Yelp Dataset Exploration2.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "# package imports\n",
    "#basics\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import ast\n",
    "\n",
    "#misc\n",
    "import gc\n",
    "import time\n",
    "import warnings\n",
    "\n",
    "\n",
    "#viz\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns \n",
    "import matplotlib.gridspec as gridspec \n",
    "import matplotlib.gridspec as gridspec \n",
    "\n",
    "#settings\n",
    "start_time=time.time()\n",
    "color = sns.color_palette()\n",
    "sns.set_style(\"dark\")\n",
    "import re\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "from sklearn.linear_model import ElasticNet, Lasso,  BayesianRidge, LassoLarsIC\n",
    "from sklearn.ensemble import RandomForestRegressor,  GradientBoostingRegressor\n",
    "\n",
    "from sklearn.pipeline import make_pipeline\n",
    "from sklearn.preprocessing import RobustScaler\n",
    "from sklearn.base import BaseEstimator, TransformerMixin, RegressorMixin, clone\n",
    "from sklearn.model_selection import KFold, cross_val_score, train_test_split\n",
    "from sklearn.metrics import mean_squared_error\n",
    "import xgboost as xgb\n",
    "from xgboost.sklearn import XGBRegressor\n",
    "from sklearn.grid_search import GridSearchCV\n",
    "\n",
    "from matplotlib.pylab import rcParams\n",
    "rcParams['figure.figsize'] = 12, 4\n",
    "from sklearn.metrics import mean_absolute_error\n",
    "from sklearn.metrics import fbeta_score, make_scorer\n",
    "from sklearn.metrics import mean_squared_error\n",
    "from math import sqrt\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Took 2.258465051651001 s\n"
     ]
    }
   ],
   "source": [
    "#importing business dataset\n",
    "business=pd.read_csv(\"yelp_academic_dataset_business.csv\")\n",
    "end_time=time.time()\n",
    "print(\"Took\",end_time-start_time,\"s\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>address</th>\n",
       "      <th>attributes</th>\n",
       "      <th>attributes.AcceptsInsurance</th>\n",
       "      <th>attributes.AgesAllowed</th>\n",
       "      <th>attributes.Alcohol</th>\n",
       "      <th>attributes.Ambience</th>\n",
       "      <th>attributes.BYOB</th>\n",
       "      <th>attributes.BYOBCorkage</th>\n",
       "      <th>attributes.BestNights</th>\n",
       "      <th>attributes.BikeParking</th>\n",
       "      <th>...</th>\n",
       "      <th>hours.Wednesday</th>\n",
       "      <th>is_open</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>name</th>\n",
       "      <th>neighborhood</th>\n",
       "      <th>postal_code</th>\n",
       "      <th>review_count</th>\n",
       "      <th>stars</th>\n",
       "      <th>state</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1314 44 Avenue NE</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>11:0-21:0</td>\n",
       "      <td>1</td>\n",
       "      <td>51.091813</td>\n",
       "      <td>-114.031675</td>\n",
       "      <td>Minhas Micro Brewery</td>\n",
       "      <td>NaN</td>\n",
       "      <td>T2E 6L6</td>\n",
       "      <td>24</td>\n",
       "      <td>4.0</td>\n",
       "      <td>AB</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>none</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>35.960734</td>\n",
       "      <td>-114.939821</td>\n",
       "      <td>CK'S BBQ &amp; Catering</td>\n",
       "      <td>NaN</td>\n",
       "      <td>89002</td>\n",
       "      <td>3</td>\n",
       "      <td>4.5</td>\n",
       "      <td>NV</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1335 rue Beaubien E</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>beer_and_wine</td>\n",
       "      <td>{'romantic': False, 'intimate': False, 'classy...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>True</td>\n",
       "      <td>...</td>\n",
       "      <td>10:0-22:0</td>\n",
       "      <td>0</td>\n",
       "      <td>45.540503</td>\n",
       "      <td>-73.599300</td>\n",
       "      <td>La Bastringue</td>\n",
       "      <td>Rosemont-La Petite-Patrie</td>\n",
       "      <td>H2G 1K7</td>\n",
       "      <td>5</td>\n",
       "      <td>4.0</td>\n",
       "      <td>QC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>211 W Monroe St</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>33.449999</td>\n",
       "      <td>-112.076979</td>\n",
       "      <td>Geico Insurance</td>\n",
       "      <td>NaN</td>\n",
       "      <td>85003</td>\n",
       "      <td>8</td>\n",
       "      <td>1.5</td>\n",
       "      <td>AZ</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2005 Alyth Place SE</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>8:0-17:0</td>\n",
       "      <td>1</td>\n",
       "      <td>51.035591</td>\n",
       "      <td>-114.027366</td>\n",
       "      <td>Action Engine</td>\n",
       "      <td>NaN</td>\n",
       "      <td>T2H 0N5</td>\n",
       "      <td>4</td>\n",
       "      <td>2.0</td>\n",
       "      <td>AB</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 61 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               address  attributes attributes.AcceptsInsurance  \\\n",
       "0    1314 44 Avenue NE         NaN                         NaN   \n",
       "1                  NaN         NaN                         NaN   \n",
       "2  1335 rue Beaubien E         NaN                         NaN   \n",
       "3      211 W Monroe St         NaN                         NaN   \n",
       "4  2005 Alyth Place SE         NaN                         NaN   \n",
       "\n",
       "  attributes.AgesAllowed attributes.Alcohol  \\\n",
       "0                    NaN                NaN   \n",
       "1                    NaN               none   \n",
       "2                    NaN      beer_and_wine   \n",
       "3                    NaN                NaN   \n",
       "4                    NaN                NaN   \n",
       "\n",
       "                                 attributes.Ambience attributes.BYOB  \\\n",
       "0                                                NaN             NaN   \n",
       "1                                                NaN             NaN   \n",
       "2  {'romantic': False, 'intimate': False, 'classy...             NaN   \n",
       "3                                                NaN             NaN   \n",
       "4                                                NaN             NaN   \n",
       "\n",
       "  attributes.BYOBCorkage attributes.BestNights attributes.BikeParking  ...   \\\n",
       "0                    NaN                   NaN                  False  ...    \n",
       "1                    NaN                   NaN                  False  ...    \n",
       "2                    NaN                   NaN                   True  ...    \n",
       "3                    NaN                   NaN                    NaN  ...    \n",
       "4                    NaN                   NaN                    NaN  ...    \n",
       "\n",
       "  hours.Wednesday is_open   latitude   longitude                  name  \\\n",
       "0       11:0-21:0       1  51.091813 -114.031675  Minhas Micro Brewery   \n",
       "1             NaN       0  35.960734 -114.939821   CK'S BBQ & Catering   \n",
       "2       10:0-22:0       0  45.540503  -73.599300         La Bastringue   \n",
       "3             NaN       1  33.449999 -112.076979       Geico Insurance   \n",
       "4        8:0-17:0       1  51.035591 -114.027366         Action Engine   \n",
       "\n",
       "                neighborhood postal_code review_count stars state  \n",
       "0                        NaN     T2E 6L6           24   4.0    AB  \n",
       "1                        NaN       89002            3   4.5    NV  \n",
       "2  Rosemont-La Petite-Patrie     H2G 1K7            5   4.0    QC  \n",
       "3                        NaN       85003            8   1.5    AZ  \n",
       "4                        NaN     T2H 0N5            4   2.0    AB  \n",
       "\n",
       "[5 rows x 61 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#take a peak\n",
    "business.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 188593 entries, 0 to 188592\n",
      "Data columns (total 61 columns):\n",
      "address                                  180970 non-null object\n",
      "attributes                               0 non-null float64\n",
      "attributes.AcceptsInsurance              11671 non-null object\n",
      "attributes.AgesAllowed                   397 non-null object\n",
      "attributes.Alcohol                       47892 non-null object\n",
      "attributes.Ambience                      47577 non-null object\n",
      "attributes.BYOB                          911 non-null object\n",
      "attributes.BYOBCorkage                   1409 non-null object\n",
      "attributes.BestNights                    6844 non-null object\n",
      "attributes.BikeParking                   84891 non-null object\n",
      "attributes.BusinessAcceptsBitcoin        12674 non-null object\n",
      "attributes.BusinessAcceptsCreditCards    140391 non-null object\n",
      "attributes.BusinessParking               103424 non-null object\n",
      "attributes.ByAppointmentOnly             45423 non-null object\n",
      "attributes.Caters                        40038 non-null object\n",
      "attributes.CoatCheck                     8531 non-null object\n",
      "attributes.Corkage                       657 non-null object\n",
      "attributes.DietaryRestrictions           138 non-null object\n",
      "attributes.DogsAllowed                   13681 non-null object\n",
      "attributes.DriveThru                     6754 non-null object\n",
      "attributes.GoodForDancing                9162 non-null object\n",
      "attributes.GoodForKids                   64931 non-null object\n",
      "attributes.GoodForMeal                   47483 non-null object\n",
      "attributes.HairSpecializesIn             1881 non-null object\n",
      "attributes.HappyHour                     9285 non-null object\n",
      "attributes.HasTV                         47533 non-null object\n",
      "attributes.Music                         8807 non-null object\n",
      "attributes.NoiseLevel                    43710 non-null object\n",
      "attributes.Open24Hours                   352 non-null object\n",
      "attributes.OutdoorSeating                54181 non-null object\n",
      "attributes.RestaurantsAttire             48182 non-null object\n",
      "attributes.RestaurantsCounterService     397 non-null object\n",
      "attributes.RestaurantsDelivery           51668 non-null object\n",
      "attributes.RestaurantsGoodForGroups      53839 non-null object\n",
      "attributes.RestaurantsPriceRange2        107120 non-null float64\n",
      "attributes.RestaurantsReservations       51363 non-null object\n",
      "attributes.RestaurantsTableService       43325 non-null object\n",
      "attributes.RestaurantsTakeOut            61206 non-null object\n",
      "attributes.Smoking                       8113 non-null object\n",
      "attributes.WheelchairAccessible          52023 non-null object\n",
      "attributes.WiFi                          49026 non-null object\n",
      "business_id                              188593 non-null object\n",
      "categories                               188052 non-null object\n",
      "city                                     188583 non-null object\n",
      "hours                                    0 non-null float64\n",
      "hours.Friday                             141796 non-null object\n",
      "hours.Monday                             132761 non-null object\n",
      "hours.Saturday                           125376 non-null object\n",
      "hours.Sunday                             93387 non-null object\n",
      "hours.Thursday                           142359 non-null object\n",
      "hours.Tuesday                            140607 non-null object\n",
      "hours.Wednesday                          141843 non-null object\n",
      "is_open                                  188593 non-null int64\n",
      "latitude                                 188587 non-null float64\n",
      "longitude                                188587 non-null float64\n",
      "name                                     188593 non-null object\n",
      "neighborhood                             68655 non-null object\n",
      "postal_code                              187912 non-null object\n",
      "review_count                             188593 non-null int64\n",
      "stars                                    188593 non-null float64\n",
      "state                                    188593 non-null object\n",
      "dtypes: float64(6), int64(2), object(53)\n",
      "memory usage: 87.8+ MB\n"
     ]
    }
   ],
   "source": [
    "business.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfkAAAEQCAYAAACz/Q2uAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3XucjnX+x/HXHDCGQWSTRKH9VEpFxTqlLZ1sOm5tBytRESHCJhYtKpUSEkpatdsBhdpQbbU6yE501qfUhgo/yZkwh98f1zVjGveMe5h7xtzez8fDY+77e32v6/5876vdz/29ru/1/SZkZ2cjIiIi8SextAMQERGR2FCSFxERiVNK8iIiInFKSV5ERCROKcmLiIjEKSV5ERGROJVc2gGIxCszaw7cA9Qg+EG9CrjD3T8Pty8ArnX3nw7gM6YB7YB1YVEiUBl4zN1H72PfY4EH3P0KM6sNzHD3FvsbS75jvwXUAzaFReWAV4C/ufsWMzsd+Iu7X1nIMc4Aurh7twjbcvcPv4PP3P2BIsaY+/2b2b8Izs0XRTmGyMFOSV4kBsysAvAycJ67LwnLrgdeNbNj3T2TIDkXh4fyJjgzqwssM7M57v5lIfvVAwzA3X8EiiXB59Hf3WeEMZUDHgH+AVzs7ulAgQk+1AioE2lDlPvvS+737+4XHeCxRA5KSvIisZEKVCPoVed4BtgMJJnZ42HZm2Z2EXAKMAgoD/wGeMrdh5hZW2AssC081hnuvnMfn10HSAC2AJjZIOASoCJQCbgDmAM8DhxlZvOBWwh6w5XNbBhwDHAkwQ+BH4Dr3X112LueGMb5Tbi9r7u/VVhA7r7bzPoCa8zseKAWMN7dTzKzVsAYIAnIJrj6sRi4G6hqZk8CT+X7HvoDD7r7SeFHtDKzK4EqwAKCXnmGmWUDNXOuluS8B+7P9/0vBK5093QzuxnoBWQCa4Ge7v5VeMVgM3AycDTwCfBnd9+6j/MhUmp0T14kBtx9AzAAmGdm35rZdKAz8Lq773L3zmHVs4HvgX5AJ3c/HWgO3Glmh4d1TgKucffGBST4283so/Bzfgo/t727/2Bm9YBzgbbu3hi4C7g7vJLQFfjG3c+PcMzWwB/d/XiCxNrNzJKBWcCQ8FiPAKcW4TvZAXxFkCTzGg6McfemwI3A7919FfBXYGGe7yr3ewDyfw91gHPCeE4BbtpHLLnff/hZAJjZ7wm+v7Pd/RSCKw8vmVlCWKUpcAFwAsEPoT9G0XSRUqMkLxIj7j4GOIKgV7gaGAgsNbOq+eplAxcDTc1sKEGvNoGg1w2wyt1XFPJRD7n7qQTJcxFBAnwrPPYK4M/AdWZ2L9CNX19dKMhb7r45fL0UqB4eH3d/Nfz7JvBZFMfKKxvYnq/seWCCmT1DkEQHFbBvYd/DdHff5u67gKfZ/1shFwDPufs6AHefBhxFkNAB5rn7TnffDXxK8L2IHLSU5EViwMxamll/d9/i7i+7+wCCe8zZ5EtAZlaJIJE2AZYQXIreTZDoAaK6HOzu24COwFnA7eGxmwDvs+cy9n15jluYHXleZ4f7ZETYNzOa2MJYUgl6wJ/ni3sSwQ+I14DzgU/MLCXCIQr7HvLGkUjw/eVICD+/fBRh5twyyCuBYOAgRP5eRA5aSvIisbEOGBzeb85xJFCVoAcIQWIqBxxHkIQHu/tcoC1QgSDhFEl4m6AfMNzMjgLaAOnhVYW3gUvzHDeDPckrGsuAnWZ2AYCZnUmQnPe5ypWZVQQeBl519+/ybXsPOC3sNd9MMJahVhHj+5OZVQh/HHQCXg3L1wGnh6+vzbdPzvef17zwWDXD2DoD64HlUcYhclBRkheJAXf/iiChjgrvlX9BcFm6s7t7WO0FgsSbRTAS/0szW0Zw6f4LoOF+fvYzwIfAA8A/gcPD435B0BuubmZp4ftfzGwxUfRI3T0DuAIYZmZLCX5MrGHvy+857g/HCiwBPgg/u1OEegOAu8NjvgUMD38ILALqm9msKJr9P4LBc0uB/xAM1IPgVsmEMIYTCG6b5HgBeNvMcgbv4e6vAQ8B/zazz8N4/+DuWVHEIHLQSdBSsyISLTO7n+DZ+rVmdjTwMVDf3TeWcmgiEoEeoRORolgBvGFmOWMGuirBixy81JMXERGJU7onLyIiEqeU5EVEROKUkryIiEiciruBd+vWbdEgAxEROaTUrJkW8TFY9eRFRETiVNz15EVE4klmZib33TeCVatWkJiYxKBBQ9m6dSsDB95OnTpHA3DZZVdyzjnnAfD996u4885+TJ/+PABr1qzhnnvuJjMzA4ABAwZRt+4xvPXWGzz99FMkJECHDpdz8cWXlk4DJaaU5EVEDmLvvrsQgIkTp7JkSTrjxo2hZcs2XH31dVxzzfW/qjtv3iu88MKzbNq0Kbfs8ccncsUVV9GmTVs++OB9HntsAn/727089th4Hn98OhUrVuT66/9I69ZtqVatWom2TWJPSV5E5CDWpk1bWrQIlkBYu3YNhx1WA/dlrFy5gnfeeZs6dY6md+9+pKZWIi2tCuPHT+bqq/f0ynv2vJ3KlYOFBzMzMylfvjxJSUk8/fQLJCcns2HDz2RnQ8WKFUulfRJbuicvInKQS05OZsSIoTz00P2cffY5nHBCI269tTcTJkyhdu2jmDp1CgAtW7beK1lXq1aN5ORkVq78jgkTHubGG2/KPebbb/+bG264hlNPPY3kZPX54pGSvIhIGTB48HD++c+Z3HffCM48sznHH38CAG3anM3XX3uh+y5Zks6dd97BkCF3U7fuMbnlZ531e1588VV2797NvHmvxDJ8KSVK8iIiB7F5815h+vQnAUhJSSExMZFBg/rzxRefAfDhh4sxO77A/ZcsSWfs2Ad48MFxHH/8iQBs27aVnj1vZteuXSQmJlKxYkUSE5UO4lHM5q43syRgCmAE6zZ3JlhLey7wdVhtors/Z2ZDgfYE60f3cffFZtYQmEawVvVnQA93z4pUN+/n6jl5EckRaWT6rl27GD16JJBNgwa/5fbb+4f3qKfx+usLqFSpEtde+2datmzN2LEP5vaSf/55PZUrpzF58jQAsrKy6N+/D61bt+HSS6+MWRt27NjBqFHD+fnn9WRkZHD99Z34zW9q8dBDo0lOTqZGjRoMGHAXlSpVzt2nQ4fzmTNnPgCdOl3D7t27qF69BgB169ZjwIC7mD17Fi+/PJvk5GQaNDgu93uQsqmg5+RjmeQvBTq4+41m1ha4nSDBV3X3B/PUa0Kw7vU5wNHATHc/w8zmAGPc/S0zewyYT7AC1l51836ukryI5PjPf97inXfeZtCgoSxZks7zz/+DhIQErr76Ok49tQkjRw6jVauzqFPnaO6+e0huAu/e/UYeffQJUlJSAMjIyKB79y785S9DaNCgIQCTJk0gPX0x7dtfHNMkLxKNgpJ8zEZauPtLZvZy+LYesBZoCpiZXULQm+8DtAIWuHs2sNLMks2sZlj37XD/V4HzAI9U193XxaodIlJ2RRqZfscdfyEpKYndu3ezfv16qlevznff/Y/TTmtKhQoVAKhTpy7Ll3/NSSedDMCMGc9y5pnNcxP8m2++TkJCAs2btyi2WDfO+rzYjhUL1S5vVNohyH6I6U0Yd88ws6eAccAMYDHQ393bAN8CQ4EqwKY8u20huKyfECbzvGUF1RURiSj/yPSkpCTWrFlNx45XsWnTRurWrUeDBg35+OMlbN++jU2bNvLZZ5/wyy87ANi9ezezZ8/imms6AvDtt8t57bX5dO3arTSbJRKVmD8z4e6dzGwg8AHQwt1/CDe9SJD8ZwNpeXZJAzYCWRHKNhdQV0SkQIMHD2f9+p+4+eYbePrpF6hV60ieffZF5s59iXHjHmLw4OFcccVV9OvXizp1jubEExtRtWowMUx6+gecemqT3GfN5817hXXr/o9evbqxZs1qkpPLUatW7WLt1YsUl5j15M2so5ndGb7dTpC0Z5nZmWHZOcCHwLvA+WaWaGZ1gUR3/wlYGt7LB7gQWFhIXRGRvUQemX4Hq1atBCA1NZXExEQ2bNjAxo0bmTjxCXr3voO1a9dSv34DANLTF/8qgd96a2+mTHmK8eMnc+GFf+BPf7pWCV4OWrHsyc8CnjSz/wDlCO6/rwLGm9kuYA1ws7tvNrOFwPsEPzp6hPv3A6aYWXlgGTDD3TMLqCsispezzvo9o0YNp0ePm8jIyKBXr75Uq3YYo0YNIzm5HCkpKQwcOIRq1arx448/0LXrnylXLpkePXrnjjRfuXIFF1zQvpRbIrJ/Yja6vrRodL2IlEUaeCcHQkvNioiIHGI0WbGIlGmzHhxe2iEU6vJ+Q0s7BDmEqScvIiISp5TkRURE4pSSvIiISJxSkhcREYlTSvIiIiJxSkleREQkTinJi4iIxCkleRERkTilyXBERESikJmZyX33jWDVqhUkJiYxaNBQsrOzGTlyGAkJCdSv34C+fQeSmJjIww8/wKeffkzFihXp3r0XjRqdlHucBQvmMXPmc0ya9GRu2YYNG+je/UaeeupZKlSoUGwxK8mLiIhE4d13FwIwceJUlixJZ9y4MWRnZ3PTTd1p0uR07r9/FAsXvk1ycjIrV65gypSn2Lx5M/363cYTT0wH4OuvnVdemU3edWM++OB9HntsHD///HOxx6zL9SIiIlFo06YtAwbcBcDatWs47LAauH/Jaac1BaB58xakpy/mu+++pVmz5iQmJlKtWjUSExNZv/4nNm3ayMSJ4+nVq9+vjpuYmMDDDz9KlSpVij1mJXkREZEoJScnM2LEUB566H7OPvscsrOzSUgIFoBLTa3Etm1bOe4444MP3icjI4Mffvie7777ll9++YV77/0bvXr1JTU19VfHPOOM5lStWi028cbkqCIiInFq8ODhrF//EzfffAM7d+7MLd++fRuVK1fmzDObs2zZ5/Tq1Y2GDY/D7AQ2bdrIqlWreOCBe9i1axffffc/xo59kN69+xXySQdOPXkREZEozJv3CtOnB4PlUlJSSExM5PjjT2DJknQAFi16j1NOOY2VK1dw2GHVefTRx7nuuk4kJCRw4okn8fTTzzN+/GSGDx/FMcccG/MED+rJi4iIROWss37PqFHD6dHjJjIyMujVqy/16h3L6NEjmTRpAvXqHUPbtueQkZHBBx+8x8svz6Z8+fL07Tuw1GJOyDvCLx6sW7clvhokIoWKl/XkN876PMaRHJhqlzcq7RCkEDVrpiVEKtflehERkTily/UiIiJ5pKd3L+0QCnX66ROjrquevIiISJyKWU/ezJKAKYABmUBnIAGYBmQDnwE93D3LzIYC7YEMoI+7LzazhtHWjVUbREREyrJY9uQvBnD3lsBfgTHhv8Hu3pog4V9iZk2As4BmwJ+ACeH+RakrIiIi+cQsybv7S8DN4dt6wFqgKfB2WPYqcC7QCljg7tnuvhJINrOaRawrIiIi+cT0nry7Z5jZU8A4YAaQ4O45j7htAaoCVYBNeXbLKS9KXREREckn5gPv3L0T8FuC+/MV82xKAzYCm8PX+cuzilBXRERE8olZkjezjmZ2Z/h2O0HSTjeztmHZhcBC4F3gfDNLNLO6QKK7/wQsLUJdERERySeWz8nPAp40s/8A5YA+wDJgipmVD1/PcPdMM1sIvE/wo6NHuH+/ItQVERGRfGKW5N19G3BVhE1nRag7DBiWr+yraOuKiIjI3jQZjoiISJxSkhcREYlTSvIiIiJxSkleREQkTinJi4iIxCkleRERkTil9eRFRCTmMjIyuOee4axevZrdu3fRqVMXGjU6mfvuG8GWLVvIyspk8OC72b59G2PHPpi73xdffMaoUQ9w/PEnMnz4XezcuZPDD6/JoEFDWbVqRcS6zZu3KI0mHpSU5EVEJObmz/8XVapUY8iQv7Fp00Y6d76Opk3PoF27CznnnHYsWZLOihXf0aJFK8aPnwzAv//9OocfXpPmzVvw8MP3067dBVx00cVMnz6N2bNncvXV10WsK3vocr2IiMTc2Wefy003dct9n5SUzKeffsy6dWvp3ftWFix4ldNOa5q7fceOHUydOok+ffoD8MknH9Gs2e8AaN68BenpiwusK3soyYuISMylpqaSmlqJ7du3MXjwQG66qTurV/9IWloVxo59lCOOqMUzzzyVW//ll2dz9tnnUq1aNQC2bdtG5cqVc4+1devWAuvKHkryIiJSItauXcNtt3Xj/PMv4rzzLqBq1Wq0atUGgJYtW/Pll1/k1l2w4FX+8IdLct9XqlSJ7du3A7B9+3bS0tIKrCt7KMmLiEjM/fzzevr27Un37rflJuTGjU/h/fffBeCjj5Zy7LENANi6dSu7d+/miCNq5e5/8sl76i5a9B6NG59aYF3ZQwPvREQk5v7+9yfZsmUL06Y9zrRpjwMwePBw7r33b7z00kwqVarM0KEjAFi1agVHHnnkr/bv1KkLI0YMY+7cF6latRpDh44ssK7skZCdnV3aMRSrdeu2xFeDRKRQsx4cXtohFOryfkOjqrdx1ucxjuTAVLu8UWmHUGLS07uXdgiFOv30iXuV1ayZlhCpri7Xi4iIxCldrhcRkWLx/PNPl3YIhbrqqutLO4QSp568iIhInFKSFxERiVO6XC8ie4k0z3irVmcBsGDBPGbOfI5Jk57k66896nnGU1JSAMjKyqJ//z60bt2GSy+9slTaJ3KoUJIXkb1Emme8Vauz+Ppr55VXZpPzVM5xx1mR5hkHmDJlIps3byq1tokcSnS5XkT2Emme8U2bNjJx4nh69eq3V/1o5xl/883XSUhI0CIiIiVESV5E9rL3POPduPfev9GrV19SU1P3qh/NPOPffruc116bT9eu3fbaX0RiIyaX682sHDAVOAaoAIwAvgfmAl+H1Sa6+3NmNhRoD2QAfdx9sZk1BKYB2cBnQA93z4pUNxbxi0gwz/igQf257LIrqVPnaFatWsUDD9zDrl27+O67/zF27IP07h306hcseJURI+7L3TdnnvEKFVJy5xmfN+8V1q37P3r16saaNatJTi5HrVq11asXiaEiJ3kzK+fuu/dR7Xpgvbt3NLMawFLgbmCMu+eO0jGzJsBZQDPgaGAmcAYwBhjs7m+Z2WPAJWa2ooC6IlLMcuYZv/32AZx++pkAPP308wCsXv0jQ4cOyk3whc0zftFFF+fOM3799Tfkbn/iiUnUqFFDCV4kxvaZ5M2sFdAWGA0sBE42s87u/lwhu70AzMjzPgNoGhzOLiHozfcBWgEL3D0bWGlmyWZWM6z7drjvq8B5gEeq6+7rom+uiEQj0jzjDz74CBUqpOxVtyjzjItIyYqmJ38/MAS4FFgPnAg8DxSY5N19K4CZpREk+8EEl+0fd/cPzewuYCiwMTxmji1AVSAhTOZ5y6oUUFdJXqSY9elzB3363BFx25FH1mby5Gm57084oRH33PPgr+pUr16DMWPGFXj8Ll1uKZY4RaRw0Qy8S3L314F2wEvu/h2QtK+dzOxo4E1gurv/A3jR3T8MN78InAZsBtLy7JZGkPizIpQVVFdEREQiiKYnn2RmZxIMeBtlZicB5QrbwcyOABYAPd39jbB4vpndFg6WOwf4EHgXGG1mDwB1gER3/8nMlppZW3d/C7iQ4MfC8kh1i9pgEdljwbT00g6hQOfdcHpphyBS5kWT5EcC/wCecPf/mdn/gN772GcQcBgwxMyGhGV9gYfNbBewBrjZ3Teb2ULgfYKrCj3Cuv2AKWZWHlgGzHD3zALqioiISAT7TPLuPguYlaeoobtn7mOf3kT+IbDXUFp3HwYMy1f2FcFI+n3WFRERkciiGV1fC3gCOA5oDfzdzG5w99WxDk5ERET2XzQD7x4FXgJ2AD8DHwGPxzIoEREROXDRJPlj3H0KkOXuu919IFA3xnGJiIjIAYomyWeZWW698Nl3zXkvIiJykIsmWc8CngGqmtktwL8JJsMRERGRg9g+k7y7jwL+BfyXYEKcyQTz0IuIiMhBLKoFatx9OjDdzI4C6ueZclZEREQOUtE8Qted4NG5XgSz1G0ys1nufmesgxMREZH9F809+S7A7cAfgTlAI4LL9iIiInIQiybJZ7v7WuBc4HV3zyCKBWpERESkdEWT5Hea2QCCaWZfCy/fb4ttWCIiInKgor1c/1ugk7tvAFoBXWMalYiIiBywaBaoccKkbma1gcfc/ctYByYiIiIHpqij65eg0fUiIiJlQlFH189Go+tFRETKhKKOrn9Do+tFRETKBo2uFxERiVMaXS8iIhKnolmgxoHbgJVmlgDcpNH1IiIiB799Jnkzaw58A7wCHAWsMrMWsQ5MREREDkw0l+vvJxh0t97dvwc6AmNjGpWIiIgcsGiWmk119y/MDAB3/5eZjSxsBzMrB0wFjgEqACOAL4BpQDbwGdDD3bPMbCjQHsgA+rj7YjNrGG3dojVXRETk0BFNT363mR1GkHCxnGxfuOsJev6tgQuB8cAYYHBYlgBcYmZNCEbtNwP+BEwI9y9KXREREYkgmiQ/AngbqGNm/wTeC8sK8wIwJM/7DKBpeByAVwluAbQCFrh7truvBJLNrGYR64qIiEgE0cxd/7KZfUkwy10ScLe7L9vHPlsBzCwNmAEMBh5w9+ywyhagKlAFWJ9n15zyhCLUXbevNoiIiByKounJA+wGFhH04iuGl84LZWZHA28C0939H0BWns1pwEZgc/g6f3lR6oqIiEgE0TxCdzewDJgFzAz/zdjHPkcAC4CB7j41LF5qZm3D1xcCC4F3gfPNLNHM6gKJ7v5TEeuKiIhIBNGMru8INHT3H4tw3EHAYcAQM8u5N98beMTMyhP8aJjh7plmthB4n+AHR4+wbj9gSpR1RUREJIJokvyqIiZ43L03QVLP76wIdYcBw/KVfRVtXREREYksmiT/hpmNJlhmdkdOobsviVlUImXc559/xsSJjzB+/GSGDr2T9euDMaNr1qymUaOTGD78HsaNG8Mnn3xEQkIiPXv2oXHjU9m4cSPDh9/Fzp07OfzwmgwaNJSUlBT+8Y/pvP76fBITE+nYsTNnnXV2KbdQRMqCaJL8DeHfP+YpywbqF3s0InHgmWeeYv78f5GSUhGA4cPvAWDz5s306tWN227rx9dff8Wnn37C5MlP8f33qxg6dBBTpz7NtGlTaNfuAi666GKmT5/G7NkzueiiDsyY8SzPPfcSO3bsoHPna5XkRSQq0SxQc2yEf0rwIgU46qg6jBx5/17lU6dO4sorr+Lwww+nZs3fkJKSwq5du9i2bRvJycHv7U8++YhmzX4HQPPmLUhPX0zFihWpVetIduzYwS+/7CAxMdqHYkTkUFdgT97MBrj7aDN7JNJ2d+8Vu7BEyq62bc9h9epfD2PZsOFn0tP/y2239QUgKSmJhIRErrvuSrZu3crAgXcBsG3bNipXrgxAamoqW7duBeA3vzmCjh3/SGZmFh073lByjRGRMq2wy/Wbwr/rC6kjIlF48803aNfufJKSkgCYN+8VatSowZgx49i+fTu33tqFk05qTKVKldi+fTsVKqSwfft20tLSWLToXdav/4nnn58DQL9+t3Hyyadw4oknlWaTRKQMKDDJu/uk8O/wnDIzqw00cPeFJRCbSNxIT19Mp05dct+npaVRsWJFkpKSSE1NpVy58uzYsZ2TTz6F999/l4suuphFi96jceNTSUurQoUKFShfvjwJCQlUrlw5t4cvIlKYfQ68M7NuQBugF7AE2GRms9z9zlgHJxIvVq5cQe3aR+W+b9fuAj799GO6dbuRzMxMzjvvAurWPYZOnbowYsQw5s59kapVqzF06EgqVqxIevpibr75BhITE2nc+FTOOKNZqbVFRMqOaEbXdyVY3vWPwBzgVoIpbpXkRQpw5JG1mTx5Wu77p59+/lfbk5KS6N9/0F77Va8eXMLPr0uXW+jS5ZZij1NE4ls0ST7b3dea2bnAc+6eYWZJsQ5M5GD27YA+pR1CoeqPfri0QxCRg0A0z+LsNLMBBDPQvWZm3YFtsQ1LREREDlQ0Sb4r8Fugk7tvIFjXvWtMoxIREZEDFs168l+SJ6m7+3UxjUhERESKRTSj6z8lmMb2V9y9cUwiEhERkWIRzcC7nnlelwf+BHwbm3BERESkuERzuf7tvO/N7HXgPWBkrIISERGRA7c/K13UAGoXdyAiIiJSvIp6Tz4BqAtMimVQIiIicuCKek8+G1jn7stiFI+IiIgUk2jWk38b+D/gZOAEYHesgxIREZEDt88kb2adgTeBM4HWwEIzuyLWgYmIiMiBieZyfV/gNHdfDWBmdYGXgZmxDExEREQOTDSj63flJHgAd1+JLtmLiIgc9ArsyZtZk/Dlx2Y2nmBEfSZwA/BuNAc3s2bAfe7eNjzeXODrcPNEd3/OzIYSLGWbAfRx98Vm1hCYRjDQ7zOgh7tnRapbpNaKiIgcQgq7XJ//cnz7PK+zgV6FHThcua4je1asawKMcfcH89RpQrC6XTPg6PAzzwDGAIPd/S0zewy4xMxWFFBXREREIigwybv7sQd47G+Ay4Hp4fumgJnZJQS9+T4EK9otcPdsYKWZJZtZzbBuzkx7rwLnAR6prruvO8A4RURE4tL+zHgXFXefya/v3S8G+rt7G4K574cCVYBNeepsAaoCCWEyz1tWUF0RERGJIGZJPoIX3f3DnNfAacBmIC1PnTRgI5AVoayguiIiIhJBgUk+HDRXnOab2Znh63OADwkG8J1vZonho3mJ7v4TsNTM2oZ1LwQWFlJXREREIiisJz8RwMzeKKbP6g48bGZvAS2BEWHPfiHwPsFAuh5h3X7AcDN7n2B52xmF1BUREZEIChtdX87MFgBNzGxO/o3u3mFfB3f374Dm4eslQIsIdYYBw/KVfUUwkn6fdUVERCSywpL8hcDvAUOz24mIiJQ5hT1C9z3wdzNbGT6vXg8o5+7LSy48ERER2V/RzF3/g5l9DtQGEs3sJ6C9u38Z29BERETkQETzCN04YLS7H+buVYERwKOxDUtEREQOVDRJ/gh3fyrnjbs/CdSMXUgiIiJSHKJJ8slmVj3njZkdTjB3vYiIiBzEorknPw5YZGbPEST3PwEPxTQqEREROWCo3897AAASaklEQVT77Mm7+2SgG8GkNKnAre4+MdaBiYiIyIGJpiePu/8b+HeMYxEREZFiVJIL1IiIiEgJUpIXERGJU/tM8mbWPdJrERERObgVeE/ezL4EPgBamtm7wDLgJsLV6UREROTgVlhPvjEwFagCDAU+BY4zs4fN7LKSCE5ERET2X2Gj6+u5+9tm9oO7XwFgZp8CbwKtgRdLIkA5tHz++WdMnPgI48dP5vvvVzFy5DASEhKoX78BffsOJDExkQkTxvLJJx+RmZlJhw6X0aHDZezYsYMHHriH1at/ZPfu3dx+e39OPPEkli37nHHjHiI7O5saNWowZMjfqFChQmk3U0SkRBSW5MeZWX2gmpkNBJYC2e4+G5hdItHJIeWZZ55i/vx/kZJSEYBx48Zw003dadLkdO6/fxQLF75NWloa33+/ikmTnmTXrl107HgVbduewwsv/JP69RswZMjdLF/+NcuXf8UJJzTivvtGMmLEfdSpczRz577E2rWrqVv3mNJtqIhICSnwcr27XwCcCGwBNgGXAQ3M7DMzm1RC8ckh5Kij6jBy5P25792/5LTTmgLQvHkL0tMX06jRydx5518BSEhIICsri+TkZBYvXkS5cuXo27cn06Y9TrNmv2PVqhVUrVqV55//Bz173szmzZuU4EXkkFLo6Hp3zwC+dPfH3L078BbQDHihBGKTQ0zbtueQnLzn4lJ2djYJCQkApKZWYtu2rVSoUIEqVaqQkZHBiBFD6dDhMlJTU9m0aSNbtmxhzJjxtGzZmvHjH2bjxo18+uknXHbZH3n44Uf58MP/kp6+uLSaJyJS4qKZ1rZ93tfuvs3dX49tWCKQmLjnP8/t27dRuXJlADZv3ky/frdx7LH16dixMwBVqlSlZcs2ALRs2Qb3ZVStWo06depw7LH1SU5Oplmz3+G+rOQbIiJSSjQZjhy0jjvOWLIkHYBFi97jlFNOY+fOX+jTpzvt23fghhu65tZt3PhUFi16F4CPP17CMcfUp3bto9ixYwfff78qLP+IY49tUPINEREpJVHNXS9SGnr27MPo0SOZNGkC9eodQ9u25zBjxrP8+OMPzJnzInPmBA94DBo0lD//uTP33juCW27pTHJyMoMHD6dcuXL85S9DGD78LrKz4aSTGtOiRatSbpWISMmJaZI3s2bAfe7e1swaAtMIlqv9DOjh7llmNhRoD2QAfdx9cVHqxjJ+KXlHHlmbyZOnAVC3bj3Gj5/8q+1XX30dV199XcR9R426f6+ypk3PYMqUvxd7nCIiZUHMLteb2QDgcSAlLBoDDHb31kACcImZNQHOIhjM9ydgwn7UFRERkQhi2ZP/BrgcmB6+bwq8Hb5+FTgPcGCBu2cDK80s2cxqFqWuu6+LYRukmPWbf1tph1CoB88fV9ohiIgUm5j15N19JrA7T1FCmKAhePa+KsGUuZvy1MkpL0pdERERiaAkR9dn5XmdBmwENoev85cXpa6IiIhEUJJJfqmZtQ1fXwgsBN4FzjezRDOrCyS6+09FrCsiIiIRlOQjdP2AKWZWnmDZ2hnunmlmC4H3CX5w9NiPuiIiIhJBTJO8u38HNA9ff0UwOj5/nWHAsHxlUdcVERGRyDTjnYiISJxSkhcREYlTSvIiIiJxSkleREQkTinJi4iIxCkleRERkTilJC8iIhKnlORFRETilJK8iIhInFKSFxERiVNK8iIiInFKSV5ERCROKcmLiIjEKSV5ERGROKUkLyIiEqeU5EVEROKUkryIiEicUpIXERGJU8mlHYDExoYNP9OlS0ceemgCU6dOYv369QCsWbOaRo1OYvjwe5gwYSyffPIRmZmZdOhwGR06XMbYsQ/y9dcOwM8/r6dy5TQmT55Wii0REZH9pSQfhzIyMhg9ehTly1cAYPjwewDYvHkzvXp147bb+rFkSTrff7+KSZOeZNeuXXTseBVt255D7979co/RvXsXBg4cXGrtEBGRA6Mkn0/eHnBWVhajR48EsmnQ4Lfcfnt/kpKSAMjKyqJ//z60bt2GSy+9kh07djB8+F1s3ryZlJSKDBlyN4cddliptGH8+Ie59NIrmD79yV+VT506iSuvvIrDDz+ctLQ0Gjb8LQAJCQlkZWWRnLznP4cZM57lzDOb06BBwxKNXUREio/uyeeRvwc8efIEbrmlBxMnTmXnzl94553/5NadMmUimzdvyn0/d+6LmJ3Ao48+zrnnnsdTTz1R4vED/Otfc6lWrRrNmv3uV+UbNvxMevp/ufDCiwGoUKECVapUISMjgxEjhtKhw2WkpqYCsHv3bmbPnsU113Qs8fhFRKT4lHhP3syWAjnZ8X/AJGAskAEscPfhZpYIPAqcAuwEurr7cjNrnr9uccaWvwc8YsRokpKS2L17N+vXr6d69eoAvPnm6yQkJNC8eYvcfa+66loyMzMBWLt2TW7dkvbKK3NISEggPX0xy5d/xYgRf+Xee8fw9ttv0q7d+blXIiC4fD9kyEBOO60pHTt2zi1PT/+AU09tQuXKlUujCSIiUkxKtCdvZikA7t42/NcZeAy4FmgFNDOzJsClQIq7/w74C/BgeIhIdYtFpB5wUlISa9aspmPHq9i0aSN169bj22+X89pr8+natdtex0hKSqJXr27MnPkcv/tdy+IKrUgmTJjC+PGTGT9+Mg0b/pbBg++mRo3DSU9fTPPme2LaufMX+vTpTvv2Hbjhhq6/OkZQt0X+Q4uISBlT0j35U4BUM1sQfvYwoIK7fwNgZvOBc4AjgXkA7r7IzE43syoF1F1SHIEV1AOuVetInn32RebOfYlx4x6ievXqrFv3f/Tq1Y01a1aTnFyOWrVq5ybFRx55jBUrvqN//948//zs4gitWKxcuYLatY/Kff/SSzP58ccfmDPnRebMeRGAQYOGUrv2UaxcuYILLmhfWqGKiEgxKekkvx14AHgcOA54FdiYZ/sWoD5QhT2X9AEyw7LNEeoWiwkTpuS+7tnzZvr3H8To0SPp2fN2jj66LqmpqSQmJnLrrb1z6z3xxCRq1KhB8+YtmD79SWrW/A0XXNCelJQUEhOTIn1MiRo/fnLu66effv5X266++jquvvq6iPvdf//YmMYlIiIlo6ST/FfAcnfPBr4ys01A3pvXaQRJPzV8nSORIMGnRagbM9dffwOjRg0jObkcKSkpDBw4pMC67dt3YMSIYbz88myysrIYNOivxR7PLzNuLPZjFqeUK6eWdggiIpJHSSf5G4GTgVvNrDZBMt9mZg2Ab4HzgeFAHeBi4PlwsN2n7r7ZzHZFqFvs8vaAJ04sOHF16XJL7uvq1WswZsy4WIQjIiKyX0o6yT8BTDOzd4BsgqSfBTwDJBGMmP/AzP4LtDOz94AEIGfod7f8dUs4fhERkTKjRJO8u+8iGB2fX/N89bIIEnr+/RflrxuN2556s6i7lKhxnc4u7RBERCQOaTIcERGROKUkLyIiEqeU5EVEROKUkryIiEicUpIXERGJU0ryIiIicUpJXkREJE4pyYuIiMQpJXkREZE4pSQvIiISp5TkRURE4pSSvIiISJxSkhcREYlTSvIiIiJxSkleREQkTinJi4iIxCkleRERkTilJC8iIhKnlORFRETilJK8iIhInFKSFxERiVPJpR1AUZlZIvAocAqwE+jq7stLNyoREZGDT1nsyV8KpLj774C/AA+WcjwiIiIHpbKY5FsB8wDcfRFweumGIyIicnBKyM7OLu0YisTMHgdmuvur4fuVQH13zyjdyERERA4uZbEnvxlIy/M+UQleRERkb2Uxyb8LXARgZs2BT0s3HBERkYNTmRtdD7wItDOz94AEoHMpxyMiInJQKnP35EuamTUD7nP3tvnKLwb+CmQAU919SimEF7VC2tEX6AKsC4tucXcv4fD2yczKAVOBY4AKwAh3n5Nn+0F/PqJoQ1k5F0nAFMCATKCzu3+TZ/tBfy4gqnaUifMBYGa/AT4E2rn7l3nKy8S5yFFIO8rEuTCzpcCm8O3/3L1znm03AbcQnIsR7v5yScRUFnvyJcbMBgAdgW35yssBDwFnhNveNbO57r6m5KPct4LaEWoC/NndPyzZqIrsemC9u3c0sxrAUmAOlKnzUWAbQmXlXFwM4O4tzawtMAa4BMrUuYBC2hEqE+cj/M4nATsilJeVc1FgO0IH/bkwsxSA/B2pcFstoBfB02ApwDtm9pq774x1XGXxnnxJ+ga4PEL5CcByd9/g7ruAd4DWJRpZ0RTUDoCmwJ1m9o6Z3VmCMRXVC8CQPO/zDrYsK+ejsDZAGTkX7v4ScHP4th6wNs/msnIu9tUOKCPnA3gAeAz4MV95mTkXoYLaAWXjXJwCpJrZAjP7dzhmLMeZwLvuvtPdNwHLgcYlEZSSfCHcfSawO8KmKuy5JAOwBahaIkHth0LaAfAs0A34PdDKzP5QYoEVgbtvdfctZpYGzAAG59lcJs7HPtoAZeRcALh7hpk9BYwjaEuOMnEuchTSDigD58PMbgDWufv8CJvLzLnYRzugDJwLYDvBD5XzCWJ9xsxyrpaX2rlQkt8/+R/jSwM2llIs+83MEoCH3f2n8Jf+K8BppRxWgczsaOBNYLq7/yPPpjJzPgpqQ1k7FwDu3gn4LTDFzCqFxWXmXOSI1I4ydD5uJBiI/BZwKvD38NIwlK1zUWA7ytC5+Ap42t2z3f0rYD1wZLit1M6F7snvn2XAcWZWHdgKtCH4BVfWVAE+M7MTCO7Z/Z5gYNhBx8yOABYAPd39jXyby8T52EcbytK56AjUcfd7CHovWQQD16CMnAvYZzvKxPlw9zY5r8ME2S3PPfcycy720Y4ycS4IfqicDNxqZrUJ4l4dblsMjAzv21cguJXyWUkEpSRfBGZ2LVDZ3SeHoz3nE1wNmeruP5RudNHL145BBD3LncAb7v6v0o2uQIOAw4AhZpZzX3sKUKkMnY99taGsnItZwJNm9h+gHNAHuNzMytr/NvbVjrJyPn5F/z9Vap4AppnZO0A2QdLvZWbL3X2OmT0CLCQ4F3e5+y8lEZQeoRMREYlTuicvIiISp5TkRURE4pSSvIiISJxSkhcREYlTSvIiIiJxSo/QicSBcArNe4AaBD/eVwF3uPvn4fYFwLXu/tMBfMY0oB17FglJBCoDj7n76H3seyzwgLtfET5DPMPdW+xvLCISHSV5kTLOzCoALwPnufuSsOx64FUzO9bdMwmSc3F4yN1zJ1Qxs7rAMjObk3fVsAjqEaz2hrv/CCjBi5QAJXmRsi8VqEbQq87xDMFUmklm9nhY9qaZXUSwkMYgoDzwG+Apdx8SrsQ2lmBWscrAGVGsklUHSCCYi5tw0pJLgIpAJeAOgpX2HgeOMrP5BMttfubulc1sGMHSu0cS/BD4Abje3Veb2RnAxDDOb8LtfYF04EngOIJZ6j4kWHo0K/qvTOTQoHvyImWcu28ABgDzzOxbM5sOdAZed/ddeda0Phv4HugHdHL304HmBKt7HR7WOQm4xt0bF5Dgbzezj8LP+Sn83Pbu/oOZ1QPOBdq6e2PgLuDu8EpCV+Abdz8/wjFbA3909+MJfmB0Cxf2mAUMCY/1CMGc5gCXAWnufirBMqoA9Yv8xYkcApTkReKAu48BjiBYs3o1MBBYamZV89XLJlhHvamZDSVYQz2BoNcNsMrdVxTyUQ+FyfVkYBHBNKNvhcdeAfwZuM7M7iVYiatyAcfJ6y133xy+XgpUD4+Pu78a/n2TPXN9vwM0Cuc4/wvB4iXLo/gckUOOkrxIGWdmLc2sv7tvcfeX3X0A0Ihg/ux2+epWIkikTYAlQH+CZYgTwipbo/lMd98GdATOAm4Pj90EeJ9gYY4FwH15jluYHXleZ4f7ZETYNzP87P8BDQkGGlYBXjezi6OJW+RQoyQvUvatAwabWas8ZUcSrFf9afg+k2ARluMIEuNgd58LtCVYFSupqB8a3iboBww3s6MIVjlLD68qvA1cmue4GeHnR2sZsNPMLgAwszMJevfZZtad4J78AncfSLAAS5Oixi9yKFCSFynjwrWrLwVGhffKvwCeBzq7u4fVXiBIvFkEI/G/NLNlBJfuvyDoGe/PZz9DMPDtAeCfwOHhcb8guCpQ3czSwve/mNlioujdu3sGcAUwzMyWEvyYWEOwJOzfCX48fGFmHxL8mHlkf+IXiXdahU5EDkpmdj/Bs/Vrzexo4GOgvrtvLOXQRMoMPUInIgerFcAbZpYzZqCrErxI0agnLyIiEqd0T15ERCROKcmLiIjEKSV5ERGROKUkLyIiEqeU5EVEROKUkryIiEic+n9HCC0/IhDX3AAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Get the distribution of the ratings\n",
    "x=business['stars'].value_counts()\n",
    "x=x.sort_index()\n",
    "#plot\n",
    "plt.figure(figsize=(8,4))\n",
    "ax= sns.barplot(x.index, x.values, alpha=0.8)\n",
    "plt.title(\"Star Rating Distribution\")\n",
    "plt.ylabel('# of businesses', fontsize=12)\n",
    "plt.xlabel('Star Ratings ', fontsize=12)\n",
    "\n",
    "#adding the text labels\n",
    "rects = ax.patches\n",
    "labels = x.values\n",
    "for rect, label in zip(rects, labels):\n",
    "    height = rect.get_height()\n",
    "    ax.text(rect.get_x() + rect.get_width()/2, height + 5, label, ha='center', va='bottom')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Get the distribution of the citys\n",
    "x=business['city'].value_counts()\n",
    "x=x.sort_values(ascending=False)\n",
    "x=x.iloc[0:20]\n",
    "plt.figure(figsize=(16,4))\n",
    "ax = sns.barplot(x.index, x.values, alpha=0.8,color=color[3])\n",
    "plt.title(\"Which city has the most reviews?\")\n",
    "locs, labels = plt.xticks()\n",
    "plt.setp(labels, rotation=45)\n",
    "plt.ylabel('# businesses', fontsize=12)\n",
    "plt.xlabel('City', fontsize=12)\n",
    "\n",
    "#adding the text labels\n",
    "rects = ax.patches\n",
    "labels = x.values\n",
    "for rect, label in zip(rects, labels):\n",
    "    height = rect.get_height()\n",
    "    ax.text(rect.get_x() + rect.get_width()/2, height + 5, label, ha='center', va='bottom')\n",
    "\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pandas.core.series.Series"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(business['categories'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0    Tours, Breweries, Pizza, Restaurants, Food, Ho...\n",
      "1    Chicken Wings, Burgers, Caterers, Street Vendo...\n",
      "2    Breakfast & Brunch, Restaurants, French, Sandw...\n",
      "3                        Insurance, Financial Services\n",
      "4    Home & Garden, Nurseries & Gardening, Shopping...\n",
      "Name: categories, dtype: object\n"
     ]
    }
   ],
   "source": [
    "print(business['categories'].head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "      category\n",
      "0        Tours\n",
      "1    Breweries\n",
      "2        Pizza\n",
      "3  Restaurants\n",
      "4         Food\n",
      "Restaurants               57173\n",
      "Shopping                  30231\n",
      "Food                      27118\n",
      "Beauty&Spas               18967\n",
      "HomeServices              18634\n",
      "Health&Medical            16157\n",
      "LocalServices             12906\n",
      "Automotive                12656\n",
      "Nightlife                 12438\n",
      "Bars                      10853\n",
      "EventPlanning&Services     9774\n",
      "ActiveLife                 9119\n",
      "Fashion                    7406\n",
      "Coffee&Tea                 6936\n",
      "Sandwiches                 6912\n",
      "HairSalons                 6825\n",
      "FastFood                   6812\n",
      "American(Traditional)      6659\n",
      "Pizza                      6603\n",
      "Home&Garden                6020\n",
      "Name: category, dtype: int64\n",
      "There are  1305  different types/categories of Businesses in Yelp!\n"
     ]
    }
   ],
   "source": [
    "# What are the popular business categories?\n",
    "strinfo = re.compile(' ')\n",
    "business_cat= business['categories'].str.cat(sep = ',')\n",
    "business_cats = strinfo.sub('', business_cat)\n",
    "#print(business_cats)\n",
    "cats=pd.DataFrame(business_cats.split(','),columns=['category'])\n",
    "print(cats.head())\n",
    "x=cats.category.value_counts()\n",
    "print(x.head(20))\n",
    "print(\"There are \",len(x),\" different types/categories of Businesses in Yelp!\")\n",
    "#prep for chart\n",
    "x=x.sort_values(ascending=False)\n",
    "x=x.iloc[0:20]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#chart\n",
    "plt.figure(figsize=(16,4))\n",
    "ax = sns.barplot(x.index, x.values, alpha=0.8)#,color=color[5])\n",
    "plt.title(\"What are the top categories?\",fontsize=25)\n",
    "locs, labels = plt.xticks()\n",
    "plt.setp(labels, rotation=80)\n",
    "plt.ylabel('# businesses', fontsize=12)\n",
    "plt.xlabel('Category', fontsize=12)\n",
    "#adding the text labels\n",
    "rects = ax.patches\n",
    "labels = x.values\n",
    "for rect, label in zip(rects, labels):\n",
    "    height = rect.get_height()\n",
    "    ax.text(rect.get_x() + rect.get_width()/2, height + 5, label, ha='center', va='bottom')\n",
    "\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                           address  attributes attributes.AcceptsInsurance  \\\n",
      "105           2450 Losee Rd, Ste E         NaN                         NaN   \n",
      "165                  6462 Losee Rd         NaN                         NaN   \n",
      "268                  3910 W Ann Rd         NaN                        True   \n",
      "270  5135 Camino Al Norte, Ste 250         NaN                         NaN   \n",
      "272                4728 W Craig Rd         NaN                         NaN   \n",
      "\n",
      "    attributes.AgesAllowed attributes.Alcohol attributes.Ambience  \\\n",
      "105                    NaN                NaN                 NaN   \n",
      "165                    NaN                NaN                 NaN   \n",
      "268                    NaN                NaN                 NaN   \n",
      "270                    NaN                NaN                 NaN   \n",
      "272                    NaN                NaN                 NaN   \n",
      "\n",
      "    attributes.BYOB attributes.BYOBCorkage attributes.BestNights  \\\n",
      "105             NaN                    NaN                   NaN   \n",
      "165             NaN                    NaN                   NaN   \n",
      "268             NaN                    NaN                   NaN   \n",
      "270             NaN                    NaN                   NaN   \n",
      "272             NaN                    NaN                   NaN   \n",
      "\n",
      "    attributes.BikeParking  ...  hours.Wednesday is_open   latitude  \\\n",
      "105                    NaN  ...        10:0-18:0       0  36.204563   \n",
      "165                    NaN  ...         7:0-19:0       1  36.277764   \n",
      "268                    NaN  ...         9:0-18:0       1  36.261768   \n",
      "270                    NaN  ...         9:0-17:0       1  36.253446   \n",
      "272                   True  ...          0:0-0:0       1  36.239341   \n",
      "\n",
      "      longitude              name neighborhood postal_code review_count stars  \\\n",
      "105 -115.136337   Unique Wireless          NaN       89030            7   5.0   \n",
      "165 -115.116317    Advanced Brows          NaN       89086            4   5.0   \n",
      "268 -115.192210     Allure Dental          NaN       89031            3   5.0   \n",
      "270 -115.153999  Superstar Movers          NaN       89031           16   3.0   \n",
      "272 -115.207056          7-Eleven    Northwest       89032            4   4.0   \n",
      "\n",
      "    state  \n",
      "105    NV  \n",
      "165    NV  \n",
      "268    NV  \n",
      "270    NV  \n",
      "272    NV  \n",
      "\n",
      "[5 rows x 61 columns]\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 1508 entries, 105 to 188561\n",
      "Data columns (total 61 columns):\n",
      "address                                  1422 non-null object\n",
      "attributes                               0 non-null float64\n",
      "attributes.AcceptsInsurance              105 non-null object\n",
      "attributes.AgesAllowed                   3 non-null object\n",
      "attributes.Alcohol                       296 non-null object\n",
      "attributes.Ambience                      285 non-null object\n",
      "attributes.BYOB                          6 non-null object\n",
      "attributes.BYOBCorkage                   10 non-null object\n",
      "attributes.BestNights                    26 non-null object\n",
      "attributes.BikeParking                   625 non-null object\n",
      "attributes.BusinessAcceptsBitcoin        117 non-null object\n",
      "attributes.BusinessAcceptsCreditCards    1199 non-null object\n",
      "attributes.BusinessParking               687 non-null object\n",
      "attributes.ByAppointmentOnly             420 non-null object\n",
      "attributes.Caters                        289 non-null object\n",
      "attributes.CoatCheck                     32 non-null object\n",
      "attributes.Corkage                       6 non-null object\n",
      "attributes.DietaryRestrictions           0 non-null object\n",
      "attributes.DogsAllowed                   58 non-null object\n",
      "attributes.DriveThru                     75 non-null object\n",
      "attributes.GoodForDancing                34 non-null object\n",
      "attributes.GoodForKids                   427 non-null object\n",
      "attributes.GoodForMeal                   304 non-null object\n",
      "attributes.HairSpecializesIn             10 non-null object\n",
      "attributes.HappyHour                     34 non-null object\n",
      "attributes.HasTV                         288 non-null object\n",
      "attributes.Music                         33 non-null object\n",
      "attributes.NoiseLevel                    257 non-null object\n",
      "attributes.Open24Hours                   2 non-null object\n",
      "attributes.OutdoorSeating                325 non-null object\n",
      "attributes.RestaurantsAttire             296 non-null object\n",
      "attributes.RestaurantsCounterService     3 non-null object\n",
      "attributes.RestaurantsDelivery           329 non-null object\n",
      "attributes.RestaurantsGoodForGroups      322 non-null object\n",
      "attributes.RestaurantsPriceRange2        717 non-null float64\n",
      "attributes.RestaurantsReservations       311 non-null object\n",
      "attributes.RestaurantsTableService       251 non-null object\n",
      "attributes.RestaurantsTakeOut            410 non-null object\n",
      "attributes.Smoking                       30 non-null object\n",
      "attributes.WheelchairAccessible          298 non-null object\n",
      "attributes.WiFi                          308 non-null object\n",
      "business_id                              1508 non-null object\n",
      "categories                               1499 non-null object\n",
      "city                                     1508 non-null object\n",
      "hours                                    0 non-null float64\n",
      "hours.Friday                             1183 non-null object\n",
      "hours.Monday                             1139 non-null object\n",
      "hours.Saturday                           1049 non-null object\n",
      "hours.Sunday                             784 non-null object\n",
      "hours.Thursday                           1179 non-null object\n",
      "hours.Tuesday                            1177 non-null object\n",
      "hours.Wednesday                          1179 non-null object\n",
      "is_open                                  1508 non-null int64\n",
      "latitude                                 1508 non-null float64\n",
      "longitude                                1508 non-null float64\n",
      "name                                     1508 non-null object\n",
      "neighborhood                             45 non-null object\n",
      "postal_code                              1504 non-null object\n",
      "review_count                             1508 non-null int64\n",
      "stars                                    1508 non-null float64\n",
      "state                                    1508 non-null object\n",
      "dtypes: float64(6), int64(2), object(53)\n",
      "memory usage: 730.4+ KB\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "business_NLasVeg = business[business['city']=='North Las Vegas']\n",
    "print(business_NLasVeg.head())\n",
    "print(business_NLasVeg.info())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Las Vegas          28865\n",
      "Phoenix            18633\n",
      "Toronto            18233\n",
      "Charlotte           9204\n",
      "Scottsdale          8822\n",
      "Calgary             7384\n",
      "Pittsburgh          6804\n",
      "Mesa                6239\n",
      "Montréal            6045\n",
      "Henderson           4815\n",
      "Tempe               4492\n",
      "Chandler            4272\n",
      "Madison             3509\n",
      "Cleveland           3506\n",
      "Glendale            3469\n",
      "Gilbert             3397\n",
      "Mississauga         2954\n",
      "Peoria              1868\n",
      "Markham             1699\n",
      "North Las Vegas     1508\n",
      "Name: city, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "x=business['city'].value_counts()\n",
    "print(x.head(20))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                           address  attributes  attributes.AcceptsInsurance  \\\n",
      "105           2450 Losee Rd, Ste E         0.0                            0   \n",
      "165                  6462 Losee Rd         0.0                            0   \n",
      "268                  3910 W Ann Rd         0.0                            1   \n",
      "270  5135 Camino Al Norte, Ste 250         0.0                            0   \n",
      "272                4728 W Craig Rd         0.0                            0   \n",
      "\n",
      "    attributes.AgesAllowed attributes.Alcohol attributes.Ambience  \\\n",
      "105                      0                  0                   0   \n",
      "165                      0                  0                   0   \n",
      "268                      0                  0                   0   \n",
      "270                      0                  0                   0   \n",
      "272                      0                  0                   0   \n",
      "\n",
      "     attributes.BYOB attributes.BYOBCorkage attributes.BestNights  \\\n",
      "105                0                      0                     0   \n",
      "165                0                      0                     0   \n",
      "268                0                      0                     0   \n",
      "270                0                      0                     0   \n",
      "272                0                      0                     0   \n",
      "\n",
      "     attributes.BikeParking  ...    hours.Wednesday  is_open   latitude  \\\n",
      "105                       0  ...          10:0-18:0        0  36.204563   \n",
      "165                       0  ...           7:0-19:0        1  36.277764   \n",
      "268                       0  ...           9:0-18:0        1  36.261768   \n",
      "270                       0  ...           9:0-17:0        1  36.253446   \n",
      "272                       1  ...            0:0-0:0        1  36.239341   \n",
      "\n",
      "      longitude              name  neighborhood  postal_code  review_count  \\\n",
      "105 -115.136337   Unique Wireless             0        89030             7   \n",
      "165 -115.116317    Advanced Brows             0        89086             4   \n",
      "268 -115.192210     Allure Dental             0        89031             3   \n",
      "270 -115.153999  Superstar Movers             0        89031            16   \n",
      "272 -115.207056          7-Eleven     Northwest        89032             4   \n",
      "\n",
      "     stars  state  \n",
      "105    5.0     NV  \n",
      "165    5.0     NV  \n",
      "268    5.0     NV  \n",
      "270    3.0     NV  \n",
      "272    4.0     NV  \n",
      "\n",
      "[5 rows x 61 columns]\n",
      "105    0\n",
      "165    0\n",
      "268    0\n",
      "270    0\n",
      "272    0\n",
      "Name: attributes.Ambience, dtype: object\n"
     ]
    }
   ],
   "source": [
    "#change all the nan value into 0\n",
    "business_NLasVeg = business_NLasVeg.fillna(0).apply(pd.to_numeric, errors = 'ignore')\n",
    "print(business_NLasVeg.head())\n",
    "print(business_NLasVeg['attributes.Ambience'].head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# check that all nulls are removed\n",
    "business_NLasVeg.isnull().sum().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.boxplot(x='attributes.WiFi', y='stars', data=business_NLasVeg);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "105    Appliances & Repair, IT Services & Computer Re...\n",
      "165    Doctors, Eyebrow Services, Tattoo Removal, Bea...\n",
      "268        Dentists, Health & Medical, General Dentistry\n",
      "270              Movers, Packing Services, Home Services\n",
      "272    Coffee & Tea, Automotive, Convenience Stores, ...\n",
      "Name: categories, dtype: object\n"
     ]
    }
   ],
   "source": [
    "print(business_NLasVeg['categories'].head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['address', 'attributes', 'attributes.AcceptsInsurance',\n",
       "       'attributes.AgesAllowed', 'attributes.Alcohol', 'attributes.Ambience',\n",
       "       'attributes.BYOB', 'attributes.BYOBCorkage', 'attributes.BestNights',\n",
       "       'attributes.BikeParking', 'attributes.BusinessAcceptsBitcoin',\n",
       "       'attributes.BusinessAcceptsCreditCards', 'attributes.BusinessParking',\n",
       "       'attributes.ByAppointmentOnly', 'attributes.Caters',\n",
       "       'attributes.CoatCheck', 'attributes.Corkage',\n",
       "       'attributes.DietaryRestrictions', 'attributes.DogsAllowed',\n",
       "       'attributes.DriveThru', 'attributes.GoodForDancing',\n",
       "       'attributes.GoodForKids', 'attributes.GoodForMeal',\n",
       "       'attributes.HairSpecializesIn', 'attributes.HappyHour',\n",
       "       'attributes.HasTV', 'attributes.Music', 'attributes.NoiseLevel',\n",
       "       'attributes.Open24Hours', 'attributes.OutdoorSeating',\n",
       "       'attributes.RestaurantsAttire', 'attributes.RestaurantsCounterService',\n",
       "       'attributes.RestaurantsDelivery', 'attributes.RestaurantsGoodForGroups',\n",
       "       'attributes.RestaurantsPriceRange2',\n",
       "       'attributes.RestaurantsReservations',\n",
       "       'attributes.RestaurantsTableService', 'attributes.RestaurantsTakeOut',\n",
       "       'attributes.Smoking', 'attributes.WheelchairAccessible',\n",
       "       'attributes.WiFi', 'business_id', 'categories', 'city', 'hours',\n",
       "       'hours.Friday', 'hours.Monday', 'hours.Saturday', 'hours.Sunday',\n",
       "       'hours.Thursday', 'hours.Tuesday', 'hours.Wednesday', 'is_open',\n",
       "       'latitude', 'longitude', 'name', 'neighborhood', 'postal_code',\n",
       "       'review_count', 'stars', 'state'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "business_NLasVeg.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "105          0\n",
       "165          0\n",
       "268          0\n",
       "270          0\n",
       "272          0\n",
       "471          0\n",
       "560          0\n",
       "902          0\n",
       "1060    casual\n",
       "1068    casual\n",
       "Name: attributes.RestaurantsAttire, dtype: object"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "business_NLasVeg[\"attributes.RestaurantsAttire\"].head(10)\n",
    "#type(business_NLasVeg[\"review_count\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 1508 entries, 105 to 188561\n",
      "Columns: 649 entries, address to Yoga\n",
      "dtypes: float64(6), int64(615), object(28)\n",
      "memory usage: 7.5+ MB\n",
      "None\n",
      "Index(['0', 'ATVRentals/Tours', 'Accessories', 'Accountants', 'ActiveLife',\n",
      "       'AdultEducation', 'AdultEntertainment', 'AerialTours',\n",
      "       'AirDuctCleaning', 'Airports',\n",
      "       ...\n",
      "       'Wigs', 'WildlifeControl', 'WindowWashing', 'WindowsInstallation',\n",
      "       'WindshieldInstallation&Repair', 'Wine&Spirits', 'WineBars',\n",
      "       'Women'sClothing', 'Wraps', 'Yoga'],\n",
      "      dtype='object', length=587)\n"
     ]
    }
   ],
   "source": [
    "#split categories columns into sub category column\n",
    "business_NLasVeg['categories_clean'] = list(map(lambda x: ''.join(x.split()),business_NLasVeg['categories'].astype(str)))\n",
    "#print(business_NLasVeg['categories_clean'])\n",
    "categories_df = business_NLasVeg.categories_clean.str.get_dummies(sep=',')\n",
    "\n",
    "business_NLasVeg = business_NLasVeg.merge(categories_df, left_index=True, right_index=True)\n",
    "print(business_NLasVeg.info())\n",
    "#print(business_NLasVeg.head())\n",
    "print(categories_df.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0    1173\n",
      "1     335\n",
      "Name: Restaurants, dtype: int64\n"
     ]
    },
    {
     "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>business_id</th>\n",
       "      <th>categories</th>\n",
       "      <th>city</th>\n",
       "      <th>hours</th>\n",
       "      <th>hours.Friday</th>\n",
       "      <th>hours.Monday</th>\n",
       "      <th>hours.Saturday</th>\n",
       "      <th>hours.Sunday</th>\n",
       "      <th>hours.Thursday</th>\n",
       "      <th>hours.Tuesday</th>\n",
       "      <th>hours.Wednesday</th>\n",
       "      <th>is_open</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>name</th>\n",
       "      <th>neighborhood</th>\n",
       "      <th>postal_code</th>\n",
       "      <th>review_count</th>\n",
       "      <th>stars</th>\n",
       "      <th>state</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1060</th>\n",
       "      <td>ywF14Amq9SLX5o8Qnx3hgA</td>\n",
       "      <td>Restaurants, Mexican</td>\n",
       "      <td>North Las Vegas</td>\n",
       "      <td>0.0</td>\n",
       "      <td>7:0-20:0</td>\n",
       "      <td>7:0-16:0</td>\n",
       "      <td>7:0-20:0</td>\n",
       "      <td>7:0-20:0</td>\n",
       "      <td>7:0-16:0</td>\n",
       "      <td>7:0-16:0</td>\n",
       "      <td>7:0-16:0</td>\n",
       "      <td>1</td>\n",
       "      <td>36.196344</td>\n",
       "      <td>-115.105675</td>\n",
       "      <td>El Menudazo</td>\n",
       "      <td>0</td>\n",
       "      <td>89030</td>\n",
       "      <td>77</td>\n",
       "      <td>4.0</td>\n",
       "      <td>NV</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1068</th>\n",
       "      <td>4MGlJYAiP5RYHzyQozAeAA</td>\n",
       "      <td>Irish, Sandwiches, Restaurants, Breakfast &amp; Br...</td>\n",
       "      <td>North Las Vegas</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0:0-0:0</td>\n",
       "      <td>0:0-0:0</td>\n",
       "      <td>0:0-0:0</td>\n",
       "      <td>0:0-0:0</td>\n",
       "      <td>0:0-0:0</td>\n",
       "      <td>0:0-0:0</td>\n",
       "      <td>0:0-0:0</td>\n",
       "      <td>1</td>\n",
       "      <td>36.261869</td>\n",
       "      <td>-115.184139</td>\n",
       "      <td>Sean Patricks</td>\n",
       "      <td>0</td>\n",
       "      <td>89031</td>\n",
       "      <td>56</td>\n",
       "      <td>3.0</td>\n",
       "      <td>NV</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3172</th>\n",
       "      <td>2qIZlW8YSLL8a7I6KkJdHQ</td>\n",
       "      <td>Mexican, Restaurants</td>\n",
       "      <td>North Las Vegas</td>\n",
       "      <td>0.0</td>\n",
       "      <td>9:0-21:0</td>\n",
       "      <td>9:0-21:0</td>\n",
       "      <td>8:0-0:0</td>\n",
       "      <td>8:0-21:0</td>\n",
       "      <td>9:0-21:0</td>\n",
       "      <td>9:0-21:0</td>\n",
       "      <td>9:0-21:0</td>\n",
       "      <td>0</td>\n",
       "      <td>36.276771</td>\n",
       "      <td>-115.114974</td>\n",
       "      <td>Lina's Mexican Restaurant</td>\n",
       "      <td>0</td>\n",
       "      <td>89086</td>\n",
       "      <td>38</td>\n",
       "      <td>4.0</td>\n",
       "      <td>NV</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3390</th>\n",
       "      <td>NQoGxONQIBEJR2iXL-xaQQ</td>\n",
       "      <td>Restaurants, Pizza</td>\n",
       "      <td>North Las Vegas</td>\n",
       "      <td>0.0</td>\n",
       "      <td>10:0-0:0</td>\n",
       "      <td>10:0-23:0</td>\n",
       "      <td>10:0-0:0</td>\n",
       "      <td>10:0-23:0</td>\n",
       "      <td>10:0-23:0</td>\n",
       "      <td>10:0-23:0</td>\n",
       "      <td>10:0-23:0</td>\n",
       "      <td>1</td>\n",
       "      <td>36.261998</td>\n",
       "      <td>-115.206747</td>\n",
       "      <td>Marco's Pizza</td>\n",
       "      <td>Northwest</td>\n",
       "      <td>89031</td>\n",
       "      <td>38</td>\n",
       "      <td>3.5</td>\n",
       "      <td>NV</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5556</th>\n",
       "      <td>Yqg9Muttvbm9s2r-HZY18Q</td>\n",
       "      <td>Salad, Vegetarian, Sandwiches, Restaurants</td>\n",
       "      <td>North Las Vegas</td>\n",
       "      <td>0.0</td>\n",
       "      <td>10:0-23:0</td>\n",
       "      <td>10:0-22:0</td>\n",
       "      <td>10:0-23:0</td>\n",
       "      <td>10:0-22:0</td>\n",
       "      <td>10:0-22:0</td>\n",
       "      <td>10:0-22:0</td>\n",
       "      <td>10:0-22:0</td>\n",
       "      <td>1</td>\n",
       "      <td>36.241020</td>\n",
       "      <td>-115.096742</td>\n",
       "      <td>Capriotti's Sandwich Shop</td>\n",
       "      <td>0</td>\n",
       "      <td>89030</td>\n",
       "      <td>50</td>\n",
       "      <td>3.5</td>\n",
       "      <td>NV</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6350</th>\n",
       "      <td>E6DmUAHfSRCEakvknot-zA</td>\n",
       "      <td>Seafood, Mexican, Restaurants</td>\n",
       "      <td>North Las Vegas</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>36.204544</td>\n",
       "      <td>-115.117071</td>\n",
       "      <td>Las Islitas Mariscos</td>\n",
       "      <td>0</td>\n",
       "      <td>89030</td>\n",
       "      <td>269</td>\n",
       "      <td>3.5</td>\n",
       "      <td>NV</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7138</th>\n",
       "      <td>P8wBSzIVgg-hw9XZ3LWphg</td>\n",
       "      <td>Restaurants, Mexican</td>\n",
       "      <td>North Las Vegas</td>\n",
       "      <td>0.0</td>\n",
       "      <td>10:0-22:0</td>\n",
       "      <td>10:0-22:0</td>\n",
       "      <td>10:0-22:0</td>\n",
       "      <td>10:0-22:0</td>\n",
       "      <td>10:0-22:0</td>\n",
       "      <td>10:0-22:0</td>\n",
       "      <td>10:0-22:0</td>\n",
       "      <td>1</td>\n",
       "      <td>36.196091</td>\n",
       "      <td>-115.112804</td>\n",
       "      <td>Las Islas Marias</td>\n",
       "      <td>0</td>\n",
       "      <td>89030</td>\n",
       "      <td>9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>NV</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7934</th>\n",
       "      <td>FoPkCFoyW1b_S8jLsKjcnQ</td>\n",
       "      <td>Restaurants, Mexican</td>\n",
       "      <td>North Las Vegas</td>\n",
       "      <td>0.0</td>\n",
       "      <td>9:0-16:0</td>\n",
       "      <td>0</td>\n",
       "      <td>9:0-16:0</td>\n",
       "      <td>9:0-16:0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>36.211229</td>\n",
       "      <td>-115.102190</td>\n",
       "      <td>Tacos Tijuana</td>\n",
       "      <td>0</td>\n",
       "      <td>89030</td>\n",
       "      <td>11</td>\n",
       "      <td>4.5</td>\n",
       "      <td>NV</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9960</th>\n",
       "      <td>swRC2ipDh0nRJZIjyXrU0w</td>\n",
       "      <td>Restaurants, American (Traditional)</td>\n",
       "      <td>North Las Vegas</td>\n",
       "      <td>0.0</td>\n",
       "      <td>10:0-15:0</td>\n",
       "      <td>10:0-15:0</td>\n",
       "      <td>10:0-15:0</td>\n",
       "      <td>0</td>\n",
       "      <td>10:0-15:0</td>\n",
       "      <td>10:0-15:0</td>\n",
       "      <td>10:0-15:0</td>\n",
       "      <td>0</td>\n",
       "      <td>36.220394</td>\n",
       "      <td>-115.124792</td>\n",
       "      <td>Jake's Filling Station</td>\n",
       "      <td>0</td>\n",
       "      <td>89030</td>\n",
       "      <td>24</td>\n",
       "      <td>5.0</td>\n",
       "      <td>NV</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10578</th>\n",
       "      <td>zRSUXYmoPraF1YSWG59gFQ</td>\n",
       "      <td>American (Traditional), Restaurants, Pubs, Nig...</td>\n",
       "      <td>North Las Vegas</td>\n",
       "      <td>0.0</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>36.240816</td>\n",
       "      <td>-115.116347</td>\n",
       "      <td>Winchell's Pub &amp; Grill</td>\n",
       "      <td>0</td>\n",
       "      <td>89030</td>\n",
       "      <td>9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>NV</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11098</th>\n",
       "      <td>45zlHi1yyrjpQS_icSi60A</td>\n",
       "      <td>Buffets, Restaurants</td>\n",
       "      <td>North Las Vegas</td>\n",
       "      <td>0.0</td>\n",
       "      <td>8:0-21:0</td>\n",
       "      <td>8:0-21:0</td>\n",
       "      <td>8:0-21:0</td>\n",
       "      <td>8:0-21:0</td>\n",
       "      <td>8:0-21:0</td>\n",
       "      <td>8:0-21:0</td>\n",
       "      <td>8:0-21:0</td>\n",
       "      <td>1</td>\n",
       "      <td>36.197476</td>\n",
       "      <td>-115.191654</td>\n",
       "      <td>Feast Buffet</td>\n",
       "      <td>0</td>\n",
       "      <td>89032</td>\n",
       "      <td>84</td>\n",
       "      <td>3.0</td>\n",
       "      <td>NV</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13611</th>\n",
       "      <td>2JU7FRlnYgdRauYyFzHSLA</td>\n",
       "      <td>Restaurants, Burgers, Chicken Wings</td>\n",
       "      <td>North Las Vegas</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>36.218130</td>\n",
       "      <td>-115.116121</td>\n",
       "      <td>Hamburger Hut 'N' Market</td>\n",
       "      <td>0</td>\n",
       "      <td>89030</td>\n",
       "      <td>147</td>\n",
       "      <td>4.0</td>\n",
       "      <td>NV</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13730</th>\n",
       "      <td>q8uVDQVlN1xoHJhrJ_ms5g</td>\n",
       "      <td>Restaurants, Mexican</td>\n",
       "      <td>North Las Vegas</td>\n",
       "      <td>0.0</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>36.196168</td>\n",
       "      <td>-115.110873</td>\n",
       "      <td>Taqueria el Palenque</td>\n",
       "      <td>0</td>\n",
       "      <td>89030</td>\n",
       "      <td>6</td>\n",
       "      <td>3.5</td>\n",
       "      <td>NV</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15139</th>\n",
       "      <td>-Sgt6wFx05Ox-O1ZSC3LOA</td>\n",
       "      <td>Mediterranean, Greek, Caterers, Restaurants, E...</td>\n",
       "      <td>North Las Vegas</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11:0-18:30</td>\n",
       "      <td>11:0-18:30</td>\n",
       "      <td>11:0-18:30</td>\n",
       "      <td>11:0-16:0</td>\n",
       "      <td>11:0-18:30</td>\n",
       "      <td>11:0-19:0</td>\n",
       "      <td>11:0-18:30</td>\n",
       "      <td>0</td>\n",
       "      <td>36.262606</td>\n",
       "      <td>-115.179417</td>\n",
       "      <td>The Little Greek</td>\n",
       "      <td>0</td>\n",
       "      <td>89031</td>\n",
       "      <td>180</td>\n",
       "      <td>4.0</td>\n",
       "      <td>NV</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15929</th>\n",
       "      <td>RCZwXZ7o_m-qLIp_18FsZw</td>\n",
       "      <td>Restaurants, Sandwiches, Fast Food</td>\n",
       "      <td>North Las Vegas</td>\n",
       "      <td>0.0</td>\n",
       "      <td>7:0-23:0</td>\n",
       "      <td>7:0-23:0</td>\n",
       "      <td>8:0-23:0</td>\n",
       "      <td>8:0-23:0</td>\n",
       "      <td>7:0-23:0</td>\n",
       "      <td>7:0-23:0</td>\n",
       "      <td>7:0-23:0</td>\n",
       "      <td>1</td>\n",
       "      <td>36.285896</td>\n",
       "      <td>-115.179198</td>\n",
       "      <td>Subway</td>\n",
       "      <td>0</td>\n",
       "      <td>89084</td>\n",
       "      <td>6</td>\n",
       "      <td>1.5</td>\n",
       "      <td>NV</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15931</th>\n",
       "      <td>sOf5D_Dw3fu5RVz3K-M6_A</td>\n",
       "      <td>Fast Food, Sandwiches, Restaurants</td>\n",
       "      <td>North Las Vegas</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>36.287400</td>\n",
       "      <td>-115.178331</td>\n",
       "      <td>Arby's</td>\n",
       "      <td>0</td>\n",
       "      <td>89084</td>\n",
       "      <td>24</td>\n",
       "      <td>2.5</td>\n",
       "      <td>NV</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17705</th>\n",
       "      <td>97-_E9JpqbUmaUg33-SOyQ</td>\n",
       "      <td>Italian, Chicken Wings, Restaurants, Pizza</td>\n",
       "      <td>North Las Vegas</td>\n",
       "      <td>0.0</td>\n",
       "      <td>10:30-0:0</td>\n",
       "      <td>10:30-23:0</td>\n",
       "      <td>10:30-0:0</td>\n",
       "      <td>10:30-23:0</td>\n",
       "      <td>10:30-23:0</td>\n",
       "      <td>10:30-23:0</td>\n",
       "      <td>10:30-23:0</td>\n",
       "      <td>1</td>\n",
       "      <td>36.268857</td>\n",
       "      <td>-115.110762</td>\n",
       "      <td>Pizza Hut</td>\n",
       "      <td>0</td>\n",
       "      <td>89081</td>\n",
       "      <td>18</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NV</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18553</th>\n",
       "      <td>VmSrPPO2WXmOKjUW7pDGsQ</td>\n",
       "      <td>Burgers, Diners, Restaurants, Breakfast &amp; Brunch</td>\n",
       "      <td>North Las Vegas</td>\n",
       "      <td>0.0</td>\n",
       "      <td>7:0-0:0</td>\n",
       "      <td>7:0-23:0</td>\n",
       "      <td>7:0-0:0</td>\n",
       "      <td>7:0-23:0</td>\n",
       "      <td>7:0-0:0</td>\n",
       "      <td>7:0-0:0</td>\n",
       "      <td>7:0-0:0</td>\n",
       "      <td>1</td>\n",
       "      <td>36.197112</td>\n",
       "      <td>-115.178532</td>\n",
       "      <td>Mega Tom's Burgers</td>\n",
       "      <td>0</td>\n",
       "      <td>89106</td>\n",
       "      <td>152</td>\n",
       "      <td>3.5</td>\n",
       "      <td>NV</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19619</th>\n",
       "      <td>FkXpr1ms8toOZ4sUWibPNA</td>\n",
       "      <td>Fast Food, Food, Burgers, Ice Cream &amp; Frozen Y...</td>\n",
       "      <td>North Las Vegas</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6:0-2:0</td>\n",
       "      <td>6:0-1:0</td>\n",
       "      <td>6:0-2:0</td>\n",
       "      <td>6:0-1:0</td>\n",
       "      <td>6:0-1:0</td>\n",
       "      <td>6:0-1:0</td>\n",
       "      <td>6:0-1:0</td>\n",
       "      <td>1</td>\n",
       "      <td>36.258881</td>\n",
       "      <td>-115.154184</td>\n",
       "      <td>Sonic Drive-In</td>\n",
       "      <td>0</td>\n",
       "      <td>89031</td>\n",
       "      <td>43</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NV</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20064</th>\n",
       "      <td>j_PL_u4SWW1LoEd2dCu0Iw</td>\n",
       "      <td>Event Planning &amp; Services, Caterers, Pizza, It...</td>\n",
       "      <td>North Las Vegas</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11:0-22:0</td>\n",
       "      <td>0</td>\n",
       "      <td>11:0-22:0</td>\n",
       "      <td>11:0-22:0</td>\n",
       "      <td>11:0-22:0</td>\n",
       "      <td>11:0-22:0</td>\n",
       "      <td>11:0-22:0</td>\n",
       "      <td>0</td>\n",
       "      <td>36.239559</td>\n",
       "      <td>-115.195117</td>\n",
       "      <td>Bambini's Pizzeria Rotisserie</td>\n",
       "      <td>0</td>\n",
       "      <td>89032</td>\n",
       "      <td>79</td>\n",
       "      <td>3.5</td>\n",
       "      <td>NV</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  business_id  \\\n",
       "1060   ywF14Amq9SLX5o8Qnx3hgA   \n",
       "1068   4MGlJYAiP5RYHzyQozAeAA   \n",
       "3172   2qIZlW8YSLL8a7I6KkJdHQ   \n",
       "3390   NQoGxONQIBEJR2iXL-xaQQ   \n",
       "5556   Yqg9Muttvbm9s2r-HZY18Q   \n",
       "6350   E6DmUAHfSRCEakvknot-zA   \n",
       "7138   P8wBSzIVgg-hw9XZ3LWphg   \n",
       "7934   FoPkCFoyW1b_S8jLsKjcnQ   \n",
       "9960   swRC2ipDh0nRJZIjyXrU0w   \n",
       "10578  zRSUXYmoPraF1YSWG59gFQ   \n",
       "11098  45zlHi1yyrjpQS_icSi60A   \n",
       "13611  2JU7FRlnYgdRauYyFzHSLA   \n",
       "13730  q8uVDQVlN1xoHJhrJ_ms5g   \n",
       "15139  -Sgt6wFx05Ox-O1ZSC3LOA   \n",
       "15929  RCZwXZ7o_m-qLIp_18FsZw   \n",
       "15931  sOf5D_Dw3fu5RVz3K-M6_A   \n",
       "17705  97-_E9JpqbUmaUg33-SOyQ   \n",
       "18553  VmSrPPO2WXmOKjUW7pDGsQ   \n",
       "19619  FkXpr1ms8toOZ4sUWibPNA   \n",
       "20064  j_PL_u4SWW1LoEd2dCu0Iw   \n",
       "\n",
       "                                              categories             city  \\\n",
       "1060                                Restaurants, Mexican  North Las Vegas   \n",
       "1068   Irish, Sandwiches, Restaurants, Breakfast & Br...  North Las Vegas   \n",
       "3172                                Mexican, Restaurants  North Las Vegas   \n",
       "3390                                  Restaurants, Pizza  North Las Vegas   \n",
       "5556          Salad, Vegetarian, Sandwiches, Restaurants  North Las Vegas   \n",
       "6350                       Seafood, Mexican, Restaurants  North Las Vegas   \n",
       "7138                                Restaurants, Mexican  North Las Vegas   \n",
       "7934                                Restaurants, Mexican  North Las Vegas   \n",
       "9960                 Restaurants, American (Traditional)  North Las Vegas   \n",
       "10578  American (Traditional), Restaurants, Pubs, Nig...  North Las Vegas   \n",
       "11098                               Buffets, Restaurants  North Las Vegas   \n",
       "13611                Restaurants, Burgers, Chicken Wings  North Las Vegas   \n",
       "13730                               Restaurants, Mexican  North Las Vegas   \n",
       "15139  Mediterranean, Greek, Caterers, Restaurants, E...  North Las Vegas   \n",
       "15929                 Restaurants, Sandwiches, Fast Food  North Las Vegas   \n",
       "15931                 Fast Food, Sandwiches, Restaurants  North Las Vegas   \n",
       "17705         Italian, Chicken Wings, Restaurants, Pizza  North Las Vegas   \n",
       "18553   Burgers, Diners, Restaurants, Breakfast & Brunch  North Las Vegas   \n",
       "19619  Fast Food, Food, Burgers, Ice Cream & Frozen Y...  North Las Vegas   \n",
       "20064  Event Planning & Services, Caterers, Pizza, It...  North Las Vegas   \n",
       "\n",
       "       hours hours.Friday hours.Monday hours.Saturday hours.Sunday  \\\n",
       "1060     0.0     7:0-20:0     7:0-16:0       7:0-20:0     7:0-20:0   \n",
       "1068     0.0      0:0-0:0      0:0-0:0        0:0-0:0      0:0-0:0   \n",
       "3172     0.0     9:0-21:0     9:0-21:0        8:0-0:0     8:0-21:0   \n",
       "3390     0.0     10:0-0:0    10:0-23:0       10:0-0:0    10:0-23:0   \n",
       "5556     0.0    10:0-23:0    10:0-22:0      10:0-23:0    10:0-22:0   \n",
       "6350     0.0            0            0              0            0   \n",
       "7138     0.0    10:0-22:0    10:0-22:0      10:0-22:0    10:0-22:0   \n",
       "7934     0.0     9:0-16:0            0       9:0-16:0     9:0-16:0   \n",
       "9960     0.0    10:0-15:0    10:0-15:0      10:0-15:0            0   \n",
       "10578    0.0            0            0              0            0   \n",
       "11098    0.0     8:0-21:0     8:0-21:0       8:0-21:0     8:0-21:0   \n",
       "13611    0.0            0            0              0            0   \n",
       "13730    0.0            0            0              0            0   \n",
       "15139    0.0   11:0-18:30   11:0-18:30     11:0-18:30    11:0-16:0   \n",
       "15929    0.0     7:0-23:0     7:0-23:0       8:0-23:0     8:0-23:0   \n",
       "15931    0.0            0            0              0            0   \n",
       "17705    0.0    10:30-0:0   10:30-23:0      10:30-0:0   10:30-23:0   \n",
       "18553    0.0      7:0-0:0     7:0-23:0        7:0-0:0     7:0-23:0   \n",
       "19619    0.0      6:0-2:0      6:0-1:0        6:0-2:0      6:0-1:0   \n",
       "20064    0.0    11:0-22:0            0      11:0-22:0    11:0-22:0   \n",
       "\n",
       "      hours.Thursday hours.Tuesday hours.Wednesday  is_open   latitude  \\\n",
       "1060        7:0-16:0      7:0-16:0        7:0-16:0        1  36.196344   \n",
       "1068         0:0-0:0       0:0-0:0         0:0-0:0        1  36.261869   \n",
       "3172        9:0-21:0      9:0-21:0        9:0-21:0        0  36.276771   \n",
       "3390       10:0-23:0     10:0-23:0       10:0-23:0        1  36.261998   \n",
       "5556       10:0-22:0     10:0-22:0       10:0-22:0        1  36.241020   \n",
       "6350               0             0               0        1  36.204544   \n",
       "7138       10:0-22:0     10:0-22:0       10:0-22:0        1  36.196091   \n",
       "7934               0             0               0        1  36.211229   \n",
       "9960       10:0-15:0     10:0-15:0       10:0-15:0        0  36.220394   \n",
       "10578              0             0               0        0  36.240816   \n",
       "11098       8:0-21:0      8:0-21:0        8:0-21:0        1  36.197476   \n",
       "13611              0             0               0        1  36.218130   \n",
       "13730              0             0               0        0  36.196168   \n",
       "15139     11:0-18:30     11:0-19:0      11:0-18:30        0  36.262606   \n",
       "15929       7:0-23:0      7:0-23:0        7:0-23:0        1  36.285896   \n",
       "15931              0             0               0        1  36.287400   \n",
       "17705     10:30-23:0    10:30-23:0      10:30-23:0        1  36.268857   \n",
       "18553        7:0-0:0       7:0-0:0         7:0-0:0        1  36.197112   \n",
       "19619        6:0-1:0       6:0-1:0         6:0-1:0        1  36.258881   \n",
       "20064      11:0-22:0     11:0-22:0       11:0-22:0        0  36.239559   \n",
       "\n",
       "        longitude                           name neighborhood  postal_code  \\\n",
       "1060  -115.105675                    El Menudazo            0        89030   \n",
       "1068  -115.184139                  Sean Patricks            0        89031   \n",
       "3172  -115.114974      Lina's Mexican Restaurant            0        89086   \n",
       "3390  -115.206747                  Marco's Pizza    Northwest        89031   \n",
       "5556  -115.096742      Capriotti's Sandwich Shop            0        89030   \n",
       "6350  -115.117071           Las Islitas Mariscos            0        89030   \n",
       "7138  -115.112804               Las Islas Marias            0        89030   \n",
       "7934  -115.102190                  Tacos Tijuana            0        89030   \n",
       "9960  -115.124792         Jake's Filling Station            0        89030   \n",
       "10578 -115.116347         Winchell's Pub & Grill            0        89030   \n",
       "11098 -115.191654                   Feast Buffet            0        89032   \n",
       "13611 -115.116121       Hamburger Hut 'N' Market            0        89030   \n",
       "13730 -115.110873           Taqueria el Palenque            0        89030   \n",
       "15139 -115.179417               The Little Greek            0        89031   \n",
       "15929 -115.179198                         Subway            0        89084   \n",
       "15931 -115.178331                         Arby's            0        89084   \n",
       "17705 -115.110762                      Pizza Hut            0        89081   \n",
       "18553 -115.178532             Mega Tom's Burgers            0        89106   \n",
       "19619 -115.154184                 Sonic Drive-In            0        89031   \n",
       "20064 -115.195117  Bambini's Pizzeria Rotisserie            0        89032   \n",
       "\n",
       "       review_count  stars state  \n",
       "1060             77    4.0    NV  \n",
       "1068             56    3.0    NV  \n",
       "3172             38    4.0    NV  \n",
       "3390             38    3.5    NV  \n",
       "5556             50    3.5    NV  \n",
       "6350            269    3.5    NV  \n",
       "7138              9    3.0    NV  \n",
       "7934             11    4.5    NV  \n",
       "9960             24    5.0    NV  \n",
       "10578             9    3.0    NV  \n",
       "11098            84    3.0    NV  \n",
       "13611           147    4.0    NV  \n",
       "13730             6    3.5    NV  \n",
       "15139           180    4.0    NV  \n",
       "15929             6    1.5    NV  \n",
       "15931            24    2.5    NV  \n",
       "17705            18    2.0    NV  \n",
       "18553           152    3.5    NV  \n",
       "19619            43    2.0    NV  \n",
       "20064            79    3.5    NV  "
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "select_df = business_NLasVeg[business_NLasVeg['Restaurants'] == 1]\n",
    "print(categories_df['Restaurants'].value_counts())\n",
    "#print(select_df.head())\n",
    "\n",
    "select_df = select_df.drop(columns = ['attributes'])\n",
    "select_df.iloc[:, 40:60].head(20)\n",
    "#print(select_df.info())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 335 entries, 1060 to 188560\n",
      "Columns: 648 entries, address to Yoga\n",
      "dtypes: float64(5), int64(615), object(28)\n",
      "memory usage: 1.7+ MB\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "print(select_df.info())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['address', 'attributes.AcceptsInsurance', 'attributes.AgesAllowed',\n",
      "       'attributes.Alcohol', 'attributes.Ambience', 'attributes.BYOB',\n",
      "       'attributes.BYOBCorkage', 'attributes.BestNights',\n",
      "       'attributes.BikeParking', 'attributes.BusinessAcceptsBitcoin',\n",
      "       ...\n",
      "       'Wigs', 'WildlifeControl', 'WindowWashing', 'WindowsInstallation',\n",
      "       'WindshieldInstallation&Repair', 'Wine&Spirits', 'WineBars',\n",
      "       'Women'sClothing', 'Wraps', 'Yoga'],\n",
      "      dtype='object', length=648)\n"
     ]
    }
   ],
   "source": [
    "print(select_df.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "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>address</th>\n",
       "      <th>attributes.AcceptsInsurance</th>\n",
       "      <th>attributes.Ambience</th>\n",
       "      <th>attributes.BYOB</th>\n",
       "      <th>attributes.BestNights</th>\n",
       "      <th>attributes.BikeParking</th>\n",
       "      <th>attributes.BusinessAcceptsBitcoin</th>\n",
       "      <th>attributes.BusinessAcceptsCreditCards</th>\n",
       "      <th>attributes.BusinessParking</th>\n",
       "      <th>attributes.ByAppointmentOnly</th>\n",
       "      <th>...</th>\n",
       "      <th>attributes.NoiseLevel_very_loud</th>\n",
       "      <th>attributes.RestaurantsAttire_0</th>\n",
       "      <th>attributes.RestaurantsAttire_casual</th>\n",
       "      <th>attributes.Smoking_0</th>\n",
       "      <th>attributes.Smoking_no</th>\n",
       "      <th>attributes.Smoking_yes</th>\n",
       "      <th>attributes.WiFi_0</th>\n",
       "      <th>attributes.WiFi_free</th>\n",
       "      <th>attributes.WiFi_no</th>\n",
       "      <th>attributes.WiFi_paid</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1060</th>\n",
       "      <td>3100 E Lake Mead</td>\n",
       "      <td>0</td>\n",
       "      <td>{'romantic': False, 'intimate': False, 'classy...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>{'garage': False, 'street': False, 'validated'...</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1068</th>\n",
       "      <td>3290 W Ann Rd</td>\n",
       "      <td>0</td>\n",
       "      <td>{'romantic': False, 'intimate': False, 'classy...</td>\n",
       "      <td>0</td>\n",
       "      <td>{'monday': False, 'tuesday': False, 'friday': ...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>{'garage': False, 'street': False, 'validated'...</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3172</th>\n",
       "      <td>6412 Losee Rd, Ste 100</td>\n",
       "      <td>0</td>\n",
       "      <td>{'romantic': False, 'intimate': False, 'classy...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>{'garage': False, 'street': False, 'validated'...</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3390</th>\n",
       "      <td>4770 W Ann Rd, Ste 6</td>\n",
       "      <td>0</td>\n",
       "      <td>{'romantic': False, 'intimate': False, 'classy...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5556</th>\n",
       "      <td>3820 E Craig Rd</td>\n",
       "      <td>0</td>\n",
       "      <td>{'romantic': False, 'intimate': False, 'classy...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>{'garage': False, 'street': False, 'validated'...</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 664 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     address  attributes.AcceptsInsurance  \\\n",
       "1060        3100 E Lake Mead                            0   \n",
       "1068           3290 W Ann Rd                            0   \n",
       "3172  6412 Losee Rd, Ste 100                            0   \n",
       "3390    4770 W Ann Rd, Ste 6                            0   \n",
       "5556         3820 E Craig Rd                            0   \n",
       "\n",
       "                                    attributes.Ambience  attributes.BYOB  \\\n",
       "1060  {'romantic': False, 'intimate': False, 'classy...                0   \n",
       "1068  {'romantic': False, 'intimate': False, 'classy...                0   \n",
       "3172  {'romantic': False, 'intimate': False, 'classy...                0   \n",
       "3390  {'romantic': False, 'intimate': False, 'classy...                0   \n",
       "5556  {'romantic': False, 'intimate': False, 'classy...                0   \n",
       "\n",
       "                                  attributes.BestNights  \\\n",
       "1060                                                  0   \n",
       "1068  {'monday': False, 'tuesday': False, 'friday': ...   \n",
       "3172                                                  0   \n",
       "3390                                                  0   \n",
       "5556                                                  0   \n",
       "\n",
       "      attributes.BikeParking  attributes.BusinessAcceptsBitcoin  \\\n",
       "1060                       1                                  0   \n",
       "1068                       1                                  0   \n",
       "3172                       0                                  0   \n",
       "3390                       0                                  0   \n",
       "5556                       0                                  0   \n",
       "\n",
       "      attributes.BusinessAcceptsCreditCards  \\\n",
       "1060                                      1   \n",
       "1068                                      1   \n",
       "3172                                      1   \n",
       "3390                                      1   \n",
       "5556                                      1   \n",
       "\n",
       "                             attributes.BusinessParking  \\\n",
       "1060  {'garage': False, 'street': False, 'validated'...   \n",
       "1068  {'garage': False, 'street': False, 'validated'...   \n",
       "3172  {'garage': False, 'street': False, 'validated'...   \n",
       "3390                                                  0   \n",
       "5556  {'garage': False, 'street': False, 'validated'...   \n",
       "\n",
       "      attributes.ByAppointmentOnly         ...           \\\n",
       "1060                             0         ...            \n",
       "1068                             0         ...            \n",
       "3172                             0         ...            \n",
       "3390                             0         ...            \n",
       "5556                             0         ...            \n",
       "\n",
       "      attributes.NoiseLevel_very_loud  attributes.RestaurantsAttire_0  \\\n",
       "1060                                0                               0   \n",
       "1068                                0                               0   \n",
       "3172                                0                               0   \n",
       "3390                                0                               0   \n",
       "5556                                0                               0   \n",
       "\n",
       "      attributes.RestaurantsAttire_casual  attributes.Smoking_0  \\\n",
       "1060                                    1                     1   \n",
       "1068                                    1                     0   \n",
       "3172                                    1                     1   \n",
       "3390                                    1                     1   \n",
       "5556                                    1                     1   \n",
       "\n",
       "      attributes.Smoking_no  attributes.Smoking_yes  attributes.WiFi_0  \\\n",
       "1060                      0                       0                  0   \n",
       "1068                      0                       1                  0   \n",
       "3172                      0                       0                  0   \n",
       "3390                      0                       0                  0   \n",
       "5556                      0                       0                  0   \n",
       "\n",
       "      attributes.WiFi_free attributes.WiFi_no attributes.WiFi_paid  \n",
       "1060                     0                  1                    0  \n",
       "1068                     0                  1                    0  \n",
       "3172                     0                  1                    0  \n",
       "3390                     1                  0                    0  \n",
       "5556                     0                  1                    0  \n",
       "\n",
       "[5 rows x 664 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# columns with non-boolean categorical values:\n",
    "cols_to_split = ['attributes.AgesAllowed', 'attributes.Alcohol', 'attributes.BYOBCorkage', \n",
    "                 'attributes.NoiseLevel', 'attributes.RestaurantsAttire', 'attributes.Smoking', 'attributes.WiFi']\n",
    "new_cat = pd.concat([pd.get_dummies(select_df[col], prefix=col, prefix_sep='_') for col in cols_to_split], axis=1)\n",
    "# keep all columns (not n-1) because 0's for all of them indicates that the data was missing (useful info)\n",
    "select_df = pd.concat([select_df, new_cat], axis=1)\n",
    "select_df.drop(cols_to_split, inplace=True, axis=1)\n",
    "select_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 335 entries, 1060 to 188560\n",
      "Columns: 664 entries, address to attributes.WiFi_paid\n",
      "dtypes: float64(5), int64(615), object(21), uint8(23)\n",
      "memory usage: 1.6+ MB\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "print(select_df.info())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['attributes.AcceptsInsurance', 'attributes.BYOB',\n",
      "       'attributes.BikeParking', 'attributes.BusinessAcceptsBitcoin',\n",
      "       'attributes.BusinessAcceptsCreditCards', 'attributes.ByAppointmentOnly',\n",
      "       'attributes.Caters', 'attributes.CoatCheck', 'attributes.Corkage',\n",
      "       'attributes.DietaryRestrictions',\n",
      "       ...\n",
      "       'attributes.NoiseLevel_very_loud', 'attributes.RestaurantsAttire_0',\n",
      "       'attributes.RestaurantsAttire_casual', 'attributes.Smoking_0',\n",
      "       'attributes.Smoking_no', 'attributes.Smoking_yes', 'attributes.WiFi_0',\n",
      "       'attributes.WiFi_free', 'attributes.WiFi_no', 'attributes.WiFi_paid'],\n",
      "      dtype='object', length=643)\n"
     ]
    }
   ],
   "source": [
    "#drop columns with non-numeric\n",
    "select_df = select_df.select_dtypes(exclude=['object'])\n",
    "print(select_df.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 335 entries, 1060 to 188560\n",
      "Columns: 643 entries, attributes.AcceptsInsurance to attributes.WiFi_paid\n",
      "dtypes: float64(5), int64(615), uint8(23)\n",
      "memory usage: 1.6 MB\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "print(select_df.info())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = select_df['stars']\n",
    "X = select_df.drop(columns = ['stars'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a1a30ad30>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(y, kde=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a1a372550>"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.boxplot(x='attributes.WiFi_free', y='stars', data= select_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(234, 642)\n",
      "(101, 642)\n"
     ]
    }
   ],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=.3, random_state=1)\n",
    "print(X_train.shape)\n",
    "print(X_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "lasso = make_pipeline(RobustScaler(), Lasso(alpha =0.0005, random_state=1))\n",
    "ENet = make_pipeline(RobustScaler(), ElasticNet(alpha=0.0005, l1_ratio=.9, random_state=3))\n",
    "GBoost = GradientBoostingRegressor(n_estimators=3000, learning_rate=0.05,\n",
    "                                   max_depth=4, max_features='sqrt',\n",
    "                                   min_samples_leaf=15, min_samples_split=10, \n",
    "                                   loss='huber', random_state =5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train RMSE:\n",
      "0.5800368673917029\n",
      "Test RMSE:\n",
      "0.9093493266424525\n"
     ]
    }
   ],
   "source": [
    "lasso.fit(X_train, y_train)\n",
    "y_pred_train = lasso.predict(X_train)\n",
    "print('Train RMSE:')\n",
    "print(np.sqrt(mean_squared_error(y_train, y_pred_train)))\n",
    "\n",
    "y_pred_test = lasso.predict(X_test)\n",
    "print('Test RMSE:')\n",
    "print(np.sqrt(mean_squared_error(y_test, y_pred_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train RMSE:\n",
      "0.5794050359223707\n",
      "Test RMSE:\n",
      "0.9160403113017858\n"
     ]
    }
   ],
   "source": [
    "ENet.fit(X_train, y_train)\n",
    "y_pred_train = ENet.predict(X_train)\n",
    "print('Train RMSE:')\n",
    "print(np.sqrt(mean_squared_error(y_train, y_pred_train)))\n",
    "\n",
    "y_pred_test = ENet.predict(X_test)\n",
    "print('Test RMSE:')\n",
    "print(np.sqrt(mean_squared_error(y_test, y_pred_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train RMSE:\n",
      "0.33384670147881473\n",
      "Test RMSE:\n",
      "0.7549843058339633\n"
     ]
    }
   ],
   "source": [
    "GBoost.fit(X_train, y_train)\n",
    "y_pred_train = GBoost.predict(X_train)\n",
    "print('Train RMSE:')\n",
    "print(np.sqrt(mean_squared_error(y_train, y_pred_train)))\n",
    "\n",
    "y_pred_test = GBoost.predict(X_test)\n",
    "print('Test RMSE:')\n",
    "print(np.sqrt(mean_squared_error(y_test, y_pred_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10-fold RMSEs:\n",
      "[0.90019571683844, 0.8035841925436111, 0.7037427164993725, 0.7736793952692828, 0.7656321614970085, 0.8589504166140622, 0.7299556990724387, 0.8239603263209999, 0.949985581295584, 0.9299393136056311]\n",
      "CV RMSE:\n",
      "0.827824111713872\n",
      "Std of CV RMSE:\n",
      "0.1326967910428538\n"
     ]
    }
   ],
   "source": [
    "#use cross-validation instead of train-test split for a better estimation of the RMSE\n",
    "kfold = KFold(n_splits=10, shuffle=True, random_state=1)\n",
    "cross_val_scores = cross_val_score(GBoost, X, y, scoring='neg_mean_squared_error', cv=kfold)\n",
    "print('10-fold RMSEs:')\n",
    "print([np.sqrt(-x) for x in cross_val_scores])\n",
    "print('CV RMSE:')\n",
    "print(np.sqrt(-np.mean(cross_val_scores))) # RMSE is the sqrt of the avg of MSEs\n",
    "print('Std of CV RMSE:')\n",
    "print(np.std(cross_val_scores))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "#parameters tuning for xgboost\n",
    "def modelfit(alg,useTrainCV=True, cv_folds=5, early_stopping_rounds=50):\n",
    "\n",
    "    if useTrainCV:\n",
    "        xgb_param = alg.get_xgb_params()\n",
    "        xgtrain =xgb.DMatrix(X_train,label=y_train)\n",
    "        xgtest = xgb.DMatrix(X_test)\n",
    "        cvresult = xgb.cv(xgb_param, xgtrain, num_boost_round=alg.get_params()['n_estimators'], nfold=cv_folds,\n",
    "                          early_stopping_rounds=early_stopping_rounds,show_stdv=False)\n",
    "        alg.set_params(n_estimators=cvresult.shape[0])#cvresult.shape[0] and alg.get_params()['n_estimators'] are same\n",
    "\n",
    "    #Fit the algorithm on the data\n",
    "    alg.fit(X_train, y_train,eval_metric='rmse')\n",
    "    #Predict training set:\n",
    "    dtrain_predictions = alg.predict(X_train)\n",
    "    #Print model report:\n",
    "    print(\"Score (Train): %f\" % mean_squared_error(y_train.values, dtrain_predictions))\n",
    "    #Predict on testing data:\n",
    "    dtest_predictions = alg.predict(X_test)\n",
    "    print(\"Score (Test): %f\" % mean_squared_error(y_test.values, dtest_predictions))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Score (Train): 0.037059\n",
      "Score (Test): 0.553299\n"
     ]
    }
   ],
   "source": [
    "xgb1 = XGBRegressor(booster='gbtree',\n",
    "                    objective= 'reg:linear',\n",
    "                    eval_metric='rmse',\n",
    "                    gamma = 0.1,\n",
    "                    min_child_weight= 1.1,\n",
    "                    max_depth= 5,\n",
    "                    subsample= 0.8,\n",
    "                    colsample_bytree= 0.8,\n",
    "                    tree_method= 'exact',\n",
    "                    learning_rate=0.1,\n",
    "                    n_estimators=100,\n",
    "                    nthread=4,\n",
    "                    scale_pos_weight=1,\n",
    "                    seed=27)\n",
    "modelfit(xgb1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 1.23 s, sys: 76.3 ms, total: 1.31 s\n",
      "Wall time: 13 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "#Grid seach for max_depth and min_child_weight tuning\n",
    "\n",
    "param_test1 = {\n",
    "    'max_depth':[3,5,7,9],\n",
    "    'min_child_weight':[1,3,5]\n",
    "}\n",
    "gsearch1 = GridSearchCV(estimator = XGBRegressor(booster='gbtree',\n",
    "                    objective= 'reg:linear',\n",
    "                    eval_metric='rmse',\n",
    "                    gamma = 0.1,\n",
    "                    min_child_weight= 1.1,\n",
    "                    max_depth= 5,\n",
    "                    subsample= 0.8,\n",
    "                    colsample_bytree= 0.8,\n",
    "                    tree_method= 'exact',\n",
    "                    learning_rate=0.1,\n",
    "                    n_estimators=100,\n",
    "                    nthread=4,\n",
    "                    scale_pos_weight=1,\n",
    "                    seed=27),\n",
    "                    param_grid = param_test1, scoring='mean_squared_error',n_jobs=4,iid=False, cv=5)\n",
    "gsearch1.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "([mean: -0.75434, std: 0.14600, params: {'max_depth': 3, 'min_child_weight': 1},\n",
       "  mean: -0.73818, std: 0.16217, params: {'max_depth': 3, 'min_child_weight': 3},\n",
       "  mean: -0.73042, std: 0.17354, params: {'max_depth': 3, 'min_child_weight': 5},\n",
       "  mean: -0.76044, std: 0.12061, params: {'max_depth': 5, 'min_child_weight': 1},\n",
       "  mean: -0.74657, std: 0.17033, params: {'max_depth': 5, 'min_child_weight': 3},\n",
       "  mean: -0.73615, std: 0.14945, params: {'max_depth': 5, 'min_child_weight': 5},\n",
       "  mean: -0.76385, std: 0.14216, params: {'max_depth': 7, 'min_child_weight': 1},\n",
       "  mean: -0.73575, std: 0.13349, params: {'max_depth': 7, 'min_child_weight': 3},\n",
       "  mean: -0.72902, std: 0.15856, params: {'max_depth': 7, 'min_child_weight': 5},\n",
       "  mean: -0.74439, std: 0.09913, params: {'max_depth': 9, 'min_child_weight': 1},\n",
       "  mean: -0.79265, std: 0.12450, params: {'max_depth': 9, 'min_child_weight': 3},\n",
       "  mean: -0.72550, std: 0.15679, params: {'max_depth': 9, 'min_child_weight': 5}],\n",
       " {'max_depth': 9, 'min_child_weight': 5},\n",
       " -0.725504265446873)"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gsearch1.grid_scores_, gsearch1.best_params_, gsearch1.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 826 ms, sys: 61.5 ms, total: 887 ms\n",
      "Wall time: 5.1 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "#try another one\n",
    "param_test1b = {\n",
    "    'min_child_weight':[6,8,10,12]\n",
    "}\n",
    "gsearch1b = GridSearchCV(estimator = XGBRegressor(booster='gbtree',\n",
    "                    objective= 'reg:linear',\n",
    "                    eval_metric='rmse',\n",
    "                    gamma = 0.1,\n",
    "                    min_child_weight= 1.1,\n",
    "                    max_depth= 7,\n",
    "                    subsample= 0.8,\n",
    "                    colsample_bytree= 0.8,\n",
    "                    tree_method= 'exact',\n",
    "                    learning_rate=0.1,\n",
    "                    n_estimators=100,\n",
    "                    nthread=4,\n",
    "                    scale_pos_weight=1,\n",
    "                    seed=27),\n",
    "                    param_grid = param_test1b, scoring='mean_squared_error',n_jobs=4,iid=False, cv=5)\n",
    "gsearch1b.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "([mean: -0.76121, std: 0.13833, params: {'min_child_weight': 6},\n",
       "  mean: -0.76389, std: 0.14727, params: {'min_child_weight': 8},\n",
       "  mean: -0.73145, std: 0.14099, params: {'min_child_weight': 10},\n",
       "  mean: -0.74546, std: 0.13577, params: {'min_child_weight': 12}],\n",
       " {'min_child_weight': 10},\n",
       " -0.7314454879291914)"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gsearch1b.grid_scores_, gsearch1b.best_params_, gsearch1b.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 837 ms, sys: 61.5 ms, total: 899 ms\n",
      "Wall time: 6.07 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "#gamma tuning\n",
    "param_test3 = {\n",
    "    'gamma':[i/10.0 for i in range(0,5)]\n",
    "}\n",
    "gsearch3 = GridSearchCV(estimator = XGBRegressor(booster='gbtree',\n",
    "                    objective= 'reg:linear',\n",
    "                    eval_metric='rmse',\n",
    "                    gamma = 0.1,\n",
    "                    min_child_weight= 10,\n",
    "                    max_depth= 7,\n",
    "                    subsample= 0.8,\n",
    "                    colsample_bytree= 0.8,\n",
    "                    tree_method= 'exact',\n",
    "                    learning_rate=0.1,\n",
    "                    n_estimators=100,\n",
    "                    nthread=4,\n",
    "                    scale_pos_weight=1,\n",
    "                    seed=27),\n",
    "                       param_grid = param_test3, scoring='mean_squared_error',n_jobs=4,iid=False, cv=5)\n",
    "gsearch3.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "([mean: -0.74424, std: 0.12212, params: {'gamma': 0.0},\n",
       "  mean: -0.73145, std: 0.14099, params: {'gamma': 0.1},\n",
       "  mean: -0.76799, std: 0.12489, params: {'gamma': 0.2},\n",
       "  mean: -0.76305, std: 0.12862, params: {'gamma': 0.3},\n",
       "  mean: -0.72498, std: 0.12775, params: {'gamma': 0.4}],\n",
       " {'gamma': 0.4},\n",
       " -0.7249840625383424)"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gsearch3.grid_scores_, gsearch3.best_params_, gsearch3.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 983 ms, sys: 82 ms, total: 1.06 s\n",
      "Wall time: 16.9 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "#subsample and colsample_bytree tuning\n",
    "param_test4 = {\n",
    "    'subsample':[i/10.0 for i in range(6,10)],\n",
    "    'colsample_bytree':[i/10.0 for i in range(6,10)]\n",
    "}\n",
    "gsearch4 = GridSearchCV(estimator = XGBRegressor(booster='gbtree',\n",
    "                    objective= 'reg:linear',\n",
    "                    eval_metric='rmse',\n",
    "                    gamma = 0.4,\n",
    "                    min_child_weight= 10,\n",
    "                    max_depth= 7,\n",
    "                    subsample= 0.8,\n",
    "                    colsample_bytree= 0.6,\n",
    "                    tree_method= 'exact',\n",
    "                    learning_rate=0.1,\n",
    "                    n_estimators=100,\n",
    "                    nthread=4,\n",
    "                    scale_pos_weight=1,\n",
    "                    seed=27),\n",
    "                       param_grid = param_test4, scoring='mean_squared_error',n_jobs=4,iid=False, cv=5)\n",
    "gsearch4.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "([mean: -0.72264, std: 0.12142, params: {'colsample_bytree': 0.6, 'subsample': 0.6},\n",
       "  mean: -0.73863, std: 0.14613, params: {'colsample_bytree': 0.6, 'subsample': 0.7},\n",
       "  mean: -0.74325, std: 0.13685, params: {'colsample_bytree': 0.6, 'subsample': 0.8},\n",
       "  mean: -0.70946, std: 0.12145, params: {'colsample_bytree': 0.6, 'subsample': 0.9},\n",
       "  mean: -0.72185, std: 0.13914, params: {'colsample_bytree': 0.7, 'subsample': 0.6},\n",
       "  mean: -0.74061, std: 0.15309, params: {'colsample_bytree': 0.7, 'subsample': 0.7},\n",
       "  mean: -0.74498, std: 0.14058, params: {'colsample_bytree': 0.7, 'subsample': 0.8},\n",
       "  mean: -0.73222, std: 0.13209, params: {'colsample_bytree': 0.7, 'subsample': 0.9},\n",
       "  mean: -0.70910, std: 0.12353, params: {'colsample_bytree': 0.8, 'subsample': 0.6},\n",
       "  mean: -0.71814, std: 0.12934, params: {'colsample_bytree': 0.8, 'subsample': 0.7},\n",
       "  mean: -0.72498, std: 0.12775, params: {'colsample_bytree': 0.8, 'subsample': 0.8},\n",
       "  mean: -0.73298, std: 0.12744, params: {'colsample_bytree': 0.8, 'subsample': 0.9},\n",
       "  mean: -0.72060, std: 0.10896, params: {'colsample_bytree': 0.9, 'subsample': 0.6},\n",
       "  mean: -0.73720, std: 0.15604, params: {'colsample_bytree': 0.9, 'subsample': 0.7},\n",
       "  mean: -0.73676, std: 0.12238, params: {'colsample_bytree': 0.9, 'subsample': 0.8},\n",
       "  mean: -0.70927, std: 0.12706, params: {'colsample_bytree': 0.9, 'subsample': 0.9}],\n",
       " {'colsample_bytree': 0.8, 'subsample': 0.6},\n",
       " -0.709101640725723)"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gsearch4.grid_scores_, gsearch4.best_params_, gsearch4.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 672 ms, sys: 62.3 ms, total: 734 ms\n",
      "Wall time: 4.8 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "#reg_alpha、reg_lambda rough tuning\n",
    "\n",
    "param_test6 = {\n",
    "    'reg_alpha':[1e-5, 1e-2, 0.1, 1, 100]\n",
    "}\n",
    "gsearch6 = GridSearchCV(estimator = XGBRegressor(booster='gbtree',\n",
    "                    objective= 'reg:linear',\n",
    "                    eval_metric='rmse',\n",
    "                    gamma = 0.4,\n",
    "                    min_child_weight= 10,\n",
    "                    max_depth= 7,\n",
    "                    subsample= 0.8,\n",
    "                    colsample_bytree= 0.6,\n",
    "                    tree_method= 'exact',\n",
    "                    learning_rate=0.1,\n",
    "                    n_estimators=100,\n",
    "                    nthread=4,\n",
    "                    scale_pos_weight=1,\n",
    "                    seed=27),\n",
    "                    param_grid = param_test6, scoring='mean_squared_error',n_jobs=4,iid=False, cv=5)\n",
    "gsearch6.fit(X_train,y_train)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "([mean: -0.74325, std: 0.13685, params: {'reg_alpha': 1e-05},\n",
       "  mean: -0.73535, std: 0.13829, params: {'reg_alpha': 0.01},\n",
       "  mean: -0.74764, std: 0.11554, params: {'reg_alpha': 0.1},\n",
       "  mean: -0.71324, std: 0.15186, params: {'reg_alpha': 1},\n",
       "  mean: -1.23661, std: 0.19673, params: {'reg_alpha': 100}],\n",
       " {'reg_alpha': 1},\n",
       " -0.7132417001396024)"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gsearch6.grid_scores_, gsearch6.best_params_, gsearch6.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 702 ms, sys: 60.8 ms, total: 763 ms\n",
      "Wall time: 4.64 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "##reg_alpha、reg_lambda fine tuning\n",
    "param_test7 = {\n",
    "    'reg_alpha':[0, 0.001, 0.005, 0.01, 0.05]\n",
    "}\n",
    "gsearch7 = GridSearchCV(estimator = XGBRegressor(booster='gbtree',\n",
    "                    objective= 'reg:linear',\n",
    "                    eval_metric='rmse',\n",
    "                    gamma = 0.4,\n",
    "                    min_child_weight= 10,\n",
    "                    max_depth= 7,\n",
    "                    subsample= 0.8,\n",
    "                    colsample_bytree= 0.6,\n",
    "                    tree_method= 'exact',\n",
    "                    learning_rate=0.1,\n",
    "                    n_estimators=100,\n",
    "                    nthread=4,\n",
    "                    scale_pos_weight=1,\n",
    "                    seed=27),\n",
    "                       param_grid = param_test7, scoring='mean_squared_error',n_jobs=4,iid=False, cv=5)\n",
    "gsearch7.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "([mean: -0.74325, std: 0.13685, params: {'reg_alpha': 0},\n",
       "  mean: -0.74322, std: 0.13684, params: {'reg_alpha': 0.001},\n",
       "  mean: -0.75105, std: 0.13057, params: {'reg_alpha': 0.005},\n",
       "  mean: -0.73535, std: 0.13829, params: {'reg_alpha': 0.01},\n",
       "  mean: -0.73670, std: 0.13679, params: {'reg_alpha': 0.05}],\n",
       " {'reg_alpha': 0.01},\n",
       " -0.735347370058071)"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gsearch7.grid_scores_, gsearch7.best_params_, gsearch7.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 2.91 s, sys: 78.8 ms, total: 2.99 s\n",
      "Wall time: 1min 3s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "#reduce learning_rate, increase n_estimators\n",
    "param_test9 = {\n",
    "    'n_estimators':[50, 100, 200, 500,1000],\n",
    "    'learning_rate':[0.001, 0.01, 0.05, 0.1,0.2]\n",
    "}\n",
    "gsearch9 = GridSearchCV(estimator = XGBRegressor(booster='gbtree',\n",
    "                    objective= 'reg:linear',\n",
    "                    eval_metric='rmse',\n",
    "                    gamma = 0.4,\n",
    "                    min_child_weight= 10,\n",
    "                    max_depth= 7,\n",
    "                    subsample= 0.8,\n",
    "                    colsample_bytree= 0.6,\n",
    "                    tree_method= 'exact',\n",
    "                    learning_rate=0.1,\n",
    "                    n_estimators=100,\n",
    "                    nthread=4,\n",
    "                    scale_pos_weight=1,\n",
    "                    reg_alpha=0.01,                           \n",
    "                    seed=27),\n",
    "                       param_grid = param_test9, scoring='mean_squared_error',n_jobs=4,iid=False, cv=5)\n",
    "gsearch9.fit(X_train,y_train)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "([mean: -7.59732, std: 0.44923, params: {'learning_rate': 0.001, 'n_estimators': 50},\n",
       "  mean: -6.94856, std: 0.42006, params: {'learning_rate': 0.001, 'n_estimators': 100},\n",
       "  mean: -5.82811, std: 0.36739, params: {'learning_rate': 0.001, 'n_estimators': 200},\n",
       "  mean: -3.52629, std: 0.24897, params: {'learning_rate': 0.001, 'n_estimators': 500},\n",
       "  mean: -1.73238, std: 0.14353, params: {'learning_rate': 0.001, 'n_estimators': 1000},\n",
       "  mean: -3.50692, std: 0.24124, params: {'learning_rate': 0.01, 'n_estimators': 50},\n",
       "  mean: -1.72382, std: 0.13514, params: {'learning_rate': 0.01, 'n_estimators': 100},\n",
       "  mean: -0.79212, std: 0.10785, params: {'learning_rate': 0.01, 'n_estimators': 200},\n",
       "  mean: -0.66393, std: 0.12769, params: {'learning_rate': 0.01, 'n_estimators': 500},\n",
       "  mean: -0.70632, std: 0.13441, params: {'learning_rate': 0.01, 'n_estimators': 1000},\n",
       "  mean: -0.69037, std: 0.11839, params: {'learning_rate': 0.05, 'n_estimators': 50},\n",
       "  mean: -0.67685, std: 0.13096, params: {'learning_rate': 0.05, 'n_estimators': 100},\n",
       "  mean: -0.71721, std: 0.13242, params: {'learning_rate': 0.05, 'n_estimators': 200},\n",
       "  mean: -0.74413, std: 0.13027, params: {'learning_rate': 0.05, 'n_estimators': 500},\n",
       "  mean: -0.75460, std: 0.13017, params: {'learning_rate': 0.05, 'n_estimators': 1000},\n",
       "  mean: -0.69668, std: 0.12919, params: {'learning_rate': 0.1, 'n_estimators': 50},\n",
       "  mean: -0.73535, std: 0.13829, params: {'learning_rate': 0.1, 'n_estimators': 100},\n",
       "  mean: -0.75353, std: 0.14043, params: {'learning_rate': 0.1, 'n_estimators': 200},\n",
       "  mean: -0.76911, std: 0.14456, params: {'learning_rate': 0.1, 'n_estimators': 500},\n",
       "  mean: -0.77553, std: 0.14402, params: {'learning_rate': 0.1, 'n_estimators': 1000},\n",
       "  mean: -0.75912, std: 0.13261, params: {'learning_rate': 0.2, 'n_estimators': 50},\n",
       "  mean: -0.78272, std: 0.13158, params: {'learning_rate': 0.2, 'n_estimators': 100},\n",
       "  mean: -0.80295, std: 0.13253, params: {'learning_rate': 0.2, 'n_estimators': 200},\n",
       "  mean: -0.81247, std: 0.12769, params: {'learning_rate': 0.2, 'n_estimators': 500},\n",
       "  mean: -0.81374, std: 0.12227, params: {'learning_rate': 0.2, 'n_estimators': 1000}],\n",
       " {'learning_rate': 0.01, 'n_estimators': 500},\n",
       " -0.6639286281047992)"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gsearch9.grid_scores_, gsearch9.best_params_, gsearch9.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "xgb9 = XGBRegressor(booster='gbtree',\n",
    "                    objective= 'reg:linear',\n",
    "                    eval_metric='rmse',\n",
    "                    gamma = 0.4,\n",
    "                    min_child_weight= 10,\n",
    "                    max_depth= 7,\n",
    "                    subsample= 0.8,\n",
    "                    colsample_bytree= 0.6,\n",
    "                    tree_method= 'exact',\n",
    "                    learning_rate=0.01,\n",
    "                    n_estimators=500,\n",
    "                    nthread=4,\n",
    "                    scale_pos_weight=1,\n",
    "                    reg_alpha=0.05,                           \n",
    "                    seed=27)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "XGBRegressor(base_score=0.5, booster='gbtree', colsample_bylevel=1,\n",
       "       colsample_bytree=0.6, eval_metric='rmse', gamma=0.4,\n",
       "       learning_rate=0.01, max_delta_step=0, max_depth=7,\n",
       "       min_child_weight=10, missing=None, n_estimators=500, n_jobs=1,\n",
       "       nthread=4, objective='reg:linear', random_state=0, reg_alpha=0.05,\n",
       "       reg_lambda=1, scale_pos_weight=1, seed=27, silent=True,\n",
       "       subsample=0.8, tree_method='exact')"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xgb9.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7442509524551613"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sqrt(mean_squared_error(xgb9.predict(X_test),y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a1a12a198>"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x936 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1, 1, figsize=(8, 13))\n",
    "xgb.plot_importance(xgb9, max_num_features=20, height=0.5, ax=ax)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}