.3  ...            0   \n",
       "131998  Mumbai Indians     top      middle  pacer  19.4  ...            0   \n",
       "131999  Mumbai Indians  middle      middle  pacer  19.5  ...            0   \n",
       "132000  Mumbai Indians  middle      middle  pacer  19.6  ...            0   \n",
       "\n",
       "        noball_runs  batsman_runs  extra_runs  total_runs  dismissal_kind  \\\n",
       "12801             0             0           0           0                   \n",
       "12802             0             0           1           1                   \n",
       "12803             0             0           0           0                   \n",
       "12804             0             0           0           0                   \n",
       "12805             0             0           0           0                   \n",
       "...             ...           ...         ...         ...             ...   \n",
       "131996            0             1           0           1                   \n",
       "131997            0             2           0           2                   \n",
       "131998            0             1           0           1         run out   \n",
       "131999            0             2           0           2                   \n",
       "132000            0             0           0           0             lbw   \n",
       "\n",
       "        wickets                                         commentary  \\\n",
       "12801         0  mcgrath to smith, no run, a peach from mcgrath...   \n",
       "12802         0  mcgrath to smith, 1 wide, and this one is wide...   \n",
       "12803         0  mcgrath to smith, no run, tight line and lengt...   \n",
       "12804         0  mcgrath to smith, no run, good length once aga...   \n",
       "12805         0  mcgrath to smith, no run, smith tries to flick...   \n",
       "...         ...                                                ...   \n",
       "131996        5  malinga to jadeja, 1 run, low full toss on mid...   \n",
       "131997        5  malinga to watson, 2 runs, yorker, just outsid...   \n",
       "131998        6  malinga to watson, 1 run, out, yorker, just ou...   \n",
       "131999        6  malinga to thakur, 2 runs, full toss on leg, a...   \n",
       "132000        7  malinga to thakur, out, mumbai indians win ipl...   \n",
       "\n",
       "        ball_length cumulative_runs  \n",
       "12801          full               0  \n",
       "12802       Unknown               1  \n",
       "12803          full               1  \n",
       "12804          full               1  \n",
       "12805          full               1  \n",
       "...             ...             ...  \n",
       "131996    full toss             143  \n",
       "131997       yorker             145  \n",
       "131998       yorker             146  \n",
       "131999    full toss             148  \n",
       "132000       yorker             148  \n",
       "\n",
       "[9186 rows x 22 columns]"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[df.match_id == \"playoffs\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "p = [\"playoffs\"]\n",
    "df.loc[~df.match_id.isin(p), \"match_id\"] = \"round-robin\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "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>season</th>\n",
       "      <th>match_id</th>\n",
       "      <th>venue</th>\n",
       "      <th>inning</th>\n",
       "      <th>batting_team</th>\n",
       "      <th>bowling_team</th>\n",
       "      <th>batsman</th>\n",
       "      <th>non_striker</th>\n",
       "      <th>bowler</th>\n",
       "      <th>ball</th>\n",
       "      <th>...</th>\n",
       "      <th>legbye_runs</th>\n",
       "      <th>noball_runs</th>\n",
       "      <th>batsman_runs</th>\n",
       "      <th>extra_runs</th>\n",
       "      <th>total_runs</th>\n",
       "      <th>dismissal_kind</th>\n",
       "      <th>wickets</th>\n",
       "      <th>commentary</th>\n",
       "      <th>ball_length</th>\n",
       "      <th>cumulative_runs</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2008</td>\n",
       "      <td>round-robin</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.1</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>kumar to ganguly, 1 leg bye, and thats runs im...</td>\n",
       "      <td>full</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2008</td>\n",
       "      <td>round-robin</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>kumar to mccullum, no run, this one nips back ...</td>\n",
       "      <td>good</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2008</td>\n",
       "      <td>round-robin</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.3</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>kumar to mccullum, 1 wide, thats wide, as prav...</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2008</td>\n",
       "      <td>round-robin</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.3</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>kumar to mccullum, no run, too straight, mccul...</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2008</td>\n",
       "      <td>round-robin</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>kumar to mccullum, no run, steps down the trac...</td>\n",
       "      <td>good</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131021</th>\n",
       "      <td>2019</td>\n",
       "      <td>round-robin</td>\n",
       "      <td>Mumbai</td>\n",
       "      <td>2</td>\n",
       "      <td>Mumbai Indians</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>15.3</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>1</td>\n",
       "      <td>prasidh krishna to sharma, no run,ê</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>122</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131022</th>\n",
       "      <td>2019</td>\n",
       "      <td>round-robin</td>\n",
       "      <td>Mumbai</td>\n",
       "      <td>2</td>\n",
       "      <td>Mumbai Indians</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>15.4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td></td>\n",
       "      <td>1</td>\n",
       "      <td>prasidh krishna to sharma, four runs, chipped ...</td>\n",
       "      <td>full</td>\n",
       "      <td>126</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131023</th>\n",
       "      <td>2019</td>\n",
       "      <td>round-robin</td>\n",
       "      <td>Mumbai</td>\n",
       "      <td>2</td>\n",
       "      <td>Mumbai Indians</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>15.5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td></td>\n",
       "      <td>1</td>\n",
       "      <td>prasidh krishna to sharma, 1 run, very full, j...</td>\n",
       "      <td>full</td>\n",
       "      <td>127</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131024</th>\n",
       "      <td>2019</td>\n",
       "      <td>round-robin</td>\n",
       "      <td>Mumbai</td>\n",
       "      <td>2</td>\n",
       "      <td>Mumbai Indians</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>15.6</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td></td>\n",
       "      <td>1</td>\n",
       "      <td>prasidh krishna to yadav, 1 run, prasidh digs ...</td>\n",
       "      <td>short</td>\n",
       "      <td>128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131025</th>\n",
       "      <td>2019</td>\n",
       "      <td>round-robin</td>\n",
       "      <td>Mumbai</td>\n",
       "      <td>2</td>\n",
       "      <td>Mumbai Indians</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>16.1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td></td>\n",
       "      <td>1</td>\n",
       "      <td>russell to yadav, six runs, and the match ends...</td>\n",
       "      <td>short</td>\n",
       "      <td>134</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>122815 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        season     match_id      venue  inning           batting_team  \\\n",
       "0         2008  round-robin  Bangalore       1  Kolkata Knight Riders   \n",
       "1         2008  round-robin  Bangalore       1  Kolkata Knight Riders   \n",
       "2         2008  round-robin  Bangalore       1  Kolkata Knight Riders   \n",
       "3         2008  round-robin  Bangalore       1  Kolkata Knight Riders   \n",
       "4         2008  round-robin  Bangalore       1  Kolkata Knight Riders   \n",
       "...        ...          ...        ...     ...                    ...   \n",
       "131021    2019  round-robin     Mumbai       2         Mumbai Indians   \n",
       "131022    2019  round-robin     Mumbai       2         Mumbai Indians   \n",
       "131023    2019  round-robin     Mumbai       2         Mumbai Indians   \n",
       "131024    2019  round-robin     Mumbai       2         Mumbai Indians   \n",
       "131025    2019  round-robin     Mumbai       2         Mumbai Indians   \n",
       "\n",
       "                       bowling_team batsman non_striker bowler  ball  ...  \\\n",
       "0       Royal Challengers Bangalore     top         top  pacer   0.1  ...   \n",
       "1       Royal Challengers Bangalore     top         top  pacer   0.2  ...   \n",
       "2       Royal Challengers Bangalore     top         top  pacer   0.3  ...   \n",
       "3       Royal Challengers Bangalore     top         top  pacer   0.3  ...   \n",
       "4       Royal Challengers Bangalore     top         top  pacer   0.4  ...   \n",
       "...                             ...     ...         ...    ...   ...  ...   \n",
       "131021        Kolkata Knight Riders     top         top  pacer  15.3  ...   \n",
       "131022        Kolkata Knight Riders     top         top  pacer  15.4  ...   \n",
       "131023        Kolkata Knight Riders     top         top  pacer  15.5  ...   \n",
       "131024        Kolkata Knight Riders     top         top  pacer  15.6  ...   \n",
       "131025        Kolkata Knight Riders     top         top  pacer  16.1  ...   \n",
       "\n",
       "        legbye_runs  noball_runs  batsman_runs  extra_runs  total_runs  \\\n",
       "0                 1            0             0           1           1   \n",
       "1                 0            0             0           0           0   \n",
       "2                 0            0             0           1           1   \n",
       "3                 0            0             0           0           0   \n",
       "4                 0            0             0           0           0   \n",
       "...             ...          ...           ...         ...         ...   \n",
       "131021            0            0             0           0           0   \n",
       "131022            0            0             4           0           4   \n",
       "131023            0            0             1           0           1   \n",
       "131024            0            0             1           0           1   \n",
       "131025            0            0             6           0           6   \n",
       "\n",
       "        dismissal_kind  wickets  \\\n",
       "0                             0   \n",
       "1                             0   \n",
       "2                             0   \n",
       "3                             0   \n",
       "4                             0   \n",
       "...                ...      ...   \n",
       "131021                        1   \n",
       "131022                        1   \n",
       "131023                        1   \n",
       "131024                        1   \n",
       "131025                        1   \n",
       "\n",
       "                                               commentary  ball_length  \\\n",
       "0       kumar to ganguly, 1 leg bye, and thats runs im...         full   \n",
       "1       kumar to mccullum, no run, this one nips back ...         good   \n",
       "2       kumar to mccullum, 1 wide, thats wide, as prav...      Unknown   \n",
       "3       kumar to mccullum, no run, too straight, mccul...      Unknown   \n",
       "4       kumar to mccullum, no run, steps down the trac...         good   \n",
       "...                                                   ...          ...   \n",
       "131021                prasidh krishna to sharma, no run,ê      Unknown   \n",
       "131022  prasidh krishna to sharma, four runs, chipped ...         full   \n",
       "131023  prasidh krishna to sharma, 1 run, very full, j...         full   \n",
       "131024  prasidh krishna to yadav, 1 run, prasidh digs ...        short   \n",
       "131025  russell to yadav, six runs, and the match ends...        short   \n",
       "\n",
       "       cumulative_runs  \n",
       "0                    1  \n",
       "1                    1  \n",
       "2                    2  \n",
       "3                    2  \n",
       "4                    2  \n",
       "...                ...  \n",
       "131021             122  \n",
       "131022             126  \n",
       "131023             127  \n",
       "131024             128  \n",
       "131025             134  \n",
       "\n",
       "[122815 rows x 22 columns]"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[df.match_id == \"round-robin\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.rename(columns={\"match_id\": \"match_type\", \"bowler\": \"bowler_type\"}, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature Engineering `runs_left_to_win` from `cumulative_runs` per ball "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/r6/q0qh39ts70582wth4g1dhjhm0000gn/T/ipykernel_8903/1057786160.py:5: SettingWithCopyWarning:\n",
      "\n",
      "\n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n"
     ]
    },
    {
     "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>season</th>\n",
       "      <th>match_type</th>\n",
       "      <th>venue</th>\n",
       "      <th>inning</th>\n",
       "      <th>batting_team</th>\n",
       "      <th>bowling_team</th>\n",
       "      <th>batsman</th>\n",
       "      <th>non_striker</th>\n",
       "      <th>bowler_type</th>\n",
       "      <th>ball</th>\n",
       "      <th>...</th>\n",
       "      <th>noball_runs</th>\n",
       "      <th>batsman_runs</th>\n",
       "      <th>extra_runs</th>\n",
       "      <th>total_runs</th>\n",
       "      <th>dismissal_kind</th>\n",
       "      <th>wickets</th>\n",
       "      <th>commentary</th>\n",
       "      <th>ball_length</th>\n",
       "      <th>cumulative_runs</th>\n",
       "      <th>runs_left_to_win</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2008</td>\n",
       "      <td>round-robin</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>kumar to ganguly, 1 leg bye, and thats runs im...</td>\n",
       "      <td>full</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2008</td>\n",
       "      <td>round-robin</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>kumar to mccullum, no run, this one nips back ...</td>\n",
       "      <td>good</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2008</td>\n",
       "      <td>round-robin</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.3</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>kumar to mccullum, 1 wide, thats wide, as prav...</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2008</td>\n",
       "      <td>round-robin</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.3</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>kumar to mccullum, no run, too straight, mccul...</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2008</td>\n",
       "      <td>round-robin</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>0</td>\n",
       "      <td>kumar to mccullum, no run, steps down the trac...</td>\n",
       "      <td>good</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   season   match_type      venue  inning           batting_team  \\\n",
       "0    2008  round-robin  Bangalore       1  Kolkata Knight Riders   \n",
       "1    2008  round-robin  Bangalore       1  Kolkata Knight Riders   \n",
       "2    2008  round-robin  Bangalore       1  Kolkata Knight Riders   \n",
       "3    2008  round-robin  Bangalore       1  Kolkata Knight Riders   \n",
       "4    2008  round-robin  Bangalore       1  Kolkata Knight Riders   \n",
       "\n",
       "                  bowling_team batsman non_striker bowler_type  ball  ...  \\\n",
       "0  Royal Challengers Bangalore     top         top       pacer   0.1  ...   \n",
       "1  Royal Challengers Bangalore     top         top       pacer   0.2  ...   \n",
       "2  Royal Challengers Bangalore     top         top       pacer   0.3  ...   \n",
       "3  Royal Challengers Bangalore     top         top       pacer   0.3  ...   \n",
       "4  Royal Challengers Bangalore     top         top       pacer   0.4  ...   \n",
       "\n",
       "   noball_runs  batsman_runs  extra_runs  total_runs  dismissal_kind  wickets  \\\n",
       "0            0             0           1           1                        0   \n",
       "1            0             0           0           0                        0   \n",
       "2            0             0           1           1                        0   \n",
       "3            0             0           0           0                        0   \n",
       "4            0             0           0           0                        0   \n",
       "\n",
       "                                          commentary ball_length  \\\n",
       "0  kumar to ganguly, 1 leg bye, and thats runs im...        full   \n",
       "1  kumar to mccullum, no run, this one nips back ...        good   \n",
       "2  kumar to mccullum, 1 wide, thats wide, as prav...     Unknown   \n",
       "3  kumar to mccullum, no run, too straight, mccul...     Unknown   \n",
       "4  kumar to mccullum, no run, steps down the trac...        good   \n",
       "\n",
       "   cumulative_runs runs_left_to_win  \n",
       "0                1              0.0  \n",
       "1                1              0.0  \n",
       "2                2              0.0  \n",
       "3                2              0.0  \n",
       "4                2              0.0  \n",
       "\n",
       "[5 rows x 23 columns]"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Find the rows where inning==1 and the inning changes from 1 to 2 (i.e. the last ball of the first innings)\n",
    "last_ball_first_innings = df[(df[\"inning\"] == 1) & (df[\"inning\"].shift(-1) == 2)]\n",
    "\n",
    "# Calculate the target value for each match\n",
    "last_ball_first_innings[\"target\"] = last_ball_first_innings[\"cumulative_runs\"] + 1\n",
    "\n",
    "# Merge the 'target' column back to the original DataFrame based on matching columns\n",
    "df = pd.merge(\n",
    "    df,\n",
    "    last_ball_first_innings[\n",
    "        [\n",
    "            \"season\",\n",
    "            \"match_type\",\n",
    "            \"venue\",\n",
    "            \"inning\",\n",
    "            \"ball\",\n",
    "            \"batting_team\",\n",
    "            \"bowling_team\",\n",
    "            \"target\",\n",
    "        ]\n",
    "    ],\n",
    "    on=[\n",
    "        \"season\",\n",
    "        \"match_type\",\n",
    "        \"venue\",\n",
    "        \"inning\",\n",
    "        \"ball\",\n",
    "        \"batting_team\",\n",
    "        \"bowling_team\",\n",
    "    ],\n",
    "    how=\"left\",\n",
    ")\n",
    "\n",
    "# now we have the target column for each ball but it is only populated for the last ball of the first innings\n",
    "\n",
    "# populate second innings target values for each ball with the target value of the last ball of the first innings\n",
    "\n",
    "# using while loop to populate target values for each ball in the second innings\n",
    "\n",
    "i = 0\n",
    "\n",
    "while i < df.shape[0]:\n",
    "    if df.loc[i, \"inning\"] == 2:\n",
    "        df.loc[i, \"target\"] = df.loc[i - 1, \"target\"]\n",
    "    i += 1\n",
    "\n",
    "# populate all first innings target values for each ball with 0\n",
    "\n",
    "df[\"target\"] = np.where((df[\"inning\"] == 1), 0, df[\"target\"])\n",
    "\n",
    "# convert `target` column to runs_left_to_win\n",
    "\n",
    "df[\"target\"] = np.where(\n",
    "    (df[\"inning\"] == 2), df[\"target\"] - df[\"cumulative_runs\"], df[\"target\"]\n",
    ")\n",
    "\n",
    "df.rename(columns={\"target\": \"runs_left_to_win\"}, inplace=True)\n",
    "\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(132007, 23)"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature Engineering `Event` Response Variable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"batsman_runs\"] = df[\"batsman_runs\"].astype(\"int64\")\n",
    "df[\"total_runs\"] = df[\"total_runs\"].astype(\"int64\")\n",
    "df[\"extra_runs\"] = df[\"extra_runs\"].astype(\"int64\")\n",
    "df[\"noball_runs\"] = df[\"noball_runs\"].astype(\"int64\")\n",
    "df[\"legbye_runs\"] = df[\"legbye_runs\"].astype(\"int64\")\n",
    "df[\"bye_runs\"] = df[\"bye_runs\"].astype(\"int64\")\n",
    "df[\"wide_runs\"] = df[\"wide_runs\"].astype(\"int64\")\n",
    "\n",
    "\n",
    "def event_pred(row):\n",
    "    if (\n",
    "        row[\"batsman_runs\"] == 0\n",
    "        and row[\"extra_runs\"] == 0\n",
    "        and row[\"total_runs\"] == 0\n",
    "        and row[\"dismissal_kind\"] != \"bowled\"\n",
    "        and row[\"dismissal_kind\"] != \"caught\"\n",
    "        and row[\"dismissal_kind\"] != \"caught and bowled\"\n",
    "        and row[\"dismissal_kind\"] != \"hit wicket\"\n",
    "        and row[\"dismissal_kind\"] != \"lbw\"\n",
    "        and row[\"dismissal_kind\"] != \"obstructing the field\"\n",
    "        and row[\"dismissal_kind\"] != \"retired hurt\"\n",
    "        and row[\"dismissal_kind\"] != \"run out\"\n",
    "        and row[\"dismissal_kind\"] != \"stumped\"\n",
    "    ):\n",
    "        return \"Dot\"\n",
    "    if (\n",
    "        row[\"batsman_runs\"] > 0\n",
    "        and row[\"extra_runs\"] == 0\n",
    "        and row[\"total_runs\"] > 0\n",
    "        and row[\"dismissal_kind\"] != \"bowled\"\n",
    "        and row[\"dismissal_kind\"] != \"caught\"\n",
    "        and row[\"dismissal_kind\"] != \"caught and bowled\"\n",
    "        and row[\"dismissal_kind\"] != \"hit wicket\"\n",
    "        and row[\"dismissal_kind\"] != \"lbw\"\n",
    "        and row[\"dismissal_kind\"] != \"obstructing the field\"\n",
    "        and row[\"dismissal_kind\"] != \"retired hurt\"\n",
    "        and row[\"dismissal_kind\"] != \"run out\"\n",
    "        and row[\"dismissal_kind\"] != \"stumped\"\n",
    "    ):\n",
    "        return \"Runs\"\n",
    "    if (\n",
    "        row[\"wide_runs\"] > 0\n",
    "        or row[\"bye_runs\"] > 0\n",
    "        or row[\"legbye_runs\"] > 0\n",
    "        or row[\"noball_runs\"] > 0\n",
    "        or row[\"batsman_runs\"] >= 0\n",
    "        and row[\"extra_runs\"] > 0\n",
    "        and row[\"total_runs\"] > 0\n",
    "        and row[\"dismissal_kind\"] != \"bowled\"\n",
    "        and row[\"dismissal_kind\"] != \"caught\"\n",
    "        and row[\"dismissal_kind\"] != \"caught and bowled\"\n",
    "        and row[\"dismissal_kind\"] != \"hit wicket\"\n",
    "        and row[\"dismissal_kind\"] != \"lbw\"\n",
    "        and row[\"dismissal_kind\"] != \"obstructing the field\"\n",
    "        and row[\"dismissal_kind\"] != \"retired hurt\"\n",
    "        and row[\"dismissal_kind\"] != \"run out\"\n",
    "        and row[\"dismissal_kind\"] != \"stumped\"\n",
    "    ):\n",
    "        return \"Extras\"\n",
    "    if (\n",
    "        row[\"dismissal_kind\"] == \"bowled\"\n",
    "        or row[\"dismissal_kind\"] == \"caught\"\n",
    "        or row[\"dismissal_kind\"] == \"caught and bowled\"\n",
    "        or row[\"dismissal_kind\"] == \"hit wicket\"\n",
    "        or row[\"dismissal_kind\"] == \"lbw\"\n",
    "        or row[\"dismissal_kind\"] == \"obstructing the field\"\n",
    "        or row[\"dismissal_kind\"] == \"retired hurt\"\n",
    "        or row[\"dismissal_kind\"] == \"run out\"\n",
    "        or row[\"dismissal_kind\"] == \"stumped\"\n",
    "    ):\n",
    "        return \"WICKET!\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"event\"] = df.apply(lambda row: event_pred(row), axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "39869"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[df.event == \"Dot\"].shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "78599"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[df.event == \"Runs\"].shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "7080"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[df.event == \"Extras\"].shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "6459"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[df.event == \"WICKET!\"].shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "season                int64\n",
       "match_type           object\n",
       "venue                object\n",
       "inning                int64\n",
       "batting_team         object\n",
       "bowling_team         object\n",
       "batsman              object\n",
       "non_striker          object\n",
       "bowler_type          object\n",
       "ball                float64\n",
       "wide_runs             int64\n",
       "bye_runs              int64\n",
       "legbye_runs           int64\n",
       "noball_runs           int64\n",
       "batsman_runs          int64\n",
       "extra_runs            int64\n",
       "total_runs            int64\n",
       "dismissal_kind       object\n",
       "wickets               int64\n",
       "commentary           object\n",
       "ball_length          object\n",
       "cumulative_runs       int64\n",
       "runs_left_to_win    float64\n",
       "event                object\n",
       "dtype: object"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Dropping Unnecessary Columns "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "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>season</th>\n",
       "      <th>match_type</th>\n",
       "      <th>venue</th>\n",
       "      <th>inning</th>\n",
       "      <th>batting_team</th>\n",
       "      <th>bowling_team</th>\n",
       "      <th>batsman</th>\n",
       "      <th>non_striker</th>\n",
       "      <th>bowler_type</th>\n",
       "      <th>ball</th>\n",
       "      <th>wickets</th>\n",
       "      <th>ball_length</th>\n",
       "      <th>cumulative_runs</th>\n",
       "      <th>runs_left_to_win</th>\n",
       "      <th>event</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2008</td>\n",
       "      <td>round-robin</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0</td>\n",
       "      <td>full</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>Extras</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2008</td>\n",
       "      <td>round-robin</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "      <td>good</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>Dot</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2008</td>\n",
       "      <td>round-robin</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.3</td>\n",
       "      <td>0</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>Extras</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2008</td>\n",
       "      <td>round-robin</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.3</td>\n",
       "      <td>0</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>Dot</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2008</td>\n",
       "      <td>round-robin</td>\n",
       "      <td>Bangalore</td>\n",
       "      <td>1</td>\n",
       "      <td>Kolkata Knight Riders</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.4</td>\n",
       "      <td>0</td>\n",
       "      <td>good</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>Dot</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   season   match_type      venue inning           batting_team  \\\n",
       "0    2008  round-robin  Bangalore      1  Kolkata Knight Riders   \n",
       "1    2008  round-robin  Bangalore      1  Kolkata Knight Riders   \n",
       "2    2008  round-robin  Bangalore      1  Kolkata Knight Riders   \n",
       "3    2008  round-robin  Bangalore      1  Kolkata Knight Riders   \n",
       "4    2008  round-robin  Bangalore      1  Kolkata Knight Riders   \n",
       "\n",
       "                  bowling_team batsman non_striker bowler_type  ball  wickets  \\\n",
       "0  Royal Challengers Bangalore     top         top       pacer   0.1        0   \n",
       "1  Royal Challengers Bangalore     top         top       pacer   0.2        0   \n",
       "2  Royal Challengers Bangalore     top         top       pacer   0.3        0   \n",
       "3  Royal Challengers Bangalore     top         top       pacer   0.3        0   \n",
       "4  Royal Challengers Bangalore     top         top       pacer   0.4        0   \n",
       "\n",
       "  ball_length  cumulative_runs  runs_left_to_win   event  \n",
       "0        full                1                 0  Extras  \n",
       "1        good                1                 0     Dot  \n",
       "2     Unknown                2                 0  Extras  \n",
       "3     Unknown                2                 0     Dot  \n",
       "4        good                2                 0     Dot  "
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[\"match_type\"] = df[\"match_type\"].astype(\"category\")\n",
    "df[\"venue\"] = df[\"venue\"].astype(\"category\")\n",
    "df[\"inning\"] = df[\"inning\"].astype(\"category\")\n",
    "df[\"batting_team\"] = df[\"batting_team\"].astype(\"category\")\n",
    "df[\"bowling_team\"] = df[\"bowling_team\"].astype(\"category\")\n",
    "df[\"batsman\"] = df[\"batsman\"].astype(\"category\")\n",
    "df[\"non_striker\"] = df[\"non_striker\"].astype(\"category\")\n",
    "df[\"bowler_type\"] = df[\"bowler_type\"].astype(\"category\")\n",
    "df[\"ball_length\"] = df[\"ball_length\"].astype(\"category\")\n",
    "df[\"ball\"] = df[\"ball\"].astype(\"float64\")\n",
    "df[\"wickets\"] = df[\"wickets\"].astype(\"int64\")\n",
    "df[\"cumulative_runs\"] = df[\"cumulative_runs\"].astype(\"int64\")\n",
    "df[\"runs_left_to_win\"] = df[\"runs_left_to_win\"].astype(\"int64\")\n",
    "\n",
    "df[\"event\"] = df[\"event\"].astype(\"object\")\n",
    "\n",
    "df.drop(labels=\"commentary\", axis=1, inplace=True)\n",
    "df.drop(labels=\"wide_runs\", axis=1, inplace=True)\n",
    "df.drop(labels=\"bye_runs\", axis=1, inplace=True)\n",
    "df.drop(labels=\"legbye_runs\", axis=1, inplace=True)\n",
    "df.drop(labels=\"noball_runs\", axis=1, inplace=True)\n",
    "df.drop(labels=\"batsman_runs\", axis=1, inplace=True)\n",
    "df.drop(labels=\"extra_runs\", axis=1, inplace=True)\n",
    "df.drop(labels=\"dismissal_kind\", axis=1, inplace=True)\n",
    "df.drop(labels=\"total_runs\", axis=1, inplace=True)\n",
    "\n",
    "#df.to_csv(\"data/clean/deliveries_clean.csv\", index=False)\n",
    "\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "season                 int64\n",
       "match_type          category\n",
       "venue               category\n",
       "inning              category\n",
       "batting_team        category\n",
       "bowling_team        category\n",
       "batsman             category\n",
       "non_striker         category\n",
       "bowler_type         category\n",
       "ball                 float64\n",
       "wickets                int64\n",
       "ball_length         category\n",
       "cumulative_runs        int64\n",
       "runs_left_to_win       int64\n",
       "event                 object\n",
       "dtype: object"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "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>season</th>\n",
       "      <th>ball</th>\n",
       "      <th>wickets</th>\n",
       "      <th>cumulative_runs</th>\n",
       "      <th>runs_left_to_win</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>132007.000000</td>\n",
       "      <td>132007.000000</td>\n",
       "      <td>132007.000000</td>\n",
       "      <td>132007.000000</td>\n",
       "      <td>132007.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2013.725265</td>\n",
       "      <td>9.547537</td>\n",
       "      <td>2.380116</td>\n",
       "      <td>75.334558</td>\n",
       "      <td>45.397244</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>3.435975</td>\n",
       "      <td>5.675536</td>\n",
       "      <td>2.061154</td>\n",
       "      <td>48.509273</td>\n",
       "      <td>58.396270</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>2008.000000</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-7.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>2011.000000</td>\n",
       "      <td>4.500000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>35.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2013.000000</td>\n",
       "      <td>9.400000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>71.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2017.000000</td>\n",
       "      <td>14.400000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>111.000000</td>\n",
       "      <td>91.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2019.000000</td>\n",
       "      <td>19.600000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>246.000000</td>\n",
       "      <td>247.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              season           ball        wickets  cumulative_runs  \\\n",
       "count  132007.000000  132007.000000  132007.000000    132007.000000   \n",
       "mean     2013.725265       9.547537       2.380116        75.334558   \n",
       "std         3.435975       5.675536       2.061154        48.509273   \n",
       "min      2008.000000       0.100000       0.000000         0.000000   \n",
       "25%      2011.000000       4.500000       1.000000        35.000000   \n",
       "50%      2013.000000       9.400000       2.000000        71.000000   \n",
       "75%      2017.000000      14.400000       4.000000       111.000000   \n",
       "max      2019.000000      19.600000      10.000000       246.000000   \n",
       "\n",
       "       runs_left_to_win  \n",
       "count     132007.000000  \n",
       "mean          45.397244  \n",
       "std           58.396270  \n",
       "min           -7.000000  \n",
       "25%            0.000000  \n",
       "50%            0.000000  \n",
       "75%           91.000000  \n",
       "max          247.000000  "
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "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>season</th>\n",
       "      <th>match_type</th>\n",
       "      <th>venue</th>\n",
       "      <th>inning</th>\n",
       "      <th>batting_team</th>\n",
       "      <th>bowling_team</th>\n",
       "      <th>batsman</th>\n",
       "      <th>non_striker</th>\n",
       "      <th>bowler_type</th>\n",
       "      <th>ball</th>\n",
       "      <th>wickets</th>\n",
       "      <th>ball_length</th>\n",
       "      <th>cumulative_runs</th>\n",
       "      <th>runs_left_to_win</th>\n",
       "      <th>event</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>21144</th>\n",
       "      <td>2010</td>\n",
       "      <td>round-robin</td>\n",
       "      <td>Chennai</td>\n",
       "      <td>2</td>\n",
       "      <td>Rajasthan Royals</td>\n",
       "      <td>Chennai Super Kings</td>\n",
       "      <td>top</td>\n",
       "      <td>top</td>\n",
       "      <td>pacer</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0</td>\n",
       "      <td>good</td>\n",
       "      <td>0</td>\n",
       "      <td>247</td>\n",
       "      <td>Dot</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       season   match_type    venue inning      batting_team  \\\n",
       "21144    2010  round-robin  Chennai      2  Rajasthan Royals   \n",
       "\n",
       "              bowling_team batsman non_striker bowler_type  ball  wickets  \\\n",
       "21144  Chennai Super Kings     top         top       pacer   0.1        0   \n",
       "\n",
       "      ball_length  cumulative_runs  runs_left_to_win event  \n",
       "21144        good                0               247   Dot  "
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# sanity check - get row where runs_left_to_win is max\n",
    "\n",
    "df.loc[df.runs_left_to_win == df.runs_left_to_win.max()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "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>season</th>\n",
       "      <th>match_type</th>\n",
       "      <th>venue</th>\n",
       "      <th>inning</th>\n",
       "      <th>batting_team</th>\n",
       "      <th>bowling_team</th>\n",
       "      <th>batsman</th>\n",
       "      <th>non_striker</th>\n",
       "      <th>bowler_type</th>\n",
       "      <th>ball</th>\n",
       "      <th>wickets</th>\n",
       "      <th>ball_length</th>\n",
       "      <th>cumulative_runs</th>\n",
       "      <th>runs_left_to_win</th>\n",
       "      <th>event</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>114201</th>\n",
       "      <td>2018</td>\n",
       "      <td>round-robin</td>\n",
       "      <td>Delhi</td>\n",
       "      <td>2</td>\n",
       "      <td>Royal Challengers Bangalore</td>\n",
       "      <td>Delhi Capitals</td>\n",
       "      <td>top</td>\n",
       "      <td>middle</td>\n",
       "      <td>pacer</td>\n",
       "      <td>18.6</td>\n",
       "      <td>5</td>\n",
       "      <td>full</td>\n",
       "      <td>190</td>\n",
       "      <td>-7</td>\n",
       "      <td>Runs</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        season   match_type  venue inning                 batting_team  \\\n",
       "114201    2018  round-robin  Delhi      2  Royal Challengers Bangalore   \n",
       "\n",
       "          bowling_team batsman non_striker bowler_type  ball  wickets  \\\n",
       "114201  Delhi Capitals     top      middle       pacer  18.6        5   \n",
       "\n",
       "       ball_length  cumulative_runs  runs_left_to_win event  \n",
       "114201        full              190                -7  Runs  "
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# sanity check - get row where runs_left_to_win is min\n",
    "\n",
    "df.loc[df.runs_left_to_win == df.runs_left_to_win.min()]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train-Test Split"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since the dataset resembles a time-series, the test set has all ball by ball data for the last season (2019) and the training set has all ball by ball data for the seasons prior to 2019. Although recent data is more relevant to the current state of the game, we believe that the training set should have more data to train the model on. Therefore, we have chosen to split the data in this manner."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/Tegveer/miniforge3/envs/ANLY590/lib/python3.9/site-packages/pandas/core/frame.py:4308: SettingWithCopyWarning:\n",
      "\n",
      "\n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# split the dataset into train and test sets by season\n",
    "\n",
    "train = df.loc[df.season <= 2018]\n",
    "\n",
    "#train.to_csv(\"data/clean/train.csv\", index=False)\n",
    "\n",
    "test = df.loc[df.season == 2019]\n",
    "\n",
    "#test.to_csv(\"data/clean/test.csv\", index=False)\n",
    "\n",
    "# drop `season` column from train and test sets\n",
    "\n",
    "train.drop(labels=\"season\", axis=1, inplace=True)\n",
    "test.drop(labels=\"season\", axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Encoding Categorical Features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "# from sklearn.preprocessing import minmax_scale\n",
    "# df[['ball','wickets']] = minmax_scale(df[['ball','wickets']])\n",
    "\n",
    "encoded_df_train = pd.get_dummies(\n",
    "    data=train,\n",
    "    columns=[\n",
    "        \"match_type\",\n",
    "        \"venue\",\n",
    "        \"inning\",\n",
    "        \"batting_team\",\n",
    "        \"bowling_team\",\n",
    "        \"batsman\",\n",
    "        \"non_striker\",\n",
    "        \"bowler_type\",\n",
    "        \"ball_length\",\n",
    "    ],\n",
    ")\n",
    "\n",
    "encoded_df_test = pd.get_dummies(\n",
    "    data=test,\n",
    "    columns=[\n",
    "        \"match_type\",\n",
    "        \"venue\",\n",
    "        \"inning\",\n",
    "        \"batting_team\",\n",
    "        \"bowling_team\",\n",
    "        \"batsman\",\n",
    "        \"non_striker\",\n",
    "        \"bowler_type\",\n",
    "        \"ball_length\",\n",
    "    ],\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['ball', 'wickets', 'cumulative_runs', 'runs_left_to_win', 'event',\n",
       "       'match_type_playoffs', 'match_type_round-robin', 'venue_Ahmedabad',\n",
       "       'venue_Bangalore', 'venue_Chennai', 'venue_Cuttack', 'venue_Delhi',\n",
       "       'venue_Dharamsala', 'venue_Hyderabad', 'venue_Indore', 'venue_Kolkata',\n",
       "       'venue_Mohali', 'venue_Mumbai', 'venue_Nagpur', 'venue_Pune',\n",
       "       'venue_Raipur', 'venue_Rajasthan', 'venue_Ranchi',\n",
       "       'venue_Visakhapatnam', 'inning_1', 'inning_2',\n",
       "       'batting_team_Chennai Super Kings', 'batting_team_Delhi Capitals',\n",
       "       'batting_team_Kings XI Punjab', 'batting_team_Kolkata Knight Riders',\n",
       "       'batting_team_Mumbai Indians', 'batting_team_Rajasthan Royals',\n",
       "       'batting_team_Royal Challengers Bangalore',\n",
       "       'batting_team_Sunrisers Hyderabad', 'bowling_team_Chennai Super Kings',\n",
       "       'bowling_team_Delhi Capitals', 'bowling_team_Kings XI Punjab',\n",
       "       'bowling_team_Kolkata Knight Riders', 'bowling_team_Mumbai Indians',\n",
       "       'bowling_team_Rajasthan Royals',\n",
       "       'bowling_team_Royal Challengers Bangalore',\n",
       "       'bowling_team_Sunrisers Hyderabad', 'batsman_middle', 'batsman_tail',\n",
       "       'batsman_top', 'non_striker_middle', 'non_striker_tail',\n",
       "       'non_striker_top', 'bowler_type_pacer', 'bowler_type_spinner',\n",
       "       'ball_length_Unknown', 'ball_length_full', 'ball_length_full toss',\n",
       "       'ball_length_good', 'ball_length_short', 'ball_length_yorker'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "encoded_df_train.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(117763, 56)\n",
      "(14244, 56)\n"
     ]
    }
   ],
   "source": [
    "print(encoded_df_train.shape)\n",
    "print(encoded_df_test.shape)\n",
    "\n",
    "# 88-12 train-test split"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Rearranging Columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "encoded_df_train = encoded_df_train[\n",
    "    [\n",
    "        \"ball\",\n",
    "        \"wickets\",\n",
    "        \"cumulative_runs\",\n",
    "        \"runs_left_to_win\",\n",
    "        \"match_type_playoffs\",\n",
    "        \"match_type_round-robin\",\n",
    "        \"venue_Ahmedabad\",\n",
    "        \"venue_Bangalore\",\n",
    "        \"venue_Chennai\",\n",
    "        \"venue_Cuttack\",\n",
    "        \"venue_Delhi\",\n",
    "        \"venue_Dharamsala\",\n",
    "        \"venue_Hyderabad\",\n",
    "        \"venue_Indore\",\n",
    "        \"venue_Kolkata\",\n",
    "        \"venue_Mohali\",\n",
    "        \"venue_Mumbai\",\n",
    "        \"venue_Nagpur\",\n",
    "        \"venue_Pune\",\n",
    "        \"venue_Raipur\",\n",
    "        \"venue_Rajasthan\",\n",
    "        \"venue_Ranchi\",\n",
    "        \"venue_Visakhapatnam\",\n",
    "        \"inning_1\",\n",
    "        \"inning_2\",\n",
    "        \"batting_team_Chennai Super Kings\",\n",
    "        \"batting_team_Delhi Capitals\",\n",
    "        \"batting_team_Kings XI Punjab\",\n",
    "        \"batting_team_Kolkata Knight Riders\",\n",
    "        \"batting_team_Mumbai Indians\",\n",
    "        \"batting_team_Rajasthan Royals\",\n",
    "        \"batting_team_Royal Challengers Bangalore\",\n",
    "        \"batting_team_Sunrisers Hyderabad\",\n",
    "        \"bowling_team_Chennai Super Kings\",\n",
    "        \"bowling_team_Delhi Capitals\",\n",
    "        \"bowling_team_Kings XI Punjab\",\n",
    "        \"bowling_team_Kolkata Knight Riders\",\n",
    "        \"bowling_team_Mumbai Indians\",\n",
    "        \"bowling_team_Rajasthan Royals\",\n",
    "        \"bowling_team_Royal Challengers Bangalore\",\n",
    "        \"bowling_team_Sunrisers Hyderabad\",\n",
    "        \"batsman_top\",\n",
    "        \"batsman_middle\",\n",
    "        \"batsman_tail\",\n",
    "        \"non_striker_middle\",\n",
    "        \"non_striker_tail\",\n",
    "        \"non_striker_top\",\n",
    "        \"bowler_type_pacer\",\n",
    "        \"bowler_type_spinner\",\n",
    "        \"ball_length_Unknown\",\n",
    "        \"ball_length_full\",\n",
    "        \"ball_length_full toss\",\n",
    "        \"ball_length_good\",\n",
    "        \"ball_length_short\",\n",
    "        \"ball_length_yorker\",\n",
    "        \"event\",\n",
    "    ]\n",
    "]\n",
    "\n",
    "encoded_df_test = encoded_df_test[\n",
    "    [\n",
    "        \"ball\",\n",
    "        \"wickets\",\n",
    "        \"cumulative_runs\",\n",
    "        \"runs_left_to_win\",\n",
    "        \"match_type_playoffs\",\n",
    "        \"match_type_round-robin\",\n",
    "        \"venue_Ahmedabad\",\n",
    "        \"venue_Bangalore\",\n",
    "        \"venue_Chennai\",\n",
    "        \"venue_Cuttack\",\n",
    "        \"venue_Delhi\",\n",
    "        \"venue_Dharamsala\",\n",
    "        \"venue_Hyderabad\",\n",
    "        \"venue_Indore\",\n",
    "        \"venue_Kolkata\",\n",
    "        \"venue_Mohali\",\n",
    "        \"venue_Mumbai\",\n",
    "        \"venue_Nagpur\",\n",
    "        \"venue_Pune\",\n",
    "        \"venue_Raipur\",\n",
    "        \"venue_Rajasthan\",\n",
    "        \"venue_Ranchi\",\n",
    "        \"venue_Visakhapatnam\",\n",
    "        \"inning_1\",\n",
    "        \"inning_2\",\n",
    "        \"batting_team_Chennai Super Kings\",\n",
    "        \"batting_team_Delhi Capitals\",\n",
    "        \"batting_team_Kings XI Punjab\",\n",
    "        \"batting_team_Kolkata Knight Riders\",\n",
    "        \"batting_team_Mumbai Indians\",\n",
    "        \"batting_team_Rajasthan Royals\",\n",
    "        \"batting_team_Royal Challengers Bangalore\",\n",
    "        \"batting_team_Sunrisers Hyderabad\",\n",
    "        \"bowling_team_Chennai Super Kings\",\n",
    "        \"bowling_team_Delhi Capitals\",\n",
    "        \"bowling_team_Kings XI Punjab\",\n",
    "        \"bowling_team_Kolkata Knight Riders\",\n",
    "        \"bowling_team_Mumbai Indians\",\n",
    "        \"bowling_team_Rajasthan Royals\",\n",
    "        \"bowling_team_Royal Challengers Bangalore\",\n",
    "        \"bowling_team_Sunrisers Hyderabad\",\n",
    "        \"batsman_top\",\n",
    "        \"batsman_middle\",\n",
    "        \"batsman_tail\",\n",
    "        \"non_striker_middle\",\n",
    "        \"non_striker_tail\",\n",
    "        \"non_striker_top\",\n",
    "        \"bowler_type_pacer\",\n",
    "        \"bowler_type_spinner\",\n",
    "        \"ball_length_Unknown\",\n",
    "        \"ball_length_full\",\n",
    "        \"ball_length_full toss\",\n",
    "        \"ball_length_good\",\n",
    "        \"ball_length_short\",\n",
    "        \"ball_length_yorker\",\n",
    "        \"event\",\n",
    "    ]\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['Extras', 'Dot', 'Runs', 'WICKET!'], dtype=object)"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "encoded_df_train[\"event\"].unique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Extracting Features and Response Variable for Training and Testing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_train shape: (117763, 55)\n",
      "y_train shape: (117763,)\n",
      "X_test shape: (14244, 55)\n",
      "y_test shape: (14244,)\n"
     ]
    }
   ],
   "source": [
    "X_train = encoded_df_train.iloc[:, :-1].values\n",
    "print(\"X_train shape:\", X_train.shape)\n",
    "\n",
    "y_train = encoded_df_train.iloc[:, -1].values\n",
    "print(\"y_train shape:\", y_train.shape)\n",
    "\n",
    "X_test = encoded_df_test.iloc[:, :-1].values\n",
    "print(\"X_test shape:\", X_test.shape)\n",
    "\n",
    "y_test = encoded_df_test.iloc[:, -1].values\n",
    "print(\"y_test shape:\", y_test.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Encoding `y_train` and `y_test` "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "encoder = LabelEncoder()\n",
    "\n",
    "y_train = encoder.fit_transform(y_train)\n",
    "\n",
    "y_test = encoder.fit_transform(y_test)\n",
    "\n",
    "y_train = y_train.reshape(\n",
    "    len(y_train),\n",
    ")\n",
    "\n",
    "y_test = y_test.reshape(\n",
    "    len(y_test),\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Perform `RepeatedKFold` Classification Routine"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 221,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|          | 0/9 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">LR Accuracy: 0.609 Weighted F1: 0.532\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 11%|█         | 1/9 [01:41<13:35, 101.94s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">Tree Accuracy: 0.491 Weighted F1: 0.495\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 33%|███▎      | 3/9 [01:46<02:26, 24.49s/it] "
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">GNB Accuracy: 0.482 Weighted F1: 0.492\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/Tegveer/miniforge3/envs/ANLY501/lib/python3.9/site-packages/joblib/externals/loky/process_executor.py:752: UserWarning:\n",
      "\n",
      "A worker stopped while some jobs were given to the executor. This can be caused by a too short worker timeout or by a memory leak.\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">RF Accuracy: 0.552 Weighted F1: 0.528\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 44%|████▍     | 4/9 [02:39<02:58, 35.79s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">AdaBoost Accuracy: 0.616 Weighted F1: 0.545\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 56%|█████▌    | 5/9 [02:59<01:59, 29.89s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">Bagging Accuracy: 0.551 Weighted F1: 0.533\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 67%|██████▋   | 6/9 [03:20<01:20, 27.00s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">GBM Accuracy: 0.623 Weighted F1: 0.545\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 78%|███████▊  | 7/9 [08:16<03:50, 115.01s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">XGB Accuracy: 0.622 Weighted F1: 0.553\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 89%|████████▉ | 8/9 [08:33<01:23, 83.69s/it] "
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">CatBoost Accuracy: 0.623 Weighted F1: 0.556\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 9/9 [11:06<00:00, 74.07s/it] "
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 5min 31s, sys: 35.4 s, total: 6min 7s\n",
      "Wall time: 11min 6s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "from sklearn.exceptions import ConvergenceWarning\n",
    "from sklearn.exceptions import DataConversionWarning\n",
    "from tqdm import tqdm  # Import tqdm\n",
    "\n",
    "# Assuming other necessary imports like LogisticRegression, DecisionTreeClassifier, etc., are already made\n",
    "\n",
    "ConvergenceWarning(\"ignore\")\n",
    "warnings.filterwarnings(action=\"ignore\", category=DataConversionWarning)\n",
    "\n",
    "\n",
    "def base_models():\n",
    "    models = dict()\n",
    "    models[\"LR\"] = LogisticRegression(verbose=0, max_iter=3000)\n",
    "    models[\"Tree\"] = DecisionTreeClassifier()\n",
    "    models[\"GNB\"] = GaussianNB()\n",
    "    models[\"RF\"] = RandomForestClassifier(verbose=0)\n",
    "    models[\"AdaBoost\"] = AdaBoostClassifier()\n",
    "    models[\"Bagging\"] = BaggingClassifier(verbose=0)\n",
    "    models[\"GBM\"] = GradientBoostingClassifier(verbose=0)\n",
    "    models[\"XGB\"] = XGBClassifier(verbosity=0)\n",
    "    models[\"CatBoost\"] = CatBoostClassifier(verbose=0)\n",
    "    return models\n",
    "\n",
    "\n",
    "# Function to evaluate the list of models\n",
    "def eval_models(model):\n",
    "    cv = RepeatedKFold(n_splits=5, n_repeats=2, random_state=1)\n",
    "    accuracy = cross_val_score(\n",
    "        model,\n",
    "        X_train,\n",
    "        y_train,\n",
    "        scoring=\"accuracy\",\n",
    "        cv=cv,\n",
    "        n_jobs=-1,\n",
    "        error_score=\"raise\",\n",
    "    )\n",
    "    weighted_f1 = cross_val_score(\n",
    "        model,\n",
    "        X_train,\n",
    "        y_train,\n",
    "        scoring=\"f1_weighted\",\n",
    "        cv=cv,\n",
    "        n_jobs=-1,\n",
    "        error_score=\"raise\",\n",
    "    )\n",
    "    return accuracy, weighted_f1\n",
    "\n",
    "\n",
    "models = base_models()\n",
    "\n",
    "# evaluate the models and store results\n",
    "results = {}\n",
    "names = list()\n",
    "accuracies = []\n",
    "weighted_f1s = []\n",
    "\n",
    "final_models = {}  # Create a dictionary to store the fitted models\n",
    "\n",
    "for name, model in tqdm(\n",
    "    models.items()\n",
    "):  # Wrap models.items() with tqdm to get a progress bar\n",
    "    accuracy, weighted_f1 = eval_models(model)\n",
    "    accuracies.extend(accuracy)\n",
    "    weighted_f1s.extend(weighted_f1)\n",
    "    names.extend([name] * len(accuracy))\n",
    "    print(\n",
    "        \">%s Accuracy: %.3f Weighted F1: %.3f\"\n",
    "        % (name, accuracy.mean(), weighted_f1.mean())\n",
    "    )\n",
    "\n",
    "    # Fit the model with the training data\n",
    "    fitted_model = model.fit(X_train, y_train)\n",
    "\n",
    "    # Save the fitted model as a pickle file\n",
    "    pkl_filename = f\"products/pickle_files/{name}.pkl\"\n",
    "    with open(pkl_filename, \"wb\") as file:\n",
    "        pickle.dump(fitted_model, file)\n",
    "\n",
    "    # Store the fitted model in the final_models dictionary\n",
    "    final_models[name] = fitted_model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 222,
   "metadata": {},
   "outputs": [],
   "source": [
    "# delete catboost_info folder\n",
    "\n",
    "shutil.rmtree(\"catboost_info\")\n",
    "\n",
    "# Save accuracy and weighted_f1 scores for each model in a dataframe\n",
    "\n",
    "results[\"Model\"] = names\n",
    "results[\"Accuracy\"] = accuracies\n",
    "results[\"Weighted_F1\"] = weighted_f1s\n",
    "\n",
    "classifmod = pd.DataFrame(results)\n",
    "\n",
    "# Generate confusion matrices and reports for each final model\n",
    "confusion_matrices = {}\n",
    "classification_reports = {}\n",
    "for name, model in final_models.items():\n",
    "    y_pred = model.predict(X_test)\n",
    "    confusion_matrices[name] = confusion_matrix(y_test, y_pred)\n",
    "    classification_reports[name] = classification_report(\n",
    "        y_test, y_pred, zero_division=0\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "for name, model in models.items():\n",
    "    pkl_filename = f\"products/pickle_files/{name}.pkl\"\n",
    "    # open pickle file\n",
    "    with open(pkl_filename, \"rb\") as infile:\n",
    "        obj = pickle.load(infile)\n",
    "\n",
    "    # convert pickle object to json object\n",
    "    json_obj = json.loads(json.dumps(obj, default=str))\n",
    "    json_filename = f\"products/json_files/{name}.json\"\n",
    "    # write the json file\n",
    "    with open(json_filename, \"w\", encoding=\"utf-8\") as outfile:\n",
    "        json.dump(json_obj, outfile, ensure_ascii=False, indent=4)\n",
    "\n",
    "# delete all pickle files for smooth git push\n",
    "\n",
    "# for name, model in models.items():\n",
    "#     pkl_filename = f\"products/pickle_files/{name}.pkl\"\n",
    "#     os.remove(pkl_filename)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training Performance Boxplots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 223,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "        <script type=\"text/javascript\">\n",
       "        window.PlotlyConfig = {MathJaxConfig: 'local'};\n",
       "        if (window.MathJax && window.MathJax.Hub && window.MathJax.Hub.Config) {window.MathJax.Hub.Config({SVG: {font: \"STIX-Web\"}});}\n",
       "        if (typeof require !== 'undefined') {\n",
       "        require.undef(\"plotly\");\n",
       "        requirejs.config({\n",
       "            paths: {\n",
       "                'plotly': ['https://cdn.plot.ly/plotly-2.20.0.min']\n",
       "            }\n",
       "        });\n",
       "        require(['plotly'], function(Plotly) {\n",
       "            window._Plotly = Plotly;\n",
       "        });\n",
       "        }\n",
       "        </script>\n",
       "        "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.plotly.v1+json": {
       "config": {
        "plotlyServerURL": "https://plotly.com"
       },
       "data": [
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>",
         "legendgroup": "LR",
         "marker": {
          "color": "#636efa"
         },
         "name": "LR",
         "notched": false,
         "offsetgroup": "LR",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "LR",
          "LR",
          "LR",
          "LR",
          "LR",
          "LR",
          "LR",
          "LR",
          "LR",
          "LR"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.6119814885577208,
          0.611556914193521,
          0.6106653080287012,
          0.6049167798913043,
          0.6081012228260869,
          0.607735744915722,
          0.6106653080287012,
          0.6084150638984418,
          0.6092900815217391,
          0.6103515625
         ],
         "y0": " ",
         "yaxis": "y"
        },
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>",
         "legendgroup": "Tree",
         "marker": {
          "color": "#EF553B"
         },
         "name": "Tree",
         "notched": false,
         "offsetgroup": "Tree",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "Tree",
          "Tree",
          "Tree",
          "Tree",
          "Tree",
          "Tree",
          "Tree",
          "Tree",
          "Tree",
          "Tree"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.4916146563070522,
          0.4945442194200314,
          0.49322803889101174,
          0.4890030570652174,
          0.48806895380434784,
          0.48902475268543283,
          0.48550078546257375,
          0.49683692098671084,
          0.486625339673913,
          0.4971976902173913
         ],
         "y0": " ",
         "yaxis": "y"
        },
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>",
         "legendgroup": "GNB",
         "marker": {
          "color": "#00cc96"
         },
         "name": "GNB",
         "notched": false,
         "offsetgroup": "GNB",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "GNB",
          "GNB",
          "GNB",
          "GNB",
          "GNB",
          "GNB",
          "GNB",
          "GNB",
          "GNB",
          "GNB"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.484609179297754,
          0.47560820277671634,
          0.48932195474037277,
          0.4794921875,
          0.4782608695652174,
          0.48452426442491403,
          0.48834543370271305,
          0.4765847238143761,
          0.48615828804347827,
          0.4782608695652174
         ],
         "y0": " ",
         "yaxis": "y"
        },
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>",
         "legendgroup": "RF",
         "marker": {
          "color": "#ab63fa"
         },
         "name": "RF",
         "notched": false,
         "offsetgroup": "RF",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "RF",
          "RF",
          "RF",
          "RF",
          "RF",
          "RF",
          "RF",
          "RF",
          "RF",
          "RF"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.5514796416592366,
          0.550078546257377,
          0.5525835350061563,
          0.5507387907608695,
          0.5491677989130435,
          0.5534326837345561,
          0.5550885237549357,
          0.5476160149450177,
          0.5524371603260869,
          0.5565981657608695
         ],
         "y0": " ",
         "yaxis": "y"
        },
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>",
         "legendgroup": "AdaBoost",
         "marker": {
          "color": "#FFA15A"
         },
         "name": "AdaBoost",
         "notched": false,
         "offsetgroup": "AdaBoost",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "AdaBoost",
          "AdaBoost",
          "AdaBoost",
          "AdaBoost",
          "AdaBoost",
          "AdaBoost",
          "AdaBoost",
          "AdaBoost",
          "AdaBoost",
          "AdaBoost"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.6188595932577591,
          0.6142741901244003,
          0.6126608075404407,
          0.6126443614130435,
          0.6157438858695652,
          0.6173311255466395,
          0.6155903706534199,
          0.6185199337663991,
          0.6165081521739131,
          0.6146824048913043
         ],
         "y0": " ",
         "yaxis": "y"
        },
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>",
         "legendgroup": "Bagging",
         "marker": {
          "color": "#19d3f3"
         },
         "name": "Bagging",
         "notched": false,
         "offsetgroup": "Bagging",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "Bagging",
          "Bagging",
          "Bagging",
          "Bagging",
          "Bagging",
          "Bagging",
          "Bagging",
          "Bagging",
          "Bagging",
          "Bagging"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.5513098119135567,
          0.5530930242431962,
          0.5519466734598565,
          0.5481912364130435,
          0.5477241847826086,
          0.5484651636734174,
          0.5509701524221967,
          0.5484227062369974,
          0.5534137228260869,
          0.556046195652174
         ],
         "y0": " ",
         "yaxis": "y"
        },
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>",
         "legendgroup": "GBM",
         "marker": {
          "color": "#FF6692"
         },
         "name": "GBM",
         "notched": false,
         "offsetgroup": "GBM",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "GBM",
          "GBM",
          "GBM",
          "GBM",
          "GBM",
          "GBM",
          "GBM",
          "GBM",
          "GBM",
          "GBM"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.6241667728102577,
          0.6256103256485374,
          0.6243790599923577,
          0.6207540760869565,
          0.6208814538043478,
          0.6214919543157984,
          0.6219589861164183,
          0.6225109327898781,
          0.6229619565217391,
          0.625
         ],
         "y0": " ",
         "yaxis": "y"
        },
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>",
         "legendgroup": "XGB",
         "marker": {
          "color": "#B6E880"
         },
         "name": "XGB",
         "notched": false,
         "offsetgroup": "XGB",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "XGB",
          "XGB",
          "XGB",
          "XGB",
          "XGB",
          "XGB",
          "XGB",
          "XGB",
          "XGB",
          "XGB"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.623020422026918,
          0.6234025389546979,
          0.6248036343565575,
          0.6185037364130435,
          0.6203294836956522,
          0.6223411030441982,
          0.6220863584256783,
          0.6212372096972785,
          0.6215183423913043,
          0.6233440896739131
         ],
         "y0": " ",
         "yaxis": "y"
        },
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>",
         "legendgroup": "CatBoost",
         "marker": {
          "color": "#FF97FF"
         },
         "name": "CatBoost",
         "notched": false,
         "offsetgroup": "CatBoost",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "CatBoost",
          "CatBoost",
          "CatBoost",
          "CatBoost",
          "CatBoost",
          "CatBoost",
          "CatBoost",
          "CatBoost",
          "CatBoost",
          "CatBoost"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.6256103256485374,
          0.6246762620472975,
          0.6238271133188978,
          0.6214334239130435,
          0.6205417798913043,
          0.6249310066658175,
          0.6222986456077783,
          0.6229355071540781,
          0.6224099864130435,
          0.6255095108695652
         ],
         "y0": " ",
         "yaxis": "y"
        }
       ],
       "layout": {
        "boxmode": "overlay",
        "legend": {
         "title": {
          "text": "Machine Learning Model"
         },
         "tracegroupgap": 0
        },
        "template": {
         "data": {
          "bar": [
           {
            "error_x": {
             "color": "#2a3f5f"
            },
            "error_y": {
             "color": "#2a3f5f"
            },
            "marker": {
             "line": {
              "color": "#E5ECF6",
              "width": 0.5
             },
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "bar"
           }
          ],
          "barpolar": [
           {
            "marker": {
             "line": {
              "color": "#E5ECF6",
              "width": 0.5
             },
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "barpolar"
           }
          ],
          "carpet": [
           {
            "aaxis": {
             "endlinecolor": "#2a3f5f",
             "gridcolor": "white",
             "linecolor": "white",
             "minorgridcolor": "white",
             "startlinecolor": "#2a3f5f"
            },
            "baxis": {
             "endlinecolor": "#2a3f5f",
             "gridcolor": "white",
             "linecolor": "white",
             "minorgridcolor": "white",
             "startlinecolor": "#2a3f5f"
            },
            "type": "carpet"
           }
          ],
          "choropleth": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "choropleth"
           }
          ],
          "contour": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "contour"
           }
          ],
          "contourcarpet": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "contourcarpet"
           }
          ],
          "heatmap": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "heatmap"
           }
          ],
          "heatmapgl": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "heatmapgl"
           }
          ],
          "histogram": [
           {
            "marker": {
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "histogram"
           }
          ],
          "histogram2d": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "histogram2d"
           }
          ],
          "histogram2dcontour": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "histogram2dcontour"
           }
          ],
          "mesh3d": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "mesh3d"
           }
          ],
          "parcoords": [
           {
            "line": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "parcoords"
           }
          ],
          "pie": [
           {
            "automargin": true,
            "type": "pie"
           }
          ],
          "scatter": [
           {
            "fillpattern": {
             "fillmode": "overlay",
             "size": 10,
             "solidity": 0.2
            },
            "type": "scatter"
           }
          ],
          "scatter3d": [
           {
            "line": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatter3d"
           }
          ],
          "scattercarpet": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattercarpet"
           }
          ],
          "scattergeo": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattergeo"
           }
          ],
          "scattergl": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattergl"
           }
          ],
          "scattermapbox": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattermapbox"
           }
          ],
          "scatterpolar": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterpolar"
           }
          ],
          "scatterpolargl": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterpolargl"
           }
          ],
          "scatterternary": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterternary"
           }
          ],
          "surface": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "surface"
           }
          ],
          "table": [
           {
            "cells": {
             "fill": {
              "color": "#EBF0F8"
             },
             "line": {
              "color": "white"
             }
            },
            "header": {
             "fill": {
              "color": "#C8D4E3"
             },
             "line": {
              "color": "white"
             }
            },
            "type": "table"
           }
          ]
         },
         "layout": {
          "annotationdefaults": {
           "arrowcolor": "#2a3f5f",
           "arrowhead": 0,
           "arrowwidth": 1
          },
          "autotypenumbers": "strict",
          "coloraxis": {
           "colorbar": {
            "outlinewidth": 0,
            "ticks": ""
           }
          },
          "colorscale": {
           "diverging": [
            [
             0,
             "#8e0152"
            ],
            [
             0.1,
             "#c51b7d"
            ],
            [
             0.2,
             "#de77ae"
            ],
            [
             0.3,
             "#f1b6da"
            ],
            [
             0.4,
             "#fde0ef"
            ],
            [
             0.5,
             "#f7f7f7"
            ],
            [
             0.6,
             "#e6f5d0"
            ],
            [
             0.7,
             "#b8e186"
            ],
            [
             0.8,
             "#7fbc41"
            ],
            [
             0.9,
             "#4d9221"
            ],
            [
             1,
             "#276419"
            ]
           ],
           "sequential": [
            [
             0,
             "#0d0887"
            ],
            [
             0.1111111111111111,
             "#46039f"
            ],
            [
             0.2222222222222222,
             "#7201a8"
            ],
            [
             0.3333333333333333,
             "#9c179e"
            ],
            [
             0.4444444444444444,
             "#bd3786"
            ],
            [
             0.5555555555555556,
             "#d8576b"
            ],
            [
             0.6666666666666666,
             "#ed7953"
            ],
            [
             0.7777777777777778,
             "#fb9f3a"
            ],
            [
             0.8888888888888888,
             "#fdca26"
            ],
            [
             1,
             "#f0f921"
            ]
           ],
           "sequentialminus": [
            [
             0,
             "#0d0887"
            ],
            [
             0.1111111111111111,
             "#46039f"
            ],
            [
             0.2222222222222222,
             "#7201a8"
            ],
            [
             0.3333333333333333,
             "#9c179e"
            ],
            [
             0.4444444444444444,
             "#bd3786"
            ],
            [
             0.5555555555555556,
             "#d8576b"
            ],
            [
             0.6666666666666666,
             "#ed7953"
            ],
            [
             0.7777777777777778,
             "#fb9f3a"
            ],
            [
             0.8888888888888888,
             "#fdca26"
            ],
            [
             1,
             "#f0f921"
            ]
           ]
          },
          "colorway": [
           "#636efa",
           "#EF553B",
           "#00cc96",
           "#ab63fa",
           "#FFA15A",
           "#19d3f3",
           "#FF6692",
           "#B6E880",
           "#FF97FF",
           "#FECB52"
          ],
          "font": {
           "color": "#2a3f5f"
          },
          "geo": {
           "bgcolor": "white",
           "lakecolor": "white",
           "landcolor": "#E5ECF6",
           "showlakes": true,
           "showland": true,
           "subunitcolor": "white"
          },
          "hoverlabel": {
           "align": "left"
          },
          "hovermode": "closest",
          "mapbox": {
           "style": "light"
          },
          "paper_bgcolor": "white",
          "plot_bgcolor": "#E5ECF6",
          "polar": {
           "angularaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "bgcolor": "#E5ECF6",
           "radialaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           }
          },
          "scene": {
           "xaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           },
           "yaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           },
           "zaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           }
          },
          "shapedefaults": {
           "line": {
            "color": "#2a3f5f"
           }
          },
          "ternary": {
           "aaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "baxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "bgcolor": "#E5ECF6",
           "caxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           }
          },
          "title": {
           "x": 0.05
          },
          "xaxis": {
           "automargin": true,
           "gridcolor": "white",
           "linecolor": "white",
           "ticks": "",
           "title": {
            "standoff": 15
           },
           "zerolinecolor": "white",
           "zerolinewidth": 2
          },
          "yaxis": {
           "automargin": true,
           "gridcolor": "white",
           "linecolor": "white",
           "ticks": "",
           "title": {
            "standoff": 15
           },
           "zerolinecolor": "white",
           "zerolinewidth": 2
          }
         }
        },
        "title": {
         "text": "Model Accuracy Performance"
        },
        "xaxis": {
         "anchor": "y",
         "categoryarray": [
          "LR",
          "Tree",
          "GNB",
          "RF",
          "AdaBoost",
          "Bagging",
          "GBM",
          "XGB",
          "CatBoost"
         ],
         "categoryorder": "array",
         "domain": [
          0,
          1
         ],
         "title": {
          "text": "Machine Learning Model"
         }
        },
        "yaxis": {
         "anchor": "x",
         "domain": [
          0,
          1
         ],
         "title": {
          "text": "Accuracy"
         }
        }
       }
      },
      "text/html": [
       "<div>                            <div id=\"e9e022f4-3f39-4a9c-ae84-c1ee3056f285\" class=\"plotly-graph-div\" style=\"height:525px; width:100%;\"></div>            <script type=\"text/javascript\">                require([\"plotly\"], function(Plotly) {                    window.PLOTLYENV=window.PLOTLYENV || {};                                    if (document.getElementById(\"e9e022f4-3f39-4a9c-ae84-c1ee3056f285\")) {                    Plotly.newPlot(                        \"e9e022f4-3f39-4a9c-ae84-c1ee3056f285\",                        [{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>\",\"legendgroup\":\"LR\",\"marker\":{\"color\":\"#636efa\"},\"name\":\"LR\",\"notched\":false,\"offsetgroup\":\"LR\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"LR\",\"LR\",\"LR\",\"LR\",\"LR\",\"LR\",\"LR\",\"LR\",\"LR\",\"LR\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.6119814885577208,0.611556914193521,0.6106653080287012,0.6049167798913043,0.6081012228260869,0.607735744915722,0.6106653080287012,0.6084150638984418,0.6092900815217391,0.6103515625],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"},{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>\",\"legendgroup\":\"Tree\",\"marker\":{\"color\":\"#EF553B\"},\"name\":\"Tree\",\"notched\":false,\"offsetgroup\":\"Tree\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"Tree\",\"Tree\",\"Tree\",\"Tree\",\"Tree\",\"Tree\",\"Tree\",\"Tree\",\"Tree\",\"Tree\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.4916146563070522,0.4945442194200314,0.49322803889101174,0.4890030570652174,0.48806895380434784,0.48902475268543283,0.48550078546257375,0.49683692098671084,0.486625339673913,0.4971976902173913],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"},{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>\",\"legendgroup\":\"GNB\",\"marker\":{\"color\":\"#00cc96\"},\"name\":\"GNB\",\"notched\":false,\"offsetgroup\":\"GNB\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"GNB\",\"GNB\",\"GNB\",\"GNB\",\"GNB\",\"GNB\",\"GNB\",\"GNB\",\"GNB\",\"GNB\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.484609179297754,0.47560820277671634,0.48932195474037277,0.4794921875,0.4782608695652174,0.48452426442491403,0.48834543370271305,0.4765847238143761,0.48615828804347827,0.4782608695652174],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"},{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>\",\"legendgroup\":\"RF\",\"marker\":{\"color\":\"#ab63fa\"},\"name\":\"RF\",\"notched\":false,\"offsetgroup\":\"RF\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"RF\",\"RF\",\"RF\",\"RF\",\"RF\",\"RF\",\"RF\",\"RF\",\"RF\",\"RF\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.5514796416592366,0.550078546257377,0.5525835350061563,0.5507387907608695,0.5491677989130435,0.5534326837345561,0.5550885237549357,0.5476160149450177,0.5524371603260869,0.5565981657608695],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"},{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>\",\"legendgroup\":\"AdaBoost\",\"marker\":{\"color\":\"#FFA15A\"},\"name\":\"AdaBoost\",\"notched\":false,\"offsetgroup\":\"AdaBoost\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"AdaBoost\",\"AdaBoost\",\"AdaBoost\",\"AdaBoost\",\"AdaBoost\",\"AdaBoost\",\"AdaBoost\",\"AdaBoost\",\"AdaBoost\",\"AdaBoost\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.6188595932577591,0.6142741901244003,0.6126608075404407,0.6126443614130435,0.6157438858695652,0.6173311255466395,0.6155903706534199,0.6185199337663991,0.6165081521739131,0.6146824048913043],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"},{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>\",\"legendgroup\":\"Bagging\",\"marker\":{\"color\":\"#19d3f3\"},\"name\":\"Bagging\",\"notched\":false,\"offsetgroup\":\"Bagging\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"Bagging\",\"Bagging\",\"Bagging\",\"Bagging\",\"Bagging\",\"Bagging\",\"Bagging\",\"Bagging\",\"Bagging\",\"Bagging\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.5513098119135567,0.5530930242431962,0.5519466734598565,0.5481912364130435,0.5477241847826086,0.5484651636734174,0.5509701524221967,0.5484227062369974,0.5534137228260869,0.556046195652174],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"},{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>\",\"legendgroup\":\"GBM\",\"marker\":{\"color\":\"#FF6692\"},\"name\":\"GBM\",\"notched\":false,\"offsetgroup\":\"GBM\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"GBM\",\"GBM\",\"GBM\",\"GBM\",\"GBM\",\"GBM\",\"GBM\",\"GBM\",\"GBM\",\"GBM\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.6241667728102577,0.6256103256485374,0.6243790599923577,0.6207540760869565,0.6208814538043478,0.6214919543157984,0.6219589861164183,0.6225109327898781,0.6229619565217391,0.625],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"},{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>\",\"legendgroup\":\"XGB\",\"marker\":{\"color\":\"#B6E880\"},\"name\":\"XGB\",\"notched\":false,\"offsetgroup\":\"XGB\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"XGB\",\"XGB\",\"XGB\",\"XGB\",\"XGB\",\"XGB\",\"XGB\",\"XGB\",\"XGB\",\"XGB\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.623020422026918,0.6234025389546979,0.6248036343565575,0.6185037364130435,0.6203294836956522,0.6223411030441982,0.6220863584256783,0.6212372096972785,0.6215183423913043,0.6233440896739131],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"},{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>\",\"legendgroup\":\"CatBoost\",\"marker\":{\"color\":\"#FF97FF\"},\"name\":\"CatBoost\",\"notched\":false,\"offsetgroup\":\"CatBoost\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"CatBoost\",\"CatBoost\",\"CatBoost\",\"CatBoost\",\"CatBoost\",\"CatBoost\",\"CatBoost\",\"CatBoost\",\"CatBoost\",\"CatBoost\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.6256103256485374,0.6246762620472975,0.6238271133188978,0.6214334239130435,0.6205417798913043,0.6249310066658175,0.6222986456077783,0.6229355071540781,0.6224099864130435,0.6255095108695652],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"}],                        {\"template\":{\"data\":{\"histogram2dcontour\":[{\"type\":\"histogram2dcontour\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"choropleth\":[{\"type\":\"choropleth\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}],\"histogram2d\":[{\"type\":\"histogram2d\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"heatmap\":[{\"type\":\"heatmap\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"heatmapgl\":[{\"type\":\"heatmapgl\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"contourcarpet\":[{\"type\":\"contourcarpet\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}],\"contour\":[{\"type\":\"contour\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"surface\":[{\"type\":\"surface\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"mesh3d\":[{\"type\":\"mesh3d\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}],\"scatter\":[{\"fillpattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2},\"type\":\"scatter\"}],\"parcoords\":[{\"type\":\"parcoords\",\"line\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scatterpolargl\":[{\"type\":\"scatterpolargl\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"bar\":[{\"error_x\":{\"color\":\"#2a3f5f\"},\"error_y\":{\"color\":\"#2a3f5f\"},\"marker\":{\"line\":{\"color\":\"#E5ECF6\",\"width\":0.5},\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"bar\"}],\"scattergeo\":[{\"type\":\"scattergeo\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scatterpolar\":[{\"type\":\"scatterpolar\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"histogram\":[{\"marker\":{\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"histogram\"}],\"scattergl\":[{\"type\":\"scattergl\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scatter3d\":[{\"type\":\"scatter3d\",\"line\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scattermapbox\":[{\"type\":\"scattermapbox\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scatterternary\":[{\"type\":\"scatterternary\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scattercarpet\":[{\"type\":\"scattercarpet\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"carpet\":[{\"aaxis\":{\"endlinecolor\":\"#2a3f5f\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"minorgridcolor\":\"white\",\"startlinecolor\":\"#2a3f5f\"},\"baxis\":{\"endlinecolor\":\"#2a3f5f\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"minorgridcolor\":\"white\",\"startlinecolor\":\"#2a3f5f\"},\"type\":\"carpet\"}],\"table\":[{\"cells\":{\"fill\":{\"color\":\"#EBF0F8\"},\"line\":{\"color\":\"white\"}},\"header\":{\"fill\":{\"color\":\"#C8D4E3\"},\"line\":{\"color\":\"white\"}},\"type\":\"table\"}],\"barpolar\":[{\"marker\":{\"line\":{\"color\":\"#E5ECF6\",\"width\":0.5},\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"barpolar\"}],\"pie\":[{\"automargin\":true,\"type\":\"pie\"}]},\"layout\":{\"autotypenumbers\":\"strict\",\"colorway\":[\"#636efa\",\"#EF553B\",\"#00cc96\",\"#ab63fa\",\"#FFA15A\",\"#19d3f3\",\"#FF6692\",\"#B6E880\",\"#FF97FF\",\"#FECB52\"],\"font\":{\"color\":\"#2a3f5f\"},\"hovermode\":\"closest\",\"hoverlabel\":{\"align\":\"left\"},\"paper_bgcolor\":\"white\",\"plot_bgcolor\":\"#E5ECF6\",\"polar\":{\"bgcolor\":\"#E5ECF6\",\"angularaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"radialaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"}},\"ternary\":{\"bgcolor\":\"#E5ECF6\",\"aaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"baxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"caxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"}},\"coloraxis\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"colorscale\":{\"sequential\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"sequentialminus\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"diverging\":[[0,\"#8e0152\"],[0.1,\"#c51b7d\"],[0.2,\"#de77ae\"],[0.3,\"#f1b6da\"],[0.4,\"#fde0ef\"],[0.5,\"#f7f7f7\"],[0.6,\"#e6f5d0\"],[0.7,\"#b8e186\"],[0.8,\"#7fbc41\"],[0.9,\"#4d9221\"],[1,\"#276419\"]]},\"xaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\",\"title\":{\"standoff\":15},\"zerolinecolor\":\"white\",\"automargin\":true,\"zerolinewidth\":2},\"yaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\",\"title\":{\"standoff\":15},\"zerolinecolor\":\"white\",\"automargin\":true,\"zerolinewidth\":2},\"scene\":{\"xaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\",\"gridwidth\":2},\"yaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\",\"gridwidth\":2},\"zaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\",\"gridwidth\":2}},\"shapedefaults\":{\"line\":{\"color\":\"#2a3f5f\"}},\"annotationdefaults\":{\"arrowcolor\":\"#2a3f5f\",\"arrowhead\":0,\"arrowwidth\":1},\"geo\":{\"bgcolor\":\"white\",\"landcolor\":\"#E5ECF6\",\"subunitcolor\":\"white\",\"showland\":true,\"showlakes\":true,\"lakecolor\":\"white\"},\"title\":{\"x\":0.05},\"mapbox\":{\"style\":\"light\"}}},\"xaxis\":{\"anchor\":\"y\",\"domain\":[0.0,1.0],\"title\":{\"text\":\"Machine Learning Model\"},\"categoryorder\":\"array\",\"categoryarray\":[\"LR\",\"Tree\",\"GNB\",\"RF\",\"AdaBoost\",\"Bagging\",\"GBM\",\"XGB\",\"CatBoost\"]},\"yaxis\":{\"anchor\":\"x\",\"domain\":[0.0,1.0],\"title\":{\"text\":\"Accuracy\"}},\"legend\":{\"title\":{\"text\":\"Machine Learning Model\"},\"tracegroupgap\":0},\"title\":{\"text\":\"Model Accuracy Performance\"},\"boxmode\":\"overlay\"},                        {\"responsive\": true}                    ).then(function(){\n",
       "                            \n",
       "var gd = document.getElementById('e9e022f4-3f39-4a9c-ae84-c1ee3056f285');\n",
       "var x = new MutationObserver(function (mutations, observer) {{\n",
       "        var display = window.getComputedStyle(gd).display;\n",
       "        if (!display || display === 'none') {{\n",
       "            console.log([gd, 'removed!']);\n",
       "            Plotly.purge(gd);\n",
       "            observer.disconnect();\n",
       "        }}\n",
       "}});\n",
       "\n",
       "// Listen for the removal of the full notebook cells\n",
       "var notebookContainer = gd.closest('#notebook-container');\n",
       "if (notebookContainer) {{\n",
       "    x.observe(notebookContainer, {childList: true});\n",
       "}}\n",
       "\n",
       "// Listen for the clearing of the current output cell\n",
       "var outputEl = gd.closest('.output');\n",
       "if (outputEl) {{\n",
       "    x.observe(outputEl, {childList: true});\n",
       "}}\n",
       "\n",
       "                        })                };                });            </script>        </div>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = px.box(\n",
    "    classifmod,\n",
    "    x=\"Model\",\n",
    "    y=\"Accuracy\",\n",
    "    color=\"Model\",\n",
    "    points=\"all\",\n",
    "    labels={\"Model\": \"Machine Learning Model\", \"Accuracy\": \"Accuracy\"},\n",
    "    title=\"Model Accuracy Performance\",\n",
    ")\n",
    "fig.show()\n",
    "fig.write_html(\"products/plots/model_train_accuracies.html\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 224,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.plotly.v1+json": {
       "config": {
        "plotlyServerURL": "https://plotly.com"
       },
       "data": [
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>",
         "legendgroup": "LR",
         "marker": {
          "color": "#636efa"
         },
         "name": "LR",
         "notched": false,
         "offsetgroup": "LR",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "LR",
          "LR",
          "LR",
          "LR",
          "LR",
          "LR",
          "LR",
          "LR",
          "LR",
          "LR"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.5330310166081167,
          0.5340882197562198,
          0.5338695590410422,
          0.5274160831589275,
          0.5306087902099951,
          0.5291600956866274,
          0.5332928009882842,
          0.5294968285334225,
          0.5328638847935039,
          0.5336852199564559
         ],
         "y0": " ",
         "yaxis": "y"
        },
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>",
         "legendgroup": "Tree",
         "marker": {
          "color": "#EF553B"
         },
         "name": "Tree",
         "notched": false,
         "offsetgroup": "Tree",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "Tree",
          "Tree",
          "Tree",
          "Tree",
          "Tree",
          "Tree",
          "Tree",
          "Tree",
          "Tree",
          "Tree"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.493658300161518,
          0.4967100917494991,
          0.49747685286945986,
          0.49499048944419577,
          0.4906545513949378,
          0.49164876791169476,
          0.49116551054719426,
          0.4981853565600421,
          0.49303806360879276,
          0.4999920492342203
         ],
         "y0": " ",
         "yaxis": "y"
        },
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>",
         "legendgroup": "GNB",
         "marker": {
          "color": "#00cc96"
         },
         "name": "GNB",
         "notched": false,
         "offsetgroup": "GNB",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "GNB",
          "GNB",
          "GNB",
          "GNB",
          "GNB",
          "GNB",
          "GNB",
          "GNB",
          "GNB",
          "GNB"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.49379418779576273,
          0.487913235018719,
          0.49830703430324885,
          0.48940554125684693,
          0.49077424505095935,
          0.4945232592393809,
          0.49833328221614365,
          0.4881203881014537,
          0.49442650087623163,
          0.4889301777080842
         ],
         "y0": " ",
         "yaxis": "y"
        },
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>",
         "legendgroup": "RF",
         "marker": {
          "color": "#ab63fa"
         },
         "name": "RF",
         "notched": false,
         "offsetgroup": "RF",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "RF",
          "RF",
          "RF",
          "RF",
          "RF",
          "RF",
          "RF",
          "RF",
          "RF",
          "RF"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.5278134518776496,
          0.5240219439259232,
          0.5280772187378432,
          0.5235097634157555,
          0.526481888527823,
          0.5287869842836027,
          0.5303750224496052,
          0.5264340593437248,
          0.5280046741238023,
          0.5326669351627846
         ],
         "y0": " ",
         "yaxis": "y"
        },
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>",
         "legendgroup": "AdaBoost",
         "marker": {
          "color": "#FFA15A"
         },
         "name": "AdaBoost",
         "notched": false,
         "offsetgroup": "AdaBoost",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "AdaBoost",
          "AdaBoost",
          "AdaBoost",
          "AdaBoost",
          "AdaBoost",
          "AdaBoost",
          "AdaBoost",
          "AdaBoost",
          "AdaBoost",
          "AdaBoost"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.5481068169657054,
          0.5457190392447224,
          0.5459703397540799,
          0.5413899875714211,
          0.5430214423881895,
          0.5400241617042195,
          0.5421946797712927,
          0.5552080140264257,
          0.540077433147427,
          0.548102398843852
         ],
         "y0": " ",
         "yaxis": "y"
        },
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>",
         "legendgroup": "Bagging",
         "marker": {
          "color": "#19d3f3"
         },
         "name": "Bagging",
         "notched": false,
         "offsetgroup": "Bagging",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "Bagging",
          "Bagging",
          "Bagging",
          "Bagging",
          "Bagging",
          "Bagging",
          "Bagging",
          "Bagging",
          "Bagging",
          "Bagging"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.5340461828217409,
          0.5283280264273261,
          0.5345599688860541,
          0.5292981693515432,
          0.5310945203005827,
          0.5343815210393914,
          0.532379275194792,
          0.5332682748134006,
          0.5337666226467009,
          0.5348626190086311
         ],
         "y0": " ",
         "yaxis": "y"
        },
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>",
         "legendgroup": "GBM",
         "marker": {
          "color": "#FF6692"
         },
         "name": "GBM",
         "notched": false,
         "offsetgroup": "GBM",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "GBM",
          "GBM",
          "GBM",
          "GBM",
          "GBM",
          "GBM",
          "GBM",
          "GBM",
          "GBM",
          "GBM"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.5451857338744811,
          0.5481054122322698,
          0.5474699384034903,
          0.5413191647969496,
          0.542751005100391,
          0.5416005218415182,
          0.5446832634024635,
          0.5438576786602591,
          0.5464732233809481,
          0.5470786274651642
         ],
         "y0": " ",
         "yaxis": "y"
        },
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>",
         "legendgroup": "XGB",
         "marker": {
          "color": "#B6E880"
         },
         "name": "XGB",
         "notched": false,
         "offsetgroup": "XGB",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "XGB",
          "XGB",
          "XGB",
          "XGB",
          "XGB",
          "XGB",
          "XGB",
          "XGB",
          "XGB",
          "XGB"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.5529375620798355,
          0.555619158490195,
          0.556785909186755,
          0.548677136354181,
          0.5508845441493477,
          0.5513959466851337,
          0.5520008378066783,
          0.5516000609909352,
          0.5538252718608446,
          0.5570678683695651
         ],
         "y0": " ",
         "yaxis": "y"
        },
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>",
         "legendgroup": "CatBoost",
         "marker": {
          "color": "#FF97FF"
         },
         "name": "CatBoost",
         "notched": false,
         "offsetgroup": "CatBoost",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "CatBoost",
          "CatBoost",
          "CatBoost",
          "CatBoost",
          "CatBoost",
          "CatBoost",
          "CatBoost",
          "CatBoost",
          "CatBoost",
          "CatBoost"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.5577635354636955,
          0.5577262536621505,
          0.555347590739337,
          0.5529293754142719,
          0.5526923021270655,
          0.5558712496332088,
          0.5538723821292479,
          0.5545615314741104,
          0.556522456969849,
          0.5589616068335584
         ],
         "y0": " ",
         "yaxis": "y"
        }
       ],
       "layout": {
        "boxmode": "overlay",
        "legend": {
         "title": {
          "text": "Machine Learning Model"
         },
         "tracegroupgap": 0
        },
        "template": {
         "data": {
          "bar": [
           {
            "error_x": {
             "color": "#2a3f5f"
            },
            "error_y": {
             "color": "#2a3f5f"
            },
            "marker": {
             "line": {
              "color": "#E5ECF6",
              "width": 0.5
             },
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "bar"
           }
          ],
          "barpolar": [
           {
            "marker": {
             "line": {
              "color": "#E5ECF6",
              "width": 0.5
             },
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "barpolar"
           }
          ],
          "carpet": [
           {
            "aaxis": {
             "endlinecolor": "#2a3f5f",
             "gridcolor": "white",
             "linecolor": "white",
             "minorgridcolor": "white",
             "startlinecolor": "#2a3f5f"
            },
            "baxis": {
             "endlinecolor": "#2a3f5f",
             "gridcolor": "white",
             "linecolor": "white",
             "minorgridcolor": "white",
             "startlinecolor": "#2a3f5f"
            },
            "type": "carpet"
           }
          ],
          "choropleth": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "choropleth"
           }
          ],
          "contour": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "contour"
           }
          ],
          "contourcarpet": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "contourcarpet"
           }
          ],
          "heatmap": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "heatmap"
           }
          ],
          "heatmapgl": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "heatmapgl"
           }
          ],
          "histogram": [
           {
            "marker": {
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "histogram"
           }
          ],
          "histogram2d": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "histogram2d"
           }
          ],
          "histogram2dcontour": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "histogram2dcontour"
           }
          ],
          "mesh3d": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "mesh3d"
           }
          ],
          "parcoords": [
           {
            "line": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "parcoords"
           }
          ],
          "pie": [
           {
            "automargin": true,
            "type": "pie"
           }
          ],
          "scatter": [
           {
            "fillpattern": {
             "fillmode": "overlay",
             "size": 10,
             "solidity": 0.2
            },
            "type": "scatter"
           }
          ],
          "scatter3d": [
           {
            "line": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatter3d"
           }
          ],
          "scattercarpet": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattercarpet"
           }
          ],
          "scattergeo": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattergeo"
           }
          ],
          "scattergl": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattergl"
           }
          ],
          "scattermapbox": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattermapbox"
           }
          ],
          "scatterpolar": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterpolar"
           }
          ],
          "scatterpolargl": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterpolargl"
           }
          ],
          "scatterternary": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterternary"
           }
          ],
          "surface": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "surface"
           }
          ],
          "table": [
           {
            "cells": {
             "fill": {
              "color": "#EBF0F8"
             },
             "line": {
              "color": "white"
             }
            },
            "header": {
             "fill": {
              "color": "#C8D4E3"
             },
             "line": {
              "color": "white"
             }
            },
            "type": "table"
           }
          ]
         },
         "layout": {
          "annotationdefaults": {
           "arrowcolor": "#2a3f5f",
           "arrowhead": 0,
           "arrowwidth": 1
          },
          "autotypenumbers": "strict",
          "coloraxis": {
           "colorbar": {
            "outlinewidth": 0,
            "ticks": ""
           }
          },
          "colorscale": {
           "diverging": [
            [
             0,
             "#8e0152"
            ],
            [
             0.1,
             "#c51b7d"
            ],
            [
             0.2,
             "#de77ae"
            ],
            [
             0.3,
             "#f1b6da"
            ],
            [
             0.4,
             "#fde0ef"
            ],
            [
             0.5,
             "#f7f7f7"
            ],
            [
             0.6,
             "#e6f5d0"
            ],
            [
             0.7,
             "#b8e186"
            ],
            [
             0.8,
             "#7fbc41"
            ],
            [
             0.9,
             "#4d9221"
            ],
            [
             1,
             "#276419"
            ]
           ],
           "sequential": [
            [
             0,
             "#0d0887"
            ],
            [
             0.1111111111111111,
             "#46039f"
            ],
            [
             0.2222222222222222,
             "#7201a8"
            ],
            [
             0.3333333333333333,
             "#9c179e"
            ],
            [
             0.4444444444444444,
             "#bd3786"
            ],
            [
             0.5555555555555556,
             "#d8576b"
            ],
            [
             0.6666666666666666,
             "#ed7953"
            ],
            [
             0.7777777777777778,
             "#fb9f3a"
            ],
            [
             0.8888888888888888,
             "#fdca26"
            ],
            [
             1,
             "#f0f921"
            ]
           ],
           "sequentialminus": [
            [
             0,
             "#0d0887"
            ],
            [
             0.1111111111111111,
             "#46039f"
            ],
            [
             0.2222222222222222,
             "#7201a8"
            ],
            [
             0.3333333333333333,
             "#9c179e"
            ],
            [
             0.4444444444444444,
             "#bd3786"
            ],
            [
             0.5555555555555556,
             "#d8576b"
            ],
            [
             0.6666666666666666,
             "#ed7953"
            ],
            [
             0.7777777777777778,
             "#fb9f3a"
            ],
            [
             0.8888888888888888,
             "#fdca26"
            ],
            [
             1,
             "#f0f921"
            ]
           ]
          },
          "colorway": [
           "#636efa",
           "#EF553B",
           "#00cc96",
           "#ab63fa",
           "#FFA15A",
           "#19d3f3",
           "#FF6692",
           "#B6E880",
           "#FF97FF",
           "#FECB52"
          ],
          "font": {
           "color": "#2a3f5f"
          },
          "geo": {
           "bgcolor": "white",
           "lakecolor": "white",
           "landcolor": "#E5ECF6",
           "showlakes": true,
           "showland": true,
           "subunitcolor": "white"
          },
          "hoverlabel": {
           "align": "left"
          },
          "hovermode": "closest",
          "mapbox": {
           "style": "light"
          },
          "paper_bgcolor": "white",
          "plot_bgcolor": "#E5ECF6",
          "polar": {
           "angularaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "bgcolor": "#E5ECF6",
           "radialaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           }
          },
          "scene": {
           "xaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           },
           "yaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           },
           "zaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           }
          },
          "shapedefaults": {
           "line": {
            "color": "#2a3f5f"
           }
          },
          "ternary": {
           "aaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "baxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "bgcolor": "#E5ECF6",
           "caxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           }
          },
          "title": {
           "x": 0.05
          },
          "xaxis": {
           "automargin": true,
           "gridcolor": "white",
           "linecolor": "white",
           "ticks": "",
           "title": {
            "standoff": 15
           },
           "zerolinecolor": "white",
           "zerolinewidth": 2
          },
          "yaxis": {
           "automargin": true,
           "gridcolor": "white",
           "linecolor": "white",
           "ticks": "",
           "title": {
            "standoff": 15
           },
           "zerolinecolor": "white",
           "zerolinewidth": 2
          }
         }
        },
        "title": {
         "text": "Model Weighted f1-Score Performance"
        },
        "xaxis": {
         "anchor": "y",
         "categoryarray": [
          "LR",
          "Tree",
          "GNB",
          "RF",
          "AdaBoost",
          "Bagging",
          "GBM",
          "XGB",
          "CatBoost"
         ],
         "categoryorder": "array",
         "domain": [
          0,
          1
         ],
         "title": {
          "text": "Machine Learning Model"
         }
        },
        "yaxis": {
         "anchor": "x",
         "domain": [
          0,
          1
         ],
         "title": {
          "text": "Weighted f1-Score"
         }
        }
       }
      },
      "text/html": [
       "<div>                            <div id=\"0666bb50-7349-4879-89dd-ba8717382bc3\" class=\"plotly-graph-div\" style=\"height:525px; width:100%;\"></div>            <script type=\"text/javascript\">                require([\"plotly\"], function(Plotly) {                    window.PLOTLYENV=window.PLOTLYENV || {};                                    if (document.getElementById(\"0666bb50-7349-4879-89dd-ba8717382bc3\")) {                    Plotly.newPlot(                        \"0666bb50-7349-4879-89dd-ba8717382bc3\",                        [{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>\",\"legendgroup\":\"LR\",\"marker\":{\"color\":\"#636efa\"},\"name\":\"LR\",\"notched\":false,\"offsetgroup\":\"LR\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"LR\",\"LR\",\"LR\",\"LR\",\"LR\",\"LR\",\"LR\",\"LR\",\"LR\",\"LR\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.5330310166081167,0.5340882197562198,0.5338695590410422,0.5274160831589275,0.5306087902099951,0.5291600956866274,0.5332928009882842,0.5294968285334225,0.5328638847935039,0.5336852199564559],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"},{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>\",\"legendgroup\":\"Tree\",\"marker\":{\"color\":\"#EF553B\"},\"name\":\"Tree\",\"notched\":false,\"offsetgroup\":\"Tree\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"Tree\",\"Tree\",\"Tree\",\"Tree\",\"Tree\",\"Tree\",\"Tree\",\"Tree\",\"Tree\",\"Tree\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.493658300161518,0.4967100917494991,0.49747685286945986,0.49499048944419577,0.4906545513949378,0.49164876791169476,0.49116551054719426,0.4981853565600421,0.49303806360879276,0.4999920492342203],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"},{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>\",\"legendgroup\":\"GNB\",\"marker\":{\"color\":\"#00cc96\"},\"name\":\"GNB\",\"notched\":false,\"offsetgroup\":\"GNB\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"GNB\",\"GNB\",\"GNB\",\"GNB\",\"GNB\",\"GNB\",\"GNB\",\"GNB\",\"GNB\",\"GNB\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.49379418779576273,0.487913235018719,0.49830703430324885,0.48940554125684693,0.49077424505095935,0.4945232592393809,0.49833328221614365,0.4881203881014537,0.49442650087623163,0.4889301777080842],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"},{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>\",\"legendgroup\":\"RF\",\"marker\":{\"color\":\"#ab63fa\"},\"name\":\"RF\",\"notched\":false,\"offsetgroup\":\"RF\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"RF\",\"RF\",\"RF\",\"RF\",\"RF\",\"RF\",\"RF\",\"RF\",\"RF\",\"RF\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.5278134518776496,0.5240219439259232,0.5280772187378432,0.5235097634157555,0.526481888527823,0.5287869842836027,0.5303750224496052,0.5264340593437248,0.5280046741238023,0.5326669351627846],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"},{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>\",\"legendgroup\":\"AdaBoost\",\"marker\":{\"color\":\"#FFA15A\"},\"name\":\"AdaBoost\",\"notched\":false,\"offsetgroup\":\"AdaBoost\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"AdaBoost\",\"AdaBoost\",\"AdaBoost\",\"AdaBoost\",\"AdaBoost\",\"AdaBoost\",\"AdaBoost\",\"AdaBoost\",\"AdaBoost\",\"AdaBoost\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.5481068169657054,0.5457190392447224,0.5459703397540799,0.5413899875714211,0.5430214423881895,0.5400241617042195,0.5421946797712927,0.5552080140264257,0.540077433147427,0.548102398843852],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"},{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>\",\"legendgroup\":\"Bagging\",\"marker\":{\"color\":\"#19d3f3\"},\"name\":\"Bagging\",\"notched\":false,\"offsetgroup\":\"Bagging\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"Bagging\",\"Bagging\",\"Bagging\",\"Bagging\",\"Bagging\",\"Bagging\",\"Bagging\",\"Bagging\",\"Bagging\",\"Bagging\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.5340461828217409,0.5283280264273261,0.5345599688860541,0.5292981693515432,0.5310945203005827,0.5343815210393914,0.532379275194792,0.5332682748134006,0.5337666226467009,0.5348626190086311],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"},{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>\",\"legendgroup\":\"GBM\",\"marker\":{\"color\":\"#FF6692\"},\"name\":\"GBM\",\"notched\":false,\"offsetgroup\":\"GBM\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"GBM\",\"GBM\",\"GBM\",\"GBM\",\"GBM\",\"GBM\",\"GBM\",\"GBM\",\"GBM\",\"GBM\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.5451857338744811,0.5481054122322698,0.5474699384034903,0.5413191647969496,0.542751005100391,0.5416005218415182,0.5446832634024635,0.5438576786602591,0.5464732233809481,0.5470786274651642],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"},{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>\",\"legendgroup\":\"XGB\",\"marker\":{\"color\":\"#B6E880\"},\"name\":\"XGB\",\"notched\":false,\"offsetgroup\":\"XGB\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"XGB\",\"XGB\",\"XGB\",\"XGB\",\"XGB\",\"XGB\",\"XGB\",\"XGB\",\"XGB\",\"XGB\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.5529375620798355,0.555619158490195,0.556785909186755,0.548677136354181,0.5508845441493477,0.5513959466851337,0.5520008378066783,0.5516000609909352,0.5538252718608446,0.5570678683695651],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"},{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>\",\"legendgroup\":\"CatBoost\",\"marker\":{\"color\":\"#FF97FF\"},\"name\":\"CatBoost\",\"notched\":false,\"offsetgroup\":\"CatBoost\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"CatBoost\",\"CatBoost\",\"CatBoost\",\"CatBoost\",\"CatBoost\",\"CatBoost\",\"CatBoost\",\"CatBoost\",\"CatBoost\",\"CatBoost\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.5577635354636955,0.5577262536621505,0.555347590739337,0.5529293754142719,0.5526923021270655,0.5558712496332088,0.5538723821292479,0.5545615314741104,0.556522456969849,0.5589616068335584],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"}],                        {\"template\":{\"data\":{\"histogram2dcontour\":[{\"type\":\"histogram2dcontour\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"choropleth\":[{\"type\":\"choropleth\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}],\"histogram2d\":[{\"type\":\"histogram2d\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"heatmap\":[{\"type\":\"heatmap\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"heatmapgl\":[{\"type\":\"heatmapgl\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"contourcarpet\":[{\"type\":\"contourcarpet\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}],\"contour\":[{\"type\":\"contour\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"surface\":[{\"type\":\"surface\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"mesh3d\":[{\"type\":\"mesh3d\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}],\"scatter\":[{\"fillpattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2},\"type\":\"scatter\"}],\"parcoords\":[{\"type\":\"parcoords\",\"line\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scatterpolargl\":[{\"type\":\"scatterpolargl\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"bar\":[{\"error_x\":{\"color\":\"#2a3f5f\"},\"error_y\":{\"color\":\"#2a3f5f\"},\"marker\":{\"line\":{\"color\":\"#E5ECF6\",\"width\":0.5},\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"bar\"}],\"scattergeo\":[{\"type\":\"scattergeo\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scatterpolar\":[{\"type\":\"scatterpolar\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"histogram\":[{\"marker\":{\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"histogram\"}],\"scattergl\":[{\"type\":\"scattergl\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scatter3d\":[{\"type\":\"scatter3d\",\"line\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scattermapbox\":[{\"type\":\"scattermapbox\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scatterternary\":[{\"type\":\"scatterternary\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scattercarpet\":[{\"type\":\"scattercarpet\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"carpet\":[{\"aaxis\":{\"endlinecolor\":\"#2a3f5f\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"minorgridcolor\":\"white\",\"startlinecolor\":\"#2a3f5f\"},\"baxis\":{\"endlinecolor\":\"#2a3f5f\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"minorgridcolor\":\"white\",\"startlinecolor\":\"#2a3f5f\"},\"type\":\"carpet\"}],\"table\":[{\"cells\":{\"fill\":{\"color\":\"#EBF0F8\"},\"line\":{\"color\":\"white\"}},\"header\":{\"fill\":{\"color\":\"#C8D4E3\"},\"line\":{\"color\":\"white\"}},\"type\":\"table\"}],\"barpolar\":[{\"marker\":{\"line\":{\"color\":\"#E5ECF6\",\"width\":0.5},\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"barpolar\"}],\"pie\":[{\"automargin\":true,\"type\":\"pie\"}]},\"layout\":{\"autotypenumbers\":\"strict\",\"colorway\":[\"#636efa\",\"#EF553B\",\"#00cc96\",\"#ab63fa\",\"#FFA15A\",\"#19d3f3\",\"#FF6692\",\"#B6E880\",\"#FF97FF\",\"#FECB52\"],\"font\":{\"color\":\"#2a3f5f\"},\"hovermode\":\"closest\",\"hoverlabel\":{\"align\":\"left\"},\"paper_bgcolor\":\"white\",\"plot_bgcolor\":\"#E5ECF6\",\"polar\":{\"bgcolor\":\"#E5ECF6\",\"angularaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"radialaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"}},\"ternary\":{\"bgcolor\":\"#E5ECF6\",\"aaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"baxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"caxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"}},\"coloraxis\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"colorscale\":{\"sequential\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"sequentialminus\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"diverging\":[[0,\"#8e0152\"],[0.1,\"#c51b7d\"],[0.2,\"#de77ae\"],[0.3,\"#f1b6da\"],[0.4,\"#fde0ef\"],[0.5,\"#f7f7f7\"],[0.6,\"#e6f5d0\"],[0.7,\"#b8e186\"],[0.8,\"#7fbc41\"],[0.9,\"#4d9221\"],[1,\"#276419\"]]},\"xaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\",\"title\":{\"standoff\":15},\"zerolinecolor\":\"white\",\"automargin\":true,\"zerolinewidth\":2},\"yaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\",\"title\":{\"standoff\":15},\"zerolinecolor\":\"white\",\"automargin\":true,\"zerolinewidth\":2},\"scene\":{\"xaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\",\"gridwidth\":2},\"yaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\",\"gridwidth\":2},\"zaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\",\"gridwidth\":2}},\"shapedefaults\":{\"line\":{\"color\":\"#2a3f5f\"}},\"annotationdefaults\":{\"arrowcolor\":\"#2a3f5f\",\"arrowhead\":0,\"arrowwidth\":1},\"geo\":{\"bgcolor\":\"white\",\"landcolor\":\"#E5ECF6\",\"subunitcolor\":\"white\",\"showland\":true,\"showlakes\":true,\"lakecolor\":\"white\"},\"title\":{\"x\":0.05},\"mapbox\":{\"style\":\"light\"}}},\"xaxis\":{\"anchor\":\"y\",\"domain\":[0.0,1.0],\"title\":{\"text\":\"Machine Learning Model\"},\"categoryorder\":\"array\",\"categoryarray\":[\"LR\",\"Tree\",\"GNB\",\"RF\",\"AdaBoost\",\"Bagging\",\"GBM\",\"XGB\",\"CatBoost\"]},\"yaxis\":{\"anchor\":\"x\",\"domain\":[0.0,1.0],\"title\":{\"text\":\"Weighted f1-Score\"}},\"legend\":{\"title\":{\"text\":\"Machine Learning Model\"},\"tracegroupgap\":0},\"title\":{\"text\":\"Model Weighted f1-Score Performance\"},\"boxmode\":\"overlay\"},                        {\"responsive\": true}                    ).then(function(){\n",
       "                            \n",
       "var gd = document.getElementById('0666bb50-7349-4879-89dd-ba8717382bc3');\n",
       "var x = new MutationObserver(function (mutations, observer) {{\n",
       "        var display = window.getComputedStyle(gd).display;\n",
       "        if (!display || display === 'none') {{\n",
       "            console.log([gd, 'removed!']);\n",
       "            Plotly.purge(gd);\n",
       "            observer.disconnect();\n",
       "        }}\n",
       "}});\n",
       "\n",
       "// Listen for the removal of the full notebook cells\n",
       "var notebookContainer = gd.closest('#notebook-container');\n",
       "if (notebookContainer) {{\n",
       "    x.observe(notebookContainer, {childList: true});\n",
       "}}\n",
       "\n",
       "// Listen for the clearing of the current output cell\n",
       "var outputEl = gd.closest('.output');\n",
       "if (outputEl) {{\n",
       "    x.observe(outputEl, {childList: true});\n",
       "}}\n",
       "\n",
       "                        })                };                });            </script>        </div>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "results[\"Model\"] = names\n",
    "results[\"Accuracy\"] = accuracies\n",
    "results[\"Weighted F1\"] = weighted_f1s\n",
    "\n",
    "classifmod = pd.DataFrame(results)\n",
    "\n",
    "fig = px.box(\n",
    "    classifmod,\n",
    "    x=\"Model\",\n",
    "    y=\"Weighted_F1\",\n",
    "    color=\"Model\",\n",
    "    points=\"all\",\n",
    "    labels={\"Model\": \"Machine Learning Model\", \"Weighted_F1\": \"Weighted f1-Score\"},\n",
    "    title=\"Model Weighted f1-Score Performance\",\n",
    ")\n",
    "fig.show()\n",
    "fig.write_html(\"products/plots/model_train_weighted_f1.html\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Testing Performance Confusion Matrices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 225,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot confusion matrices\n",
    "for name, matrix in confusion_matrices.items():\n",
    "    plt.figure(figsize=(5, 5))\n",
    "    sns.heatmap(\n",
    "        matrix,\n",
    "        annot=True,\n",
    "        fmt=\"d\",\n",
    "        cmap=\"YlGnBu\",\n",
    "        cbar=False,\n",
    "        xticklabels=[\"Extras\", \"Dot\", \"Runs\", \"Wicket\"],\n",
    "        yticklabels=[\"Extras\", \"Dot\", \"Runs\", \"Wicket\"],\n",
    "    )\n",
    "    plt.xlabel(\"Predicted\")\n",
    "    plt.ylabel(\"Actual\")\n",
    "    plt.title(f\"Confusion Matrix - {name}\")\n",
    "    plt.savefig(f\"products/plots/confusion_matrix_{name}.png\")\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In our multi-class classification project of predicting ball outcomes in cricket, we evaluated several algorithms, including boosting  (Adaboost, XGBoost, Gradient Boosting, and CatBoost), Gaussian Naive Bayes, Logistic Regression, and Decision Trees. Each of these algorithms exhibit unique strengths and weaknesses, and the choice between them involves striking a balance between different aspects of performance that are discussed below.\n",
    "\n",
    "**Accuracy vs. Precision and Recall**:\n",
    "\n",
    "Accuracy, while an essential metric, might not be the sole measure of a model's effectiveness, especially in this imbalanced dataset, where two classes, Extras and Runs, significantly outnumber the others, Dots and Wickets. As can be seen from the confusion matrices above, majority classes dominate the accuracy metric, leading to a misrespresentative view of holistic model performance. In cricket, especially in the scenario of betting, correctly predicting the minority class, Wickets, is of paramount importance.\n",
    "\n",
    "Precision and recall are crucial metrics in such cases. Precision measures the accuracy of positive predictions, while recall gauges the model's ability to find all the relevant cases in a dataset. From the confusion matrices, we observed that Naive Bayes and Decision Tree (DTree) models exhibited better precision and recall for minority classes. This means that when these models predict a wicket or a boundary ball, they are more likely to be correct. In a business or real-world setting, this could translate to better decision-making and assist organizations that are trying to track illegal betting activity.\n",
    "\n",
    "**Assessing Trade-Offs and Choosing the Right Model**:\n",
    "\n",
    "Machine learning models often involve trade-offs. Boosting algorithms, while achieving high overall accuracy and weighted F1 scores, might sometimes sacrifice precision and recall for minority classes in favor of generalizing patterns across the majority class. On the other hand, Naive Bayes and Decision Tree models, being less complex, can capture intricate details in minority classes but might not generalize as well on the majority class.\n",
    "\n",
    "Therefore, choosing the right model depends on the specific objectives of our prediction task and, eventually, the user who will be utilizing our deployed application. If the focus is on correctly identifying wickets and boundary balls, we might prioritize models like Naive Bayes or Decision Trees. However, if a more balanced performance across all classes is desired, as showcased by Weighted F1-Score, the boosting algorithms could be a better choice.\n",
    "\n",
    "**Next Steps**:\n",
    "\n",
    "Next, we shall implement a simple `LGBMClassifier` to see if we can improve the performance of our models. We shall also try tuning the parameters of selected boosting models using `RandomizedSearchCV` to obtain better generalization performance on the minority classes. Fianlly, we will build a stacked classifier with Naive Bayes and Decision Tree as the base estimators and CatBoost as the meta-estimator. We shall then compare the performance of these models with the ones we have already built."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## LightGBM Experiment "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[LightGBM] [Info] Auto-choosing row-wise multi-threading, the overhead of testing was 0.001597 seconds.\n",
      "You can set `force_row_wise=true` to remove the overhead.\n",
      "And if memory is not enough, you can set `force_col_wise=true`.\n",
      "[LightGBM] [Info] Total Bins 489\n",
      "[LightGBM] [Info] Number of data points in the train set: 117757, number of used features: 54\n",
      "[LightGBM] [Info] Start training from score -1.196053\n",
      "[LightGBM] [Info] Start training from score -2.908271\n",
      "[LightGBM] [Info] Start training from score -0.521043\n",
      "[LightGBM] [Info] Start training from score -3.013009\n",
      "Accuracy Score:  0.6345127773097444\n",
      "Weighted F1 Score:  0.5581512338764404\n",
      "Classification Report: \n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "           0       0.58      0.21      0.31      4260\n",
      "           1       0.00      0.00      0.00       654\n",
      "           2       0.64      0.94      0.76      8661\n",
      "           3       0.73      0.04      0.08       669\n",
      "\n",
      "    accuracy                           0.63     14244\n",
      "   macro avg       0.49      0.30      0.29     14244\n",
      "weighted avg       0.60      0.63      0.56     14244\n",
      "\n",
      "Confusion Matrix: \n",
      " [[ 886    0 3369    5]\n",
      " [  44    0  610    0]\n",
      " [ 531    0 8125    5]\n",
      " [  64    0  578   27]]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Program Files\\Python310\\lib\\site-packages\\sklearn\\metrics\\_classification.py:1469: UndefinedMetricWarning:\n",
      "\n",
      "Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "\n",
      "c:\\Program Files\\Python310\\lib\\site-packages\\sklearn\\metrics\\_classification.py:1469: UndefinedMetricWarning:\n",
      "\n",
      "Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "\n",
      "c:\\Program Files\\Python310\\lib\\site-packages\\sklearn\\metrics\\_classification.py:1469: UndefinedMetricWarning:\n",
      "\n",
      "Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score, f1_score\n",
    "from lightgbm import LGBMClassifier\n",
    "\n",
    "LGBM = LGBMClassifier()\n",
    "\n",
    "LGBM.fit(X_train, y_train)\n",
    "\n",
    "y_pred = LGBM.predict(X_test)\n",
    "\n",
    "print(\"Accuracy Score: \", accuracy_score(y_test, y_pred))\n",
    "\n",
    "print(\"Weighted F1 Score: \", f1_score(y_test, y_pred, average=\"weighted\"))\n",
    "\n",
    "print(\"Classification Report: \\n\", classification_report(y_test, y_pred))\n",
    "\n",
    "print(\"Confusion Matrix: \\n\", confusion_matrix(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Hyperparameter Tuning With `RandomizedSearchCV`\n",
    "\n",
    "## XGBoost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/Tegveer/miniforge3/envs/ANLY501/lib/python3.9/site-packages/joblib/externals/loky/process_executor.py:752: UserWarning:\n",
      "\n",
      "A worker stopped while some jobs were given to the executor. This can be caused by a too short worker timeout or by a memory leak.\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The best parameters after tuning are:  {'learning_rate': 1, 'gamma': 2, 'booster': 'dart'}\n",
      "The best model after tuning looks like:  XGBClassifier(base_score=None, booster='dart', callbacks=None,\n",
      "              colsample_bylevel=None, colsample_bynode=None,\n",
      "              colsample_bytree=None, device=None, early_stopping_rounds=None,\n",
      "              enable_categorical=False, eval_metric=None, feature_types=None,\n",
      "              gamma=2, grow_policy=None, importance_type=None,\n",
      "              interaction_constraints=None, learning_rate=1, max_bin=None,\n",
      "              max_cat_threshold=None, max_cat_to_onehot=None,\n",
      "              max_delta_step=None, max_depth=None, max_leaves=None,\n",
      "              min_child_weight=None, missing=nan, monotone_constraints=None,\n",
      "              multi_strategy=None, n_estimators=None, n_jobs=None,\n",
      "              num_parallel_tree=None, objective='multi:softprob', ...)\n",
      "Grid-Search Tuned XGBoost Test Set Accuracy: 55.88 %\n"
     ]
    }
   ],
   "source": [
    "# defining parameter range\n",
    "param_search = {\n",
    "    #'max_depth': [2, 5, 10, 20],\n",
    "    #'max_leaves': [0, 1, 3, 5],\n",
    "    \"learning_rate\": [1, 0.1, 0.01],\n",
    "    #'n_estimators': [50, 100],\n",
    "    \"booster\": [\"gbtree\", \"gblinear\", \"dart\"],\n",
    "    \"gamma\": [0, 0.5, 2, 5],\n",
    "    #'min_child_weight': [0, 1, 3, 5],\n",
    "}\n",
    "\n",
    "search = RandomizedSearchCV(\n",
    "    XGBClassifier(verbosity=0),\n",
    "    param_search,\n",
    "    refit=True,\n",
    "    verbose=0,\n",
    "    n_iter=25,\n",
    "    scoring=\"f1_weighted\",\n",
    "    cv=5,\n",
    "    n_jobs=-1,\n",
    ")\n",
    "\n",
    "search.fit(X_train, y_train)\n",
    "\n",
    "# print best parameter after tuning\n",
    "print(\"The best parameters after tuning are: \", search.best_params_)\n",
    "\n",
    "# print how our model looks after hyper-parameter tuning\n",
    "print(\"The best model after tuning looks like: \", search.best_estimator_)\n",
    "\n",
    "# Save to file in the current working directory\n",
    "pkl_filename = \"products/pickle_files/XGB_Rand_Tuned.pkl\"\n",
    "with open(pkl_filename, \"wb\") as file:\n",
    "    pickle.dump(search, file)\n",
    "\n",
    "# Load from file\n",
    "with open(pkl_filename, \"rb\") as file:\n",
    "    pickle_model = pickle.load(file)\n",
    "    score = pickle_model.score(X_test, y_test)\n",
    "    print(\"Grid-Search Tuned XGBoost Test Set Accuracy: {0:.2f} %\".format(100 * score))\n",
    "    search_predictions = pickle_model.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "conf_matrix = confusion_matrix(y_test, search_predictions)\n",
    "disp = ConfusionMatrixDisplay(\n",
    "    confusion_matrix=conf_matrix,\n",
    "    display_labels=[\"Extras\", \"Dot\", \"Runs\", \"Wicket\"],\n",
    ")\n",
    "\n",
    "# save the plot\n",
    "\n",
    "disp.plot(cmap=\"YlGnBu\", values_format=\"d\", ax=None)\n",
    "plt.title(\n",
    "    \"Test Set Confusion Matrix for Random-Search Tuned XGBoost\",\n",
    "    fontsize=15,\n",
    "    color=\"#336699\",\n",
    "    loc=\"center\",\n",
    ")\n",
    "plt.savefig(\"products/plots/confusion_matrix_XGB_Rand_Tuned.png\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## AdaBoost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/Tegveer/miniforge3/envs/ANLY501/lib/python3.9/site-packages/sklearn/model_selection/_search.py:307: UserWarning:\n",
      "\n",
      "The total space of parameters 1 is smaller than n_iter=25. Running 1 iterations. For exhaustive searches, use GridSearchCV.\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The best parameters after tuning are:  {'n_estimators': 500, 'learning_rate': 1}\n",
      "The best model after tuning looks like:  AdaBoostClassifier(learning_rate=1, n_estimators=500)\n",
      "Grid-Search Tuned XGBoost Test Set Accuracy: 55.67 %\n"
     ]
    }
   ],
   "source": [
    "# defining parameter range\n",
    "param_search = {\n",
    "    \"learning_rate\": [1, 0.1, 0.01],\n",
    "    \"n_estimators\": [50, 100, 200],\n",
    "    \"estimator\": [DecisionTreeClassifier(), Sup],\n",
    "}\n",
    "\n",
    "search = RandomizedSearchCV(\n",
    "    AdaBoostClassifier(),\n",
    "    param_search,\n",
    "    refit=True,\n",
    "    verbose=0,\n",
    "    n_iter=25,\n",
    "    scoring=\"f1_weighted\",\n",
    "    cv=5,\n",
    "    n_jobs=-1,\n",
    ")\n",
    "\n",
    "search.fit(X_train, y_train)\n",
    "\n",
    "# print best parameter after tuning\n",
    "print(\"The best parameters after tuning are: \", search.best_params_)\n",
    "\n",
    "# print how our model looks after hyper-parameter tuning\n",
    "print(\"The best model after tuning looks like: \", search.best_estimator_)\n",
    "\n",
    "# Save to file in the current working directory\n",
    "pkl_filename = \"products/pickle_files/Ada_Rand_Tuned.pkl\"\n",
    "with open(pkl_filename, \"wb\") as file:\n",
    "    pickle.dump(search, file)\n",
    "\n",
    "# Load from file\n",
    "with open(pkl_filename, \"rb\") as file:\n",
    "    pickle_model = pickle.load(file)\n",
    "    score = pickle_model.score(X_test, y_test)\n",
    "    print(\"Grid-Search Tuned XGBoost Test Set Accuracy: {0:.2f} %\".format(100 * score))\n",
    "    search_predictions = pickle_model.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "conf_matrix = confusion_matrix(y_test, search_predictions)\n",
    "disp = ConfusionMatrixDisplay(\n",
    "    confusion_matrix=conf_matrix,\n",
    "    display_labels=[\"Extras\", \"Dot\", \"Runs\", \"Wicket\"],\n",
    ")\n",
    "\n",
    "# save the plot\n",
    "\n",
    "disp.plot(cmap=\"YlGnBu\", values_format=\"d\", ax=None)\n",
    "plt.title(\n",
    "    \"Test Set Confusion Matrix for Random-Search Tuned AdaBoost\",\n",
    "    fontsize=15,\n",
    "    color=\"#336699\",\n",
    "    loc=\"center\",\n",
    ")\n",
    "plt.savefig(\"products/plots/confusion_matrix_Ada_Rand_Tuned.png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# remove all pickle files for smooth git push\n",
    "\n",
    "for name, model in models.items():\n",
    "    pkl_filename = f\"products/pickle_files/{name}.pkl\"\n",
    "    os.remove(pkl_filename)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Hyperparameter tuning, unfortunately, resulted in worse performing models. The default parameters of the boosting algorithms seem to be the best fit for our data. Let's now build a `StackingClassifier` with Naive Bayes and Decision Tree as the base estimators and CatBoost as the meta-estimator."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Creating The Stacked Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_stacking():\n",
    "    # define the base models\n",
    "    level0 = list()\n",
    "    level0.append((\"GNB\", GaussianNB()))\n",
    "    level0.append((\"Tree\", DecisionTreeClassifier()))\n",
    "    # define meta learner model\n",
    "\n",
    "    level1 = CatBoostClassifier(verbose=0)\n",
    "\n",
    "    # define the stacking ensemble\n",
    "    model = StackingClassifier(estimators=level0, final_estimator=level1, cv=10)\n",
    "\n",
    "    return model\n",
    "\n",
    "\n",
    "def base_models():\n",
    "    models = dict()\n",
    "    models[\"GNB\"] = GaussianNB()\n",
    "    models[\"Tree\"] = DecisionTreeClassifier()\n",
    "    models[\"XGB\"] = XGBClassifier(verbosity=0)\n",
    "    models[\"Stacked Model\"] = get_stacking()\n",
    "    return models\n",
    "\n",
    "\n",
    "# Function to evaluate the list of models\n",
    "def eval_models(model):\n",
    "    cv = RepeatedKFold(n_splits=5, n_repeats=2, random_state=1)\n",
    "    accuracy = cross_val_score(\n",
    "        model,\n",
    "        X_train,\n",
    "        y_train,\n",
    "        scoring=\"accuracy\",\n",
    "        cv=cv,\n",
    "        n_jobs=-1,\n",
    "        error_score=\"raise\",\n",
    "    )\n",
    "    weighted_f1 = cross_val_score(\n",
    "        model,\n",
    "        X_train,\n",
    "        y_train,\n",
    "        scoring=\"f1_weighted\",\n",
    "        cv=cv,\n",
    "        n_jobs=-1,\n",
    "        error_score=\"raise\",\n",
    "    )\n",
    "    return accuracy, weighted_f1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">GNB Accuracy: 0.483 Weighted F1: 0.493\n",
      ">Tree Accuracy: 0.491 Weighted F1: 0.494\n",
      ">XGB Accuracy: 0.622 Weighted F1: 0.553\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/Tegveer/miniforge3/envs/ANLY501/lib/python3.9/site-packages/sklearn/preprocessing/_label.py:153: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n",
      "/Users/Tegveer/miniforge3/envs/ANLY501/lib/python3.9/site-packages/sklearn/preprocessing/_label.py:153: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n",
      "/Users/Tegveer/miniforge3/envs/ANLY501/lib/python3.9/site-packages/sklearn/preprocessing/_label.py:153: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n",
      "/Users/Tegveer/miniforge3/envs/ANLY501/lib/python3.9/site-packages/sklearn/preprocessing/_label.py:153: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n",
      "/Users/Tegveer/miniforge3/envs/ANLY501/lib/python3.9/site-packages/sklearn/preprocessing/_label.py:153: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n",
      "/Users/Tegveer/miniforge3/envs/ANLY501/lib/python3.9/site-packages/sklearn/preprocessing/_label.py:153: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n",
      "/Users/Tegveer/miniforge3/envs/ANLY501/lib/python3.9/site-packages/sklearn/preprocessing/_label.py:153: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n",
      "/Users/Tegveer/miniforge3/envs/ANLY501/lib/python3.9/site-packages/sklearn/preprocessing/_label.py:153: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n",
      "/Users/Tegveer/miniforge3/envs/ANLY501/lib/python3.9/site-packages/sklearn/preprocessing/_label.py:153: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n",
      "/Users/Tegveer/miniforge3/envs/ANLY501/lib/python3.9/site-packages/sklearn/preprocessing/_label.py:153: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n",
      "/Users/Tegveer/miniforge3/envs/ANLY501/lib/python3.9/site-packages/sklearn/preprocessing/_label.py:153: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n",
      "/Users/Tegveer/miniforge3/envs/ANLY501/lib/python3.9/site-packages/sklearn/preprocessing/_label.py:153: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n",
      "/Users/Tegveer/miniforge3/envs/ANLY501/lib/python3.9/site-packages/sklearn/preprocessing/_label.py:153: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n",
      "/Users/Tegveer/miniforge3/envs/ANLY501/lib/python3.9/site-packages/sklearn/preprocessing/_label.py:153: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n",
      "/Users/Tegveer/miniforge3/envs/ANLY501/lib/python3.9/site-packages/sklearn/preprocessing/_label.py:153: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n",
      "/Users/Tegveer/miniforge3/envs/ANLY501/lib/python3.9/site-packages/sklearn/preprocessing/_label.py:153: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n",
      "/Users/Tegveer/miniforge3/envs/ANLY501/lib/python3.9/site-packages/sklearn/preprocessing/_label.py:153: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n",
      "/Users/Tegveer/miniforge3/envs/ANLY501/lib/python3.9/site-packages/sklearn/preprocessing/_label.py:153: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n",
      "/Users/Tegveer/miniforge3/envs/ANLY501/lib/python3.9/site-packages/sklearn/preprocessing/_label.py:153: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">Stacked Model Accuracy: 0.597 Weighted F1: 0.507\n",
      "CPU times: user 630 ms, sys: 460 ms, total: 1.09 s\n",
      "Wall time: 3min 11s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/Tegveer/miniforge3/envs/ANLY501/lib/python3.9/site-packages/sklearn/preprocessing/_label.py:153: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "models = base_models()\n",
    "# evaluate the models and store results\n",
    "results = {}\n",
    "names = list()\n",
    "accuracies = []\n",
    "weighted_f1s = []\n",
    "\n",
    "for name, model in models.items():\n",
    "    accuracy, weighted_f1 = eval_models(model)\n",
    "    accuracies.extend(accuracy)\n",
    "    weighted_f1s.extend(weighted_f1)\n",
    "    names.extend([name] * len(accuracy))\n",
    "    print(\n",
    "        \">%s Accuracy: %.3f Weighted F1: %.3f\"\n",
    "        % (name, accuracy.mean(), weighted_f1.mean())\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Store Results "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 158,
   "metadata": {},
   "outputs": [],
   "source": [
    "results[\"Model\"] = names\n",
    "results[\"Accuracy\"] = accuracies\n",
    "results[\"Weighted_F1\"] = weighted_f1s\n",
    "\n",
    "classifmod1 = pd.DataFrame(results)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualize Train Set Accuracies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.plotly.v1+json": {
       "config": {
        "plotlyServerURL": "https://plotly.com"
       },
       "data": [
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>",
         "legendgroup": "GNB",
         "marker": {
          "color": "#636efa"
         },
         "name": "GNB",
         "notched": false,
         "offsetgroup": "GNB",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "GNB",
          "GNB",
          "GNB",
          "GNB",
          "GNB",
          "GNB",
          "GNB",
          "GNB",
          "GNB",
          "GNB"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.48063858695652173,
          0.4782608695652174,
          0.4919111715001486,
          0.47747441722219863,
          0.4830368137234088,
          0.48191236413043476,
          0.48276154891304346,
          0.49114687274425717,
          0.4781537938941022,
          0.4809562226657042
         ],
         "y0": " ",
         "yaxis": "y"
        },
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>",
         "legendgroup": "Tree",
         "marker": {
          "color": "#EF553B"
         },
         "name": "Tree",
         "notched": false,
         "offsetgroup": "Tree",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "Tree",
          "Tree",
          "Tree",
          "Tree",
          "Tree",
          "Tree",
          "Tree",
          "Tree",
          "Tree",
          "Tree"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.48904551630434784,
          0.48904551630434784,
          0.49662434716147935,
          0.4910619506602692,
          0.4928877754660099,
          0.48934273097826086,
          0.48713485054347827,
          0.49157148316419685,
          0.4909770285762813,
          0.49182624941616065
         ],
         "y0": " ",
         "yaxis": "y"
        },
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>",
         "legendgroup": "XGB",
         "marker": {
          "color": "#00cc96"
         },
         "name": "XGB",
         "notched": false,
         "offsetgroup": "XGB",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "XGB",
          "XGB",
          "XGB",
          "XGB",
          "XGB",
          "XGB",
          "XGB",
          "XGB",
          "XGB",
          "XGB"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.6230893342391305,
          0.6224099864130435,
          0.6204407456158975,
          0.6231582523035115,
          0.6203133624899155,
          0.6266559103260869,
          0.625,
          0.6166192518364401,
          0.622478875631608,
          0.6170863232983738
         ],
         "y0": " ",
         "yaxis": "y"
        },
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>",
         "legendgroup": "Stacked Model",
         "marker": {
          "color": "#ab63fa"
         },
         "name": "Stacked Model",
         "notched": false,
         "offsetgroup": "Stacked Model",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "Stacked Model",
          "Stacked Model",
          "Stacked Model",
          "Stacked Model",
          "Stacked Model",
          "Stacked Model",
          "Stacked Model",
          "Stacked Model",
          "Stacked Model",
          "Stacked Model"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.5955757472826086,
          0.600968070652174,
          0.5953887308394548,
          0.5985308479470086,
          0.5981911596110568,
          0.5978260869565217,
          0.5979534646739131,
          0.5949641204195151,
          0.5989129973249544,
          0.5927986072778226
         ],
         "y0": " ",
         "yaxis": "y"
        }
       ],
       "layout": {
        "boxmode": "overlay",
        "legend": {
         "title": {
          "text": "Machine Learning Model"
         },
         "tracegroupgap": 0
        },
        "template": {
         "data": {
          "bar": [
           {
            "error_x": {
             "color": "#2a3f5f"
            },
            "error_y": {
             "color": "#2a3f5f"
            },
            "marker": {
             "line": {
              "color": "#E5ECF6",
              "width": 0.5
             },
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "bar"
           }
          ],
          "barpolar": [
           {
            "marker": {
             "line": {
              "color": "#E5ECF6",
              "width": 0.5
             },
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "barpolar"
           }
          ],
          "carpet": [
           {
            "aaxis": {
             "endlinecolor": "#2a3f5f",
             "gridcolor": "white",
             "linecolor": "white",
             "minorgridcolor": "white",
             "startlinecolor": "#2a3f5f"
            },
            "baxis": {
             "endlinecolor": "#2a3f5f",
             "gridcolor": "white",
             "linecolor": "white",
             "minorgridcolor": "white",
             "startlinecolor": "#2a3f5f"
            },
            "type": "carpet"
           }
          ],
          "choropleth": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "choropleth"
           }
          ],
          "contour": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "contour"
           }
          ],
          "contourcarpet": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "contourcarpet"
           }
          ],
          "heatmap": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "heatmap"
           }
          ],
          "heatmapgl": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "heatmapgl"
           }
          ],
          "histogram": [
           {
            "marker": {
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "histogram"
           }
          ],
          "histogram2d": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "histogram2d"
           }
          ],
          "histogram2dcontour": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "histogram2dcontour"
           }
          ],
          "mesh3d": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "mesh3d"
           }
          ],
          "parcoords": [
           {
            "line": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "parcoords"
           }
          ],
          "pie": [
           {
            "automargin": true,
            "type": "pie"
           }
          ],
          "scatter": [
           {
            "fillpattern": {
             "fillmode": "overlay",
             "size": 10,
             "solidity": 0.2
            },
            "type": "scatter"
           }
          ],
          "scatter3d": [
           {
            "line": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatter3d"
           }
          ],
          "scattercarpet": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattercarpet"
           }
          ],
          "scattergeo": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattergeo"
           }
          ],
          "scattergl": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattergl"
           }
          ],
          "scattermapbox": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattermapbox"
           }
          ],
          "scatterpolar": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterpolar"
           }
          ],
          "scatterpolargl": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterpolargl"
           }
          ],
          "scatterternary": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterternary"
           }
          ],
          "surface": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "surface"
           }
          ],
          "table": [
           {
            "cells": {
             "fill": {
              "color": "#EBF0F8"
             },
             "line": {
              "color": "white"
             }
            },
            "header": {
             "fill": {
              "color": "#C8D4E3"
             },
             "line": {
              "color": "white"
             }
            },
            "type": "table"
           }
          ]
         },
         "layout": {
          "annotationdefaults": {
           "arrowcolor": "#2a3f5f",
           "arrowhead": 0,
           "arrowwidth": 1
          },
          "autotypenumbers": "strict",
          "coloraxis": {
           "colorbar": {
            "outlinewidth": 0,
            "ticks": ""
           }
          },
          "colorscale": {
           "diverging": [
            [
             0,
             "#8e0152"
            ],
            [
             0.1,
             "#c51b7d"
            ],
            [
             0.2,
             "#de77ae"
            ],
            [
             0.3,
             "#f1b6da"
            ],
            [
             0.4,
             "#fde0ef"
            ],
            [
             0.5,
             "#f7f7f7"
            ],
            [
             0.6,
             "#e6f5d0"
            ],
            [
             0.7,
             "#b8e186"
            ],
            [
             0.8,
             "#7fbc41"
            ],
            [
             0.9,
             "#4d9221"
            ],
            [
             1,
             "#276419"
            ]
           ],
           "sequential": [
            [
             0,
             "#0d0887"
            ],
            [
             0.1111111111111111,
             "#46039f"
            ],
            [
             0.2222222222222222,
             "#7201a8"
            ],
            [
             0.3333333333333333,
             "#9c179e"
            ],
            [
             0.4444444444444444,
             "#bd3786"
            ],
            [
             0.5555555555555556,
             "#d8576b"
            ],
            [
             0.6666666666666666,
             "#ed7953"
            ],
            [
             0.7777777777777778,
             "#fb9f3a"
            ],
            [
             0.8888888888888888,
             "#fdca26"
            ],
            [
             1,
             "#f0f921"
            ]
           ],
           "sequentialminus": [
            [
             0,
             "#0d0887"
            ],
            [
             0.1111111111111111,
             "#46039f"
            ],
            [
             0.2222222222222222,
             "#7201a8"
            ],
            [
             0.3333333333333333,
             "#9c179e"
            ],
            [
             0.4444444444444444,
             "#bd3786"
            ],
            [
             0.5555555555555556,
             "#d8576b"
            ],
            [
             0.6666666666666666,
             "#ed7953"
            ],
            [
             0.7777777777777778,
             "#fb9f3a"
            ],
            [
             0.8888888888888888,
             "#fdca26"
            ],
            [
             1,
             "#f0f921"
            ]
           ]
          },
          "colorway": [
           "#636efa",
           "#EF553B",
           "#00cc96",
           "#ab63fa",
           "#FFA15A",
           "#19d3f3",
           "#FF6692",
           "#B6E880",
           "#FF97FF",
           "#FECB52"
          ],
          "font": {
           "color": "#2a3f5f"
          },
          "geo": {
           "bgcolor": "white",
           "lakecolor": "white",
           "landcolor": "#E5ECF6",
           "showlakes": true,
           "showland": true,
           "subunitcolor": "white"
          },
          "hoverlabel": {
           "align": "left"
          },
          "hovermode": "closest",
          "mapbox": {
           "style": "light"
          },
          "paper_bgcolor": "white",
          "plot_bgcolor": "#E5ECF6",
          "polar": {
           "angularaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "bgcolor": "#E5ECF6",
           "radialaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           }
          },
          "scene": {
           "xaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           },
           "yaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           },
           "zaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           }
          },
          "shapedefaults": {
           "line": {
            "color": "#2a3f5f"
           }
          },
          "ternary": {
           "aaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "baxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "bgcolor": "#E5ECF6",
           "caxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           }
          },
          "title": {
           "x": 0.05
          },
          "xaxis": {
           "automargin": true,
           "gridcolor": "white",
           "linecolor": "white",
           "ticks": "",
           "title": {
            "standoff": 15
           },
           "zerolinecolor": "white",
           "zerolinewidth": 2
          },
          "yaxis": {
           "automargin": true,
           "gridcolor": "white",
           "linecolor": "white",
           "ticks": "",
           "title": {
            "standoff": 15
           },
           "zerolinecolor": "white",
           "zerolinewidth": 2
          }
         }
        },
        "title": {
         "text": "Model Accuracy Performance"
        },
        "xaxis": {
         "anchor": "y",
         "categoryarray": [
          "GNB",
          "Tree",
          "XGB",
          "Stacked Model"
         ],
         "categoryorder": "array",
         "domain": [
          0,
          1
         ],
         "title": {
          "text": "Machine Learning Model"
         }
        },
        "yaxis": {
         "anchor": "x",
         "domain": [
          0,
          1
         ],
         "title": {
          "text": "Accuracy"
         }
        }
       }
      },
      "text/html": [
       "<div>                            <div id=\"cce56d44-8739-45d2-91d5-a1e59cc8b7e2\" class=\"plotly-graph-div\" style=\"height:525px; width:100%;\"></div>            <script type=\"text/javascript\">                require([\"plotly\"], function(Plotly) {                    window.PLOTLYENV=window.PLOTLYENV || {};                                    if (document.getElementById(\"cce56d44-8739-45d2-91d5-a1e59cc8b7e2\")) {                    Plotly.newPlot(                        \"cce56d44-8739-45d2-91d5-a1e59cc8b7e2\",                        [{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>\",\"legendgroup\":\"GNB\",\"marker\":{\"color\":\"#636efa\"},\"name\":\"GNB\",\"notched\":false,\"offsetgroup\":\"GNB\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"GNB\",\"GNB\",\"GNB\",\"GNB\",\"GNB\",\"GNB\",\"GNB\",\"GNB\",\"GNB\",\"GNB\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.48063858695652173,0.4782608695652174,0.4919111715001486,0.47747441722219863,0.4830368137234088,0.48191236413043476,0.48276154891304346,0.49114687274425717,0.4781537938941022,0.4809562226657042],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"},{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>\",\"legendgroup\":\"Tree\",\"marker\":{\"color\":\"#EF553B\"},\"name\":\"Tree\",\"notched\":false,\"offsetgroup\":\"Tree\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"Tree\",\"Tree\",\"Tree\",\"Tree\",\"Tree\",\"Tree\",\"Tree\",\"Tree\",\"Tree\",\"Tree\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.48904551630434784,0.48904551630434784,0.49662434716147935,0.4910619506602692,0.4928877754660099,0.48934273097826086,0.48713485054347827,0.49157148316419685,0.4909770285762813,0.49182624941616065],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"},{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>\",\"legendgroup\":\"XGB\",\"marker\":{\"color\":\"#00cc96\"},\"name\":\"XGB\",\"notched\":false,\"offsetgroup\":\"XGB\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"XGB\",\"XGB\",\"XGB\",\"XGB\",\"XGB\",\"XGB\",\"XGB\",\"XGB\",\"XGB\",\"XGB\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.6230893342391305,0.6224099864130435,0.6204407456158975,0.6231582523035115,0.6203133624899155,0.6266559103260869,0.625,0.6166192518364401,0.622478875631608,0.6170863232983738],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"},{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Accuracy=%{y}<extra></extra>\",\"legendgroup\":\"Stacked Model\",\"marker\":{\"color\":\"#ab63fa\"},\"name\":\"Stacked Model\",\"notched\":false,\"offsetgroup\":\"Stacked Model\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"Stacked Model\",\"Stacked Model\",\"Stacked Model\",\"Stacked Model\",\"Stacked Model\",\"Stacked Model\",\"Stacked Model\",\"Stacked Model\",\"Stacked Model\",\"Stacked Model\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.5955757472826086,0.600968070652174,0.5953887308394548,0.5985308479470086,0.5981911596110568,0.5978260869565217,0.5979534646739131,0.5949641204195151,0.5989129973249544,0.5927986072778226],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"}],                        {\"template\":{\"data\":{\"histogram2dcontour\":[{\"type\":\"histogram2dcontour\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"choropleth\":[{\"type\":\"choropleth\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}],\"histogram2d\":[{\"type\":\"histogram2d\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"heatmap\":[{\"type\":\"heatmap\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"heatmapgl\":[{\"type\":\"heatmapgl\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"contourcarpet\":[{\"type\":\"contourcarpet\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}],\"contour\":[{\"type\":\"contour\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"surface\":[{\"type\":\"surface\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"mesh3d\":[{\"type\":\"mesh3d\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}],\"scatter\":[{\"fillpattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2},\"type\":\"scatter\"}],\"parcoords\":[{\"type\":\"parcoords\",\"line\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scatterpolargl\":[{\"type\":\"scatterpolargl\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"bar\":[{\"error_x\":{\"color\":\"#2a3f5f\"},\"error_y\":{\"color\":\"#2a3f5f\"},\"marker\":{\"line\":{\"color\":\"#E5ECF6\",\"width\":0.5},\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"bar\"}],\"scattergeo\":[{\"type\":\"scattergeo\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scatterpolar\":[{\"type\":\"scatterpolar\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"histogram\":[{\"marker\":{\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"histogram\"}],\"scattergl\":[{\"type\":\"scattergl\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scatter3d\":[{\"type\":\"scatter3d\",\"line\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scattermapbox\":[{\"type\":\"scattermapbox\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scatterternary\":[{\"type\":\"scatterternary\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scattercarpet\":[{\"type\":\"scattercarpet\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"carpet\":[{\"aaxis\":{\"endlinecolor\":\"#2a3f5f\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"minorgridcolor\":\"white\",\"startlinecolor\":\"#2a3f5f\"},\"baxis\":{\"endlinecolor\":\"#2a3f5f\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"minorgridcolor\":\"white\",\"startlinecolor\":\"#2a3f5f\"},\"type\":\"carpet\"}],\"table\":[{\"cells\":{\"fill\":{\"color\":\"#EBF0F8\"},\"line\":{\"color\":\"white\"}},\"header\":{\"fill\":{\"color\":\"#C8D4E3\"},\"line\":{\"color\":\"white\"}},\"type\":\"table\"}],\"barpolar\":[{\"marker\":{\"line\":{\"color\":\"#E5ECF6\",\"width\":0.5},\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"barpolar\"}],\"pie\":[{\"automargin\":true,\"type\":\"pie\"}]},\"layout\":{\"autotypenumbers\":\"strict\",\"colorway\":[\"#636efa\",\"#EF553B\",\"#00cc96\",\"#ab63fa\",\"#FFA15A\",\"#19d3f3\",\"#FF6692\",\"#B6E880\",\"#FF97FF\",\"#FECB52\"],\"font\":{\"color\":\"#2a3f5f\"},\"hovermode\":\"closest\",\"hoverlabel\":{\"align\":\"left\"},\"paper_bgcolor\":\"white\",\"plot_bgcolor\":\"#E5ECF6\",\"polar\":{\"bgcolor\":\"#E5ECF6\",\"angularaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"radialaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"}},\"ternary\":{\"bgcolor\":\"#E5ECF6\",\"aaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"baxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"caxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"}},\"coloraxis\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"colorscale\":{\"sequential\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"sequentialminus\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"diverging\":[[0,\"#8e0152\"],[0.1,\"#c51b7d\"],[0.2,\"#de77ae\"],[0.3,\"#f1b6da\"],[0.4,\"#fde0ef\"],[0.5,\"#f7f7f7\"],[0.6,\"#e6f5d0\"],[0.7,\"#b8e186\"],[0.8,\"#7fbc41\"],[0.9,\"#4d9221\"],[1,\"#276419\"]]},\"xaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\",\"title\":{\"standoff\":15},\"zerolinecolor\":\"white\",\"automargin\":true,\"zerolinewidth\":2},\"yaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\",\"title\":{\"standoff\":15},\"zerolinecolor\":\"white\",\"automargin\":true,\"zerolinewidth\":2},\"scene\":{\"xaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\",\"gridwidth\":2},\"yaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\",\"gridwidth\":2},\"zaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\",\"gridwidth\":2}},\"shapedefaults\":{\"line\":{\"color\":\"#2a3f5f\"}},\"annotationdefaults\":{\"arrowcolor\":\"#2a3f5f\",\"arrowhead\":0,\"arrowwidth\":1},\"geo\":{\"bgcolor\":\"white\",\"landcolor\":\"#E5ECF6\",\"subunitcolor\":\"white\",\"showland\":true,\"showlakes\":true,\"lakecolor\":\"white\"},\"title\":{\"x\":0.05},\"mapbox\":{\"style\":\"light\"}}},\"xaxis\":{\"anchor\":\"y\",\"domain\":[0.0,1.0],\"title\":{\"text\":\"Machine Learning Model\"},\"categoryorder\":\"array\",\"categoryarray\":[\"GNB\",\"Tree\",\"XGB\",\"Stacked Model\"]},\"yaxis\":{\"anchor\":\"x\",\"domain\":[0.0,1.0],\"title\":{\"text\":\"Accuracy\"}},\"legend\":{\"title\":{\"text\":\"Machine Learning Model\"},\"tracegroupgap\":0},\"title\":{\"text\":\"Model Accuracy Performance\"},\"boxmode\":\"overlay\"},                        {\"responsive\": true}                    ).then(function(){\n",
       "                            \n",
       "var gd = document.getElementById('cce56d44-8739-45d2-91d5-a1e59cc8b7e2');\n",
       "var x = new MutationObserver(function (mutations, observer) {{\n",
       "        var display = window.getComputedStyle(gd).display;\n",
       "        if (!display || display === 'none') {{\n",
       "            console.log([gd, 'removed!']);\n",
       "            Plotly.purge(gd);\n",
       "            observer.disconnect();\n",
       "        }}\n",
       "}});\n",
       "\n",
       "// Listen for the removal of the full notebook cells\n",
       "var notebookContainer = gd.closest('#notebook-container');\n",
       "if (notebookContainer) {{\n",
       "    x.observe(notebookContainer, {childList: true});\n",
       "}}\n",
       "\n",
       "// Listen for the clearing of the current output cell\n",
       "var outputEl = gd.closest('.output');\n",
       "if (outputEl) {{\n",
       "    x.observe(outputEl, {childList: true});\n",
       "}}\n",
       "\n",
       "                        })                };                });            </script>        </div>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = px.box(\n",
    "    classifmod1,\n",
    "    x=\"Model\",\n",
    "    y=\"Accuracy\",\n",
    "    color=\"Model\",\n",
    "    points=\"all\",\n",
    "    labels={\"Model\": \"Machine Learning Model\", \"Accuracy\": \"Accuracy\"},\n",
    "    title=\"Model Accuracy Performance\",\n",
    ")\n",
    "fig.show()\n",
    "fig.write_html(\"products/plots/stacked_train_accuracies.html\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualize Train Set Weighted f1-Scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.plotly.v1+json": {
       "config": {
        "plotlyServerURL": "https://plotly.com"
       },
       "data": [
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>",
         "legendgroup": "GNB",
         "marker": {
          "color": "#636efa"
         },
         "name": "GNB",
         "notched": false,
         "offsetgroup": "GNB",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "GNB",
          "GNB",
          "GNB",
          "GNB",
          "GNB",
          "GNB",
          "GNB",
          "GNB",
          "GNB",
          "GNB"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.49021277923226114,
          0.48816174819698194,
          0.5016366850721734,
          0.4890656186573671,
          0.4937436255807292,
          0.4931294872578772,
          0.4949186105771983,
          0.4991549262764149,
          0.4883404620898014,
          0.4896592647365606
         ],
         "y0": " ",
         "yaxis": "y"
        },
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>",
         "legendgroup": "Tree",
         "marker": {
          "color": "#EF553B"
         },
         "name": "Tree",
         "notched": false,
         "offsetgroup": "Tree",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "Tree",
          "Tree",
          "Tree",
          "Tree",
          "Tree",
          "Tree",
          "Tree",
          "Tree",
          "Tree",
          "Tree"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.49395591647842946,
          0.4913186019958329,
          0.49635265212463026,
          0.4936003617148166,
          0.49728753381826846,
          0.4915021325543078,
          0.4963355793833316,
          0.4923241383014452,
          0.4949583200031113,
          0.49293209331283633
         ],
         "y0": " ",
         "yaxis": "y"
        },
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>",
         "legendgroup": "XGB",
         "marker": {
          "color": "#00cc96"
         },
         "name": "XGB",
         "notched": false,
         "offsetgroup": "XGB",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "XGB",
          "XGB",
          "XGB",
          "XGB",
          "XGB",
          "XGB",
          "XGB",
          "XGB",
          "XGB",
          "XGB"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.5559134871731306,
          0.5536083694618409,
          0.5512097489754343,
          0.5551107100753496,
          0.5516901629328173,
          0.5590246002838755,
          0.5580970876343704,
          0.5472750973416809,
          0.555179334886721,
          0.5469519120204224
         ],
         "y0": " ",
         "yaxis": "y"
        },
        {
         "alignmentgroup": "True",
         "boxpoints": "all",
         "hovertemplate": "Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>",
         "legendgroup": "Stacked Model",
         "marker": {
          "color": "#ab63fa"
         },
         "name": "Stacked Model",
         "notched": false,
         "offsetgroup": "Stacked Model",
         "orientation": "v",
         "showlegend": true,
         "type": "box",
         "x": [
          "Stacked Model",
          "Stacked Model",
          "Stacked Model",
          "Stacked Model",
          "Stacked Model",
          "Stacked Model",
          "Stacked Model",
          "Stacked Model",
          "Stacked Model",
          "Stacked Model"
         ],
         "x0": " ",
         "xaxis": "x",
         "y": [
          0.5078471990699265,
          0.5049234732497797,
          0.5069081786582854,
          0.5082227301871328,
          0.5082061631095763,
          0.5047138232159096,
          0.5086445787017143,
          0.5075823903883498,
          0.5086454103430422,
          0.5041613087296903
         ],
         "y0": " ",
         "yaxis": "y"
        }
       ],
       "layout": {
        "boxmode": "overlay",
        "legend": {
         "title": {
          "text": "Machine Learning Model"
         },
         "tracegroupgap": 0
        },
        "template": {
         "data": {
          "bar": [
           {
            "error_x": {
             "color": "#2a3f5f"
            },
            "error_y": {
             "color": "#2a3f5f"
            },
            "marker": {
             "line": {
              "color": "#E5ECF6",
              "width": 0.5
             },
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "bar"
           }
          ],
          "barpolar": [
           {
            "marker": {
             "line": {
              "color": "#E5ECF6",
              "width": 0.5
             },
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "barpolar"
           }
          ],
          "carpet": [
           {
            "aaxis": {
             "endlinecolor": "#2a3f5f",
             "gridcolor": "white",
             "linecolor": "white",
             "minorgridcolor": "white",
             "startlinecolor": "#2a3f5f"
            },
            "baxis": {
             "endlinecolor": "#2a3f5f",
             "gridcolor": "white",
             "linecolor": "white",
             "minorgridcolor": "white",
             "startlinecolor": "#2a3f5f"
            },
            "type": "carpet"
           }
          ],
          "choropleth": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "choropleth"
           }
          ],
          "contour": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "contour"
           }
          ],
          "contourcarpet": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "contourcarpet"
           }
          ],
          "heatmap": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "heatmap"
           }
          ],
          "heatmapgl": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "heatmapgl"
           }
          ],
          "histogram": [
           {
            "marker": {
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "histogram"
           }
          ],
          "histogram2d": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "histogram2d"
           }
          ],
          "histogram2dcontour": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "histogram2dcontour"
           }
          ],
          "mesh3d": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "mesh3d"
           }
          ],
          "parcoords": [
           {
            "line": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "parcoords"
           }
          ],
          "pie": [
           {
            "automargin": true,
            "type": "pie"
           }
          ],
          "scatter": [
           {
            "fillpattern": {
             "fillmode": "overlay",
             "size": 10,
             "solidity": 0.2
            },
            "type": "scatter"
           }
          ],
          "scatter3d": [
           {
            "line": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatter3d"
           }
          ],
          "scattercarpet": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattercarpet"
           }
          ],
          "scattergeo": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattergeo"
           }
          ],
          "scattergl": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattergl"
           }
          ],
          "scattermapbox": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattermapbox"
           }
          ],
          "scatterpolar": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterpolar"
           }
          ],
          "scatterpolargl": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterpolargl"
           }
          ],
          "scatterternary": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterternary"
           }
          ],
          "surface": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "surface"
           }
          ],
          "table": [
           {
            "cells": {
             "fill": {
              "color": "#EBF0F8"
             },
             "line": {
              "color": "white"
             }
            },
            "header": {
             "fill": {
              "color": "#C8D4E3"
             },
             "line": {
              "color": "white"
             }
            },
            "type": "table"
           }
          ]
         },
         "layout": {
          "annotationdefaults": {
           "arrowcolor": "#2a3f5f",
           "arrowhead": 0,
           "arrowwidth": 1
          },
          "autotypenumbers": "strict",
          "coloraxis": {
           "colorbar": {
            "outlinewidth": 0,
            "ticks": ""
           }
          },
          "colorscale": {
           "diverging": [
            [
             0,
             "#8e0152"
            ],
            [
             0.1,
             "#c51b7d"
            ],
            [
             0.2,
             "#de77ae"
            ],
            [
             0.3,
             "#f1b6da"
            ],
            [
             0.4,
             "#fde0ef"
            ],
            [
             0.5,
             "#f7f7f7"
            ],
            [
             0.6,
             "#e6f5d0"
            ],
            [
             0.7,
             "#b8e186"
            ],
            [
             0.8,
             "#7fbc41"
            ],
            [
             0.9,
             "#4d9221"
            ],
            [
             1,
             "#276419"
            ]
           ],
           "sequential": [
            [
             0,
             "#0d0887"
            ],
            [
             0.1111111111111111,
             "#46039f"
            ],
            [
             0.2222222222222222,
             "#7201a8"
            ],
            [
             0.3333333333333333,
             "#9c179e"
            ],
            [
             0.4444444444444444,
             "#bd3786"
            ],
            [
             0.5555555555555556,
             "#d8576b"
            ],
            [
             0.6666666666666666,
             "#ed7953"
            ],
            [
             0.7777777777777778,
             "#fb9f3a"
            ],
            [
             0.8888888888888888,
             "#fdca26"
            ],
            [
             1,
             "#f0f921"
            ]
           ],
           "sequentialminus": [
            [
             0,
             "#0d0887"
            ],
            [
             0.1111111111111111,
             "#46039f"
            ],
            [
             0.2222222222222222,
             "#7201a8"
            ],
            [
             0.3333333333333333,
             "#9c179e"
            ],
            [
             0.4444444444444444,
             "#bd3786"
            ],
            [
             0.5555555555555556,
             "#d8576b"
            ],
            [
             0.6666666666666666,
             "#ed7953"
            ],
            [
             0.7777777777777778,
             "#fb9f3a"
            ],
            [
             0.8888888888888888,
             "#fdca26"
            ],
            [
             1,
             "#f0f921"
            ]
           ]
          },
          "colorway": [
           "#636efa",
           "#EF553B",
           "#00cc96",
           "#ab63fa",
           "#FFA15A",
           "#19d3f3",
           "#FF6692",
           "#B6E880",
           "#FF97FF",
           "#FECB52"
          ],
          "font": {
           "color": "#2a3f5f"
          },
          "geo": {
           "bgcolor": "white",
           "lakecolor": "white",
           "landcolor": "#E5ECF6",
           "showlakes": true,
           "showland": true,
           "subunitcolor": "white"
          },
          "hoverlabel": {
           "align": "left"
          },
          "hovermode": "closest",
          "mapbox": {
           "style": "light"
          },
          "paper_bgcolor": "white",
          "plot_bgcolor": "#E5ECF6",
          "polar": {
           "angularaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "bgcolor": "#E5ECF6",
           "radialaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           }
          },
          "scene": {
           "xaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           },
           "yaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           },
           "zaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           }
          },
          "shapedefaults": {
           "line": {
            "color": "#2a3f5f"
           }
          },
          "ternary": {
           "aaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "baxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "bgcolor": "#E5ECF6",
           "caxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           }
          },
          "title": {
           "x": 0.05
          },
          "xaxis": {
           "automargin": true,
           "gridcolor": "white",
           "linecolor": "white",
           "ticks": "",
           "title": {
            "standoff": 15
           },
           "zerolinecolor": "white",
           "zerolinewidth": 2
          },
          "yaxis": {
           "automargin": true,
           "gridcolor": "white",
           "linecolor": "white",
           "ticks": "",
           "title": {
            "standoff": 15
           },
           "zerolinecolor": "white",
           "zerolinewidth": 2
          }
         }
        },
        "title": {
         "text": "Model Weighted f1-Score Performance"
        },
        "xaxis": {
         "anchor": "y",
         "categoryarray": [
          "GNB",
          "Tree",
          "XGB",
          "Stacked Model"
         ],
         "categoryorder": "array",
         "domain": [
          0,
          1
         ],
         "title": {
          "text": "Machine Learning Model"
         }
        },
        "yaxis": {
         "anchor": "x",
         "domain": [
          0,
          1
         ],
         "title": {
          "text": "Weighted f1-Score"
         }
        }
       }
      },
      "text/html": [
       "<div>                            <div id=\"4cdf09a7-77a3-4c4b-81e1-4e420b918932\" class=\"plotly-graph-div\" style=\"height:525px; width:100%;\"></div>            <script type=\"text/javascript\">                require([\"plotly\"], function(Plotly) {                    window.PLOTLYENV=window.PLOTLYENV || {};                                    if (document.getElementById(\"4cdf09a7-77a3-4c4b-81e1-4e420b918932\")) {                    Plotly.newPlot(                        \"4cdf09a7-77a3-4c4b-81e1-4e420b918932\",                        [{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>\",\"legendgroup\":\"GNB\",\"marker\":{\"color\":\"#636efa\"},\"name\":\"GNB\",\"notched\":false,\"offsetgroup\":\"GNB\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"GNB\",\"GNB\",\"GNB\",\"GNB\",\"GNB\",\"GNB\",\"GNB\",\"GNB\",\"GNB\",\"GNB\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.49021277923226114,0.48816174819698194,0.5016366850721734,0.4890656186573671,0.4937436255807292,0.4931294872578772,0.4949186105771983,0.4991549262764149,0.4883404620898014,0.4896592647365606],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"},{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>\",\"legendgroup\":\"Tree\",\"marker\":{\"color\":\"#EF553B\"},\"name\":\"Tree\",\"notched\":false,\"offsetgroup\":\"Tree\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"Tree\",\"Tree\",\"Tree\",\"Tree\",\"Tree\",\"Tree\",\"Tree\",\"Tree\",\"Tree\",\"Tree\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.49395591647842946,0.4913186019958329,0.49635265212463026,0.4936003617148166,0.49728753381826846,0.4915021325543078,0.4963355793833316,0.4923241383014452,0.4949583200031113,0.49293209331283633],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"},{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>\",\"legendgroup\":\"XGB\",\"marker\":{\"color\":\"#00cc96\"},\"name\":\"XGB\",\"notched\":false,\"offsetgroup\":\"XGB\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"XGB\",\"XGB\",\"XGB\",\"XGB\",\"XGB\",\"XGB\",\"XGB\",\"XGB\",\"XGB\",\"XGB\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.5559134871731306,0.5536083694618409,0.5512097489754343,0.5551107100753496,0.5516901629328173,0.5590246002838755,0.5580970876343704,0.5472750973416809,0.555179334886721,0.5469519120204224],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"},{\"alignmentgroup\":\"True\",\"boxpoints\":\"all\",\"hovertemplate\":\"Machine Learning Model=%{x}<br>Weighted f1-Score=%{y}<extra></extra>\",\"legendgroup\":\"Stacked Model\",\"marker\":{\"color\":\"#ab63fa\"},\"name\":\"Stacked Model\",\"notched\":false,\"offsetgroup\":\"Stacked Model\",\"orientation\":\"v\",\"showlegend\":true,\"x\":[\"Stacked Model\",\"Stacked Model\",\"Stacked Model\",\"Stacked Model\",\"Stacked Model\",\"Stacked Model\",\"Stacked Model\",\"Stacked Model\",\"Stacked Model\",\"Stacked Model\"],\"x0\":\" \",\"xaxis\":\"x\",\"y\":[0.5078471990699265,0.5049234732497797,0.5069081786582854,0.5082227301871328,0.5082061631095763,0.5047138232159096,0.5086445787017143,0.5075823903883498,0.5086454103430422,0.5041613087296903],\"y0\":\" \",\"yaxis\":\"y\",\"type\":\"box\"}],                        {\"template\":{\"data\":{\"histogram2dcontour\":[{\"type\":\"histogram2dcontour\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"choropleth\":[{\"type\":\"choropleth\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}],\"histogram2d\":[{\"type\":\"histogram2d\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"heatmap\":[{\"type\":\"heatmap\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"heatmapgl\":[{\"type\":\"heatmapgl\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"contourcarpet\":[{\"type\":\"contourcarpet\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}],\"contour\":[{\"type\":\"contour\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"surface\":[{\"type\":\"surface\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"mesh3d\":[{\"type\":\"mesh3d\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}],\"scatter\":[{\"fillpattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2},\"type\":\"scatter\"}],\"parcoords\":[{\"type\":\"parcoords\",\"line\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scatterpolargl\":[{\"type\":\"scatterpolargl\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"bar\":[{\"error_x\":{\"color\":\"#2a3f5f\"},\"error_y\":{\"color\":\"#2a3f5f\"},\"marker\":{\"line\":{\"color\":\"#E5ECF6\",\"width\":0.5},\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"bar\"}],\"scattergeo\":[{\"type\":\"scattergeo\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scatterpolar\":[{\"type\":\"scatterpolar\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"histogram\":[{\"marker\":{\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"histogram\"}],\"scattergl\":[{\"type\":\"scattergl\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scatter3d\":[{\"type\":\"scatter3d\",\"line\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scattermapbox\":[{\"type\":\"scattermapbox\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scatterternary\":[{\"type\":\"scatterternary\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scattercarpet\":[{\"type\":\"scattercarpet\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"carpet\":[{\"aaxis\":{\"endlinecolor\":\"#2a3f5f\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"minorgridcolor\":\"white\",\"startlinecolor\":\"#2a3f5f\"},\"baxis\":{\"endlinecolor\":\"#2a3f5f\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"minorgridcolor\":\"white\",\"startlinecolor\":\"#2a3f5f\"},\"type\":\"carpet\"}],\"table\":[{\"cells\":{\"fill\":{\"color\":\"#EBF0F8\"},\"line\":{\"color\":\"white\"}},\"header\":{\"fill\":{\"color\":\"#C8D4E3\"},\"line\":{\"color\":\"white\"}},\"type\":\"table\"}],\"barpolar\":[{\"marker\":{\"line\":{\"color\":\"#E5ECF6\",\"width\":0.5},\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"barpolar\"}],\"pie\":[{\"automargin\":true,\"type\":\"pie\"}]},\"layout\":{\"autotypenumbers\":\"strict\",\"colorway\":[\"#636efa\",\"#EF553B\",\"#00cc96\",\"#ab63fa\",\"#FFA15A\",\"#19d3f3\",\"#FF6692\",\"#B6E880\",\"#FF97FF\",\"#FECB52\"],\"font\":{\"color\":\"#2a3f5f\"},\"hovermode\":\"closest\",\"hoverlabel\":{\"align\":\"left\"},\"paper_bgcolor\":\"white\",\"plot_bgcolor\":\"#E5ECF6\",\"polar\":{\"bgcolor\":\"#E5ECF6\",\"angularaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"radialaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"}},\"ternary\":{\"bgcolor\":\"#E5ECF6\",\"aaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"baxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"caxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"}},\"coloraxis\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"colorscale\":{\"sequential\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"sequentialminus\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"diverging\":[[0,\"#8e0152\"],[0.1,\"#c51b7d\"],[0.2,\"#de77ae\"],[0.3,\"#f1b6da\"],[0.4,\"#fde0ef\"],[0.5,\"#f7f7f7\"],[0.6,\"#e6f5d0\"],[0.7,\"#b8e186\"],[0.8,\"#7fbc41\"],[0.9,\"#4d9221\"],[1,\"#276419\"]]},\"xaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\",\"title\":{\"standoff\":15},\"zerolinecolor\":\"white\",\"automargin\":true,\"zerolinewidth\":2},\"yaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\",\"title\":{\"standoff\":15},\"zerolinecolor\":\"white\",\"automargin\":true,\"zerolinewidth\":2},\"scene\":{\"xaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\",\"gridwidth\":2},\"yaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\",\"gridwidth\":2},\"zaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\",\"gridwidth\":2}},\"shapedefaults\":{\"line\":{\"color\":\"#2a3f5f\"}},\"annotationdefaults\":{\"arrowcolor\":\"#2a3f5f\",\"arrowhead\":0,\"arrowwidth\":1},\"geo\":{\"bgcolor\":\"white\",\"landcolor\":\"#E5ECF6\",\"subunitcolor\":\"white\",\"showland\":true,\"showlakes\":true,\"lakecolor\":\"white\"},\"title\":{\"x\":0.05},\"mapbox\":{\"style\":\"light\"}}},\"xaxis\":{\"anchor\":\"y\",\"domain\":[0.0,1.0],\"title\":{\"text\":\"Machine Learning Model\"},\"categoryorder\":\"array\",\"categoryarray\":[\"GNB\",\"Tree\",\"XGB\",\"Stacked Model\"]},\"yaxis\":{\"anchor\":\"x\",\"domain\":[0.0,1.0],\"title\":{\"text\":\"Weighted f1-Score\"}},\"legend\":{\"title\":{\"text\":\"Machine Learning Model\"},\"tracegroupgap\":0},\"title\":{\"text\":\"Model Weighted f1-Score Performance\"},\"boxmode\":\"overlay\"},                        {\"responsive\": true}                    ).then(function(){\n",
       "                            \n",
       "var gd = document.getElementById('4cdf09a7-77a3-4c4b-81e1-4e420b918932');\n",
       "var x = new MutationObserver(function (mutations, observer) {{\n",
       "        var display = window.getComputedStyle(gd).display;\n",
       "        if (!display || display === 'none') {{\n",
       "            console.log([gd, 'removed!']);\n",
       "            Plotly.purge(gd);\n",
       "            observer.disconnect();\n",
       "        }}\n",
       "}});\n",
       "\n",
       "// Listen for the removal of the full notebook cells\n",
       "var notebookContainer = gd.closest('#notebook-container');\n",
       "if (notebookContainer) {{\n",
       "    x.observe(notebookContainer, {childList: true});\n",
       "}}\n",
       "\n",
       "// Listen for the clearing of the current output cell\n",
       "var outputEl = gd.closest('.output');\n",
       "if (outputEl) {{\n",
       "    x.observe(outputEl, {childList: true});\n",
       "}}\n",
       "\n",
       "                        })                };                });            </script>        </div>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = px.box(\n",
    "    classifmod1,\n",
    "    x=\"Model\",\n",
    "    y=\"Weighted_F1\",\n",
    "    color=\"Model\",\n",
    "    points=\"all\",\n",
    "    labels={\"Model\": \"Machine Learning Model\", \"Weighted_F1\": \"Weighted f1-Score\"},\n",
    "    title=\"Model Weighted f1-Score Performance\",\n",
    ")\n",
    "fig.show()\n",
    "fig.write_html(\"products/plots/stacked_train_weighted_f1.html\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Pickle Stacked Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test score: 61.12 %\n"
     ]
    }
   ],
   "source": [
    "import pickle\n",
    "\n",
    "level0 = list()\n",
    "level0.append((\"GNB\", GaussianNB()))\n",
    "level0.append((\"Tree\", DecisionTreeClassifier()))\n",
    "level1 = XGBClassifier(verbosity=0)\n",
    "model = StackingClassifier(estimators=level0, final_estimator=level1, cv=5)\n",
    "model.fit(X_train, y_train)\n",
    "\n",
    "# Save to file in the current working directory\n",
    "pkl_filename = \"products/pickle_files/stacked.pkl\"\n",
    "with open(pkl_filename, \"wb\") as file:\n",
    "    pickle.dump(model, file)\n",
    "\n",
    "# Load from file\n",
    "with open(pkl_filename, \"rb\") as file:\n",
    "    pickle_model = pickle.load(file)\n",
    "    score = pickle_model.score(X_test, y_test)\n",
    "    print(\"Test score: {0:.2f} %\".format(100 * score))\n",
    "    Y_predict = pickle_model.predict(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Test Set Confusion Matrix for Stacked Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import ConfusionMatrixDisplay\n",
    "from seaborn import set_palette\n",
    "\n",
    "conf_matrix = confusion_matrix(y_test, Y_predict)\n",
    "disp = ConfusionMatrixDisplay(\n",
    "    confusion_matrix=conf_matrix,\n",
    "    display_labels=[\"Extras\", \"Dot\", \"Runs\", \"Wicket\"],\n",
    ")\n",
    "\n",
    "# save the plot\n",
    "\n",
    "disp.plot(cmap=\"YlGnBu\", values_format=\"d\", ax=None)\n",
    "plt.title(\n",
    "    \"Test Set Confusion Matrix for Stacked Classifier\",\n",
    "    fontsize=15,\n",
    "    color=\"#336699\",\n",
    "    loc=\"center\",\n",
    ")\n",
    "plt.savefig(\"products/plots/confusion_matrix_stacked.png\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# XGBoost, Naive Bayes, Tree Pickle Files for Deployment"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preprocess Training Data Only (No Test Data Split)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "# split the dataset into train and test sets by season\n",
    "train = df\n",
    "\n",
    "# drop `season` column from train and test sets\n",
    "train.drop(labels=\"season\", axis=1, inplace=True)\n",
    "\n",
    "encoded_df_train = pd.get_dummies(\n",
    "    data=train,\n",
    "    columns=[\n",
    "        \"match_type\",\n",
    "        \"venue\",\n",
    "        \"inning\",\n",
    "        \"batting_team\",\n",
    "        \"bowling_team\",\n",
    "        \"batsman\",\n",
    "        \"non_striker\",\n",
    "        \"bowler_type\",\n",
    "        \"ball_length\",\n",
    "    ],\n",
    ")\n",
    "\n",
    "encoded_df_train = encoded_df_train[\n",
    "    [\n",
    "        \"ball\",\n",
    "        \"wickets\",\n",
    "        \"cumulative_runs\",\n",
    "        \"runs_left_to_win\",\n",
    "        \"match_type_playoffs\",\n",
    "        \"match_type_round-robin\",\n",
    "        \"venue_Ahmedabad\",\n",
    "        \"venue_Bangalore\",\n",
    "        \"venue_Chennai\",\n",
    "        \"venue_Cuttack\",\n",
    "        \"venue_Delhi\",\n",
    "        \"venue_Dharamsala\",\n",
    "        \"venue_Hyderabad\",\n",
    "        \"venue_Indore\",\n",
    "        \"venue_Kolkata\",\n",
    "        \"venue_Mohali\",\n",
    "        \"venue_Mumbai\",\n",
    "        \"venue_Nagpur\",\n",
    "        \"venue_Pune\",\n",
    "        \"venue_Raipur\",\n",
    "        \"venue_Rajasthan\",\n",
    "        \"venue_Ranchi\",\n",
    "        \"venue_Visakhapatnam\",\n",
    "        \"inning_1\",\n",
    "        \"inning_2\",\n",
    "        \"batting_team_Chennai Super Kings\",\n",
    "        \"batting_team_Delhi Capitals\",\n",
    "        \"batting_team_Kings XI Punjab\",\n",
    "        \"batting_team_Kolkata Knight Riders\",\n",
    "        \"batting_team_Mumbai Indians\",\n",
    "        \"batting_team_Rajasthan Royals\",\n",
    "        \"batting_team_Royal Challengers Bangalore\",\n",
    "        \"batting_team_Sunrisers Hyderabad\",\n",
    "        \"bowling_team_Chennai Super Kings\",\n",
    "        \"bowling_team_Delhi Capitals\",\n",
    "        \"bowling_team_Kings XI Punjab\",\n",
    "        \"bowling_team_Kolkata Knight Riders\",\n",
    "        \"bowling_team_Mumbai Indians\",\n",
    "        \"bowling_team_Rajasthan Royals\",\n",
    "        \"bowling_team_Royal Challengers Bangalore\",\n",
    "        \"bowling_team_Sunrisers Hyderabad\",\n",
    "        \"batsman_top\",\n",
    "        \"batsman_middle\",\n",
    "        \"batsman_tail\",\n",
    "        \"non_striker_middle\",\n",
    "        \"non_striker_tail\",\n",
    "        \"non_striker_top\",\n",
    "        \"bowler_type_pacer\",\n",
    "        \"bowler_type_spinner\",\n",
    "        \"ball_length_Unknown\",\n",
    "        \"ball_length_full\",\n",
    "        \"ball_length_full toss\",\n",
    "        \"ball_length_good\",\n",
    "        \"ball_length_short\",\n",
    "        \"ball_length_yorker\",\n",
    "        \"event\",\n",
    "    ]\n",
    "]\n",
    "\n",
    "X_train = encoded_df_train.iloc[:, :-1].values\n",
    "\n",
    "y_train = encoded_df_train.iloc[:, -1].values\n",
    "\n",
    "encoder = LabelEncoder()\n",
    "\n",
    "y_train = encoder.fit_transform(y_train)\n",
    "\n",
    "y_train = y_train.reshape(\n",
    "    len(y_train),\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Perform `RepeatedKFold` Classification Routine"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|          | 0/3 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">Tree Accuracy: 0.492 Weighted F1: 0.496\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 67%|██████▋   | 2/3 [00:07<00:03,  3.40s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">GNB Accuracy: 0.485 Weighted F1: 0.496\n",
      ">XGB Accuracy: 0.624 Weighted F1: 0.555\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 3/3 [00:29<00:00,  9.76s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 12.2 s, sys: 4.49 s, total: 16.7 s\n",
      "Wall time: 29.4 s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "from sklearn.exceptions import ConvergenceWarning\n",
    "from sklearn.exceptions import DataConversionWarning\n",
    "from tqdm import tqdm  # Import tqdm\n",
    "# Assuming other necessary imports like LogisticRegression, DecisionTreeClassifier, etc., are already made\n",
    "\n",
    "ConvergenceWarning('ignore')\n",
    "warnings.filterwarnings(action='ignore', category=DataConversionWarning)\n",
    "\n",
    "def base_models():\n",
    "    models = dict()\n",
    "    models[\"Tree\"] = DecisionTreeClassifier()\n",
    "    models[\"GNB\"] = GaussianNB()\n",
    "    models[\"XGB\"] = XGBClassifier(verbosity=0)\n",
    "    return models\n",
    "\n",
    "# Function to evaluate the list of models\n",
    "def eval_models(model):\n",
    "    cv = RepeatedKFold(n_splits=5, n_repeats=2, random_state=1)\n",
    "    accuracy = cross_val_score(model, X_train, y_train, scoring='accuracy', cv=cv, n_jobs=-1, error_score='raise')\n",
    "    weighted_f1 = cross_val_score(model, X_train, y_train, scoring='f1_weighted', cv=cv, n_jobs=-1, error_score='raise')\n",
    "    return accuracy, weighted_f1\n",
    "\n",
    "models = base_models()\n",
    "\n",
    "# evaluate the models and store results\n",
    "results = {}\n",
    "names = list()\n",
    "accuracies = []\n",
    "weighted_f1s = []\n",
    "\n",
    "final_models = {}  # Create a dictionary to store the fitted models\n",
    "\n",
    "for name, model in tqdm(models.items()):  # Wrap models.items() with tqdm to get a progress bar\n",
    "    accuracy, weighted_f1 = eval_models(model)\n",
    "    accuracies.extend(accuracy)\n",
    "    weighted_f1s.extend(weighted_f1)\n",
    "    names.extend([name] * len(accuracy))\n",
    "    print('>%s Accuracy: %.3f Weighted F1: %.3f' % (name, accuracy.mean(), weighted_f1.mean()))\n",
    "\n",
    "    # Fit the model with the training data\n",
    "    fitted_model = model.fit(X_train, y_train)\n",
    "\n",
    "    # Save the fitted model as a pickle file\n",
    "    joblib_filename = f\"products/pickle_files/{name}_deploy.joblib\"\n",
    "    with open(joblib_filename, 'wb') as file:\n",
    "        joblib.dump(fitted_model, file)\n",
    "\n",
    "# Store the fitted model in the final_models dictionary\n",
    "    final_models[name] = fitted_model"
   ]
  }
 ],
 "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": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}