4-Information gain - mutual information In Regression.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature Selection-Information gain - mutual information In Regression Problem Statements"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Mutual Information \n",
    "Estimate mutual information for a continuous target variable.\n",
    "\n",
    "Mutual information (MI) between two random variables is a non-negative value, which measures the dependency between the variables. It is equal to zero if and only if two random variables are independent, and higher values mean higher dependency.\n",
    "\n",
    "The function relies on nonparametric methods based on entropy estimation from k-nearest neighbors distances \n",
    "\n",
    "Mutual information is calculated between two variables and measures the reduction in uncertainty for one variable given a known value of the other variable.\n",
    "\n",
    "Inshort\n",
    "\n",
    "A quantity called mutual information measures the amount of information one can obtain from one random variable given another.\n",
    "\n",
    "The mutual information between two random variables X and Y can be stated formally as follows:\n",
    "\n",
    "<b>I(X ; Y) = H(X) – H(X | Y)<b>\n",
    "Where I(X ; Y) is the mutual information for X and Y, H(X) is the entropy for X and H(X | Y) is the conditional entropy for X given Y. The result has the units of bits."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Youtube Videos\n",
    "\n",
    "Entropy: https://www.youtube.com/watch?v=1IQOtJ4NI_0\n",
    "        \n",
    "Information Gain: https://www.youtube.com/watch?v=FuTRucXB9rA\n",
    "        \n",
    "Gini Impurity: https://www.youtube.com/watch?v=5aIFgrrTqOw\n",
    "\n",
    "Statistical test: https://www.youtube.com/watch?v=4-rxTA_5_xA\n",
    "\n",
    "https://www.youtube.com/watch?v=YrhlQB3mQFI"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "#https://www.kaggle.com/c/house-prices-advanced-regression-techniques/data\n",
    "import pandas as pd\n",
    "housing_df=pd.read_csv('housing_data.csv')"
   ]
  },
  {
   "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>Id</th>\n",
       "      <th>MSSubClass</th>\n",
       "      <th>MSZoning</th>\n",
       "      <th>LotFrontage</th>\n",
       "      <th>LotArea</th>\n",
       "      <th>Street</th>\n",
       "      <th>Alley</th>\n",
       "      <th>LotShape</th>\n",
       "      <th>LandContour</th>\n",
       "      <th>Utilities</th>\n",
       "      <th>...</th>\n",
       "      <th>PoolArea</th>\n",
       "      <th>PoolQC</th>\n",
       "      <th>Fence</th>\n",
       "      <th>MiscFeature</th>\n",
       "      <th>MiscVal</th>\n",
       "      <th>MoSold</th>\n",
       "      <th>YrSold</th>\n",
       "      <th>SaleType</th>\n",
       "      <th>SaleCondition</th>\n",
       "      <th>SalePrice</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>60</td>\n",
       "      <td>RL</td>\n",
       "      <td>65.0</td>\n",
       "      <td>8450</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Reg</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2008</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "      <td>208500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>20</td>\n",
       "      <td>RL</td>\n",
       "      <td>80.0</td>\n",
       "      <td>9600</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Reg</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>2007</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "      <td>181500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>60</td>\n",
       "      <td>RL</td>\n",
       "      <td>68.0</td>\n",
       "      <td>11250</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>IR1</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>2008</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "      <td>223500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>70</td>\n",
       "      <td>RL</td>\n",
       "      <td>60.0</td>\n",
       "      <td>9550</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>IR1</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2006</td>\n",
       "      <td>WD</td>\n",
       "      <td>Abnorml</td>\n",
       "      <td>140000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>60</td>\n",
       "      <td>RL</td>\n",
       "      <td>84.0</td>\n",
       "      <td>14260</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>IR1</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>12</td>\n",
       "      <td>2008</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "      <td>250000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 81 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Id  MSSubClass MSZoning  LotFrontage  LotArea Street Alley LotShape  \\\n",
       "0   1          60       RL         65.0     8450   Pave   NaN      Reg   \n",
       "1   2          20       RL         80.0     9600   Pave   NaN      Reg   \n",
       "2   3          60       RL         68.0    11250   Pave   NaN      IR1   \n",
       "3   4          70       RL         60.0     9550   Pave   NaN      IR1   \n",
       "4   5          60       RL         84.0    14260   Pave   NaN      IR1   \n",
       "\n",
       "  LandContour Utilities  ... PoolArea PoolQC Fence MiscFeature MiscVal MoSold  \\\n",
       "0         Lvl    AllPub  ...        0    NaN   NaN         NaN       0      2   \n",
       "1         Lvl    AllPub  ...        0    NaN   NaN         NaN       0      5   \n",
       "2         Lvl    AllPub  ...        0    NaN   NaN         NaN       0      9   \n",
       "3         Lvl    AllPub  ...        0    NaN   NaN         NaN       0      2   \n",
       "4         Lvl    AllPub  ...        0    NaN   NaN         NaN       0     12   \n",
       "\n",
       "  YrSold  SaleType  SaleCondition  SalePrice  \n",
       "0   2008        WD         Normal     208500  \n",
       "1   2007        WD         Normal     181500  \n",
       "2   2008        WD         Normal     223500  \n",
       "3   2006        WD        Abnorml     140000  \n",
       "4   2008        WD         Normal     250000  \n",
       "\n",
       "[5 rows x 81 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 1460 entries, 0 to 1459\n",
      "Data columns (total 81 columns):\n",
      " #   Column         Non-Null Count  Dtype  \n",
      "---  ------         --------------  -----  \n",
      " 0   Id             1460 non-null   int64  \n",
      " 1   MSSubClass     1460 non-null   int64  \n",
      " 2   MSZoning       1460 non-null   object \n",
      " 3   LotFrontage    1201 non-null   float64\n",
      " 4   LotArea        1460 non-null   int64  \n",
      " 5   Street         1460 non-null   object \n",
      " 6   Alley          91 non-null     object \n",
      " 7   LotShape       1460 non-null   object \n",
      " 8   LandContour    1460 non-null   object \n",
      " 9   Utilities      1460 non-null   object \n",
      " 10  LotConfig      1460 non-null   object \n",
      " 11  LandSlope      1460 non-null   object \n",
      " 12  Neighborhood   1460 non-null   object \n",
      " 13  Condition1     1460 non-null   object \n",
      " 14  Condition2     1460 non-null   object \n",
      " 15  BldgType       1460 non-null   object \n",
      " 16  HouseStyle     1460 non-null   object \n",
      " 17  OverallQual    1460 non-null   int64  \n",
      " 18  OverallCond    1460 non-null   int64  \n",
      " 19  YearBuilt      1460 non-null   int64  \n",
      " 20  YearRemodAdd   1460 non-null   int64  \n",
      " 21  RoofStyle      1460 non-null   object \n",
      " 22  RoofMatl       1460 non-null   object \n",
      " 23  Exterior1st    1460 non-null   object \n",
      " 24  Exterior2nd    1460 non-null   object \n",
      " 25  MasVnrType     1452 non-null   object \n",
      " 26  MasVnrArea     1452 non-null   float64\n",
      " 27  ExterQual      1460 non-null   object \n",
      " 28  ExterCond      1460 non-null   object \n",
      " 29  Foundation     1460 non-null   object \n",
      " 30  BsmtQual       1423 non-null   object \n",
      " 31  BsmtCond       1423 non-null   object \n",
      " 32  BsmtExposure   1422 non-null   object \n",
      " 33  BsmtFinType1   1423 non-null   object \n",
      " 34  BsmtFinSF1     1460 non-null   int64  \n",
      " 35  BsmtFinType2   1422 non-null   object \n",
      " 36  BsmtFinSF2     1460 non-null   int64  \n",
      " 37  BsmtUnfSF      1460 non-null   int64  \n",
      " 38  TotalBsmtSF    1460 non-null   int64  \n",
      " 39  Heating        1460 non-null   object \n",
      " 40  HeatingQC      1460 non-null   object \n",
      " 41  CentralAir     1460 non-null   object \n",
      " 42  Electrical     1459 non-null   object \n",
      " 43  1stFlrSF       1460 non-null   int64  \n",
      " 44  2ndFlrSF       1460 non-null   int64  \n",
      " 45  LowQualFinSF   1460 non-null   int64  \n",
      " 46  GrLivArea      1460 non-null   int64  \n",
      " 47  BsmtFullBath   1460 non-null   int64  \n",
      " 48  BsmtHalfBath   1460 non-null   int64  \n",
      " 49  FullBath       1460 non-null   int64  \n",
      " 50  HalfBath       1460 non-null   int64  \n",
      " 51  BedroomAbvGr   1460 non-null   int64  \n",
      " 52  KitchenAbvGr   1460 non-null   int64  \n",
      " 53  KitchenQual    1460 non-null   object \n",
      " 54  TotRmsAbvGrd   1460 non-null   int64  \n",
      " 55  Functional     1460 non-null   object \n",
      " 56  Fireplaces     1460 non-null   int64  \n",
      " 57  FireplaceQu    770 non-null    object \n",
      " 58  GarageType     1379 non-null   object \n",
      " 59  GarageYrBlt    1379 non-null   float64\n",
      " 60  GarageFinish   1379 non-null   object \n",
      " 61  GarageCars     1460 non-null   int64  \n",
      " 62  GarageArea     1460 non-null   int64  \n",
      " 63  GarageQual     1379 non-null   object \n",
      " 64  GarageCond     1379 non-null   object \n",
      " 65  PavedDrive     1460 non-null   object \n",
      " 66  WoodDeckSF     1460 non-null   int64  \n",
      " 67  OpenPorchSF    1460 non-null   int64  \n",
      " 68  EnclosedPorch  1460 non-null   int64  \n",
      " 69  3SsnPorch      1460 non-null   int64  \n",
      " 70  ScreenPorch    1460 non-null   int64  \n",
      " 71  PoolArea       1460 non-null   int64  \n",
      " 72  PoolQC         7 non-null      object \n",
      " 73  Fence          281 non-null    object \n",
      " 74  MiscFeature    54 non-null     object \n",
      " 75  MiscVal        1460 non-null   int64  \n",
      " 76  MoSold         1460 non-null   int64  \n",
      " 77  YrSold         1460 non-null   int64  \n",
      " 78  SaleType       1460 non-null   object \n",
      " 79  SaleCondition  1460 non-null   object \n",
      " 80  SalePrice      1460 non-null   int64  \n",
      "dtypes: float64(3), int64(35), object(43)\n",
      "memory usage: 924.0+ KB\n"
     ]
    }
   ],
   "source": [
    "housing_df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Id                 0\n",
       "MSSubClass         0\n",
       "MSZoning           0\n",
       "LotFrontage      259\n",
       "LotArea            0\n",
       "                ... \n",
       "MoSold             0\n",
       "YrSold             0\n",
       "SaleType           0\n",
       "SaleCondition      0\n",
       "SalePrice          0\n",
       "Length: 81, dtype: int64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing_df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "numeric_lst=['int16', 'int32', 'int64', 'float16', 'float32', 'float64']\n",
    "numerical_cols = list(housing_df.select_dtypes(include=numeric_lst).columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Id',\n",
       " 'MSSubClass',\n",
       " 'LotFrontage',\n",
       " 'LotArea',\n",
       " 'OverallQual',\n",
       " 'OverallCond',\n",
       " 'YearBuilt',\n",
       " 'YearRemodAdd',\n",
       " 'MasVnrArea',\n",
       " 'BsmtFinSF1',\n",
       " 'BsmtFinSF2',\n",
       " 'BsmtUnfSF',\n",
       " 'TotalBsmtSF',\n",
       " '1stFlrSF',\n",
       " '2ndFlrSF',\n",
       " 'LowQualFinSF',\n",
       " 'GrLivArea',\n",
       " 'BsmtFullBath',\n",
       " 'BsmtHalfBath',\n",
       " 'FullBath',\n",
       " 'HalfBath',\n",
       " 'BedroomAbvGr',\n",
       " 'KitchenAbvGr',\n",
       " 'TotRmsAbvGrd',\n",
       " 'Fireplaces',\n",
       " 'GarageYrBlt',\n",
       " 'GarageCars',\n",
       " 'GarageArea',\n",
       " 'WoodDeckSF',\n",
       " 'OpenPorchSF',\n",
       " 'EnclosedPorch',\n",
       " '3SsnPorch',\n",
       " 'ScreenPorch',\n",
       " 'PoolArea',\n",
       " 'MiscVal',\n",
       " 'MoSold',\n",
       " 'YrSold',\n",
       " 'SalePrice']"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "numerical_cols"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "housing_df=housing_df[numerical_cols]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>Id</th>\n",
       "      <th>MSSubClass</th>\n",
       "      <th>LotFrontage</th>\n",
       "      <th>LotArea</th>\n",
       "      <th>OverallQual</th>\n",
       "      <th>OverallCond</th>\n",
       "      <th>YearBuilt</th>\n",
       "      <th>YearRemodAdd</th>\n",
       "      <th>MasVnrArea</th>\n",
       "      <th>BsmtFinSF1</th>\n",
       "      <th>...</th>\n",
       "      <th>WoodDeckSF</th>\n",
       "      <th>OpenPorchSF</th>\n",
       "      <th>EnclosedPorch</th>\n",
       "      <th>3SsnPorch</th>\n",
       "      <th>ScreenPorch</th>\n",
       "      <th>PoolArea</th>\n",
       "      <th>MiscVal</th>\n",
       "      <th>MoSold</th>\n",
       "      <th>YrSold</th>\n",
       "      <th>SalePrice</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>60</td>\n",
       "      <td>65.0</td>\n",
       "      <td>8450</td>\n",
       "      <td>7</td>\n",
       "      <td>5</td>\n",
       "      <td>2003</td>\n",
       "      <td>2003</td>\n",
       "      <td>196.0</td>\n",
       "      <td>706</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>61</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>2008</td>\n",
       "      <td>208500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>20</td>\n",
       "      <td>80.0</td>\n",
       "      <td>9600</td>\n",
       "      <td>6</td>\n",
       "      <td>8</td>\n",
       "      <td>1976</td>\n",
       "      <td>1976</td>\n",
       "      <td>0.0</td>\n",
       "      <td>978</td>\n",
       "      <td>...</td>\n",
       "      <td>298</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>5</td>\n",
       "      <td>2007</td>\n",
       "      <td>181500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>60</td>\n",
       "      <td>68.0</td>\n",
       "      <td>11250</td>\n",
       "      <td>7</td>\n",
       "      <td>5</td>\n",
       "      <td>2001</td>\n",
       "      <td>2002</td>\n",
       "      <td>162.0</td>\n",
       "      <td>486</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>42</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>2008</td>\n",
       "      <td>223500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>70</td>\n",
       "      <td>60.0</td>\n",
       "      <td>9550</td>\n",
       "      <td>7</td>\n",
       "      <td>5</td>\n",
       "      <td>1915</td>\n",
       "      <td>1970</td>\n",
       "      <td>0.0</td>\n",
       "      <td>216</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>35</td>\n",
       "      <td>272</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2006</td>\n",
       "      <td>140000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>60</td>\n",
       "      <td>84.0</td>\n",
       "      <td>14260</td>\n",
       "      <td>8</td>\n",
       "      <td>5</td>\n",
       "      <td>2000</td>\n",
       "      <td>2000</td>\n",
       "      <td>350.0</td>\n",
       "      <td>655</td>\n",
       "      <td>...</td>\n",
       "      <td>192</td>\n",
       "      <td>84</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>12</td>\n",
       "      <td>2008</td>\n",
       "      <td>250000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 38 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Id  MSSubClass  LotFrontage  LotArea  OverallQual  OverallCond  YearBuilt  \\\n",
       "0   1          60         65.0     8450            7            5       2003   \n",
       "1   2          20         80.0     9600            6            8       1976   \n",
       "2   3          60         68.0    11250            7            5       2001   \n",
       "3   4          70         60.0     9550            7            5       1915   \n",
       "4   5          60         84.0    14260            8            5       2000   \n",
       "\n",
       "   YearRemodAdd  MasVnrArea  BsmtFinSF1  ...  WoodDeckSF  OpenPorchSF  \\\n",
       "0          2003       196.0         706  ...           0           61   \n",
       "1          1976         0.0         978  ...         298            0   \n",
       "2          2002       162.0         486  ...           0           42   \n",
       "3          1970         0.0         216  ...           0           35   \n",
       "4          2000       350.0         655  ...         192           84   \n",
       "\n",
       "   EnclosedPorch  3SsnPorch  ScreenPorch  PoolArea  MiscVal  MoSold  YrSold  \\\n",
       "0              0          0            0         0        0       2    2008   \n",
       "1              0          0            0         0        0       5    2007   \n",
       "2              0          0            0         0        0       9    2008   \n",
       "3            272          0            0         0        0       2    2006   \n",
       "4              0          0            0         0        0      12    2008   \n",
       "\n",
       "   SalePrice  \n",
       "0     208500  \n",
       "1     181500  \n",
       "2     223500  \n",
       "3     140000  \n",
       "4     250000  \n",
       "\n",
       "[5 rows x 38 columns]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "housing_df=housing_df.drop(\"Id\",axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "### It is always a good practice to split train and test data to avoid\n",
    "#overfitting\n",
    "from sklearn.model_selection import train_test_split\n",
    "X_train,X_test,y_train,y_test=train_test_split(housing_df.drop(labels=['SalePrice'], axis=1),\n",
    "    housing_df['SalePrice'],\n",
    "    test_size=0.3,\n",
    "    random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>MSSubClass</th>\n",
       "      <th>LotFrontage</th>\n",
       "      <th>LotArea</th>\n",
       "      <th>OverallQual</th>\n",
       "      <th>OverallCond</th>\n",
       "      <th>YearBuilt</th>\n",
       "      <th>YearRemodAdd</th>\n",
       "      <th>MasVnrArea</th>\n",
       "      <th>BsmtFinSF1</th>\n",
       "      <th>BsmtFinSF2</th>\n",
       "      <th>...</th>\n",
       "      <th>GarageArea</th>\n",
       "      <th>WoodDeckSF</th>\n",
       "      <th>OpenPorchSF</th>\n",
       "      <th>EnclosedPorch</th>\n",
       "      <th>3SsnPorch</th>\n",
       "      <th>ScreenPorch</th>\n",
       "      <th>PoolArea</th>\n",
       "      <th>MiscVal</th>\n",
       "      <th>MoSold</th>\n",
       "      <th>YrSold</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>64</th>\n",
       "      <td>60</td>\n",
       "      <td>NaN</td>\n",
       "      <td>9375</td>\n",
       "      <td>7</td>\n",
       "      <td>5</td>\n",
       "      <td>1997</td>\n",
       "      <td>1998</td>\n",
       "      <td>573.0</td>\n",
       "      <td>739</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>645</td>\n",
       "      <td>576</td>\n",
       "      <td>36</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>2009</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>682</th>\n",
       "      <td>120</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2887</td>\n",
       "      <td>6</td>\n",
       "      <td>5</td>\n",
       "      <td>1996</td>\n",
       "      <td>1997</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1003</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>431</td>\n",
       "      <td>307</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>11</td>\n",
       "      <td>2008</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>960</th>\n",
       "      <td>20</td>\n",
       "      <td>50.0</td>\n",
       "      <td>7207</td>\n",
       "      <td>5</td>\n",
       "      <td>7</td>\n",
       "      <td>1958</td>\n",
       "      <td>2008</td>\n",
       "      <td>0.0</td>\n",
       "      <td>696</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>117</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>2010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1384</th>\n",
       "      <td>50</td>\n",
       "      <td>60.0</td>\n",
       "      <td>9060</td>\n",
       "      <td>6</td>\n",
       "      <td>5</td>\n",
       "      <td>1939</td>\n",
       "      <td>1950</td>\n",
       "      <td>0.0</td>\n",
       "      <td>204</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>280</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>10</td>\n",
       "      <td>2009</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1100</th>\n",
       "      <td>30</td>\n",
       "      <td>60.0</td>\n",
       "      <td>8400</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>1920</td>\n",
       "      <td>1950</td>\n",
       "      <td>0.0</td>\n",
       "      <td>290</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>246</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>2009</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>763</th>\n",
       "      <td>60</td>\n",
       "      <td>82.0</td>\n",
       "      <td>9430</td>\n",
       "      <td>8</td>\n",
       "      <td>5</td>\n",
       "      <td>1999</td>\n",
       "      <td>1999</td>\n",
       "      <td>673.0</td>\n",
       "      <td>1163</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>856</td>\n",
       "      <td>0</td>\n",
       "      <td>128</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>180</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>2009</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>835</th>\n",
       "      <td>20</td>\n",
       "      <td>60.0</td>\n",
       "      <td>9600</td>\n",
       "      <td>4</td>\n",
       "      <td>7</td>\n",
       "      <td>1950</td>\n",
       "      <td>1995</td>\n",
       "      <td>0.0</td>\n",
       "      <td>442</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>436</td>\n",
       "      <td>290</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>2010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1216</th>\n",
       "      <td>90</td>\n",
       "      <td>68.0</td>\n",
       "      <td>8930</td>\n",
       "      <td>6</td>\n",
       "      <td>5</td>\n",
       "      <td>1978</td>\n",
       "      <td>1978</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>539</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>4</td>\n",
       "      <td>2010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>559</th>\n",
       "      <td>120</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3196</td>\n",
       "      <td>7</td>\n",
       "      <td>5</td>\n",
       "      <td>2003</td>\n",
       "      <td>2004</td>\n",
       "      <td>18.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>420</td>\n",
       "      <td>143</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "      <td>2006</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>684</th>\n",
       "      <td>60</td>\n",
       "      <td>58.0</td>\n",
       "      <td>16770</td>\n",
       "      <td>7</td>\n",
       "      <td>5</td>\n",
       "      <td>1998</td>\n",
       "      <td>1998</td>\n",
       "      <td>30.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>486</td>\n",
       "      <td>0</td>\n",
       "      <td>81</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>2010</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1022 rows × 36 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      MSSubClass  LotFrontage  LotArea  OverallQual  OverallCond  YearBuilt  \\\n",
       "64            60          NaN     9375            7            5       1997   \n",
       "682          120          NaN     2887            6            5       1996   \n",
       "960           20         50.0     7207            5            7       1958   \n",
       "1384          50         60.0     9060            6            5       1939   \n",
       "1100          30         60.0     8400            2            5       1920   \n",
       "...          ...          ...      ...          ...          ...        ...   \n",
       "763           60         82.0     9430            8            5       1999   \n",
       "835           20         60.0     9600            4            7       1950   \n",
       "1216          90         68.0     8930            6            5       1978   \n",
       "559          120          NaN     3196            7            5       2003   \n",
       "684           60         58.0    16770            7            5       1998   \n",
       "\n",
       "      YearRemodAdd  MasVnrArea  BsmtFinSF1  BsmtFinSF2  ...  GarageArea  \\\n",
       "64            1998       573.0         739           0  ...         645   \n",
       "682           1997         0.0        1003           0  ...         431   \n",
       "960           2008         0.0         696           0  ...           0   \n",
       "1384          1950         0.0         204           0  ...         280   \n",
       "1100          1950         0.0         290           0  ...         246   \n",
       "...            ...         ...         ...         ...  ...         ...   \n",
       "763           1999       673.0        1163           0  ...         856   \n",
       "835           1995         0.0         442           0  ...         436   \n",
       "1216          1978         0.0           0           0  ...         539   \n",
       "559           2004        18.0           0           0  ...         420   \n",
       "684           1998        30.0           0           0  ...         486   \n",
       "\n",
       "      WoodDeckSF  OpenPorchSF  EnclosedPorch  3SsnPorch  ScreenPorch  \\\n",
       "64           576           36              0          0            0   \n",
       "682          307            0              0          0            0   \n",
       "960          117            0              0          0            0   \n",
       "1384           0            0              0          0            0   \n",
       "1100           0            0              0          0            0   \n",
       "...          ...          ...            ...        ...          ...   \n",
       "763            0          128              0          0          180   \n",
       "835          290            0              0          0            0   \n",
       "1216           0            0              0          0            0   \n",
       "559          143           20              0          0            0   \n",
       "684            0           81              0          0            0   \n",
       "\n",
       "      PoolArea  MiscVal  MoSold  YrSold  \n",
       "64           0        0       2    2009  \n",
       "682          0        0      11    2008  \n",
       "960          0        0       2    2010  \n",
       "1384         0        0      10    2009  \n",
       "1100         0        0       1    2009  \n",
       "...        ...      ...     ...     ...  \n",
       "763          0        0       7    2009  \n",
       "835          0        0       2    2010  \n",
       "1216         0        0       4    2010  \n",
       "559          0        0      10    2006  \n",
       "684          0        0       6    2010  \n",
       "\n",
       "[1022 rows x 36 columns]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "MSSubClass         0\n",
       "LotFrontage      189\n",
       "LotArea            0\n",
       "OverallQual        0\n",
       "OverallCond        0\n",
       "YearBuilt          0\n",
       "YearRemodAdd       0\n",
       "MasVnrArea         5\n",
       "BsmtFinSF1         0\n",
       "BsmtFinSF2         0\n",
       "BsmtUnfSF          0\n",
       "TotalBsmtSF        0\n",
       "1stFlrSF           0\n",
       "2ndFlrSF           0\n",
       "LowQualFinSF       0\n",
       "GrLivArea          0\n",
       "BsmtFullBath       0\n",
       "BsmtHalfBath       0\n",
       "FullBath           0\n",
       "HalfBath           0\n",
       "BedroomAbvGr       0\n",
       "KitchenAbvGr       0\n",
       "TotRmsAbvGrd       0\n",
       "Fireplaces         0\n",
       "GarageYrBlt       54\n",
       "GarageCars         0\n",
       "GarageArea         0\n",
       "WoodDeckSF         0\n",
       "OpenPorchSF        0\n",
       "EnclosedPorch      0\n",
       "3SsnPorch          0\n",
       "ScreenPorch        0\n",
       "PoolArea           0\n",
       "MiscVal            0\n",
       "MoSold             0\n",
       "YrSold             0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.25011168, 0.21292648, 0.16620992, 0.52201246, 0.10677582,\n",
       "       0.34019504, 0.23982053, 0.0929883 , 0.16041656, 0.0087294 ,\n",
       "       0.10935919, 0.38436417, 0.31182785, 0.23201755, 0.00355761,\n",
       "       0.48241304, 0.02894393, 0.00354248, 0.28790693, 0.08102465,\n",
       "       0.04768536, 0.02125749, 0.18564212, 0.18172804, 0.26496226,\n",
       "       0.35773488, 0.36916149, 0.07218483, 0.17044817, 0.00822029,\n",
       "       0.00196552, 0.01673325, 0.        , 0.00152766, 0.        ,\n",
       "       0.00811775])"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.feature_selection import mutual_info_regression\n",
    "# determine the mutual information\n",
    "mutual_info = mutual_info_regression(X_train.fillna(0), y_train)\n",
    "mutual_info"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "OverallQual      0.522012\n",
       "GrLivArea        0.482413\n",
       "TotalBsmtSF      0.384364\n",
       "GarageArea       0.369161\n",
       "GarageCars       0.357735\n",
       "YearBuilt        0.340195\n",
       "1stFlrSF         0.311828\n",
       "FullBath         0.287907\n",
       "GarageYrBlt      0.264962\n",
       "MSSubClass       0.250112\n",
       "YearRemodAdd     0.239821\n",
       "2ndFlrSF         0.232018\n",
       "LotFrontage      0.212926\n",
       "TotRmsAbvGrd     0.185642\n",
       "Fireplaces       0.181728\n",
       "OpenPorchSF      0.170448\n",
       "LotArea          0.166210\n",
       "BsmtFinSF1       0.160417\n",
       "BsmtUnfSF        0.109359\n",
       "OverallCond      0.106776\n",
       "MasVnrArea       0.092988\n",
       "HalfBath         0.081025\n",
       "WoodDeckSF       0.072185\n",
       "BedroomAbvGr     0.047685\n",
       "BsmtFullBath     0.028944\n",
       "KitchenAbvGr     0.021257\n",
       "ScreenPorch      0.016733\n",
       "BsmtFinSF2       0.008729\n",
       "EnclosedPorch    0.008220\n",
       "YrSold           0.008118\n",
       "LowQualFinSF     0.003558\n",
       "BsmtHalfBath     0.003542\n",
       "3SsnPorch        0.001966\n",
       "MiscVal          0.001528\n",
       "MoSold           0.000000\n",
       "PoolArea         0.000000\n",
       "dtype: float64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mutual_info = pd.Series(mutual_info)\n",
    "mutual_info.index = X_train.columns\n",
    "mutual_info.sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "mutual_info.sort_values(ascending=False).plot.bar(figsize=(15,5))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_selection import SelectPercentile"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SelectPercentile(percentile=20,\n",
       "                 score_func=<function mutual_info_regression at 0x000001796953DE58>)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Selecting the top 20 percentile\n",
    "selected_top_columns = SelectPercentile(mutual_info_regression, percentile=20)\n",
    "selected_top_columns.fit(X_train.fillna(0), y_train)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([False, False, False,  True, False,  True, False, False, False,\n",
       "       False, False,  True,  True, False, False,  True, False, False,\n",
       "       False, False, False, False, False, False, False,  True,  True,\n",
       "       False, False, False, False, False, False, False, False, False])"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "selected_top_columns.get_support()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['OverallQual', 'YearBuilt', 'TotalBsmtSF', '1stFlrSF', 'GrLivArea',\n",
       "       'GarageCars', 'GarageArea'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.columns[selected_top_columns.get_support()]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Difference Between Information Gain And Mutual Information\n",
    "I(X ; Y) = H(X) – H(X | Y)\n",
    "and\n",
    "IG(S, a) = H(S) – H(S | a)\n",
    "\n",
    "\n",
    "As such, mutual information is sometimes used as a synonym for \n",
    "information gain. Technically, they calculate the same quantity\n",
    "if applied to the same data."
   ]
  },
  {
   "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": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}