subset_selection_methods_final.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from matplotlib import pyplot as plt\n",
    "from scipy.stats import linregress\n",
    "import numpy as np\n",
    "import itertools\n",
    "import time\n",
    "import seaborn as sns\n",
    "import statsmodels.api as sm\n",
    "from sklearn import linear_model\n",
    "from sklearn.metrics import mean_squared_error\n",
    "\n",
    "%matplotlib inline\n",
    "plt.style.use('ggplot')"
   ]
  },
  {
   "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>playerID</th>\n",
       "      <th>Player Name</th>\n",
       "      <th>weight</th>\n",
       "      <th>height</th>\n",
       "      <th>bats</th>\n",
       "      <th>throws</th>\n",
       "      <th>Season</th>\n",
       "      <th>League</th>\n",
       "      <th>teamID</th>\n",
       "      <th>Team</th>\n",
       "      <th>...</th>\n",
       "      <th>RBI</th>\n",
       "      <th>SB</th>\n",
       "      <th>BB</th>\n",
       "      <th>SO</th>\n",
       "      <th>salary</th>\n",
       "      <th>IBB</th>\n",
       "      <th>HBP</th>\n",
       "      <th>SH</th>\n",
       "      <th>SF</th>\n",
       "      <th>PA</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>aardsda01</td>\n",
       "      <td>David Aardsma</td>\n",
       "      <td>215</td>\n",
       "      <td>75</td>\n",
       "      <td>R</td>\n",
       "      <td>R</td>\n",
       "      <td>2004</td>\n",
       "      <td>NL</td>\n",
       "      <td>SFN</td>\n",
       "      <td>San Francisco Giants</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>300000</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>aardsda01</td>\n",
       "      <td>David Aardsma</td>\n",
       "      <td>215</td>\n",
       "      <td>75</td>\n",
       "      <td>R</td>\n",
       "      <td>R</td>\n",
       "      <td>2007</td>\n",
       "      <td>AL</td>\n",
       "      <td>CHA</td>\n",
       "      <td>Chicago White Sox</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>387500</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>aardsda01</td>\n",
       "      <td>David Aardsma</td>\n",
       "      <td>215</td>\n",
       "      <td>75</td>\n",
       "      <td>R</td>\n",
       "      <td>R</td>\n",
       "      <td>2008</td>\n",
       "      <td>AL</td>\n",
       "      <td>BOS</td>\n",
       "      <td>Boston Red Sox</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>403250</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>aardsda01</td>\n",
       "      <td>David Aardsma</td>\n",
       "      <td>215</td>\n",
       "      <td>75</td>\n",
       "      <td>R</td>\n",
       "      <td>R</td>\n",
       "      <td>2009</td>\n",
       "      <td>AL</td>\n",
       "      <td>SEA</td>\n",
       "      <td>Seattle Mariners</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>419000</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>4</th>\n",
       "      <td>aardsda01</td>\n",
       "      <td>David Aardsma</td>\n",
       "      <td>215</td>\n",
       "      <td>75</td>\n",
       "      <td>R</td>\n",
       "      <td>R</td>\n",
       "      <td>2012</td>\n",
       "      <td>AL</td>\n",
       "      <td>NYA</td>\n",
       "      <td>New York Yankees</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>500000</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>5 rows × 28 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    playerID    Player Name  weight  height bats throws  Season League teamID  \\\n",
       "0  aardsda01  David Aardsma     215      75    R      R    2004     NL    SFN   \n",
       "1  aardsda01  David Aardsma     215      75    R      R    2007     AL    CHA   \n",
       "2  aardsda01  David Aardsma     215      75    R      R    2008     AL    BOS   \n",
       "3  aardsda01  David Aardsma     215      75    R      R    2009     AL    SEA   \n",
       "4  aardsda01  David Aardsma     215      75    R      R    2012     AL    NYA   \n",
       "\n",
       "                   Team  ... RBI  SB  BB  SO  salary  IBB  HBP  SH  SF  PA  \n",
       "0  San Francisco Giants  ...   0   0   0   0  300000    0    0   0   0   0  \n",
       "1     Chicago White Sox  ...   0   0   0   0  387500    0    0   0   0   0  \n",
       "2        Boston Red Sox  ...   0   0   0   1  403250    0    0   0   0   1  \n",
       "3      Seattle Mariners  ...   0   0   0   0  419000    0    0   0   0   0  \n",
       "4      New York Yankees  ...   0   0   0   0  500000    0    0   0   0   0  \n",
       "\n",
       "[5 rows x 28 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv('MLB Stats.csv')\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0        False\n",
       "1        False\n",
       "2        False\n",
       "3        False\n",
       "4        False\n",
       "         ...  \n",
       "28288    False\n",
       "28289    False\n",
       "28290    False\n",
       "28291    False\n",
       "28292     True\n",
       "Name: Season, Length: 28293, dtype: bool"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "is_2016 =  data['Season']==2016\n",
    "is_2016"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>playerID</th>\n",
       "      <th>Player Name</th>\n",
       "      <th>weight</th>\n",
       "      <th>height</th>\n",
       "      <th>bats</th>\n",
       "      <th>throws</th>\n",
       "      <th>Season</th>\n",
       "      <th>League</th>\n",
       "      <th>teamID</th>\n",
       "      <th>Team</th>\n",
       "      <th>...</th>\n",
       "      <th>RBI</th>\n",
       "      <th>SB</th>\n",
       "      <th>BB</th>\n",
       "      <th>SO</th>\n",
       "      <th>salary</th>\n",
       "      <th>IBB</th>\n",
       "      <th>HBP</th>\n",
       "      <th>SH</th>\n",
       "      <th>SF</th>\n",
       "      <th>PA</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>abadfe01</td>\n",
       "      <td>Fernando Abad</td>\n",
       "      <td>220</td>\n",
       "      <td>73</td>\n",
       "      <td>L</td>\n",
       "      <td>L</td>\n",
       "      <td>2016</td>\n",
       "      <td>AL</td>\n",
       "      <td>BOS</td>\n",
       "      <td>Boston Red Sox</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1250000</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>16</th>\n",
       "      <td>abadfe01</td>\n",
       "      <td>Fernando Abad</td>\n",
       "      <td>220</td>\n",
       "      <td>73</td>\n",
       "      <td>L</td>\n",
       "      <td>L</td>\n",
       "      <td>2016</td>\n",
       "      <td>AL</td>\n",
       "      <td>MIN</td>\n",
       "      <td>Minnesota Twins</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1250000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>81</th>\n",
       "      <td>abreujo02</td>\n",
       "      <td>Jose Abreu</td>\n",
       "      <td>255</td>\n",
       "      <td>75</td>\n",
       "      <td>R</td>\n",
       "      <td>R</td>\n",
       "      <td>2016</td>\n",
       "      <td>AL</td>\n",
       "      <td>CHA</td>\n",
       "      <td>Chicago White Sox</td>\n",
       "      <td>...</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>47</td>\n",
       "      <td>125</td>\n",
       "      <td>11666667</td>\n",
       "      <td>7</td>\n",
       "      <td>15</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>702</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>117</th>\n",
       "      <td>ackledu01</td>\n",
       "      <td>Dustin Ackley</td>\n",
       "      <td>205</td>\n",
       "      <td>73</td>\n",
       "      <td>L</td>\n",
       "      <td>R</td>\n",
       "      <td>2016</td>\n",
       "      <td>AL</td>\n",
       "      <td>NYA</td>\n",
       "      <td>New York Yankees</td>\n",
       "      <td>...</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>9</td>\n",
       "      <td>3200000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>124</th>\n",
       "      <td>adamecr01</td>\n",
       "      <td>Cristhian Adames</td>\n",
       "      <td>185</td>\n",
       "      <td>72</td>\n",
       "      <td>B</td>\n",
       "      <td>R</td>\n",
       "      <td>2016</td>\n",
       "      <td>NL</td>\n",
       "      <td>COL</td>\n",
       "      <td>Colorado Rockies</td>\n",
       "      <td>...</td>\n",
       "      <td>17</td>\n",
       "      <td>2</td>\n",
       "      <td>24</td>\n",
       "      <td>47</td>\n",
       "      <td>509500</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>256</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 28 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      playerID       Player Name  weight  height bats throws  Season League  \\\n",
       "15    abadfe01     Fernando Abad     220      73    L      L    2016     AL   \n",
       "16    abadfe01     Fernando Abad     220      73    L      L    2016     AL   \n",
       "81   abreujo02        Jose Abreu     255      75    R      R    2016     AL   \n",
       "117  ackledu01     Dustin Ackley     205      73    L      R    2016     AL   \n",
       "124  adamecr01  Cristhian Adames     185      72    B      R    2016     NL   \n",
       "\n",
       "    teamID               Team  ...  RBI  SB  BB   SO    salary  IBB  HBP  SH  \\\n",
       "15     BOS     Boston Red Sox  ...    0   0   0    0   1250000    0    0   0   \n",
       "16     MIN    Minnesota Twins  ...    0   0   0    1   1250000    0    0   0   \n",
       "81     CHA  Chicago White Sox  ...  100   0  47  125  11666667    7   15   0   \n",
       "117    NYA   New York Yankees  ...    4   0   8    9   3200000    0    0   0   \n",
       "124    COL   Colorado Rockies  ...   17   2  24   47    509500    0    4   3   \n",
       "\n",
       "     SF   PA  \n",
       "15    0    0  \n",
       "16    0    1  \n",
       "81    9  702  \n",
       "117   1   70  \n",
       "124   0  256  \n",
       "\n",
       "[5 rows x 28 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_2016 = data[is_2016]\n",
    "data_2016.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>playerID</th>\n",
       "      <th>Player Name</th>\n",
       "      <th>weight</th>\n",
       "      <th>height</th>\n",
       "      <th>bats</th>\n",
       "      <th>throws</th>\n",
       "      <th>Season</th>\n",
       "      <th>League</th>\n",
       "      <th>teamID</th>\n",
       "      <th>Team</th>\n",
       "      <th>...</th>\n",
       "      <th>RBI</th>\n",
       "      <th>SB</th>\n",
       "      <th>BB</th>\n",
       "      <th>SO</th>\n",
       "      <th>salary</th>\n",
       "      <th>IBB</th>\n",
       "      <th>HBP</th>\n",
       "      <th>SH</th>\n",
       "      <th>SF</th>\n",
       "      <th>PA</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>81</th>\n",
       "      <td>abreujo02</td>\n",
       "      <td>Jose Abreu</td>\n",
       "      <td>255</td>\n",
       "      <td>75</td>\n",
       "      <td>R</td>\n",
       "      <td>R</td>\n",
       "      <td>2016</td>\n",
       "      <td>AL</td>\n",
       "      <td>CHA</td>\n",
       "      <td>Chicago White Sox</td>\n",
       "      <td>...</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>47</td>\n",
       "      <td>125</td>\n",
       "      <td>11666667</td>\n",
       "      <td>7</td>\n",
       "      <td>15</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>702</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>117</th>\n",
       "      <td>ackledu01</td>\n",
       "      <td>Dustin Ackley</td>\n",
       "      <td>205</td>\n",
       "      <td>73</td>\n",
       "      <td>L</td>\n",
       "      <td>R</td>\n",
       "      <td>2016</td>\n",
       "      <td>AL</td>\n",
       "      <td>NYA</td>\n",
       "      <td>New York Yankees</td>\n",
       "      <td>...</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>9</td>\n",
       "      <td>3200000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>124</th>\n",
       "      <td>adamecr01</td>\n",
       "      <td>Cristhian Adames</td>\n",
       "      <td>185</td>\n",
       "      <td>72</td>\n",
       "      <td>B</td>\n",
       "      <td>R</td>\n",
       "      <td>2016</td>\n",
       "      <td>NL</td>\n",
       "      <td>COL</td>\n",
       "      <td>Colorado Rockies</td>\n",
       "      <td>...</td>\n",
       "      <td>17</td>\n",
       "      <td>2</td>\n",
       "      <td>24</td>\n",
       "      <td>47</td>\n",
       "      <td>509500</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>132</th>\n",
       "      <td>adamsma01</td>\n",
       "      <td>Matt Adams</td>\n",
       "      <td>245</td>\n",
       "      <td>75</td>\n",
       "      <td>L</td>\n",
       "      <td>R</td>\n",
       "      <td>2016</td>\n",
       "      <td>NL</td>\n",
       "      <td>SLN</td>\n",
       "      <td>St. Louis Cardinals</td>\n",
       "      <td>...</td>\n",
       "      <td>54</td>\n",
       "      <td>0</td>\n",
       "      <td>25</td>\n",
       "      <td>81</td>\n",
       "      <td>1650000</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>328</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>162</th>\n",
       "      <td>adriaeh01</td>\n",
       "      <td>Ehire Adrianza</td>\n",
       "      <td>170</td>\n",
       "      <td>73</td>\n",
       "      <td>B</td>\n",
       "      <td>R</td>\n",
       "      <td>2016</td>\n",
       "      <td>NL</td>\n",
       "      <td>SFN</td>\n",
       "      <td>San Francisco Giants</td>\n",
       "      <td>...</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>13</td>\n",
       "      <td>512500</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>71</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 28 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      playerID       Player Name  weight  height bats throws  Season League  \\\n",
       "81   abreujo02        Jose Abreu     255      75    R      R    2016     AL   \n",
       "117  ackledu01     Dustin Ackley     205      73    L      R    2016     AL   \n",
       "124  adamecr01  Cristhian Adames     185      72    B      R    2016     NL   \n",
       "132  adamsma01        Matt Adams     245      75    L      R    2016     NL   \n",
       "162  adriaeh01    Ehire Adrianza     170      73    B      R    2016     NL   \n",
       "\n",
       "    teamID                  Team  ...  RBI  SB  BB   SO    salary  IBB  HBP  \\\n",
       "81     CHA     Chicago White Sox  ...  100   0  47  125  11666667    7   15   \n",
       "117    NYA      New York Yankees  ...    4   0   8    9   3200000    0    0   \n",
       "124    COL      Colorado Rockies  ...   17   2  24   47    509500    0    4   \n",
       "132    SLN   St. Louis Cardinals  ...   54   0  25   81   1650000    1    2   \n",
       "162    SFN  San Francisco Giants  ...    7   0   2   13    512500    0    2   \n",
       "\n",
       "     SH  SF   PA  \n",
       "81    0   9  702  \n",
       "117   0   1   70  \n",
       "124   3   0  256  \n",
       "132   0   3  328  \n",
       "162   4   0   71  \n",
       "\n",
       "[5 rows x 28 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_2016 = data_2016.drop_duplicates(subset=['playerID'], keep=False)\n",
    "data_2016.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def fit_linear_reg(X,Y):\n",
    "    #Fit linear regression model and return RSS and R squared values\n",
    "    model_k = linear_model.LinearRegression(fit_intercept = True)\n",
    "    model_k.fit(X,Y)\n",
    "    RSS = mean_squared_error(Y,model_k.predict(X)) * len(Y)\n",
    "    R_squared = model_k.score(X,Y)\n",
    "    return RSS, R_squared"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:13: TqdmDeprecationWarning: Please use `tqdm.notebook.trange` instead of `tqdm.tnrange`\n",
      "  del sys.path[0]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "48c7f2ca40644bceb53dbf4e25ba6ae1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Loop...', max=18.0, style=ProgressStyle(description_width…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "#Importing tqdm for the progress bar\n",
    "from tqdm import tnrange, tqdm_notebook\n",
    "\n",
    "#Initialization variables\n",
    "Y = data_2016.salary\n",
    "X = data_2016.drop(columns = ['salary', 'playerID', 'Player Name', 'bats', 'throws', 'Season', 'League', 'teamID',\n",
    "                           'Team', 'Franchise'], axis = 1)\n",
    "k = 18\n",
    "RSS_list, R_squared_list, feature_list = [],[], []\n",
    "numb_features = []\n",
    "\n",
    "#Looping over k = 1 to k = 11 features in X\n",
    "for k in tnrange(1,len(X.columns) + 1, desc = 'Loop...'):\n",
    "\n",
    "    #Looping over all possible combinations: from 18 choose k\n",
    "    for combo in itertools.combinations(X.columns,k):\n",
    "        tmp_result = fit_linear_reg(X[list(combo)],Y)   #Store temp result \n",
    "        RSS_list.append(tmp_result[0])                  #Append lists\n",
    "        R_squared_list.append(tmp_result[1])\n",
    "        feature_list.append(combo)\n",
    "        numb_features.append(len(combo))   \n",
    "\n",
    "#Store in DataFrame\n",
    "df = pd.DataFrame({'numb_features': numb_features,'RSS': RSS_list, 'R_squared':R_squared_list,'features':feature_list})"
   ]
  },
  {
   "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>numb_features</th>\n",
       "      <th>RSS</th>\n",
       "      <th>R_squared</th>\n",
       "      <th>features</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1</td>\n",
       "      <td>2.460153e+16</td>\n",
       "      <td>0.070481</td>\n",
       "      <td>(RBI,)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>2</td>\n",
       "      <td>2.391391e+16</td>\n",
       "      <td>0.096461</td>\n",
       "      <td>(weight, RBI)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>273</th>\n",
       "      <td>3</td>\n",
       "      <td>2.344846e+16</td>\n",
       "      <td>0.114047</td>\n",
       "      <td>(weight, RBI, SO)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1590</th>\n",
       "      <td>4</td>\n",
       "      <td>2.303185e+16</td>\n",
       "      <td>0.129788</td>\n",
       "      <td>(weight, RBI, BB, SO)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5295</th>\n",
       "      <td>5</td>\n",
       "      <td>2.262757e+16</td>\n",
       "      <td>0.145063</td>\n",
       "      <td>(weight, AB, 3B, SO, PA)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14668</th>\n",
       "      <td>6</td>\n",
       "      <td>2.215402e+16</td>\n",
       "      <td>0.162955</td>\n",
       "      <td>(weight, G, AB, 3B, SO, PA)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36087</th>\n",
       "      <td>7</td>\n",
       "      <td>2.190074e+16</td>\n",
       "      <td>0.172525</td>\n",
       "      <td>(weight, G, AB, 2B, 3B, SO, PA)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>83714</th>\n",
       "      <td>8</td>\n",
       "      <td>2.168366e+16</td>\n",
       "      <td>0.180727</td>\n",
       "      <td>(height, G, AB, 2B, 3B, RBI, SO, PA)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>108024</th>\n",
       "      <td>9</td>\n",
       "      <td>2.151299e+16</td>\n",
       "      <td>0.187175</td>\n",
       "      <td>(weight, height, G, AB, 2B, 3B, RBI, SO, PA)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>157567</th>\n",
       "      <td>10</td>\n",
       "      <td>2.140313e+16</td>\n",
       "      <td>0.191326</td>\n",
       "      <td>(weight, height, G, AB, 2B, 3B, RBI, BB, SO, SH)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>201907</th>\n",
       "      <td>11</td>\n",
       "      <td>2.136218e+16</td>\n",
       "      <td>0.192873</td>\n",
       "      <td>(weight, height, G, AB, 2B, 3B, RBI, SB, BB, S...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>233624</th>\n",
       "      <td>12</td>\n",
       "      <td>2.133236e+16</td>\n",
       "      <td>0.194000</td>\n",
       "      <td>(weight, height, G, AB, 2B, 3B, RBI, SB, SO, H...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>251393</th>\n",
       "      <td>13</td>\n",
       "      <td>2.131800e+16</td>\n",
       "      <td>0.194543</td>\n",
       "      <td>(weight, height, G, AB, 2B, 3B, RBI, SB, BB, S...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>258894</th>\n",
       "      <td>14</td>\n",
       "      <td>2.130811e+16</td>\n",
       "      <td>0.194916</td>\n",
       "      <td>(weight, height, G, AB, H, 2B, 3B, RBI, SB, BB...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>261441</th>\n",
       "      <td>15</td>\n",
       "      <td>2.130212e+16</td>\n",
       "      <td>0.195143</td>\n",
       "      <td>(weight, height, G, AB, H, 2B, 3B, HR, RBI, SB...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>261971</th>\n",
       "      <td>16</td>\n",
       "      <td>2.130196e+16</td>\n",
       "      <td>0.195149</td>\n",
       "      <td>(weight, height, G, AB, R, H, 2B, 3B, HR, RBI,...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>262126</th>\n",
       "      <td>17</td>\n",
       "      <td>2.130195e+16</td>\n",
       "      <td>0.195149</td>\n",
       "      <td>(weight, height, G, AB, R, H, 2B, 3B, HR, RBI,...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>262142</th>\n",
       "      <td>18</td>\n",
       "      <td>2.130195e+16</td>\n",
       "      <td>0.195149</td>\n",
       "      <td>(weight, height, G, AB, R, H, 2B, 3B, HR, RBI,...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        numb_features           RSS  R_squared  \\\n",
       "9                   1  2.460153e+16   0.070481   \n",
       "26                  2  2.391391e+16   0.096461   \n",
       "273                 3  2.344846e+16   0.114047   \n",
       "1590                4  2.303185e+16   0.129788   \n",
       "5295                5  2.262757e+16   0.145063   \n",
       "14668               6  2.215402e+16   0.162955   \n",
       "36087               7  2.190074e+16   0.172525   \n",
       "83714               8  2.168366e+16   0.180727   \n",
       "108024              9  2.151299e+16   0.187175   \n",
       "157567             10  2.140313e+16   0.191326   \n",
       "201907             11  2.136218e+16   0.192873   \n",
       "233624             12  2.133236e+16   0.194000   \n",
       "251393             13  2.131800e+16   0.194543   \n",
       "258894             14  2.130811e+16   0.194916   \n",
       "261441             15  2.130212e+16   0.195143   \n",
       "261971             16  2.130196e+16   0.195149   \n",
       "262126             17  2.130195e+16   0.195149   \n",
       "262142             18  2.130195e+16   0.195149   \n",
       "\n",
       "                                                 features  \n",
       "9                                                  (RBI,)  \n",
       "26                                          (weight, RBI)  \n",
       "273                                     (weight, RBI, SO)  \n",
       "1590                                (weight, RBI, BB, SO)  \n",
       "5295                             (weight, AB, 3B, SO, PA)  \n",
       "14668                         (weight, G, AB, 3B, SO, PA)  \n",
       "36087                     (weight, G, AB, 2B, 3B, SO, PA)  \n",
       "83714                (height, G, AB, 2B, 3B, RBI, SO, PA)  \n",
       "108024       (weight, height, G, AB, 2B, 3B, RBI, SO, PA)  \n",
       "157567   (weight, height, G, AB, 2B, 3B, RBI, BB, SO, SH)  \n",
       "201907  (weight, height, G, AB, 2B, 3B, RBI, SB, BB, S...  \n",
       "233624  (weight, height, G, AB, 2B, 3B, RBI, SB, SO, H...  \n",
       "251393  (weight, height, G, AB, 2B, 3B, RBI, SB, BB, S...  \n",
       "258894  (weight, height, G, AB, H, 2B, 3B, RBI, SB, BB...  \n",
       "261441  (weight, height, G, AB, H, 2B, 3B, HR, RBI, SB...  \n",
       "261971  (weight, height, G, AB, R, H, 2B, 3B, HR, RBI,...  \n",
       "262126  (weight, height, G, AB, R, H, 2B, 3B, HR, RBI,...  \n",
       "262142  (weight, height, G, AB, R, H, 2B, 3B, HR, RBI,...  "
      ]
     },
     "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>numb_features</th>\n",
       "      <th>RSS</th>\n",
       "      <th>R_squared</th>\n",
       "      <th>features</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1</td>\n",
       "      <td>2.460153e+16</td>\n",
       "      <td>0.070481</td>\n",
       "      <td>(RBI,)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>2</td>\n",
       "      <td>2.391391e+16</td>\n",
       "      <td>0.096461</td>\n",
       "      <td>(weight, RBI)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>273</th>\n",
       "      <td>3</td>\n",
       "      <td>2.344846e+16</td>\n",
       "      <td>0.114047</td>\n",
       "      <td>(weight, RBI, SO)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1590</th>\n",
       "      <td>4</td>\n",
       "      <td>2.303185e+16</td>\n",
       "      <td>0.129788</td>\n",
       "      <td>(weight, RBI, BB, SO)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5295</th>\n",
       "      <td>5</td>\n",
       "      <td>2.262757e+16</td>\n",
       "      <td>0.145063</td>\n",
       "      <td>(weight, AB, 3B, SO, PA)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14668</th>\n",
       "      <td>6</td>\n",
       "      <td>2.215402e+16</td>\n",
       "      <td>0.162955</td>\n",
       "      <td>(weight, G, AB, 3B, SO, PA)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36087</th>\n",
       "      <td>7</td>\n",
       "      <td>2.190074e+16</td>\n",
       "      <td>0.172525</td>\n",
       "      <td>(weight, G, AB, 2B, 3B, SO, PA)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>83714</th>\n",
       "      <td>8</td>\n",
       "      <td>2.168366e+16</td>\n",
       "      <td>0.180727</td>\n",
       "      <td>(height, G, AB, 2B, 3B, RBI, SO, PA)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>108024</th>\n",
       "      <td>9</td>\n",
       "      <td>2.151299e+16</td>\n",
       "      <td>0.187175</td>\n",
       "      <td>(weight, height, G, AB, 2B, 3B, RBI, SO, PA)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>157567</th>\n",
       "      <td>10</td>\n",
       "      <td>2.140313e+16</td>\n",
       "      <td>0.191326</td>\n",
       "      <td>(weight, height, G, AB, 2B, 3B, RBI, BB, SO, SH)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>201907</th>\n",
       "      <td>11</td>\n",
       "      <td>2.136218e+16</td>\n",
       "      <td>0.192873</td>\n",
       "      <td>(weight, height, G, AB, 2B, 3B, RBI, SB, BB, S...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>233624</th>\n",
       "      <td>12</td>\n",
       "      <td>2.133236e+16</td>\n",
       "      <td>0.194000</td>\n",
       "      <td>(weight, height, G, AB, 2B, 3B, RBI, SB, SO, H...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>251393</th>\n",
       "      <td>13</td>\n",
       "      <td>2.131800e+16</td>\n",
       "      <td>0.194543</td>\n",
       "      <td>(weight, height, G, AB, 2B, 3B, RBI, SB, BB, S...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>258894</th>\n",
       "      <td>14</td>\n",
       "      <td>2.130811e+16</td>\n",
       "      <td>0.194916</td>\n",
       "      <td>(weight, height, G, AB, H, 2B, 3B, RBI, SB, BB...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>261441</th>\n",
       "      <td>15</td>\n",
       "      <td>2.130212e+16</td>\n",
       "      <td>0.195143</td>\n",
       "      <td>(weight, height, G, AB, H, 2B, 3B, HR, RBI, SB...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>261971</th>\n",
       "      <td>16</td>\n",
       "      <td>2.130196e+16</td>\n",
       "      <td>0.195149</td>\n",
       "      <td>(weight, height, G, AB, R, H, 2B, 3B, HR, RBI,...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>262126</th>\n",
       "      <td>17</td>\n",
       "      <td>2.130195e+16</td>\n",
       "      <td>0.195149</td>\n",
       "      <td>(weight, height, G, AB, R, H, 2B, 3B, HR, RBI,...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>262142</th>\n",
       "      <td>18</td>\n",
       "      <td>2.130195e+16</td>\n",
       "      <td>0.195149</td>\n",
       "      <td>(weight, height, G, AB, R, H, 2B, 3B, HR, RBI,...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        numb_features           RSS  R_squared  \\\n",
       "9                   1  2.460153e+16   0.070481   \n",
       "26                  2  2.391391e+16   0.096461   \n",
       "273                 3  2.344846e+16   0.114047   \n",
       "1590                4  2.303185e+16   0.129788   \n",
       "5295                5  2.262757e+16   0.145063   \n",
       "14668               6  2.215402e+16   0.162955   \n",
       "36087               7  2.190074e+16   0.172525   \n",
       "83714               8  2.168366e+16   0.180727   \n",
       "108024              9  2.151299e+16   0.187175   \n",
       "157567             10  2.140313e+16   0.191326   \n",
       "201907             11  2.136218e+16   0.192873   \n",
       "233624             12  2.133236e+16   0.194000   \n",
       "251393             13  2.131800e+16   0.194543   \n",
       "258894             14  2.130811e+16   0.194916   \n",
       "261441             15  2.130212e+16   0.195143   \n",
       "261971             16  2.130196e+16   0.195149   \n",
       "262126             17  2.130195e+16   0.195149   \n",
       "262142             18  2.130195e+16   0.195149   \n",
       "\n",
       "                                                 features  \n",
       "9                                                  (RBI,)  \n",
       "26                                          (weight, RBI)  \n",
       "273                                     (weight, RBI, SO)  \n",
       "1590                                (weight, RBI, BB, SO)  \n",
       "5295                             (weight, AB, 3B, SO, PA)  \n",
       "14668                         (weight, G, AB, 3B, SO, PA)  \n",
       "36087                     (weight, G, AB, 2B, 3B, SO, PA)  \n",
       "83714                (height, G, AB, 2B, 3B, RBI, SO, PA)  \n",
       "108024       (weight, height, G, AB, 2B, 3B, RBI, SO, PA)  \n",
       "157567   (weight, height, G, AB, 2B, 3B, RBI, BB, SO, SH)  \n",
       "201907  (weight, height, G, AB, 2B, 3B, RBI, SB, BB, S...  \n",
       "233624  (weight, height, G, AB, 2B, 3B, RBI, SB, SO, H...  \n",
       "251393  (weight, height, G, AB, 2B, 3B, RBI, SB, BB, S...  \n",
       "258894  (weight, height, G, AB, H, 2B, 3B, RBI, SB, BB...  \n",
       "261441  (weight, height, G, AB, H, 2B, 3B, HR, RBI, SB...  \n",
       "261971  (weight, height, G, AB, R, H, 2B, 3B, HR, RBI,...  \n",
       "262126  (weight, height, G, AB, R, H, 2B, 3B, HR, RBI,...  \n",
       "262142  (weight, height, G, AB, R, H, 2B, 3B, HR, RBI,...  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_min = df[df.groupby('numb_features')['RSS'].transform(min) == df['RSS']]\n",
    "df_max = df[df.groupby('numb_features')['R_squared'].transform(max) == df['R_squared']]\n",
    "display(df_min.head(18))\n",
    "display(df_max.head(18))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>numb_features</th>\n",
       "      <th>RSS</th>\n",
       "      <th>R_squared</th>\n",
       "      <th>features</th>\n",
       "      <th>min_RSS</th>\n",
       "      <th>max_R_squared</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>2.571075e+16</td>\n",
       "      <td>0.028571</td>\n",
       "      <td>(weight,)</td>\n",
       "      <td>2.460153e+16</td>\n",
       "      <td>0.070481</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>2.619389e+16</td>\n",
       "      <td>0.010317</td>\n",
       "      <td>(height,)</td>\n",
       "      <td>2.460153e+16</td>\n",
       "      <td>0.070481</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>2.577244e+16</td>\n",
       "      <td>0.026240</td>\n",
       "      <td>(G,)</td>\n",
       "      <td>2.460153e+16</td>\n",
       "      <td>0.070481</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>2.513295e+16</td>\n",
       "      <td>0.050402</td>\n",
       "      <td>(AB,)</td>\n",
       "      <td>2.460153e+16</td>\n",
       "      <td>0.070481</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>2.511591e+16</td>\n",
       "      <td>0.051046</td>\n",
       "      <td>(R,)</td>\n",
       "      <td>2.460153e+16</td>\n",
       "      <td>0.070481</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   numb_features           RSS  R_squared   features       min_RSS  \\\n",
       "0              1  2.571075e+16   0.028571  (weight,)  2.460153e+16   \n",
       "1              1  2.619389e+16   0.010317  (height,)  2.460153e+16   \n",
       "2              1  2.577244e+16   0.026240       (G,)  2.460153e+16   \n",
       "3              1  2.513295e+16   0.050402      (AB,)  2.460153e+16   \n",
       "4              1  2.511591e+16   0.051046       (R,)  2.460153e+16   \n",
       "\n",
       "   max_R_squared  \n",
       "0       0.070481  \n",
       "1       0.070481  \n",
       "2       0.070481  \n",
       "3       0.070481  \n",
       "4       0.070481  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['min_RSS'] = df.groupby('numb_features')['RSS'].transform(min)\n",
    "df['max_R_squared'] = df.groupby('numb_features')['R_squared'].transform(max)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\IPython\\core\\pylabtools.py:132: UserWarning: Creating legend with loc=\"best\" can be slow with large amounts of data.\n",
      "  fig.canvas.print_figure(bytes_io, **kw)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x432 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize = (16,6))\n",
    "ax = fig.add_subplot(1, 2, 1)\n",
    "\n",
    "ax.scatter(df.numb_features,df.RSS, alpha = .2, color = 'darkblue' )\n",
    "ax.set_xlabel('# Features')\n",
    "ax.set_ylabel('RSS')\n",
    "ax.set_title('RSS - Best subset selection')\n",
    "ax.plot(df.numb_features,df.min_RSS,color = 'r', label = 'Best subset')\n",
    "ax.legend()\n",
    "\n",
    "ax = fig.add_subplot(1, 2, 2)\n",
    "ax.scatter(df.numb_features,df.R_squared, alpha = .2, color = 'darkblue' )\n",
    "ax.plot(df.numb_features,df.max_R_squared,color = 'r', label = 'Best subset')\n",
    "ax.set_xlabel('# Features')\n",
    "ax.set_ylabel('R squared')\n",
    "ax.set_title('R_squared - Best subset selection')\n",
    "ax.legend()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "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>features</th>\n",
       "      <th>RSS</th>\n",
       "      <th>R_squared</th>\n",
       "      <th>numb_features</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>(weight,)</td>\n",
       "      <td>2.571075e+16</td>\n",
       "      <td>0.028571</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>(height,)</td>\n",
       "      <td>2.619389e+16</td>\n",
       "      <td>0.010317</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>(G,)</td>\n",
       "      <td>2.577244e+16</td>\n",
       "      <td>0.026240</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>(AB,)</td>\n",
       "      <td>2.513295e+16</td>\n",
       "      <td>0.050402</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>(R,)</td>\n",
       "      <td>2.511591e+16</td>\n",
       "      <td>0.051046</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>262138</th>\n",
       "      <td>(weight, height, G, R, H, 2B, 3B, HR, RBI, SB,...</td>\n",
       "      <td>2.130195e+16</td>\n",
       "      <td>0.195149</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>262139</th>\n",
       "      <td>(weight, height, AB, R, H, 2B, 3B, HR, RBI, SB...</td>\n",
       "      <td>2.171544e+16</td>\n",
       "      <td>0.179526</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>262140</th>\n",
       "      <td>(weight, G, AB, R, H, 2B, 3B, HR, RBI, SB, BB,...</td>\n",
       "      <td>2.145061e+16</td>\n",
       "      <td>0.189532</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>262141</th>\n",
       "      <td>(height, G, AB, R, H, 2B, 3B, HR, RBI, SB, BB,...</td>\n",
       "      <td>2.147949e+16</td>\n",
       "      <td>0.188441</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>262142</th>\n",
       "      <td>(weight, height, G, AB, R, H, 2B, 3B, HR, RBI,...</td>\n",
       "      <td>2.130195e+16</td>\n",
       "      <td>0.195149</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>262143 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 features           RSS  \\\n",
       "0                                               (weight,)  2.571075e+16   \n",
       "1                                               (height,)  2.619389e+16   \n",
       "2                                                    (G,)  2.577244e+16   \n",
       "3                                                   (AB,)  2.513295e+16   \n",
       "4                                                    (R,)  2.511591e+16   \n",
       "...                                                   ...           ...   \n",
       "262138  (weight, height, G, R, H, 2B, 3B, HR, RBI, SB,...  2.130195e+16   \n",
       "262139  (weight, height, AB, R, H, 2B, 3B, HR, RBI, SB...  2.171544e+16   \n",
       "262140  (weight, G, AB, R, H, 2B, 3B, HR, RBI, SB, BB,...  2.145061e+16   \n",
       "262141  (height, G, AB, R, H, 2B, 3B, HR, RBI, SB, BB,...  2.147949e+16   \n",
       "262142  (weight, height, G, AB, R, H, 2B, 3B, HR, RBI,...  2.130195e+16   \n",
       "\n",
       "        R_squared  numb_features  \n",
       "0        0.028571              1  \n",
       "1        0.010317              1  \n",
       "2        0.026240              1  \n",
       "3        0.050402              1  \n",
       "4        0.051046              1  \n",
       "...           ...            ...  \n",
       "262138   0.195149             17  \n",
       "262139   0.179526             17  \n",
       "262140   0.189532             17  \n",
       "262141   0.188441             17  \n",
       "262142   0.195149             18  \n",
       "\n",
       "[262143 rows x 4 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1 = pd.concat([pd.DataFrame({'features':feature_list}),pd.DataFrame({'RSS':RSS_list, 'R_squared': R_squared_list})], axis=1, join='inner')\n",
    "df1['numb_features'] = df1['features'].str.len()\n",
    "df1"
   ]
  },
  {
   "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>features</th>\n",
       "      <th>RSS</th>\n",
       "      <th>R_squared</th>\n",
       "      <th>numb_features</th>\n",
       "      <th>C_p</th>\n",
       "      <th>AIC</th>\n",
       "      <th>BIC</th>\n",
       "      <th>R_squared_adj</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>83714</th>\n",
       "      <td>(height, G, AB, 2B, 3B, RBI, SO, PA)</td>\n",
       "      <td>2.168366e+16</td>\n",
       "      <td>0.180727</td>\n",
       "      <td>8</td>\n",
       "      <td>3.048839e+13</td>\n",
       "      <td>1.013324</td>\n",
       "      <td>1.063821</td>\n",
       "      <td>0.171598</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>72274</th>\n",
       "      <td>(weight, G, AB, 2B, 3B, RBI, SO, PA)</td>\n",
       "      <td>2.169271e+16</td>\n",
       "      <td>0.180385</td>\n",
       "      <td>8</td>\n",
       "      <td>3.050083e+13</td>\n",
       "      <td>1.013738</td>\n",
       "      <td>1.064235</td>\n",
       "      <td>0.171253</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>86641</th>\n",
       "      <td>(height, G, 2B, 3B, RBI, BB, SO, PA)</td>\n",
       "      <td>2.169938e+16</td>\n",
       "      <td>0.180133</td>\n",
       "      <td>8</td>\n",
       "      <td>3.051001e+13</td>\n",
       "      <td>1.014043</td>\n",
       "      <td>1.064540</td>\n",
       "      <td>0.170998</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75201</th>\n",
       "      <td>(weight, G, 2B, 3B, RBI, BB, SO, PA)</td>\n",
       "      <td>2.170101e+16</td>\n",
       "      <td>0.180071</td>\n",
       "      <td>8</td>\n",
       "      <td>3.051225e+13</td>\n",
       "      <td>1.014117</td>\n",
       "      <td>1.064614</td>\n",
       "      <td>0.170936</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36087</th>\n",
       "      <td>(weight, G, AB, 2B, 3B, SO, PA)</td>\n",
       "      <td>2.190074e+16</td>\n",
       "      <td>0.172525</td>\n",
       "      <td>7</td>\n",
       "      <td>3.070422e+13</td>\n",
       "      <td>1.020498</td>\n",
       "      <td>1.064683</td>\n",
       "      <td>0.164469</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                   features           RSS  R_squared  \\\n",
       "83714  (height, G, AB, 2B, 3B, RBI, SO, PA)  2.168366e+16   0.180727   \n",
       "72274  (weight, G, AB, 2B, 3B, RBI, SO, PA)  2.169271e+16   0.180385   \n",
       "86641  (height, G, 2B, 3B, RBI, BB, SO, PA)  2.169938e+16   0.180133   \n",
       "75201  (weight, G, 2B, 3B, RBI, BB, SO, PA)  2.170101e+16   0.180071   \n",
       "36087       (weight, G, AB, 2B, 3B, SO, PA)  2.190074e+16   0.172525   \n",
       "\n",
       "       numb_features           C_p       AIC       BIC  R_squared_adj  \n",
       "83714              8  3.048839e+13  1.013324  1.063821       0.171598  \n",
       "72274              8  3.050083e+13  1.013738  1.064235       0.171253  \n",
       "86641              8  3.051001e+13  1.014043  1.064540       0.170998  \n",
       "75201              8  3.051225e+13  1.014117  1.064614       0.170936  \n",
       "36087              7  3.070422e+13  1.020498  1.064683       0.164469  "
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m = len(Y)\n",
    "p = 18\n",
    "hat_sigma_squared = (1/(m - p -1)) * min(df1['RSS'])\n",
    "\n",
    "#Computing\n",
    "df1['C_p'] = (1/m) * (df1['RSS'] + 2 * df1['numb_features'] * hat_sigma_squared )\n",
    "df1['AIC'] = (1/(m*hat_sigma_squared)) * (df1['RSS'] + 2 * df1['numb_features'] * hat_sigma_squared )\n",
    "df1['BIC'] = (1/(m*hat_sigma_squared)) * (df1['RSS'] +  np.log(m) * df1['numb_features'] * hat_sigma_squared )\n",
    "df1['R_squared_adj'] = 1 - ( (1 - df1['R_squared'])*(m-1)/(m-df1['numb_features'] -1))\n",
    "df_BIC = df1.sort_values(by=['BIC'])\n",
    "df_BIC.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "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>features</th>\n",
       "      <th>RSS</th>\n",
       "      <th>R_squared</th>\n",
       "      <th>numb_features</th>\n",
       "      <th>C_p</th>\n",
       "      <th>AIC</th>\n",
       "      <th>BIC</th>\n",
       "      <th>R_squared_adj</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>201907</th>\n",
       "      <td>(weight, height, G, AB, 2B, 3B, RBI, SB, BB, S...</td>\n",
       "      <td>2.136218e+16</td>\n",
       "      <td>0.192873</td>\n",
       "      <td>11</td>\n",
       "      <td>3.029451e+13</td>\n",
       "      <td>1.006880</td>\n",
       "      <td>1.076314</td>\n",
       "      <td>0.180456</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>233624</th>\n",
       "      <td>(weight, height, G, AB, 2B, 3B, RBI, SB, SO, H...</td>\n",
       "      <td>2.133236e+16</td>\n",
       "      <td>0.194000</td>\n",
       "      <td>12</td>\n",
       "      <td>3.033626e+13</td>\n",
       "      <td>1.008268</td>\n",
       "      <td>1.084013</td>\n",
       "      <td>0.180454</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>233598</th>\n",
       "      <td>(weight, height, G, AB, 2B, 3B, RBI, SB, BB, S...</td>\n",
       "      <td>2.133310e+16</td>\n",
       "      <td>0.193972</td>\n",
       "      <td>12</td>\n",
       "      <td>3.033727e+13</td>\n",
       "      <td>1.008302</td>\n",
       "      <td>1.084047</td>\n",
       "      <td>0.180425</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205213</th>\n",
       "      <td>(weight, height, G, 2B, 3B, RBI, SB, BB, SO, S...</td>\n",
       "      <td>2.136536e+16</td>\n",
       "      <td>0.192753</td>\n",
       "      <td>11</td>\n",
       "      <td>3.029888e+13</td>\n",
       "      <td>1.007025</td>\n",
       "      <td>1.076459</td>\n",
       "      <td>0.180334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>235839</th>\n",
       "      <td>(weight, height, G, 2B, 3B, RBI, SB, BB, SO, H...</td>\n",
       "      <td>2.133945e+16</td>\n",
       "      <td>0.193732</td>\n",
       "      <td>12</td>\n",
       "      <td>3.034601e+13</td>\n",
       "      <td>1.008592</td>\n",
       "      <td>1.084338</td>\n",
       "      <td>0.180181</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 features           RSS  \\\n",
       "201907  (weight, height, G, AB, 2B, 3B, RBI, SB, BB, S...  2.136218e+16   \n",
       "233624  (weight, height, G, AB, 2B, 3B, RBI, SB, SO, H...  2.133236e+16   \n",
       "233598  (weight, height, G, AB, 2B, 3B, RBI, SB, BB, S...  2.133310e+16   \n",
       "205213  (weight, height, G, 2B, 3B, RBI, SB, BB, SO, S...  2.136536e+16   \n",
       "235839  (weight, height, G, 2B, 3B, RBI, SB, BB, SO, H...  2.133945e+16   \n",
       "\n",
       "        R_squared  numb_features           C_p       AIC       BIC  \\\n",
       "201907   0.192873             11  3.029451e+13  1.006880  1.076314   \n",
       "233624   0.194000             12  3.033626e+13  1.008268  1.084013   \n",
       "233598   0.193972             12  3.033727e+13  1.008302  1.084047   \n",
       "205213   0.192753             11  3.029888e+13  1.007025  1.076459   \n",
       "235839   0.193732             12  3.034601e+13  1.008592  1.084338   \n",
       "\n",
       "        R_squared_adj  \n",
       "201907       0.180456  \n",
       "233624       0.180454  \n",
       "233598       0.180425  \n",
       "205213       0.180334  \n",
       "235839       0.180181  "
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_R2 = df2.sort_values(by=['R_squared_adj'], ascending=False)\n",
    "df_R2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "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>features</th>\n",
       "      <th>RSS</th>\n",
       "      <th>R_squared</th>\n",
       "      <th>numb_features</th>\n",
       "      <th>C_p</th>\n",
       "      <th>AIC</th>\n",
       "      <th>BIC</th>\n",
       "      <th>R_squared_adj</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>157567</th>\n",
       "      <td>(weight, height, G, AB, 2B, 3B, RBI, BB, SO, SH)</td>\n",
       "      <td>2.140313e+16</td>\n",
       "      <td>0.191326</td>\n",
       "      <td>10</td>\n",
       "      <td>3.026806e+13</td>\n",
       "      <td>1.006001</td>\n",
       "      <td>1.069122</td>\n",
       "      <td>0.180032</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>161193</th>\n",
       "      <td>(weight, height, G, 2B, 3B, RBI, BB, SO, SH, PA)</td>\n",
       "      <td>2.140694e+16</td>\n",
       "      <td>0.191182</td>\n",
       "      <td>10</td>\n",
       "      <td>3.027330e+13</td>\n",
       "      <td>1.006175</td>\n",
       "      <td>1.069297</td>\n",
       "      <td>0.179886</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>201907</th>\n",
       "      <td>(weight, height, G, AB, 2B, 3B, RBI, SB, BB, S...</td>\n",
       "      <td>2.136218e+16</td>\n",
       "      <td>0.192873</td>\n",
       "      <td>11</td>\n",
       "      <td>3.029451e+13</td>\n",
       "      <td>1.006880</td>\n",
       "      <td>1.076314</td>\n",
       "      <td>0.180456</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205213</th>\n",
       "      <td>(weight, height, G, 2B, 3B, RBI, SB, BB, SO, S...</td>\n",
       "      <td>2.136536e+16</td>\n",
       "      <td>0.192753</td>\n",
       "      <td>11</td>\n",
       "      <td>3.029888e+13</td>\n",
       "      <td>1.007025</td>\n",
       "      <td>1.076459</td>\n",
       "      <td>0.180334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>157586</th>\n",
       "      <td>(weight, height, G, AB, 2B, 3B, RBI, SO, HBP, PA)</td>\n",
       "      <td>2.142846e+16</td>\n",
       "      <td>0.190369</td>\n",
       "      <td>10</td>\n",
       "      <td>3.030290e+13</td>\n",
       "      <td>1.007159</td>\n",
       "      <td>1.070281</td>\n",
       "      <td>0.179061</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 features           RSS  \\\n",
       "157567   (weight, height, G, AB, 2B, 3B, RBI, BB, SO, SH)  2.140313e+16   \n",
       "161193   (weight, height, G, 2B, 3B, RBI, BB, SO, SH, PA)  2.140694e+16   \n",
       "201907  (weight, height, G, AB, 2B, 3B, RBI, SB, BB, S...  2.136218e+16   \n",
       "205213  (weight, height, G, 2B, 3B, RBI, SB, BB, SO, S...  2.136536e+16   \n",
       "157586  (weight, height, G, AB, 2B, 3B, RBI, SO, HBP, PA)  2.142846e+16   \n",
       "\n",
       "        R_squared  numb_features           C_p       AIC       BIC  \\\n",
       "157567   0.191326             10  3.026806e+13  1.006001  1.069122   \n",
       "161193   0.191182             10  3.027330e+13  1.006175  1.069297   \n",
       "201907   0.192873             11  3.029451e+13  1.006880  1.076314   \n",
       "205213   0.192753             11  3.029888e+13  1.007025  1.076459   \n",
       "157586   0.190369             10  3.030290e+13  1.007159  1.070281   \n",
       "\n",
       "        R_squared_adj  \n",
       "157567       0.180032  \n",
       "161193       0.179886  \n",
       "201907       0.180456  \n",
       "205213       0.180334  \n",
       "157586       0.179061  "
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_Cp = df1.sort_values(by=['C_p'])\n",
    "df_Cp.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "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>features</th>\n",
       "      <th>RSS</th>\n",
       "      <th>R_squared</th>\n",
       "      <th>numb_features</th>\n",
       "      <th>C_p</th>\n",
       "      <th>AIC</th>\n",
       "      <th>BIC</th>\n",
       "      <th>R_squared_adj</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>157567</th>\n",
       "      <td>(weight, height, G, AB, 2B, 3B, RBI, BB, SO, SH)</td>\n",
       "      <td>2.140313e+16</td>\n",
       "      <td>0.191326</td>\n",
       "      <td>10</td>\n",
       "      <td>3.026806e+13</td>\n",
       "      <td>1.006001</td>\n",
       "      <td>1.069122</td>\n",
       "      <td>0.180032</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>161193</th>\n",
       "      <td>(weight, height, G, 2B, 3B, RBI, BB, SO, SH, PA)</td>\n",
       "      <td>2.140694e+16</td>\n",
       "      <td>0.191182</td>\n",
       "      <td>10</td>\n",
       "      <td>3.027330e+13</td>\n",
       "      <td>1.006175</td>\n",
       "      <td>1.069297</td>\n",
       "      <td>0.179886</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>201907</th>\n",
       "      <td>(weight, height, G, AB, 2B, 3B, RBI, SB, BB, S...</td>\n",
       "      <td>2.136218e+16</td>\n",
       "      <td>0.192873</td>\n",
       "      <td>11</td>\n",
       "      <td>3.029451e+13</td>\n",
       "      <td>1.006880</td>\n",
       "      <td>1.076314</td>\n",
       "      <td>0.180456</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205213</th>\n",
       "      <td>(weight, height, G, 2B, 3B, RBI, SB, BB, SO, S...</td>\n",
       "      <td>2.136536e+16</td>\n",
       "      <td>0.192753</td>\n",
       "      <td>11</td>\n",
       "      <td>3.029888e+13</td>\n",
       "      <td>1.007025</td>\n",
       "      <td>1.076459</td>\n",
       "      <td>0.180334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>157586</th>\n",
       "      <td>(weight, height, G, AB, 2B, 3B, RBI, SO, HBP, PA)</td>\n",
       "      <td>2.142846e+16</td>\n",
       "      <td>0.190369</td>\n",
       "      <td>10</td>\n",
       "      <td>3.030290e+13</td>\n",
       "      <td>1.007159</td>\n",
       "      <td>1.070281</td>\n",
       "      <td>0.179061</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 features           RSS  \\\n",
       "157567   (weight, height, G, AB, 2B, 3B, RBI, BB, SO, SH)  2.140313e+16   \n",
       "161193   (weight, height, G, 2B, 3B, RBI, BB, SO, SH, PA)  2.140694e+16   \n",
       "201907  (weight, height, G, AB, 2B, 3B, RBI, SB, BB, S...  2.136218e+16   \n",
       "205213  (weight, height, G, 2B, 3B, RBI, SB, BB, SO, S...  2.136536e+16   \n",
       "157586  (weight, height, G, AB, 2B, 3B, RBI, SO, HBP, PA)  2.142846e+16   \n",
       "\n",
       "        R_squared  numb_features           C_p       AIC       BIC  \\\n",
       "157567   0.191326             10  3.026806e+13  1.006001  1.069122   \n",
       "161193   0.191182             10  3.027330e+13  1.006175  1.069297   \n",
       "201907   0.192873             11  3.029451e+13  1.006880  1.076314   \n",
       "205213   0.192753             11  3.029888e+13  1.007025  1.076459   \n",
       "157586   0.190369             10  3.030290e+13  1.007159  1.070281   \n",
       "\n",
       "        R_squared_adj  \n",
       "157567       0.180032  \n",
       "161193       0.179886  \n",
       "201907       0.180456  \n",
       "205213       0.180334  \n",
       "157586       0.179061  "
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_AIC = df1.sort_values(by=['AIC'])\n",
    "df_AIC.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}