bs_ml_training.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Training Machine Learning Models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pycaret.classification import *\n",
    "from sklearn.model_selection import train_test_split\n",
    "import pandas as pd\n",
    "import json"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 278987 entries, 0 to 278986\n",
      "Data columns (total 58 columns):\n",
      " #   Column                                 Non-Null Count   Dtype         \n",
      "---  ------                                 --------------   -----         \n",
      " 0   battleTime                             278987 non-null  datetime64[ns]\n",
      " 1   playertag                              278987 non-null  object        \n",
      " 2   event_mode                             278987 non-null  object        \n",
      " 3   event_map                              278987 non-null  object        \n",
      " 4   battle_type                            278987 non-null  object        \n",
      " 5   battle_result                          278987 non-null  object        \n",
      " 6   battle_duration                        278987 non-null  int16         \n",
      " 7   battle_trophyChange                    259972 non-null  Float64       \n",
      " 8   battle_team1_player1_tag               278987 non-null  object        \n",
      " 9   battle_team1_player1_name              278987 non-null  object        \n",
      " 10  battle_team1_player1_brawler_id        278987 non-null  Float64       \n",
      " 11  battle_team1_player1_brawler_name      278987 non-null  object        \n",
      " 12  battle_team1_player1_brawler_power     278987 non-null  Float64       \n",
      " 13  battle_team1_player1_brawler_trophies  278987 non-null  Float64       \n",
      " 14  battle_team1_player2_tag               278987 non-null  object        \n",
      " 15  battle_team1_player2_name              278987 non-null  object        \n",
      " 16  battle_team1_player2_brawler_id        278987 non-null  Float64       \n",
      " 17  battle_team1_player2_brawler_name      278987 non-null  object        \n",
      " 18  battle_team1_player2_brawler_power     278987 non-null  Float64       \n",
      " 19  battle_team1_player2_brawler_trophies  278987 non-null  Float64       \n",
      " 20  battle_team1_player3_tag               278987 non-null  object        \n",
      " 21  battle_team1_player3_name              278987 non-null  object        \n",
      " 22  battle_team1_player3_brawler_id        278987 non-null  Float64       \n",
      " 23  battle_team1_player3_brawler_name      278987 non-null  object        \n",
      " 24  battle_team1_player3_brawler_power     278987 non-null  Float64       \n",
      " 25  battle_team1_player3_brawler_trophies  278987 non-null  Float64       \n",
      " 26  battle_team2_player1_tag               278987 non-null  object        \n",
      " 27  battle_team2_player1_name              278987 non-null  object        \n",
      " 28  battle_team2_player1_brawler_id        278987 non-null  Float64       \n",
      " 29  battle_team2_player1_brawler_name      278987 non-null  object        \n",
      " 30  battle_team2_player1_brawler_power     278987 non-null  Float64       \n",
      " 31  battle_team2_player1_brawler_trophies  278987 non-null  Float64       \n",
      " 32  battle_team2_player2_tag               278987 non-null  object        \n",
      " 33  battle_team2_player2_name              278987 non-null  object        \n",
      " 34  battle_team2_player2_brawler_id        278987 non-null  Float64       \n",
      " 35  battle_team2_player2_brawler_name      278987 non-null  object        \n",
      " 36  battle_team2_player2_brawler_power     278987 non-null  Float64       \n",
      " 37  battle_team2_player2_brawler_trophies  278987 non-null  Float64       \n",
      " 38  battle_team2_player3_tag               278987 non-null  object        \n",
      " 39  battle_team2_player3_name              278987 non-null  object        \n",
      " 40  battle_team2_player3_brawler_id        278987 non-null  Float64       \n",
      " 41  battle_team2_player3_brawler_name      278987 non-null  object        \n",
      " 42  battle_team2_player3_brawler_power     278987 non-null  Float64       \n",
      " 43  battle_team2_player3_brawler_trophies  278987 non-null  Float64       \n",
      " 44  player_team                            278987 non-null  int64         \n",
      " 45  winner_team                            278987 non-null  int64         \n",
      " 46  avg_brawler_trophies_team1             278987 non-null  float64       \n",
      " 47  avg_brawler_trophies_team2             278987 non-null  float64       \n",
      " 48  min_brawler_trophies_team1             278987 non-null  float64       \n",
      " 49  max_brawler_trophies_team1             278987 non-null  float64       \n",
      " 50  min_brawler_trophies_team2             278987 non-null  float64       \n",
      " 51  max_brawler_trophies_team2             278987 non-null  float64       \n",
      " 52  battle_power_team1                     278987 non-null  Float64       \n",
      " 53  battle_power_team2                     278987 non-null  Float64       \n",
      " 54  battle_power_diff                      278987 non-null  Float64       \n",
      " 55  avg_brawler_trophies_diff              278987 non-null  float64       \n",
      " 56  min_brawler_trophies_diff              278987 non-null  float64       \n",
      " 57  max_brawler_trophies_diff              278987 non-null  float64       \n",
      "dtypes: Float64(22), datetime64[ns](1), float64(9), int16(1), int64(2), object(23)\n",
      "memory usage: 127.7+ MB\n"
     ]
    }
   ],
   "source": [
    "df = pd.read_parquet('datasets/teams/battlelog_train.parquet')\n",
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [],
   "source": [
    "# subset columns by feature importance\n",
    "\n",
    "df = df[[\n",
    "\t'event_mode',\n",
    "\t'event_map',\n",
    "\t'battle_team1_player1_brawler_name',\n",
    "\t'battle_team1_player2_brawler_name',\n",
    "\t'battle_team1_player3_brawler_name',\n",
    "\t'battle_team2_player1_brawler_name',\n",
    "\t'battle_team2_player2_brawler_name',\n",
    "\t'battle_team2_player3_brawler_name',\n",
    "\t'avg_brawler_trophies_diff',\n",
    "\t# 'avg_highestTrophies_diff',\n",
    "\t# 'avg_trophies_diff',\n",
    "\t# 'avg_team_victories_diff',\n",
    "\t# 'avg_expPoints_diff',\n",
    "\t'max_brawler_trophies_diff',\n",
    "\t# 'max_highestTrophies_diff',\n",
    "\t# 'max_trophies_diff',\n",
    "\t# 'max_team_victories_diff',\n",
    "\t# 'max_expPoints_diff',\n",
    "\t'min_brawler_trophies_diff',\n",
    "\t# 'min_highestTrophies_diff',\n",
    "\t# 'min_trophies_diff',\n",
    "\t# 'min_team_victories_diff',\n",
    "\t# 'min_expPoints_diff',\n",
    "\t'battle_power_diff',\n",
    "\t'winner_team',\n",
    "]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "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>event_mode</th>\n",
       "      <th>event_map</th>\n",
       "      <th>avg_brawler_trophies_diff</th>\n",
       "      <th>max_brawler_trophies_diff</th>\n",
       "      <th>min_brawler_trophies_diff</th>\n",
       "      <th>battle_power_diff</th>\n",
       "      <th>winner_team</th>\n",
       "      <th>T1_8-BIT</th>\n",
       "      <th>T1_AMBER</th>\n",
       "      <th>T1_ASH</th>\n",
       "      <th>...</th>\n",
       "      <th>T2_SANDY</th>\n",
       "      <th>T2_SHELLY</th>\n",
       "      <th>T2_SPIKE</th>\n",
       "      <th>T2_SPROUT</th>\n",
       "      <th>T2_SQUEAK</th>\n",
       "      <th>T2_STU</th>\n",
       "      <th>T2_SURGE</th>\n",
       "      <th>T2_TARA</th>\n",
       "      <th>T2_TICK</th>\n",
       "      <th>T2_WILLOW</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>brawlBall</td>\n",
       "      <td>Super Beach</td>\n",
       "      <td>0.147371</td>\n",
       "      <td>0.105546</td>\n",
       "      <td>0.186610</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>brawlBall</td>\n",
       "      <td>Super Beach</td>\n",
       "      <td>-0.078622</td>\n",
       "      <td>-0.157831</td>\n",
       "      <td>0.012148</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>brawlBall</td>\n",
       "      <td>Super Beach</td>\n",
       "      <td>0.030869</td>\n",
       "      <td>0.052697</td>\n",
       "      <td>-0.020846</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>brawlBall</td>\n",
       "      <td>Super Beach</td>\n",
       "      <td>-0.007010</td>\n",
       "      <td>-0.018387</td>\n",
       "      <td>0.028386</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>brawlBall</td>\n",
       "      <td>Super Beach</td>\n",
       "      <td>0.068858</td>\n",
       "      <td>0.038340</td>\n",
       "      <td>0.106025</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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>129185</th>\n",
       "      <td>knockout</td>\n",
       "      <td>New Perspective</td>\n",
       "      <td>-0.034374</td>\n",
       "      <td>-0.031272</td>\n",
       "      <td>-0.016926</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>129186</th>\n",
       "      <td>heist</td>\n",
       "      <td>Snaked Assault</td>\n",
       "      <td>0.003576</td>\n",
       "      <td>0.010302</td>\n",
       "      <td>0.009662</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>129187</th>\n",
       "      <td>heist</td>\n",
       "      <td>Snaked Assault</td>\n",
       "      <td>0.008475</td>\n",
       "      <td>-0.028988</td>\n",
       "      <td>0.059325</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>129188</th>\n",
       "      <td>knockout</td>\n",
       "      <td>Riverside</td>\n",
       "      <td>0.011111</td>\n",
       "      <td>-0.005038</td>\n",
       "      <td>0.050325</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>129189</th>\n",
       "      <td>knockout</td>\n",
       "      <td>Riverside</td>\n",
       "      <td>-0.003571</td>\n",
       "      <td>0.247011</td>\n",
       "      <td>-0.674219</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>129190 rows × 143 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       event_mode        event_map  avg_brawler_trophies_diff  \\\n",
       "0       brawlBall      Super Beach                   0.147371   \n",
       "1       brawlBall      Super Beach                  -0.078622   \n",
       "2       brawlBall      Super Beach                   0.030869   \n",
       "3       brawlBall      Super Beach                  -0.007010   \n",
       "4       brawlBall      Super Beach                   0.068858   \n",
       "...           ...              ...                        ...   \n",
       "129185   knockout  New Perspective                  -0.034374   \n",
       "129186      heist   Snaked Assault                   0.003576   \n",
       "129187      heist   Snaked Assault                   0.008475   \n",
       "129188   knockout        Riverside                   0.011111   \n",
       "129189   knockout        Riverside                  -0.003571   \n",
       "\n",
       "        max_brawler_trophies_diff  min_brawler_trophies_diff  \\\n",
       "0                        0.105546                   0.186610   \n",
       "1                       -0.157831                   0.012148   \n",
       "2                        0.052697                  -0.020846   \n",
       "3                       -0.018387                   0.028386   \n",
       "4                        0.038340                   0.106025   \n",
       "...                           ...                        ...   \n",
       "129185                  -0.031272                  -0.016926   \n",
       "129186                   0.010302                   0.009662   \n",
       "129187                  -0.028988                   0.059325   \n",
       "129188                  -0.005038                   0.050325   \n",
       "129189                   0.247011                  -0.674219   \n",
       "\n",
       "        battle_power_diff  winner_team  T1_8-BIT  T1_AMBER  T1_ASH  ...  \\\n",
       "0                     0.0            1         1         0       0  ...   \n",
       "1                     0.0            1         0         0       0  ...   \n",
       "2                     0.0            2         1         0       0  ...   \n",
       "3                     0.0            1         1         0       0  ...   \n",
       "4                     0.0            2         0         0       0  ...   \n",
       "...                   ...          ...       ...       ...     ...  ...   \n",
       "129185               -1.0            1         0         0       0  ...   \n",
       "129186               -1.0            2         0         1       0  ...   \n",
       "129187               -1.0            2         1         0       0  ...   \n",
       "129188               -1.0            2         0         0       0  ...   \n",
       "129189                1.0            2         0         0       0  ...   \n",
       "\n",
       "        T2_SANDY  T2_SHELLY  T2_SPIKE  T2_SPROUT  T2_SQUEAK  T2_STU  T2_SURGE  \\\n",
       "0              0          0         0          0          0       0         0   \n",
       "1              0          0         0          0          0       0         0   \n",
       "2              0          0         1          0          0       0         0   \n",
       "3              0          0         0          0          0       0         0   \n",
       "4              0          0         0          0          0       0         0   \n",
       "...          ...        ...       ...        ...        ...     ...       ...   \n",
       "129185         0          0         0          0          0       0         0   \n",
       "129186         0          0         0          0          0       0         0   \n",
       "129187         0          0         0          0          0       0         0   \n",
       "129188         0          0         0          0          0       0         0   \n",
       "129189         0          0         0          0          0       0         0   \n",
       "\n",
       "        T2_TARA  T2_TICK  T2_WILLOW  \n",
       "0             0        0          0  \n",
       "1             0        0          0  \n",
       "2             0        0          0  \n",
       "3             1        0          0  \n",
       "4             0        0          0  \n",
       "...         ...      ...        ...  \n",
       "129185        0        0          0  \n",
       "129186        0        0          0  \n",
       "129187        0        0          0  \n",
       "129188        0        0          0  \n",
       "129189        0        0          0  \n",
       "\n",
       "[129190 rows x 143 columns]"
      ]
     },
     "execution_count": 104,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# dummy brawler name\n",
    "dft1 = pd.get_dummies(data=df['battle_team1_player1_brawler_name'], prefix='T1') + \\\n",
    "\tpd.get_dummies(data=df['battle_team1_player2_brawler_name'], prefix='T1') + \\\n",
    "\tpd.get_dummies(data=df['battle_team1_player3_brawler_name'], prefix='T1')\n",
    "\n",
    "dft2 = pd.get_dummies(data=df['battle_team2_player1_brawler_name'], prefix='T2') + \\\n",
    "\tpd.get_dummies(data=df['battle_team2_player2_brawler_name'], prefix='T2') + \\\n",
    "\tpd.get_dummies(data=df['battle_team2_player3_brawler_name'], prefix='T2')\n",
    "\n",
    "df = pd.concat([df, dft1, dft2], axis=1)\n",
    "df = df.drop(columns=[\n",
    "\t'battle_team1_player1_brawler_name',\n",
    "\t'battle_team1_player2_brawler_name',\n",
    "\t'battle_team1_player3_brawler_name',\n",
    "\t'battle_team2_player1_brawler_name',\n",
    "\t'battle_team2_player2_brawler_name',\n",
    "\t'battle_team2_player3_brawler_name',\n",
    "])\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [],
   "source": [
    "# subset by gamemode\n",
    "event_mode = 'brawlBall'\n",
    "df = df[df['event_mode'] == event_mode]\n",
    "df = df.drop(columns=['event_mode'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train:  (62905, 142)\n",
      "test:  (20969, 142)\n"
     ]
    }
   ],
   "source": [
    "def split_data(data, test_size, random_state):\n",
    "\t\"\"\"Funcion para dividir el dataset en train y test\"\"\"\n",
    "\ttrain, test = train_test_split(\n",
    "\t\t\tdata,\n",
    "\t\t\ttest_size=test_size,\n",
    "\t\t\trandom_state=random_state\n",
    "\t\t\t)\n",
    "\t\t\n",
    "\tprint('train: ', train.shape)\n",
    "\tprint('test: ', test.shape)\n",
    "\n",
    "\treturn train, test\n",
    "\n",
    "seed=14697\n",
    "\n",
    "train, test = split_data(df, test_size = 0.25, random_state=seed)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_b84a1_row10_col1 {\n",
       "  background-color: lightgreen;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_b84a1\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_b84a1_level0_col0\" class=\"col_heading level0 col0\" >Description</th>\n",
       "      <th id=\"T_b84a1_level0_col1\" class=\"col_heading level0 col1\" >Value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_b84a1_level0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "      <td id=\"T_b84a1_row0_col0\" class=\"data row0 col0\" >Session id</td>\n",
       "      <td id=\"T_b84a1_row0_col1\" class=\"data row0 col1\" >2557</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b84a1_level0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "      <td id=\"T_b84a1_row1_col0\" class=\"data row1 col0\" >Target</td>\n",
       "      <td id=\"T_b84a1_row1_col1\" class=\"data row1 col1\" >winner_team</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b84a1_level0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "      <td id=\"T_b84a1_row2_col0\" class=\"data row2 col0\" >Target type</td>\n",
       "      <td id=\"T_b84a1_row2_col1\" class=\"data row2 col1\" >Binary</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b84a1_level0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
       "      <td id=\"T_b84a1_row3_col0\" class=\"data row3 col0\" >Target mapping</td>\n",
       "      <td id=\"T_b84a1_row3_col1\" class=\"data row3 col1\" >1: 0, 2: 1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b84a1_level0_row4\" class=\"row_heading level0 row4\" >4</th>\n",
       "      <td id=\"T_b84a1_row4_col0\" class=\"data row4 col0\" >Original data shape</td>\n",
       "      <td id=\"T_b84a1_row4_col1\" class=\"data row4 col1\" >(62905, 142)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b84a1_level0_row5\" class=\"row_heading level0 row5\" >5</th>\n",
       "      <td id=\"T_b84a1_row5_col0\" class=\"data row5 col0\" >Transformed data shape</td>\n",
       "      <td id=\"T_b84a1_row5_col1\" class=\"data row5 col1\" >(62905, 146)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b84a1_level0_row6\" class=\"row_heading level0 row6\" >6</th>\n",
       "      <td id=\"T_b84a1_row6_col0\" class=\"data row6 col0\" >Transformed train set shape</td>\n",
       "      <td id=\"T_b84a1_row6_col1\" class=\"data row6 col1\" >(44033, 146)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b84a1_level0_row7\" class=\"row_heading level0 row7\" >7</th>\n",
       "      <td id=\"T_b84a1_row7_col0\" class=\"data row7 col0\" >Transformed test set shape</td>\n",
       "      <td id=\"T_b84a1_row7_col1\" class=\"data row7 col1\" >(18872, 146)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b84a1_level0_row8\" class=\"row_heading level0 row8\" >8</th>\n",
       "      <td id=\"T_b84a1_row8_col0\" class=\"data row8 col0\" >Numeric features</td>\n",
       "      <td id=\"T_b84a1_row8_col1\" class=\"data row8 col1\" >140</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b84a1_level0_row9\" class=\"row_heading level0 row9\" >9</th>\n",
       "      <td id=\"T_b84a1_row9_col0\" class=\"data row9 col0\" >Categorical features</td>\n",
       "      <td id=\"T_b84a1_row9_col1\" class=\"data row9 col1\" >1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b84a1_level0_row10\" class=\"row_heading level0 row10\" >10</th>\n",
       "      <td id=\"T_b84a1_row10_col0\" class=\"data row10 col0\" >Preprocess</td>\n",
       "      <td id=\"T_b84a1_row10_col1\" class=\"data row10 col1\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b84a1_level0_row11\" class=\"row_heading level0 row11\" >11</th>\n",
       "      <td id=\"T_b84a1_row11_col0\" class=\"data row11 col0\" >Imputation type</td>\n",
       "      <td id=\"T_b84a1_row11_col1\" class=\"data row11 col1\" >simple</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b84a1_level0_row12\" class=\"row_heading level0 row12\" >12</th>\n",
       "      <td id=\"T_b84a1_row12_col0\" class=\"data row12 col0\" >Numeric imputation</td>\n",
       "      <td id=\"T_b84a1_row12_col1\" class=\"data row12 col1\" >mean</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b84a1_level0_row13\" class=\"row_heading level0 row13\" >13</th>\n",
       "      <td id=\"T_b84a1_row13_col0\" class=\"data row13 col0\" >Categorical imputation</td>\n",
       "      <td id=\"T_b84a1_row13_col1\" class=\"data row13 col1\" >mode</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b84a1_level0_row14\" class=\"row_heading level0 row14\" >14</th>\n",
       "      <td id=\"T_b84a1_row14_col0\" class=\"data row14 col0\" >Maximum one-hot encoding</td>\n",
       "      <td id=\"T_b84a1_row14_col1\" class=\"data row14 col1\" >500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b84a1_level0_row15\" class=\"row_heading level0 row15\" >15</th>\n",
       "      <td id=\"T_b84a1_row15_col0\" class=\"data row15 col0\" >Encoding method</td>\n",
       "      <td id=\"T_b84a1_row15_col1\" class=\"data row15 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b84a1_level0_row16\" class=\"row_heading level0 row16\" >16</th>\n",
       "      <td id=\"T_b84a1_row16_col0\" class=\"data row16 col0\" >Fold Generator</td>\n",
       "      <td id=\"T_b84a1_row16_col1\" class=\"data row16 col1\" >StratifiedKFold</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b84a1_level0_row17\" class=\"row_heading level0 row17\" >17</th>\n",
       "      <td id=\"T_b84a1_row17_col0\" class=\"data row17 col0\" >Fold Number</td>\n",
       "      <td id=\"T_b84a1_row17_col1\" class=\"data row17 col1\" >10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b84a1_level0_row18\" class=\"row_heading level0 row18\" >18</th>\n",
       "      <td id=\"T_b84a1_row18_col0\" class=\"data row18 col0\" >CPU Jobs</td>\n",
       "      <td id=\"T_b84a1_row18_col1\" class=\"data row18 col1\" >-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b84a1_level0_row19\" class=\"row_heading level0 row19\" >19</th>\n",
       "      <td id=\"T_b84a1_row19_col0\" class=\"data row19 col0\" >Use GPU</td>\n",
       "      <td id=\"T_b84a1_row19_col1\" class=\"data row19 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b84a1_level0_row20\" class=\"row_heading level0 row20\" >20</th>\n",
       "      <td id=\"T_b84a1_row20_col0\" class=\"data row20 col0\" >Log Experiment</td>\n",
       "      <td id=\"T_b84a1_row20_col1\" class=\"data row20 col1\" >MlflowLogger</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b84a1_level0_row21\" class=\"row_heading level0 row21\" >21</th>\n",
       "      <td id=\"T_b84a1_row21_col0\" class=\"data row21 col0\" >Experiment Name</td>\n",
       "      <td id=\"T_b84a1_row21_col1\" class=\"data row21 col1\" >clf-default-name</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b84a1_level0_row22\" class=\"row_heading level0 row22\" >22</th>\n",
       "      <td id=\"T_b84a1_row22_col0\" class=\"data row22 col0\" >USI</td>\n",
       "      <td id=\"T_b84a1_row22_col1\" class=\"data row22 col1\" >5b33</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x162327a17e0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# setup model\n",
    "session_1 = setup(\n",
    "\tdata = train,\n",
    "\ttarget = 'winner_team',\n",
    "\t# fix_imbalance = True,\n",
    "    # feature_selection= True,\n",
    "\t# remove_outliers=True,\n",
    "\tlog_experiment = True,\n",
    "    use_gpu=False,\n",
    "    max_encoding_ohe=500,\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_310fe th {\n",
       "  text-align: left;\n",
       "}\n",
       "#T_310fe_row0_col0, #T_310fe_row0_col4, #T_310fe_row1_col0, #T_310fe_row1_col1, #T_310fe_row1_col2, #T_310fe_row1_col3, #T_310fe_row1_col5, #T_310fe_row1_col6, #T_310fe_row1_col7, #T_310fe_row2_col0, #T_310fe_row2_col1, #T_310fe_row2_col2, #T_310fe_row2_col3, #T_310fe_row2_col4, #T_310fe_row2_col5, #T_310fe_row2_col6, #T_310fe_row2_col7, #T_310fe_row3_col0, #T_310fe_row3_col1, #T_310fe_row3_col2, #T_310fe_row3_col3, #T_310fe_row3_col4, #T_310fe_row3_col5, #T_310fe_row3_col6, #T_310fe_row3_col7, #T_310fe_row4_col0, #T_310fe_row4_col1, #T_310fe_row4_col2, #T_310fe_row4_col3, #T_310fe_row4_col4, #T_310fe_row4_col5, #T_310fe_row4_col6, #T_310fe_row4_col7, #T_310fe_row5_col0, #T_310fe_row5_col1, #T_310fe_row5_col2, #T_310fe_row5_col3, #T_310fe_row5_col4, #T_310fe_row5_col5, #T_310fe_row5_col6, #T_310fe_row5_col7, #T_310fe_row6_col0, #T_310fe_row6_col1, #T_310fe_row6_col2, #T_310fe_row6_col3, #T_310fe_row6_col4, #T_310fe_row6_col5, #T_310fe_row6_col6, #T_310fe_row6_col7, #T_310fe_row7_col0, #T_310fe_row7_col1, #T_310fe_row7_col2, #T_310fe_row7_col3, #T_310fe_row7_col4, #T_310fe_row7_col5, #T_310fe_row7_col6, #T_310fe_row7_col7, #T_310fe_row8_col0, #T_310fe_row8_col1, #T_310fe_row8_col2, #T_310fe_row8_col3, #T_310fe_row8_col4, #T_310fe_row8_col5, #T_310fe_row8_col6, #T_310fe_row8_col7, #T_310fe_row9_col0, #T_310fe_row9_col1, #T_310fe_row9_col2, #T_310fe_row9_col3, #T_310fe_row9_col4, #T_310fe_row9_col5, #T_310fe_row9_col6, #T_310fe_row9_col7, #T_310fe_row10_col0, #T_310fe_row10_col1, #T_310fe_row10_col2, #T_310fe_row10_col3, #T_310fe_row10_col4, #T_310fe_row10_col5, #T_310fe_row10_col6, #T_310fe_row10_col7, #T_310fe_row11_col0, #T_310fe_row11_col1, #T_310fe_row11_col2, #T_310fe_row11_col3, #T_310fe_row11_col4, #T_310fe_row11_col5, #T_310fe_row11_col6, #T_310fe_row11_col7, #T_310fe_row12_col0, #T_310fe_row12_col1, #T_310fe_row12_col2, #T_310fe_row12_col3, #T_310fe_row12_col4, #T_310fe_row12_col5, #T_310fe_row12_col6, #T_310fe_row12_col7, #T_310fe_row13_col0, #T_310fe_row13_col1, #T_310fe_row13_col2, #T_310fe_row13_col3, #T_310fe_row13_col4, #T_310fe_row13_col5, #T_310fe_row13_col6, #T_310fe_row13_col7 {\n",
       "  text-align: left;\n",
       "}\n",
       "#T_310fe_row0_col1, #T_310fe_row0_col2, #T_310fe_row0_col3, #T_310fe_row0_col5, #T_310fe_row0_col6, #T_310fe_row0_col7, #T_310fe_row1_col4 {\n",
       "  text-align: left;\n",
       "  background-color: yellow;\n",
       "}\n",
       "#T_310fe_row0_col8, #T_310fe_row1_col8, #T_310fe_row2_col8, #T_310fe_row3_col8, #T_310fe_row4_col8, #T_310fe_row5_col8, #T_310fe_row7_col8, #T_310fe_row8_col8, #T_310fe_row9_col8, #T_310fe_row10_col8, #T_310fe_row11_col8, #T_310fe_row12_col8, #T_310fe_row13_col8 {\n",
       "  text-align: left;\n",
       "  background-color: lightgrey;\n",
       "}\n",
       "#T_310fe_row6_col8 {\n",
       "  text-align: left;\n",
       "  background-color: yellow;\n",
       "  background-color: lightgrey;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_310fe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_310fe_level0_col0\" class=\"col_heading level0 col0\" >Model</th>\n",
       "      <th id=\"T_310fe_level0_col1\" class=\"col_heading level0 col1\" >Accuracy</th>\n",
       "      <th id=\"T_310fe_level0_col2\" class=\"col_heading level0 col2\" >AUC</th>\n",
       "      <th id=\"T_310fe_level0_col3\" class=\"col_heading level0 col3\" >Recall</th>\n",
       "      <th id=\"T_310fe_level0_col4\" class=\"col_heading level0 col4\" >Prec.</th>\n",
       "      <th id=\"T_310fe_level0_col5\" class=\"col_heading level0 col5\" >F1</th>\n",
       "      <th id=\"T_310fe_level0_col6\" class=\"col_heading level0 col6\" >Kappa</th>\n",
       "      <th id=\"T_310fe_level0_col7\" class=\"col_heading level0 col7\" >MCC</th>\n",
       "      <th id=\"T_310fe_level0_col8\" class=\"col_heading level0 col8\" >TT (Sec)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_310fe_level0_row0\" class=\"row_heading level0 row0\" >lightgbm</th>\n",
       "      <td id=\"T_310fe_row0_col0\" class=\"data row0 col0\" >Light Gradient Boosting Machine</td>\n",
       "      <td id=\"T_310fe_row0_col1\" class=\"data row0 col1\" >0.7035</td>\n",
       "      <td id=\"T_310fe_row0_col2\" class=\"data row0 col2\" >0.7789</td>\n",
       "      <td id=\"T_310fe_row0_col3\" class=\"data row0 col3\" >0.6875</td>\n",
       "      <td id=\"T_310fe_row0_col4\" class=\"data row0 col4\" >0.7010</td>\n",
       "      <td id=\"T_310fe_row0_col5\" class=\"data row0 col5\" >0.6941</td>\n",
       "      <td id=\"T_310fe_row0_col6\" class=\"data row0 col6\" >0.4066</td>\n",
       "      <td id=\"T_310fe_row0_col7\" class=\"data row0 col7\" >0.4067</td>\n",
       "      <td id=\"T_310fe_row0_col8\" class=\"data row0 col8\" >18.1760</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_310fe_level0_row1\" class=\"row_heading level0 row1\" >rf</th>\n",
       "      <td id=\"T_310fe_row1_col0\" class=\"data row1 col0\" >Random Forest Classifier</td>\n",
       "      <td id=\"T_310fe_row1_col1\" class=\"data row1 col1\" >0.7001</td>\n",
       "      <td id=\"T_310fe_row1_col2\" class=\"data row1 col2\" >0.7710</td>\n",
       "      <td id=\"T_310fe_row1_col3\" class=\"data row1 col3\" >0.6739</td>\n",
       "      <td id=\"T_310fe_row1_col4\" class=\"data row1 col4\" >0.7016</td>\n",
       "      <td id=\"T_310fe_row1_col5\" class=\"data row1 col5\" >0.6874</td>\n",
       "      <td id=\"T_310fe_row1_col6\" class=\"data row1 col6\" >0.3995</td>\n",
       "      <td id=\"T_310fe_row1_col7\" class=\"data row1 col7\" >0.3998</td>\n",
       "      <td id=\"T_310fe_row1_col8\" class=\"data row1 col8\" >24.6080</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_310fe_level0_row2\" class=\"row_heading level0 row2\" >gbc</th>\n",
       "      <td id=\"T_310fe_row2_col0\" class=\"data row2 col0\" >Gradient Boosting Classifier</td>\n",
       "      <td id=\"T_310fe_row2_col1\" class=\"data row2 col1\" >0.6957</td>\n",
       "      <td id=\"T_310fe_row2_col2\" class=\"data row2 col2\" >0.7682</td>\n",
       "      <td id=\"T_310fe_row2_col3\" class=\"data row2 col3\" >0.6759</td>\n",
       "      <td id=\"T_310fe_row2_col4\" class=\"data row2 col4\" >0.6942</td>\n",
       "      <td id=\"T_310fe_row2_col5\" class=\"data row2 col5\" >0.6849</td>\n",
       "      <td id=\"T_310fe_row2_col6\" class=\"data row2 col6\" >0.3908</td>\n",
       "      <td id=\"T_310fe_row2_col7\" class=\"data row2 col7\" >0.3910</td>\n",
       "      <td id=\"T_310fe_row2_col8\" class=\"data row2 col8\" >26.1490</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_310fe_level0_row3\" class=\"row_heading level0 row3\" >ada</th>\n",
       "      <td id=\"T_310fe_row3_col0\" class=\"data row3 col0\" >Ada Boost Classifier</td>\n",
       "      <td id=\"T_310fe_row3_col1\" class=\"data row3 col1\" >0.6906</td>\n",
       "      <td id=\"T_310fe_row3_col2\" class=\"data row3 col2\" >0.7617</td>\n",
       "      <td id=\"T_310fe_row3_col3\" class=\"data row3 col3\" >0.6771</td>\n",
       "      <td id=\"T_310fe_row3_col4\" class=\"data row3 col4\" >0.6864</td>\n",
       "      <td id=\"T_310fe_row3_col5\" class=\"data row3 col5\" >0.6817</td>\n",
       "      <td id=\"T_310fe_row3_col6\" class=\"data row3 col6\" >0.3808</td>\n",
       "      <td id=\"T_310fe_row3_col7\" class=\"data row3 col7\" >0.3809</td>\n",
       "      <td id=\"T_310fe_row3_col8\" class=\"data row3 col8\" >24.3430</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_310fe_level0_row4\" class=\"row_heading level0 row4\" >lr</th>\n",
       "      <td id=\"T_310fe_row4_col0\" class=\"data row4 col0\" >Logistic Regression</td>\n",
       "      <td id=\"T_310fe_row4_col1\" class=\"data row4 col1\" >0.6904</td>\n",
       "      <td id=\"T_310fe_row4_col2\" class=\"data row4 col2\" >0.7626</td>\n",
       "      <td id=\"T_310fe_row4_col3\" class=\"data row4 col3\" >0.6768</td>\n",
       "      <td id=\"T_310fe_row4_col4\" class=\"data row4 col4\" >0.6863</td>\n",
       "      <td id=\"T_310fe_row4_col5\" class=\"data row4 col5\" >0.6815</td>\n",
       "      <td id=\"T_310fe_row4_col6\" class=\"data row4 col6\" >0.3804</td>\n",
       "      <td id=\"T_310fe_row4_col7\" class=\"data row4 col7\" >0.3804</td>\n",
       "      <td id=\"T_310fe_row4_col8\" class=\"data row4 col8\" >19.6950</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_310fe_level0_row5\" class=\"row_heading level0 row5\" >lda</th>\n",
       "      <td id=\"T_310fe_row5_col0\" class=\"data row5 col0\" >Linear Discriminant Analysis</td>\n",
       "      <td id=\"T_310fe_row5_col1\" class=\"data row5 col1\" >0.6900</td>\n",
       "      <td id=\"T_310fe_row5_col2\" class=\"data row5 col2\" >0.7625</td>\n",
       "      <td id=\"T_310fe_row5_col3\" class=\"data row5 col3\" >0.6764</td>\n",
       "      <td id=\"T_310fe_row5_col4\" class=\"data row5 col4\" >0.6858</td>\n",
       "      <td id=\"T_310fe_row5_col5\" class=\"data row5 col5\" >0.6810</td>\n",
       "      <td id=\"T_310fe_row5_col6\" class=\"data row5 col6\" >0.3795</td>\n",
       "      <td id=\"T_310fe_row5_col7\" class=\"data row5 col7\" >0.3796</td>\n",
       "      <td id=\"T_310fe_row5_col8\" class=\"data row5 col8\" >18.5870</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_310fe_level0_row6\" class=\"row_heading level0 row6\" >ridge</th>\n",
       "      <td id=\"T_310fe_row6_col0\" class=\"data row6 col0\" >Ridge Classifier</td>\n",
       "      <td id=\"T_310fe_row6_col1\" class=\"data row6 col1\" >0.6899</td>\n",
       "      <td id=\"T_310fe_row6_col2\" class=\"data row6 col2\" >0.0000</td>\n",
       "      <td id=\"T_310fe_row6_col3\" class=\"data row6 col3\" >0.6764</td>\n",
       "      <td id=\"T_310fe_row6_col4\" class=\"data row6 col4\" >0.6857</td>\n",
       "      <td id=\"T_310fe_row6_col5\" class=\"data row6 col5\" >0.6810</td>\n",
       "      <td id=\"T_310fe_row6_col6\" class=\"data row6 col6\" >0.3794</td>\n",
       "      <td id=\"T_310fe_row6_col7\" class=\"data row6 col7\" >0.3794</td>\n",
       "      <td id=\"T_310fe_row6_col8\" class=\"data row6 col8\" >17.5490</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_310fe_level0_row7\" class=\"row_heading level0 row7\" >svm</th>\n",
       "      <td id=\"T_310fe_row7_col0\" class=\"data row7 col0\" >SVM - Linear Kernel</td>\n",
       "      <td id=\"T_310fe_row7_col1\" class=\"data row7 col1\" >0.6855</td>\n",
       "      <td id=\"T_310fe_row7_col2\" class=\"data row7 col2\" >0.0000</td>\n",
       "      <td id=\"T_310fe_row7_col3\" class=\"data row7 col3\" >0.6710</td>\n",
       "      <td id=\"T_310fe_row7_col4\" class=\"data row7 col4\" >0.6828</td>\n",
       "      <td id=\"T_310fe_row7_col5\" class=\"data row7 col5\" >0.6757</td>\n",
       "      <td id=\"T_310fe_row7_col6\" class=\"data row7 col6\" >0.3706</td>\n",
       "      <td id=\"T_310fe_row7_col7\" class=\"data row7 col7\" >0.3718</td>\n",
       "      <td id=\"T_310fe_row7_col8\" class=\"data row7 col8\" >18.5270</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_310fe_level0_row8\" class=\"row_heading level0 row8\" >et</th>\n",
       "      <td id=\"T_310fe_row8_col0\" class=\"data row8 col0\" >Extra Trees Classifier</td>\n",
       "      <td id=\"T_310fe_row8_col1\" class=\"data row8 col1\" >0.6842</td>\n",
       "      <td id=\"T_310fe_row8_col2\" class=\"data row8 col2\" >0.7511</td>\n",
       "      <td id=\"T_310fe_row8_col3\" class=\"data row8 col3\" >0.6583</td>\n",
       "      <td id=\"T_310fe_row8_col4\" class=\"data row8 col4\" >0.6844</td>\n",
       "      <td id=\"T_310fe_row8_col5\" class=\"data row8 col5\" >0.6711</td>\n",
       "      <td id=\"T_310fe_row8_col6\" class=\"data row8 col6\" >0.3677</td>\n",
       "      <td id=\"T_310fe_row8_col7\" class=\"data row8 col7\" >0.3680</td>\n",
       "      <td id=\"T_310fe_row8_col8\" class=\"data row8 col8\" >29.0670</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_310fe_level0_row9\" class=\"row_heading level0 row9\" >knn</th>\n",
       "      <td id=\"T_310fe_row9_col0\" class=\"data row9 col0\" >K Neighbors Classifier</td>\n",
       "      <td id=\"T_310fe_row9_col1\" class=\"data row9 col1\" >0.6419</td>\n",
       "      <td id=\"T_310fe_row9_col2\" class=\"data row9 col2\" >0.6876</td>\n",
       "      <td id=\"T_310fe_row9_col3\" class=\"data row9 col3\" >0.6350</td>\n",
       "      <td id=\"T_310fe_row9_col4\" class=\"data row9 col4\" >0.6339</td>\n",
       "      <td id=\"T_310fe_row9_col5\" class=\"data row9 col5\" >0.6344</td>\n",
       "      <td id=\"T_310fe_row9_col6\" class=\"data row9 col6\" >0.2836</td>\n",
       "      <td id=\"T_310fe_row9_col7\" class=\"data row9 col7\" >0.2836</td>\n",
       "      <td id=\"T_310fe_row9_col8\" class=\"data row9 col8\" >20.2080</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_310fe_level0_row10\" class=\"row_heading level0 row10\" >dt</th>\n",
       "      <td id=\"T_310fe_row10_col0\" class=\"data row10 col0\" >Decision Tree Classifier</td>\n",
       "      <td id=\"T_310fe_row10_col1\" class=\"data row10 col1\" >0.6190</td>\n",
       "      <td id=\"T_310fe_row10_col2\" class=\"data row10 col2\" >0.6189</td>\n",
       "      <td id=\"T_310fe_row10_col3\" class=\"data row10 col3\" >0.6130</td>\n",
       "      <td id=\"T_310fe_row10_col4\" class=\"data row10 col4\" >0.6103</td>\n",
       "      <td id=\"T_310fe_row10_col5\" class=\"data row10 col5\" >0.6116</td>\n",
       "      <td id=\"T_310fe_row10_col6\" class=\"data row10 col6\" >0.2377</td>\n",
       "      <td id=\"T_310fe_row10_col7\" class=\"data row10 col7\" >0.2377</td>\n",
       "      <td id=\"T_310fe_row10_col8\" class=\"data row10 col8\" >20.6300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_310fe_level0_row11\" class=\"row_heading level0 row11\" >nb</th>\n",
       "      <td id=\"T_310fe_row11_col0\" class=\"data row11 col0\" >Naive Bayes</td>\n",
       "      <td id=\"T_310fe_row11_col1\" class=\"data row11 col1\" >0.6115</td>\n",
       "      <td id=\"T_310fe_row11_col2\" class=\"data row11 col2\" >0.6555</td>\n",
       "      <td id=\"T_310fe_row11_col3\" class=\"data row11 col3\" >0.6698</td>\n",
       "      <td id=\"T_310fe_row11_col4\" class=\"data row11 col4\" >0.5909</td>\n",
       "      <td id=\"T_310fe_row11_col5\" class=\"data row11 col5\" >0.6278</td>\n",
       "      <td id=\"T_310fe_row11_col6\" class=\"data row11 col6\" >0.2247</td>\n",
       "      <td id=\"T_310fe_row11_col7\" class=\"data row11 col7\" >0.2267</td>\n",
       "      <td id=\"T_310fe_row11_col8\" class=\"data row11 col8\" >19.0590</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_310fe_level0_row12\" class=\"row_heading level0 row12\" >qda</th>\n",
       "      <td id=\"T_310fe_row12_col0\" class=\"data row12 col0\" >Quadratic Discriminant Analysis</td>\n",
       "      <td id=\"T_310fe_row12_col1\" class=\"data row12 col1\" >0.5819</td>\n",
       "      <td id=\"T_310fe_row12_col2\" class=\"data row12 col2\" >0.5980</td>\n",
       "      <td id=\"T_310fe_row12_col3\" class=\"data row12 col3\" >0.4203</td>\n",
       "      <td id=\"T_310fe_row12_col4\" class=\"data row12 col4\" >0.6076</td>\n",
       "      <td id=\"T_310fe_row12_col5\" class=\"data row12 col5\" >0.4793</td>\n",
       "      <td id=\"T_310fe_row12_col6\" class=\"data row12 col6\" >0.1578</td>\n",
       "      <td id=\"T_310fe_row12_col7\" class=\"data row12 col7\" >0.1703</td>\n",
       "      <td id=\"T_310fe_row12_col8\" class=\"data row12 col8\" >23.1720</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_310fe_level0_row13\" class=\"row_heading level0 row13\" >dummy</th>\n",
       "      <td id=\"T_310fe_row13_col0\" class=\"data row13 col0\" >Dummy Classifier</td>\n",
       "      <td id=\"T_310fe_row13_col1\" class=\"data row13 col1\" >0.5106</td>\n",
       "      <td id=\"T_310fe_row13_col2\" class=\"data row13 col2\" >0.5000</td>\n",
       "      <td id=\"T_310fe_row13_col3\" class=\"data row13 col3\" >0.0000</td>\n",
       "      <td id=\"T_310fe_row13_col4\" class=\"data row13 col4\" >0.0000</td>\n",
       "      <td id=\"T_310fe_row13_col5\" class=\"data row13 col5\" >0.0000</td>\n",
       "      <td id=\"T_310fe_row13_col6\" class=\"data row13 col6\" >0.0000</td>\n",
       "      <td id=\"T_310fe_row13_col7\" class=\"data row13 col7\" >0.0000</td>\n",
       "      <td id=\"T_310fe_row13_col8\" class=\"data row13 col8\" >17.7240</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x162328992a0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#comparación de modelos\n",
    "model = compare_models()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [],
   "source": [
    "# model = create_model('lightgbm')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_4b58c_row10_col0, #T_4b58c_row10_col1, #T_4b58c_row10_col2, #T_4b58c_row10_col3, #T_4b58c_row10_col4, #T_4b58c_row10_col5, #T_4b58c_row10_col6 {\n",
       "  background: yellow;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_4b58c\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_4b58c_level0_col0\" class=\"col_heading level0 col0\" >Accuracy</th>\n",
       "      <th id=\"T_4b58c_level0_col1\" class=\"col_heading level0 col1\" >AUC</th>\n",
       "      <th id=\"T_4b58c_level0_col2\" class=\"col_heading level0 col2\" >Recall</th>\n",
       "      <th id=\"T_4b58c_level0_col3\" class=\"col_heading level0 col3\" >Prec.</th>\n",
       "      <th id=\"T_4b58c_level0_col4\" class=\"col_heading level0 col4\" >F1</th>\n",
       "      <th id=\"T_4b58c_level0_col5\" class=\"col_heading level0 col5\" >Kappa</th>\n",
       "      <th id=\"T_4b58c_level0_col6\" class=\"col_heading level0 col6\" >MCC</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th class=\"index_name level0\" >Fold</th>\n",
       "      <th class=\"blank col0\" >&nbsp;</th>\n",
       "      <th class=\"blank col1\" >&nbsp;</th>\n",
       "      <th class=\"blank col2\" >&nbsp;</th>\n",
       "      <th class=\"blank col3\" >&nbsp;</th>\n",
       "      <th class=\"blank col4\" >&nbsp;</th>\n",
       "      <th class=\"blank col5\" >&nbsp;</th>\n",
       "      <th class=\"blank col6\" >&nbsp;</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_4b58c_level0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "      <td id=\"T_4b58c_row0_col0\" class=\"data row0 col0\" >0.7128</td>\n",
       "      <td id=\"T_4b58c_row0_col1\" class=\"data row0 col1\" >0.7844</td>\n",
       "      <td id=\"T_4b58c_row0_col2\" class=\"data row0 col2\" >0.7039</td>\n",
       "      <td id=\"T_4b58c_row0_col3\" class=\"data row0 col3\" >0.7076</td>\n",
       "      <td id=\"T_4b58c_row0_col4\" class=\"data row0 col4\" >0.7057</td>\n",
       "      <td id=\"T_4b58c_row0_col5\" class=\"data row0 col5\" >0.4252</td>\n",
       "      <td id=\"T_4b58c_row0_col6\" class=\"data row0 col6\" >0.4252</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4b58c_level0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "      <td id=\"T_4b58c_row1_col0\" class=\"data row1 col0\" >0.6973</td>\n",
       "      <td id=\"T_4b58c_row1_col1\" class=\"data row1 col1\" >0.7660</td>\n",
       "      <td id=\"T_4b58c_row1_col2\" class=\"data row1 col2\" >0.6896</td>\n",
       "      <td id=\"T_4b58c_row1_col3\" class=\"data row1 col3\" >0.6912</td>\n",
       "      <td id=\"T_4b58c_row1_col4\" class=\"data row1 col4\" >0.6904</td>\n",
       "      <td id=\"T_4b58c_row1_col5\" class=\"data row1 col5\" >0.3943</td>\n",
       "      <td id=\"T_4b58c_row1_col6\" class=\"data row1 col6\" >0.3943</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4b58c_level0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "      <td id=\"T_4b58c_row2_col0\" class=\"data row2 col0\" >0.7023</td>\n",
       "      <td id=\"T_4b58c_row2_col1\" class=\"data row2 col1\" >0.7819</td>\n",
       "      <td id=\"T_4b58c_row2_col2\" class=\"data row2 col2\" >0.6849</td>\n",
       "      <td id=\"T_4b58c_row2_col3\" class=\"data row2 col3\" >0.7002</td>\n",
       "      <td id=\"T_4b58c_row2_col4\" class=\"data row2 col4\" >0.6925</td>\n",
       "      <td id=\"T_4b58c_row2_col5\" class=\"data row2 col5\" >0.4041</td>\n",
       "      <td id=\"T_4b58c_row2_col6\" class=\"data row2 col6\" >0.4042</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4b58c_level0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
       "      <td id=\"T_4b58c_row3_col0\" class=\"data row3 col0\" >0.7057</td>\n",
       "      <td id=\"T_4b58c_row3_col1\" class=\"data row3 col1\" >0.7823</td>\n",
       "      <td id=\"T_4b58c_row3_col2\" class=\"data row3 col2\" >0.6854</td>\n",
       "      <td id=\"T_4b58c_row3_col3\" class=\"data row3 col3\" >0.7050</td>\n",
       "      <td id=\"T_4b58c_row3_col4\" class=\"data row3 col4\" >0.6951</td>\n",
       "      <td id=\"T_4b58c_row3_col5\" class=\"data row3 col5\" >0.4107</td>\n",
       "      <td id=\"T_4b58c_row3_col6\" class=\"data row3 col6\" >0.4109</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4b58c_level0_row4\" class=\"row_heading level0 row4\" >4</th>\n",
       "      <td id=\"T_4b58c_row4_col0\" class=\"data row4 col0\" >0.6979</td>\n",
       "      <td id=\"T_4b58c_row4_col1\" class=\"data row4 col1\" >0.7755</td>\n",
       "      <td id=\"T_4b58c_row4_col2\" class=\"data row4 col2\" >0.6914</td>\n",
       "      <td id=\"T_4b58c_row4_col3\" class=\"data row4 col3\" >0.6914</td>\n",
       "      <td id=\"T_4b58c_row4_col4\" class=\"data row4 col4\" >0.6914</td>\n",
       "      <td id=\"T_4b58c_row4_col5\" class=\"data row4 col5\" >0.3956</td>\n",
       "      <td id=\"T_4b58c_row4_col6\" class=\"data row4 col6\" >0.3956</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4b58c_level0_row5\" class=\"row_heading level0 row5\" >5</th>\n",
       "      <td id=\"T_4b58c_row5_col0\" class=\"data row5 col0\" >0.6993</td>\n",
       "      <td id=\"T_4b58c_row5_col1\" class=\"data row5 col1\" >0.7696</td>\n",
       "      <td id=\"T_4b58c_row5_col2\" class=\"data row5 col2\" >0.6910</td>\n",
       "      <td id=\"T_4b58c_row5_col3\" class=\"data row5 col3\" >0.6935</td>\n",
       "      <td id=\"T_4b58c_row5_col4\" class=\"data row5 col4\" >0.6922</td>\n",
       "      <td id=\"T_4b58c_row5_col5\" class=\"data row5 col5\" >0.3983</td>\n",
       "      <td id=\"T_4b58c_row5_col6\" class=\"data row5 col6\" >0.3983</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4b58c_level0_row6\" class=\"row_heading level0 row6\" >6</th>\n",
       "      <td id=\"T_4b58c_row6_col0\" class=\"data row6 col0\" >0.6991</td>\n",
       "      <td id=\"T_4b58c_row6_col1\" class=\"data row6 col1\" >0.7820</td>\n",
       "      <td id=\"T_4b58c_row6_col2\" class=\"data row6 col2\" >0.6840</td>\n",
       "      <td id=\"T_4b58c_row6_col3\" class=\"data row6 col3\" >0.6959</td>\n",
       "      <td id=\"T_4b58c_row6_col4\" class=\"data row6 col4\" >0.6899</td>\n",
       "      <td id=\"T_4b58c_row6_col5\" class=\"data row6 col5\" >0.3977</td>\n",
       "      <td id=\"T_4b58c_row6_col6\" class=\"data row6 col6\" >0.3977</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4b58c_level0_row7\" class=\"row_heading level0 row7\" >7</th>\n",
       "      <td id=\"T_4b58c_row7_col0\" class=\"data row7 col0\" >0.6982</td>\n",
       "      <td id=\"T_4b58c_row7_col1\" class=\"data row7 col1\" >0.7738</td>\n",
       "      <td id=\"T_4b58c_row7_col2\" class=\"data row7 col2\" >0.6724</td>\n",
       "      <td id=\"T_4b58c_row7_col3\" class=\"data row7 col3\" >0.6993</td>\n",
       "      <td id=\"T_4b58c_row7_col4\" class=\"data row7 col4\" >0.6856</td>\n",
       "      <td id=\"T_4b58c_row7_col5\" class=\"data row7 col5\" >0.3956</td>\n",
       "      <td id=\"T_4b58c_row7_col6\" class=\"data row7 col6\" >0.3959</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4b58c_level0_row8\" class=\"row_heading level0 row8\" >8</th>\n",
       "      <td id=\"T_4b58c_row8_col0\" class=\"data row8 col0\" >0.7179</td>\n",
       "      <td id=\"T_4b58c_row8_col1\" class=\"data row8 col1\" >0.7881</td>\n",
       "      <td id=\"T_4b58c_row8_col2\" class=\"data row8 col2\" >0.7089</td>\n",
       "      <td id=\"T_4b58c_row8_col3\" class=\"data row8 col3\" >0.7129</td>\n",
       "      <td id=\"T_4b58c_row8_col4\" class=\"data row8 col4\" >0.7109</td>\n",
       "      <td id=\"T_4b58c_row8_col5\" class=\"data row8 col5\" >0.4355</td>\n",
       "      <td id=\"T_4b58c_row8_col6\" class=\"data row8 col6\" >0.4355</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4b58c_level0_row9\" class=\"row_heading level0 row9\" >9</th>\n",
       "      <td id=\"T_4b58c_row9_col0\" class=\"data row9 col0\" >0.6998</td>\n",
       "      <td id=\"T_4b58c_row9_col1\" class=\"data row9 col1\" >0.7742</td>\n",
       "      <td id=\"T_4b58c_row9_col2\" class=\"data row9 col2\" >0.6755</td>\n",
       "      <td id=\"T_4b58c_row9_col3\" class=\"data row9 col3\" >0.7002</td>\n",
       "      <td id=\"T_4b58c_row9_col4\" class=\"data row9 col4\" >0.6876</td>\n",
       "      <td id=\"T_4b58c_row9_col5\" class=\"data row9 col5\" >0.3988</td>\n",
       "      <td id=\"T_4b58c_row9_col6\" class=\"data row9 col6\" >0.3990</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4b58c_level0_row10\" class=\"row_heading level0 row10\" >Mean</th>\n",
       "      <td id=\"T_4b58c_row10_col0\" class=\"data row10 col0\" >0.7030</td>\n",
       "      <td id=\"T_4b58c_row10_col1\" class=\"data row10 col1\" >0.7778</td>\n",
       "      <td id=\"T_4b58c_row10_col2\" class=\"data row10 col2\" >0.6887</td>\n",
       "      <td id=\"T_4b58c_row10_col3\" class=\"data row10 col3\" >0.6997</td>\n",
       "      <td id=\"T_4b58c_row10_col4\" class=\"data row10 col4\" >0.6941</td>\n",
       "      <td id=\"T_4b58c_row10_col5\" class=\"data row10 col5\" >0.4056</td>\n",
       "      <td id=\"T_4b58c_row10_col6\" class=\"data row10 col6\" >0.4057</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4b58c_level0_row11\" class=\"row_heading level0 row11\" >Std</th>\n",
       "      <td id=\"T_4b58c_row11_col0\" class=\"data row11 col0\" >0.0067</td>\n",
       "      <td id=\"T_4b58c_row11_col1\" class=\"data row11 col1\" >0.0067</td>\n",
       "      <td id=\"T_4b58c_row11_col2\" class=\"data row11 col2\" >0.0107</td>\n",
       "      <td id=\"T_4b58c_row11_col3\" class=\"data row11 col3\" >0.0068</td>\n",
       "      <td id=\"T_4b58c_row11_col4\" class=\"data row11 col4\" >0.0076</td>\n",
       "      <td id=\"T_4b58c_row11_col5\" class=\"data row11 col5\" >0.0134</td>\n",
       "      <td id=\"T_4b58c_row11_col6\" class=\"data row11 col6\" >0.0134</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x1622d31eec0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 10 folds for each of 10 candidates, totalling 100 fits\n",
      "Original model was better than the tuned model, hence it will be returned. NOTE: The display metrics are for the tuned model (not the original one).\n"
     ]
    }
   ],
   "source": [
    "# tunear el mejor modelo\n",
    "model_tuned = tune_model(model, optimize = 'F1', choose_better=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [],
   "source": [
    "# finalizar modelo\n",
    "model_finalized = finalize_model(model_tuned)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Transformation Pipeline and Model Successfully Saved\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(Pipeline(memory=FastMemory(location=C:\\Users\\alniquia\\AppData\\Local\\Temp\\joblib),\n",
       "          steps=[('label_encoding',\n",
       "                  TransformerWrapperWithInverse(exclude=None, include=None,\n",
       "                                                transformer=LabelEncoder())),\n",
       "                 ('numerical_imputer',\n",
       "                  TransformerWrapper(exclude=None,\n",
       "                                     include=['avg_brawler_trophies_diff',\n",
       "                                              'max_brawler_trophies_diff',\n",
       "                                              'min_brawler_trophies_diff',\n",
       "                                              'battle_power...\n",
       "                  LGBMClassifier(boosting_type='gbdt', class_weight=None,\n",
       "                                 colsample_bytree=1.0, importance_type='split',\n",
       "                                 learning_rate=0.1, max_depth=-1,\n",
       "                                 min_child_samples=20, min_child_weight=0.001,\n",
       "                                 min_split_gain=0.0, n_estimators=100, n_jobs=-1,\n",
       "                                 num_leaves=31, objective=None,\n",
       "                                 random_state=2557, reg_alpha=0.0,\n",
       "                                 reg_lambda=0.0, silent='warn', subsample=1.0,\n",
       "                                 subsample_for_bin=200000, subsample_freq=0))],\n",
       "          verbose=False),\n",
       " 'models/bs_predictor_brawlBall.pkl')"
      ]
     },
     "execution_count": 112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "save_model(model_finalized, 'models/bs_predictor_' + event_mode)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>event_map</th>\n",
       "      <th>avg_brawler_trophies_diff</th>\n",
       "      <th>max_brawler_trophies_diff</th>\n",
       "      <th>min_brawler_trophies_diff</th>\n",
       "      <th>battle_power_diff</th>\n",
       "      <th>T1_8-BIT</th>\n",
       "      <th>T1_AMBER</th>\n",
       "      <th>T1_ASH</th>\n",
       "      <th>T1_BARLEY</th>\n",
       "      <th>T1_BEA</th>\n",
       "      <th>...</th>\n",
       "      <th>T2_SPROUT</th>\n",
       "      <th>T2_SQUEAK</th>\n",
       "      <th>T2_STU</th>\n",
       "      <th>T2_SURGE</th>\n",
       "      <th>T2_TARA</th>\n",
       "      <th>T2_TICK</th>\n",
       "      <th>T2_WILLOW</th>\n",
       "      <th>winner_team</th>\n",
       "      <th>prediction_label</th>\n",
       "      <th>prediction_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>57102</th>\n",
       "      <td>Pinhole Punt</td>\n",
       "      <td>-0.041070</td>\n",
       "      <td>-0.010917</td>\n",
       "      <td>-0.049714</td>\n",
       "      <td>0.0</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.5082</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38482</th>\n",
       "      <td>Super Beach</td>\n",
       "      <td>-0.005692</td>\n",
       "      <td>0.007639</td>\n",
       "      <td>-0.033310</td>\n",
       "      <td>-3.0</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0.6706</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9568</th>\n",
       "      <td>Super Beach</td>\n",
       "      <td>-0.009444</td>\n",
       "      <td>-0.007335</td>\n",
       "      <td>-0.021158</td>\n",
       "      <td>4.0</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.8010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43483</th>\n",
       "      <td>Super Beach</td>\n",
       "      <td>-0.150787</td>\n",
       "      <td>-0.165928</td>\n",
       "      <td>-0.178248</td>\n",
       "      <td>-2.0</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>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>2</td>\n",
       "      <td>0.6556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>102972</th>\n",
       "      <td>Backyard Bowl</td>\n",
       "      <td>0.021368</td>\n",
       "      <td>-0.016099</td>\n",
       "      <td>0.105361</td>\n",
       "      <td>-1.0</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.6457</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 144 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            event_map  avg_brawler_trophies_diff  max_brawler_trophies_diff  \\\n",
       "57102    Pinhole Punt                  -0.041070                  -0.010917   \n",
       "38482     Super Beach                  -0.005692                   0.007639   \n",
       "9568      Super Beach                  -0.009444                  -0.007335   \n",
       "43483     Super Beach                  -0.150787                  -0.165928   \n",
       "102972  Backyard Bowl                   0.021368                  -0.016099   \n",
       "\n",
       "        min_brawler_trophies_diff  battle_power_diff  T1_8-BIT  T1_AMBER  \\\n",
       "57102                   -0.049714                0.0         0         0   \n",
       "38482                   -0.033310               -3.0         0         0   \n",
       "9568                    -0.021158                4.0         0         0   \n",
       "43483                   -0.178248               -2.0         0         0   \n",
       "102972                   0.105361               -1.0         0         0   \n",
       "\n",
       "        T1_ASH  T1_BARLEY  T1_BEA  ...  T2_SPROUT  T2_SQUEAK  T2_STU  \\\n",
       "57102        0          0       0  ...          0          0       0   \n",
       "38482        0          0       0  ...          0          0       0   \n",
       "9568         0          0       0  ...          0          0       0   \n",
       "43483        0          0       0  ...          0          0       0   \n",
       "102972       0          0       0  ...          0          0       0   \n",
       "\n",
       "        T2_SURGE  T2_TARA  T2_TICK  T2_WILLOW  winner_team  prediction_label  \\\n",
       "57102          0        0        0          0            2                 1   \n",
       "38482          0        1        0          0            2                 2   \n",
       "9568           1        0        0          0            2                 1   \n",
       "43483          0        0        1          0            1                 2   \n",
       "102972         0        0        0          0            1                 1   \n",
       "\n",
       "        prediction_score  \n",
       "57102             0.5082  \n",
       "38482             0.6706  \n",
       "9568              0.8010  \n",
       "43483             0.6556  \n",
       "102972            0.6457  \n",
       "\n",
       "[5 rows x 144 columns]"
      ]
     },
     "execution_count": 113,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions = predict_model(model_finalized, data = test)\n",
    "predictions['winner_team'] = predictions['winner_team'] + 1\n",
    "predictions.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'accuracy': 0.7042777433353998, 'f1': 0.7125840092699883, 'precision': 0.7140070592606353, 'recall': 0.7111666204089185}\n"
     ]
    }
   ],
   "source": [
    "def metrics_capturing(df):\n",
    "\t\"\"\"Captures the metrics of a classification model\"\"\"\n",
    "\t# from sklearn.metrics import accuracy_score, f1_score, precision_score, recall_score, classification_report, confusion_matrix\n",
    "\tfrom sklearn.metrics import accuracy_score, f1_score, precision_score, recall_score\n",
    "\n",
    "\taccuracy = accuracy_score(df['winner_team'], df['prediction_label'])\n",
    "\tf1 = f1_score(df['winner_team'], df['prediction_label'])\n",
    "\tprecision = precision_score(df['winner_team'], df['prediction_label'])\n",
    "\trecall = recall_score(df['winner_team'], df['prediction_label'])\n",
    "\t# report = classification_report(df['winner_team'], df['prediction_label'])\n",
    "\t# matrix = confusion_matrix(df['winner_team'], df['prediction_label'])\n",
    "\n",
    "\t# return accuracy, f1, precision, recall, report, matrix\n",
    "\treturn accuracy, f1, precision, recall\n",
    "\n",
    "accuracy, f1, precision, recall = metrics_capturing(predictions)\n",
    "\n",
    "# save metrics\n",
    "metrics = {\n",
    "\t\t'accuracy': accuracy,\n",
    "\t\t'f1': f1,\n",
    "\t\t'precision': precision,\n",
    "\t\t'recall': recall,\n",
    "}\n",
    "\n",
    "print(metrics)\n",
    "\n",
    "with open('resources/bs_metrics.json', 'r') as f:\n",
    "\tdata = json.load(f)\n",
    "\tdata[event_mode] = metrics\n",
    "\n",
    "with open('resources/bs_metrics.json', 'w') as f:\n",
    "\tjson.dump(data, f, indent=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "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>feature</th>\n",
       "      <th>importance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>max_brawler_trophies_diff</td>\n",
       "      <td>375</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>avg_brawler_trophies_diff</td>\n",
       "      <td>329</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>battle_power_diff</td>\n",
       "      <td>264</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>min_brawler_trophies_diff</td>\n",
       "      <td>263</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>110</th>\n",
       "      <td>T2_HANK</td>\n",
       "      <td>75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>127</th>\n",
       "      <td>T2_PENNY</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>122</th>\n",
       "      <td>T2_MR._P</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>T1_LOLA</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>T1_MEG</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>101</th>\n",
       "      <td>T2_EVE</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>145 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                       feature  importance\n",
       "6    max_brawler_trophies_diff         375\n",
       "5    avg_brawler_trophies_diff         329\n",
       "8            battle_power_diff         264\n",
       "7    min_brawler_trophies_diff         263\n",
       "110                    T2_HANK          75\n",
       "..                         ...         ...\n",
       "127                   T2_PENNY           0\n",
       "122                   T2_MR._P           0\n",
       "47                     T1_LOLA           0\n",
       "52                      T1_MEG           0\n",
       "101                     T2_EVE           0\n",
       "\n",
       "[145 rows x 2 columns]"
      ]
     },
     "execution_count": 115,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# get feature importance\n",
    "feature_importance = pd.DataFrame({\n",
    "\t\t'feature': model_finalized.booster_.feature_name(),\n",
    "\t\t'importance': model_finalized.feature_importances_\n",
    "\t\t})\n",
    "\n",
    "feature_importance.sort_values(by='importance', ascending=False, inplace=True)\n",
    "\n",
    "feature_importance.to_json('resources/bs_feature_importance.json', orient='records')\n",
    "\n",
    "feature_importance"
   ]
  }
 ],
 "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": "3.10.10"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}