predicting_house_sales.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import libraries\n",
    "import warnings\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "from sklearn.pipeline import make_pipeline\n",
    "from sklearn.impute import SimpleImputer\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from category_encoders import OneHotEncoder\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import accuracy_score\n",
    "from ipywidgets import Dropdown, FloatSlider, IntSlider, interact\n",
    "\n",
    "warnings.simplefilter(action=\"ignore\", category=FutureWarning)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Prepare Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Import"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def wrangle(filepath):\n",
    "    # read csv file\n",
    "    df=pd.read_csv(filepath)\n",
    "    \n",
    "    # remove low cardinality\n",
    "    df.drop(columns=['bus_ter'],inplace=True)\n",
    "    \n",
    "    # remove outliers in n_hot_rooms & rainfall\n",
    "    low = df['rainfall'].quantile(0.1)\n",
    "    mask_rainfall = df['rainfall'].between(low, df['rainfall'].max())\n",
    "    \n",
    "    high = df['n_hot_rooms'].quantile(0.9)\n",
    "    mask_rooms = df['n_hot_rooms'].between(df['n_hot_rooms'].min(), high)\n",
    "\n",
    "    df=df[mask_rooms & mask_rainfall]\n",
    "\n",
    "    # variable transformation\n",
    "    df['avg_dis']=(df['dist1']+df['dist2']+df['dist3']+df['dist4'])/4\n",
    "    df.drop(columns=['dist1','dist2','dist3','dist4'],inplace=True)\n",
    "    \n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>price</th>\n",
       "      <th>resid_area</th>\n",
       "      <th>air_qual</th>\n",
       "      <th>room_num</th>\n",
       "      <th>age</th>\n",
       "      <th>teachers</th>\n",
       "      <th>poor_prop</th>\n",
       "      <th>airport</th>\n",
       "      <th>n_hos_beds</th>\n",
       "      <th>n_hot_rooms</th>\n",
       "      <th>waterbody</th>\n",
       "      <th>rainfall</th>\n",
       "      <th>parks</th>\n",
       "      <th>Sold</th>\n",
       "      <th>avg_dis</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>24.0</td>\n",
       "      <td>32.31</td>\n",
       "      <td>0.538</td>\n",
       "      <td>6.575</td>\n",
       "      <td>65.2</td>\n",
       "      <td>24.7</td>\n",
       "      <td>4.98</td>\n",
       "      <td>YES</td>\n",
       "      <td>5.480</td>\n",
       "      <td>11.1920</td>\n",
       "      <td>River</td>\n",
       "      <td>23</td>\n",
       "      <td>0.049347</td>\n",
       "      <td>0</td>\n",
       "      <td>4.0875</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>21.6</td>\n",
       "      <td>37.07</td>\n",
       "      <td>0.469</td>\n",
       "      <td>6.421</td>\n",
       "      <td>78.9</td>\n",
       "      <td>22.2</td>\n",
       "      <td>9.14</td>\n",
       "      <td>NO</td>\n",
       "      <td>7.332</td>\n",
       "      <td>12.1728</td>\n",
       "      <td>Lake</td>\n",
       "      <td>42</td>\n",
       "      <td>0.046146</td>\n",
       "      <td>1</td>\n",
       "      <td>4.9675</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>33.4</td>\n",
       "      <td>32.18</td>\n",
       "      <td>0.458</td>\n",
       "      <td>6.998</td>\n",
       "      <td>45.8</td>\n",
       "      <td>21.3</td>\n",
       "      <td>2.94</td>\n",
       "      <td>YES</td>\n",
       "      <td>9.268</td>\n",
       "      <td>11.2672</td>\n",
       "      <td>Lake</td>\n",
       "      <td>45</td>\n",
       "      <td>0.047151</td>\n",
       "      <td>0</td>\n",
       "      <td>6.0650</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>36.2</td>\n",
       "      <td>32.18</td>\n",
       "      <td>0.458</td>\n",
       "      <td>7.147</td>\n",
       "      <td>54.2</td>\n",
       "      <td>21.3</td>\n",
       "      <td>5.33</td>\n",
       "      <td>NO</td>\n",
       "      <td>8.824</td>\n",
       "      <td>11.2896</td>\n",
       "      <td>Lake</td>\n",
       "      <td>55</td>\n",
       "      <td>0.039474</td>\n",
       "      <td>0</td>\n",
       "      <td>6.0625</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>28.7</td>\n",
       "      <td>32.18</td>\n",
       "      <td>0.458</td>\n",
       "      <td>6.430</td>\n",
       "      <td>58.7</td>\n",
       "      <td>21.3</td>\n",
       "      <td>5.21</td>\n",
       "      <td>YES</td>\n",
       "      <td>7.174</td>\n",
       "      <td>14.2296</td>\n",
       "      <td>None</td>\n",
       "      <td>53</td>\n",
       "      <td>0.045910</td>\n",
       "      <td>1</td>\n",
       "      <td>6.0600</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   price  resid_area  air_qual  room_num   age  teachers  poor_prop airport  \\\n",
       "0   24.0       32.31     0.538     6.575  65.2      24.7       4.98     YES   \n",
       "1   21.6       37.07     0.469     6.421  78.9      22.2       9.14      NO   \n",
       "3   33.4       32.18     0.458     6.998  45.8      21.3       2.94     YES   \n",
       "4   36.2       32.18     0.458     7.147  54.2      21.3       5.33      NO   \n",
       "5   28.7       32.18     0.458     6.430  58.7      21.3       5.21     YES   \n",
       "\n",
       "   n_hos_beds  n_hot_rooms waterbody  rainfall     parks  Sold  avg_dis  \n",
       "0       5.480      11.1920     River        23  0.049347     0   4.0875  \n",
       "1       7.332      12.1728      Lake        42  0.046146     1   4.9675  \n",
       "3       9.268      11.2672      Lake        45  0.047151     0   6.0650  \n",
       "4       8.824      11.2896      Lake        55  0.039474     0   6.0625  \n",
       "5       7.174      14.2296      None        53  0.045910     1   6.0600  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df=wrangle('House_Price.csv')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(422, 15)\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 422 entries, 0 to 505\n",
      "Data columns (total 15 columns):\n",
      " #   Column       Non-Null Count  Dtype  \n",
      "---  ------       --------------  -----  \n",
      " 0   price        422 non-null    float64\n",
      " 1   resid_area   422 non-null    float64\n",
      " 2   air_qual     422 non-null    float64\n",
      " 3   room_num     422 non-null    float64\n",
      " 4   age          422 non-null    float64\n",
      " 5   teachers     422 non-null    float64\n",
      " 6   poor_prop    422 non-null    float64\n",
      " 7   airport      422 non-null    object \n",
      " 8   n_hos_beds   416 non-null    float64\n",
      " 9   n_hot_rooms  422 non-null    float64\n",
      " 10  waterbody    422 non-null    object \n",
      " 11  rainfall     422 non-null    int64  \n",
      " 12  parks        422 non-null    float64\n",
      " 13  Sold         422 non-null    int64  \n",
      " 14  avg_dis      422 non-null    float64\n",
      "dtypes: float64(11), int64(2), object(2)\n",
      "memory usage: 52.8+ KB\n"
     ]
    }
   ],
   "source": [
    "print(df.shape)\n",
    "df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Explore"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "price          0.000000\n",
       "resid_area     0.000000\n",
       "air_qual       0.000000\n",
       "room_num       0.000000\n",
       "age            0.000000\n",
       "teachers       0.000000\n",
       "poor_prop      0.000000\n",
       "airport        0.000000\n",
       "n_hos_beds     0.014218\n",
       "n_hot_rooms    0.000000\n",
       "waterbody      0.000000\n",
       "rainfall       0.000000\n",
       "parks          0.000000\n",
       "Sold           0.000000\n",
       "avg_dis        0.000000\n",
       "dtype: float64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# missing values\n",
    "df.isnull().sum()/len(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "airport      2\n",
       "waterbody    4\n",
       "dtype: int64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.select_dtypes('object').nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>price</th>\n",
       "      <th>resid_area</th>\n",
       "      <th>air_qual</th>\n",
       "      <th>room_num</th>\n",
       "      <th>age</th>\n",
       "      <th>teachers</th>\n",
       "      <th>poor_prop</th>\n",
       "      <th>n_hos_beds</th>\n",
       "      <th>n_hot_rooms</th>\n",
       "      <th>rainfall</th>\n",
       "      <th>parks</th>\n",
       "      <th>Sold</th>\n",
       "      <th>avg_dis</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>422.000000</td>\n",
       "      <td>422.000000</td>\n",
       "      <td>422.000000</td>\n",
       "      <td>422.000000</td>\n",
       "      <td>422.000000</td>\n",
       "      <td>422.000000</td>\n",
       "      <td>422.000000</td>\n",
       "      <td>416.000000</td>\n",
       "      <td>422.000000</td>\n",
       "      <td>422.000000</td>\n",
       "      <td>422.000000</td>\n",
       "      <td>422.000000</td>\n",
       "      <td>422.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>21.846919</td>\n",
       "      <td>41.654147</td>\n",
       "      <td>0.561990</td>\n",
       "      <td>6.247384</td>\n",
       "      <td>70.832227</td>\n",
       "      <td>21.464455</td>\n",
       "      <td>13.247749</td>\n",
       "      <td>7.913582</td>\n",
       "      <td>12.480463</td>\n",
       "      <td>40.469194</td>\n",
       "      <td>0.055099</td>\n",
       "      <td>0.452607</td>\n",
       "      <td>3.719923</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>9.024018</td>\n",
       "      <td>6.955601</td>\n",
       "      <td>0.119163</td>\n",
       "      <td>0.705062</td>\n",
       "      <td>27.776848</td>\n",
       "      <td>2.140330</td>\n",
       "      <td>7.355683</td>\n",
       "      <td>1.510874</td>\n",
       "      <td>1.545135</td>\n",
       "      <td>11.731669</td>\n",
       "      <td>0.010759</td>\n",
       "      <td>0.498340</td>\n",
       "      <td>2.140696</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>5.000000</td>\n",
       "      <td>30.460000</td>\n",
       "      <td>0.385000</td>\n",
       "      <td>3.561000</td>\n",
       "      <td>2.900000</td>\n",
       "      <td>18.000000</td>\n",
       "      <td>1.730000</td>\n",
       "      <td>5.268000</td>\n",
       "      <td>10.057600</td>\n",
       "      <td>22.000000</td>\n",
       "      <td>0.033292</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.127500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>16.025000</td>\n",
       "      <td>35.640000</td>\n",
       "      <td>0.453000</td>\n",
       "      <td>5.875500</td>\n",
       "      <td>48.625000</td>\n",
       "      <td>19.800000</td>\n",
       "      <td>7.280000</td>\n",
       "      <td>6.632000</td>\n",
       "      <td>11.166400</td>\n",
       "      <td>30.000000</td>\n",
       "      <td>0.046922</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.991875</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>20.550000</td>\n",
       "      <td>39.900000</td>\n",
       "      <td>0.538000</td>\n",
       "      <td>6.183500</td>\n",
       "      <td>81.650000</td>\n",
       "      <td>20.900000</td>\n",
       "      <td>12.335000</td>\n",
       "      <td>8.012000</td>\n",
       "      <td>12.206400</td>\n",
       "      <td>40.000000</td>\n",
       "      <td>0.054361</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.078750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>24.800000</td>\n",
       "      <td>48.100000</td>\n",
       "      <td>0.647000</td>\n",
       "      <td>6.594500</td>\n",
       "      <td>95.000000</td>\n",
       "      <td>22.600000</td>\n",
       "      <td>17.600000</td>\n",
       "      <td>9.149000</td>\n",
       "      <td>14.112200</td>\n",
       "      <td>51.000000</td>\n",
       "      <td>0.061719</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>5.113125</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>50.000000</td>\n",
       "      <td>57.740000</td>\n",
       "      <td>0.871000</td>\n",
       "      <td>8.780000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>27.400000</td>\n",
       "      <td>37.970000</td>\n",
       "      <td>10.876000</td>\n",
       "      <td>15.153600</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>0.086711</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>12.127500</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            price  resid_area    air_qual    room_num         age    teachers  \\\n",
       "count  422.000000  422.000000  422.000000  422.000000  422.000000  422.000000   \n",
       "mean    21.846919   41.654147    0.561990    6.247384   70.832227   21.464455   \n",
       "std      9.024018    6.955601    0.119163    0.705062   27.776848    2.140330   \n",
       "min      5.000000   30.460000    0.385000    3.561000    2.900000   18.000000   \n",
       "25%     16.025000   35.640000    0.453000    5.875500   48.625000   19.800000   \n",
       "50%     20.550000   39.900000    0.538000    6.183500   81.650000   20.900000   \n",
       "75%     24.800000   48.100000    0.647000    6.594500   95.000000   22.600000   \n",
       "max     50.000000   57.740000    0.871000    8.780000  100.000000   27.400000   \n",
       "\n",
       "        poor_prop  n_hos_beds  n_hot_rooms    rainfall       parks  \\\n",
       "count  422.000000  416.000000   422.000000  422.000000  422.000000   \n",
       "mean    13.247749    7.913582    12.480463   40.469194    0.055099   \n",
       "std      7.355683    1.510874     1.545135   11.731669    0.010759   \n",
       "min      1.730000    5.268000    10.057600   22.000000    0.033292   \n",
       "25%      7.280000    6.632000    11.166400   30.000000    0.046922   \n",
       "50%     12.335000    8.012000    12.206400   40.000000    0.054361   \n",
       "75%     17.600000    9.149000    14.112200   51.000000    0.061719   \n",
       "max     37.970000   10.876000    15.153600   60.000000    0.086711   \n",
       "\n",
       "             Sold     avg_dis  \n",
       "count  422.000000  422.000000  \n",
       "mean     0.452607    3.719923  \n",
       "std      0.498340    2.140696  \n",
       "min      0.000000    1.127500  \n",
       "25%      0.000000    1.991875  \n",
       "50%      0.000000    3.078750  \n",
       "75%      1.000000    5.113125  \n",
       "max      1.000000   12.127500  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.boxplot(y='n_hot_rooms',data=df);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\USER\\Anaconda3\\lib\\site-packages\\matplotlib\\axes\\_axes.py:6462: UserWarning: The 'normed' kwarg is deprecated, and has been replaced by the 'density' kwarg.\n",
      "  warnings.warn(\"The 'normed' kwarg is deprecated, and has been \"\n",
      "C:\\Users\\USER\\Anaconda3\\lib\\site-packages\\matplotlib\\axes\\_axes.py:6462: UserWarning: The 'normed' kwarg is deprecated, and has been replaced by the 'density' kwarg.\n",
      "  warnings.warn(\"The 'normed' kwarg is deprecated, and has been \"\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.jointplot(y='rainfall',x='Sold',data=df);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "corr = df.select_dtypes('number').drop(columns='Sold').corr()\n",
    "sns.heatmap(corr);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Split Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "X=df.loc[:,df.columns!='Sold']\n",
    "y=df['Sold']\n",
    "X_train,X_test,y_train,y_test = train_test_split(X,y,test_size=0.2,random_state=42)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Build Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Baseline Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Baseline Accuracy: 0.55\n"
     ]
    }
   ],
   "source": [
    "# baseline \n",
    "acc_baseline = y_train.value_counts(normalize=True).max()\n",
    "print('Baseline Accuracy:', round(acc_baseline,2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Iterate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Pipeline(steps=[('onehotencoder',\n",
       "                 OneHotEncoder(cols=['airport', 'waterbody'],\n",
       "                               use_cat_names=True)),\n",
       "                ('simpleimputer', SimpleImputer()),\n",
       "                ('logisticregression', LogisticRegression(max_iter=1000))])"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model=make_pipeline(OneHotEncoder(use_cat_names=True),SimpleImputer(),LogisticRegression(max_iter=1000))\n",
    "model.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Evalute"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training Accuracy: 0.69\n",
      "Test Accuracy: 0.71\n"
     ]
    }
   ],
   "source": [
    "acc_train = model.score(X_train, y_train)\n",
    "acc_test = model.score(X_test, y_test)\n",
    "print('Training Accuracy:', round(acc_train, 2))\n",
    "print('Test Accuracy:', round(acc_test, 2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Communicate Results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "features = model.named_steps['onehotencoder'].get_feature_names_out()\n",
    "importances = model.named_steps['logisticregression'].coef_[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "odds_ratio = pd.Series(np.exp(importances), index=features).sort_values()\n",
    "odds_ratio.plot(kind='bar');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "def make_prediction(age,air_qual,airport,avg_dis,n_hos_beds,n_hot_rooms,parks,poor_prop,price,rainfall,resid_area,room_num,teachers,waterbody):\n",
    "    data={\n",
    "        'age':age,\n",
    " 'air_qual':air_qual,\n",
    " 'airport': airport,\n",
    " 'avg_dis':avg_dis,\n",
    " 'n_hos_beds': n_hos_beds,\n",
    " 'n_hot_rooms': n_hot_rooms,\n",
    " 'parks': parks,\n",
    " 'poor_prop': poor_prop,\n",
    " 'price': price,\n",
    " 'rainfall': rainfall,\n",
    " 'resid_area': resid_area,\n",
    " 'room_num':room_num,\n",
    " 'teachers': teachers,\n",
    "    'waterbody'  :waterbody,  \n",
    "    }\n",
    "    df=pd.DataFrame(data,index=[0])\n",
    "    prediction = model.predict(df).round(2)[0]\n",
    "    return f\"Apartment will sell or not: {prediction}\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c5889bf4664241a49bdbe0bcc6580a79",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "interactive(children=(FloatSlider(value=71.41602373887245, description='age', min=2.9, step=0.01), FloatSlider…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "interact(\n",
    "    make_prediction,\n",
    "    rainfall=IntSlider(\n",
    "        min=X_train[\"rainfall\"].min(),\n",
    "        max=X_train[\"rainfall\"].max(),\n",
    "        value=X_train[\"rainfall\"].mean(),\n",
    "    ),\n",
    "    age=FloatSlider(\n",
    "        min=X_train[\"age\"].min(),\n",
    "        max=X_train[\"age\"].max(),\n",
    "        step=0.01,\n",
    "        value=X_train[\"age\"].mean(),\n",
    "    ),\n",
    "    air_qual=FloatSlider(\n",
    "        min=X_train[\"air_qual\"].min(),\n",
    "        max=X_train[\"air_qual\"].max(),\n",
    "        step=0.01,\n",
    "        value=X_train[\"air_qual\"].mean(),\n",
    "    ),\n",
    "    avg_dis=FloatSlider(\n",
    "        min=X_train[\"avg_dis\"].min(),\n",
    "        max=X_train[\"avg_dis\"].max(),\n",
    "        step=0.01,\n",
    "        value=X_train[\"avg_dis\"].mean(),\n",
    "    ),\n",
    "    n_hos_beds=FloatSlider(\n",
    "        min=X_train[\"n_hos_beds\"].min(),\n",
    "        max=X_train[\"n_hos_beds\"].max(),\n",
    "        step=0.01,\n",
    "        value=X_train[\"n_hos_beds\"].mean(),\n",
    "    ),\n",
    "    n_hot_rooms=FloatSlider(\n",
    "        min=X_train[\"n_hot_rooms\"].min(),\n",
    "        max=X_train[\"n_hot_rooms\"].max(),\n",
    "        step=0.01,\n",
    "        value=X_train[\"n_hot_rooms\"].mean(),\n",
    "    ),parks=FloatSlider(\n",
    "        min=X_train[\"parks\"].min(),\n",
    "        max=X_train[\"parks\"].max(),\n",
    "        step=0.01,\n",
    "        value=X_train[\"parks\"].mean(),\n",
    "    ),poor_prop=FloatSlider(\n",
    "        min=X_train[\"poor_prop\"].min(),\n",
    "        max=X_train[\"poor_prop\"].max(),\n",
    "        step=0.01,\n",
    "        value=X_train[\"poor_prop\"].mean(),\n",
    "    ),price=FloatSlider(\n",
    "        min=X_train[\"price\"].min(),\n",
    "        max=X_train[\"price\"].max(),\n",
    "        step=0.01,\n",
    "        value=X_train[\"price\"].mean(),\n",
    "    ),resid_area=FloatSlider(\n",
    "        min=X_train[\"resid_area\"].min(),\n",
    "        max=X_train[\"resid_area\"].max(),\n",
    "        step=0.01,\n",
    "        value=X_train[\"resid_area\"].mean(),\n",
    "    ),room_num=FloatSlider(\n",
    "        min=X_train[\"room_num\"].min(),\n",
    "        max=X_train[\"room_num\"].max(),\n",
    "        step=0.01,\n",
    "        value=X_train[\"room_num\"].mean(),\n",
    "    ),teachers=FloatSlider(\n",
    "        min=X_train[\"teachers\"].min(),\n",
    "        max=X_train[\"teachers\"].max(),\n",
    "        step=0.01,\n",
    "        value=X_train[\"teachers\"].mean(),\n",
    "    ),\n",
    "    airport=Dropdown(options=sorted(X_train[\"airport\"].unique())),\n",
    "    waterbody=Dropdown(options=sorted(X_train[\"waterbody\"].unique())),\n",
    ");"
   ]
  },
  {
   "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}