Code/05 - Yelp Further Feature Engineering and Modeling.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Further Feature Engineering and Modeling\n",
    "\n",
    "This journal represents a humbling attempt to best our own model. Aggressive KMeans clustering initially led us to believe we had a high performing model, but we later realized that it was merely training itself to be specific to LA county.\n",
    "\n",
    "We will walk you through our thought processess and discuss each model's shortcomings, as well as recommendations depending on the goal of your Yelp app."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import StandardScaler, OneHotEncoder, PolynomialFeatures\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.neighbors import KNeighborsRegressor\n",
    "from sklearn.tree import DecisionTreeRegressor\n",
    "from sklearn.ensemble import BaggingRegressor, RandomForestRegressor, AdaBoostRegressor\n",
    "from sklearn.svm import SVR\n",
    "from sklearn.metrics import mean_squared_error\n",
    "\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.metrics import silhouette_samples, silhouette_score\n",
    "from sklearn.linear_model import Lasso\n",
    "from sklearn.linear_model import LassoCV\n",
    "from sklearn.cluster import KMeans\n",
    "\n",
    "from xgboost import XGBRegressor\n",
    "\n",
    "from sklearn.preprocessing import FunctionTransformer\n",
    "from sklearn.pipeline import FeatureUnion\n",
    "from sklearn.base import BaseEstimator, TransformerMixin\n",
    "from matplotlib.ticker import FuncFormatter\n",
    "from collections import defaultdict\n",
    "import pickle\n",
    "import folium"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>price</th>\n",
       "      <th>review_count</th>\n",
       "      <th>rating</th>\n",
       "      <th>zip_code</th>\n",
       "      <th>city</th>\n",
       "      <th>alias</th>\n",
       "      <th>category</th>\n",
       "      <th>med_agi</th>\n",
       "      <th>avg_agi</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>zzvkb4J_sQAk8N20OWdZhg</td>\n",
       "      <td>33.911154</td>\n",
       "      <td>-117.881259</td>\n",
       "      <td>2</td>\n",
       "      <td>1257.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>92821</td>\n",
       "      <td>Brea</td>\n",
       "      <td>lucilles-smokehouse-bar-b-que-brea</td>\n",
       "      <td>['bbq', 'smokehouse', 'southern']</td>\n",
       "      <td>86070</td>\n",
       "      <td>106163</td>\n",
       "      <td>170</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>zWs68lBY2PyNOTRJhxXapg</td>\n",
       "      <td>33.911959</td>\n",
       "      <td>-117.867374</td>\n",
       "      <td>1</td>\n",
       "      <td>22.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>92821</td>\n",
       "      <td>Brea</td>\n",
       "      <td>subway-restaurants-brea-4</td>\n",
       "      <td>['sandwiches']</td>\n",
       "      <td>86070</td>\n",
       "      <td>106163</td>\n",
       "      <td>170</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>zOYNwoDjpg6Ehwn0RKX9qA</td>\n",
       "      <td>33.911283</td>\n",
       "      <td>-117.867449</td>\n",
       "      <td>2</td>\n",
       "      <td>367.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>92821</td>\n",
       "      <td>Brea</td>\n",
       "      <td>nekter-juice-bar-brea</td>\n",
       "      <td>['juicebars', 'acaibowls', 'icecream']</td>\n",
       "      <td>86070</td>\n",
       "      <td>106163</td>\n",
       "      <td>170</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>zNNLl821AlzApHYqRxNALw</td>\n",
       "      <td>33.911431</td>\n",
       "      <td>-117.865941</td>\n",
       "      <td>1</td>\n",
       "      <td>84.0</td>\n",
       "      <td>2.5</td>\n",
       "      <td>92821</td>\n",
       "      <td>Brea</td>\n",
       "      <td>arbys-brea-2</td>\n",
       "      <td>['sandwiches', 'hotdogs']</td>\n",
       "      <td>86070</td>\n",
       "      <td>106163</td>\n",
       "      <td>170</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>zJ4pvxtpo3Y5LvylLXmCzA</td>\n",
       "      <td>33.932860</td>\n",
       "      <td>-117.916253</td>\n",
       "      <td>2</td>\n",
       "      <td>231.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>92821</td>\n",
       "      <td>Brea</td>\n",
       "      <td>portofino-trattoria-brea</td>\n",
       "      <td>['italian']</td>\n",
       "      <td>86070</td>\n",
       "      <td>106163</td>\n",
       "      <td>170</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       id   latitude   longitude  price  review_count  rating  \\\n",
       "0  zzvkb4J_sQAk8N20OWdZhg  33.911154 -117.881259      2        1257.0     3.5   \n",
       "1  zWs68lBY2PyNOTRJhxXapg  33.911959 -117.867374      1          22.0     3.0   \n",
       "2  zOYNwoDjpg6Ehwn0RKX9qA  33.911283 -117.867449      2         367.0     3.5   \n",
       "3  zNNLl821AlzApHYqRxNALw  33.911431 -117.865941      1          84.0     2.5   \n",
       "4  zJ4pvxtpo3Y5LvylLXmCzA  33.932860 -117.916253      2         231.0     4.0   \n",
       "\n",
       "   zip_code  city                               alias  \\\n",
       "0     92821  Brea  lucilles-smokehouse-bar-b-que-brea   \n",
       "1     92821  Brea           subway-restaurants-brea-4   \n",
       "2     92821  Brea               nekter-juice-bar-brea   \n",
       "3     92821  Brea                        arbys-brea-2   \n",
       "4     92821  Brea            portofino-trattoria-brea   \n",
       "\n",
       "                                 category  med_agi  avg_agi  count  \n",
       "0       ['bbq', 'smokehouse', 'southern']    86070   106163    170  \n",
       "1                          ['sandwiches']    86070   106163    170  \n",
       "2  ['juicebars', 'acaibowls', 'icecream']    86070   106163    170  \n",
       "3               ['sandwiches', 'hotdogs']    86070   106163    170  \n",
       "4                             ['italian']    86070   106163    170  "
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df= pd.read_csv('../data/03_data_income.csv')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "id               object\n",
       "latitude        float64\n",
       "longitude       float64\n",
       "price             int64\n",
       "review_count    float64\n",
       "rating          float64\n",
       "zip_code         object\n",
       "city             object\n",
       "alias            object\n",
       "category         object\n",
       "med_agi           int64\n",
       "avg_agi           int64\n",
       "count             int64\n",
       "dtype: object"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['zip_code'] = df['zip_code'].astype(str)\n",
    "df.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_pivot = df.pivot_table(index='zip_code').reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# df_pivot.to_csv('../data/data_zipcode_j.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Making custom transformers: Adjusted Price\n",
    "\n",
    "Different types of businesses will have a different distribution of price ranges depending on the product or service they provide. For example, a taco shop with a price rating of '$$' is actually relatively expensive, whereas a sushi restaurant with the same rating is quite common. We hypothesize that a neighborhood with expensive taco shops might be more affluent than a neighborhood with average-priced sushi restaurants; more generally, neighborhoods with disproportionately expensive businessess might be more wealthy than those with just objectively expensive businesses."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1.2374581939799332\n",
      "2.0346938775510206\n"
     ]
    }
   ],
   "source": [
    "print(df[df['category'].str.contains('tacos')]['price'].mean())\n",
    "print(df[df['category'].str.contains('sushi')]['price'].mean())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We introduce the idea of an adjusted price, which has the goal of scaling a listing's price relative to the overall expensiveness of its parent category. We will calculate the adjusted price by looking at businesses category-by-category and standard scaling the listed price to the mean of that category.\n",
    "\n",
    "Since we do not want to have any data leakage, we can only apply this transformation on our training data set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "metadata": {},
   "outputs": [],
   "source": [
    "# https://towardsdatascience.com/custom-transformers-and-ml-data-pipelines-with-python-20ea2a7adb65\n",
    "# https://www.slideshare.net/PyData/julie-michelman-pandas-pipelines-and-custom-transformers\n",
    "# https://gist.github.com/tdpetrou/6a97304dd4452a53be98e4f4e93196e6\n",
    "# https://stackoverflow.com/questions/44408861/\n",
    "\n",
    "class CategoryPricer(BaseEstimator, TransformerMixin):\n",
    "    \n",
    "    def __init__(self):\n",
    "        self.pcv = None\n",
    "        \n",
    "    def fit(self, X, y=None):\n",
    "        self.pcv = CountVectorizer().fit(X['category'])\n",
    "        print('cv fit')\n",
    "        return self\n",
    "            \n",
    "    def transform(self, X, y=None):     \n",
    "        \n",
    "        catpri_cols = ['cat_pri_' + s for s in self.pcv.get_feature_names()]\n",
    "        print('cv cols')\n",
    "        C = self.pcv.transform(X['category']).toarray()\n",
    "        print('cv transform')\n",
    "        for i in range(len(C)):\n",
    "            C[i] = C[i] * X['price'].values[i]\n",
    "        print('cv * price')\n",
    "        Cdf = pd.DataFrame(C, columns=catpri_cols, index=X.index)\n",
    "        print('cdf')\n",
    "        Cdf.replace(0, np.NaN, inplace=True)\n",
    "        print('cdf replace 0 w nan')\n",
    "        X = X.join(Cdf)\n",
    "        print('X join cdf')\n",
    "        return X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "class StandardizerAndCategoryReducer(BaseEstimator, TransformerMixin):\n",
    "    \n",
    "    def __init__(self):\n",
    "        self.sc = None\n",
    "        \n",
    "    def fit(self, X, y=None):\n",
    "        num_cols = [col for col in X.describe().columns if 'cluster' not in col]\n",
    "        self.sc = StandardScaler().fit(X[num_cols])\n",
    "        return self\n",
    "            \n",
    "    def transform(self, X, y=None):\n",
    "        num_cols = [col for col in X.describe().columns if 'cluster' not in col]\n",
    "        cat_cols = [col for col in X.columns if X[col].dtype == 'O']\n",
    "        cluster_cols = [col for col in X.columns if 'cluster' in col]\n",
    "        \n",
    "        S = self.sc.transform(X[num_cols])\n",
    "        Sdf = pd.DataFrame(S, columns=num_cols, index=X.index)\n",
    "        catpri_cols = [col for col in X.columns if 'cat_pri_' in col]\n",
    "        \n",
    "        Sdf.loc[:, 'price_adj'] = Sdf.loc[:, catpri_cols].mean(axis=1)\n",
    "        Sdf.drop(columns=catpri_cols, inplace=True)\n",
    "        \n",
    "        X = X.loc[:, cat_cols + cluster_cols].join(Sdf)\n",
    "        X.drop(columns='category', inplace=True)\n",
    "        \n",
    "        return X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "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>price</th>\n",
       "      <th>review_count</th>\n",
       "      <th>rating</th>\n",
       "      <th>category</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>33.911154</td>\n",
       "      <td>-117.881259</td>\n",
       "      <td>2</td>\n",
       "      <td>1257.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>['bbq', 'smokehouse', 'southern']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>33.911959</td>\n",
       "      <td>-117.867374</td>\n",
       "      <td>1</td>\n",
       "      <td>22.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>['sandwiches']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>33.911283</td>\n",
       "      <td>-117.867449</td>\n",
       "      <td>2</td>\n",
       "      <td>367.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>['juicebars', 'acaibowls', 'icecream']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>33.911431</td>\n",
       "      <td>-117.865941</td>\n",
       "      <td>1</td>\n",
       "      <td>84.0</td>\n",
       "      <td>2.5</td>\n",
       "      <td>['sandwiches', 'hotdogs']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>33.932860</td>\n",
       "      <td>-117.916253</td>\n",
       "      <td>2</td>\n",
       "      <td>231.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>['italian']</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    latitude   longitude  price  review_count  rating  \\\n",
       "0  33.911154 -117.881259      2        1257.0     3.5   \n",
       "1  33.911959 -117.867374      1          22.0     3.0   \n",
       "2  33.911283 -117.867449      2         367.0     3.5   \n",
       "3  33.911431 -117.865941      1          84.0     2.5   \n",
       "4  33.932860 -117.916253      2         231.0     4.0   \n",
       "\n",
       "                                 category  \n",
       "0       ['bbq', 'smokehouse', 'southern']  \n",
       "1                          ['sandwiches']  \n",
       "2  ['juicebars', 'acaibowls', 'icecream']  \n",
       "3               ['sandwiches', 'hotdogs']  \n",
       "4                             ['italian']  "
      ]
     },
     "execution_count": 108,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = df.drop(columns=['avg_agi', 'med_agi', \n",
    "                     'count', 'city', 'id', 'alias', 'zip_code'])\n",
    "y = df['med_agi']\n",
    "X.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "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>price</th>\n",
       "      <th>review_count</th>\n",
       "      <th>rating</th>\n",
       "      <th>price_adj</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.865918</td>\n",
       "      <td>1.797864</td>\n",
       "      <td>0.898905</td>\n",
       "      <td>2.401713</td>\n",
       "      <td>-0.106237</td>\n",
       "      <td>0.273958</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-0.860828</td>\n",
       "      <td>1.866601</td>\n",
       "      <td>-0.833936</td>\n",
       "      <td>-0.523979</td>\n",
       "      <td>-0.721385</td>\n",
       "      <td>-0.593050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.865098</td>\n",
       "      <td>1.866229</td>\n",
       "      <td>0.898905</td>\n",
       "      <td>0.293319</td>\n",
       "      <td>-0.106237</td>\n",
       "      <td>1.440286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.864164</td>\n",
       "      <td>1.873697</td>\n",
       "      <td>-0.833936</td>\n",
       "      <td>-0.377103</td>\n",
       "      <td>-1.336532</td>\n",
       "      <td>-0.419023</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-0.728723</td>\n",
       "      <td>1.624631</td>\n",
       "      <td>0.898905</td>\n",
       "      <td>-0.028862</td>\n",
       "      <td>0.508910</td>\n",
       "      <td>0.364028</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>-0.871440</td>\n",
       "      <td>1.858230</td>\n",
       "      <td>0.898905</td>\n",
       "      <td>-0.388947</td>\n",
       "      <td>0.508910</td>\n",
       "      <td>1.608113</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>-0.824289</td>\n",
       "      <td>1.600982</td>\n",
       "      <td>-0.833936</td>\n",
       "      <td>-0.400792</td>\n",
       "      <td>-0.721385</td>\n",
       "      <td>-0.414354</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>-0.742374</td>\n",
       "      <td>1.737415</td>\n",
       "      <td>0.898905</td>\n",
       "      <td>0.025624</td>\n",
       "      <td>0.508910</td>\n",
       "      <td>1.172865</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>-0.730867</td>\n",
       "      <td>1.710399</td>\n",
       "      <td>-0.833936</td>\n",
       "      <td>-0.542931</td>\n",
       "      <td>-2.566826</td>\n",
       "      <td>-0.340126</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>-0.860725</td>\n",
       "      <td>1.790306</td>\n",
       "      <td>0.898905</td>\n",
       "      <td>2.039259</td>\n",
       "      <td>-0.106237</td>\n",
       "      <td>0.364028</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>-0.778370</td>\n",
       "      <td>1.724259</td>\n",
       "      <td>-0.833936</td>\n",
       "      <td>0.288581</td>\n",
       "      <td>0.508910</td>\n",
       "      <td>-1.621097</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>-0.828971</td>\n",
       "      <td>1.629473</td>\n",
       "      <td>-0.833936</td>\n",
       "      <td>-0.559514</td>\n",
       "      <td>-0.721385</td>\n",
       "      <td>-0.340126</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>-0.733089</td>\n",
       "      <td>1.734330</td>\n",
       "      <td>-0.833936</td>\n",
       "      <td>-0.441065</td>\n",
       "      <td>-0.721385</td>\n",
       "      <td>-0.702398</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>-0.862345</td>\n",
       "      <td>1.776256</td>\n",
       "      <td>0.898905</td>\n",
       "      <td>2.745215</td>\n",
       "      <td>0.508910</td>\n",
       "      <td>0.707288</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>-0.815977</td>\n",
       "      <td>1.778724</td>\n",
       "      <td>0.898905</td>\n",
       "      <td>1.030072</td>\n",
       "      <td>-0.106237</td>\n",
       "      <td>0.266283</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>-0.732741</td>\n",
       "      <td>1.625957</td>\n",
       "      <td>-0.833936</td>\n",
       "      <td>-0.109408</td>\n",
       "      <td>-0.106237</td>\n",
       "      <td>-1.497484</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>-0.816202</td>\n",
       "      <td>1.696624</td>\n",
       "      <td>0.898905</td>\n",
       "      <td>1.454120</td>\n",
       "      <td>0.508910</td>\n",
       "      <td>0.482678</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>-0.819416</td>\n",
       "      <td>1.878189</td>\n",
       "      <td>2.631746</td>\n",
       "      <td>-0.526348</td>\n",
       "      <td>-0.106237</td>\n",
       "      <td>2.928767</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>-0.815007</td>\n",
       "      <td>1.699076</td>\n",
       "      <td>-0.833936</td>\n",
       "      <td>0.857137</td>\n",
       "      <td>0.508910</td>\n",
       "      <td>-0.965307</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>-0.812059</td>\n",
       "      <td>1.713099</td>\n",
       "      <td>0.898905</td>\n",
       "      <td>1.266971</td>\n",
       "      <td>-0.106237</td>\n",
       "      <td>0.438199</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    latitude  longitude     price  review_count    rating  price_adj\n",
       "0  -0.865918   1.797864  0.898905      2.401713 -0.106237   0.273958\n",
       "1  -0.860828   1.866601 -0.833936     -0.523979 -0.721385  -0.593050\n",
       "2  -0.865098   1.866229  0.898905      0.293319 -0.106237   1.440286\n",
       "3  -0.864164   1.873697 -0.833936     -0.377103 -1.336532  -0.419023\n",
       "4  -0.728723   1.624631  0.898905     -0.028862  0.508910   0.364028\n",
       "5  -0.871440   1.858230  0.898905     -0.388947  0.508910   1.608113\n",
       "6  -0.824289   1.600982 -0.833936     -0.400792 -0.721385  -0.414354\n",
       "7  -0.742374   1.737415  0.898905      0.025624  0.508910   1.172865\n",
       "8  -0.730867   1.710399 -0.833936     -0.542931 -2.566826  -0.340126\n",
       "9  -0.860725   1.790306  0.898905      2.039259 -0.106237   0.364028\n",
       "10 -0.778370   1.724259 -0.833936      0.288581  0.508910  -1.621097\n",
       "11 -0.828971   1.629473 -0.833936     -0.559514 -0.721385  -0.340126\n",
       "12 -0.733089   1.734330 -0.833936     -0.441065 -0.721385  -0.702398\n",
       "13 -0.862345   1.776256  0.898905      2.745215  0.508910   0.707288\n",
       "14 -0.815977   1.778724  0.898905      1.030072 -0.106237   0.266283\n",
       "15 -0.732741   1.625957 -0.833936     -0.109408 -0.106237  -1.497484\n",
       "16 -0.816202   1.696624  0.898905      1.454120  0.508910   0.482678\n",
       "17 -0.819416   1.878189  2.631746     -0.526348 -0.106237   2.928767\n",
       "18 -0.815007   1.699076 -0.833936      0.857137  0.508910  -0.965307\n",
       "19 -0.812059   1.713099  0.898905      1.266971 -0.106237   0.438199"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "preprocessor = Pipeline([\n",
    "    ('catpri', CategoryPricer()),\n",
    "    ('sacr', StandardizerAndCategoryReducer())\n",
    "])\n",
    "Z_train = preprocessor.fit_transform(X)\n",
    "Z_train.head(20)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Transfer learning: KMeans"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "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>price</th>\n",
       "      <th>review_count</th>\n",
       "      <th>rating</th>\n",
       "      <th>category</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>33.911154</td>\n",
       "      <td>-117.881259</td>\n",
       "      <td>2</td>\n",
       "      <td>1257.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>['bbq', 'smokehouse', 'southern']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>33.911959</td>\n",
       "      <td>-117.867374</td>\n",
       "      <td>1</td>\n",
       "      <td>22.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>['sandwiches']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>33.911283</td>\n",
       "      <td>-117.867449</td>\n",
       "      <td>2</td>\n",
       "      <td>367.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>['juicebars', 'acaibowls', 'icecream']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>33.911431</td>\n",
       "      <td>-117.865941</td>\n",
       "      <td>1</td>\n",
       "      <td>84.0</td>\n",
       "      <td>2.5</td>\n",
       "      <td>['sandwiches', 'hotdogs']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>33.932860</td>\n",
       "      <td>-117.916253</td>\n",
       "      <td>2</td>\n",
       "      <td>231.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>['italian']</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    latitude   longitude  price  review_count  rating  \\\n",
       "0  33.911154 -117.881259      2        1257.0     3.5   \n",
       "1  33.911959 -117.867374      1          22.0     3.0   \n",
       "2  33.911283 -117.867449      2         367.0     3.5   \n",
       "3  33.911431 -117.865941      1          84.0     2.5   \n",
       "4  33.932860 -117.916253      2         231.0     4.0   \n",
       "\n",
       "                                 category  \n",
       "0       ['bbq', 'smokehouse', 'southern']  \n",
       "1                          ['sandwiches']  \n",
       "2  ['juicebars', 'acaibowls', 'icecream']  \n",
       "3               ['sandwiches', 'hotdogs']  \n",
       "4                             ['italian']  "
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = df.drop(columns=['avg_agi', 'med_agi', \n",
    "                     'count', 'city', 'id', 'alias', 'zip_code'])\n",
    "y = df['med_agi']\n",
    "X.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [],
   "source": [
    "def inertia_silhouetter(df, features, somelist):\n",
    "    \n",
    "    scores = []\n",
    "\n",
    "    for k in somelist:\n",
    "    \n",
    "        cluster_data = df[features]\n",
    "\n",
    "        # instantiate KMeans\n",
    "        km = KMeans(n_clusters=k)\n",
    "        km.fit(cluster_data)\n",
    "\n",
    "        inertia = km.inertia_\n",
    "        sil = silhouette_score(location_data, km.labels_)\n",
    "        scores.append((k, inertia, sil))\n",
    "        \n",
    "    scores_df = pd.DataFrame(scores, columns=['k', 'inertia', 'silhouette'])\n",
    "        \n",
    "    fig, axes = plt.subplots(1, 2, figsize=(13, 8))\n",
    "    axes[0].plot(scores_df.k, scores_df.inertia)\n",
    "    axes[0].set_title('inertia vs. k')\n",
    "    axes[1].plot(scores_df.k, scores_df.silhouette);\n",
    "    axes[1].set_title('silhouette score vs. k')\n",
    "    \n",
    "    return scores_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "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>k</th>\n",
       "      <th>inertia</th>\n",
       "      <th>silhouette</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>1067.743127</td>\n",
       "      <td>0.394059</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>12</td>\n",
       "      <td>142.692225</td>\n",
       "      <td>0.429196</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>22</td>\n",
       "      <td>70.407543</td>\n",
       "      <td>0.421550</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>32</td>\n",
       "      <td>47.364154</td>\n",
       "      <td>0.413496</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>42</td>\n",
       "      <td>35.249139</td>\n",
       "      <td>0.413371</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>52</td>\n",
       "      <td>27.718518</td>\n",
       "      <td>0.416892</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>62</td>\n",
       "      <td>22.060894</td>\n",
       "      <td>0.428224</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>72</td>\n",
       "      <td>18.570411</td>\n",
       "      <td>0.424308</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>82</td>\n",
       "      <td>15.457919</td>\n",
       "      <td>0.443793</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>92</td>\n",
       "      <td>13.703331</td>\n",
       "      <td>0.439017</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>102</td>\n",
       "      <td>12.113171</td>\n",
       "      <td>0.438257</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      k      inertia  silhouette\n",
       "0     2  1067.743127    0.394059\n",
       "1    12   142.692225    0.429196\n",
       "2    22    70.407543    0.421550\n",
       "3    32    47.364154    0.413496\n",
       "4    42    35.249139    0.413371\n",
       "5    52    27.718518    0.416892\n",
       "6    62    22.060894    0.428224\n",
       "7    72    18.570411    0.424308\n",
       "8    82    15.457919    0.443793\n",
       "9    92    13.703331    0.439017\n",
       "10  102    12.113171    0.438257"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 936x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "inertia_silhouetter(X, ['latitude', 'longitude'], range(2, 103, 10))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "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>k</th>\n",
       "      <th>inertia</th>\n",
       "      <th>silhouette</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8</td>\n",
       "      <td>232.272010</td>\n",
       "      <td>0.405710</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>9</td>\n",
       "      <td>204.939568</td>\n",
       "      <td>0.420037</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>10</td>\n",
       "      <td>180.289107</td>\n",
       "      <td>0.437783</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>11</td>\n",
       "      <td>159.250030</td>\n",
       "      <td>0.431647</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>12</td>\n",
       "      <td>142.498311</td>\n",
       "      <td>0.431963</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>13</td>\n",
       "      <td>129.104179</td>\n",
       "      <td>0.407683</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>14</td>\n",
       "      <td>118.730938</td>\n",
       "      <td>0.425332</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    k     inertia  silhouette\n",
       "0   8  232.272010    0.405710\n",
       "1   9  204.939568    0.420037\n",
       "2  10  180.289107    0.437783\n",
       "3  11  159.250030    0.431647\n",
       "4  12  142.498311    0.431963\n",
       "5  13  129.104179    0.407683\n",
       "6  14  118.730938    0.425332"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAv0AAAHiCAYAAAB7rnPHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOzdd3gc1dn+8e+j3ottSbjIlovkXrBNDQFCNWBKAiHkTSHlF0IqScAB8oYEAgmhBN6QQhLSe4CQxJTQW+iRjG3c5aKVu2VrZcmS1c/vjx2ZxUiWZEuaLffnunR5d3Zm9l6Ddx/NnvMcc84hIiIiIiKxK8HvACIiIiIiMrhU9IuIiIiIxDgV/SIiIiIiMU5Fv4iIiIhIjFPRLyIiIiIS41T0i4iIiIjEOBX9ElHMbKWZnTqEz/dvM7t8qJ6vN2ZWZWZn+J1DRGQgmNk3zOyX3u0SM3NmluTdf97M/p+/CaWvzOy3ZnaL3znk8CX5HUAknHNu+mCd28xuBCY55z4a9nznDNbziYjEO+fc9/zOcDAzqwL+n3Puae9+CbAJSHbOtfuXTGRw6Uq/xIWuK0siIiLRRJ9fMlBU9EtECR/eYmY3mtn9ZvZ7M2vwhv7MD9t3lJn93cxqzGyTmX057LEbzexBM/ujmdUDVwLfAD5kZvvMbJm334Gvl81sopk9a2Z7zGy3mf3JzPJ6yPkzM7vzoG3/MrOvebevNbOtXu61Znb6YfxdTPFe12X9PVZEZCj19J7nvRf/8RCHjjOzl73jnjSzEWHnvMB736/z3qunhj3mzGxS2P13DD0xs4VmttQ79hUzm+Vt/wMwFnjY+yz4OvCid1idt+0Eb99PmdlqMwua2RNmNq6H157mfdbs8Z7vv2ZW5D02zMx+Y2bbvPP8M+y4z5jZejOrNbPFZjbqoNf3BTOrBCq9bVPM7Clv/7VmdmkPeS4zs/KDtn3VzBZ7t881s1Xe3/lWM7vmEP99umVm2Wb2nJndY2bW3+PFHyr6JdJdAPwVyAMWAz8GMLME4GFgGTAaOB34ipmdHXbshcCD3rG/Ar4H/M05l+Wcm93NcxlwKzAKmAoUAzf2kOvPhH6BMC9PPnAW8Fczmwx8ETjGOZcNnA1U9edFm9lc4EngS865v/bnWBGRoXSE73n/A3wSKARSgGu8c5YBfwG+AhQAjxEq1FP6kGcu8Gvgs8Bw4OfAYjNLdc59DKgGzvc+C24HTvYOzfO2vWpmFxG6UPQB7/n/4+XpzuVALqHPjOGELjLt9x77A5ABTPde491extMIfd5cCowEAoQ+68JdBBwHTDOzTOApQp89hcCHgZ+aWXdDYhcDk82sNGzb/3jHQujz8LPef6sZwLM9vK5umdlw4BngZefcl51zrj/Hi39U9Euke8k595hzroPQm2dXsX4MUOCc+45zrtU5txG4Dwi/Kv6qc+6fzrlO59x+euGcW++ce8o51+KcqwHuAk7pYff/AA54r3f/Eu/5tgEdQCqhN+pk51yVc25DP17zewm9aV/unHukH8eJiPjhSN7zfuOcW+e9R98PzPG2fwh41HtPbgPuBNKBE/twzs8AP3fOve6c63DO/Q5oAY7vx2v6LHCrc261N87/e8CcHq72txEq9id5z1fhnKs3s5HAOcCVzrmgc67NOfeCd8xHgF8755Y451qA64ETLDS/oMutzrla7+9mIVDlnPuNc67dObcE+Duhz553cM41Af8i9IsBXvE/hdDnSlfeaWaW4+Va0o+/l1HAC8ADzrlv9uM4iQAq+iXS7Qi73QSkWWh84zhglPdVap2Z1RG6KlMUtv/m/jyRmRWa2V+9rzvrgT8CI7rb17uy8Ve8N1VCV1H+5D22ntDVqRuBXd45R3V3nh5cCbzinHuuP/lFRPxwhO95B7/HZ3m3RxG6+t31HJ2E3tNH9+Gc44CrD/p8KPbO2VfjgB+GHV9L6Nvg7p7/D8AThL7p3WZmt5tZsvectc65YDfHHPz69gF7Djp/+GfYOOC4g17TR4Cjesj/Z975+fRP75cBgIuBc4GAmb3QNZypj84j9MvXz/pxjEQIFf0SrTYDm5xzeWE/2c65c8P2Ofgrx96+grzV22eWcy4H+CihN/me/AW4xLvycxyhqy6hJ3Luz865kwi9UTvgtj69qpArgbFmdnc/jhER8c0Rvud1Z5t3LgC8oZTFwFZvUxOhYTNdwovfzcB3D/p8yHDOdQ3P6ctnw2ZCQ2DCz5HunHvl4B29K/g3OeemEfomYiHwce8cw6z7uWEHv75MQt8WbA3bJzzXZuCFg/JkOec+1825ITQ8dISZzSFU/HcN7cE591/n3IWEhgn9k9A3LH11H/A48JiXWaKIin6JVm8A9RaaPJZuZolmNsPMjjnEMTuBEm8+QHeygX2EJnONBhYdKoBz7k2gBvgl8IRzrg5C41vN7DQzSwWaCY3t7OjHa2sAFgAnm9n3+3GciMiQG4D3vO7cD5xnZqd7V82vJjREp6voXgr8j/fev4B3DsW8D7jSzI6zkEwzO8/Msr3HdwITwvavAToP2vYz4PquMfNmlmtmH+wuqJm9z8xmmlkiUE9o+EyHc2478G9CY+/zzSzZzLrmD/wZ+KSZzfH+3r4HvO6cq+rh7+MRoMzMPuadJ9nMjrGwyc3hvCFJDwJ3AMMIzQfAzFLM7CNmlusNm6qn//+tvgisBR4xs/R+His+UtEvUckb438+ofGfm4DdhIrv3EMc9oD35x4z624M403AXGAv8CjwUB+i/AU4g7CrKITGtn7fy7SD0NWUbwB4b7Yrezup9wvEmcA5ZnZzH3KIiPilx/e8w+WcW0vo29Yfeec9n9Dk21Zvl6u8bV3DXP4Zdmw5oXH9PwaCwHrgE2GnvxX4pjdM5hpv2Mt3gZe9bcc75/5B6NuKv3rDPVcQGp/fnaMIFdj1wGpCY967OhZ9jNAvAWuAXYSGQeGcewa4gdA3xNuBibxzTtrBfx8NhJpFXEboW4IdXr7Uno4h9Ll0BqHx9+HrD3wMqLK3O9t9FMDMxlqoe9HYQ5yza3jrFYS+ffiXmaUdan+JHKZJ1yIiIiIisU1X+kVEREREYpyKfhERERGRGKeiX0REREQkxqnoFxERERGJcSr6RURERERiXJLfAQBGjBjhSkpK/I4hIhKxKioqdjvnCvzO4Td9XoiIHFpPnxcRUfSXlJRQXl7udwwRkYhlZgG/M0QCfV6IiBxaT58XGt4jIiIiIhLjVPSLiIiIiMQ4Ff0iIiIiIjFORb+IiIiISIxT0S8iIiIiEuNU9IuIiIiIxDgV/SIiIiIiMU5Fv4iIiIhIjFPRLyIiIiIS41T0i4iIiIjEOBX9IiIiIiIxTkW/iIiIiEiMU9EvIiIiIhLjVPSLiIiIiMQ4Ff0iIiIiIjFORb+IiIiISIyL6qLfOcfm2ia/Y4iIiEg/dHY6vyOIxJ2oLvr/958ruPjeV9jf2uF3FBEREemDjk7HyXc8x+2Pr/E7ikhcieqi/6I5o9nV0MLvXq3yO4qIiIj0wbqdDWwJ7ufeFzZQEQj6HUckbkR10X/s+GGcOrmAe5/fQH1zm99xREREpBflXqGfn5HCogeX0dymb+tFhkJUF/0A15w1mb3727jvxY1+RxEREZFeLAkEKchO5YeXzWFjTSN3P73O70gicSHqi/4Zo3M5b9ZIfvXSJnbva/E7joiIiBxCRSDI/HH5vLe0gMuOKea+FzeydHOd37FEYl7UF/0AV59ZRkt7Jz99boPfUURERKQHuxqaqa5tYt64fAC+cd5UinLSWPTAMlraNcxHZDDFRNE/oSCLS+aO4Y+vBdhat9/vOCIiItKNJd54/rle0Z+Tlsz3PjCTyl37+NEz6/2MJhLzYqLoB/jyGaUA3PN0pc9JREREpDvlVUFSkhKYMSr3wLb3TS7k4rljuPeFDazYutfHdCKxLWaK/tF56Xz0+HE8uGQLG2r2+R1HREREDlJRHWT2mFxSkt5Zfnxr4TSGZ6ZwzQPLaG3v9CmdSGyLmaIf4PPvm0hqUgJ3PaVOACIiIpGkua2DFVv3Mm/csHc9lpuRzHffP5M1Oxr46fMa5iMyGGKq6B+RlcqnTxrPo8u36ytCERGRCPLW1r20dbgDk3gPdua0Ii6cM4ofP7ue1dvrhzidSOyLqaIf4DMnTyA3PZk7n1zrdxQRERHxlFeFJvH2VPQD3Hj+dPIykln04DLaOjTMR2QgxVzRn5OWzOdOncjza2t4Y1Ot33FERESEUH/+CSMyGZaZ0uM++Zkp3HzhDFZsrecXWnRTZEDFXNEPcPkJJRRmp3LHE2twzvkdR0REJK4551hSHTzkVf4u58wcyXkzR/LDpytZt7NhCNKJxIeYLPrTUxL50uml/LcqyPPravyOIyIiEtc27W6ktrG1T0U/wE0XTicrLYlFDy6nXcN8RAZETBb9AB+aX0zxsHTufGItnZ262i8iIuKXcm9RrvklfSv6R2SlctMF01m2uY5fvrRpMKOJxI2YLfpTkhL42pllrNxWz79X7PA7joiISNxaEgiSm57MhBFZfT5m4ayRnD29iLueWsf6XVp/R+RIxWzRD3DB7NGUFWXxg6fW6utBERERn5QHQuP5ExKsz8eYGTdfNIOMlES+/uAyOvStvcgRiemiPzHBuPqsyWysaeShJVv9jiMiIhJ36ppaWb9rX5/H84crzE7j2+dPY0l1Hb95WcN8RI5ETBf9AGdNK2J2cR7/9/Q6mts6/I4jIiISV96srgMO3Z//UC6aM5rTpxRy55NrqdrdOJDRROJKzBf9ZsbXz57Mtr3N/Pn1ar/jiIiIxJXyQC2JCcbsMXmHdbyZ8d33zyQ5MYGv/325mnOIHKaYL/oB3jNpBCdOHM5PnltPY0u733FERETiRkUgyPRROaSnJB72OY7KTeOGhdN4Y1Mtf3gtMIDpROJHXBT9AIvOnsyexlZ+rdZfIiIiQ6Kto5Olm+sOe2hPuA/OG8PJZQXc9vgaNtc2DUA6kfgSN0X/0WPzOXNaEb94cSN1Ta1+xxEREYl5q7fX09zWOSBFv5nx/Q/MJMGMa/++HOc0zEekP+Km6Ae45qzJ7Gtt52cvbPQ7ioiISMwrrwotyjUQRT/AqLx0vnHuVF7ZsIc/v6F5eiL9EVdF/+Sjsrlozmh++8omdtU3+x1HREQkplVUBxmdl87I3PQBO+eHjy3mPZOGc+tja9hat3/AzisS6+Kq6Af4yhmltHc4fvTser+jiIiIxCznHBVVwQG7yt8lNMxnFp3OcZ2G+Yj0WdwV/eOGZ3LZscX85Y1qqvdoIpCIiMhg2La3mR31zQNe9AMUD8vg+nOm8J/K3dxfvnnAzy8Si+Ku6Af40mmlJCYY//f0Or+jiIiIxKTyqlpg4MbzH+wjx43j+AnDuOWR1Wzfq2E+Ir2Jy6K/KCeNT5xYwj+WbmXdzga/44iIiMScJYEgGSmJTDkqe1DOn5Bg3HbxLNo7Hd946C0N8xHpRVwW/QBXnjKRrJQk7nxird9RREREYk55IMjRY/NIShy8UmPc8EwWnT2Z59bW8NCSrYP2PCKxIG6L/vzMFD5z8gSeXLWTpZvr/I4jIhLRzGyBma01s/Vmdt0h9rvEzJyZzffuH2tmS72fZWb2/rB9q8zsLe+x8qF4HTI0GlvaWb29nnljB2doT7hPnFjC/HH53PTwSnXmEzmEuC36AT510niGZ6boar+IyCGYWSLwE+AcYBrwYTOb1s1+2cCXgdfDNq8A5jvn5gALgJ+bWVLY4+9zzs1xzs0ftBcgQ27p5jo6HcwrGTboz5WQYNx+ySxa2jv533+u0DAfkR7EddGflZrE5983iZfW7+aV9bv9jiMiEqmOBdY75zY651qBvwIXdrPfzcDtwIHLrc65Judcu3c3DVBFFgcqAkHM4OixeUPyfBMKsrj6rDKeWrWTxcu2DclzikSbuC76AT5y3FhG5aZx+xNrdXVARKR7o4HwvohbvG0HmNnRQLFz7pGDDzaz48xsJfAWcGXYLwEOeNLMKszsisGJLn6oCASZXJRNTlrykD3np0+awJziPG5cvJKahpYhe16RaBH3RX9aciJXnVHK0s11PLVqp99xREQikXWz7cBVEjNLAO4Gru7uYOfc68656cAxwPVmluY99B7n3FxCw4a+YGYnd/vkZleYWbmZldfU1BzJ65Ah0NnpWFIdZO4gtersSWKCcccls2hs6eDbi1cM6XOLRIO4L/oBLp47hvEjMvnBk+vo6NTVfhGRg2wBisPujwHCx1BkAzOA582sCjgeWNw1mbeLc2410Ojti3Num/fnLuAfhIYRvYtz7hfOufnOufkFBQUD8oJk8FTu2kdDc/uQTOI9WGlRNledUcpjb+3gsbe2D/nzi0QyFf1AUmICXzuzjLU7G3hYYwFFRA72X6DUzMabWQpwGbC460Hn3F7n3AjnXIlzrgR4DbjAOVfuHZMEYGbjgMlAlZllehN/MbNM4CxCk34lypUHQotyzS8Z+qIf4LMnT2Dm6Fxu+OcKahtbfckgEolU9HvOmzmSaSNzuOupdbS2d/odR0QkYnhj8L8IPAGsBu53zq00s++Y2QW9HH4SsMzMlhK6mv9559xuoAh4ycyWAW8AjzrnHh+8VyFDpSIQZERWCmOHZfjy/EmJCdzxwVnUN7fx7cUrfckgEomSet8lPiQkGIvOnswnf/tf7i/fzEePH+d3JBGRiOGcewx47KBt3+ph31PDbv8B+EM3+2wEZg9sSokEFYEg88blY9bdVJChMeWoHL50Wil3PbWOhbNGcvb0o3zLIhIper3Sb2bFZvacma02s5VmdpW3/Q4zW2Nmy83sH2aWF3bM9d4CLmvN7OzBfAED6dTJBRxTks89z1Syv7XD7zgiIiJRpaahhcCeJuYN8STe7nzu1IlMG5nD//5jBXVNGuYj0pfhPe3A1c65qYQmZ33BW5TlKWCGc24WsA64HsB77DJgOqGFWH7qLewS8cyMRWdPYVdDC79/tcrvOCIiIlGlIhAEYN64wV+UqzfJ3jCfuqZWvvPwKr/jiPiu16LfObfdObfEu91AaDznaOfck2G9ll8j1M0BQgu2/NU51+Kc2wSsp4eODJHo2PHDOKWsgHtf2EB9c5vfcURERKLGkuogKUkJzBid43cUAKaPyuXzp07koTe38uwateWW+NavibxmVgIczTuXWAf4FPBv73avi7h454rYvsuLzp5MXVMbv3xxo99RREREokZ5VS2zRueSmhQ5X/B/8bRSJhdlc/1Db7F3vy7mSfzqc9FvZlnA34GvOOfqw7b/L6EhQH/q2tTN4e9qfh/JfZdnjM7lvJkj+eVLm9i9T6v6iYiI9Ka5rYMVW+sjYjx/uJSk0DCf3fta+e6jGuYj8atPRb+ZJRMq+P/knHsobPvlwELgI865rsK+t0VcosJXzyyjua2De5/f4HcUERGRiLdi615aOzojrugHmDUmjytOnsD95Vt4YV1kjS4QGSp96d5jwK+A1c65u8K2LwCuJbQAS1PYIYuBy8ws1czGA6WEejBHlUmFWVwybwx/eC3Atrr9fscRERGJaF2TeOdGYNEPcNXppUwqzOL6vy+nQXP2JA715Ur/e4CPAaeZ2VLv51zgx4SWXn/K2/YzAOfcSuB+YBXwOPAF51xU9r+86owycHDPM5V+RxEREYlo5YEg40dkMiIr1e8o3UpLTuT2S2axo76ZW/+9xu84IkOu18W5nHMv0f04/ce62dZ1zHeB7x5BrogwOi+djxw/lt+/GuCKkycwoSDL70giIiIRxznHkkCQUycX+h3lkOaOzefTJ43nvv9sYuHMkZw4aYTfkUSGTL+698Sjz586idSkBO56ap3fUURERCJS1Z4m9jS2Mr8kMof2hLv6rMmMH5HJ1/++nMaW9t4PEIkRKvp7UZCdyqfeM55Hlm9n5ba9fscRERGJOG8vyhX5RX/XMJ+tdfu57XEN85H4oaK/Dz5z8gRy05O584m1fkcRERGJOBWBIDlpSUyKkmGwx5QM4xMnlvD7VwO8tnGP33FEhoSK/j7ITU/mylMm8tzaGv5bVet3HBERkYhSEahl7rh8EhK6mwIYmRadPZmxwzK49u/L2d8alf1GRPpFRX8ffeLEEgqyU7nj8bW8vSSBiIhIfNu7v411O/cxb2zkD+0Jl5GSxG0XzyKwp4k79E2+xAEV/X2UnpLIl0+bxBtVtVrYQ0RExLOk2hvPHwWTeA92wsThfOz4cfzmlU2U65t8iXEq+vvhQ8eMpXhYOnc8sZbOTl3tFxERWRIIkphgzCnO8zvKYbnunCmMyk3n6w8up7lNw3wkdqno74eUpAS+ekYZK7fV8+8VO/yOIyIi4rvyqiDTRuaQkdLr0j8RKTM1NMxn4+5G7lZ7bolhKvr76cI5oyktzOIHT62lvaPT7zgiIiK+ae/oZOnmuqho1XkoJ5WO4MPHFnPffzbypjdcSSTWqOjvp8QE4+qzJrOxppGH3tzqdxwRERHfrN7ewP62jqgv+gGuP3cqRTlpfP3B5bS0a5iPxB4V/Yfh7OlFzB6Tyw+frtQbg4iIxK2KQGjyaywU/Tlpydz6gZlU7trHPc9U+h1HZMCp6D8MZsais6ewtW4/f3692u84IiIivigPBBmVm8aovHS/owyIUycXcsm8MfzshY28tWWv33FEBpSK/sP0nknDOWHCcH7y3HoaW9r9jiMiIjLklgSCzI2Bq/zhbjhvGsMzU1j04DJa2zV3T2KHiv7DZGYsWjCZ3fta+c3Lm/yOIyIiMqS21e1n295m5sdY0Z+bkcz33j+TNTsa+PFz6/2OIzJgVPQfgblj8zljahE/f3EjdU2tfscREREZMhUBb1GuccN8TjLwzphWxPuPHs1Pn1vPym0a5iOxQUX/Ebrm7DL2tbTzsxc2+h1FRERkyFQEgqQnJzJ1ZLbfUQbFt8+fRl5GCoseWE6bWnRLDFDRf4SmHJXDhbNH8dtXNrGrvtnvOCIiIkOiIhBkTnEeSYmxWUrkZaRwy0UzWLW9np89v8HvOCJHLDb/pQ6xr55ZRnuH09g/ERGJC02t7azaXh8TrToPZcGMo1g4ayT3PFvJ2h0NfscROSIq+gfAuOGZfOiYYv7yRjWba5v8jiMiIjKolm6uo6PTMa8ktot+gJsumE5OWjKLHlxGu4b5SBRT0T9AvnRaKQlm3P30Or+jiIiIDKol3iTeucWxX/QPz0rlpguns3zLXu77j7r1SfRS0T9AjspN4/ITS/jHm1tZt1NfAYqISOwqDwQpK8oiNyPZ7yhD4ryZI1kw/Sjufnod63ft8zuOyGFR0T+APnfKRDJTkvjBk2v9jiIiIjIoOjsdSwLBmB/PH87MuPmiGWSkJLLowWV0dDq/I4n0m4r+AZSfmcJn3juBJ1buZNnmOr/jiIiIDLj1Nfuob26Pyf78h1KQncqN50/nzeq6qF+Us7PTEWxsZUPNPsqranlu7S72t3b4HUsGWZLfAWLNp987nt+9WsUdT6zlj//vOL/jiIiIDKi3F+WKnyv9XS6cM4pHlm/jjifWcvrUIsaPyPQ7Es456pvbCTa2UtvUSl1TK7WNbQfuBxtbqW1sJdjU9WcbdU2tHPxlxTfPm8r/e+8Ef16EDAkV/QMsKzWJz586kVseXc0r63dz4qQRfkcSEREZMOVVQYZnplAyPMPvKEPOzPju+2dy5l0v8PUHl/G3K04gIcEG7PzOOZpaOw4q0g9VxIcK+PYehhslJxr5GSkMy0whPyOFKUflkJ+ZzLCMFPIz395+zQPLWLWtfsBeh0QmFf2D4KPHj+NXL23ijifX8tDE4ZgN3BuCiIiIn5ZUB5k7Lj9uP9uKctL41vnTueaBZfzu1So++Z7xPe7b3BYq4N9RxDe2UtsUKuKDTe8u6lvbu28Lmphg5Gckk+8V7BNGZDFvXArDMpPfLuwzU0IFfUYK+ZnJZKUm9em/0+Sjslm3S01IYp2K/kGQlpzIVaeXct1Db/H06l2cOa3I70giIiJHbPe+FjbtbuSyY4r9juKri+eO5pHl27j98bXUNbWxd3/bO67MBxvbqG1sZX9bz+Pk8zLevuI+Oi+dmaNz3i7aw//0bmenJQ3otwrhyoqy+dPrATo73aA9h/hPRf8guXjeGH7+4kbufGItp08p1D8iERGJekvieDx/ODPj1g/M5KKfvMwPn6kkOy3pwFCZwuw0yoqy3zWEZljm21flc9OTSUqMnF4qZUVZNLd1siW4n7FxOGwrXqjoHyTJiQl89cwyvvyXN3l4+TYunDPa70giIiJHpCIQJCUxgRmjc/2O4ruRuem8dO1pOAcpSZFTwB+O0qJsANbtbFDRH8Oi+//SCLdw5kimjszhrqfW0aalu0VEJMpVBILMGJ1DWnKi31EiQnJiQtQX/ACTCrMAqNTCYzEt+v9PjWAJCcais8sI7Gni/vLNfscRERE5bC3tHSzfupf5JfHVnz8e5KQlMzI3jcqdmswby1T0D7L3TS5k/rh87nmmkuZDTOgRERGJZCu21tPa3sncsfE9nj9WlRapg0+sU9E/yMyMRWdPZmd9C79/tcrvOCIiIodFk3hjW1lhFut37aOzh57/Ev1U9A+B4yYM5+SyAn76/AYamtv8jiMiItJv5YFaxg3PoCA71e8oMghKwzr4SGxS0T9EFp01mbqmNu77zya/o4iIiPSLc46KQB3zNLQnZoV38JHYpKJ/iMwck8u5M4/iV//ZyJ59LX7HERER6bPq2iZ272thXomK/lhV6nXw0bj+2KWifwh97cwy9rd1cO/zG/yOIiIi0mcVGs8f87IPdPBR285YpaJ/CE0qzObiuWP4/WsBttVpzJyIiESH8kCQ7NQkygqz/Y4ig6i0KJtKXemPWSr6h9hVZ5SCgx89W+l3FBERkT5ZEghy9Lh8EhLM7ygyiNTBJ7ap6B9iY/Iz+J/jxnJ/+RY21ugrNBERiWx797exdmcD8zW0J+aVFWXT3NbJ5mCT31FkEKjo98EX3jeJlMQE7n5aV/tFRCSyLd1ch3Mazx8PJhV5k3k1rj8mqej3QUF2Kp86qYSHl21j1bZ6v+OIiIj0qKKqlgSDOcV5fkeRQdbVwYPjoUsAACAASURBVEfj+mOTin6fXHHyRHLSkrjzybV+RxEREelRRXWQqSNzyExN8juKDLLstGRGqYNPzFLR75Pc9GSuPHUiz67ZRXlVrd9xRERE3qW9o5M3q+s0nj+OlBZla4GuGKWi30efOLGEEVmp3P7EWpzTTHkRiVxmtsDM1prZejO77hD7XWJmzszme/ePNbOl3s8yM3t/f88p/lmzo4Gm1g7mquiPG6VeB58OdfCJOSr6fZSRksSXT5/EG5tqebFyt99xRES6ZWaJwE+Ac4BpwIfNbFo3+2UDXwZeD9u8ApjvnJsDLAB+bmZJfT2n+KtrUa75JcN8TiJDpawom5b2Traog0/MUdHvs8uOGcuY/HRuf3yN+uKKSKQ6FljvnNvonGsF/gpc2M1+NwO3A81dG5xzTc65du9uGtD1RtfXc4qPKgJBjspJY1Rumt9RZIiUqoNPzFLR77OUpAQWnT2ZldvqeaBis99xRES6MxoIf4Pa4m07wMyOBoqdc48cfLCZHWdmK4G3gCu9XwJ6Paf4ryIQZN64fMy0KFe8KC0Krbqscf2xR0V/BLhg9ijmjcvnjifWUt/c5nccEZGDdVfxHfhq0swSgLuBq7s72Dn3unNuOnAMcL2ZpfV2znc8udkVZlZuZuU1NTX9Di+HZ8feZrbW7Vd//jiTlZrEqNw01u/Slf5Yo6I/ApgZN54/nT2NrdyjBbtEJPJsAYrD7o8BtoXdzwZmAM+bWRVwPLC4azJvF+fcaqDR27e3c4Yf9wvn3Hzn3PyCgoIjfCnSV13j+VX0xx918IlNKvojxMwxuVw6r5jfvlKl365FJNL8Fyg1s/FmlgJcBizuetA5t9c5N8I5V+KcKwFeAy5wzpV7xyQBmNk4YDJQ1ds5xX/lgVrSkhOYNirH7ygyxMqK1MEnFqnojyCLFkwmPTmRmx9ZpRaeIhIxvDH4XwSeAFYD9zvnVprZd8zsgl4OPwlYZmZLgX8An3fO7e7pnIP3KqS/lgSCzB6TR3KiSoV4U1oY6uCzuVYdfGKJlteLICOyUrnqjFJueXQ1z67ZxelTi/yOJCICgHPuMeCxg7Z9q4d9Tw27/QfgD309p0SG/a0drNxWz2dPmeB3FPFBVwefyl37KBmR6XMaGSj69T3CfPyEEiYUZHLzI6toae/wO46IiMShZVvqaO90Gs8fp9TBJzap6I8wKUkJfGvhNKr2NPGbl6v8jiMiInGoaxLv3LEq+uNRVmoSo/PSqVTRH1NU9EegUycXcvqUQn70TCW76pt7P0BERGQAVQSCTCrMIi8jxe8o4pNJhVlaoCvGqOiPUN9cOI3Wjk5ue3yt31FERCSOdHY6KgJB5mtoT1wrK8piQ406+MQSFf0RavyITD510nj+vmQLb1YH/Y4jIiJxYuPufezd38ZcFf1xrbRIHXxijYr+CPal00opyE7lxodX0anftEVEZAiUV4UuNOlKf3wr02TemKOiP4JlpSZx7YIpLNtcx0NvbvU7joiIxIGKQJBhmSmMV6vGuDap8O22nRIbei36zazYzJ4zs9VmttLMrvK2DzOzp8ys0vsz39tuZnaPma03s+VmNnewX0Qs+8DRo5ldnMdtj69hX0u733FERCTGVQSCzB2bj5n5HUV8pA4+sacvV/rbgaudc1OB44EvmNk04DrgGedcKfCMdx/gHKDU+7kCuHfAU8eRhATjxvOnUdPQwo+erfQ7joiIxLDaxlY27m5Uf34BQot0qYNP7Oi16HfObXfOLfFuNxBaLn00cCHwO2+33wEXebcvBH7vQl4D8sxs5IAnjyNHj83n4rlj+PVLm9i0u9HvOCIiEqO6+vOr6BcIjetXB5/Y0a8x/WZWAhwNvA4UOee2Q+gXA6DQ2200sDnssC3eNjkC1y6YTEpiArc8ssrvKCIiEqMqAkGSE41ZY3L9jiIRYFJhljr4xJA+F/1mlgX8HfiKc67+ULt2s+1dvyKa2RVmVm5m5TU1NX2NEbcKc9L40umlPLNmF8+v3eV3HBERiUFLAkGmj8olLTnR7ygSAdTBJ7b0qeg3s2RCBf+fnHMPeZt3dg3b8f7sqkS3AMVhh48Bth18TufcL5xz851z8wsKCg43f1z55HtKKBmewXceWUVre6ffcUREJIa0tneybEudWnXKAergE1v60r3HgF8Bq51zd4U9tBi43Lt9OfCvsO0f97r4HA/s7RoGJEcmNSmRb50/jY01jfz+1Sq/44iISAxZuW0vLe2dGs8vB3R18NGV/tjQlyv97wE+BpxmZku9n3OB7wNnmlklcKZ3H+AxYCOwHrgP+PzAx45fp00p4tTJBfzw6UpqGlr8jiMiIjFCk3ilO6VFWVSqg09MSOptB+fcS3Q/Th/g9G72d8AXjjCXHMINC6dx9t0vcucTa7ntkll+xxERkRhQEQhSPCydwpw0v6NIBCkryubVDXvo6HQkJmjthmimFXmj0MSCLD75nhLur9jMW1v2+h1HRESinHOO8kCQ+eOG+R1FIkyp18GnWh18op6K/ij1pdNLGZ6Zwo0PryT05YqIiMjh2RLcT01DC3M1tEcOUqoOPjFDRX+UyklL5utnT6EiEORfS9/VHElERKTPygO1AOrcI+9S6nXwWa8OPlFPRX8Uu2TeGGaNyeXWf6+msaXd7zgiIhKlKgJBslOTDvRlF+mSqQ4+MUNFfxRLSDC+ff50dta38NPn1/sdR0REolR5VZA5Y/M0UVO6VVaUxTp18Il6Kvqj3Lxx+bz/6NHc959NVO/RJBsREemfhuY21u5sUKtO6VFpUTYbavbR0ak5hNFMRX8MuO6cKSQlGLc8usrvKCIiEmXerK7DOdS5R3pUWphFqzr4RD0V/TGgKCeNL7xvEk+u2slLlbv9jiMiIlGkIhAkwWB2ca7fUSRClamDT0xQ0R8jPn3SeMYOy+Cmh1fS1tHpdxwREYkSFYEgk4/KITst2e8oEqEmeR18KlX0RzUV/TEiLTmRb543lcpd+/jjawG/44iISBTo6HS8WR1Uq045pLc7+GgybzRT0R9DzpxWxHtLR3D3U+vYs6/F7zgiIhLh1u5ooLG1Q5N4pVdlRVlUqld/VFPRH0PMjG8tnEZjawc/eGqd33FERCTCVXiLcqnol96UqYNP1FPRH2NKi7L5+Anj+Msb1azcttfvOCIiEsEqAkEKs1MZk5/udxSJcJO8Dj6BPY1+R5HDpKI/Bn3ljDLyM1K4afEqnNNv5CIi0r3yQJD5JfmYaVEuObSuDj4a4hO9VPTHoNz0ZK45azJvVNXyyPLtfscREZEItLO+mS3B/cwdq6E90jt18Il+Kvpj1IeOKWb6qBxufWw1+1s7/I4jIiIRpiIQBGB+iRblkt5lpiYxJl8dfKKZiv4YlZhgfPv86Wzb28y9L2zwO46IiESYikCQ1KQEpo3M8TuKRInSwiwt0BXFVPTHsGPHD+P82aP4+Qsb2BLU0tkiIvK28kCQ2cV5pCSpFJC+KSvKZuPuRtq1CGhU0r/0GHf9OVMwg+89ttrvKCIiEiGa2zpYuXWvWnVKv5QWZdPa3kl1rS4kRiMV/TFuVF46nz91Eo+9tYNXNuz2O46IiESAZZvraO90WolX+qWsKDSZV+P6o5OK/jhwxckTGJOfznceXqWv5EREhIrq0CRede6R/phYoA4+0UxFfxxIS07kf8+dypodDfzljWq/44iIiM8qqoJMLMgkPzPF7ygSRbo6+KhXf3RS0R8nFsw4ihMmDOcHT62jrqnV7zgiIuIT5xwV1UGN55fDUlaUrQ4+UUpFf5wwM759wTTq97dx11Pr/I4jIiI+2VDTSF1Tm4p+OSylRVlsrFEHn2ikoj+OTDkqh48eP44/vhZgzY56v+OIiIgPlniLcs0bp0W5pP9KC7Np7egkoA4+UUdFf5z52pll5KQnc9PiVTjn/I4jIiJDrCIQJC8jmQkjMv2OIlGoq4NPpTr4RB0V/XEmLyOFq88s49WNe3h8xQ6/44iIyBArD9Qyb2w+CQnmdxSJQpMK1cEnWqnoj0MfPnYsU47K5pZHV9Pc1uF3HBERGSLBxlY21DQyV+P55TBlpIQ6+KxTB5+oo6I/DiUlJvDt86eztW4/v3hxo99xRERkiCzx+vNrUS45EmVF2brSH4VU9MepEyYO59yZR/HT59ezrW6/33FERGQIVASCJCUYs8bk+R1Fopg6+EQnFf1x7BvnTsU5uPXfa/yOIiIiQ6A8EGT66FzSUxL9jiJRrEwdfKKSiv44NiY/g8+eMpGHl23jjU21fscREZFB1NbRybLNdcwbq6E9cmRKizSZNxqp6I9znztlIqNy07hx8Uo6OtXCU0QkVq3cVk9LeyfzS1T0y5F5u4OPJvNGExX9cS49JZHrz53Kqu31/O2/m/2OIyIig6TiwKJcKvrlyGSkJFE8TB18oo2KfmHhrJEcO34Ydz65lr1NbX7HERGRQVARqGVMfjpFOWl+R5EYUFaoDj7RRkW/YGZ8+/xp1DW18n/PrPM7joiIDDDnHBWBoK7yy4CZpA4+g6K5rYOv/W3pgfa6A0lFvwAwfVQulx07lt+/GtBv7iIiMWZLcD8761vUn18GjDr4DI7lW/by0Jtb2bOvdcDPraJfDrjmrMlkpiTynUdW4Zwm9YrI28xsgZmtNbP1ZnbdIfa7xMycmc337p9pZhVm9pb352lh+z7vnXOp91M4FK8lHnVdNdRKvDJQyoqyAXXwGWiDOfdGRb8cMCwzha+eWcZ/Knfz1KqdfscRkQhhZonAT4BzgGnAh81sWjf7ZQNfBl4P27wbON85NxO4HPjDQYd9xDk3x/vZNSgvQCivCpKZkshkr1ATOVITCzMBWKcOPgOqIlDLhIJMhmWmDPi5VfTLO3z0+HGUFmZxy6OraW7r8DuOiESGY4H1zrmNzrlW4K/Ahd3sdzNwO9DctcE596Zzbpt3dyWQZmapgx1Y3qkiEOTosfkkJepjXwbGgQ4+utI/YLrm3gzWMDz965d3SE5M4NvnT6e6tolfvbTJ7zgiEhlGA+E9fbd42w4ws6OBYufcI4c4z8XAm865lrBtv/GG9txgZtbdQWZ2hZmVm1l5TU3NYb6E+LWvpZ01O+o1tEcGXFlhNuvVtnPAbKhpJNjUNmgT7lX0y7ucVDqCs6YV8ZPn1rNjb3PvB4hIrOuuGD8w8cfMEoC7gat7PIHZdOA24LNhmz/iDft5r/fzse6Odc79wjk33zk3v6Cg4DDix7el1XV0OjSJVwZcaVG2OvgMoIpALQDzxg0blPOr6JduffO8abR3Om57fI3fUUTEf1uA4rD7Y4BtYfezgRnA82ZWBRwPLA6bzDsG+Afwcefchq6DnHNbvT8bgD8TGkYkA6wiEMQM5ozN8zuKxJjSwixaOzqp2qMOPgOhIhAkPyOZiQWZg3J+Ff3SrbHDM/jMe8fzjze3HphJLiJx679AqZmNN7MU4DJgcdeDzrm9zrkRzrkS51wJ8BpwgXOu3MzygEeB651zL3cdY2ZJZjbCu50MLARWDN1Lih/lgVomF2WTk5bsdxSJMergM7DKvbU0ehjpeMRU9EuPPn/qJIpyUrnp4ZV0dqqFp0i8cs61A18EngBWA/c751aa2XfM7IJeDv8iMAm44aDWnKnAE2a2HFgKbAXuG7xXEZ86Oh1Lq+u0KJcMikmFWZhBpcb1H7HaxlY21jQO2tAegKRBO7NEvczUJK4/Zypf+dtSHqzYwqXHFPd+kIjEJOfcY8BjB237Vg/7nhp2+xbglh5OO2+g8kn31u1soKGlnfklKvpl4KWnJFKcn6EOPgNgMPvzd9GVfjmkC+eMYt64fG5/Yg31zW1+xxERkX44UEiMHbyrhxLfSguzqFSv/iNWHqglOdGYNSZ30J5DRb8ckplx4/nT2dPYyo+eqfQ7joiI9ENFIEhBdirFw9L9jiIxqrQom42796mDzxFaEggyY3QuacmJg/YcKvqlVzPH5HLpvGJ+83IVG2r027yISLSoCASZN3bwJgaKlBVl0dbh1MHnCLS0d7Bsy95Bb6urol/6ZNGCyaQnJ3LzI6v8jiIiIn2wq6GZ6tomjeeXQaUOPkduxdZ6Wts7B33CvYp+6ZMRWalcdUYpz6+t4dk1O/2OIyIivVjijefXSrwymCYWhDr4rNO4/sM22ItydVHRL3328RNKmFCQyc2PrKa1XWP3REQiWXlVkJSkBGaMGryJgSJdHXwqd+lK/+EqrwoybngGBdmpg/o8Kvqlz1KSEvjWwmls2t3Ib17e5HccERE5hIrqILPH5JKSpI96GVxlRergc7iccyypDg7JWhp6J5B+OXVyIadPKeRHz65nV0Oz33FERKQbzW0drNi6V0N7ZEhMKgx18GlTB59+C+xpYve+VuYP8tAeUNEvh+GbC6fR0t7B7Y+v9TuKiIh0462te2nrcENSSIh0dfAJ7Gn0O0rUKR+CRbm6qOiXfhs/IpNPnTSeByu2sHRznd9xRETkIOVV3iTesXk+J5F48HYHHw3x6a+KQC05aUmUFmYN+nOp6JfD8qXTSinITuXGxSvp7HR+xxERkTAVgSATRmQyPGtwJwaKgDr4HInyqiBzx+WTkDD4a2mo6JfDkpWaxLULprB0cx3/eHOr33FERMTTNTFQ4/llqHR18FmnDj79srepjcpd+wZ9Ua4uKvrlsH3g6NHMLs7j+4+vYV9Lu99xREQE2LS7kdrG1iErJESgq4OPiv7+WFLdNZ5/aObeqOiXw5aQYNx4/jRqGlr48bPr/Y4jIiKEhvbA0EwMFOlSWpTNpt2N6uDTD+WBWhITjNnFQ7OWhop+OSJHj83n4rlj+PVLm6jarVn7IiJ+qwgEyU1PZmLB4E8MFOmiDj79V14VZPqoHDJSkobk+Xot+s3s12a2y8xWhG2bY2avmdlSMys3s2O97WZm95jZejNbbmZzBzO8RIZrF0wmOdG45dFVfkcREYl7FYEgc8fmDcnEQJEupYWhDj6azNs3bR2dLNtSN6TfyPXlSv9vgQUHbbsduMk5Nwf4lncf4Byg1Pu5Arh3YGJKJCvMSeNLp5fy9OpdvLCuxu84IiJxq66pNTQxsET9+WVovd3BR+P6+2LVtnqa2zqHdC2NXot+59yLQO3Bm4Ec73YusM27fSHwexfyGpBnZiMHKqxErk++p4SS4RncuHgl+1s7/I4jIhKX3qwOrZ0yd6zG88vQSk9JZOywDCp36Up/X3QtyjW/JLKu9HfnK8AdZrYZuBO43ts+Gtgctt8Wb5vEuNSkRL73/pls2t3I7U+s8TuOiEhc6poYOKdYi3LJ0CstzFYHnz6qCNQyOi+dopy0IXvOwy36Pwd81TlXDHwV+JW3vbsBhN2u3GRmV3jzAcprajQkJBacOGkEl58wjt+8XMUrG3b7HUdEJO5UBEITA9NTEv2OInGotChLHXz6wDlHeVVwSK/yw+EX/ZcDD3m3HwCO9W5vAYrD9hvD20N/3sE59wvn3Hzn3PyCgoLDjCGR5rpzpjJ+RCaLHlhOQ3Ob33FEROJGW0cnSzcP7cRAkXDq4NM3W4L72dXQMuRraRxu0b8NOMW7fRpQ6d1eDHzc6+JzPLDXObf9CDNKFElPSeTOD85m+9793PLIar/jiIjEjdXbQxMDVfSLX9TBp2/eXktjaCfc96Vl51+AV4HJZrbFzD4NfAb4gZktA75HqFMPwGPARmA9cB/w+UFJLRFt3rh8PnvKRP5Wvpln1+z0O46ISFwor9KiXOIvdfDpm/JALVmpSUw+KntIn7fX1QCccx/u4aF53ezrgC8caSiJfl85o5Tn1uzi2r+/xZNfySc/M8XvSCIiMa2iOsjovHRG5qb7HUXi1IEOPrrSf0jlVUGOHptH4hCvpaEVeWVQpCYl8oNLZ1PX1MoN/1rR+wEiInLYnHNUVAWZq6v84rPSwmwqd+lKf0/qm9tYu7PBl2/kVPTLoJk+KperTi/lkeXbeXhZt/O5RURkAGzb28yO+uYhnxgocrAydfA5pKXVdTjHkC7K1UVFvwyqK0+ZyOziPG741wp21Tf7HUdEJCa9PTFQRb/4q9Tr4FO1Wx18ulMeCJJgMGfs0K+loaJfBlVSYgI/+OBs9rd2cP1DbxGa9iEiIgOpoqqWjJREpgzxxECRg6mDz6FVBGqZOjKHrNRep9UOOBX9MugmFWZx7YIpPLNmFw+Ub/E7johIzKmoDjKnOI+kRH2si78mFWaRYGhcfzfaOzp5s9q/tTT07iBD4hMnlnD8hGF855FVbK5t8juOiEjMaGxpZ/X2Bo3nl4iQlqwOPj1Zs6OBptYOFf0S2xISjDsumY1zjq8/uJzOTg3zEREZCMs219HR6dS5RyLGpMJs9ervRtfcm/klQz+JF1T0yxAqHpbBDQun8erGPfzu1Sq/44iIxITyQBAzOHqsin6JDF0dfFrb1cEnXHkgyMjcNEbn+bOWhop+GVIfOqaY06YU8v1/r2FDjb76ExE5UhWBIGWF2eSmJ/sdRQSAsqJs2jsdgT3q4BOuoqrW12/kVPTLkDIzvv+BmaQlJ3L1/ctoVx9fEZHD1tnpWFIdZF6JrvJL5CgtygLUwSfctrr9bNvr71oaKvplyBXmpHHzRTNYurmOn7+40e84IiJRq3LXPhqa25mnoT0SQSYWhDr4aFz/28q7xvP7sChXFxX94osLZo/ivFkj+b+n17FqW73fcUREolJ5oBaA+brSLxHkQAcfte08YEkgSEZKIlNH+reWhop+8c3NF84gNz2Fr92/lJb2Dr/jiIhEnYpAkBFZKYwdluF3FJF3KC3KVtvOMOWBWt/X0lDRL74ZlpnCbRfPZM2OBu55ptLvOCIiUaciEGTu2HzMzO8oIu9QWqgOPl261tLwqz9/FxX94qvTpxZx6fwx3Pv8BpZUB/2OIyISNWoaWgjsadLQHolIXR18qtTBh6XeWhoq+iXu3bBwGiNz07nm/mXsb9UwHxGRvuha6MfvQkKkO10dfDTEB8qrQmtp+L2Anop+8V12WjJ3XDKLjbsbue3xNX7HERGJCkuqg6QkJjBjdK7fUUTeRR183lZRHWRyUTY5af6upaGiXyLCiZNG8IkTS/jtK1W8sn6333FERCJeRSDIzDG5pCYl+h1F5F3UwSeko9PxZiAYEd/IqeiXiHHtgilMGJHJogeX09Dc5nccEZGI1dzWwVtb9vq60I9Ib0qLsuN+ga51OxtoaGlX0S8SLj0lkTsvnc32vfu5+ZFVfscREYlYK7ftpbWj0/cxwiKHUlaURVWcd/CJhEW5uqjol4gyd2w+V54ykfvLt/DM6p1+xxERiUjlVZrEK5FPHXygoqqWguxUioel+x1FRb9EnqvOKGXKUdlc99BbBBtb/Y4jIhJxKgJBSoZnMCIr1e8oIj2aVBjq4BPPk3krqoPMHxcZa2mo6JeIk5qUyF2XzqGuqZVv/muF33FERCKKc46KQJB5ETBcQORQ3u7gE5/j+nfVN7O5dn/EfCOnol8i0rRROXzljDIeXb6dh5dt8zuOiEjECOxpYk9ja8QUEiI9SUtOZNzwTNbHaQef8ghbS0NFv0Ssz548gTnFedzwrxXsqm/2O46ISEQ4MDFQK/FKFCgtzIrbK/3lVUFSkxKYPioy1tJQ0S8RKykxgR9cOpvmtg6ue+gtnHN+RxIR8V1FIEhOWhKTCrL8jiLSq9I47uBTEahldnEeKUmRUW5HRgqRHkwsyOLaBVN4ds0u7i/f7HccERHfVQRqmTsun4QE/ycGivSmq4PPpt3x1cFnf2sHK7fVR9RaGir6JeJdfkIJJ0wYznceXsXm2ia/44iI+Gbv/jbW7dzHvLGRU0iIHEppYTZA3K3Mu2xLHe2dLqKG4anol4iXkGDcfskszIxFDy6js1PDfESGmpktMLO1ZrbezK47xH6XmJkzs/ne/TPNrMLM3vL+PC1s33ne9vVmdo9FQk+7CLekOrImBor0ZkJBZlx28Knw5t7MjaBf0FX0S1QoHpbBDQun8trGWn77SpXfcUTiipklAj8BzgGmAR82s2nd7JcNfBl4PWzzbuB859xM4HLgD2GP3QtcAZR6PwsG5QXEkCWBIIkJxuziPL+jiPRJVwefyjjr1V9eVcukwizyMlL8jnKAin6JGpfOL+a0KYXc9vgaNtTE1xUDEZ8dC6x3zm10zrUCfwUu7Ga/m4HbgQPttpxzbzrnuvrurgTSzCzVzEYCOc65V11olv7vgYsG9VXEgPKqIFNHZpOZmuR3FJE+C3XwiZ+iv7PTsaS6LqLG84OKfokiZsb3PzCT9JREvnb/Mto74q8TgIhPRgPhM+m3eNsOMLOjgWLn3COHOM/FwJvOuRbv+C2HOqe8U0t7B0s31zFfi3JJlCkryqZqT1PcdPDZULOPvfvbIm4Ynop+iSqFOWncfOEMlm2u4+cvbvQ7jki86G6s/YHJNWaWANwNXN3jCcymA7cBn+3LOQ869gozKzez8pqamj6HjjX/3RRkf1sHJ5eN8DuKSL+UFmXREUcdfN5eSyOyfkFX0S9R5/zZo1g4ayT/9/Q6Vm2r9zuOSDzYAhSH3R8DhC+VnQ3MAJ43syrgeGBx2GTeMcA/gI875zaEnXPMIc55gHPuF865+c65+QUFBQPwcqLT82t3kZKYwPEThvsdRaRfujr4xMsQn/KqIMMzUygZnuF3lHdQ0S9R6eYLZ5CXkcLX7l9KS3uH33FEYt1/gVIzG29mKcBlwOKuB51ze51zI5xzJc65EuA14ALnXLmZ5QGPAtc7514OO2Y70GBmx3tdez4O/GsIX1PUeWFdDcdNGEZGisbzS3Tp6uBTuSs+5uN1raURaQ3JVPRLVMrPTOH7H5jJmh0N/PDpSr/jiMQ051w78EXgCWA1cL9zbqWZfcfMLujl8C8Ck4AbzGyp91PoPfY54JfAemAD8O/BeQXRb2vdfip37eOUsvj9xZJKWAAAIABJREFUpkOiV1pyIiVx0sFn974WqvY0RdwkXgBdLpCodfrUIi6dP4afvbCB06cWRdyEGZFY4px7DHjsoG3f6mHfU8Nu3wLc0sN+5YSGBUkvXlgbmsugol+i1aQ46eBTcWA8f+TVJLrSL1HthoXTGJmbzjUPLGN/q4b5iEhsemHdLkbnpTOpMMvvKCKHpauDT6wPya0IBElJTGD6qFy/o7yLin6JatlpydzxwVls2t3IbY+v8TuOiMiAa23v5OX1ezi5rCDixgiL9FVXB5+q3U1+RxlU5VW1zByTS1pyot9R3kVFv0S9EyeO4BMnlvDbV6r4/+3deXyddZ33/9cne7M1bTbaJmm6JGUpXQO2ZevoqEUZUIQKomw6rSIj9z2iwu3oPcL48xYVBBUBETriNoCgyIiyjA1bgSZASylkoU3adMtJc5K0abN/f3/kHIwlbbNf1znn/Xw8+mhz5eT0fbH0+vTK93p/X6xt8jqOiMiYenVHkIOdPaycp6U9ErlK86O/waeju5ctu9p8uZ4fNPRLlPjaqhOZnZPGVx7eTFtHt9dxRETGTHl1gIQ4Y8UcVXVK5JqVE2rwieKhf8uuVrp6+3z7jKGGfokKk5Li+f7qhexpPczNf9zqdRwRkTGzvipAWfEUMlISvY4iMmLhBp/qfdFb2xnelEtDv8g4W1I0hS+snMNDlQ08vXWf13FEREZtX1sHb+1p45zSvOO/WMTnSvLTqWmM3jv9FXVBZuWkkZ2e7HWUQWnol6hy3QdKOfGEDG545A2a27u8jiMiMirl1arqlOhRkhe9DT7OOV7dEfTtXX7Q0C9RJikhjts+uYjWw1184w9bvI4jIjIq5dUB8jKSOWlahtdRREYt3OCzvand6yhjbltTO83tXb59iBc09EsUOmlaJv/rH0v57817eGzTbq/jiIiMSE9vH89VBzhHVZ0SJf7W4BN96/r9vClXmIZ+iUprz57N4qIsvvH7Lexr6/A6jojIsG1qaKGto4eV87SeX6LD7Nz+Bp/aKGzwqawLkpWayOwc/26gp6FfolJCfBw/uHghnT293PC7zTjnvI4kIjIs5VUB4gzOnJvjdRSRMZGcEL0NPhX1zSwpmkJcnH+/K6ehX6LW7Nx0blh1In+tCvBfG3d6HUdEZFjWVwdYXDSFyamq6pToUZKfTnWUNfgE27t4J9Du64d4QUO/RLnLlxezfHY2Nz++lZ3N0b31t4hEj6aDnWxuaGWlWnskypTmZ1AfZQ0+767n19Av4p24OON7Fy/AzLj+oU309WmZj4j433M1oarOeRr6JbqU5GdEXYNP5Y4gifHGwsIsr6Mck4Z+iXoFU1L55nkn8/L2Zta9WOd1HBGR4yqvCpCdlsT86ZO9jiIypkry+h90jaZ1/ZV1QU6ZPpmUxHivoxyThn6JCReXFfCBE/P47p/fprYxev6gEZHo09fneLamibNLc339UKDISMzOTSM+zqiJkgafrp4+NjW0+H49P2jolxhhZnznwlOZlBTPlx/aRE9vn9eRREQG9cauVprbu1ippT0ShZIT4pmZnUpNlNzp37K7lc6ePt+v5wcN/RJD8jJT+I+PzWfTzhbuKn/H6zgiIoMqrw5gquqUKFaalxE1DT6Vdf0P8S718aZcYRr6Jaact2A6/7RwOrc/U8Obu1u9jiMi8h7rqxpZMGMy2enJXkcRGRcl+elR0+BTWR+kaGoqeRkpXkc5Lg39EnNuOv8UslKT+PKDm6LiDxwRiR4th7p4fWcL52gXXoli4QafbYHIbvBxzlFRH4yIpT2goV9i0JS0JL77iVN5e+8Bfvh0jddxRETe9VxNE30OzlE/v0Sx0vz+Bp+aCC/W2NF8iKaDnSzR0C/iX+8/MZ9PlhVyd/k7726qISLitfLqAJMnJbLI533fIqMxKyc6GnwqQuv5yyJgPT9o6JcY9m/nncS0yZO4/qFNHOrq8TqOiMQ45xzl1QHOKskhXlWdEsXCDT7VET70V+4IkpGSQGlehtdRhkRDv8SsjJREvn/xQrY3tfPdJ972Oo6IxLite9oIHOjU0h6JCaV5GRFf21lZF2RJ0ZSI2U/juEO/md1nZo1mtuWI4/9iZlVm9qaZ3TLg+I1mVhv63IfHI7TIWFk+J5urzijmPzfU80Jtk9dxRCSGlVcHAK3nl9hQmp9O3f72iC3UaD3cTXXjgYh5iBeGdqd/HbBq4AEz+wfgAmCBc+4U4Puh4ycDlwCnhL7mTjPz957EEvO+tupEZuem8ZWHNtHW0e11HBGJUeurApw8LZO8TP9X/4mM1tz8DPocEdvg8+qOIM4RETvxhh136HfOPQs0H3H4C8D/c851hl7TGDp+AfBb51ync247UAucPoZ5RcZcSmI8P7h4IXvbOrj5j1u9jiMiMaito5tX64PahVdiRrjBJ1LX9VfWBYmPMxYVRc5D9yNd018KnGVmL5tZuZmdFjo+A9g54HUNoWPvYWZrzKzCzCoCgcAIY4iMjcVFU7hm5Vweqmzg6a37vI4jIjHmxdomevqclvZIzPhbg09kruuvrA9y8rRMUpMSvI4yZCMd+hOAKcAy4CvAg2ZmwGBPMrjB3sA5d49zrsw5V5abqz/kxHtf+kAJJ03L5IZH3qC5vcvrOCISQ8qrA2QkJ0RM37fIaCUnxFOcnUpNY+Td6e/u7eP1nS0RtbQHRj70NwCPuH6vAH1ATuh44YDXFQC7RxdRZGIkJcRx6+qFtB7u4obfbaavb9C/r4qIjCnnHOVVAc6Ym0NivEr1JHaURGiDz1t72jjc3RszQ//vgfcDmFkpkAQ0AY8Bl5hZspnNAkqAV8YiqMhEOGlaJjecexJPbt3H95+s8jqOiMSAmsaD7G7t4Byt55cYE27w6eiOrAafSNuUK+y4C5HM7DfASiDHzBqA/wvcB9wXqvHsAq5wzjngTTN7ENgK9ABfdM5F1r9JiXlXn1FMbeNB7lz/DsU5aawuKzz+F4mIjFB5lao6JTaVhBp8tje1c9K0TK/jDFllfZAZWZOYNnmS11GG5bhDv3Pu0qN86tNHef23gW+PJpSIl8yMmy44hYbgIf7PI29QkDWJFXNzvI4lIlFqfXUjpfnpTM+KrAFCZLRK8/t3sq3edyBihn7nHBX1zbxvVrbXUYZNiwdFBpEYH8dPLlvC7Nw01v6yktoIfNBIRPyvvbOHjduDrJyX53UUkQlXnJMacQ0+u1oOs6+tM+KW9oCGfpGjykxJ5L4rTyM5IZ6r1m2k6WCn15FEJMq8tG0/Xb19WtojMSnc4BNJXf2V9f3r+ZcUaegXiSoFU1K594oyAgc6WfOLioh72EhE/G19VYDUpPiIvGsoMhZK8zOobYycO/0VdUHSkuI58YQMr6MMm4Z+keNYVJjFbasX8eqOFq5/aJOqPEVkTDjnWF/dyIo52SQnxHsdR8QTJXmR1eBTUR9kcdEUEiKwXjfyEot44NxTp3HjuSfy+OY9/OApVXmKyOhtb2pnZ/NhLe2RmBZu8NkWaPc6ynEd6Oimam9bxPXzh2noFxmiNWfP5tLTC/nJX9/hwYqdXscRkQhXXh2u6tRDvBK7wg0+kbAz7+s7W+hzkdfPH3bcyk4R6ddf5TmfhuBhVXmKyKitrwowOyeNouxUr6OIeGZWTlrENPhU1AWJs/5lv5FId/pFhiFc5TkrJ43P/7Iyoh4+EhH/6Oju5aVt+7ULr8S8pIS4iGnwqawPMu+ETDJSEr2OMiIa+kWGKVzlmZQQx1XrXmG/qjxFZJhe3t5MZ4+qOkWgf4lPjc9vovX09vHajiBlEbqeHzT0i4xI4dRUfnZ5GY1tnax5oDJiWgdExB/WVzWSnBDHstmRt6unyFgryc+g3ucNPlX7DtDe1Rux6/lBQ7/IiC0umsJtn1xEZX1QVZ4iMizl1QGWzc4mJVFVnSKl+em+b/AJb8oVqc09oKFfZFQ+cuo0bghVed76VLXXcUQkAuxsPsS2QLuW9oiElOT5v8Gnoi5IfmYyM7ImeR1lxNTeIzJKa8+eTV1TOz/+ay0zs1O5uKzQ60gi4mPrQ1WdK/UQrwjQ3+CTEGe+fpi3sj5I2cypmJnXUUZMd/pFRsnMuPlj8zlzbg7/59E32PDOfq8jiYiPlVc1Ujh1ErNy0ryOIuILSQlxFOekUe3T2s49rYfZ1XI4opf2gIZ+kTERrvIszk5j7QMVqvIUkUF19vTy4jv7Oac0N6LvGIqMtdL8dN9eO8Pr+SP5IV7Q0C8yZiZP+luV59XrNqrKU0Teo7IuyKGuXlZqF16RvzM3z78NPhV1QSYlxnPStEyvo4yKhn6RMRSu8tzX1qEqTxF5j/XVAZLi41g+R1WdIgOFG3zeCfjvbn9lfZCFhZNJjI/ssTmy04v40MAqz688vFlVniLyrvKqAKfNmkJasno0RAYqze9v8PHbEp/2zh627mmjbOZUr6OMmoZ+kXHwkVOn8bVVJ/LHTbu57WlVeYoI7G45TNW+A6rqFBlEcbY/G3w2NbTQ2+dYGuHr+UGVnSLj5vPn9Fd5/uh/apmZncZFSwu8jiQiHnr23apOrecXOZJfG3wq64KYwZKiyB/6dadfZJyYGf/x8fmcMTebGx/ZrCpPiWhmtsrMqsys1sxuOMbrLjIzZ2ZloY+zzeyvZnbQzH58xGvXh97z9dCPqJ6Gy6sDTJucQkleutdRRHypND+dGp/d6a+oD1Kal8HkSYleRxk1Df0i4ygxPo47L1vKzOw0Pv/LSl8+oCRyPGYWD/wEOBc4GbjUzE4e5HUZwJeAlwcc7gC+AVx/lLe/zDm3KPSjcWyT+0d3bx/P1zSpqlPkGEryMtjRfMg3JRh9fY5XdwRZEuH9/GEa+kXG2eRJidx/5WkkxBlXr9tIc3uX15FEhut0oNY5t8051wX8FrhgkNfdDNxC/6APgHOu3Tn3/MBjsei1HS0c6OzRLrwix1Diswaf6sYDHOjooUxDv4gMVeHUVH52RRl7WztY84sK39zFEBmiGcDOAR83hI69y8wWA4XOuceH+d73h5b2fMOi+Bb4+qpGEuKMFXNzvI4i4lvhBp8an6zrj5ZNucI09ItMkCVFU7h19SIq6oN87XebcU5VnhIxBhvG3/0P2MzigNuALw/zfS9zzp0KnBX68ZlBf3OzNWZWYWYVgUBgmL+FP5RXB1gycwqZKZG/LlhkvPitwaeyLkhOejJFU1O9jjImNPSLTKCPLpjGVz48jz+8vpvbnq7xOo7IUDUAhQM+LgB2D/g4A5gPrDezOmAZ8Fj4Yd6jcc7tCv18APg1/cuIBnvdPc65MudcWW5u5C2PaTzQwZu721TVKXIcSQlxzMpJo8YnXf0V9UGWzsyKmudwNPSLTLBrVs5hdVkBdzxTw+8qG7yOIzIUG4ESM5tlZknAJcBj4U8651qdcznOuWLnXDHwEnC+c67iaG9oZglmlhP6dSJwHrBlPE/CK89WNwFo6BcZghKfNPg0HuhgR/OhqNiUK0w9/SITzMz49sdPpSF4mBse2cyMKZNYNjvb61giR+Wc6zGza4G/APHAfc65N83sJqDCOffYsb4+dPc/E0gys48BHwLqgb+EBv544GngZ+N4Gp4prw6Qm5HMKdMzvY4i4nsleRk8sWUvHd29pCTGe5ajsq5/PX80bMoVpqFfxAOJ8XH89LKlXPjTF1j7QCWPXLOCObnq7hb/cs79CfjTEce+eZTXrjzi4+KjvO3SscjmZ719judqAnzgxPyoWSIgMp5K8zNwDmobDzJ/xmTPclTWB0lOiGP+dO8yjDUt7xHxyOTURO6/8nRVeYpEsU0NLbQc6lZVp8gQleb33wCr9Xhdf0V9kIUFWSQlRM+oHD1nIhKBirJTuefyMva0drD2gQo6e1TlKRJN1lcFiDM4q0RVnSJDMdMHDT4d3b28ubs1ajblCtPQL+KxpTOncOvqhWysC/LVh1XlKRJNyqsDLCrMIis1yesoIhEh3OBT7WFX/6adLXT3uqjZlCtMQ7+ID5y3YPq7VZ4/VJWnSFRobu9ic0ML55TmeR1FJKKU5mdQ0+jdnf6K0KZcSzX0i8h4uGblHC5eWsDtz9TwyKuq8hSJdM/VBHAOztF6fpFhmZuXzo7mQ57tXv9qfZA5uWlMSYuu79Bp6BfxiXCV5/LZ2Xztd5t5edt+ryOJyCiUVwWYmpbEAg8bSEQi0cAGn4nW1+eo3BGMqn7+MA39Ij6SlBDHXZ9eStHUVNb+spLtTe1eRxKREejrc5RXBzirJIe4OFV1igxHuMHHiyU+25oO0nKoO+qW9oCGfhHfCVd5xplx1f2vEFSVp0jEeXN3G/vbu1TVKTICxTn9DT41HjzMWxGFm3KFaegX8aGi7FR+dvlSdrd2sEZVniIRZ31VIwBnlWjoFxmuxHjvGnwq6oNMTUtidk7ahP/e401Dv4hPLZ05lR9c3F/lecPv3lCVp0gEKa8OcOqMyeSkJ3sdRSQiedXg82p9kCVFU6JyB20N/SI+9k8Lp3P9h0p59LVd3P6MqjxFIkHroW5e3RHU0h6RUSjJ72/wOdw1cd/p3n+wk21N7ZRF4dIe0NAv4ntf/Ie5XLS0gB8+XcOjr6nKU8Tvnq9tos/BOaUa+kVGKtzg805g4pb4VEZpP3+Yhn4RnzMz/r9wlefDb/DK9mavI4nIMZRXN5KZksCiwiyvo4hErJK8iW/wqawPkhQfx6lRWrOroV8kAoSrPAumTmLNAxWq8hTxKefCVZ25JMTrEisyUsU5aSTG24Q+zFtRH2T+jExSEuMn7PecSPoTSSRC9Fd5nkacGVev26gqTxEfenvvAfa1dWoXXpFRCjf41OybmDv9nT29vNHQSllx9G3KFaahXySCzMxO457PLGVXy2HWPlCpKk8Rn1lfFQC0nl9kLJTkZ1AzQbvybtnVSldvX9Su5wcN/SIRp6x4Kt+/eCGv1DWrylPEZ8qrGznxhAzyM1O8jiIS8UryJq7BJ7wp15IiDf0i4iPnL5zOlz/YX+V5xzO1XscREeBgZw8VdUFWzsvzOopIVJjIBp+K+iDF2ankZkTv3hoa+kUi1LXvn8snlhRw29PV/P61XV7HEYl5L9Q20dPntLRHZIyU5vc3+FSP87p+5xyv1gdZOjN61/MDJHgdQERGxsz4zoWnsqvlEF99eDPTsyZx+qzo/gNLxM/KqwOkJydE9ZpgkYk0M7u/wWe81/XX7T/E/vauqN2UK0x3+kUi2LtVnlMmsfaBCupU5SniCecc5VUBVszJJilBl1aRsTBRDT4Vdf3735RF+V/Y9SeTSITLSk3i/qtOA+AqVXmKeOKdwEF2tRxWVafIGCvJzxj3rv7K+iCZKQnMyU0f19/Haxr6RaLAzOw0fnZ5GbuCh1n7S1V5ikw0VXWKjI/SvAx2Bse3waeiPsjSmVOIi7Nx+z38QEO/SJQoK57K9y5ewCvbm7lRVZ4iE6q8OsDcvHQKpqR6HUUkqpTkp49rg0/LoS5qGw9G9aZcYRr6RaLIBYtm8K8fLOWR13bxo/9RlafIRDjU1cPL25pZqbv8ImNuvBt8Xt3R388fCw/gq71HJMr8y/vnUre/nVufqqZoaiofWzzD60giUe2lbfvp6u3Ten6RcRBu8Bmvdf0VdUES4oyFBVnj8v5+oqFfJMqEqzx3txzmXx98ne7ePi4uK/Q6lkjUKq8KMCkxntNiYHmAyERLjI9jdk46tY3jc6e/oj7IKdMzmZQUPy7v7yda3iMShZIT4rnvytM4Y24OX3l4Mz9/frvXkUSi1vrqAMvnZJOSGP1Dg4gX5uanj8ud/q6ePjbtbIn6TbnCNPSLRKnUpATuvaKMc+efwM2Pb+XWp6r1cK/IGKtraqd+/yG19oiMo/Fq8Nm6p43Onr6o35QrTEO/SBRLTojnR5cuZnVZAXc8U8O3/riVvj4N/iJjpby6v6pzpdbzi4yb0lCDT+0Y78wb3pQrFh7iBa3pF4l6CfFxfPcTC8hMSeTe57fTdribWy5aQEK8/s4vMlrrqxopzk5lZnaa11FEolZJfgYANY0HOLVg8pi9b2V9kIIpk8jPTBmz9/Sz4171zew+M2s0sy2DfO56M3NmlhP62MzsDjOrNbPNZrZkPEKLyPCYGV//6Elc/6H+Os8v/OpVOrq1gZfIaHR097Jh235WzsvzOopIVJuZnTrmDT7OOSrqg5TFyF1+GNrynnXAqiMPmlkh8EFgx4DD5wIloR9rgJ+OPqKIjAUz49r3l3DTBafw1NZ9XHX/Rg529ngdSyRivbK9mY7uPq3nFxln4QafmjHs6t/ZfJjAgU6WxlDr1nGHfufcs0DzIJ+6DfgqMHCB8AXAL1y/l4AsM5s2JklFZExcvryY2z65kFfqmrnsZy8RbO/yOpJIRCqvDpCUEMey2dleRxGJeiX56VSPYW1n5Y7+0VZ3+o/DzM4HdjnnNh3xqRnAzgEfN4SOiYiPfHxxAXd/eilv7T3A6rs3sLe1w+tIIhFnfVUj75s1NSb6vUW8VpKXQUPwMIe6xuY71BV1QTKSEygNPS8QC4Y99JtZKvB14JuDfXqQY4NWhZjZGjOrMLOKQCAw3BgiMkr/eHI+/3nV6exuOczFd79I/f52ryOJRIydzYd4J9CupT0iEyTc4PNO49hcqyrrgywqyiI+brDRNTqN5E7/HGAWsMnM6oAC4FUzO4H+O/sDt/4sAHYP9ibOuXucc2XOubLcXP2hKeKF5XOy+c2aZRzs6OGiuzZQtXd8djwUiTbP1oSrOvUQr8hECDf4VI/Buv7Ww91U7TtAWYxsyhU27KHfOfeGcy7POVfsnCumf9Bf4pzbCzwGXB5q8VkGtDrn9oxtZBEZSwsKsnhw7XLiDFbfvYFXdwS9jiTie+urAszImsScXFV1ikyE4nCDzxis639tRxDniJlNucKGUtn5G2ADMM/MGszss8d4+Z+AbUAt8DPgmjFJKSLjqiQ/g4c/v4Ks1EQ+fe/LPF/T5HUkEd/q6unjxdomVs7LxSx2lgaIeCkh1OBTOwa1na/WB4mPMxYVZo1BssgxlPaeS51z05xzic65Aufcz4/4fLFzrin0a+ec+6Jzbo5z7lTnXMV4BReRsVU4NZWH1i6naGoqV6/byJ+37PU6kogvVdQ3097Vq/X8IhNsrBp8KuqDnDQtg7Tk2NqjVltyisi78jJT+K81y5k/I5NrflXJQxU7j/9FIjGmvDpAYryxYm6O11FEYkppfgY7m0fX4NPT28frO1tYWhRbS3tAQ7+IHGFyaiK//Nz7OGNuDl95eDP3Pb/d60givlJeFaBs5lTSY+wuoYjXSvPTgdE1+Ly15wCHunpjalOuMA39IvIeqUkJ3HtFGatOOYGbHt/KrU9V49yg7bsiMWVvawdv7z3AOfO0tEdkos3NG32DT0V97G3KFaahX0QGlZwQz48/tZiLlxZwxzM1fOuPW+nr0+Avse3Z6nBVp4Z+kYlWnJ1KUnzcqNb1V9YHmT45helZk8YwWWTQ9yZF5KgS4uO45aIFTJ6UyL3Pb6eto5tbPrGAhHjdL5DYtL66kRMyU5gXQ7t4ivhFQnwcs3PTqBlFg09lfZCyGFzaAxr6ReQ4zIyvf/QkJk9K5AdPVXOwo4c7Ll1MSmK819FEJlRPbx/P1TTxkfnTVNUp4pGS/Axe3zmy/WR2tRxmT2sHS4tiq6ozTLfrROS4zIx/+UAJ3zr/FJ7cuo+r123kYOfI2xNEItFrO1s40NGj9fwiHirJSx9xg09FXWg9f4ze6dfQLyJDdsWKYm5dvZCXtzdz2b0vE2zv8jqSyIQprwoQH2ecoapOEc+EG3xqG4e/xKeyPkhqUjwnnhCby/M09IvIsFy4pIC7Pr2Ut/a08cl7NrCvrcPrSCITorw6wJKiLCZPSvQ6ikjMKskPN/iMbOhfXJQVs8+lxeZZi8iofPDkfNZddRq7goe56K4X2bH/kNeRRMZV4EAnb+xq1S68Ih6bObW/wadmmA0+Bzt7eGtPG0tnxubSHtDQLyIjtGJODr/+52Uc6OjhortepGrv6LdGF/Gr52rCVZ15HicRiW0jbfB5fUcLfQ6WxmA/f5iGfhEZsYWFWTy4djlmsPruDby2Y2SNCiJ+t74qQE56EidPy/Q6ikjMK8nPGPYGXRX1zZjB4hht7gEN/SIySqX5GTz8+RVkpSZy2b0v80Jtk9eRZByY2SozqzKzWjO74Rivu8jMnJmVhT7ONrO/mtlBM/vxEa9damZvhN7zDvNpD2Zvn+O5mgBnl+QSF+fLiCIxpTQvnYbgYdqH0SJXWR9kXn4GmSmx+0yOhn4RGbXCqak8tHY5RVNTuer+jfx5y16vI8kYMrN44CfAucDJwKVmdvIgr8sAvgS8POBwB/AN4PpB3vqnwBqgJPRj1dgmHxubG1oIHupWVaeIT5SEGnzeCQxtiU9vn+O1HS2UFcfu0h7Q0C8iYyQvM4XfrlnGKTMyueZXlTxc2eB1JBk7pwO1zrltzrku4LfABYO87mbgFvoHfQCcc+3OuecHHgMws2lApnNug3POAb8APjZeJzAa5dUBzOCsEg39In4w3Aafqr0HONjZE9Pr+UFDv4iMoazUJH752fexYk4O1z+0ifue3+51JBkbM4CdAz5uCB17l5ktBgqdc48P4z0H/s3wPe/pF+XVARYWZDE1LcnrKCLCgAafIa7rr6wPbcoVw809oKFfRMZYWnICP7+yjFWnnMBNj2/ltqeq6b+RKxFssIXs7/5LNbM44Dbgy2P1nn/3QrM1ZlZhZhWBQGAYv8XoBdu7eH1ni6o6RXwk3OAz1Id5K+qD5GUkUzBl0jgn8zcN/SIy5pIT4vnxpxZz0dICbn+mhm/9cSt9fRr8I1gDUDjg4wJg94CPM4D5wHozqwOWAY+FH+Y9xnsWHOM93+Wcu8c5V+acK8vNndjh+7naJpyDlVrPL+IrJfkZ1AwwbGYFAAATpElEQVRxV97K+iBlxVPwaVfAhNHQLyLjIiE+jls+sYDPnjmLdS/W8ZWHN9PT2+d1LBmZjUCJmc0ysyTgEuCx8Cedc63OuRznXLFzrhh4CTjfOVdxtDd0zu0BDpjZslBrz+XAH8b1LEZgfVUjWamJLCiI3Zo/ET8aaoPPvrYOGoKHY3pTrrAErwOISPSKizP+7aMnMXlSIrc+Vc2Bjm7uuHQxKYnxXkeTYXDO9ZjZtcBfgHjgPufcm2Z2E1DhnHvsWF8fuvufCSSZ2ceADznntgJfANYBk4AnQj98o6/P8Wx1E2eV5BKvqk4RXwk/zFvbeJCFhUf/S3lFXf/+MbH+EC9o6BeRcWZmfOkDJWSmJPDvf9zK1es2cs/lZaQn64+fSOKc+xPwpyOOffMor115xMfFR3ldBf3Lgnxp6542mg52slLr+UV8pzRU21lzvKG/vpmUxDhOma6N9bS8R0QmxJVnzOLW1Qt5eXszl937Mi2HuryOJHJM5dX9Dw2fraFfxHeKhtjgU1kfZGFBFonxGnn1T0BEJsyFSwr46WVLeGtPG6vv3sC+to7jf5GIR8qrAsyfkUluRrLXUUTkCENp8DnU1cObu9tiflOuMA39IjKhPnTKCay76jR2BQ9z8V0b2LH/kNeRRN6j9XA3lTuCquoU8bHS/IxjbtC1aWcrvX0u5vv5wzT0i8iEWzEnh1/98zLaOrq56K4Xqdo7tK5lkYnyYm0TvX2Oc0rzvI4iIkdRkpfOrpajN/iEN+VaXKT2LdDQLyIeWVSYxYNrl2MGq+/ewGs7gl5HEnnX+qoAGSkJLNGwIOJbAxt8BlNRH6QkL52sVO2mDRr6RcRDpfkZPPz5FUyelMhl977MC7VNXkcSwTlHeXWAM+fmkKCH/0R8K9zgM9i6/r4+x6uhTbmkn/40ExFPFU5N5eHPL6dwSipX3b+Rv7y51+tIEuOq9x1kb1uHduEV8bmiqakkJcQNujNvbeAgbR092pRrAA39IuK5vMwU/mvtMk6enskXflnJw5UNXkeSGLa+qhFQVaeI3yXExzE7J23Q2s7wplxl2pTrXRr6RcQXslKT+NXn3sfyOdlc/9Am7n9hu9eRJEaVVwc48YQMpk2e5HUUETmOozX4VNQ3k52WxMzsVA9S+ZOGfhHxjbTkBO678jQ+fEo+3/rjVn74dDXOOa9jSQw52NnDxrpmVXWKRIjS/MEbfCrrgyydOQUz8yiZ/2joFxFfSU6I5yefWsJFSwv44dM13PT4Vvr6NPjLxNjwzn66e52GfpEIEW7wGbiuP3Cgk/r9h/QQ7xESvA4gInKkhPg4bvnEAjJTErnvhe0EDnTynQtPJSMl0etoEuXKqxtJTYqnrFgP/4lEgpK8/gafmn0HWFTYX7FbWd+/nl8P8f49Df0i4ktxccY3zjuJ3Ixkvv9kFZsbWrn9kkUsLtKdGxkfzjnWVwVYMSeHpAR9I1wkEszMTntPg09lfTNJCXHMn5HpYTL/0Z9qIuJbZsYXVs7hwbXL6O1zXHzXBu5cX6vlPjIutjW10xA8rKpOkQgSH2fMyU3/u67+ivogC2ZMJjkh3sNk/qOhX0R8b+nMqfzpurNYNf8EbvlzFZ/++cvsa+vwOpZEmfVVAQCt5xeJMKX56dSEGnw6unvZsquVpVrP/x4a+kUkIkyelMiPLl3MLZ9YwGs7Wlj1w2d5eus+r2NJFCmvDjA7N43Cqar4E4kkJXl/a/DZ3NBKd6+jTOv530NDv4hEDDNj9WmFPP6lM5meNYnP/aKCb/5hCx3dvV5Hkwh3uKuXl7btZ2VpntdRRGSYBjb4/O0hXt3pP5KGfhGJOHNy03nkmhV87sxZ/GJDPRf8+IW/W88pMlwvbd9PV08f52g9v0jEKQ0N/dX7DlBZ38zs3DSmpiV5nMp/NPSLSERKTojn3847mXVXncb+9k7+6UfP88BL9drMS0akvCpASmIc75ulJQEikaZoamp/g8++A/2bcqnlbVAa+kUkoq2cl8cT153NstnZfOP3W1j7QCXB9i6vY0mEKa8OsGx2NimJavsQiTThBp8nt+4jeKhbm3IdhYZ+EYl4uRnJ3H/lafzbR0/ir1WNnHv7c2x4Z7/XsSRC1O9vZ3tTOyvV2iMSsUrz06nffwjQplxHo6FfRKJCXJzxubNm8+g1Z5CaFM+n7n2J7/3lbbp7+7yOJj5XXh2q6pynh3hFIlV4XX9WaiJzctM8TuNPGvpFJKrMnzGZP/7LmaxeWshP/voOq+/ewM7mQ17HEh8rrwpQNDWV4mxVdYpEqrl56QAsLZqCmXmcxp809ItI1ElLTuC7Fy3gx59aTG3jQT5y+3P84fVdXscSH+ro7uXFd/azcl6uBgWRCHbiCf13+suKtbTnaDT0i0jUOm/BdJ647ixKT8jgut++zpcf3MTBzh6vY4mPVNQFOdzdq114RSLczOw07v7MUj6zfKbXUXxLQ7+IRLWCKan815plfOkDJTz6WgPn3fEcmxtavI4lPlFe3UhSfBzL52R7HUVERunDp5xAenKC1zF8S0O/iES9hPg4/vWDpfx2zXK6evq48M4Xubv8Hfr61Okf69ZXBTh91lRSkzQoiEh009AvIjHj9FlTeeK6s/ngyfl854m3ueL+V2hs6/A6lnhkV8thahoPammPiMQEDf0iElMmpyZy52VL+M6Fp7KxrplVtz/H/7y9z+tY4oHyqv6qzpXzNPSLSPTT0C8iMcfMuPT0Ih7/lzPJz0zh6nUV/Ptjb9LR3et1NJlA5dWNTJ+c8m7Vn4hINNPQLyIxa25eBo9es4IrVxSz7sU6Pn7ni9Q2HvA6lkyA7t4+Xqjdzznz8lTVKSIxQUO/iMS0lMR4/v38U7jvyjL2tXVw3o+e5zev7MA5PeQbzSrrgxzs7NF6fhGJGRr6RUSA95+Yz5+vO4uymVO58ZE3uOZXr9J6qNvrWDJOyqsDJMQZZ8xVVaeIxAYN/SIiIXmZKfzi6tO58dwTeWrrPs69/Vle2d7sdSwZB+urAiydOYWMlESvo4iITAgN/SIiA8TFGWvPmcMj16wgKSGOS+7ZwK1PVdPT2+d1NBkj+9o6eGtPG+eotUdEYoiGfhGRQSwoyOLxL53FxxcXcMczNXzynpfY2XzI61gyBsqrQ1WdpXkeJxERmTga+kVEjiI9OYEfrF7I7ZcsonrvAT5yx3M8vnm317FklMqrA+RlJHPStAyvo4iITBgN/SIix3HBohn86bqzmJuXzrW/fo2vPryJQ109XseSEejp7eP5mibOKc1VVaeIxBQN/SIiQ1A4NZUH1y7n2n+Yy0OVDZx3x/Ns2dXqdSwZpk0NLbQe7tZ6fhGJORr6RUSGKDE+jus/PI9ffe59HOrq5eN3vsC9z22jr0+d/pGivCpAnMGZc3O8jiIiMqGOO/Sb2X1m1mhmWwYc+56ZvW1mm83sUTPLGvC5G82s1syqzOzD4xVcRMQrK+bk8MR1Z7FyXh7/8d9vcdW6jQQOdHodS4ZgfXWAxUVTyEpN8jqKiMiEGsqd/nXAqiOOPQXMd84tAKqBGwHM7GTgEuCU0NfcaWbxY5ZWRMQnpqQlcc9nlnLzx+bz0rb9nHv7s++2wog/NR3sZHNDq3bhFZGYdNyh3zn3LNB8xLEnnXPhp9heAgpCv74A+K1zrtM5tx2oBU4fw7wiIr5hZnxm2Uweu/ZMstOSueK+V/iPx7fS2dPrdTQZxHM1oapOrecXkRg0Fmv6rwaeCP16BrBzwOcaQsdERKLWvBMy+MO1Z3D58pnc+/x2LrzzRd4JHPQ6lhyhvCpAdloS86dP9jqKiMiEG9XQb2ZfB3qAX4UPDfKyQZ9wM7M1ZlZhZhWBgL4lLiKRLSUxnpsumM/PLi9jd8thzrvjeR7cuBPn9JCvH/T1OZ6taeLs0lzi4lTVKSKxZ8RDv5ldAZwHXOb+dlVrAAoHvKwAGHQnG+fcPc65MudcWW6uvtUqItHhgyfn88R1Z7O4KIuv/m4z1/7mNVoPd3sdK+a9sauV5vYurecXkZg1oqHfzFYBXwPOd84N3Jf+MeASM0s2s1lACfDK6GOKiESOEyan8MBn38dXV83jL1v28pHbn6Oirvn4Xyjjprw6gBmcVaKqThGJTUOp7PwNsAGYZ2YNZvZZ4MdABvCUmb1uZncBOOfeBB4EtgJ/Br7onNMTbSISc+LjjGtWzuXhL6wgPs5YffcGfvRMjdexYtb6qkYWzJhMdnqy11FERDyRcLwXOOcuHeTwz4/x+m8D3x5NKBGRaLGoMIv//tKZfPMPb3KoW/dAvNDb55ialsTioileRxER8cxxh34RERmdjJREbvvkIu3c65H4OOPeK07zOoaIiKfGorJTRESGQK0xIiLiFQ39IiIiIiJRTkO/iIiIiEiU09AvIiLHZWarzKzKzGrN7IZjvO4iM3NmVjbg2I2hr6sysw8POF5nZm+EWuAqxvscRERimR7kFRGRYzKzeOAnwAfp34Rxo5k95pzbesTrMoAvAS8POHYycAlwCjAdeNrMSgfUOf+Dc65pAk5DRCSm6U6/iIgcz+lArXNum3OuC/gtcMEgr7sZuAXoGHDsAuC3zrlO59x2oDb0fiIiMoE09IuIyPHMAHYO+LghdOxdZrYYKHTOPT6Mr3XAk2ZWaWZrxjayiIgMpOU9IiJyPIN1jb676YCZxQG3AVcO82vPcM7tNrM8+nd4f9s59+x73qD/LwRrAIqKioYZXUREQHf6RUTk+BqAwgEfFwC7B3ycAcwH1ptZHbAMeCz0MO9Rv9Y5F/65EXiUoyz7cc7d45wrc86V5ebmjskJiYjEGg39IiJyPBuBEjObZWZJ9D+Y+1j4k865VudcjnOu2DlXDLwEnO+cqwi97hIzSzazWUAJ8IqZpYUe/MXM0oAPAVsm9rRERGKHlveIiMgxOed6zOxa4C9APHCfc+5NM7sJqHDOPXaMr33TzB4EtgI9wBedc71mlg88ambQfy36tXPuz+N+MiIiMUpDv4iIHJdz7k/An4449s2jvHblER9/G/j2Ece2AQvHNqWIiByNlveIiIiIiEQ5Df0iIiIiIlFOQ7+IiIiISJTT0C8iIiIiEuU09IuIiIiIRDkN/SIiIiIiUc6cc8d/1XiHMAsA9SP88hygaQzjeEXn4S86D3/RecBM51zMb0er6wWg8/AbnYe/6DyOcr3wxdA/GmZW4Zwr8zrHaOk8/EXn4S86DxkL0fLPX+fhLzoPf9F5HJ2W94iIiIiIRDkN/SIiIiIiUS4ahv57vA4wRnQe/qLz8Bedh4yFaPnnr/PwF52Hv+g8jiLi1/SLiIiIiMixRcOdfhEREREROYaIHvrN7H+b2ZtmtsXMfmNmKV5nGgkzuy50Dm+a2f/yOs9Qmdl9ZtZoZlsGHJtqZk+ZWU3o5yleZhyKo5zHxaF/H31mFhEtAEc5j++Z2dtmttnMHjWzLC8zDsVRzuPm0Dm8bmZPmtl0LzMOxWDnMeBz15uZM7McL7LFGl0rvBcN1wtdK/xF14rhi9ih38xmAF8Cypxz84F44BJvUw2fmc0H/hk4HVgInGdmJd6mGrJ1wKojjt0APOOcKwGeCX3sd+t473lsAS4Enp3wNCO3jveex1PAfOfcAqAauHGiQ43AOt57Ht9zzi1wzi0CHge+OeGphm8d7z0PzKwQ+CCwY6IDxSJdK3xjHZF/vViHrhV+sg5dK4YlYof+kARgkpklAKnAbo/zjMRJwEvOuUPOuR6gHPi4x5mGxDn3LNB8xOELgP8M/fo/gY9NaKgRGOw8nHNvOeeqPIo0Ikc5jydD/10BvAQUTHiwYTrKebQN+DAN8P3DSEf5/wPgNuCrRMA5RBFdKzwWDdcLXSv8RdeK4YvYod85twv4Pv1/A9oDtDrnnvQ21YhsAc42s2wzSwU+AhR6nGk08p1zewBCP+d5nEf+5mrgCa9DjJSZfdvMdgKXERl3b97DzM4HdjnnNnmdJVboWuFrul74k64VHhuva0XEDv2htX8XALOA6UCamX3a21TD55x7C/gu/d9a+zOwCeg55heJDJOZfZ3+/65+5XWWkXLOfd05V0j/OVzrdZ7hCg1qXydCL0KRStcKkaHTtcJ743mtiNihH/hHYLtzLuCc6wYeAVZ4nGlEnHM/d84tcc6dTf+3eGq8zjQK+8xsGkDo50aP88Q8M7sCOA+4zEVHR++vgU94HWIE5tA/eG4yszr6v33+qpmd4Gmq6KdrhX/peuEjulb4xrhdKyJ56N8BLDOzVDMz4APAWx5nGhEzywv9XET/A0G/8TbRqDwGXBH69RXAHzzMEvPMbBXwNeB859whr/OM1BEPLJ4PvO1VlpFyzr3hnMtzzhU754qBBmCJc26vx9Gina4V/qXrhU/oWuEf43mtiOjNuczsW8An6f9W1GvA55xznd6mGj4zew7IBrqBf3XOPeNxpCExs98AK4EcYB/wf4HfAw8CRfRfbC92zg32gIpvHOU8moEfAblAC/C6c+7DXmUciqOcx41AMrA/9LKXnHOf9yTgEB3lPD4CzAP6gHrg86G12r412Hk4534+4PN19DfKNHkSMIboWuG9aLhe6FrhL7pWjOD3iuShX0REREREji+Sl/eIiIiIiMgQaOgXEREREYlyGvpFRERERKKchn4RERERkSinoV9EREREJMpp6BcRERERiXIa+kVEREREopyGfhERERGRKPf/AyZnJMQ0NREjAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 936x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "inertia_silhouetter(X, ['latitude', 'longitude'], range(8, 15, 1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We will use a _k_ of 10."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Coarse model selection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [],
   "source": [
    "estimator_dict = {\n",
    "    'lr': LinearRegression(),\n",
    "    'knn': KNeighborsRegressor(),\n",
    "    'bag': BaggingRegressor(),\n",
    "    'cart': DecisionTreeRegressor(),\n",
    "    'rf': RandomForestRegressor(),\n",
    "    'ada': AdaBoostRegressor(),\n",
    "    'svr': SVR(),\n",
    "    'xgb': XGBRegressor()\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "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>review_count</th>\n",
       "      <th>rating</th>\n",
       "      <th>category</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>1257.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>['bbq', 'smokehouse', 'southern']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>22.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>['sandwiches']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>367.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>['juicebars', 'acaibowls', 'icecream']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>84.0</td>\n",
       "      <td>2.5</td>\n",
       "      <td>['sandwiches', 'hotdogs']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2</td>\n",
       "      <td>231.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>['italian']</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   price  review_count  rating                                category\n",
       "0      2        1257.0     3.5       ['bbq', 'smokehouse', 'southern']\n",
       "1      1          22.0     3.0                          ['sandwiches']\n",
       "2      2         367.0     3.5  ['juicebars', 'acaibowls', 'icecream']\n",
       "3      1          84.0     2.5               ['sandwiches', 'hotdogs']\n",
       "4      2         231.0     4.0                             ['italian']"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = df.drop(columns=['avg_agi', 'med_agi', \n",
    "                     'count', 'city', 'id', 'alias', 'zip_code',\n",
    "                     'latitude', 'longitude'])\n",
    "y = df['avg_agi']\n",
    "X.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = df['avg_agi']\n",
    "\n",
    "estimators = ['lr', 'knn', 'bag', 'cart', 'rf', 'ada', 'svr', 'xgb']\n",
    "\n",
    "# we will try to cluster the data based on geocoordinates\n",
    "cluster_data = df[[\"latitude\", \"longitude\"]]\n",
    "\n",
    "# this was taken from the silhouette score\n",
    "km = KMeans(n_clusters=10, random_state = 42)\n",
    "km.fit(cluster_data)\n",
    "\n",
    "# this is so we can call the onehot columns 'cluster_xx'\n",
    "cluster_df = pd.DataFrame(km.labels_.reshape(-1, 1), columns=['cluster'])\n",
    "\n",
    "oh = OneHotEncoder(categories='auto', sparse=False)\n",
    "oh.fit(cluster_df)\n",
    "dummy_cols = oh.get_feature_names(['cluster'])\n",
    "\n",
    "# create the cluster dummy matrix to concatenate with the original features\n",
    "dummy_matrix = pd.DataFrame(\n",
    "    oh.transform(km.labels_.reshape(-1, 1)),\n",
    "    columns = dummy_cols,\n",
    "    index = X.index)\n",
    "\n",
    "X_dummy = pd.concat([X,dummy_matrix], axis=1)\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "        X_dummy.copy(), y.copy(), random_state=42, test_size=0.33)\n",
    "\n",
    "# custom pipeline for adjusted price\n",
    "prep = Pipeline([\n",
    "    ('catpri', CategoryPricer()),\n",
    "    ('sacr', StandardizerAndCategoryReducer())\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "lr\n",
      "0.2819749656894984\n",
      "0.2891596129112597\n",
      "\n",
      "knn\n",
      "0.4488774483156829\n",
      "0.20483105371767682\n",
      "\n",
      "bag\n",
      "0.8474910037952382\n",
      "0.18507568456534063\n",
      "\n",
      "cart\n",
      "0.9938869921642322\n",
      "-0.3474579919802514\n",
      "\n",
      "rf\n",
      "0.8493291027494249\n",
      "0.18367827962481295\n",
      "\n",
      "ada\n",
      "0.030740620083166137\n",
      "0.047399690262929776\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/Jelly/anaconda3/lib/python3.7/site-packages/sklearn/svm/base.py:193: FutureWarning: The default value of gamma will change from 'auto' to 'scale' in version 0.22 to account better for unscaled features. Set gamma explicitly to 'auto' or 'scale' to avoid this warning.\n",
      "  \"avoid this warning.\", FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "svr\n",
      "-0.05352503270866604\n",
      "-0.05856842260281425\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/Jelly/anaconda3/lib/python3.7/site-packages/xgboost/core.py:587: FutureWarning: Series.base is deprecated and will be removed in a future version\n",
      "  if getattr(data, 'base', None) is not None and \\\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[21:28:44] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "xgb\n",
      "0.32925472018608093\n",
      "0.31676913272584384\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# loop through each estimator\n",
    "for estimator in estimators:\n",
    "\n",
    "    pipe = Pipeline([\n",
    "        ('prep', prep),\n",
    "        (estimator, estimator_dict[estimator])\n",
    "    ])\n",
    "\n",
    "    pipe.fit(X_train, y_train)\n",
    "    \n",
    "    print(estimator)\n",
    "    print(pipe.score(X_train, y_train))\n",
    "    print(pipe.score(X_test, y_test))\n",
    "    print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "lr\n",
      "0.29625940232579717\n",
      "0.31622218555420956\n",
      "\n",
      "knn\n",
      "0.46461762889192365\n",
      "0.2254974358935108\n",
      "\n",
      "bag\n",
      "0.8543861517550183\n",
      "0.19822271279506412\n",
      "\n",
      "cart\n",
      "0.9931941631172392\n",
      "-0.3259959287110987\n",
      "\n",
      "rf\n",
      "0.852761991353548\n",
      "0.21921475185452524\n",
      "\n",
      "ada\n",
      "0.061987826951699465\n",
      "0.07979101016194068\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/Jelly/anaconda3/lib/python3.7/site-packages/sklearn/svm/base.py:193: FutureWarning: The default value of gamma will change from 'auto' to 'scale' in version 0.22 to account better for unscaled features. Set gamma explicitly to 'auto' or 'scale' to avoid this warning.\n",
      "  \"avoid this warning.\", FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "svr\n",
      "-0.002293654523634725\n",
      "-0.004475268027521517\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/Jelly/anaconda3/lib/python3.7/site-packages/xgboost/core.py:587: FutureWarning: Series.base is deprecated and will be removed in a future version\n",
      "  if getattr(data, 'base', None) is not None and \\\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[21:30:16] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "xgb\n",
      "0.3366242648335461\n",
      "0.3421459380777879\n",
      "\n"
     ]
    }
   ],
   "source": [
    "y = df['med_agi']\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "        X_dummy.copy(), y.copy(), random_state=42, test_size=0.33)\n",
    "\n",
    "# loop through each estimator\n",
    "for estimator in estimators:\n",
    "\n",
    "    pipe = Pipeline([\n",
    "        ('prep', prep),\n",
    "        (estimator, estimator_dict[estimator])\n",
    "    ])\n",
    "\n",
    "    pipe.fit(X_train, y_train)\n",
    "    \n",
    "    print(estimator)\n",
    "    print(pipe.score(X_train, y_train))\n",
    "    print(pipe.score(X_test, y_test))\n",
    "    print()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We will stick with using median income as our target variable.  \n",
    "\n",
    "`XGBRegressor()` seems the most balanced but is objectively not very good. We will attempt to shift the bias/variance from the higher-performing decision tree models, `BaggingRegressor()` and `RandomForestRegressor()`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [],
   "source": [
    "score_df = pd.DataFrame(columns=['params', \n",
    "                                 'rmse_train', \n",
    "                                 'rmse_test', \n",
    "                                 'train', \n",
    "                                 'test'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "        X_dummy.copy(), y.copy(), random_state=42, test_size=0.33)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "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>review_count</th>\n",
       "      <th>rating</th>\n",
       "      <th>category</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>cluster_9</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>10872</th>\n",
       "      <td>2</td>\n",
       "      <td>278.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>['cajun', 'thai', 'seafood']</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",
       "      <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>7592</th>\n",
       "      <td>1</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.5</td>\n",
       "      <td>['lounges']</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",
       "      <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>2448</th>\n",
       "      <td>1</td>\n",
       "      <td>108.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>['coffee']</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",
       "      <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>9033</th>\n",
       "      <td>1</td>\n",
       "      <td>27.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>['burgers']</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>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>14560</th>\n",
       "      <td>2</td>\n",
       "      <td>152.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>['cajun']</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>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       price  review_count  rating                      category  cluster_0  \\\n",
       "10872      2         278.0     3.5  ['cajun', 'thai', 'seafood']        0.0   \n",
       "7592       1           4.0     4.5                   ['lounges']        0.0   \n",
       "2448       1         108.0     4.0                    ['coffee']        0.0   \n",
       "9033       1          27.0     4.0                   ['burgers']        0.0   \n",
       "14560      2         152.0     4.0                     ['cajun']        1.0   \n",
       "\n",
       "       cluster_1  cluster_2  cluster_3  cluster_4  cluster_5  cluster_6  \\\n",
       "10872        0.0        1.0        0.0        0.0        0.0        0.0   \n",
       "7592         1.0        0.0        0.0        0.0        0.0        0.0   \n",
       "2448         1.0        0.0        0.0        0.0        0.0        0.0   \n",
       "9033         0.0        0.0        0.0        0.0        0.0        0.0   \n",
       "14560        0.0        0.0        0.0        0.0        0.0        0.0   \n",
       "\n",
       "       cluster_7  cluster_8  cluster_9  \n",
       "10872        0.0        0.0        0.0  \n",
       "7592         0.0        0.0        0.0  \n",
       "2448         0.0        0.0        0.0  \n",
       "9033         1.0        0.0        0.0  \n",
       "14560        0.0        0.0        0.0  "
      ]
     },
     "execution_count": 146,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "price             int64\n",
       "review_count    float64\n",
       "rating          float64\n",
       "category         object\n",
       "cluster_0       float64\n",
       "cluster_1       float64\n",
       "cluster_2       float64\n",
       "cluster_3       float64\n",
       "cluster_4       float64\n",
       "cluster_5       float64\n",
       "cluster_6       float64\n",
       "cluster_7       float64\n",
       "cluster_8       float64\n",
       "cluster_9       float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 150,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "10872    47981\n",
       "7592     74279\n",
       "2448     90399\n",
       "9033     39498\n",
       "14560    74965\n",
       "Name: med_agi, dtype: int64"
      ]
     },
     "execution_count": 153,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "metadata": {},
   "outputs": [],
   "source": [
    "Z_train = prep.fit_transform(X_train, y_train)\n",
    "Z_test = prep.transform(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "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>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>cluster_9</th>\n",
       "      <th>price</th>\n",
       "      <th>review_count</th>\n",
       "      <th>rating</th>\n",
       "      <th>price_adj</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>10872</th>\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",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.888695</td>\n",
       "      <td>0.084782</td>\n",
       "      <td>-0.103699</td>\n",
       "      <td>0.431800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7592</th>\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",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-0.836753</td>\n",
       "      <td>-0.579782</td>\n",
       "      <td>1.129366</td>\n",
       "      <td>-1.771436</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2448</th>\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",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-0.836753</td>\n",
       "      <td>-0.327539</td>\n",
       "      <td>0.512833</td>\n",
       "      <td>-0.560244</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9033</th>\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>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>-0.836753</td>\n",
       "      <td>-0.523997</td>\n",
       "      <td>0.512833</td>\n",
       "      <td>-0.594259</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14560</th>\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>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.888695</td>\n",
       "      <td>-0.220820</td>\n",
       "      <td>0.512833</td>\n",
       "      <td>0.230940</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       cluster_0  cluster_1  cluster_2  cluster_3  cluster_4  cluster_5  \\\n",
       "10872        0.0        0.0        1.0        0.0        0.0        0.0   \n",
       "7592         0.0        1.0        0.0        0.0        0.0        0.0   \n",
       "2448         0.0        1.0        0.0        0.0        0.0        0.0   \n",
       "9033         0.0        0.0        0.0        0.0        0.0        0.0   \n",
       "14560        1.0        0.0        0.0        0.0        0.0        0.0   \n",
       "\n",
       "       cluster_6  cluster_7  cluster_8  cluster_9     price  review_count  \\\n",
       "10872        0.0        0.0        0.0        0.0  0.888695      0.084782   \n",
       "7592         0.0        0.0        0.0        0.0 -0.836753     -0.579782   \n",
       "2448         0.0        0.0        0.0        0.0 -0.836753     -0.327539   \n",
       "9033         0.0        1.0        0.0        0.0 -0.836753     -0.523997   \n",
       "14560        0.0        0.0        0.0        0.0  0.888695     -0.220820   \n",
       "\n",
       "         rating  price_adj  \n",
       "10872 -0.103699   0.431800  \n",
       "7592   1.129366  -1.771436  \n",
       "2448   0.512833  -0.560244  \n",
       "9033   0.512833  -0.594259  \n",
       "14560  0.512833   0.230940  "
      ]
     },
     "execution_count": 137,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Z_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "cluster_0       0\n",
       "cluster_1       0\n",
       "cluster_2       0\n",
       "cluster_3       0\n",
       "cluster_4       0\n",
       "cluster_5       0\n",
       "cluster_6       0\n",
       "cluster_7       0\n",
       "cluster_8       0\n",
       "cluster_9       0\n",
       "price           0\n",
       "review_count    0\n",
       "rating          0\n",
       "price_adj       0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 138,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.isnan(Z_train).sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "cluster_0       0\n",
       "cluster_1       0\n",
       "cluster_2       0\n",
       "cluster_3       0\n",
       "cluster_4       0\n",
       "cluster_5       0\n",
       "cluster_6       0\n",
       "cluster_7       0\n",
       "cluster_8       0\n",
       "cluster_9       0\n",
       "price           0\n",
       "review_count    0\n",
       "rating          0\n",
       "price_adj       0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 143,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.isinf(Z_train).sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/Jelly/anaconda3/lib/python3.7/site-packages/sklearn/ensemble/forest.py:245: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n",
      "  \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(bootstrap=True, criterion='mse', max_depth=None,\n",
       "                      max_features='auto', max_leaf_nodes=None,\n",
       "                      min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "                      min_samples_leaf=1, min_samples_split=2,\n",
       "                      min_weight_fraction_leaf=0.0, n_estimators=10,\n",
       "                      n_jobs=None, oob_score=False, random_state=None,\n",
       "                      verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 154,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf = RandomForestRegressor()\n",
    "rf.fit(Z_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.20598389755233484"
      ]
     },
     "execution_count": 160,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf.score(Z_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 2 folds for each of 2 candidates, totalling 4 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=4)]: Using backend LokyBackend with 4 concurrent workers.\n",
      "[Parallel(n_jobs=4)]: Done   2 out of   4 | elapsed:    5.9s remaining:    5.9s\n"
     ]
    },
    {
     "ename": "ValueError",
     "evalue": "Input contains NaN, infinity or a value too large for dtype('float32').",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31m_RemoteTraceback\u001b[0m                          Traceback (most recent call last)",
      "\u001b[0;31m_RemoteTraceback\u001b[0m: \n\"\"\"\nTraceback (most recent call last):\n  File \"/Users/Jelly/anaconda3/lib/python3.7/site-packages/joblib/externals/loky/process_executor.py\", line 418, in _process_worker\n    r = call_item()\n  File \"/Users/Jelly/anaconda3/lib/python3.7/site-packages/joblib/externals/loky/process_executor.py\", line 272, in __call__\n    return self.fn(*self.args, **self.kwargs)\n  File \"/Users/Jelly/anaconda3/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 567, in __call__\n    return self.func(*args, **kwargs)\n  File \"/Users/Jelly/anaconda3/lib/python3.7/site-packages/joblib/parallel.py\", line 225, in __call__\n    for func, args, kwargs in self.items]\n  File \"/Users/Jelly/anaconda3/lib/python3.7/site-packages/joblib/parallel.py\", line 225, in <listcomp>\n    for func, args, kwargs in self.items]\n  File \"/Users/Jelly/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 556, in _fit_and_score\n    test_scores = _score(estimator, X_test, y_test, scorer, is_multimetric)\n  File \"/Users/Jelly/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 599, in _score\n    return _multimetric_score(estimator, X_test, y_test, scorer)\n  File \"/Users/Jelly/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 629, in _multimetric_score\n    score = scorer(estimator, X_test, y_test)\n  File \"/Users/Jelly/anaconda3/lib/python3.7/site-packages/sklearn/metrics/scorer.py\", line 240, in _passthrough_scorer\n    return estimator.score(*args, **kwargs)\n  File \"/Users/Jelly/anaconda3/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 116, in <lambda>\n    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)\n  File \"/Users/Jelly/anaconda3/lib/python3.7/site-packages/sklearn/pipeline.py\", line 602, in score\n    return self.steps[-1][-1].score(Xt, y, **score_params)\n  File \"/Users/Jelly/anaconda3/lib/python3.7/site-packages/sklearn/base.py\", line 408, in score\n    y_pred = self.predict(X)\n  File \"/Users/Jelly/anaconda3/lib/python3.7/site-packages/sklearn/ensemble/forest.py\", line 693, in predict\n    X = self._validate_X_predict(X)\n  File \"/Users/Jelly/anaconda3/lib/python3.7/site-packages/sklearn/ensemble/forest.py\", line 359, in _validate_X_predict\n    return self.estimators_[0]._validate_X_predict(X, check_input=True)\n  File \"/Users/Jelly/anaconda3/lib/python3.7/site-packages/sklearn/tree/tree.py\", line 391, in _validate_X_predict\n    X = check_array(X, dtype=DTYPE, accept_sparse=\"csr\")\n  File \"/Users/Jelly/anaconda3/lib/python3.7/site-packages/sklearn/utils/validation.py\", line 542, in check_array\n    allow_nan=force_all_finite == 'allow-nan')\n  File \"/Users/Jelly/anaconda3/lib/python3.7/site-packages/sklearn/utils/validation.py\", line 56, in _assert_all_finite\n    raise ValueError(msg_err.format(type_err, X.dtype))\nValueError: Input contains NaN, infinity or a value too large for dtype('float32').\n\"\"\"",
      "\nThe above exception was the direct cause of the following exception:\n",
      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-162-7376e3f486e9>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m     16\u001b[0m )\n\u001b[1;32m     17\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 18\u001b[0;31m \u001b[0mgrid\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_train\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_train\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     19\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     20\u001b[0m \u001b[0mpredictions\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mgrid\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpredict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_test\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/Jelly/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_search.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, X, y, groups, **fit_params)\u001b[0m\n\u001b[1;32m    686\u001b[0m                 \u001b[0;32mreturn\u001b[0m \u001b[0mresults\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    687\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 688\u001b[0;31m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_run_search\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mevaluate_candidates\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    689\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    690\u001b[0m         \u001b[0;31m# For multi-metric evaluation, store the best_index_, best_params_ and\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/Jelly/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_search.py\u001b[0m in \u001b[0;36m_run_search\u001b[0;34m(self, evaluate_candidates)\u001b[0m\n\u001b[1;32m   1147\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_run_search\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevaluate_candidates\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1148\u001b[0m         \u001b[0;34m\"\"\"Search all candidates in param_grid\"\"\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1149\u001b[0;31m         \u001b[0mevaluate_candidates\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mParameterGrid\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mparam_grid\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1150\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1151\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/Jelly/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_search.py\u001b[0m in \u001b[0;36mevaluate_candidates\u001b[0;34m(candidate_params)\u001b[0m\n\u001b[1;32m    665\u001b[0m                                \u001b[0;32mfor\u001b[0m \u001b[0mparameters\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mtrain\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtest\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    666\u001b[0m                                in product(candidate_params,\n\u001b[0;32m--> 667\u001b[0;31m                                           cv.split(X, y, groups)))\n\u001b[0m\u001b[1;32m    668\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    669\u001b[0m                 \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mout\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m<\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/Jelly/anaconda3/lib/python3.7/site-packages/joblib/parallel.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, iterable)\u001b[0m\n\u001b[1;32m    932\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    933\u001b[0m             \u001b[0;32mwith\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_backend\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mretrieval_context\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 934\u001b[0;31m                 \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mretrieve\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    935\u001b[0m             \u001b[0;31m# Make sure that we get a last message telling us we are done\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    936\u001b[0m             \u001b[0melapsed_time\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m-\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_start_time\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/Jelly/anaconda3/lib/python3.7/site-packages/joblib/parallel.py\u001b[0m in \u001b[0;36mretrieve\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    831\u001b[0m             \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    832\u001b[0m                 \u001b[0;32mif\u001b[0m \u001b[0mgetattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_backend\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'supports_timeout'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;32mFalse\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 833\u001b[0;31m                     \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_output\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mextend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mjob\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtimeout\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtimeout\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    834\u001b[0m                 \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    835\u001b[0m                     \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_output\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mextend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mjob\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/Jelly/anaconda3/lib/python3.7/site-packages/joblib/_parallel_backends.py\u001b[0m in \u001b[0;36mwrap_future_result\u001b[0;34m(future, timeout)\u001b[0m\n\u001b[1;32m    519\u001b[0m         AsyncResults.get from multiprocessing.\"\"\"\n\u001b[1;32m    520\u001b[0m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 521\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mfuture\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mresult\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtimeout\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtimeout\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    522\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mLokyTimeoutError\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    523\u001b[0m             \u001b[0;32mraise\u001b[0m \u001b[0mTimeoutError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/Jelly/anaconda3/lib/python3.7/concurrent/futures/_base.py\u001b[0m in \u001b[0;36mresult\u001b[0;34m(self, timeout)\u001b[0m\n\u001b[1;32m    430\u001b[0m                 \u001b[0;32mraise\u001b[0m \u001b[0mCancelledError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    431\u001b[0m             \u001b[0;32melif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_state\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0mFINISHED\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 432\u001b[0;31m                 \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__get_result\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    433\u001b[0m             \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    434\u001b[0m                 \u001b[0;32mraise\u001b[0m \u001b[0mTimeoutError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/Jelly/anaconda3/lib/python3.7/concurrent/futures/_base.py\u001b[0m in \u001b[0;36m__get_result\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    382\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m__get_result\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    383\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_exception\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 384\u001b[0;31m             \u001b[0;32mraise\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_exception\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    385\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    386\u001b[0m             \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_result\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mValueError\u001b[0m: Input contains NaN, infinity or a value too large for dtype('float32')."
     ]
    }
   ],
   "source": [
    "pipe_params = {\n",
    "    'rf__n_estimators': [10, 100]\n",
    "}\n",
    "\n",
    "pipe = Pipeline([\n",
    "    ('prep', prep),\n",
    "    ('rf', RandomForestRegressor())\n",
    "])\n",
    "    \n",
    "grid = GridSearchCV(\n",
    "    pipe,\n",
    "    param_grid=pipe_params,\n",
    "    cv=2,\n",
    "    verbose=1,\n",
    "    n_jobs = 4\n",
    ")\n",
    "\n",
    "grid.fit(X_train, y_train)\n",
    "\n",
    "predictions = grid.predict(X_test)\n",
    "\n",
    "mse_train = mean_squared_error(y_true = y_train,\n",
    "                              y_pred = model.predict(X_train))\n",
    "mse_test = mean_squared_error(y_true = y_test,\n",
    "                              y_pred = model.predict(X_test))\n",
    "rmse_train = mse_train ** 0.5\n",
    "rmse_test = mse_test ** 0.5\n",
    "\n",
    "scores = {}\n",
    "\n",
    "scores['params'] = grid.best_params_\n",
    "scores['rmse_train'] = rmse_train\n",
    "scores['rmse_test'] = rmse_test\n",
    "scores['train'] = grid.score(X_train, y_train)\n",
    "scores['test'] = grid.score(X_test, y_test)\n",
    "\n",
    "score_df = score_df.append(scores, ignore_index=True)\n",
    "score_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def pipeliner(estimator, k=10, ):\n",
    "    \n",
    "    # these dictionaries will allow us to pass string arguments in order to instantiate different objects\n",
    "    trans_dict = {\n",
    "        'cvect': CountVectorizer(),\n",
    "        'tvect': TfidfVectorizer()\n",
    "    }\n",
    "    \n",
    "    est_dict = {\n",
    "        'mnb': MultinomialNB(),\n",
    "        'logreg': LogisticRegression(solver='saga'),\n",
    "        'rand': RandomForestClassifier(n_estimators=10),\n",
    "        'extra': ExtraTreesClassifier(n_estimators=10)\n",
    "    }\n",
    "    \n",
    "    # this is the actual pipeline object\n",
    "    pipe = Pipeline([\n",
    "        (transformer, eval(f'trans_dict[\"{transformer}\"]')),\n",
    "        ('dense', DenseTransformer()),\n",
    "        (estimator, eval(f'est_dict[\"{estimator}\"]'))\n",
    "    ])\n",
    "    \n",
    "    return pipe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "scores = defaultdict(list)\n",
    "\n",
    "def transfer_tune(X, y, k):\n",
    "    \n",
    "    # get location data from un-pivoted dataframe\n",
    "    location_data = X[[\"latitude\", \"longitude\"]]\n",
    "    \n",
    "    # instantiate KMeans\n",
    "    km = KMeans(n_clusters=k)\n",
    "    \n",
    "    # fit\n",
    "    km.fit(location_data)\n",
    "    \n",
    "    inertia = km.inertia_\n",
    "    sil = silhouette_score(location_data, km.labels_)\n",
    "    \n",
    "    # insert 'cluster' column with predicted cluster based on k\n",
    "    X.loc[:, 'cluster'] = km.predict(location_data)\n",
    "    \n",
    "    # create dummy columns for cluster\n",
    "    X_dummy = pd.get_dummies(columns=['cluster'], data=X)\n",
    "    \n",
    "    X_dummy.drop(columns=[\"latitude\", \"longitude\"], inplace=True)\n",
    "    \n",
    "    # train-test split on new dummy dataframe\n",
    "    X_train, X_test, y_train, y_test = train_test_split(\n",
    "        X_dummy.copy(), y.copy(), random_state=111, test_size=0.3\n",
    "    )\n",
    "    \n",
    "    prep = Pipeline([\n",
    "        ('catpri', CategoryPricer()),\n",
    "        ('sacr', StandardizerAndCategoryReducer())\n",
    "    ])\n",
    "    \n",
    "    Z_train = prep.fit_transform(X_train)\n",
    "    Z_test = prep.transform(X_test)\n",
    "    \n",
    "    model = LinearRegression()\n",
    "    model.fit(Z_train, y_train)\n",
    "    r2 = model.score(Z_test, y_test)\n",
    "    print(r2)\n",
    "    # print score vs. k\n",
    "    return inertia, sil, r2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "def transfer_tune_iterator(X, y, somelist):\n",
    "    \n",
    "    k_dict = defaultdict(list)\n",
    "    \n",
    "    for k in somelist:\n",
    "        k_dict[k] = transfer_tune(X, y, k)\n",
    "        \n",
    "    k_df = pd.DataFrame(k_dict)\n",
    "    \n",
    "    return k_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.2691859914910215\n",
      "0.31355618944724606\n",
      "0.32774203963905457\n",
      "0.35766191887673415\n",
      "0.3468018295770292\n",
      "0.36479997556650967\n",
      "0.36086732975668057\n",
      "0.3996988450011499\n",
      "0.42167982777381396\n",
      "0.40900124605119137\n",
      "0.41668150370643775\n",
      "0.45952220706231384\n"
     ]
    }
   ],
   "source": [
    "k_df = transfer_tune_iterator(X, y, range(2, 203, 20))\n",
    "k_df = k_df.T.rename(columns={0: 'i', 1: 's', 2: 'r2'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1a2f4c8f98>]"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "plt.plot(k_df['i'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1a2e883c18>]"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "plt.plot(k_df['s'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "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>price</th>\n",
       "      <th>review_count</th>\n",
       "      <th>rating</th>\n",
       "      <th>med_agi</th>\n",
       "      <th>avg_agi</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>latitude</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.218405</td>\n",
       "      <td>0.012324</td>\n",
       "      <td>-0.013928</td>\n",
       "      <td>-0.003688</td>\n",
       "      <td>-0.010076</td>\n",
       "      <td>-0.001085</td>\n",
       "      <td>-0.092896</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>longitude</th>\n",
       "      <td>-0.218405</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.141744</td>\n",
       "      <td>-0.055043</td>\n",
       "      <td>-0.048951</td>\n",
       "      <td>-0.162352</td>\n",
       "      <td>-0.271891</td>\n",
       "      <td>-0.045835</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>price</th>\n",
       "      <td>0.012324</td>\n",
       "      <td>-0.141744</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.292298</td>\n",
       "      <td>0.208114</td>\n",
       "      <td>0.182626</td>\n",
       "      <td>0.237329</td>\n",
       "      <td>0.103833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>review_count</th>\n",
       "      <td>-0.013928</td>\n",
       "      <td>-0.055043</td>\n",
       "      <td>0.292298</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.166327</td>\n",
       "      <td>0.080533</td>\n",
       "      <td>0.110508</td>\n",
       "      <td>0.119049</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rating</th>\n",
       "      <td>-0.003688</td>\n",
       "      <td>-0.048951</td>\n",
       "      <td>0.208114</td>\n",
       "      <td>0.166327</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.030914</td>\n",
       "      <td>0.067282</td>\n",
       "      <td>0.026384</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>med_agi</th>\n",
       "      <td>-0.010076</td>\n",
       "      <td>-0.162352</td>\n",
       "      <td>0.182626</td>\n",
       "      <td>0.080533</td>\n",
       "      <td>0.030914</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.940798</td>\n",
       "      <td>-0.206464</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>avg_agi</th>\n",
       "      <td>-0.001085</td>\n",
       "      <td>-0.271891</td>\n",
       "      <td>0.237329</td>\n",
       "      <td>0.110508</td>\n",
       "      <td>0.067282</td>\n",
       "      <td>0.940798</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.143934</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>-0.092896</td>\n",
       "      <td>-0.045835</td>\n",
       "      <td>0.103833</td>\n",
       "      <td>0.119049</td>\n",
       "      <td>0.026384</td>\n",
       "      <td>-0.206464</td>\n",
       "      <td>-0.143934</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              latitude  longitude     price  review_count    rating   med_agi  \\\n",
       "latitude      1.000000  -0.218405  0.012324     -0.013928 -0.003688 -0.010076   \n",
       "longitude    -0.218405   1.000000 -0.141744     -0.055043 -0.048951 -0.162352   \n",
       "price         0.012324  -0.141744  1.000000      0.292298  0.208114  0.182626   \n",
       "review_count -0.013928  -0.055043  0.292298      1.000000  0.166327  0.080533   \n",
       "rating       -0.003688  -0.048951  0.208114      0.166327  1.000000  0.030914   \n",
       "med_agi      -0.010076  -0.162352  0.182626      0.080533  0.030914  1.000000   \n",
       "avg_agi      -0.001085  -0.271891  0.237329      0.110508  0.067282  0.940798   \n",
       "count        -0.092896  -0.045835  0.103833      0.119049  0.026384 -0.206464   \n",
       "\n",
       "               avg_agi     count  \n",
       "latitude     -0.001085 -0.092896  \n",
       "longitude    -0.271891 -0.045835  \n",
       "price         0.237329  0.103833  \n",
       "review_count  0.110508  0.119049  \n",
       "rating        0.067282  0.026384  \n",
       "med_agi       0.940798 -0.206464  \n",
       "avg_agi       1.000000 -0.143934  \n",
       "count        -0.143934  1.000000  "
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "prep = Pipeline([\n",
    "    ('catpri', CategoryPricer()),\n",
    "    ('sacr', StandardizerAndCategoryReducer())\n",
    "])\n",
    "\n",
    "Z_train = prep.fit_transform(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "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>review_count</th>\n",
       "      <th>rating</th>\n",
       "      <th>price_adj</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>11130</th>\n",
       "      <td>-0.834990</td>\n",
       "      <td>-0.579423</td>\n",
       "      <td>-3.186599</td>\n",
       "      <td>-0.960900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22783</th>\n",
       "      <td>-0.834990</td>\n",
       "      <td>-0.364631</td>\n",
       "      <td>-0.724279</td>\n",
       "      <td>-0.602790</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25724</th>\n",
       "      <td>0.899315</td>\n",
       "      <td>-0.192797</td>\n",
       "      <td>1.122461</td>\n",
       "      <td>0.897473</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1210</th>\n",
       "      <td>0.899315</td>\n",
       "      <td>-0.381337</td>\n",
       "      <td>1.122461</td>\n",
       "      <td>0.875899</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13717</th>\n",
       "      <td>0.899315</td>\n",
       "      <td>0.079273</td>\n",
       "      <td>1.122461</td>\n",
       "      <td>0.737111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7443</th>\n",
       "      <td>-0.834990</td>\n",
       "      <td>0.189055</td>\n",
       "      <td>1.122461</td>\n",
       "      <td>-0.555462</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4182</th>\n",
       "      <td>0.899315</td>\n",
       "      <td>-0.207117</td>\n",
       "      <td>0.506881</td>\n",
       "      <td>0.434370</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4820</th>\n",
       "      <td>0.899315</td>\n",
       "      <td>-0.488733</td>\n",
       "      <td>0.506881</td>\n",
       "      <td>1.579880</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10196</th>\n",
       "      <td>-0.834990</td>\n",
       "      <td>-0.505439</td>\n",
       "      <td>-1.339859</td>\n",
       "      <td>-0.459670</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12116</th>\n",
       "      <td>0.899315</td>\n",
       "      <td>-0.407590</td>\n",
       "      <td>1.738041</td>\n",
       "      <td>1.701330</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>21084 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          price  review_count    rating  price_adj\n",
       "11130 -0.834990     -0.579423 -3.186599  -0.960900\n",
       "22783 -0.834990     -0.364631 -0.724279  -0.602790\n",
       "25724  0.899315     -0.192797  1.122461   0.897473\n",
       "1210   0.899315     -0.381337  1.122461   0.875899\n",
       "13717  0.899315      0.079273  1.122461   0.737111\n",
       "...         ...           ...       ...        ...\n",
       "7443  -0.834990      0.189055  1.122461  -0.555462\n",
       "4182   0.899315     -0.207117  0.506881   0.434370\n",
       "4820   0.899315     -0.488733  0.506881   1.579880\n",
       "10196 -0.834990     -0.505439 -1.339859  -0.459670\n",
       "12116  0.899315     -0.407590  1.738041   1.701330\n",
       "\n",
       "[21084 rows x 4 columns]"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Z_train"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## just going to see if i can make anything right now"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>price</th>\n",
       "      <th>review_count</th>\n",
       "      <th>rating</th>\n",
       "      <th>zip_code</th>\n",
       "      <th>city</th>\n",
       "      <th>alias</th>\n",
       "      <th>category</th>\n",
       "      <th>med_agi</th>\n",
       "      <th>avg_agi</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>zzvkb4J_sQAk8N20OWdZhg</td>\n",
       "      <td>33.911154</td>\n",
       "      <td>-117.881259</td>\n",
       "      <td>2</td>\n",
       "      <td>1257.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>92821</td>\n",
       "      <td>Brea</td>\n",
       "      <td>lucilles-smokehouse-bar-b-que-brea</td>\n",
       "      <td>['bbq', 'smokehouse', 'southern']</td>\n",
       "      <td>86070</td>\n",
       "      <td>106163</td>\n",
       "      <td>170</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>zWs68lBY2PyNOTRJhxXapg</td>\n",
       "      <td>33.911959</td>\n",
       "      <td>-117.867374</td>\n",
       "      <td>1</td>\n",
       "      <td>22.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>92821</td>\n",
       "      <td>Brea</td>\n",
       "      <td>subway-restaurants-brea-4</td>\n",
       "      <td>['sandwiches']</td>\n",
       "      <td>86070</td>\n",
       "      <td>106163</td>\n",
       "      <td>170</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>zOYNwoDjpg6Ehwn0RKX9qA</td>\n",
       "      <td>33.911283</td>\n",
       "      <td>-117.867449</td>\n",
       "      <td>2</td>\n",
       "      <td>367.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>92821</td>\n",
       "      <td>Brea</td>\n",
       "      <td>nekter-juice-bar-brea</td>\n",
       "      <td>['juicebars', 'acaibowls', 'icecream']</td>\n",
       "      <td>86070</td>\n",
       "      <td>106163</td>\n",
       "      <td>170</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>zNNLl821AlzApHYqRxNALw</td>\n",
       "      <td>33.911431</td>\n",
       "      <td>-117.865941</td>\n",
       "      <td>1</td>\n",
       "      <td>84.0</td>\n",
       "      <td>2.5</td>\n",
       "      <td>92821</td>\n",
       "      <td>Brea</td>\n",
       "      <td>arbys-brea-2</td>\n",
       "      <td>['sandwiches', 'hotdogs']</td>\n",
       "      <td>86070</td>\n",
       "      <td>106163</td>\n",
       "      <td>170</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>zJ4pvxtpo3Y5LvylLXmCzA</td>\n",
       "      <td>33.932860</td>\n",
       "      <td>-117.916253</td>\n",
       "      <td>2</td>\n",
       "      <td>231.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>92821</td>\n",
       "      <td>Brea</td>\n",
       "      <td>portofino-trattoria-brea</td>\n",
       "      <td>['italian']</td>\n",
       "      <td>86070</td>\n",
       "      <td>106163</td>\n",
       "      <td>170</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       id   latitude   longitude  price  review_count  rating  \\\n",
       "0  zzvkb4J_sQAk8N20OWdZhg  33.911154 -117.881259      2        1257.0     3.5   \n",
       "1  zWs68lBY2PyNOTRJhxXapg  33.911959 -117.867374      1          22.0     3.0   \n",
       "2  zOYNwoDjpg6Ehwn0RKX9qA  33.911283 -117.867449      2         367.0     3.5   \n",
       "3  zNNLl821AlzApHYqRxNALw  33.911431 -117.865941      1          84.0     2.5   \n",
       "4  zJ4pvxtpo3Y5LvylLXmCzA  33.932860 -117.916253      2         231.0     4.0   \n",
       "\n",
       "  zip_code  city                               alias  \\\n",
       "0    92821  Brea  lucilles-smokehouse-bar-b-que-brea   \n",
       "1    92821  Brea           subway-restaurants-brea-4   \n",
       "2    92821  Brea               nekter-juice-bar-brea   \n",
       "3    92821  Brea                        arbys-brea-2   \n",
       "4    92821  Brea            portofino-trattoria-brea   \n",
       "\n",
       "                                 category  med_agi  avg_agi  count  \n",
       "0       ['bbq', 'smokehouse', 'southern']    86070   106163    170  \n",
       "1                          ['sandwiches']    86070   106163    170  \n",
       "2  ['juicebars', 'acaibowls', 'icecream']    86070   106163    170  \n",
       "3               ['sandwiches', 'hotdogs']    86070   106163    170  \n",
       "4                             ['italian']    86070   106163    170  "
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "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>review_count</th>\n",
       "      <th>rating</th>\n",
       "      <th>category</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>1257.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>['bbq', 'smokehouse', 'southern']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>22.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>['sandwiches']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>367.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>['juicebars', 'acaibowls', 'icecream']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>84.0</td>\n",
       "      <td>2.5</td>\n",
       "      <td>['sandwiches', 'hotdogs']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2</td>\n",
       "      <td>231.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>['italian']</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   price  review_count  rating                                category\n",
       "0      2        1257.0     3.5       ['bbq', 'smokehouse', 'southern']\n",
       "1      1          22.0     3.0                          ['sandwiches']\n",
       "2      2         367.0     3.5  ['juicebars', 'acaibowls', 'icecream']\n",
       "3      1          84.0     2.5               ['sandwiches', 'hotdogs']\n",
       "4      2         231.0     4.0                             ['italian']"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = df.drop(columns=['avg_agi', 'med_agi', 'count', 'city', 'id', 'alias', 'zip_code', 'longitude', 'latitude'])\n",
    "y = df['med_agi']\n",
    "X.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## with price_adj, bagging"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "grid_params = {\n",
    "    'bagger__n_estimators': [120,140,160],\n",
    "    }\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<function Pipeline.score at 0x1a33620e18>\n"
     ]
    }
   ],
   "source": [
    "location_data = df[[\"latitude\", \"longitude\"]]\n",
    "km = KMeans(n_clusters=300, random_state = 42)\n",
    "km.fit(location_data)\n",
    "X.loc[:, \"cluster\"] = km.predict(location_data)\n",
    "X_dummy = pd.get_dummies(columns=[\"cluster\"], data=X)\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "        X_dummy.copy(), y.copy(), random_state=42, test_size=0.4\n",
    ")\n",
    "\n",
    "prep = Pipeline([\n",
    "    ('catpri', CategoryPricer()),\n",
    "    ('sacr', StandardizerAndCategoryReducer())\n",
    "])\n",
    "\n",
    "pipe = Pipeline([\n",
    "    ('prep', prep),\n",
    "    ('bagger', BaggingRegressor())\n",
    "])\n",
    "\n",
    "# grid = GridSearchCV(pipe,\n",
    "#                     param_grid=grid_params,\n",
    "#                     cv=5,\n",
    "#                     verbose=1)\n",
    "\n",
    "pipe.fit(X_train, y_train)\n",
    "score = pipe.score(X_test)\n",
    "# score = grid.best_score_\n",
    "\n",
    "print (score)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9354680046585758"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pipe.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_bag = pipe.predict(X_test)\n",
    "residuals_bag = y_test - y_pred_bag"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 7))\n",
    "plt.scatter(y_pred_bag, residuals_bag, s=5)\n",
    "plt.axhline(0, color='r')\n",
    "plt.title('Residual vs. Predicted Med. Income',size = 18)\n",
    "plt.ylabel(\"Residual (1000 USD)\", fontsize = 14)\n",
    "plt.xlabel(\"Predicted Med. Income (1000 USD)\", fontsize = 14)\n",
    "plt.gca().get_xaxis().set_major_formatter(FuncFormatter(lambda x, p: format(int(x/1000), ',')))\n",
    "plt.gca().get_yaxis().set_major_formatter(FuncFormatter(lambda y, p: format(int(y/1000), ',')))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## with price_adj, ADAboost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "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>price</th>\n",
       "      <th>review_count</th>\n",
       "      <th>rating</th>\n",
       "      <th>category</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>33.911154</td>\n",
       "      <td>-117.881259</td>\n",
       "      <td>2</td>\n",
       "      <td>1257.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>['bbq', 'smokehouse', 'southern']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>33.911959</td>\n",
       "      <td>-117.867374</td>\n",
       "      <td>1</td>\n",
       "      <td>22.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>['sandwiches']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>33.911283</td>\n",
       "      <td>-117.867449</td>\n",
       "      <td>2</td>\n",
       "      <td>367.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>['juicebars', 'acaibowls', 'icecream']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>33.911431</td>\n",
       "      <td>-117.865941</td>\n",
       "      <td>1</td>\n",
       "      <td>84.0</td>\n",
       "      <td>2.5</td>\n",
       "      <td>['sandwiches', 'hotdogs']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>33.932860</td>\n",
       "      <td>-117.916253</td>\n",
       "      <td>2</td>\n",
       "      <td>231.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>['italian']</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    latitude   longitude  price  review_count  rating  \\\n",
       "0  33.911154 -117.881259      2        1257.0     3.5   \n",
       "1  33.911959 -117.867374      1          22.0     3.0   \n",
       "2  33.911283 -117.867449      2         367.0     3.5   \n",
       "3  33.911431 -117.865941      1          84.0     2.5   \n",
       "4  33.932860 -117.916253      2         231.0     4.0   \n",
       "\n",
       "                                 category  \n",
       "0       ['bbq', 'smokehouse', 'southern']  \n",
       "1                          ['sandwiches']  \n",
       "2  ['juicebars', 'acaibowls', 'icecream']  \n",
       "3               ['sandwiches', 'hotdogs']  \n",
       "4                             ['italian']  "
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = df.drop(columns=['avg_agi', 'med_agi', 'count', 'city', 'id', 'alias', 'zip_code'])\n",
    "y = df['med_agi']\n",
    "X.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "latitude        0\n",
       "longitude       0\n",
       "price           0\n",
       "review_count    0\n",
       "rating          0\n",
       "category        0\n",
       "cluster         0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/Jelly/anaconda3/lib/python3.7/site-packages/sklearn/ensemble/forest.py:245: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n",
      "  \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9403246454434404\n"
     ]
    }
   ],
   "source": [
    "location_data = df[[\"latitude\", \"longitude\"]]\n",
    "km = KMeans(n_clusters=300, random_state = 42)\n",
    "km.fit(location_data)\n",
    "X.loc[:, \"cluster\"] = km.predict(location_data)\n",
    "X_dummy = pd.get_dummies(columns=[\"cluster\"], data=X)\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "        X_dummy.copy(), y.copy(), random_state=42, test_size=0.3\n",
    ")\n",
    "\n",
    "prep = Pipeline([\n",
    "    ('catpri', CategoryPricer()),\n",
    "    ('sacr', StandardizerAndCategoryReducer())\n",
    "])\n",
    "\n",
    "pipe = Pipeline(steps=[\n",
    "    ('prep', prep),\n",
    "    ('rf', RandomForestRegressor())\n",
    "])\n",
    "\n",
    "# grid = GridSearchCV(pipe,\n",
    "#                     param_grid=grid_params,\n",
    "#                     cv=5,\n",
    "#                     verbose=1)\n",
    "\n",
    "pipe.fit(X_train, y_train)\n",
    "score = pipe.score(X_test, y_test)\n",
    "# score = grid.best_score_\n",
    "\n",
    "print (score)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## with price_adj"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "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>price</th>\n",
       "      <th>review_count</th>\n",
       "      <th>rating</th>\n",
       "      <th>category</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>33.911154</td>\n",
       "      <td>-117.881259</td>\n",
       "      <td>2</td>\n",
       "      <td>1257.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>['bbq', 'smokehouse', 'southern']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>33.911959</td>\n",
       "      <td>-117.867374</td>\n",
       "      <td>1</td>\n",
       "      <td>22.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>['sandwiches']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>33.911283</td>\n",
       "      <td>-117.867449</td>\n",
       "      <td>2</td>\n",
       "      <td>367.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>['juicebars', 'acaibowls', 'icecream']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>33.911431</td>\n",
       "      <td>-117.865941</td>\n",
       "      <td>1</td>\n",
       "      <td>84.0</td>\n",
       "      <td>2.5</td>\n",
       "      <td>['sandwiches', 'hotdogs']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>33.932860</td>\n",
       "      <td>-117.916253</td>\n",
       "      <td>2</td>\n",
       "      <td>231.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>['italian']</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    latitude   longitude  price  review_count  rating  \\\n",
       "0  33.911154 -117.881259      2        1257.0     3.5   \n",
       "1  33.911959 -117.867374      1          22.0     3.0   \n",
       "2  33.911283 -117.867449      2         367.0     3.5   \n",
       "3  33.911431 -117.865941      1          84.0     2.5   \n",
       "4  33.932860 -117.916253      2         231.0     4.0   \n",
       "\n",
       "                                 category  \n",
       "0       ['bbq', 'smokehouse', 'southern']  \n",
       "1                          ['sandwiches']  \n",
       "2  ['juicebars', 'acaibowls', 'icecream']  \n",
       "3               ['sandwiches', 'hotdogs']  \n",
       "4                             ['italian']  "
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = df.drop(columns=['avg_agi', 'med_agi', 'count', 'city', 'id', 'alias', 'zip_code'])\n",
    "y = df['med_agi']\n",
    "X.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "X['price', 'review_count', 'rating', 'category']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.03790874504228936\n"
     ]
    }
   ],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    X[['price', 'review_count', 'rating', 'category']].copy(),\n",
    "    y.copy(), random_state=42, test_size=0.3\n",
    ")\n",
    "\n",
    "prep = Pipeline([\n",
    "    ('catpri', CategoryPricer()),\n",
    "    ('sacr', StandardizerAndCategoryReducer())\n",
    "])\n",
    "\n",
    "Z_train = prep.fit_transform(X_train)\n",
    "Z_test = prep.transform(X_test)\n",
    "\n",
    "lr = LinearRegression()\n",
    "lr.fit(Z_train, y_train)\n",
    "r2 = lr.score(Z_test, y_test)\n",
    "print (r2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8114838329043217\n"
     ]
    }
   ],
   "source": [
    "location_data = df[[\"latitude\", \"longitude\"]]\n",
    "km = KMeans(n_clusters=300, random_state = 42)\n",
    "km.fit(location_data)\n",
    "X.loc[:, \"cluster\"] = km.predict(location_data)\n",
    "X_dummy = pd.get_dummies(columns=[\"cluster\"], data=X)\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "        X_dummy.copy(), y.copy(), random_state=42, test_size=0.3\n",
    ")\n",
    "\n",
    "prep = Pipeline([\n",
    "    ('catpri', CategoryPricer()),\n",
    "    ('sacr', StandardizerAndCategoryReducer())\n",
    "])\n",
    "\n",
    "Z_train = prep.fit_transform(X_train)\n",
    "Z_test = prep.transform(X_test)\n",
    "\n",
    "lr = LinearRegression()\n",
    "lr.fit(Z_train, y_train)\n",
    "r2 = lr.score(Z_test, y_test)\n",
    "print (r2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## without price-adj"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "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>price</th>\n",
       "      <th>review_count</th>\n",
       "      <th>rating</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>33.911154</td>\n",
       "      <td>-117.881259</td>\n",
       "      <td>2</td>\n",
       "      <td>1257.0</td>\n",
       "      <td>3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>33.911959</td>\n",
       "      <td>-117.867374</td>\n",
       "      <td>1</td>\n",
       "      <td>22.0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>33.911283</td>\n",
       "      <td>-117.867449</td>\n",
       "      <td>2</td>\n",
       "      <td>367.0</td>\n",
       "      <td>3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>33.911431</td>\n",
       "      <td>-117.865941</td>\n",
       "      <td>1</td>\n",
       "      <td>84.0</td>\n",
       "      <td>2.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>33.932860</td>\n",
       "      <td>-117.916253</td>\n",
       "      <td>2</td>\n",
       "      <td>231.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    latitude   longitude  price  review_count  rating\n",
       "0  33.911154 -117.881259      2        1257.0     3.5\n",
       "1  33.911959 -117.867374      1          22.0     3.0\n",
       "2  33.911283 -117.867449      2         367.0     3.5\n",
       "3  33.911431 -117.865941      1          84.0     2.5\n",
       "4  33.932860 -117.916253      2         231.0     4.0"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X1 = df.drop(columns=['avg_agi', 'med_agi', 'count', 'city', 'id', 'alias', 'zip_code', 'category'])\n",
    "y = df['med_agi']\n",
    "X1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8115538253044035\n"
     ]
    }
   ],
   "source": [
    "location_data = df[[\"latitude\", \"longitude\"]]\n",
    "km = KMeans(n_clusters=300, random_state = 42)\n",
    "km.fit(location_data)\n",
    "X1.loc[:, \"cluster\"] = km.predict(location_data)\n",
    "X_dummy = pd.get_dummies(columns=[\"cluster\"], data=X1)\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "        X_dummy.copy(), y.copy(), random_state=42, test_size=0.3\n",
    ")\n",
    "\n",
    "lr = LinearRegression()\n",
    "lr.fit(X_train, y_train)\n",
    "r2 = lr.score(X_test, y_test)\n",
    "print (r2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 4.44852614e+04, -1.03403188e+05,  1.14929374e+03,  2.06255848e-02,\n",
       "       -9.53445379e+01, -6.33667779e+03, -6.42351063e+03,  3.69774779e+04,\n",
       "        6.51687135e+04, -1.08790357e+04, -5.13034181e+04, -2.12215318e+04,\n",
       "       -4.29201192e+04,  1.36027720e+04, -3.72277364e+03, -1.75272058e+03,\n",
       "       -1.44130888e+04, -1.94787090e+04, -1.00265555e+04, -5.30803274e+04,\n",
       "       -1.92542314e+04,  3.79311732e+04,  4.16897150e+04,  1.24566348e+04,\n",
       "        4.52485035e+04, -4.32340108e+04, -1.40372118e+04,  3.60288575e+04,\n",
       "        1.80654425e+04, -4.20777857e+04, -3.82244010e+04,  3.83808251e+04,\n",
       "       -1.31303794e+05, -3.84864101e+04, -1.46141329e+03, -1.53983794e+04,\n",
       "       -1.88729929e+04, -2.75434350e+04, -3.32062536e+04,  8.41806708e+02,\n",
       "        2.05543891e+04,  7.42842318e+03,  3.63818400e+04,  1.14832235e+04,\n",
       "        4.15714920e+04, -7.52516492e+03,  3.47723555e+04, -9.23068090e+03,\n",
       "        2.03356167e+04, -3.56242798e+04,  8.29697498e+02,  2.65531285e+04,\n",
       "       -1.83286409e+04, -3.54920012e+04,  1.29960109e+04, -3.25453960e+04,\n",
       "        1.51889227e+04, -2.03862846e+04, -2.52723757e+04,  1.11505841e+03,\n",
       "       -1.28071033e+04,  2.12950342e+04,  5.84366846e+03,  6.12571183e+04,\n",
       "       -3.22660288e+04, -8.55594310e+03,  1.48926026e+03, -6.03549176e+03,\n",
       "        3.23226526e+04, -3.51246466e+04,  1.91282337e+04,  3.55217094e+04,\n",
       "       -7.79652428e+03, -1.00318422e+04, -4.13645021e+04, -1.53015580e+04,\n",
       "        4.05066259e+04, -2.45888385e+04,  2.54694518e+04, -1.70772205e+04,\n",
       "       -2.53748511e+04, -1.58594198e+04, -9.87774689e+03,  8.16534785e+03,\n",
       "       -4.24607482e+04, -1.94141111e+04,  4.51548586e+04,  5.67730778e+04,\n",
       "       -4.88550324e+04,  1.81136027e+04,  3.57732885e+04,  4.65876752e+04,\n",
       "       -3.44135916e+04, -4.05501750e+04, -5.25901114e+04,  4.09331710e+04,\n",
       "        4.57357959e+04,  3.50312299e+04, -2.50438490e+04,  7.62534430e+03,\n",
       "       -1.20763737e+04,  1.17902987e+04,  5.16714508e+03, -5.98402921e+04,\n",
       "       -4.35168986e+04, -1.20541943e+04, -3.54200737e+04,  7.68190154e+04,\n",
       "        1.75843245e+04, -1.18380946e+04, -5.31333481e+04, -5.13800880e+03,\n",
       "        3.17159404e+04, -2.41313908e+04,  2.67571084e+04,  3.15973461e+04,\n",
       "       -1.14582019e+04, -1.82498317e+04,  3.22189094e+04, -3.94579138e+04,\n",
       "       -2.86715131e+04, -2.44997100e+04,  4.70904327e+04,  7.29036826e+04,\n",
       "       -2.26811375e+04, -4.22547818e+04, -9.95624725e+03,  4.92416582e+04,\n",
       "        2.04269439e+04, -1.31858938e+04, -1.83173173e+04,  6.68828264e+04,\n",
       "        3.60721377e+04,  1.37972032e+04,  4.93198967e+04,  5.53663031e+04,\n",
       "       -5.87644382e+04, -2.84015927e+04,  8.48705825e+03,  9.29209461e+02,\n",
       "       -1.22691200e+04, -3.72044757e+04, -3.81195792e+04,  3.98931058e+03,\n",
       "       -2.13738785e+04,  4.22728001e+04,  3.33440220e+04,  1.43616717e+04,\n",
       "        2.04121174e+04,  3.80994315e+04, -1.46773626e+04, -1.43619964e+04,\n",
       "       -3.53672879e+04, -2.53585352e+04, -2.14729919e+04, -1.69912092e+04,\n",
       "       -5.81122774e+03, -3.53806342e+04, -2.66012273e+04, -3.60931753e+04,\n",
       "       -1.37596114e+04, -5.25355168e+04, -4.27018002e+04, -8.77950750e+03,\n",
       "       -3.31180243e+04, -4.83606992e+04, -4.11928184e+04, -4.80912479e+04,\n",
       "        8.54500823e+03, -2.02525337e+02,  8.97145553e+03, -5.25911782e+04,\n",
       "       -2.47895918e+04,  2.96083191e+04,  1.89930638e+04,  6.96469939e+04,\n",
       "        2.56481907e+04,  1.53693960e+04, -9.54161481e+03,  2.66612706e+02,\n",
       "        5.72872997e+03, -2.31605596e+04,  4.27701711e+04,  4.45424187e+04,\n",
       "        5.64966580e+03, -3.41908938e+03,  8.32593068e+04,  4.00615932e+04,\n",
       "        2.44522761e+04, -5.03944334e+04,  3.00387413e+04, -4.73028378e+03,\n",
       "        3.44674361e+04, -7.21033336e+03,  2.10019554e+04, -3.42026292e+03,\n",
       "       -1.04273598e+04, -2.37739365e+03,  2.27604443e+04, -9.46729832e+03,\n",
       "       -3.78454512e+04,  2.32788441e+04, -2.89833144e+04,  3.43131460e+03,\n",
       "        1.02217975e+04, -5.03781383e+04, -2.21628996e+04,  2.03090623e+04,\n",
       "       -2.33363689e+03,  4.54840268e+04, -3.88891883e+03,  1.85371923e+04,\n",
       "       -3.89095503e+04, -5.20880052e+04, -4.40541193e+04,  5.39415721e+04,\n",
       "        3.97991128e+04,  2.05758026e+03,  1.58876464e+04,  4.47217385e+04,\n",
       "       -2.39708415e+04, -3.36253166e+04, -5.45443136e+03,  2.48701238e+04,\n",
       "        4.86086001e+03,  1.77483207e+04,  1.00440375e+05,  3.94526028e+04,\n",
       "        5.57776276e+04, -2.19179870e+04,  2.68604777e+04, -2.75672983e+04,\n",
       "        1.81252998e+04,  2.53529526e+04, -2.36585401e+04,  8.27100304e+04,\n",
       "        9.83675395e+03, -3.70999850e+04, -2.46970176e+04, -7.33369910e+03,\n",
       "       -3.20025431e+04,  6.69731241e+04, -1.28696315e+03,  1.11407909e+03,\n",
       "        1.20565689e+04,  2.83086637e+04, -3.04137477e+04,  2.26689067e+04,\n",
       "        5.07787723e+04, -4.24203988e+04, -3.55142231e+04, -3.62168094e+04,\n",
       "       -2.82345088e+04, -9.78414991e+03, -2.04014333e+04,  2.71072569e+04,\n",
       "        5.07725681e+04,  3.30770615e+04,  3.14678570e+04, -5.01950211e+04,\n",
       "        1.87304058e+04, -2.05615676e+04,  8.22302252e+04,  9.55182267e+03,\n",
       "        1.83166633e+04, -5.06343019e+03,  5.21108556e+04,  1.45519152e-11,\n",
       "        1.58666473e+04,  2.62572879e+04,  3.62956422e+04,  3.05326755e+04,\n",
       "       -2.68245661e+04,  2.39982161e+04, -3.50284446e+04, -6.14559953e+03,\n",
       "       -4.40380238e+03, -1.41104655e+04, -6.68176996e+03, -3.17530140e+03,\n",
       "       -4.22973117e+04, -4.22507944e+04,  1.83543731e+04, -4.09985620e+04,\n",
       "       -1.45581072e+04, -4.19568837e+04,  5.19840226e+04, -4.47196274e+04,\n",
       "       -4.90070913e+04, -1.45936074e+03,  6.01504627e+04,  2.49415892e+04,\n",
       "        9.11384851e+03,  8.45368809e+04, -3.87711753e+04, -4.55634195e+03,\n",
       "       -3.17087652e+04, -2.08630787e+04, -1.67620651e+04, -5.84379642e+03,\n",
       "       -1.85735838e+04, -2.35497157e+03, -5.92578660e+03,  7.00274990e+04,\n",
       "       -4.18918340e+04])"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lr.coef_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [],
   "source": [
    "pd.set_option('display.max_rows', 500)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['cluster_0',\n",
       " 'cluster_1',\n",
       " 'cluster_2',\n",
       " 'cluster_3',\n",
       " 'cluster_4',\n",
       " 'cluster_5',\n",
       " 'cluster_6',\n",
       " 'cluster_7',\n",
       " 'cluster_8',\n",
       " 'cluster_9',\n",
       " 'cluster_10',\n",
       " 'cluster_11',\n",
       " 'cluster_12',\n",
       " 'cluster_13',\n",
       " 'cluster_14',\n",
       " 'cluster_15',\n",
       " 'cluster_16',\n",
       " 'cluster_17',\n",
       " 'cluster_18',\n",
       " 'cluster_19',\n",
       " 'cluster_20',\n",
       " 'cluster_21',\n",
       " 'cluster_22',\n",
       " 'cluster_23',\n",
       " 'cluster_24',\n",
       " 'cluster_25',\n",
       " 'cluster_26',\n",
       " 'cluster_27',\n",
       " 'cluster_28',\n",
       " 'cluster_29',\n",
       " 'cluster_30',\n",
       " 'cluster_31',\n",
       " 'cluster_32',\n",
       " 'cluster_33',\n",
       " 'cluster_34',\n",
       " 'cluster_35',\n",
       " 'cluster_36',\n",
       " 'cluster_37',\n",
       " 'cluster_38',\n",
       " 'cluster_39',\n",
       " 'cluster_40',\n",
       " 'cluster_41',\n",
       " 'cluster_42',\n",
       " 'cluster_43',\n",
       " 'cluster_44',\n",
       " 'cluster_45',\n",
       " 'cluster_46',\n",
       " 'cluster_47',\n",
       " 'cluster_48',\n",
       " 'cluster_49',\n",
       " 'cluster_50',\n",
       " 'cluster_51',\n",
       " 'cluster_52',\n",
       " 'cluster_53',\n",
       " 'cluster_54',\n",
       " 'cluster_55',\n",
       " 'cluster_56',\n",
       " 'cluster_57',\n",
       " 'cluster_58',\n",
       " 'cluster_59',\n",
       " 'cluster_60',\n",
       " 'cluster_61',\n",
       " 'cluster_62',\n",
       " 'cluster_63',\n",
       " 'cluster_64',\n",
       " 'cluster_65',\n",
       " 'cluster_66',\n",
       " 'cluster_67',\n",
       " 'cluster_68',\n",
       " 'cluster_69',\n",
       " 'cluster_70',\n",
       " 'cluster_71',\n",
       " 'cluster_72',\n",
       " 'cluster_73',\n",
       " 'cluster_74',\n",
       " 'cluster_75',\n",
       " 'cluster_76',\n",
       " 'cluster_77',\n",
       " 'cluster_78',\n",
       " 'cluster_79',\n",
       " 'cluster_80',\n",
       " 'cluster_81',\n",
       " 'cluster_82',\n",
       " 'cluster_83',\n",
       " 'cluster_84',\n",
       " 'cluster_85',\n",
       " 'cluster_86',\n",
       " 'cluster_87',\n",
       " 'cluster_88',\n",
       " 'cluster_89',\n",
       " 'cluster_90',\n",
       " 'cluster_91',\n",
       " 'cluster_92',\n",
       " 'cluster_93',\n",
       " 'cluster_94',\n",
       " 'cluster_95',\n",
       " 'cluster_96',\n",
       " 'cluster_97',\n",
       " 'cluster_98',\n",
       " 'cluster_99',\n",
       " 'cluster_100',\n",
       " 'cluster_101',\n",
       " 'cluster_102',\n",
       " 'cluster_103',\n",
       " 'cluster_104',\n",
       " 'cluster_105',\n",
       " 'cluster_106',\n",
       " 'cluster_107',\n",
       " 'cluster_108',\n",
       " 'cluster_109',\n",
       " 'cluster_110',\n",
       " 'cluster_111',\n",
       " 'cluster_112',\n",
       " 'cluster_113',\n",
       " 'cluster_114',\n",
       " 'cluster_115',\n",
       " 'cluster_116',\n",
       " 'cluster_117',\n",
       " 'cluster_118',\n",
       " 'cluster_119',\n",
       " 'cluster_120',\n",
       " 'cluster_121',\n",
       " 'cluster_122',\n",
       " 'cluster_123',\n",
       " 'cluster_124',\n",
       " 'cluster_125',\n",
       " 'cluster_126',\n",
       " 'cluster_127',\n",
       " 'cluster_128',\n",
       " 'cluster_129',\n",
       " 'cluster_130',\n",
       " 'cluster_131',\n",
       " 'cluster_132',\n",
       " 'cluster_133',\n",
       " 'cluster_134',\n",
       " 'cluster_135',\n",
       " 'cluster_136',\n",
       " 'cluster_137',\n",
       " 'cluster_138',\n",
       " 'cluster_139',\n",
       " 'cluster_140',\n",
       " 'cluster_141',\n",
       " 'cluster_142',\n",
       " 'cluster_143',\n",
       " 'cluster_144',\n",
       " 'cluster_145',\n",
       " 'cluster_146',\n",
       " 'cluster_147',\n",
       " 'cluster_148',\n",
       " 'cluster_149',\n",
       " 'cluster_150',\n",
       " 'cluster_151',\n",
       " 'cluster_152',\n",
       " 'cluster_153',\n",
       " 'cluster_154',\n",
       " 'cluster_155',\n",
       " 'cluster_156',\n",
       " 'cluster_157',\n",
       " 'cluster_158',\n",
       " 'cluster_159',\n",
       " 'cluster_160',\n",
       " 'cluster_161',\n",
       " 'cluster_162',\n",
       " 'cluster_163',\n",
       " 'cluster_164',\n",
       " 'cluster_165',\n",
       " 'cluster_166',\n",
       " 'cluster_167',\n",
       " 'cluster_168',\n",
       " 'cluster_169',\n",
       " 'cluster_170',\n",
       " 'cluster_171',\n",
       " 'cluster_172',\n",
       " 'cluster_173',\n",
       " 'cluster_174',\n",
       " 'cluster_175',\n",
       " 'cluster_176',\n",
       " 'cluster_177',\n",
       " 'cluster_178',\n",
       " 'cluster_179',\n",
       " 'cluster_180',\n",
       " 'cluster_181',\n",
       " 'cluster_182',\n",
       " 'cluster_183',\n",
       " 'cluster_184',\n",
       " 'cluster_185',\n",
       " 'cluster_186',\n",
       " 'cluster_187',\n",
       " 'cluster_188',\n",
       " 'cluster_189',\n",
       " 'cluster_190',\n",
       " 'cluster_191',\n",
       " 'cluster_192',\n",
       " 'cluster_193',\n",
       " 'cluster_194',\n",
       " 'cluster_195',\n",
       " 'cluster_196',\n",
       " 'cluster_197',\n",
       " 'cluster_198',\n",
       " 'cluster_199',\n",
       " 'cluster_200',\n",
       " 'cluster_201',\n",
       " 'cluster_202',\n",
       " 'cluster_203',\n",
       " 'cluster_204',\n",
       " 'cluster_205',\n",
       " 'cluster_206',\n",
       " 'cluster_207',\n",
       " 'cluster_208',\n",
       " 'cluster_209',\n",
       " 'cluster_210',\n",
       " 'cluster_211',\n",
       " 'cluster_212',\n",
       " 'cluster_213',\n",
       " 'cluster_214',\n",
       " 'cluster_215',\n",
       " 'cluster_216',\n",
       " 'cluster_217',\n",
       " 'cluster_218',\n",
       " 'cluster_219',\n",
       " 'cluster_220',\n",
       " 'cluster_221',\n",
       " 'cluster_222',\n",
       " 'cluster_223',\n",
       " 'cluster_224',\n",
       " 'cluster_225',\n",
       " 'cluster_226',\n",
       " 'cluster_227',\n",
       " 'cluster_228',\n",
       " 'cluster_229',\n",
       " 'cluster_230',\n",
       " 'cluster_231',\n",
       " 'cluster_232',\n",
       " 'cluster_233',\n",
       " 'cluster_234',\n",
       " 'cluster_235',\n",
       " 'cluster_236',\n",
       " 'cluster_237',\n",
       " 'cluster_238',\n",
       " 'cluster_239',\n",
       " 'cluster_240',\n",
       " 'cluster_241',\n",
       " 'cluster_242',\n",
       " 'cluster_243',\n",
       " 'cluster_244',\n",
       " 'cluster_245',\n",
       " 'cluster_246',\n",
       " 'cluster_247',\n",
       " 'cluster_248',\n",
       " 'cluster_249',\n",
       " 'cluster_250',\n",
       " 'cluster_251',\n",
       " 'cluster_252',\n",
       " 'cluster_253',\n",
       " 'cluster_254',\n",
       " 'cluster_255',\n",
       " 'cluster_256',\n",
       " 'cluster_257',\n",
       " 'cluster_258',\n",
       " 'cluster_259',\n",
       " 'cluster_260',\n",
       " 'cluster_261',\n",
       " 'cluster_262',\n",
       " 'cluster_263',\n",
       " 'cluster_264',\n",
       " 'cluster_265',\n",
       " 'cluster_266',\n",
       " 'cluster_267',\n",
       " 'cluster_268',\n",
       " 'cluster_269',\n",
       " 'cluster_270',\n",
       " 'cluster_271',\n",
       " 'cluster_272',\n",
       " 'cluster_273',\n",
       " 'cluster_274',\n",
       " 'cluster_275',\n",
       " 'cluster_276',\n",
       " 'cluster_277',\n",
       " 'cluster_278',\n",
       " 'cluster_279',\n",
       " 'cluster_280',\n",
       " 'cluster_281',\n",
       " 'cluster_282',\n",
       " 'cluster_283',\n",
       " 'cluster_284',\n",
       " 'cluster_285',\n",
       " 'cluster_286',\n",
       " 'cluster_287',\n",
       " 'cluster_288',\n",
       " 'cluster_289',\n",
       " 'cluster_290',\n",
       " 'cluster_291',\n",
       " 'cluster_292',\n",
       " 'cluster_293',\n",
       " 'cluster_294',\n",
       " 'cluster_295',\n",
       " 'cluster_296',\n",
       " 'cluster_297',\n",
       " 'cluster_298',\n",
       " 'cluster_299',\n",
       " 'price',\n",
       " 'review_count',\n",
       " 'rating',\n",
       " 'price_adj']"
      ]
     },
     "execution_count": 163,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(Z_train.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "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>feature</th>\n",
       "      <th>coef</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>price</td>\n",
       "      <td>4729.443837</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>review_count</td>\n",
       "      <td>728.994039</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>cluster</td>\n",
       "      <td>-16.855430</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>rating</td>\n",
       "      <td>-323.233277</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>price_adj</td>\n",
       "      <td>-936.956755</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>latitude</td>\n",
       "      <td>-1007.834407</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>longitude</td>\n",
       "      <td>-3617.984978</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        feature         coef\n",
       "3         price  4729.443837\n",
       "4  review_count   728.994039\n",
       "0       cluster   -16.855430\n",
       "5        rating  -323.233277\n",
       "6     price_adj  -936.956755\n",
       "1      latitude -1007.834407\n",
       "2     longitude -3617.984978"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "coef_dict = dict(zip(Z_train.columns, lr.coef_))\n",
    "coef_df = pd.DataFrame(coef_dict.items(), columns=['feature', 'coef'])\n",
    "coef_df.sort_values(by='coef', ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = lr.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "residuals = y_test - y_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 7))\n",
    "plt.scatter(y_pred, residuals, s=5)\n",
    "plt.axhline(0, color='r')\n",
    "plt.title('Residual vs. Predicted Med. Income',size = 18)\n",
    "plt.ylabel(\"Residual (1000 USD)\", fontsize = 14)\n",
    "plt.xlabel(\"Predicted Med. Income (1000 USD)\", fontsize = 14)\n",
    "plt.gca().get_xaxis().set_major_formatter(FuncFormatter(lambda x, p: format(int(x/1000), ',')))\n",
    "plt.gca().get_yaxis().set_major_formatter(FuncFormatter(lambda y, p: format(int(y/1000), ',')))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "metadata": {},
   "outputs": [],
   "source": [
    "pickle.dump(lr, open('../flask_app_complete/model.p', 'wb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "location_data = df[[\"latitude\", \"longitude\"]]\n",
    "km = KMeans(n_clusters=300, random_state = 42)\n",
    "km.fit(location_data)\n",
    "\n",
    "plt.figure(figsize=(10, 10))\n",
    "plt.scatter(df.longitude, df.latitude, c=km.labels_, s=8, cmap=\"tab20\")\n",
    "plt.xlabel(\"longitude\", fontsize = 20)\n",
    "plt.ylabel(\"latitude\", fontsize = 20)\n",
    "plt.title('Model with Clusters', fontsize = 24);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "metadata": {},
   "outputs": [],
   "source": [
    "import json"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "metadata": {},
   "outputs": [],
   "source": [
    "county_bounds = json.loads(open('../Assets/la_county_coordinates.json').read())\n",
    "\n",
    "la_map = folium.Map(location=[34.3390, -118.2359], zoom_start=10)\n",
    "folium.GeoJson(county_bounds).add_to(la_map)\n",
    "\n",
    "la_map.save('../Assets/la_map.html')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "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>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>cluster_9</th>\n",
       "      <th>...</th>\n",
       "      <th>cluster_294</th>\n",
       "      <th>cluster_295</th>\n",
       "      <th>cluster_296</th>\n",
       "      <th>cluster_297</th>\n",
       "      <th>cluster_298</th>\n",
       "      <th>cluster_299</th>\n",
       "      <th>price</th>\n",
       "      <th>review_count</th>\n",
       "      <th>rating</th>\n",
       "      <th>price_adj</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>18413</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</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>-0.836631</td>\n",
       "      <td>-0.197074</td>\n",
       "      <td>-0.103384</td>\n",
       "      <td>-0.562599</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14782</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</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>-0.836631</td>\n",
       "      <td>-0.165788</td>\n",
       "      <td>0.513285</td>\n",
       "      <td>-0.577336</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9960</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</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>0.889436</td>\n",
       "      <td>-0.283711</td>\n",
       "      <td>-1.336723</td>\n",
       "      <td>0.623610</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3468</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</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>0.889436</td>\n",
       "      <td>0.828132</td>\n",
       "      <td>0.513285</td>\n",
       "      <td>0.413986</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>20005</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</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>-0.836631</td>\n",
       "      <td>-0.170601</td>\n",
       "      <td>0.513285</td>\n",
       "      <td>-1.288784</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 304 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       cluster_0  cluster_1  cluster_2  cluster_3  cluster_4  cluster_5  \\\n",
       "18413          0          0          0          0          0          0   \n",
       "14782          0          0          0          0          0          0   \n",
       "9960           0          0          0          0          0          0   \n",
       "3468           0          0          0          0          0          0   \n",
       "20005          0          0          0          0          0          0   \n",
       "\n",
       "       cluster_6  cluster_7  cluster_8  cluster_9  ...  cluster_294  \\\n",
       "18413          0          0          0          0  ...            0   \n",
       "14782          0          0          0          0  ...            0   \n",
       "9960           0          0          0          0  ...            0   \n",
       "3468           0          0          0          0  ...            0   \n",
       "20005          0          0          0          0  ...            0   \n",
       "\n",
       "       cluster_295  cluster_296  cluster_297  cluster_298  cluster_299  \\\n",
       "18413            0            0            0            0            0   \n",
       "14782            0            0            0            0            0   \n",
       "9960             0            0            0            0            0   \n",
       "3468             0            0            0            0            0   \n",
       "20005            0            0            0            0            0   \n",
       "\n",
       "          price  review_count    rating  price_adj  \n",
       "18413 -0.836631     -0.197074 -0.103384  -0.562599  \n",
       "14782 -0.836631     -0.165788  0.513285  -0.577336  \n",
       "9960   0.889436     -0.283711 -1.336723   0.623610  \n",
       "3468   0.889436      0.828132  0.513285   0.413986  \n",
       "20005 -0.836631     -0.170601  0.513285  -1.288784  \n",
       "\n",
       "[5 rows x 304 columns]"
      ]
     },
     "execution_count": 131,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Z_cluster = pd.concat([Z_train,Z_test],axis = 0, sort = False)\n",
    "Z_cluster.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "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>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>cluster_9</th>\n",
       "      <th>...</th>\n",
       "      <th>cluster_294</th>\n",
       "      <th>cluster_295</th>\n",
       "      <th>cluster_296</th>\n",
       "      <th>cluster_297</th>\n",
       "      <th>cluster_298</th>\n",
       "      <th>cluster_299</th>\n",
       "      <th>price</th>\n",
       "      <th>review_count</th>\n",
       "      <th>rating</th>\n",
       "      <th>price_adj</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>18413</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</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>-0.836631</td>\n",
       "      <td>-0.197074</td>\n",
       "      <td>-0.103384</td>\n",
       "      <td>-0.562599</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14782</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</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>-0.836631</td>\n",
       "      <td>-0.165788</td>\n",
       "      <td>0.513285</td>\n",
       "      <td>-0.577336</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9960</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</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>0.889436</td>\n",
       "      <td>-0.283711</td>\n",
       "      <td>-1.336723</td>\n",
       "      <td>0.623610</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3468</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</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>0.889436</td>\n",
       "      <td>0.828132</td>\n",
       "      <td>0.513285</td>\n",
       "      <td>0.413986</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>20005</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</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>-0.836631</td>\n",
       "      <td>-0.170601</td>\n",
       "      <td>0.513285</td>\n",
       "      <td>-1.288784</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 304 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       cluster_0  cluster_1  cluster_2  cluster_3  cluster_4  cluster_5  \\\n",
       "18413          0          0          0          0          0          0   \n",
       "14782          0          0          0          0          0          0   \n",
       "9960           0          0          0          0          0          0   \n",
       "3468           0          0          0          0          0          0   \n",
       "20005          0          0          0          0          0          0   \n",
       "\n",
       "       cluster_6  cluster_7  cluster_8  cluster_9  ...  cluster_294  \\\n",
       "18413          0          0          0          0  ...            0   \n",
       "14782          0          0          0          0  ...            0   \n",
       "9960           0          0          0          0  ...            0   \n",
       "3468           0          0          0          0  ...            0   \n",
       "20005          0          0          0          0  ...            0   \n",
       "\n",
       "       cluster_295  cluster_296  cluster_297  cluster_298  cluster_299  \\\n",
       "18413            0            0            0            0            0   \n",
       "14782            0            0            0            0            0   \n",
       "9960             0            0            0            0            0   \n",
       "3468             0            0            0            0            0   \n",
       "20005            0            0            0            0            0   \n",
       "\n",
       "          price  review_count    rating  price_adj  \n",
       "18413 -0.836631     -0.197074 -0.103384  -0.562599  \n",
       "14782 -0.836631     -0.165788  0.513285  -0.577336  \n",
       "9960   0.889436     -0.283711 -1.336723   0.623610  \n",
       "3468   0.889436      0.828132  0.513285   0.413986  \n",
       "20005 -0.836631     -0.170601  0.513285  -1.288784  \n",
       "\n",
       "[5 rows x 304 columns]"
      ]
     },
     "execution_count": 132,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Z_cluster.sort_index()\n",
    "Z_cluster.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {},
   "outputs": [],
   "source": [
    "Z_cluster = pd.concat([df['zip_code'], Z_cluster.sort_index()], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {},
   "outputs": [],
   "source": [
    "Z_cluster['zip_code'] = Z_cluster['zip_code'].astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [],
   "source": [
    "Z_cluster = Z_cluster.astype(float)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.        , 0.        , 0.        , 0.        , 0.8427673 ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.00314465, 0.00314465, 0.        ,\n",
       "        0.        , 0.03144654, 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.00314465, 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.01572327, 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.10062893,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.16209942, 0.02972904, 0.20107217, 0.47677058]])"
      ]
     },
     "execution_count": 136,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Z_cluster.pivot_table(index='zip_code')[Z_cluster.pivot_table(index='zip_code').index == 90012].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "304"
      ]
     },
     "execution_count": 162,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(list(Z_cluster.pivot_table(index='zip_code')[Z_cluster.pivot_table(index='zip_code').index == 90012]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "304"
      ]
     },
     "execution_count": 164,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(list(Z_train.columns))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "metadata": {},
   "outputs": [],
   "source": [
    "bleh = Z_cluster.pivot_table(index='zip_code')[Z_cluster.pivot_table(index='zip_code').index == 90012].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([98382.47688809])"
      ]
     },
     "execution_count": 165,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_model.predict(bleh)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {},
   "outputs": [],
   "source": [
    "Z_cluster['zip_code'] = Z_cluster['zip_code'].astype(int)\n",
    "Z_cluster = Z_cluster.pivot_table(index='zip_code')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Z_cluster.to_csv('../data/Z_cluster.csv', index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-1.45139458e+04,  9.88163569e+03,  1.44374270e+04,  2.49872908e+04,\n",
       "        3.33403159e+04, -3.13005142e+04, -2.08267563e+04, -4.08333569e+04,\n",
       "       -9.77879382e+03,  1.88881475e+04,  5.39097690e+02,  3.20916880e+03,\n",
       "        4.33672150e+04, -2.11042900e+04, -2.17140656e+04,  2.94385017e+04,\n",
       "       -1.59584415e+04,  2.40872327e+04,  1.56050358e+04,  9.46470362e+03,\n",
       "       -1.08032551e+03, -3.01484306e+04,  5.45363816e+03,  2.85212953e+04,\n",
       "       -9.16359178e+03, -2.82177699e+04,  1.33230642e+04, -3.14608665e+04,\n",
       "       -2.99148309e+04,  1.29395464e+04,  3.25676579e+04, -2.80881577e+04,\n",
       "       -3.10983550e+04, -9.68894335e+03, -1.64649616e+04,  2.08069697e+04,\n",
       "        1.92070365e+03,  1.30221355e+03, -2.61395232e+03,  6.73376915e+03,\n",
       "        1.36410337e+04, -3.49471165e+03, -1.71186884e+04,  6.64030932e+04,\n",
       "       -2.89454173e+04,  1.23453369e+04, -1.98429257e+04,  7.45636200e+03,\n",
       "       -3.75951529e+04,  6.48296559e+03,  7.72206771e+03, -4.33635424e+03,\n",
       "       -7.01959960e+03,  8.53517821e+03,  1.70871051e+04, -6.07756014e+03,\n",
       "       -3.79088608e+03,  1.89283995e+04,  1.85641219e+04,  2.14974274e+04,\n",
       "       -2.53506486e+04,  1.99071468e+04, -1.90829786e+04, -1.49161958e+03,\n",
       "       -2.76654133e+04, -1.10284753e+04, -1.06626558e+04, -2.10947636e+04,\n",
       "        1.79831263e+04, -3.41029126e+04, -9.06376927e+03,  1.20982112e+04,\n",
       "        1.29924518e+04,  3.12130101e+04, -2.42260794e+04, -1.63066043e+04,\n",
       "        2.79728889e+04, -2.78609997e+04,  6.67273627e+04, -3.03598861e+04,\n",
       "        5.86382461e+03,  5.40737775e+04,  1.17305551e+04, -2.74951081e+03,\n",
       "        2.52719504e+04,  1.11679556e+04,  8.63205619e+03, -3.54900225e+04,\n",
       "       -1.80541559e+04, -1.44035884e+04,  1.15857492e+04,  1.78099497e+04,\n",
       "        2.29513863e+04, -1.35292927e+04,  2.22202095e+04, -2.01658514e+04,\n",
       "       -7.44478367e+03,  1.09789236e+03, -1.25322092e+04, -4.02933858e+04,\n",
       "        4.69357213e+04, -3.42648274e+04,  2.59157175e+04, -2.72482840e+03,\n",
       "        4.89587913e+03, -2.16419563e+04, -1.98553458e+04,  2.01682144e+03,\n",
       "        3.56278493e+04,  9.14312219e+03,  9.65418954e+03, -1.92057778e+04,\n",
       "       -3.66850897e+03, -1.48645033e+03, -3.85041489e+04, -5.79424666e+03,\n",
       "       -3.16310187e+04, -5.06703272e+03,  1.03831067e+05,  5.75555632e+03,\n",
       "       -2.70815744e+04, -2.14011553e+04,  2.05934429e+04,  1.27711325e+03,\n",
       "       -1.38614199e+04, -9.78776059e+02,  2.48091246e+04,  1.29572827e+04,\n",
       "       -1.22963123e+04, -1.44481778e+03,  6.01132504e+04, -1.33409882e+04,\n",
       "        1.14334673e+04,  2.29989581e+03,  2.52916799e+04, -8.80439851e+03,\n",
       "       -1.73329033e+04, -9.34412023e+03, -1.17710902e+04,  2.82854812e+04,\n",
       "       -8.23665706e+03, -7.59912932e+03, -2.32057785e+02, -4.15112260e+03,\n",
       "        4.34735327e+03,  5.24457826e+02, -1.95153368e+04,  1.11901503e+04,\n",
       "       -1.50882429e+04, -1.73143140e+04, -1.33288941e+04, -1.78468089e+04,\n",
       "       -2.09670023e+04, -1.50354901e+04, -2.86722185e+04,  4.45277856e+03,\n",
       "       -1.71444771e+04, -3.01845302e+03, -2.18451527e+04, -3.23644986e+04,\n",
       "       -4.44010492e+04,  1.62661330e+04, -4.90146152e+04, -8.68514254e+03,\n",
       "        5.24619693e+04,  2.85689694e+04, -2.36880735e+04, -2.51821077e+04,\n",
       "       -1.49073643e+03, -7.01517859e+03,  2.50006795e+04,  4.54129526e+04,\n",
       "       -6.54459557e+03, -1.00114557e+04,  2.20131359e+04,  7.34685350e+02,\n",
       "       -2.78862401e+04,  9.38009008e+03,  1.31399912e+04, -5.18580595e+03,\n",
       "       -2.00137598e+04,  8.59630606e+04, -1.62375879e+04,  9.93115831e+03,\n",
       "       -3.11584670e+04, -1.62906938e+03, -1.71131148e+04,  2.71889830e+02,\n",
       "       -1.02586929e+04, -1.32963896e+03, -1.29054295e+04,  3.85053059e+03,\n",
       "        8.64474221e+03,  1.75430775e+04,  1.34365713e+04, -2.65293607e+04,\n",
       "        3.05603738e+04, -2.18814336e+04, -1.17260170e+04, -6.19217942e+03,\n",
       "       -8.99145024e+03,  4.38146339e+04,  3.36920510e+04, -5.52806376e+03,\n",
       "        1.62529459e+04, -2.44350269e+04,  2.39609500e+04, -3.67113135e+04,\n",
       "       -2.60384333e+04, -8.81704732e+03,  8.49856808e+04,  6.91967451e+03,\n",
       "       -1.16086596e+04, -9.92767264e+03,  2.44091050e+04, -2.82959117e+04,\n",
       "        2.91337259e+02,  1.12708022e+04, -7.21221617e+03, -1.37942597e+04,\n",
       "       -1.05628663e+04,  8.92601098e+04,  2.12296886e+04,  1.56523808e+04,\n",
       "        1.45984421e+04,  1.80502134e+04,  2.28388579e+04,  3.43321388e+03,\n",
       "        6.28622848e+04, -1.99339151e+04,  8.60982894e+04, -1.08559153e+04,\n",
       "        2.87431399e+02, -2.82385599e+04,  3.94486855e+04, -2.48880687e+04,\n",
       "        7.76557387e+04,  1.61566008e+04, -1.78095030e+04,  3.05420981e+04,\n",
       "        2.24055196e+03,  9.90238202e+03,  1.85513206e+04,  2.00130880e+04,\n",
       "       -2.15077692e+04, -2.16890507e+04, -2.39048184e+04, -2.40108760e+04,\n",
       "       -2.72848924e+04, -2.86750820e+04,  2.19551403e+04,  2.76220025e+04,\n",
       "       -1.63776030e+04,  4.97289584e+04, -2.15024664e+04, -9.80104837e+03,\n",
       "        2.45691523e+04,  2.13962103e+04, -2.53249981e+04, -4.54221197e+03,\n",
       "        1.44471489e+03,  1.13733505e+04,  1.31603455e+17,  3.18025474e+04,\n",
       "       -1.70425511e+04, -1.13858604e+04, -5.56934804e+03, -5.11581607e+03,\n",
       "       -5.44733221e+03, -3.61029063e+04,  1.92389609e+04, -1.95021172e+04,\n",
       "       -2.40181001e+04, -2.13809273e+04,  1.71709528e+04,  2.19780833e+03,\n",
       "       -1.26268515e+04,  1.52803764e+04, -3.89172310e+04, -2.27176973e+04,\n",
       "       -3.29983440e+04, -2.86681092e+03, -3.87748807e+04, -2.59159668e+04,\n",
       "       -2.23110609e+04,  5.28345528e+03, -1.91656920e+03, -6.93926823e+03,\n",
       "        3.24907418e+04, -3.01274034e+04, -2.25906627e+03,  1.66808048e+03,\n",
       "       -2.59252723e+04, -8.77052262e+02,  4.51801971e+04, -2.93894283e+04,\n",
       "       -3.10310788e+03, -1.99488078e+04,  2.49466561e+04, -3.65344680e+04,\n",
       "        4.35039856e+02,  2.03559714e+01, -6.16659067e+01,  2.72479901e+02])"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lr.coef_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>avg_agi</th>\n",
       "      <th>count</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>med_agi</th>\n",
       "      <th>price</th>\n",
       "      <th>rating</th>\n",
       "      <th>review_count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>zip_code</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",
       "      <td>90012</td>\n",
       "      <td>61644</td>\n",
       "      <td>318</td>\n",
       "      <td>34.056503</td>\n",
       "      <td>-118.241029</td>\n",
       "      <td>38786</td>\n",
       "      <td>1.578616</td>\n",
       "      <td>3.746855</td>\n",
       "      <td>442.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          avg_agi  count   latitude   longitude  med_agi     price    rating  \\\n",
       "zip_code                                                                       \n",
       "90012       61644    318  34.056503 -118.241029    38786  1.578616  3.746855   \n",
       "\n",
       "          review_count  \n",
       "zip_code                \n",
       "90012            442.0  "
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.pivot_table(index='zip_code')[df.pivot_table(index='zip_code').index == '90012']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_model = pickle.load(open('../flask_app_complete/model.p','rb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearRegression(copy_X=True, fit_intercept=True, n_jobs=None, normalize=False)"
      ]
     },
     "execution_count": 159,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_2 = test_model.predict(Z_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "metadata": {},
   "outputs": [],
   "source": [
    "residuals_2 = y_test - y_pred_2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Residual Mean')"
      ]
     },
     "execution_count": 168,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "plt.scatter(y_pred_2, residuals_2)\n",
    "plt.axhline(0, color='r')\n",
    "plt.title('Residual Mean',size = 20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "Z_cluster.drop()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# https://stackoverflow.com/questions/49735683/\n",
    "df_2 = df_2.groupby('zip_code').filter(lambda x : len(x) > 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "train, test = category_pricer(df_2, state=111)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "train['zip_code'].dtype"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "drop_columns = ['id', 'city', 'alias', 'med_agi', 'avg_agi', 'count']\n",
    "\n",
    "X_train = train.drop(columns=drop_columns)\n",
    "y_train = train['med_agi']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_pivot = df_2.pivot_table(index=['zip_code'])\n",
    "df_pivot['price*review_count'] = df_pivot['price']*df_pivot['review_count']\n",
    "df_pivot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "X.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_pivot = train.pivot_table(index = ['zip_code'])\n",
    "train_pivot['price*review_count'] = train_pivot['price']*df_pivot['review_count']\n",
    "train_pivot.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "lr = LinearRegression()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.scatter(x=train_pivot['price_sc'], y=train_pivot['avg_agi'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "lr.fit(train_pivot[['price_sc']], train_pivot['avg_agi'], sample_weight=train_pivot['count'])\n",
    "print(lr.score(train_pivot[['price_sc']], train_pivot['avg_agi'], sample_weight=train_pivot['count']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.scatter(x=train_pivot['price_adj'], y=train_pivot['avg_agi'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "lr.fit(train_pivot[['price_adj']], train_pivot['avg_agi'], sample_weight=train_pivot['count'])\n",
    "print(lr.score(train_pivot[['price_adj']], train_pivot['avg_agi'], sample_weight=train_pivot['count']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.scatter(x=train_pivot['price_sc'], y=train_pivot['med_agi'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "lr.fit(train_pivot[['price_sc']], train_pivot['med_agi'], sample_weight=train_pivot['count'])\n",
    "print(lr.score(train_pivot[['price_sc']], train_pivot['med_agi'], sample_weight=train_pivot['count']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.scatter(x=train_pivot['price_adj'], y=train_pivot['med_agi'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "lr.fit(train_pivot[['price_adj']], train_pivot['med_agi'], sample_weight=train_pivot['count'])\n",
    "print(lr.score(train_pivot[['price_adj']], train_pivot['med_agi'], sample_weight=train_pivot['count']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_pivot = test.pivot_table(index = ['zip_code'])\n",
    "test_pivot['price*review_count'] = test_pivot['price']*df_pivot['review_count']\n",
    "test_pivot.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## trying different models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = train_pivot.drop(columns=['avg_agi', 'latitude', 'longitude', 'med_agi'])\n",
    "y_train_avg = train_pivot['avg_agi']\n",
    "y_train_med = train_pivot['med_agi']\n",
    "\n",
    "X_test = test_pivot.drop(columns=['avg_agi', 'latitude', 'longitude', 'med_agi'])\n",
    "y_test_avg = test_pivot['avg_agi']\n",
    "y_test_med = test_pivot['med_agi']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "linear_reg = LinearRegression()\n",
    "linear_reg.fit(X_train, y_train_med)\n",
    "\n",
    "knn_reg = KNeighborsRegressor()\n",
    "knn_reg.fit(X_train, y_train_med)\n",
    "\n",
    "cart_reg = DecisionTreeRegressor()\n",
    "cart_reg.fit(X_train, y_train_med)\n",
    "\n",
    "bagged_reg = BaggingRegressor()\n",
    "bagged_reg.fit(X_train, y_train_med)\n",
    "\n",
    "random_forest_reg = RandomForestRegressor()\n",
    "random_forest_reg.fit(X_train, y_train_med)\n",
    "\n",
    "adaboost_reg = AdaBoostRegressor()\n",
    "adaboost_reg.fit(X_train, y_train_med)\n",
    "\n",
    "support_vector_reg = SVR()\n",
    "support_vector_reg.fit(X_train, y_train_med)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def rmse_score(model, X_train, X_test, y_train, y_test):\n",
    "    mse_train = mean_squared_error(y_true = y_train,\n",
    "                                  y_pred = model.predict(X_train))\n",
    "    mse_test = mean_squared_error(y_true = y_test,\n",
    "                                  y_pred = model.predict(X_test))\n",
    "    rmse_train = mse_train ** 0.5\n",
    "    rmse_test = mse_test ** 0.5\n",
    "\n",
    "    return (rmse_train, rmse_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "rmse_score(linear_reg, X_train, X_test, y_train_med, y_test_med)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(linear_reg.score(X_train, y_train_med))\n",
    "print(linear_reg.score(X_test, y_test_med))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## clustering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_coordinates(zipcode):\n",
    "    '''\n",
    "    Converts a five-digit USPS zip code to geographic coordinates.\n",
    "    \n",
    "    Arguments:\n",
    "        zipcode:            a five-digit string or integer\n",
    "        \n",
    "    Returns:\n",
    "        a tuple containing (longitude, latitude), in decimal degrees\n",
    "    ''' \n",
    "    zipcode = str(zipcode)\n",
    "    zip_coords = json.loads(open('../Assets/zip_code_coordinates.json').read())\n",
    "    if len(zipcode) == 5 and zipcode.isnumeric():\n",
    "        latitude = zip_coords[zipcode]['latitude']\n",
    "        longitude = zip_coords[zipcode]['longitude']\n",
    "        \n",
    "    return longitude, latitude"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "transfer_tune_iterator(X, y, range(20, 50, 1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## median agi"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = train_pivot.drop(columns=['avg_agi','med_agi','latitude','longitude'])\n",
    "#features = list(X.columns)\n",
    "y_train = df_dummy['med_agi']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_test = test_pivot.drop(columns=['avg_agi','med_agi','latitude','longitude'])\n",
    "#features = list(X.columns)\n",
    "y_test = df_dummy['med_agi']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def pipeliner(transformer, estimator):\n",
    "    \n",
    "    # these dictionaries will allow us to pass string arguments in order to instantiate different objects\n",
    "    trans_dict = {\n",
    "        'cvect': CountVectorizer(),\n",
    "        'tvect': TfidfVectorizer()\n",
    "    }\n",
    "    \n",
    "    est_dict = {\n",
    "        'mnb': MultinomialNB(),\n",
    "        'logreg': LogisticRegression(solver='saga'),\n",
    "        'rand': RandomForestClassifier(n_estimators=10),\n",
    "        'extra': ExtraTreesClassifier(n_estimators=10)\n",
    "    }\n",
    "    \n",
    "    # this is the actual pipeline object\n",
    "    pipe = Pipeline([\n",
    "        (transformer, eval(f'trans_dict[\"{transformer}\"]')),\n",
    "        ('dense', DenseTransformer()),\n",
    "        (estimator, eval(f'est_dict[\"{estimator}\"]'))\n",
    "    ])\n",
    "    \n",
    "    return pipe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def fit_grid_and_track(pipe, split, feature, params, twox, score_df):\n",
    "    \n",
    "    t = time.time()\n",
    "    \n",
    "    x_train, x_test, y_train, y_test = split\n",
    "    \n",
    "    # this condition allows us to calculate top features for the 0 class\n",
    "    if not twox:\n",
    "        y_train = 1 - y_train\n",
    "        y_test = 1 - y_test\n",
    "    \n",
    "    # instantiate gridsearch.\n",
    "    grid = GridSearchCV(pipe, params, cv=5, n_jobs=4)\n",
    "    grid.fit(x_train[feature], y_train)\n",
    "    \n",
    "    best = grid.best_estimator_\n",
    "    bestparams = grid.best_params_\n",
    "    train_score = best.score(x_train[feature], y_train)\n",
    "    test_score = best.score(x_test[feature], y_test)\n",
    "    \n",
    "    t = time.time() - t\n",
    "    \n",
    "    # only print out stats for the 1 class so we can see what's going on\n",
    "    if twox:\n",
    "        \n",
    "        results = {}\n",
    "        \n",
    "        # store each results dictionary into our score DataFrame\n",
    "        pipe_name = '-'.join([step for step in best.named_steps if step != 'dense'])\n",
    "        results['pipeline'] = pipe_name\n",
    "        results['feature'] = feature\n",
    "        results['params'] = bestparams\n",
    "        results['test'] = test_score\n",
    "        results['overfit'] = train_score - test_score\n",
    "        \n",
    "        score_df = score_df.append(results, ignore_index=True)\n",
    "        print(f'{pipe_name}: {feature}  -- {t}')\n",
    "        print(f'train score: {train_score}')\n",
    "        print(f'test score: {test_score}')\n",
    "        print('--------------------------')\n",
    "        print()\n",
    "        \n",
    "    return (best, score_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# this function puts everything together. \n",
    "# still unsure if it's more clear like this or as one gigantic function.\n",
    "# you can be the judge\n",
    "\n",
    "def pipe_gridder(transformer, estimator, x, y, feature, params, scores):\n",
    "    \n",
    "    split = train_test_split(x, y, stratify=y, random_state=111)\n",
    "    \n",
    "    pipe = pipeliner(transformer, estimator)\n",
    "    \n",
    "    return fit_grid_and_track(pipe, split, feature, params, 1, scores)[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [],
   "source": [
    "def capstone_deadline():\n",
    "    from datetime import datetime\n",
    "    from IPython.display import Image, display, HTML    \n",
    "    \n",
    "    due_date = datetime(2019, 12, 3, 23, 59)\n",
    "    time_left = due_date - datetime.now()\n",
    "    hours_left = round(time_left.total_seconds()/3600,2)\n",
    "    deadline = due_date.strftime('%B %-d, %Y - %I:%M %p')\n",
    "    print(f'You have {hours_left} hours left until {deadline}')\n",
    "    print(\"Good Luck!\")\n",
    "    display(Image('https://media.giphy.com/media/F4a3ugD96Xl28/giphy.gif'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "You have 626.82 hours left until December 3, 2019 - 11:59 PM\n",
      "Good Luck!\n"
     ]
    },
    {
     "data": {
      "image/gif": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "capstone_deadline()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Trying different models\n",
    " - linear regression\n",
    " - knn regression\n",
    " - decision tree\n",
    " - bagging\n",
    " - random_forest\n",
    " - ada boost\n",
    " - support_vector"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "|model|RMSE Train|RMSE Test|R2 Train|R2 Test|pipeline R2 test|\n",
    "| --- | --- | --- | --- | ---|---|\n",
    "|Linear Regression| 0.414|0.44|0.54|0.56|0.43|\n",
    "|KNN Regression| 0.39|0.51|0.56|0.31|0.33|\n",
    "|Decision Tree Regression| 8.11e05|0.56|0.99|0.16|0.03|\n",
    "|Baggin Regression| 0.19|0.46|0.89|0.45|0.54|\n",
    "|Random Forest Regression| 0.17|0.42|0.92|0.53|0.536|\n",
    "|Adaboost Regression| 0.29|0.42|0.77|0.53|0.543|\n",
    "|support vector Regression| 0.17|0.62|0.92|-0.02|-0.02|"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}