Stratified sampling + modelling.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "# importing the required dependencies\n",
    "\n",
    "import pandas as pd # python's library for data manipulation and preprocessing\n",
    "import numpy as np  # python's library for number crunching\n",
    "\n",
    "import matplotlib            # python's library for visualisation\n",
    "import matplotlib.pyplot as plt\n",
    "plt.rcParams['axes.labelsize'] = 14\n",
    "plt.rcParams['xtick.labelsize'] = 12\n",
    "plt.rcParams['ytick.labelsize'] = 12\n",
    "\n",
    "\n",
    "import seaborn as sns        # also python's library for visualisations\n",
    "color = sns.color_palette()\n",
    "sns.set_style('darkgrid')\n",
    "\n",
    "\n",
    "import sklearn                #python's machine learning library\n",
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "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>longitude</th>\n",
       "      <th>latitude</th>\n",
       "      <th>housing_median_age</th>\n",
       "      <th>total_rooms</th>\n",
       "      <th>total_bedrooms</th>\n",
       "      <th>population</th>\n",
       "      <th>households</th>\n",
       "      <th>median_income</th>\n",
       "      <th>median_house_value</th>\n",
       "      <th>ocean_proximity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-122.23</td>\n",
       "      <td>37.88</td>\n",
       "      <td>41.0</td>\n",
       "      <td>880.0</td>\n",
       "      <td>129.0</td>\n",
       "      <td>322.0</td>\n",
       "      <td>126.0</td>\n",
       "      <td>8.3252</td>\n",
       "      <td>452600.0</td>\n",
       "      <td>NEAR BAY</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-122.22</td>\n",
       "      <td>37.86</td>\n",
       "      <td>21.0</td>\n",
       "      <td>7099.0</td>\n",
       "      <td>1106.0</td>\n",
       "      <td>2401.0</td>\n",
       "      <td>1138.0</td>\n",
       "      <td>8.3014</td>\n",
       "      <td>358500.0</td>\n",
       "      <td>NEAR BAY</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-122.24</td>\n",
       "      <td>37.85</td>\n",
       "      <td>52.0</td>\n",
       "      <td>1467.0</td>\n",
       "      <td>190.0</td>\n",
       "      <td>496.0</td>\n",
       "      <td>177.0</td>\n",
       "      <td>7.2574</td>\n",
       "      <td>352100.0</td>\n",
       "      <td>NEAR BAY</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-122.25</td>\n",
       "      <td>37.85</td>\n",
       "      <td>52.0</td>\n",
       "      <td>1274.0</td>\n",
       "      <td>235.0</td>\n",
       "      <td>558.0</td>\n",
       "      <td>219.0</td>\n",
       "      <td>5.6431</td>\n",
       "      <td>341300.0</td>\n",
       "      <td>NEAR BAY</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-122.25</td>\n",
       "      <td>37.85</td>\n",
       "      <td>52.0</td>\n",
       "      <td>1627.0</td>\n",
       "      <td>280.0</td>\n",
       "      <td>565.0</td>\n",
       "      <td>259.0</td>\n",
       "      <td>3.8462</td>\n",
       "      <td>342200.0</td>\n",
       "      <td>NEAR BAY</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   longitude  latitude  housing_median_age  total_rooms  total_bedrooms  \\\n",
       "0    -122.23     37.88                41.0        880.0           129.0   \n",
       "1    -122.22     37.86                21.0       7099.0          1106.0   \n",
       "2    -122.24     37.85                52.0       1467.0           190.0   \n",
       "3    -122.25     37.85                52.0       1274.0           235.0   \n",
       "4    -122.25     37.85                52.0       1627.0           280.0   \n",
       "\n",
       "   population  households  median_income  median_house_value ocean_proximity  \n",
       "0       322.0       126.0         8.3252            452600.0        NEAR BAY  \n",
       "1      2401.0      1138.0         8.3014            358500.0        NEAR BAY  \n",
       "2       496.0       177.0         7.2574            352100.0        NEAR BAY  \n",
       "3       558.0       219.0         5.6431            341300.0        NEAR BAY  \n",
       "4       565.0       259.0         3.8462            342200.0        NEAR BAY  "
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing = pd.read_csv('housing.csv')   # reading the data into a pandas dataframe\n",
    "housing.head()                         # calling the first five rows of the dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 20640 entries, 0 to 20639\n",
      "Data columns (total 10 columns):\n",
      "longitude             20640 non-null float64\n",
      "latitude              20640 non-null float64\n",
      "housing_median_age    20640 non-null float64\n",
      "total_rooms           20640 non-null float64\n",
      "total_bedrooms        20433 non-null float64\n",
      "population            20640 non-null float64\n",
      "households            20640 non-null float64\n",
      "median_income         20640 non-null float64\n",
      "median_house_value    20640 non-null float64\n",
      "ocean_proximity       20640 non-null object\n",
      "dtypes: float64(9), object(1)\n",
      "memory usage: 1.6+ MB\n"
     ]
    }
   ],
   "source": [
    "housing.info()  # exploratory analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x275fe13320>"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2768be88d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "correlation_matrix = housing.corr()   #creating a heatmap of the attributes in the dataset\n",
    "plt.subplots(figsize=(8,6))\n",
    "sns.heatmap(correlation_matrix, center=0, annot=True, linewidths=.3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "median_house_value    1.000000\n",
       "median_income         0.688075\n",
       "total_rooms           0.134153\n",
       "housing_median_age    0.105623\n",
       "households            0.065843\n",
       "total_bedrooms        0.049686\n",
       "population           -0.024650\n",
       "longitude            -0.045967\n",
       "latitude             -0.144160\n",
       "Name: median_house_value, dtype: float64"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corr = housing.corr()      # showing correlations by target variable\n",
    "corr['median_house_value'].sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2768de9e48>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(housing.median_income) # showing the distribution of the median_income variable in the dataset\n",
    "plt.show()                                  # check the distplot params though. see how you can make it prettier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0      822\n",
       "2.0     6581\n",
       "3.0     7236\n",
       "4.0     3639\n",
       "5.0     1423\n",
       "6.0      532\n",
       "7.0      189\n",
       "8.0      105\n",
       "9.0       50\n",
       "10.0      14\n",
       "11.0      49\n",
       "Name: median_income_category, dtype: int64"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Divide by 1.5 to limit the number of income categories\n",
    "housing[\"median_income_category\"] = np.ceil(housing[\"median_income\"] / 1.5)\n",
    "\n",
    "# showing the frequency of each category\n",
    "housing.median_income_category.value_counts().sort_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Label those above 5 as 5\n",
    "housing[\"median_income_category\"].where(housing[\"median_income_category\"] < 5, 5.0, inplace=True)\n",
    "\n",
    "\n",
    "#housing.median_income_category.value_counts().sort_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>longitude</th>\n",
       "      <th>latitude</th>\n",
       "      <th>housing_median_age</th>\n",
       "      <th>total_rooms</th>\n",
       "      <th>total_bedrooms</th>\n",
       "      <th>population</th>\n",
       "      <th>households</th>\n",
       "      <th>median_income</th>\n",
       "      <th>median_house_value</th>\n",
       "      <th>ocean_proximity</th>\n",
       "      <th>median_income_category</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>5241</th>\n",
       "      <td>-118.39</td>\n",
       "      <td>34.12</td>\n",
       "      <td>29.0</td>\n",
       "      <td>6447.0</td>\n",
       "      <td>1012.0</td>\n",
       "      <td>2184.0</td>\n",
       "      <td>960.0</td>\n",
       "      <td>8.2816</td>\n",
       "      <td>500001.0</td>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10970</th>\n",
       "      <td>-117.86</td>\n",
       "      <td>33.77</td>\n",
       "      <td>39.0</td>\n",
       "      <td>4159.0</td>\n",
       "      <td>655.0</td>\n",
       "      <td>1669.0</td>\n",
       "      <td>651.0</td>\n",
       "      <td>4.6111</td>\n",
       "      <td>240300.0</td>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20351</th>\n",
       "      <td>-119.05</td>\n",
       "      <td>34.21</td>\n",
       "      <td>27.0</td>\n",
       "      <td>4357.0</td>\n",
       "      <td>926.0</td>\n",
       "      <td>2110.0</td>\n",
       "      <td>876.0</td>\n",
       "      <td>3.0119</td>\n",
       "      <td>218200.0</td>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6568</th>\n",
       "      <td>-118.15</td>\n",
       "      <td>34.20</td>\n",
       "      <td>52.0</td>\n",
       "      <td>1786.0</td>\n",
       "      <td>306.0</td>\n",
       "      <td>1018.0</td>\n",
       "      <td>322.0</td>\n",
       "      <td>4.1518</td>\n",
       "      <td>182100.0</td>\n",
       "      <td>INLAND</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13285</th>\n",
       "      <td>-117.68</td>\n",
       "      <td>34.07</td>\n",
       "      <td>32.0</td>\n",
       "      <td>1775.0</td>\n",
       "      <td>314.0</td>\n",
       "      <td>1067.0</td>\n",
       "      <td>302.0</td>\n",
       "      <td>4.0375</td>\n",
       "      <td>121300.0</td>\n",
       "      <td>INLAND</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       longitude  latitude  housing_median_age  total_rooms  total_bedrooms  \\\n",
       "5241     -118.39     34.12                29.0       6447.0          1012.0   \n",
       "10970    -117.86     33.77                39.0       4159.0           655.0   \n",
       "20351    -119.05     34.21                27.0       4357.0           926.0   \n",
       "6568     -118.15     34.20                52.0       1786.0           306.0   \n",
       "13285    -117.68     34.07                32.0       1775.0           314.0   \n",
       "\n",
       "       population  households  median_income  median_house_value  \\\n",
       "5241       2184.0       960.0         8.2816            500001.0   \n",
       "10970      1669.0       651.0         4.6111            240300.0   \n",
       "20351      2110.0       876.0         3.0119            218200.0   \n",
       "6568       1018.0       322.0         4.1518            182100.0   \n",
       "13285      1067.0       302.0         4.0375            121300.0   \n",
       "\n",
       "      ocean_proximity  median_income_category  \n",
       "5241        <1H OCEAN                     5.0  \n",
       "10970       <1H OCEAN                     4.0  \n",
       "20351       <1H OCEAN                     3.0  \n",
       "6568           INLAND                     3.0  \n",
       "13285          INLAND                     3.0  "
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# importing the split class\n",
    "from sklearn.model_selection import StratifiedShuffleSplit\n",
    "\n",
    "# instantiating the class and aplitting \n",
    "split = StratifiedShuffleSplit(n_splits=1, test_size=0.2, random_state=42)\n",
    "\n",
    "# smiple loop to stratify by category and store the values in strat_train and test sets\n",
    "for train_index, test_index in split.split(housing, housing[\"median_income_category\"]):\n",
    "    strat_train_set = housing.loc[train_index]\n",
    "    strat_test_set = housing.loc[test_index]\n",
    "    \n",
    "# view the first five rows of the dataframe\n",
    "strat_test_set.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "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>Overall</th>\n",
       "      <th>Random</th>\n",
       "      <th>Stratified</th>\n",
       "      <th>Rand. %error</th>\n",
       "      <th>Strat. %error</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1.0</th>\n",
       "      <td>0.039826</td>\n",
       "      <td>0.040213</td>\n",
       "      <td>0.039729</td>\n",
       "      <td>0.973236</td>\n",
       "      <td>-0.243309</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2.0</th>\n",
       "      <td>0.318847</td>\n",
       "      <td>0.324370</td>\n",
       "      <td>0.318798</td>\n",
       "      <td>1.732260</td>\n",
       "      <td>-0.015195</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3.0</th>\n",
       "      <td>0.350581</td>\n",
       "      <td>0.358527</td>\n",
       "      <td>0.350533</td>\n",
       "      <td>2.266446</td>\n",
       "      <td>-0.013820</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4.0</th>\n",
       "      <td>0.176308</td>\n",
       "      <td>0.167393</td>\n",
       "      <td>0.176357</td>\n",
       "      <td>-5.056334</td>\n",
       "      <td>0.027480</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5.0</th>\n",
       "      <td>0.114438</td>\n",
       "      <td>0.109496</td>\n",
       "      <td>0.114583</td>\n",
       "      <td>-4.318374</td>\n",
       "      <td>0.127011</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Overall    Random  Stratified  Rand. %error  Strat. %error\n",
       "1.0  0.039826  0.040213    0.039729      0.973236      -0.243309\n",
       "2.0  0.318847  0.324370    0.318798      1.732260      -0.015195\n",
       "3.0  0.350581  0.358527    0.350533      2.266446      -0.013820\n",
       "4.0  0.176308  0.167393    0.176357     -5.056334       0.027480\n",
       "5.0  0.114438  0.109496    0.114583     -4.318374       0.127011"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# custom function to create a comparable dataframe\n",
    "\n",
    "def income_cat_props(data):\n",
    "    return data['median_income_category'].value_counts() / len(data)\n",
    "\n",
    "\n",
    "train_set, test_set = train_test_split(housing, test_size=0.2, random_state=42)\n",
    "\n",
    "compare_props = pd.DataFrame({\n",
    "    \"Overall\": income_cat_props(housing),\n",
    "    \"Stratified\": income_cat_props(strat_test_set),\n",
    "    \"Random\": income_cat_props(test_set),\n",
    "}).sort_index()\n",
    "\n",
    "\n",
    "compare_props[\"Rand. %error\"] = 100 * compare_props[\"Random\"] / compare_props[\"Overall\"] - 100\n",
    "compare_props[\"Strat. %error\"] = 100 * compare_props[\"Stratified\"] / compare_props[\"Overall\"] - 100\n",
    "\n",
    "compare_props"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Running some modelling to compare the stratified and random sampling metrics"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "# Starting with the stratified sampling metrics. Random sampling would be in another notebook for simplicity and ease of understanding"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "for set_ in (strat_train_set, strat_test_set):    # dropping the unwanted medium_income_category for bith strat train and test sets\n",
    "    set_.drop(\"median_income_category\", axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "17606    286600.0\n",
       "18632    340600.0\n",
       "14650    196900.0\n",
       "3230      46300.0\n",
       "3555     254500.0\n",
       "Name: median_house_value, dtype: float64"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing = strat_train_set.drop('median_house_value', axis=1)    # creating the train set\n",
    "housing_labels =  strat_train_set['median_house_value'].copy()  # creating the test set\n",
    "housing_labels.head()                                           # viewing the first five values in the test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "longitude               0\n",
       "latitude                0\n",
       "housing_median_age      0\n",
       "total_rooms             0\n",
       "total_bedrooms        158\n",
       "population              0\n",
       "households              0\n",
       "median_income           0\n",
       "ocean_proximity         0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rows_missing_data = housing.isnull().any()  # checking the dataframe for missing values\n",
    "housing.isnull().sum()                       # 158 misssing rows in the training set. the others have been stratified to the test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "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>longitude</th>\n",
       "      <th>latitude</th>\n",
       "      <th>housing_median_age</th>\n",
       "      <th>total_rooms</th>\n",
       "      <th>total_bedrooms</th>\n",
       "      <th>population</th>\n",
       "      <th>households</th>\n",
       "      <th>median_income</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>17606</th>\n",
       "      <td>-121.89</td>\n",
       "      <td>37.29</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1568.0</td>\n",
       "      <td>351.0</td>\n",
       "      <td>710.0</td>\n",
       "      <td>339.0</td>\n",
       "      <td>2.7042</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18632</th>\n",
       "      <td>-121.93</td>\n",
       "      <td>37.05</td>\n",
       "      <td>14.0</td>\n",
       "      <td>679.0</td>\n",
       "      <td>108.0</td>\n",
       "      <td>306.0</td>\n",
       "      <td>113.0</td>\n",
       "      <td>6.4214</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14650</th>\n",
       "      <td>-117.20</td>\n",
       "      <td>32.77</td>\n",
       "      <td>31.0</td>\n",
       "      <td>1952.0</td>\n",
       "      <td>471.0</td>\n",
       "      <td>936.0</td>\n",
       "      <td>462.0</td>\n",
       "      <td>2.8621</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3230</th>\n",
       "      <td>-119.61</td>\n",
       "      <td>36.31</td>\n",
       "      <td>25.0</td>\n",
       "      <td>1847.0</td>\n",
       "      <td>371.0</td>\n",
       "      <td>1460.0</td>\n",
       "      <td>353.0</td>\n",
       "      <td>1.8839</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3555</th>\n",
       "      <td>-118.59</td>\n",
       "      <td>34.23</td>\n",
       "      <td>17.0</td>\n",
       "      <td>6592.0</td>\n",
       "      <td>1525.0</td>\n",
       "      <td>4459.0</td>\n",
       "      <td>1463.0</td>\n",
       "      <td>3.0347</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       longitude  latitude  housing_median_age  total_rooms  total_bedrooms  \\\n",
       "17606    -121.89     37.29                38.0       1568.0           351.0   \n",
       "18632    -121.93     37.05                14.0        679.0           108.0   \n",
       "14650    -117.20     32.77                31.0       1952.0           471.0   \n",
       "3230     -119.61     36.31                25.0       1847.0           371.0   \n",
       "3555     -118.59     34.23                17.0       6592.0          1525.0   \n",
       "\n",
       "       population  households  median_income  \n",
       "17606       710.0       339.0         2.7042  \n",
       "18632       306.0       113.0         6.4214  \n",
       "14650       936.0       462.0         2.8621  \n",
       "3230       1460.0       353.0         1.8839  \n",
       "3555       4459.0      1463.0         3.0347  "
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# creating an sklearn IMPUTER class for replacing missing values.\n",
    "\n",
    "from sklearn.preprocessing import Imputer\n",
    "# instantiate the class\n",
    "imputer = Imputer(strategy='median')\n",
    "\n",
    "#since the housing dataset has a categorical column you should drop it since the imputer class\n",
    "#works on only numerical or quantitative variables.\n",
    "\n",
    "housing_num = housing.drop('ocean_proximity', axis=1)\n",
    "housing_num.head()     # viewing the first 5 rows"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Imputer(axis=0, copy=True, missing_values='NaN', strategy='median', verbose=0)"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# after dropping the categorical column the next thing is to fit the training data\n",
    "\n",
    "imputer.fit(housing_num)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-118.51  ,   34.26  ,   29.    , 2119.5   ,  433.    , 1164.    ,\n",
       "        408.    ,    3.5409])"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "imputer.statistics_  # all the median values of all categories are stored in statistics_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = imputer.transform(housing_num) # its in an array that needs to be converted to a dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "longitude             False\n",
       "latitude              False\n",
       "housing_median_age    False\n",
       "total_rooms           False\n",
       "total_bedrooms        False\n",
       "population            False\n",
       "households            False\n",
       "median_income         False\n",
       "dtype: bool"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing_transformed = pd.DataFrame(X, columns = housing_num.columns)\n",
    "housing_transformed.isnull().any() # check for missing values in columns. NONE!!!!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#  FROM HERE ON WE WOULD BE HANDLING THE TEXT AND CATEGORICAL COLUMNS "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# BECAUSE WE CANT JUST PASS CATEGORICAL VALUES TO A MACHINE LEARNING PREDICTOR"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#  AND TO THIS WE WOULD USE THE SCIKIT-LEARN LABELENCODER CLASS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['<1H OCEAN', 'INLAND', 'ISLAND', 'NEAR BAY', 'NEAR OCEAN'],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import LabelEncoder\n",
    "encoder = LabelEncoder()   # we first encode the values with numbers be4 they are fed to model\n",
    "housing_cat = housing['ocean_proximity']\n",
    "housing_cat_enc = encoder.fit_transform(housing_cat)\n",
    "housing_cat_enc\n",
    "encoder.classes_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    7276\n",
       "1    5263\n",
       "4    2124\n",
       "3    1847\n",
       "2       2\n",
       "dtype: int64"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Series(housing_cat_enc).value_counts()  # loading the encoded result to a pandas series to check the unique value counts"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### fromm the cell above we notice that there a only five classes and some classes are more than the others. this could lead to bias if its not checked. we need to create a binary category for each of the class. OneHotEnccoder to the rescue.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### And OneHotEncoder can only work if we have already encoded the categorical column or columns with LabelEncoder."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1., 0., 0., 0., 0.],\n",
       "       [1., 0., 0., 0., 0.],\n",
       "       [0., 0., 0., 0., 1.],\n",
       "       ...,\n",
       "       [0., 1., 0., 0., 0.],\n",
       "       [1., 0., 0., 0., 0.],\n",
       "       [0., 0., 0., 1., 0.]])"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import OneHotEncoder\n",
    "\n",
    "encoder = OneHotEncoder(sparse=False) #pass the sparse=false arg to output a numpy array\n",
    "housing_cat_1hot = encoder.fit_transform(housing_cat_enc.reshape(-1, 1))\n",
    "housing_cat_1hot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-1.15604281,  0.77194962,  0.74333089, ..., -0.63621141,\n",
       "        -0.42069842, -0.61493744],\n",
       "       [-1.17602483,  0.6596948 , -1.1653172 , ..., -0.99833135,\n",
       "        -1.02222705,  1.33645936],\n",
       "       [ 1.18684903, -1.34218285,  0.18664186, ..., -0.43363936,\n",
       "        -0.0933178 , -0.5320456 ],\n",
       "       ...,\n",
       "       [ 1.58648943, -0.72478134, -1.56295222, ...,  0.60790363,\n",
       "         0.71315642, -0.3167053 ],\n",
       "       [ 0.78221312, -0.85106801,  0.18664186, ..., -0.05717804,\n",
       "        -0.37545069,  0.09812139],\n",
       "       [-1.43579109,  0.99645926,  1.85670895, ..., -0.13515931,\n",
       "         0.3777909 , -0.15779865]])"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "# creating a pipeeline for all the numerical attriibutes of the dataset first\n",
    "num_pipeline = Pipeline([\n",
    "    ('imputer', Imputer(strategy='median')),\n",
    "    ('std_scaler', StandardScaler())\n",
    "])\n",
    "\n",
    "houaing_num_transform=(num_pipeline.fit_transform(housing_num))\n",
    "houaing_num_transform"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.base import BaseEstimator, TransformerMixin\n",
    "\n",
    "# so whats happening here is that i want to pass my dataframe to the pipeline i created earlier to fully automate the\n",
    "#the process, so i create a custom transformation below that accepts the numerical values of the dataset still in df \n",
    "# form and coverts it to a numPy array since Scikit-Learn doesn't handle DataFrames yet\n",
    "\n",
    "# dataframeselector would also make it poossible for u to select the categorical or numerical columns and which\n",
    "# to pass into the dataframe.\n",
    "\n",
    "#its still a process as i would still have to create a transformer for the categorical attributes too. and then combine\n",
    "# those transformers later using sklearns FeatureUnion class\n",
    "\n",
    "class DataFrameSelector(BaseEstimator, TransformerMixin):   \n",
    "    def __init__(self, attribute_names):\n",
    "        self.attribute_names = attribute_names\n",
    "    def fit(self, X, y=None):\n",
    "        return self\n",
    "    def transform(self, X):\n",
    "        return X[self.attribute_names].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1., 0., 0., 0., 0.],\n",
       "       [1., 0., 0., 0., 0.],\n",
       "       [0., 0., 0., 0., 1.],\n",
       "       ...,\n",
       "       [0., 1., 0., 0., 0.],\n",
       "       [1., 0., 0., 0., 0.],\n",
       "       [0., 0., 0., 1., 0.]])"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "num_attribs = list(housing_num)   # selecting only the numerical features\n",
    "cat_attribs = [\"ocean_prox\"]       # selecting only the categorical features\n",
    " \n",
    "num_pipeline = Pipeline([\n",
    "        ('selector', DataFrameSelector(num_attribs)),\n",
    "        ('imputer', Imputer(strategy=\"median\")),\n",
    "        ('std_scaler', StandardScaler()),\n",
    "    ])\n",
    "\n",
    "\n",
    "\n",
    "housing['ocean_prox'] = housing_cat_enc\n",
    "model_housing = housing.drop('ocean_proximity', axis=1)\n",
    "\n",
    "cat_pipeline = Pipeline([\n",
    "        ('selector', DataFrameSelector(cat_attribs)),\n",
    "        ('one_hot_encoder', OneHotEncoder(sparse=False)),\n",
    "    ])\n",
    "\n",
    "housingprep= cat_pipeline.fit_transform(model_housing)\n",
    "housingprep"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "# combining the categorical and numerical pipeline together using FeatureUnion\n",
    "from sklearn.pipeline import FeatureUnion\n",
    "full_pipeline = FeatureUnion(transformer_list=[\n",
    "        (\"num_pipeline\", num_pipeline),\n",
    "        (\"cat_pipeline\", cat_pipeline),\n",
    "    ])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(16512, 13)"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final_housing = full_pipeline.fit_transform(model_housing)\n",
    "final_housing.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Building a quick model with the stratified sampling generated trainset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearRegression(copy_X=True, fit_intercept=True, n_jobs=1, normalize=False)"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "\n",
    "lr = LinearRegression()\n",
    "lr.fit(final_housing, housing_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "69050.98178244587"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# LinearRegrassion Algorithm \n",
    "#rmse = 69036\n",
    "#r2 = 64\n",
    "\n",
    "from sklearn.metrics import mean_squared_error, r2_score\n",
    "model_predictions = lr.predict(final_housing)\n",
    "mse = mean_squared_error(housing_labels, model_predictions)\n",
    "rmse = np.sqrt(mse)\n",
    "rmse\n",
    "\n",
    "#r2 score\n",
    "#acc = r2_score(housing_labels, model_predictions)\n",
    "#acc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "# rmse of 69050 not too good, but the best when compared with our brother random sampling split"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Now let's check with our low bias reserved test-set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# we already have a pipeline. so its just to fit and transform the test set in the pipeline. \n",
    "# this is the importance of pipeline in a machine modelling project. imagine how much time it would cost us to go over\n",
    "# all the feature transformation again\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(4128, 13)"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing_test = strat_test_set.drop('median_house_value', axis=1)    # creating the train set(for the reserved testset)\n",
    "housing_test_labels =  strat_test_set['median_house_value'].copy()  # creating the test set((for the reserved testset))\n",
    "#housing_test_labels.head()  \n",
    "\n",
    "\n",
    "# we have to label encode this one. we cant use the encoded values from the testset as they are of different shapes\n",
    "\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "encoder = LabelEncoder()   # we first encode the values with numbers be4 they are fed to model\n",
    "housing_cat = housing_test['ocean_proximity']\n",
    "housing_cat_enc = encoder.fit_transform(housing_cat)\n",
    "housing_cat_enc\n",
    "encoder.classes_\n",
    "\n",
    "housing_test['ocean_prox'] = housing_cat_enc\n",
    "housing_test1 = housing_test.drop('ocean_proximity', axis=1)\n",
    "\n",
    "# passing the test data to the pipeline!!\n",
    "\n",
    "final_housing_test = full_pipeline.fit_transform(housing_test1)\n",
    "final_housing_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearRegression(copy_X=True, fit_intercept=True, n_jobs=1, normalize=False)"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Modelling the testset data\n",
    "\n",
    "from sklearn.linear_model import LinearRegression\n",
    "\n",
    "lr = LinearRegression()\n",
    "lr.fit(final_housing_test, housing_test_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "67271.20449806108"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# LinearRegrassion Algorithm \n",
    "#rmse = 69036\n",
    "#r2 = 64\n",
    "\n",
    "from sklearn.metrics import mean_squared_error, r2_score\n",
    "model_predictions = lr.predict(final_housing_test)\n",
    "mse = mean_squared_error(housing_test_labels, model_predictions)\n",
    "rmse = np.sqrt(mse)\n",
    "rmse\n",
    "\n",
    "#r2 score\n",
    "#acc = r2_score(housing_labels, model_predictions)\n",
    "#acc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "#  an even lower rmse!!!!! MISSION ACCOMPLISHED "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "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": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}