{
 "cells": [
  {
   "cell_type": "raw",
   "metadata": {},
   "source": [
    "<script>\n",
    "  function code_toggle() {\n",
    "    if (code_shown){\n",
    "      $('div.input').hide('500');\n",
    "      $('#toggleButton').val('Show Code')\n",
    "    } else {\n",
    "      $('div.input').show('500');\n",
    "      $('#toggleButton').val('Hide Code')\n",
    "    }\n",
    "    code_shown = !code_shown\n",
    "  }\n",
    "\n",
    "  $( document ).ready(function(){\n",
    "    code_shown=false;\n",
    "    $('div.input').hide()\n",
    "  });\n",
    "</script>\n",
    "<div align=\"right\">\n",
    "    <form action=\"javascript:code_toggle()\"><input type=\"submit\" id=\"toggleButton\" value=\"Show Code\"></form>\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Introduction\n",
    "Thus far, I have written about ways to transform and encode numeric features, and to geocode and develop clusters of resale flats. These two steps generated new categorical features. In this post, I will demonstrate several ways to encode categorical features for our machine learning model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import modules\n",
    "import category_encoders as ce\n",
    "from IPython.display import Image  \n",
    "import matplotlib as mpl\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import pydotplus\n",
    "import seaborn.apionly as sns\n",
    "from sklearn import tree\n",
    "from sklearn.cluster import KMeans\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from sklearn.tree import DecisionTreeRegressor\n",
    "import warnings\n",
    "\n",
    "# Settings\n",
    "%matplotlib inline\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "# Read data\n",
    "hdb = pd.read_csv('resale-flat-prices-based-on-registration-date-from-jan-2015-onwards.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Recap on New Features Generated\n",
    "  \n",
    "## Numeric Features\n",
    "In my [first post](https://chrischow.github.io/dataandstuff/2018-09-08-hdb-feature-engineering-i/) on feature engineering, we looked at (1) transformation of non-normal features and (2) binning. Specifically, for (1), we performed a log transformation of the target (resale prices); for (2), we looked at fixed-width binning, quantile binning, and decision tree binning, with the latter two being preferred. For this post, I will use decision tree binning to convert floor area into a categorical feature for the purpose of demonstrating categorical feature encoding techniques.\n",
    "  \n",
    "First, we re-fit the decision tree to obtain the criteria for the respective bins."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 2,
     "metadata": {
      "image/png": {
       "width": 750
      }
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Prepare data\n",
    "y_train = hdb.resale_price\n",
    "X_train = hdb[['floor_area_sqm']]\n",
    "\n",
    "# Configure decision tree regressor\n",
    "dt_model = DecisionTreeRegressor(\n",
    "    criterion = 'mse',\n",
    "    max_depth = 4,\n",
    "    min_samples_leaf = 6500,\n",
    "    random_state = 123\n",
    ")\n",
    "\n",
    "# Fit data\n",
    "dt_model.fit(X_train, y_train)\n",
    "\n",
    "# Plot\n",
    "dot_data = tree.export_graphviz(\n",
    "    dt_model, feature_names=['floor_area'],\n",
    "    out_file=None, filled=True,\n",
    "    rounded=True, precision = 0\n",
    ")\n",
    "\n",
    "# Draw graph\n",
    "graph = pydotplus.graph_from_dot_data(dot_data)  \n",
    "\n",
    "# Show graph\n",
    "Image(graph.create_png(), width = 750)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, we extract the leaf node IDs using the `apply` function from our decision tree regressor model. We then rename the categories in order of their predicted mean value."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Extract end nodes\n",
    "hdb['floor_area'] = dt_model.apply(X_train)\n",
    "\n",
    "# Re-name end nodes\n",
    "hdb['floor_area'][hdb.floor_area == 2] = 1\n",
    "hdb['floor_area'][hdb.floor_area == 3] = 2\n",
    "hdb['floor_area'][hdb.floor_area == 7] = 3\n",
    "hdb['floor_area'][hdb.floor_area == 8] = 4\n",
    "hdb['floor_area'][hdb.floor_area == 9] = 5\n",
    "hdb['floor_area'][hdb.floor_area == 11] = 6\n",
    "hdb['floor_area'][hdb.floor_area == 12] = 7\n",
    "hdb['floor_area'] = 'C' + hdb.floor_area.astype('str')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Geocoding and Clusters\n",
    "In my [second post](http://www.google.com) on feature engineering, we looked at how we could extract much more information from the addresses provided in the dataset. We constructed addresses using the block numbers and street names, ran them through the [HERE API](https://developer.here.com/) to obtain geographic coordinates, and ran those through K-Means clustering to obtain clusters for each of the 26 towns in the dataset. Below, we re-generate and attach the clusters to the main dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load coordinates\n",
    "map_latlon = pd.read_csv('latlon_data.csv')\n",
    "\n",
    "# Set index to address\n",
    "map_latlon = map_latlon.rename({'address': 'search_address'})\n",
    "map_latlon = map_latlon.set_index('address')\n",
    "\n",
    "# Separate the mappings\n",
    "map_lat = map_latlon['lat']\n",
    "map_lon = map_latlon['lon']\n",
    "\n",
    "# Create search address feature\n",
    "hdb['search_address'] = hdb.block + '+' + hdb.street_name.str.replace(' ', '+') + '+SINGAPORE'\n",
    "\n",
    "# Map coordinates to main dataset\n",
    "hdb['lat'] = hdb.search_address.map(map_lat)\n",
    "hdb['lon'] = hdb.search_address.map(map_lon)\n",
    "\n",
    "# Optimal clusters\n",
    "clust_results = [7, 5, 6, 7, 7, 4, 5, 6, 5, 5, 3, 6, 5, 7, 7, 6, 6, 5, 5, 8, 5, 5, 4, 2, 1, 2]\n",
    "\n",
    "# Create dataframe\n",
    "disp_clust = pd.DataFrame(\n",
    "    [hdb.town.value_counts().index, clust_results], index = ['Town', 'Clusters']\n",
    ").T.set_index('Town')\n",
    "\n",
    "# Get list of towns\n",
    "all_towns = hdb.town.value_counts().index\n",
    "\n",
    "# Initialise cluster feature\n",
    "hdb['cluster'] = 0\n",
    "\n",
    "# Loop through\n",
    "for town in all_towns:\n",
    "    \n",
    "    # Extract town data\n",
    "    temp_dat = hdb[['lat', 'lon']][hdb.town == town]\n",
    "    temp_dat = temp_dat.reset_index(drop = True)\n",
    "\n",
    "    # Normalise\n",
    "    temp_mm = MinMaxScaler()\n",
    "    temp_mm.fit(temp_dat)\n",
    "    temp_dat_scaled = pd.DataFrame(temp_mm.transform(temp_dat), columns = ['lat', 'lon'])\n",
    "    \n",
    "    # Get optimal clusters\n",
    "    opt_clust = disp_clust.loc[town][0]\n",
    "    \n",
    "    # Fit optimal clusters:\n",
    "    temp_km = KMeans(n_clusters = opt_clust, random_state = 123)\n",
    "    temp_km.fit(temp_dat_scaled)\n",
    "    \n",
    "    # Attach labels\n",
    "    hdb['cluster'][hdb.town == town] = temp_km.labels_ + 1\n",
    "\n",
    "# Rename cluster feature\n",
    "hdb['cluster'] = hdb.town.str.replace(' ', '_').replace('/', '_') + hdb.cluster.astype('str')\n",
    "\n",
    "# Save data\n",
    "hdb.to_csv('hdb_categorical.csv',index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "With that, we have generated two categorical features that I will use to demonstrate several categorical feature encoding techniques:  \n",
    "  \n",
    "* `floor_area`: Decision tree bins for floor area\n",
    "* `cluster`: Cluster within towns  \n",
    "  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Before we begin, let's clean up the dataset for easy previewing:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Re-load data\n",
    "hdb = pd.read_csv('hdb_categorical.csv')\n",
    "\n",
    "# Delete unwanted columns\n",
    "hdb = hdb.drop(['month', 'flat_type', 'block', 'street_name', 'storey_range', 'flat_model',\n",
    "              'lease_commence_date', 'remaining_lease', 'search_address', 'lat', 'lon'], axis = 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Categorical Feature Encoding Techniques\n",
    "There are two types of categorical features: **nominal** and **ordinal** features. Nominal features do not have any order to them. An example of this would be the clusters we developed. Ang Mo Kio Clusters 1 and 2 have no ordinal relation other than the arbitrary numbering I gave them. Ordinal features have some meaningful order behind them. Our decision tree bins can be thought of as an ordinal feature because bin 1 contains resale flats with the smallest floor area, while bin 7 contains resale flats with the largest floor area."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Ordinal Encoding\n",
    "The first and most intuitive way to encode a categorical feature is to assign integers to each category. This applies mainly to **ordinal** features, since there is some hierarchy among the categories. Since we have 7 bins for floor area, we can encode it with the numbers 1 to 7. A decision tree model can process this by making splits between 1 and 7, or by singling out categories (e.g. `floor_cat_ordinal == 7 vs. floor_cat_ordinal != 7`).  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>town</th>\n",
       "      <th>floor_area_sqm</th>\n",
       "      <th>resale_price</th>\n",
       "      <th>floor_area</th>\n",
       "      <th>cluster</th>\n",
       "      <th>fa_ord</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ANG MO KIO</td>\n",
       "      <td>60.0</td>\n",
       "      <td>255000.0</td>\n",
       "      <td>C1</td>\n",
       "      <td>ANG_MO_KIO3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ANG MO KIO</td>\n",
       "      <td>69.0</td>\n",
       "      <td>285000.0</td>\n",
       "      <td>C2</td>\n",
       "      <td>ANG_MO_KIO1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>ANG MO KIO</td>\n",
       "      <td>92.0</td>\n",
       "      <td>385000.0</td>\n",
       "      <td>C3</td>\n",
       "      <td>ANG_MO_KIO5</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>118</th>\n",
       "      <td>BEDOK</td>\n",
       "      <td>103.0</td>\n",
       "      <td>418000.0</td>\n",
       "      <td>C4</td>\n",
       "      <td>BEDOK5</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57</th>\n",
       "      <td>ANG MO KIO</td>\n",
       "      <td>110.0</td>\n",
       "      <td>755000.0</td>\n",
       "      <td>C5</td>\n",
       "      <td>ANG_MO_KIO4</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>ANG MO KIO</td>\n",
       "      <td>125.0</td>\n",
       "      <td>533000.0</td>\n",
       "      <td>C6</td>\n",
       "      <td>ANG_MO_KIO1</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53</th>\n",
       "      <td>ANG MO KIO</td>\n",
       "      <td>138.0</td>\n",
       "      <td>580000.0</td>\n",
       "      <td>C7</td>\n",
       "      <td>ANG_MO_KIO1</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           town  floor_area_sqm  resale_price floor_area      cluster  fa_ord\n",
       "0    ANG MO KIO            60.0      255000.0         C1  ANG_MO_KIO3       1\n",
       "2    ANG MO KIO            69.0      285000.0         C2  ANG_MO_KIO1       2\n",
       "32   ANG MO KIO            92.0      385000.0         C3  ANG_MO_KIO5       3\n",
       "118       BEDOK           103.0      418000.0         C4       BEDOK5       4\n",
       "57   ANG MO KIO           110.0      755000.0         C5  ANG_MO_KIO4       5\n",
       "51   ANG MO KIO           125.0      533000.0         C6  ANG_MO_KIO1       6\n",
       "53   ANG MO KIO           138.0      580000.0         C7  ANG_MO_KIO1       7"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Reset data\n",
    "df = hdb.copy()\n",
    "\n",
    "# Get categories\n",
    "all_cats = sorted(df.floor_area.unique())\n",
    "\n",
    "# Convert floor_cat to ordinal feature\n",
    "df['fa_ord'] = df.floor_area.astype('category', ordered = True, categories = all_cats).cat.codes + 1\n",
    "\n",
    "# Preview\n",
    "df.groupby('fa_ord').head(1).sort_values(by = 'fa_ord')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note that we could also employ the same approach for a nominal feature. However, the numbers assigned would be purely arbitrary. As such, there may not be much intuition in applying ordinal encoding to nominal features."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## One-Hot Encoding\n",
    "The second way to encode a categorical feature is using one-hot encoding. This involves creating a new feature for every category, and setting each feature to 1 if the observation corresponds to that category and 0 otherwise. In other words, we create one binary feature per category."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>floor_area</th>\n",
       "      <th>floor_area_C1</th>\n",
       "      <th>floor_area_C2</th>\n",
       "      <th>floor_area_C3</th>\n",
       "      <th>floor_area_C4</th>\n",
       "      <th>floor_area_C5</th>\n",
       "      <th>floor_area_C6</th>\n",
       "      <th>floor_area_C7</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>C1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>C2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>C3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>118</th>\n",
       "      <td>C4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57</th>\n",
       "      <td>C5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>C6</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53</th>\n",
       "      <td>C7</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",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    floor_area  floor_area_C1  floor_area_C2  floor_area_C3  floor_area_C4  \\\n",
       "0           C1              1              0              0              0   \n",
       "2           C2              0              1              0              0   \n",
       "32          C3              0              0              1              0   \n",
       "118         C4              0              0              0              1   \n",
       "57          C5              0              0              0              0   \n",
       "51          C6              0              0              0              0   \n",
       "53          C7              0              0              0              0   \n",
       "\n",
       "     floor_area_C5  floor_area_C6  floor_area_C7  \n",
       "0                0              0              0  \n",
       "2                0              0              0  \n",
       "32               0              0              0  \n",
       "118              0              0              0  \n",
       "57               1              0              0  \n",
       "51               0              1              0  \n",
       "53               0              0              1  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Reset data\n",
    "df = hdb.copy()\n",
    "\n",
    "# One-hot encoding\n",
    "df = pd.concat([df, pd.get_dummies(df[['floor_area']])], axis = 1)\n",
    "\n",
    "# Drop unused columns\n",
    "df = df.drop(['town', 'floor_area_sqm', 'resale_price', 'cluster'], axis = 1)\n",
    "\n",
    "# View\n",
    "df.groupby('floor_area').head(1).sort_values(by = 'floor_area')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The downside to using this technique is that we end up adding many sparse features (many zeros) to the dataset. Imagine what would happen if we were to apply this approach to nominal features: we would have 134 new features to analyse. Fortunately, there is a neat way to generate fewer binary features: binary encoding."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Binary Encoding\n",
    "Binary encoding converts integers into their bitstrings, and generates one binary feature per digit. I'll demonstrate this on the cluster feature to show how effective it is on categorical features with high cardinality."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>town</th>\n",
       "      <th>cluster</th>\n",
       "      <th>cluster_0</th>\n",
       "      <th>cluster_1</th>\n",
       "      <th>cluster_2</th>\n",
       "      <th>cluster_3</th>\n",
       "      <th>cluster_4</th>\n",
       "      <th>cluster_5</th>\n",
       "      <th>cluster_6</th>\n",
       "      <th>cluster_7</th>\n",
       "      <th>cluster_8</th>\n",
       "      <th>binary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ANG MO KIO</td>\n",
       "      <td>ANG_MO_KIO3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>000000001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59</th>\n",
       "      <td>BEDOK</td>\n",
       "      <td>BEDOK5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>000000110</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>148</th>\n",
       "      <td>BISHAN</td>\n",
       "      <td>BISHAN4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>000001101</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>173</th>\n",
       "      <td>BUKIT BATOK</td>\n",
       "      <td>BUKIT_BATOK2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>000010001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>234</th>\n",
       "      <td>BUKIT MERAH</td>\n",
       "      <td>BUKIT_MERAH5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>000010100</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            town       cluster  cluster_0  cluster_1  cluster_2  cluster_3  \\\n",
       "0     ANG MO KIO   ANG_MO_KIO3          0          0          0          0   \n",
       "59         BEDOK        BEDOK5          0          0          0          0   \n",
       "148       BISHAN       BISHAN4          0          0          0          0   \n",
       "173  BUKIT BATOK  BUKIT_BATOK2          0          0          0          0   \n",
       "234  BUKIT MERAH  BUKIT_MERAH5          0          0          0          0   \n",
       "\n",
       "     cluster_4  cluster_5  cluster_6  cluster_7  cluster_8     binary  \n",
       "0            0          0          0          0          1  000000001  \n",
       "59           0          0          1          1          0  000000110  \n",
       "148          0          1          1          0          1  000001101  \n",
       "173          1          0          0          0          1  000010001  \n",
       "234          1          0          1          0          0  000010100  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Reset data\n",
    "df = hdb.copy()\n",
    "\n",
    "# Set up binary encoder\n",
    "hdb_binary = ce.binary.BinaryEncoder()\n",
    "\n",
    "# Calculate digits:\n",
    "# hdb_binary.calc_required_digits(X = df[['cluster']], col = 'cluster')\n",
    "# Output: 9\n",
    "\n",
    "# Fit, transform, and append features\n",
    "df = pd.concat([df, hdb_binary.fit_transform(df[['cluster']])], axis = 1)\n",
    "\n",
    "# Drop unused columns\n",
    "df = df.drop(['floor_area_sqm', 'resale_price', 'floor_area'], axis = 1)\n",
    "\n",
    "# Add binary number\n",
    "df['binary'] = df.cluster_0.astype('str') + \\\n",
    "    df.cluster_1.astype('str') + \\\n",
    "    df.cluster_2.astype('str') + \\\n",
    "    df.cluster_3.astype('str') + \\\n",
    "    df.cluster_4.astype('str') + \\\n",
    "    df.cluster_5.astype('str') + \\\n",
    "    df.cluster_6.astype('str') + \\\n",
    "    df.cluster_7.astype('str') + \\\n",
    "    df.cluster_8.astype('str')\n",
    "\n",
    "# View\n",
    "df.groupby('town').head(1).head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Using binary encoding, we convert a feature with 134 categories into only 8 binary features. The columns `cluster_1` to `cluster_8` correspond to the binary version of the integer assigned to the cluster feature (captured in the `binary` feature for demonstration purposes only). In this case, `ANG_MO_KIO3` has been coded arbitrarily as the number 1, while `BISHAN4` has been coded as the number 13."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Contrast Coding\n",
    "Contrast coding is about comparing the mean of the target (resale price) for a given category (e.g. category 1 in floor area) against the mean of the means of the target for all categories. That sounds confusing, but it really isn't. First, let's calculate the mean of the target for each floor area category:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>resale_price</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>floor_area</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>C1</th>\n",
       "      <td>304573.420675</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C2</th>\n",
       "      <td>352248.416638</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C3</th>\n",
       "      <td>455199.014743</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C4</th>\n",
       "      <td>417628.500689</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C5</th>\n",
       "      <td>473160.796673</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C6</th>\n",
       "      <td>535978.644477</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C7</th>\n",
       "      <td>622046.397215</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             resale_price\n",
       "floor_area               \n",
       "C1          304573.420675\n",
       "C2          352248.416638\n",
       "C3          455199.014743\n",
       "C4          417628.500689\n",
       "C5          473160.796673\n",
       "C6          535978.644477\n",
       "C7          622046.397215"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Reset data\n",
    "df = hdb.copy()\n",
    "\n",
    "# Calculate target means across floor area categories\n",
    "floor_area_means = pd.DataFrame(df.groupby('floor_area').resale_price.mean())\n",
    "\n",
    "# View\n",
    "floor_area_means"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, we calculate the grand mean - the mean of the means above - and subtract it from the individual means:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>resale_price</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>floor_area</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>C1</th>\n",
       "      <td>-146974.463770</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C2</th>\n",
       "      <td>-99299.467806</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C3</th>\n",
       "      <td>3651.130299</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C4</th>\n",
       "      <td>-33919.383756</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C5</th>\n",
       "      <td>21612.912229</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C6</th>\n",
       "      <td>84430.760033</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C7</th>\n",
       "      <td>170498.512771</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             resale_price\n",
       "floor_area               \n",
       "C1         -146974.463770\n",
       "C2          -99299.467806\n",
       "C3            3651.130299\n",
       "C4          -33919.383756\n",
       "C5           21612.912229\n",
       "C6           84430.760033\n",
       "C7          170498.512771"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Calculate grand mean\n",
    "grand_mean = floor_area_means.mean()\n",
    "\n",
    "# Subtract\n",
    "floor_area_means - grand_mean"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The numbers above are essentially the coefficients on the encoded features when we run a linear regression of resale price against them. What goes on behind the scenes is the following: we simply generate *n - 1* features to represent the comparison of their target means to the grand mean, with one category omitted from the comparison (in this case, category 4, which has many instances of `-1`)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>floor_area</th>\n",
       "      <th>col_floor_area_0</th>\n",
       "      <th>col_floor_area_1</th>\n",
       "      <th>col_floor_area_2</th>\n",
       "      <th>col_floor_area_3</th>\n",
       "      <th>col_floor_area_4</th>\n",
       "      <th>col_floor_area_5</th>\n",
       "      <th>col_floor_area_6</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>C1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>C2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>C3</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>118</th>\n",
       "      <td>C4</td>\n",
       "      <td>1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57</th>\n",
       "      <td>C5</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>C6</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53</th>\n",
       "      <td>C7</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    floor_area  col_floor_area_0  col_floor_area_1  col_floor_area_2  \\\n",
       "0           C1               1.0               1.0               0.0   \n",
       "2           C2               1.0               0.0               1.0   \n",
       "32          C3               1.0               0.0               0.0   \n",
       "118         C4               1.0              -1.0              -1.0   \n",
       "57          C5               1.0               0.0               0.0   \n",
       "51          C6               1.0               0.0               0.0   \n",
       "53          C7               1.0               0.0               0.0   \n",
       "\n",
       "     col_floor_area_3  col_floor_area_4  col_floor_area_5  col_floor_area_6  \n",
       "0                 0.0               0.0               0.0               0.0  \n",
       "2                 0.0               0.0               0.0               0.0  \n",
       "32                1.0               0.0               0.0               0.0  \n",
       "118              -1.0              -1.0              -1.0              -1.0  \n",
       "57                0.0               0.0               0.0               1.0  \n",
       "51                0.0               1.0               0.0               0.0  \n",
       "53                0.0               0.0               1.0               0.0  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Set up sum encoder\n",
    "hdb_sum = ce.sum_coding.SumEncoder()\n",
    "\n",
    "# Fit, transform, and append features\n",
    "df = pd.concat([df, hdb_sum.fit_transform(X = df[['floor_area']], y = df.resale_price)], axis = 1)\n",
    "\n",
    "# Remove unused columns\n",
    "df = df.drop(['town', 'floor_area_sqm', 'resale_price', 'cluster'], axis = 1)\n",
    "\n",
    "# View\n",
    "df.groupby('floor_area').head(1).sort_values(by = 'floor_area')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are several other techniques that involve comparison of means. Helmert encoding, Reverse Helmert Encoding, Backward Difference Encoding, and Forward Difference Encoding all compare the target mean of one category with the mean of the target for other categories or groups of categories. Understanding Contrast Encoding is sufficient to understand the core concept of comparing means. Hence, we need not dive into the above techniques. Nevertheless, it is worth testing them out on your own if you wish to experiment with more categorical encoding schemes."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Stats Encoding\n",
    "Stats encoding requires us to replace each category with some statistic about the target. For example, we could add the mean, standard deviation, and coefficient of variation of the target for each floor area category.\n",
    "  \n",
    "***Note:** I performed the log transformation on resale prices before calculating the statistics.*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>floor_area</th>\n",
       "      <th>log_price</th>\n",
       "      <th>fa_means</th>\n",
       "      <th>fa_std</th>\n",
       "      <th>fa_cv</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>C1</td>\n",
       "      <td>12.449019</td>\n",
       "      <td>12.612001</td>\n",
       "      <td>0.167384</td>\n",
       "      <td>0.013272</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>C2</td>\n",
       "      <td>12.560244</td>\n",
       "      <td>12.751954</td>\n",
       "      <td>0.194041</td>\n",
       "      <td>0.015217</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>C3</td>\n",
       "      <td>12.860999</td>\n",
       "      <td>12.996433</td>\n",
       "      <td>0.243862</td>\n",
       "      <td>0.018764</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>118</th>\n",
       "      <td>C4</td>\n",
       "      <td>12.943237</td>\n",
       "      <td>12.921575</td>\n",
       "      <td>0.194310</td>\n",
       "      <td>0.015038</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57</th>\n",
       "      <td>C5</td>\n",
       "      <td>13.534473</td>\n",
       "      <td>13.039018</td>\n",
       "      <td>0.224441</td>\n",
       "      <td>0.017213</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>C6</td>\n",
       "      <td>13.186277</td>\n",
       "      <td>13.163824</td>\n",
       "      <td>0.229672</td>\n",
       "      <td>0.017447</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53</th>\n",
       "      <td>C7</td>\n",
       "      <td>13.270783</td>\n",
       "      <td>13.323070</td>\n",
       "      <td>0.186554</td>\n",
       "      <td>0.014002</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    floor_area  log_price   fa_means    fa_std     fa_cv\n",
       "0           C1  12.449019  12.612001  0.167384  0.013272\n",
       "2           C2  12.560244  12.751954  0.194041  0.015217\n",
       "32          C3  12.860999  12.996433  0.243862  0.018764\n",
       "118         C4  12.943237  12.921575  0.194310  0.015038\n",
       "57          C5  13.534473  13.039018  0.224441  0.017213\n",
       "51          C6  13.186277  13.163824  0.229672  0.017447\n",
       "53          C7  13.270783  13.323070  0.186554  0.014002"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Reset data\n",
    "df = hdb.copy()\n",
    "\n",
    "# Log transform resale price\n",
    "df['log_price'] = np.log(df.resale_price)\n",
    "\n",
    "# Calculate means\n",
    "fa_means = df.groupby('floor_area').log_price.mean()\n",
    "\n",
    "# Calculate standard deviations\n",
    "fa_std = df.groupby('floor_area').log_price.std()\n",
    "\n",
    "# Calculate coefficients of variation\n",
    "fa_cv = fa_std / fa_means\n",
    "\n",
    "# Create new features\n",
    "df['fa_means'] = df.floor_area.map(fa_means)\n",
    "df['fa_std'] = df.floor_area.map(fa_std)\n",
    "df['fa_cv'] = df.floor_area.map(fa_cv)\n",
    "\n",
    "# Drop unused features\n",
    "df = df.drop(['town', 'floor_area_sqm', 'cluster', 'resale_price'], axis = 1)\n",
    "\n",
    "# View\n",
    "df.groupby('floor_area').head(1).sort_values('floor_area')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Stats encoding enables us to convert categorical features back into numeric features and save on the total number of features added to the dataset, especially when the categorical feature being converted is of high cardinality. For example, using the same computations above, the cluster feature is expanded to only three columns:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>town</th>\n",
       "      <th>cluster</th>\n",
       "      <th>log_price</th>\n",
       "      <th>cl_means</th>\n",
       "      <th>cl_std</th>\n",
       "      <th>cl_cv</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ANG MO KIO</td>\n",
       "      <td>ANG_MO_KIO3</td>\n",
       "      <td>12.449019</td>\n",
       "      <td>12.806781</td>\n",
       "      <td>0.292011</td>\n",
       "      <td>0.022801</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59</th>\n",
       "      <td>BEDOK</td>\n",
       "      <td>BEDOK5</td>\n",
       "      <td>12.413087</td>\n",
       "      <td>12.776089</td>\n",
       "      <td>0.274420</td>\n",
       "      <td>0.021479</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>148</th>\n",
       "      <td>BISHAN</td>\n",
       "      <td>BISHAN4</td>\n",
       "      <td>12.879017</td>\n",
       "      <td>13.062051</td>\n",
       "      <td>0.351218</td>\n",
       "      <td>0.026888</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>173</th>\n",
       "      <td>BUKIT BATOK</td>\n",
       "      <td>BUKIT_BATOK2</td>\n",
       "      <td>12.449019</td>\n",
       "      <td>12.792429</td>\n",
       "      <td>0.263350</td>\n",
       "      <td>0.020586</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>234</th>\n",
       "      <td>BUKIT MERAH</td>\n",
       "      <td>BUKIT_MERAH5</td>\n",
       "      <td>12.542545</td>\n",
       "      <td>13.133030</td>\n",
       "      <td>0.424531</td>\n",
       "      <td>0.032325</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            town       cluster  log_price   cl_means    cl_std     cl_cv\n",
       "0     ANG MO KIO   ANG_MO_KIO3  12.449019  12.806781  0.292011  0.022801\n",
       "59         BEDOK        BEDOK5  12.413087  12.776089  0.274420  0.021479\n",
       "148       BISHAN       BISHAN4  12.879017  13.062051  0.351218  0.026888\n",
       "173  BUKIT BATOK  BUKIT_BATOK2  12.449019  12.792429  0.263350  0.020586\n",
       "234  BUKIT MERAH  BUKIT_MERAH5  12.542545  13.133030  0.424531  0.032325"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Reset data\n",
    "df = hdb.copy()\n",
    "\n",
    "# Log transform resale price\n",
    "df['log_price'] = np.log(df.resale_price)\n",
    "\n",
    "# Calculate means\n",
    "cl_means = df.groupby('cluster').log_price.mean()\n",
    "\n",
    "# Calculate standard deviations\n",
    "cl_std = df.groupby('cluster').log_price.std()\n",
    "\n",
    "# Calculate coefficients of variation\n",
    "cl_cv = cl_std / cl_means\n",
    "\n",
    "# Create new features\n",
    "df['cl_means'] = df.cluster.map(cl_means)\n",
    "df['cl_std'] = df.cluster.map(cl_std)\n",
    "df['cl_cv'] = df.cluster.map(cl_cv)\n",
    "\n",
    "# Drop unused features\n",
    "df = df.drop(['floor_area_sqm', 'floor_area', 'resale_price'], axis = 1)\n",
    "\n",
    "# View\n",
    "df.groupby('town').head(1).sort_values('cluster').head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Feature Interactions\n",
    "In this section, we explore meaningful interactions between features using decision trees. I demonstrate this using flat type and floor area categories. To begin, we need to group some flat types together because there are two categories that have very few observations (1-room and multi-generation flats). We attach them to the next closest categories (2-room and executive flats)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>median</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>flat_type</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1 ROOM</th>\n",
       "      <td>28.0</td>\n",
       "      <td>31.000000</td>\n",
       "      <td>31.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2 ROOM</th>\n",
       "      <td>725.0</td>\n",
       "      <td>45.366897</td>\n",
       "      <td>45.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3 ROOM</th>\n",
       "      <td>17886.0</td>\n",
       "      <td>68.408750</td>\n",
       "      <td>67.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4 ROOM</th>\n",
       "      <td>28612.0</td>\n",
       "      <td>95.887191</td>\n",
       "      <td>94.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5 ROOM</th>\n",
       "      <td>16885.0</td>\n",
       "      <td>118.262718</td>\n",
       "      <td>119.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EXECUTIVE</th>\n",
       "      <td>5618.0</td>\n",
       "      <td>144.000000</td>\n",
       "      <td>145.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MULTI-GENERATION</th>\n",
       "      <td>13.0</td>\n",
       "      <td>162.384615</td>\n",
       "      <td>164.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    count        mean  median\n",
       "flat_type                                    \n",
       "1 ROOM               28.0   31.000000    31.0\n",
       "2 ROOM              725.0   45.366897    45.0\n",
       "3 ROOM            17886.0   68.408750    67.0\n",
       "4 ROOM            28612.0   95.887191    94.0\n",
       "5 ROOM            16885.0  118.262718   119.0\n",
       "EXECUTIVE          5618.0  144.000000   145.0\n",
       "MULTI-GENERATION     13.0  162.384615   164.0"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Re-load data\n",
    "hdb = pd.read_csv('hdb_categorical.csv')\n",
    "\n",
    "# View\n",
    "hdb.pivot_table(\n",
    "    values = 'floor_area_sqm',\n",
    "    index = 'flat_type',\n",
    "    aggfunc = [len, np.mean, np.median]\n",
    ").rename(columns = {'len': 'count'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Attach 1-room flat to 2-room flat category\n",
    "hdb['flat_type'][hdb.flat_type == '1 ROOM'] = '2 ROOM'\n",
    "\n",
    "# Attach 1-room flat to 2-room flat category\n",
    "hdb['flat_type'][hdb.flat_type == 'MULTI-GENERATION'] = 'EXECUTIVE'\n",
    "\n",
    "# View\n",
    "hdb.pivot_table(\n",
    "    values = 'floor_area_sqm',\n",
    "    index = 'flat_type',\n",
    "    aggfunc = [len, np.mean, np.median]\n",
    ").rename(columns = {'len': 'count'})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Combining the 5 flat types and 7 floor area categories, we obtain 20 combined categories. This is because some flat types and floor area categories are not compatible e.g. 1-room and C7, the largest floor area category. I performed one-hot encoding on the features to convert them to a usable form for the decision tree. Then, I set the`min_samples_leaf` parameter, which is the minimum number of samples in an end node, to be 1/35th of the data (about 2,000). I also increase `max_depth`, the maximum depth of the tree, to 20 to allow for more end nodes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 78,
     "metadata": {
      "image/png": {
       "width": 750
      }
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create new feature for flat type X floor area\n",
    "hdb['type_X_floor'] = hdb.flat_type + '_' + hdb.floor_area\n",
    "\n",
    "# Prepare data\n",
    "X_train = pd.get_dummies(hdb[['type_X_floor']])\n",
    "y_train = hdb.resale_price\n",
    "\n",
    "# Configure decision tree regressor\n",
    "dt_interact = DecisionTreeRegressor(\n",
    "    criterion = 'mse',\n",
    "    max_depth = 20,\n",
    "    min_samples_leaf = 2000,\n",
    "    random_state = 123\n",
    ")\n",
    "\n",
    "# Fit data\n",
    "dt_interact.fit(X_train, y_train)\n",
    "\n",
    "# Plot\n",
    "dt_data = tree.export_graphviz(\n",
    "    dt_interact, feature_names=X_train.columns,\n",
    "    out_file=None, filled=True,\n",
    "    rounded=True, precision = 0\n",
    ")\n",
    "\n",
    "# Draw graph\n",
    "int_graph = pydotplus.graph_from_dot_data(dt_data)  \n",
    "\n",
    "# Show graph\n",
    "Image(int_graph.create_png(), width = 750)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The decision tree model gave us 9 categories:  \n",
    "  \n",
    "1. 3-Room flat with floor area category C1\n",
    "2. Executive flat with floor area category C7\n",
    "3. 5-Room flat with floor area category C6\n",
    "4. 3-Room flat with floor area category C2\n",
    "5. 4-Room flat with floor area category C4\n",
    "6. 4-Room flat with floor area category C2\n",
    "7. 5-Room flat with floor area category C5\n",
    "8. 4-Room flat with floor area category C3\n",
    "9. Others  \n",
    "  \n",
    "See the table below for the summary statistics of resale price for each category."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>min</th>\n",
       "      <th>mean</th>\n",
       "      <th>median</th>\n",
       "      <th>max</th>\n",
       "      <th>std</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>type_X_floor_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3-room and C1</th>\n",
       "      <td>11962.0</td>\n",
       "      <td>170000.0</td>\n",
       "      <td>308547.119702</td>\n",
       "      <td>300000.0</td>\n",
       "      <td>645000.00</td>\n",
       "      <td>54117.039961</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3-room and C2</th>\n",
       "      <td>5712.0</td>\n",
       "      <td>206000.0</td>\n",
       "      <td>336330.989279</td>\n",
       "      <td>323000.0</td>\n",
       "      <td>888888.88</td>\n",
       "      <td>66381.179991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4-room and C2</th>\n",
       "      <td>2700.0</td>\n",
       "      <td>255000.0</td>\n",
       "      <td>385922.618519</td>\n",
       "      <td>360000.0</td>\n",
       "      <td>742000.00</td>\n",
       "      <td>85023.593735</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4-room and C3</th>\n",
       "      <td>15949.0</td>\n",
       "      <td>225000.0</td>\n",
       "      <td>455238.890135</td>\n",
       "      <td>418000.0</td>\n",
       "      <td>1028000.00</td>\n",
       "      <td>125347.254027</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4-room and C4</th>\n",
       "      <td>8244.0</td>\n",
       "      <td>275000.0</td>\n",
       "      <td>412737.461640</td>\n",
       "      <td>393000.0</td>\n",
       "      <td>852000.00</td>\n",
       "      <td>81862.680409</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5-room and C5</th>\n",
       "      <td>5497.0</td>\n",
       "      <td>315000.0</td>\n",
       "      <td>485573.205683</td>\n",
       "      <td>443888.0</td>\n",
       "      <td>1145000.00</td>\n",
       "      <td>132101.267003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5-room and C6</th>\n",
       "      <td>9996.0</td>\n",
       "      <td>300000.0</td>\n",
       "      <td>537967.401725</td>\n",
       "      <td>495000.0</td>\n",
       "      <td>1180000.00</td>\n",
       "      <td>138279.756453</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Exec and C7</th>\n",
       "      <td>5277.0</td>\n",
       "      <td>390000.0</td>\n",
       "      <td>631651.504620</td>\n",
       "      <td>610000.0</td>\n",
       "      <td>1160000.00</td>\n",
       "      <td>112003.601242</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Other</th>\n",
       "      <td>4430.0</td>\n",
       "      <td>175000.0</td>\n",
       "      <td>462390.614898</td>\n",
       "      <td>440000.0</td>\n",
       "      <td>1150000.00</td>\n",
       "      <td>166154.386757</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   count       min           mean    median         max  \\\n",
       "type_X_floor_id                                                           \n",
       "3-room and C1    11962.0  170000.0  308547.119702  300000.0   645000.00   \n",
       "3-room and C2     5712.0  206000.0  336330.989279  323000.0   888888.88   \n",
       "4-room and C2     2700.0  255000.0  385922.618519  360000.0   742000.00   \n",
       "4-room and C3    15949.0  225000.0  455238.890135  418000.0  1028000.00   \n",
       "4-room and C4     8244.0  275000.0  412737.461640  393000.0   852000.00   \n",
       "5-room and C5     5497.0  315000.0  485573.205683  443888.0  1145000.00   \n",
       "5-room and C6     9996.0  300000.0  537967.401725  495000.0  1180000.00   \n",
       "Exec and C7       5277.0  390000.0  631651.504620  610000.0  1160000.00   \n",
       "Other             4430.0  175000.0  462390.614898  440000.0  1150000.00   \n",
       "\n",
       "                           std  \n",
       "type_X_floor_id                 \n",
       "3-room and C1     54117.039961  \n",
       "3-room and C2     66381.179991  \n",
       "4-room and C2     85023.593735  \n",
       "4-room and C3    125347.254027  \n",
       "4-room and C4     81862.680409  \n",
       "5-room and C5    132101.267003  \n",
       "5-room and C6    138279.756453  \n",
       "Exec and C7      112003.601242  \n",
       "Other            166154.386757  "
      ]
     },
     "execution_count": 119,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Extract end nodes\n",
    "hdb['type_X_floor_id'] = dt_interact.apply(X_train)\n",
    "\n",
    "# Re-name end nodes\n",
    "hdb['type_X_floor_id'][hdb.type_X_floor_id == 8] = 'Other'\n",
    "hdb['type_X_floor_id'][hdb.type_X_floor_id == 9] = '4-room and C3'\n",
    "hdb['type_X_floor_id'][hdb.type_X_floor_id == 10] = '5-room and C5'\n",
    "hdb['type_X_floor_id'][hdb.type_X_floor_id == 11] = '4-room and C2'\n",
    "hdb['type_X_floor_id'][hdb.type_X_floor_id == 12] = '4-room and C4'\n",
    "hdb['type_X_floor_id'][hdb.type_X_floor_id == 13] = '3-room and C2'\n",
    "hdb['type_X_floor_id'][hdb.type_X_floor_id == 14] = '5-room and C6'\n",
    "hdb['type_X_floor_id'][hdb.type_X_floor_id == 15] = 'Exec and C7'\n",
    "hdb['type_X_floor_id'][hdb.type_X_floor_id == 16] = '3-room and C1'\n",
    "\n",
    "# Summarise\n",
    "int_sum = hdb.pivot_table(\n",
    "    values = 'resale_price',\n",
    "    index = 'type_X_floor_id',\n",
    "    aggfunc = [len, np.min, np.mean, np.median, np.max, np.std]\n",
    ").rename(columns = {'len': 'count', 'amin': 'min', 'amax': 'max'})\n",
    "\n",
    "# View\n",
    "int_sum"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To evaluate the fit of the model, we compute the RMSEs of the different features. Performing the computations for the interactive feature categories, we find that the weighted average of RMSEs was **73.22%** of the RMSE for the dataset ($144,174)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>counts</th>\n",
       "      <th>rmse</th>\n",
       "      <th>weight</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>type_X_floor_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3-room and C1</th>\n",
       "      <td>11962.0</td>\n",
       "      <td>54114.777874</td>\n",
       "      <td>0.171456</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3-room and C2</th>\n",
       "      <td>5712.0</td>\n",
       "      <td>66375.369059</td>\n",
       "      <td>0.081873</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4-room and C2</th>\n",
       "      <td>2700.0</td>\n",
       "      <td>85007.847167</td>\n",
       "      <td>0.038700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4-room and C3</th>\n",
       "      <td>15949.0</td>\n",
       "      <td>125343.324338</td>\n",
       "      <td>0.228604</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4-room and C4</th>\n",
       "      <td>8244.0</td>\n",
       "      <td>81857.715273</td>\n",
       "      <td>0.118165</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5-room and C5</th>\n",
       "      <td>5497.0</td>\n",
       "      <td>132089.250696</td>\n",
       "      <td>0.078791</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5-room and C6</th>\n",
       "      <td>9996.0</td>\n",
       "      <td>138272.839525</td>\n",
       "      <td>0.143277</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Exec and C7</th>\n",
       "      <td>5277.0</td>\n",
       "      <td>111992.988308</td>\n",
       "      <td>0.075637</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Other</th>\n",
       "      <td>4430.0</td>\n",
       "      <td>166135.632381</td>\n",
       "      <td>0.063497</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  counts           rmse    weight\n",
       "type_X_floor_id                                  \n",
       "3-room and C1    11962.0   54114.777874  0.171456\n",
       "3-room and C2     5712.0   66375.369059  0.081873\n",
       "4-room and C2     2700.0   85007.847167  0.038700\n",
       "4-room and C3    15949.0  125343.324338  0.228604\n",
       "4-room and C4     8244.0   81857.715273  0.118165\n",
       "5-room and C5     5497.0  132089.250696  0.078791\n",
       "5-room and C6     9996.0  138272.839525  0.143277\n",
       "Exec and C7       5277.0  111992.988308  0.075637\n",
       "Other             4430.0  166135.632381  0.063497"
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Define function to get square root of mean\n",
    "def rmse(x):\n",
    "    \n",
    "    return np.sqrt(np.mean(x))\n",
    "\n",
    "# Calculate total MSE\n",
    "hdb['tss'] = (hdb.resale_price - np.mean(hdb.resale_price)) ** 2\n",
    "total_rmse = rmse(hdb.tss)\n",
    "\n",
    "# Calculate RMSE for interactive feature\n",
    "# Extract means\n",
    "int_means = int_sum['mean']\n",
    "\n",
    "# Map to existing data\n",
    "hdb['int_means'] = hdb.type_X_floor_id.map(int_means)\n",
    "\n",
    "# Calculate squared errors\n",
    "hdb['int_se'] = (hdb.resale_price - hdb.int_means) ** 2\n",
    "\n",
    "# Summarise\n",
    "int_rmse = hdb.pivot_table(\n",
    "    values = 'int_se',\n",
    "    index = 'type_X_floor_id',\n",
    "    aggfunc = [len, rmse]\n",
    ").rename(columns = {'len': 'counts'})\n",
    "\n",
    "# Calculate weights\n",
    "int_rmse['weight'] = int_rmse.counts/hdb.shape[0]\n",
    "\n",
    "# View\n",
    "int_rmse"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7322241485397311"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Compute weighted average\n",
    "np.sum(int_rmse.rmse * int_rmse.weight) / total_rmse"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Performing the same computations for floor area and flat type give us similar RMSEs of **72.39%** and **73.68%** respectively."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7238646528694407"
      ]
     },
     "execution_count": 115,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Calculate RMSE for floor area\n",
    "# Summarise\n",
    "fa_sum = hdb.pivot_table(\n",
    "    values = 'resale_price',\n",
    "    index = 'floor_area',\n",
    "    aggfunc = [len, np.min, np.mean, np.median, np.max, np.std]\n",
    ").rename(columns = {'len': 'count', 'amin': 'min', 'amax': 'max'})\n",
    "\n",
    "# Extract means\n",
    "fa_means = fa_sum['mean']\n",
    "\n",
    "# Map to existing data\n",
    "hdb['fa_means'] = hdb.floor_area.map(fa_means)\n",
    "\n",
    "# Calculate squared errors\n",
    "hdb['fa_se'] = (hdb.resale_price - hdb.fa_means) ** 2\n",
    "\n",
    "# Summarise\n",
    "fa_rmse = hdb.pivot_table(\n",
    "    values = 'fa_se',\n",
    "    index = 'floor_area',\n",
    "    aggfunc = [len, rmse]\n",
    ").rename(columns = {'len': 'counts'})\n",
    "\n",
    "# Calculate weights\n",
    "fa_rmse['weight'] = fa_rmse.counts/hdb.shape[0]\n",
    "\n",
    "# Compute weighted average\n",
    "np.sum(fa_rmse.rmse * fa_rmse.weight) / total_rmse"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7368107185512195"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Calculate RMSE for flat type\n",
    "# Summarise\n",
    "ft_sum = hdb.pivot_table(\n",
    "    values = 'resale_price',\n",
    "    index = 'flat_type',\n",
    "    aggfunc = [len, np.min, np.mean, np.median, np.max, np.std]\n",
    ").rename(columns = {'len': 'count', 'amin': 'min', 'amax': 'max'})\n",
    "\n",
    "# Extract means\n",
    "ft_means = ft_sum['mean']\n",
    "\n",
    "# Map to existing data\n",
    "hdb['ft_means'] = hdb.flat_type.map(ft_means)\n",
    "\n",
    "# Calculate squared errors\n",
    "hdb['ft_se'] = (hdb.resale_price - hdb.ft_means) ** 2\n",
    "\n",
    "# Summarise\n",
    "ft_rmse = hdb.pivot_table(\n",
    "    values = 'ft_se',\n",
    "    index = 'flat_type',\n",
    "    aggfunc = [len, rmse]\n",
    ").rename(columns = {'len': 'counts'})\n",
    "\n",
    "# Calculate weights\n",
    "ft_rmse['weight'] = ft_rmse.counts/hdb.shape[0]\n",
    "\n",
    "# Compute weighted average\n",
    "np.sum(ft_rmse.rmse * ft_rmse.weight) / total_rmse"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Overall, creating an interaction feature between flat type and floor area did not reduce the RMSE of the dataset, although it helped us to reduce 12 binary features from flat type and floor area into 9 from the interaction feature (more like (1) 12 into 20 through one-hot encoding and (2) 20 into 9 using a decision tree). The cost of using this interaction feature would be a loss of information because we end up compressing 20 features into 9. We also end up being unable to assess the importance of flat type and floor area independently. These costs in intuition and flexibility must be weighed against the *potential* increase in prediction accuracy."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Conclusion\n",
    "In this post, I demonstrated 5 techniques for encoding categorical features and showed how pairs of categorical features can be combined to form interaction features. Like the choice of technique for the binning of numeric features, there is no way to choose a categorical feature encoding or interaction scheme without testing all of them out as part of the overall machine learning pipeline. Only through cross validation can we select the scheme that will perform best on unseen data.  \n",
    "  \n",
    "With that, we have come to the end of this subseries on *(the first, and hopefully, last round of)* feature engineering. We will have to re-visit the feature engineering phase if the machine learning models are unable to satisfactorily detect patterns in the data. I hope this subseries has been of use. To encourage you to dive deeper in this area, I quote Andrew Ng, the former Chief Scientist of Baidu:  \n",
    "  \n",
    "> *\"Coming up with features is difficult, time-consuming, requires expert knowledge. 'Applied machine learning' is basically feature engineering.\"*  \n",
    "> -Andrew Ng"
   ]
  }
 ],
 "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": "3.6.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}