ML zoomcamp week 3.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "852af9a8",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.model_selection import train_test_split\n",
    "from IPython.display import display\n",
    "from sklearn.metrics import mutual_info_score\n",
    "from sklearn.feature_extraction import DictVectorizer\n",
    "from sklearn.linear_model import LogisticRegression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "bb5d9633",
   "metadata": {},
   "outputs": [],
   "source": [
    "# New York City Airbnb Open Data\n",
    "data = pd.read_csv(\"https://raw.githubusercontent.com/alexeygrigorev/datasets/master/AB_NYC_2019.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "796b21b9",
   "metadata": {},
   "outputs": [],
   "source": [
    "features = ['neighbourhood_group',\n",
    "'room_type',\n",
    "'latitude',\n",
    "'longitude',\n",
    "'price',\n",
    "'minimum_nights',\n",
    "'number_of_reviews',\n",
    "'reviews_per_month',\n",
    "'calculated_host_listings_count',\n",
    "'availability_365']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "cc6e309b",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_full = data[features].fillna(0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "6fb90936",
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>neighbourhood_group</th>\n",
       "      <td>Brooklyn</td>\n",
       "      <td>Manhattan</td>\n",
       "      <td>Manhattan</td>\n",
       "      <td>Brooklyn</td>\n",
       "      <td>Manhattan</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>room_type</th>\n",
       "      <td>Private room</td>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>Private room</td>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>Entire home/apt</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>latitude</th>\n",
       "      <td>40.64749</td>\n",
       "      <td>40.75362</td>\n",
       "      <td>40.80902</td>\n",
       "      <td>40.68514</td>\n",
       "      <td>40.79851</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>longitude</th>\n",
       "      <td>-73.97237</td>\n",
       "      <td>-73.98377</td>\n",
       "      <td>-73.9419</td>\n",
       "      <td>-73.95976</td>\n",
       "      <td>-73.94399</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>price</th>\n",
       "      <td>149</td>\n",
       "      <td>225</td>\n",
       "      <td>150</td>\n",
       "      <td>89</td>\n",
       "      <td>80</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>minimum_nights</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>number_of_reviews</th>\n",
       "      <td>9</td>\n",
       "      <td>45</td>\n",
       "      <td>0</td>\n",
       "      <td>270</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>reviews_per_month</th>\n",
       "      <td>0.21</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.64</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>calculated_host_listings_count</th>\n",
       "      <td>6</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>availability_365</th>\n",
       "      <td>365</td>\n",
       "      <td>355</td>\n",
       "      <td>365</td>\n",
       "      <td>194</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                           0                1             2  \\\n",
       "neighbourhood_group                 Brooklyn        Manhattan     Manhattan   \n",
       "room_type                       Private room  Entire home/apt  Private room   \n",
       "latitude                            40.64749         40.75362      40.80902   \n",
       "longitude                          -73.97237        -73.98377      -73.9419   \n",
       "price                                    149              225           150   \n",
       "minimum_nights                             1                1             3   \n",
       "number_of_reviews                          9               45             0   \n",
       "reviews_per_month                       0.21             0.38           0.0   \n",
       "calculated_host_listings_count             6                2             1   \n",
       "availability_365                         365              355           365   \n",
       "\n",
       "                                              3                4  \n",
       "neighbourhood_group                    Brooklyn        Manhattan  \n",
       "room_type                       Entire home/apt  Entire home/apt  \n",
       "latitude                               40.68514         40.79851  \n",
       "longitude                             -73.95976        -73.94399  \n",
       "price                                        89               80  \n",
       "minimum_nights                                1               10  \n",
       "number_of_reviews                           270                9  \n",
       "reviews_per_month                          4.64              0.1  \n",
       "calculated_host_listings_count                1                1  \n",
       "availability_365                            194                0  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_full.head().T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "cfcb0c1e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    Manhattan\n",
       "dtype: object"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Question 1\n",
    "df_full['neighbourhood_group'].mode()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "1bf4ccbe",
   "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>room_type</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>price</th>\n",
       "      <th>minimum_nights</th>\n",
       "      <th>number_of_reviews</th>\n",
       "      <th>reviews_per_month</th>\n",
       "      <th>calculated_host_listings_count</th>\n",
       "      <th>availability_365</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>neighbourhood_group</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></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>Bronx</th>\n",
       "      <td>1091</td>\n",
       "      <td>1091</td>\n",
       "      <td>1091</td>\n",
       "      <td>1091</td>\n",
       "      <td>1091</td>\n",
       "      <td>1091</td>\n",
       "      <td>1091</td>\n",
       "      <td>1091</td>\n",
       "      <td>1091</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Brooklyn</th>\n",
       "      <td>20104</td>\n",
       "      <td>20104</td>\n",
       "      <td>20104</td>\n",
       "      <td>20104</td>\n",
       "      <td>20104</td>\n",
       "      <td>20104</td>\n",
       "      <td>20104</td>\n",
       "      <td>20104</td>\n",
       "      <td>20104</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Manhattan</th>\n",
       "      <td>21661</td>\n",
       "      <td>21661</td>\n",
       "      <td>21661</td>\n",
       "      <td>21661</td>\n",
       "      <td>21661</td>\n",
       "      <td>21661</td>\n",
       "      <td>21661</td>\n",
       "      <td>21661</td>\n",
       "      <td>21661</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Queens</th>\n",
       "      <td>5666</td>\n",
       "      <td>5666</td>\n",
       "      <td>5666</td>\n",
       "      <td>5666</td>\n",
       "      <td>5666</td>\n",
       "      <td>5666</td>\n",
       "      <td>5666</td>\n",
       "      <td>5666</td>\n",
       "      <td>5666</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Staten Island</th>\n",
       "      <td>373</td>\n",
       "      <td>373</td>\n",
       "      <td>373</td>\n",
       "      <td>373</td>\n",
       "      <td>373</td>\n",
       "      <td>373</td>\n",
       "      <td>373</td>\n",
       "      <td>373</td>\n",
       "      <td>373</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     room_type  latitude  longitude  price  minimum_nights  \\\n",
       "neighbourhood_group                                                          \n",
       "Bronx                     1091      1091       1091   1091            1091   \n",
       "Brooklyn                 20104     20104      20104  20104           20104   \n",
       "Manhattan                21661     21661      21661  21661           21661   \n",
       "Queens                    5666      5666       5666   5666            5666   \n",
       "Staten Island              373       373        373    373             373   \n",
       "\n",
       "                     number_of_reviews  reviews_per_month  \\\n",
       "neighbourhood_group                                         \n",
       "Bronx                             1091               1091   \n",
       "Brooklyn                         20104              20104   \n",
       "Manhattan                        21661              21661   \n",
       "Queens                            5666               5666   \n",
       "Staten Island                      373                373   \n",
       "\n",
       "                     calculated_host_listings_count  availability_365  \n",
       "neighbourhood_group                                                    \n",
       "Bronx                                          1091              1091  \n",
       "Brooklyn                                      20104             20104  \n",
       "Manhattan                                     21661             21661  \n",
       "Queens                                         5666              5666  \n",
       "Staten Island                                   373               373  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_full.groupby(['neighbourhood_group']).count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "e9e1d56d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Data prep\n",
    "df_full_train, df_test = train_test_split(df_full, test_size=0.2, random_state = 42)\n",
    "df_train, df_val = train_test_split(df_full_train, test_size=0.25, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "9bada587",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_train = df_train.reset_index(drop=True)\n",
    "df_val = df_val.reset_index(drop=True)\n",
    "df_test = df_test.reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "316994a5",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train = df_train.price.values\n",
    "y_val = df_val.price.values\n",
    "y_test = df_test.price.values\n",
    "\n",
    "del df_train['price']\n",
    "del df_val['price']\n",
    "del df_test['price']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "c522fce9",
   "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>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>minimum_nights</th>\n",
       "      <th>number_of_reviews</th>\n",
       "      <th>reviews_per_month</th>\n",
       "      <th>calculated_host_listings_count</th>\n",
       "      <th>availability_365</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>latitude</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.080301</td>\n",
       "      <td>0.027441</td>\n",
       "      <td>-0.006246</td>\n",
       "      <td>-0.007159</td>\n",
       "      <td>0.019375</td>\n",
       "      <td>-0.005891</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>longitude</th>\n",
       "      <td>0.080301</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.060660</td>\n",
       "      <td>0.055084</td>\n",
       "      <td>0.134642</td>\n",
       "      <td>-0.117041</td>\n",
       "      <td>0.083666</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>minimum_nights</th>\n",
       "      <td>0.027441</td>\n",
       "      <td>-0.060660</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.076020</td>\n",
       "      <td>-0.120703</td>\n",
       "      <td>0.118647</td>\n",
       "      <td>0.138901</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>number_of_reviews</th>\n",
       "      <td>-0.006246</td>\n",
       "      <td>0.055084</td>\n",
       "      <td>-0.076020</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.590374</td>\n",
       "      <td>-0.073167</td>\n",
       "      <td>0.174477</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>reviews_per_month</th>\n",
       "      <td>-0.007159</td>\n",
       "      <td>0.134642</td>\n",
       "      <td>-0.120703</td>\n",
       "      <td>0.590374</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.048767</td>\n",
       "      <td>0.165376</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>calculated_host_listings_count</th>\n",
       "      <td>0.019375</td>\n",
       "      <td>-0.117041</td>\n",
       "      <td>0.118647</td>\n",
       "      <td>-0.073167</td>\n",
       "      <td>-0.048767</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.225913</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>availability_365</th>\n",
       "      <td>-0.005891</td>\n",
       "      <td>0.083666</td>\n",
       "      <td>0.138901</td>\n",
       "      <td>0.174477</td>\n",
       "      <td>0.165376</td>\n",
       "      <td>0.225913</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                latitude  longitude  minimum_nights  \\\n",
       "latitude                        1.000000   0.080301        0.027441   \n",
       "longitude                       0.080301   1.000000       -0.060660   \n",
       "minimum_nights                  0.027441  -0.060660        1.000000   \n",
       "number_of_reviews              -0.006246   0.055084       -0.076020   \n",
       "reviews_per_month              -0.007159   0.134642       -0.120703   \n",
       "calculated_host_listings_count  0.019375  -0.117041        0.118647   \n",
       "availability_365               -0.005891   0.083666        0.138901   \n",
       "\n",
       "                                number_of_reviews  reviews_per_month  \\\n",
       "latitude                                -0.006246          -0.007159   \n",
       "longitude                                0.055084           0.134642   \n",
       "minimum_nights                          -0.076020          -0.120703   \n",
       "number_of_reviews                        1.000000           0.590374   \n",
       "reviews_per_month                        0.590374           1.000000   \n",
       "calculated_host_listings_count          -0.073167          -0.048767   \n",
       "availability_365                         0.174477           0.165376   \n",
       "\n",
       "                                calculated_host_listings_count  \\\n",
       "latitude                                              0.019375   \n",
       "longitude                                            -0.117041   \n",
       "minimum_nights                                        0.118647   \n",
       "number_of_reviews                                    -0.073167   \n",
       "reviews_per_month                                    -0.048767   \n",
       "calculated_host_listings_count                        1.000000   \n",
       "availability_365                                      0.225913   \n",
       "\n",
       "                                availability_365  \n",
       "latitude                               -0.005891  \n",
       "longitude                               0.083666  \n",
       "minimum_nights                          0.138901  \n",
       "number_of_reviews                       0.174477  \n",
       "reviews_per_month                       0.165376  \n",
       "calculated_host_listings_count          0.225913  \n",
       "availability_365                        1.000000  "
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Question 2\n",
    "numerical = ['latitude',\n",
    "'longitude',\n",
    "'minimum_nights',\n",
    "'number_of_reviews',\n",
    "'reviews_per_month',\n",
    "'calculated_host_listings_count',\n",
    "'availability_365']\n",
    "df_corr = pd.DataFrame()\n",
    "df_train_corr = df_train[numerical]\n",
    "for col in df_train_corr.columns:\n",
    "    df_corr[col] = df_train_corr.corrwith(df_train_corr[col])\n",
    "df_corr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "d4ce0cea",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Question 3\n",
    "avg_price = 152\n",
    "above_avg_train = (y_train >= avg_price).astype(int)\n",
    "above_avg_val = (y_val >= avg_price).astype(int)\n",
    "above_avg_test = (y_test >= avg_price).astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "67aae7d1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.05\n",
      "0.14\n"
     ]
    }
   ],
   "source": [
    "categorical = ['neighbourhood_group',\n",
    "'room_type']\n",
    "score_neighbour = mutual_info_score(above_avg_train, df_train[categorical[0]])\n",
    "score_roomtype = mutual_info_score(above_avg_train, df_train[categorical[1]])\n",
    "print(round(score_neighbour, 2))\n",
    "print(round(score_roomtype, 2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "a39a45ed",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Question 4\n",
    "dv = DictVectorizer(sparse=False)\n",
    "train_dict = df_train[categorical + numerical].to_dict(orient='records')\n",
    "X_train = dv.fit_transform(train_dict)\n",
    "\n",
    "val_dict = df_val[categorical + numerical].to_dict(orient='records')\n",
    "X_val = dv.transform(val_dict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "221cdb60",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = LogisticRegression(solver='lbfgs', C=1.0, random_state=42, max_iter = 5000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "e1e94ef5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression(max_iter=5000, random_state=42)"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(X_train, above_avg_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "76276716",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = model.predict_proba(X_val)[:, 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "0430e701",
   "metadata": {},
   "outputs": [],
   "source": [
    "avg_price_pred = (y_pred >= 0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "05d2468e",
   "metadata": {},
   "outputs": [],
   "source": [
    "score_original = (above_avg_val == avg_price_pred).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "39b04d3b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Removed: latitude\n",
      "Score: -0.058084\n",
      "Removed: longitude\n",
      "Score: -0.058288\n",
      "Removed: minimum_nights\n",
      "Score: -0.062174\n",
      "Removed: number_of_reviews\n",
      "Score: -0.062788\n",
      "Removed: reviews_per_month\n",
      "Score: -0.062072\n",
      "Removed: calculated_host_listings_count\n",
      "Score: -0.060947\n",
      "Removed: availability_365\n",
      "Score: -0.052562\n",
      "Removed: neighbourhood_group\n",
      "Score: -0.022293\n",
      "Removed: room_type\n",
      "Score: 0.0\n"
     ]
    }
   ],
   "source": [
    "# Question 5\n",
    "columns = numerical + categorical\n",
    "\n",
    "for i in range(0, len(columns)):\n",
    "    c = columns.copy()\n",
    "    removed_col = c.pop(i)\n",
    "    \n",
    "    # Train model without i-th element\n",
    "    dv = DictVectorizer(sparse=False)\n",
    "    train_dict = df_train[c].to_dict(orient='records')\n",
    "    X_train = dv.fit_transform(train_dict)\n",
    "\n",
    "    val_dict = df_val[c].to_dict(orient='records')\n",
    "    X_val = dv.transform(val_dict)\n",
    "    \n",
    "    model = LogisticRegression(solver='lbfgs', C=1.0, random_state=42, max_iter = 5000)\n",
    "    model.fit(X_train, above_avg_train)\n",
    "    \n",
    "    y_pred = model.predict_proba(X_val)[:, 1]\n",
    "    avg_price_pred = (y_pred >= 0.5)\n",
    "    \n",
    "    score = (above_avg_val == avg_price_pred).mean()\n",
    "    score_diff = round(score_original - score, 6)\n",
    "    print(\"Removed: {}\".format(removed_col))\n",
    "    print(\"Score: {}\".format(score_diff))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "c7060505",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Question 6\n",
    "\n",
    "# Data prep (copy from earlier)\n",
    "df_full_train, df_test = train_test_split(df_full, test_size=0.2, random_state = 42)\n",
    "df_train, df_val = train_test_split(df_full_train, test_size=0.25, random_state=42)\n",
    "\n",
    "df_train = df_train.reset_index(drop=True)\n",
    "df_val = df_val.reset_index(drop=True)\n",
    "df_test = df_test.reset_index(drop=True)\n",
    "\n",
    "y_train = df_train.price.values\n",
    "y_val = df_val.price.values\n",
    "y_test = df_test.price.values\n",
    "\n",
    "del df_train['price']\n",
    "del df_val['price']\n",
    "del df_test['price']\n",
    "\n",
    "y_train = np.log1p(y_train)\n",
    "y_val = np.log1p(y_val)\n",
    "y_test = np.log1p(y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "ac1c7567",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import Ridge\n",
    "from sklearn.metrics import mean_squared_error"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "dc752eff",
   "metadata": {},
   "outputs": [],
   "source": [
    "dv = DictVectorizer(sparse=False)\n",
    "train_dict = df_train[categorical + numerical].to_dict(orient='records')\n",
    "X_train = dv.fit_transform(train_dict)\n",
    "\n",
    "val_dict = df_val[categorical + numerical].to_dict(orient='records')\n",
    "X_val = dv.transform(val_dict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "id": "ee792b22",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Alpha: 0 \t RMSE: 0.2472\n",
      "Alpha: 0.01 \t RMSE: 0.2471\n",
      "Alpha: 0.1 \t RMSE: 0.2471\n",
      "Alpha: 1 \t RMSE: 0.2471\n",
      "Alpha: 10 \t RMSE: 0.2479\n"
     ]
    }
   ],
   "source": [
    "alpha_list = [0, 0.01, 0.1, 1, 10]\n",
    "for a in alpha_list:\n",
    "    clf = Ridge(alpha=a)\n",
    "    clf.fit(X_train, y_train)\n",
    "    y_pred = clf.predict(X_val)\n",
    "    rmse = mean_squared_error(y_pred, y_val)\n",
    "    print(\"Alpha: {} \\t RMSE: {}\".format(a, round(rmse, 4)))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.8.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}