cross_validation_kmeans.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Setting up and reading data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/rh/anaconda/root/envs/PUI2016_Python2/lib/python2.7/site-packages/IPython/html.py:14: ShimWarning: The `IPython.html` package has been deprecated. You should import from `notebook` instead. `IPython.html.widgets` has moved to `ipywidgets`.\n",
      "  \"`IPython.html.widgets` has moved to `ipywidgets`.\", ShimWarning)\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "#import seaborn\n",
    "from sklearn import naive_bayes\n",
    "from sklearn import svm\n",
    "from sklearn.cross_validation import train_test_split\n",
    "from pandas.tools.plotting import scatter_matrix\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.ensemble import AdaBoostClassifier\n",
    "from sklearn.ensemble import GradientBoostingClassifier\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.ensemble import BaggingClassifier\n",
    "from IPython.display import clear_output\n",
    "import csv\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import geopandas as gpd\n",
    "%matplotlib inline\n",
    "import csv\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from sklearn import naive_bayes\n",
    "from sklearn import svm\n",
    "from sklearn.cross_validation import train_test_split\n",
    "from sklearn import cluster\n",
    "#pd.options.display.max_rows = 999\n",
    "#pd.set_option('display.max_rows', 500)\n",
    "#pd.set_option('display.expand_frame_repr', False)\n",
    "from IPython.display import display\n",
    "pd.options.display.max_columns = None\n",
    "import seaborn as sns\n",
    "from sklearn.cluster import SpectralClustering\n",
    "from sklearn.cluster import KMeans\n",
    "from sklearn.cluster import AgglomerativeClustering\n",
    "from sklearn.neighbors import NearestNeighbors\n",
    "import time\n",
    "%matplotlib inline "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df = gpd.GeoDataFrame.from_file('Data/jp-clustered.geojson')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Entire_home/apt</th>\n",
       "      <th>Median_income</th>\n",
       "      <th>Median_rent</th>\n",
       "      <th>Private_room</th>\n",
       "      <th>Shared_room</th>\n",
       "      <th>accesstosubway</th>\n",
       "      <th>availability_365</th>\n",
       "      <th>beer_count</th>\n",
       "      <th>boro_ct_2010</th>\n",
       "      <th>calculated_host_listings_count</th>\n",
       "      <th>coffee_count</th>\n",
       "      <th>connectivityScore</th>\n",
       "      <th>geometry</th>\n",
       "      <th>host_id</th>\n",
       "      <th>host_name</th>\n",
       "      <th>id</th>\n",
       "      <th>index_left</th>\n",
       "      <th>kmeans</th>\n",
       "      <th>kmeansCat</th>\n",
       "      <th>last_review</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>minimum_nights</th>\n",
       "      <th>name</th>\n",
       "      <th>neighbourhood</th>\n",
       "      <th>neighbourhood_group</th>\n",
       "      <th>number_of_reviews</th>\n",
       "      <th>price</th>\n",
       "      <th>reviews_per_month</th>\n",
       "      <th>room_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>21855.0</td>\n",
       "      <td>1287.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>8.368050</td>\n",
       "      <td>242</td>\n",
       "      <td>0</td>\n",
       "      <td>2002400</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>POINT (-73.86406089014145 40.81274651012037)</td>\n",
       "      <td>525710</td>\n",
       "      <td>Libby</td>\n",
       "      <td>14880182</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2017-02-08</td>\n",
       "      <td>40.812747</td>\n",
       "      <td>-73.864061</td>\n",
       "      <td>2</td>\n",
       "      <td>Community Living in the Heart &amp; Soul of NYC (3C)</td>\n",
       "      <td>Clason Point</td>\n",
       "      <td>Bronx</td>\n",
       "      <td>9</td>\n",
       "      <td>59</td>\n",
       "      <td>1.53</td>\n",
       "      <td>Private room</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>21855.0</td>\n",
       "      <td>1287.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>8.380864</td>\n",
       "      <td>365</td>\n",
       "      <td>0</td>\n",
       "      <td>2000200</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>POINT (-73.86381389361847 40.81293405229447)</td>\n",
       "      <td>525710</td>\n",
       "      <td>Libby</td>\n",
       "      <td>15359150</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2017-02-10</td>\n",
       "      <td>40.812934</td>\n",
       "      <td>-73.863814</td>\n",
       "      <td>2</td>\n",
       "      <td>Private Room for 2, Community House NYC (2A)</td>\n",
       "      <td>Clason Point</td>\n",
       "      <td>Bronx</td>\n",
       "      <td>2</td>\n",
       "      <td>43</td>\n",
       "      <td>0.42</td>\n",
       "      <td>Private room</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>21855.0</td>\n",
       "      <td>1287.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>8.406552</td>\n",
       "      <td>336</td>\n",
       "      <td>0</td>\n",
       "      <td>2000200</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>POINT (-73.86370502873359 40.81359965937455)</td>\n",
       "      <td>525710</td>\n",
       "      <td>Libby</td>\n",
       "      <td>15359306</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2017-01-09</td>\n",
       "      <td>40.813600</td>\n",
       "      <td>-73.863705</td>\n",
       "      <td>1</td>\n",
       "      <td>King sized bed + futon in Large Private Room  ...</td>\n",
       "      <td>Clason Point</td>\n",
       "      <td>Bronx</td>\n",
       "      <td>6</td>\n",
       "      <td>76</td>\n",
       "      <td>1.37</td>\n",
       "      <td>Private room</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Entire_home/apt  Median_income  Median_rent  Private_room  Shared_room  \\\n",
       "0                0        21855.0       1287.0             1            0   \n",
       "1                0        21855.0       1287.0             1            0   \n",
       "2                0        21855.0       1287.0             1            0   \n",
       "\n",
       "   accesstosubway  availability_365  beer_count boro_ct_2010  \\\n",
       "0        8.368050               242           0      2002400   \n",
       "1        8.380864               365           0      2000200   \n",
       "2        8.406552               336           0      2000200   \n",
       "\n",
       "   calculated_host_listings_count  coffee_count  connectivityScore  \\\n",
       "0                              10             0                0.0   \n",
       "1                              10             0                0.0   \n",
       "2                              10             0                0.0   \n",
       "\n",
       "                                       geometry  host_id host_name        id  \\\n",
       "0  POINT (-73.86406089014145 40.81274651012037)   525710     Libby  14880182   \n",
       "1  POINT (-73.86381389361847 40.81293405229447)   525710     Libby  15359150   \n",
       "2  POINT (-73.86370502873359 40.81359965937455)   525710     Libby  15359306   \n",
       "\n",
       "   index_left  kmeans kmeansCat last_review   latitude  longitude  \\\n",
       "0         1.0       1         1  2017-02-08  40.812747 -73.864061   \n",
       "1         1.0       1         1  2017-02-10  40.812934 -73.863814   \n",
       "2         1.0       1         1  2017-01-09  40.813600 -73.863705   \n",
       "\n",
       "   minimum_nights                                               name  \\\n",
       "0               2   Community Living in the Heart & Soul of NYC (3C)   \n",
       "1               2       Private Room for 2, Community House NYC (2A)   \n",
       "2               1  King sized bed + futon in Large Private Room  ...   \n",
       "\n",
       "  neighbourhood neighbourhood_group  number_of_reviews  price  \\\n",
       "0  Clason Point               Bronx                  9     59   \n",
       "1  Clason Point               Bronx                  2     43   \n",
       "2  Clason Point               Bronx                  6     76   \n",
       "\n",
       "   reviews_per_month     room_type  \n",
       "0               1.53  Private room  \n",
       "1               0.42  Private room  \n",
       "2               1.37  Private room  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Entire_home/apt                     int64\n",
       "Median_income                     float64\n",
       "Median_rent                       float64\n",
       "Private_room                        int64\n",
       "Shared_room                         int64\n",
       "accesstosubway                    float64\n",
       "availability_365                    int64\n",
       "beer_count                          int64\n",
       "boro_ct_2010                       object\n",
       "calculated_host_listings_count      int64\n",
       "coffee_count                        int64\n",
       "connectivityScore                 float64\n",
       "geometry                           object\n",
       "host_id                             int64\n",
       "host_name                          object\n",
       "id                                 object\n",
       "index_left                        float64\n",
       "kmeans                              int64\n",
       "kmeansCat                          object\n",
       "last_review                        object\n",
       "latitude                          float64\n",
       "longitude                         float64\n",
       "minimum_nights                      int64\n",
       "name                               object\n",
       "neighbourhood                      object\n",
       "neighbourhood_group                object\n",
       "number_of_reviews                   int64\n",
       "price                               int64\n",
       "reviews_per_month                 float64\n",
       "room_type                          object\n",
       "dtype: object"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Inspecting data and cleaning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Entire_home/apt</th>\n",
       "      <th>Median_income</th>\n",
       "      <th>Median_rent</th>\n",
       "      <th>Private_room</th>\n",
       "      <th>Shared_room</th>\n",
       "      <th>accesstosubway</th>\n",
       "      <th>availability_365</th>\n",
       "      <th>beer_count</th>\n",
       "      <th>boro_ct_2010</th>\n",
       "      <th>calculated_host_listings_count</th>\n",
       "      <th>coffee_count</th>\n",
       "      <th>connectivityScore</th>\n",
       "      <th>geometry</th>\n",
       "      <th>host_id</th>\n",
       "      <th>host_name</th>\n",
       "      <th>id</th>\n",
       "      <th>index_left</th>\n",
       "      <th>kmeansCat</th>\n",
       "      <th>last_review</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>minimum_nights</th>\n",
       "      <th>name</th>\n",
       "      <th>neighbourhood</th>\n",
       "      <th>neighbourhood_group</th>\n",
       "      <th>number_of_reviews</th>\n",
       "      <th>price</th>\n",
       "      <th>reviews_per_month</th>\n",
       "      <th>room_type</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>kmeans</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10906</td>\n",
       "      <td>10906</td>\n",
       "      <td>10894</td>\n",
       "      <td>10906</td>\n",
       "      <td>10906</td>\n",
       "      <td>10906</td>\n",
       "      <td>10906</td>\n",
       "      <td>10906</td>\n",
       "      <td>10904</td>\n",
       "      <td>10906</td>\n",
       "      <td>10906</td>\n",
       "      <td>10906</td>\n",
       "      <td>10906</td>\n",
       "      <td>10906</td>\n",
       "      <td>10901</td>\n",
       "      <td>10906</td>\n",
       "      <td>10906</td>\n",
       "      <td>10906</td>\n",
       "      <td>8587</td>\n",
       "      <td>10906</td>\n",
       "      <td>10906</td>\n",
       "      <td>10906</td>\n",
       "      <td>10906</td>\n",
       "      <td>10906</td>\n",
       "      <td>10906</td>\n",
       "      <td>10906</td>\n",
       "      <td>10906</td>\n",
       "      <td>8587</td>\n",
       "      <td>10906</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>20366</td>\n",
       "      <td>20366</td>\n",
       "      <td>20366</td>\n",
       "      <td>20366</td>\n",
       "      <td>20366</td>\n",
       "      <td>20366</td>\n",
       "      <td>20366</td>\n",
       "      <td>20366</td>\n",
       "      <td>20364</td>\n",
       "      <td>20366</td>\n",
       "      <td>20366</td>\n",
       "      <td>20366</td>\n",
       "      <td>20366</td>\n",
       "      <td>20366</td>\n",
       "      <td>20357</td>\n",
       "      <td>20366</td>\n",
       "      <td>20366</td>\n",
       "      <td>20366</td>\n",
       "      <td>15795</td>\n",
       "      <td>20366</td>\n",
       "      <td>20366</td>\n",
       "      <td>20366</td>\n",
       "      <td>20365</td>\n",
       "      <td>20366</td>\n",
       "      <td>20366</td>\n",
       "      <td>20366</td>\n",
       "      <td>20366</td>\n",
       "      <td>15795</td>\n",
       "      <td>20366</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4312</td>\n",
       "      <td>4312</td>\n",
       "      <td>4269</td>\n",
       "      <td>4312</td>\n",
       "      <td>4312</td>\n",
       "      <td>4312</td>\n",
       "      <td>4312</td>\n",
       "      <td>4312</td>\n",
       "      <td>4312</td>\n",
       "      <td>4312</td>\n",
       "      <td>4312</td>\n",
       "      <td>4312</td>\n",
       "      <td>4312</td>\n",
       "      <td>4312</td>\n",
       "      <td>4308</td>\n",
       "      <td>4312</td>\n",
       "      <td>4312</td>\n",
       "      <td>4312</td>\n",
       "      <td>3329</td>\n",
       "      <td>4312</td>\n",
       "      <td>4312</td>\n",
       "      <td>4312</td>\n",
       "      <td>4312</td>\n",
       "      <td>4312</td>\n",
       "      <td>4312</td>\n",
       "      <td>4312</td>\n",
       "      <td>4312</td>\n",
       "      <td>3329</td>\n",
       "      <td>4312</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "      <td>672</td>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "      <td>563</td>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "      <td>563</td>\n",
       "      <td>740</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        Entire_home/apt  Median_income  Median_rent  Private_room  \\\n",
       "kmeans                                                              \n",
       "0                 10906          10906        10894         10906   \n",
       "1                 20366          20366        20366         20366   \n",
       "2                  4312           4312         4269          4312   \n",
       "3                   740            740          672           740   \n",
       "\n",
       "        Shared_room  accesstosubway  availability_365  beer_count  \\\n",
       "kmeans                                                              \n",
       "0             10906           10906             10906       10906   \n",
       "1             20366           20366             20366       20366   \n",
       "2              4312            4312              4312        4312   \n",
       "3               740             740               740         740   \n",
       "\n",
       "        boro_ct_2010  calculated_host_listings_count  coffee_count  \\\n",
       "kmeans                                                               \n",
       "0              10904                           10906         10906   \n",
       "1              20364                           20366         20366   \n",
       "2               4312                            4312          4312   \n",
       "3                740                             740           740   \n",
       "\n",
       "        connectivityScore  geometry  host_id  host_name     id  index_left  \\\n",
       "kmeans                                                                       \n",
       "0                   10906     10906    10906      10901  10906       10906   \n",
       "1                   20366     20366    20366      20357  20366       20366   \n",
       "2                    4312      4312     4312       4308   4312        4312   \n",
       "3                     740       740      740        740    740         740   \n",
       "\n",
       "        kmeansCat  last_review  latitude  longitude  minimum_nights   name  \\\n",
       "kmeans                                                                       \n",
       "0           10906         8587     10906      10906           10906  10906   \n",
       "1           20366        15795     20366      20366           20366  20365   \n",
       "2            4312         3329      4312       4312            4312   4312   \n",
       "3             740          563       740        740             740    740   \n",
       "\n",
       "        neighbourhood  neighbourhood_group  number_of_reviews  price  \\\n",
       "kmeans                                                                 \n",
       "0               10906                10906              10906  10906   \n",
       "1               20366                20366              20366  20366   \n",
       "2                4312                 4312               4312   4312   \n",
       "3                 740                  740                740    740   \n",
       "\n",
       "        reviews_per_month  room_type  \n",
       "kmeans                                \n",
       "0                    8587      10906  \n",
       "1                   15795      20366  \n",
       "2                    3329       4312  \n",
       "3                     563        740  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby(['kmeans']).count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index([               u'Entire_home/apt',                  u'Median_income',\n",
       "                          u'Median_rent',                   u'Private_room',\n",
       "                          u'Shared_room',                 u'accesstosubway',\n",
       "                     u'availability_365',                     u'beer_count',\n",
       "                         u'boro_ct_2010', u'calculated_host_listings_count',\n",
       "                         u'coffee_count',              u'connectivityScore',\n",
       "                             u'geometry',                        u'host_id',\n",
       "                            u'host_name',                             u'id',\n",
       "                           u'index_left',                         u'kmeans',\n",
       "                            u'kmeansCat',                    u'last_review',\n",
       "                             u'latitude',                      u'longitude',\n",
       "                       u'minimum_nights',                           u'name',\n",
       "                        u'neighbourhood',            u'neighbourhood_group',\n",
       "                    u'number_of_reviews',                          u'price',\n",
       "                    u'reviews_per_month',                      u'room_type'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df_1 = df[[ u'kmeans', u'beer_count', u'coffee_count', u'connectivityScore', u'Median_income', u'price']]\n",
    "            #u'reviews_per_month']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>kmeans</th>\n",
       "      <th>beer_count</th>\n",
       "      <th>coffee_count</th>\n",
       "      <th>connectivityScore</th>\n",
       "      <th>Median_income</th>\n",
       "      <th>price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>36324.000000</td>\n",
       "      <td>36324.000000</td>\n",
       "      <td>36324.000000</td>\n",
       "      <td>36324.000000</td>\n",
       "      <td>36324.000000</td>\n",
       "      <td>36324.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.859212</td>\n",
       "      <td>0.405600</td>\n",
       "      <td>6.242429</td>\n",
       "      <td>79.558997</td>\n",
       "      <td>50235.823808</td>\n",
       "      <td>128.687562</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.693277</td>\n",
       "      <td>0.773165</td>\n",
       "      <td>7.359190</td>\n",
       "      <td>65.852461</td>\n",
       "      <td>36489.584342</td>\n",
       "      <td>84.574034</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>4291.000000</td>\n",
       "      <td>10.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>37.000000</td>\n",
       "      <td>23142.000000</td>\n",
       "      <td>70.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>61.000000</td>\n",
       "      <td>39357.000000</td>\n",
       "      <td>100.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>66060.000000</td>\n",
       "      <td>165.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>3.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>35.000000</td>\n",
       "      <td>396.000000</td>\n",
       "      <td>340545.000000</td>\n",
       "      <td>500.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             kmeans    beer_count  coffee_count  connectivityScore  \\\n",
       "count  36324.000000  36324.000000  36324.000000       36324.000000   \n",
       "mean       0.859212      0.405600      6.242429          79.558997   \n",
       "std        0.693277      0.773165      7.359190          65.852461   \n",
       "min        0.000000      0.000000      0.000000           0.000000   \n",
       "25%        0.000000      0.000000      1.000000          37.000000   \n",
       "50%        1.000000      0.000000      3.000000          61.000000   \n",
       "75%        1.000000      1.000000     10.000000         100.000000   \n",
       "max        3.000000      5.000000     35.000000         396.000000   \n",
       "\n",
       "       Median_income         price  \n",
       "count   36324.000000  36324.000000  \n",
       "mean    50235.823808    128.687562  \n",
       "std     36489.584342     84.574034  \n",
       "min      4291.000000     10.000000  \n",
       "25%     23142.000000     70.000000  \n",
       "50%     39357.000000    100.000000  \n",
       "75%     66060.000000    165.000000  \n",
       "max    340545.000000    500.000000  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_1.set_index(df['id'], inplace=True)\n",
    "df_1 = df_1\n",
    "df_1.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/rh/anaconda/root/envs/PUI2016_Python2/lib/python2.7/site-packages/ipykernel/__main__.py:1: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  if __name__ == '__main__':\n"
     ]
    }
   ],
   "source": [
    "df_1.dropna(inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>beer_count</th>\n",
       "      <th>coffee_count</th>\n",
       "      <th>connectivityScore</th>\n",
       "      <th>Median_income</th>\n",
       "      <th>price</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>kmeans</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10906</td>\n",
       "      <td>10906</td>\n",
       "      <td>10906</td>\n",
       "      <td>10906</td>\n",
       "      <td>10906</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>20366</td>\n",
       "      <td>20366</td>\n",
       "      <td>20366</td>\n",
       "      <td>20366</td>\n",
       "      <td>20366</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4312</td>\n",
       "      <td>4312</td>\n",
       "      <td>4312</td>\n",
       "      <td>4312</td>\n",
       "      <td>4312</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "      <td>740</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        beer_count  coffee_count  connectivityScore  Median_income  price\n",
       "kmeans                                                                   \n",
       "0            10906         10906              10906          10906  10906\n",
       "1            20366         20366              20366          20366  20366\n",
       "2             4312          4312               4312           4312   4312\n",
       "3              740           740                740            740    740"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_1.groupby(['kmeans']).count()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Splitting data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_x = df_1.ix[:,1:]\n",
    "df_y = df_1.ix[:,0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "id\n",
       "14880182    1\n",
       "15359150    1\n",
       "15359306    1\n",
       "15244537    1\n",
       "15346892    1\n",
       "Name: kmeans, dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_y.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X = df_x.as_matrix()\n",
    "Y = df_y.as_matrix()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Normalize"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "#Xtmax = np.amax(X,axis=0)\n",
    "#X = np.divide(X,Xtmax)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Splitting to train and test (70:30)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "X_train, X_test, Y_train, Y_test = train_test_split(X, Y, train_size=0.7, random_state=3)\n",
    "ts = Y_test.shape[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Cross Validation on KMeans"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "99\n"
     ]
    }
   ],
   "source": [
    "# --------------------\n",
    "# Decision Tree Classifier\n",
    "# --------------------\n",
    "DTC_results = []\n",
    "for j in range(100):\n",
    "    clear_output()\n",
    "    print j\n",
    "    X_train, X_test, Y_train, Y_test = train_test_split(X, Y, train_size=0.7)\n",
    "    ts = Y_test.shape[0]\n",
    "    # Train\n",
    "    dtc = DecisionTreeClassifier(max_depth=10)\n",
    "    dtc.fit(X_train, Y_train)\n",
    "\n",
    "    # Predicting\n",
    "    dtc_pred = dtc.predict(X_test)\n",
    "\n",
    "    # Finding mispredicted samples\n",
    "    dtc_verror = np.asarray([int(dtc_pred[i] != Y_test[i]) for i in range(0,ts)])\n",
    "    dtc_error = np.sum(dtc_verror)\n",
    "    dtc_ccidx = np.where(dtc_verror == 0)\n",
    "    dtc_mcidx = np.where(dtc_verror == 1)\n",
    "\n",
    "    DTC_results.append(dtc_error/float(ts))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('miss-classified %age is ', 0.0)\n"
     ]
    }
   ],
   "source": [
    "print ('miss-classified %age is ', (sum(DTC_results)/len(DTC_results)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "99\n"
     ]
    }
   ],
   "source": [
    "# --------------------\n",
    "# Bagging Classifier\n",
    "# --------------------\n",
    "bagb_results = []\n",
    "for j in range(100):\n",
    "    clear_output()\n",
    "    print j\n",
    "    X_train, X_test, Y_train, Y_test = train_test_split(X, Y, train_size=0.7)\n",
    "    ts = Y_test.shape[0]\n",
    "    bagb = BaggingClassifier(DecisionTreeClassifier(max_depth=10), n_estimators=20)\n",
    "#adab = AdaBoostClassifier(DecisionTreeClassifier(max_depth=2), n_estimators=20,learning_rate=1.5,algorithm=\"SAMME\")\n",
    "    bagb.fit(X_train,Y_train)\n",
    "\n",
    "# Predicting\n",
    "    bagb_pred = bagb.predict(X_test)\n",
    "\n",
    "# Finding mispredicted samples\n",
    "    bagb_verror = np.asarray([int(bagb_pred[i] != Y_test[i]) for i in range(0,ts)])\n",
    "    bagb_error = np.sum(bagb_verror)\n",
    "    bagb_ccidx = np.where(bagb_verror == 0)\n",
    "    bagb_mcidx = np.where(bagb_verror == 1)\n",
    "\n",
    "    bagb_results.append(bagb_error/float(ts))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('miss-classified %age is ', 0.0)\n"
     ]
    }
   ],
   "source": [
    "print ('miss-classified %age is ', (sum(bagb_results)/len(bagb_results)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "99\n"
     ]
    }
   ],
   "source": [
    "# --------------------\n",
    "# AdaBoost Classifier\n",
    "# --------------------\n",
    "\n",
    "adab_results = []\n",
    "for j in range(100):\n",
    "    clear_output()\n",
    "    print j\n",
    "    X_train, X_test, Y_train, Y_test = train_test_split(X, Y, train_size=0.7)\n",
    "    ts = Y_test.shape[0]\n",
    "    adab = AdaBoostClassifier(DecisionTreeClassifier(max_depth=10), n_estimators=20)\n",
    "    #adab = AdaBoostClassifier(DecisionTreeClassifier(max_depth=10), n_estimators=20,learning_rate=1.2,algorithm=\"SAMME\")\n",
    "    adab.fit(X_train,Y_train)\n",
    "\n",
    "    # Predicting\n",
    "    adab_pred = adab.predict(X_test)\n",
    "\n",
    "    # Finding mispredicted samples\n",
    "    adab_verror = np.asarray([int(adab_pred[i] != Y_test[i]) for i in range(0,ts)])\n",
    "    adab_error = np.sum(adab_verror)\n",
    "    adab_ccidx = np.where(adab_verror == 0)\n",
    "    adab_mcidx = np.where(adab_verror == 1)\n",
    "\n",
    "    adab_results.append(adab_error/float(ts))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('miss-classified %age is ', 0.0)\n"
     ]
    }
   ],
   "source": [
    "print ('miss-classified %age is ', (sum(adab_results)/len(adab_results)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "99\n"
     ]
    }
   ],
   "source": [
    "# --------------------\n",
    "# Gradient Boosting Classifier\n",
    "# --------------------\n",
    "gradb_results = []\n",
    "for j in range(100):\n",
    "    clear_output()\n",
    "    print j\n",
    "    X_train, X_test, Y_train, Y_test = train_test_split(X, Y, train_size=0.7)\n",
    "    ts = Y_test.shape[0]\n",
    "    gradb = GradientBoostingClassifier(max_depth=10, n_estimators=20)\n",
    "    gradb.fit(X_train,Y_train)\n",
    "\n",
    "    # Predicting\n",
    "    gradb_pred = gradb.predict(X_test)\n",
    "\n",
    "    # Finding mispredicted samples\n",
    "    gradb_verror = np.asarray([int(gradb_pred[i] != Y_test[i]) for i in range(0,ts)])\n",
    "    gradb_error = np.sum(gradb_verror)\n",
    "    gradb_ccidx = np.where(gradb_verror == 0)\n",
    "    gradb_mcidx = np.where(gradb_verror == 1)\n",
    "\n",
    "\n",
    "    gradb_results.append(gradb_error/float(ts))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('miss-classified %age is ', 0.0)\n"
     ]
    }
   ],
   "source": [
    "print ('miss-classified %age is ', (sum(gradb_results)/len(gradb_results)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "99\n"
     ]
    }
   ],
   "source": [
    "# --------------------\n",
    "# Random Forest Classifier\n",
    "# --------------------\n",
    "rdf_results = []\n",
    "for j in range(100):\n",
    "    clear_output()\n",
    "    print j\n",
    "    X_train, X_test, Y_train, Y_test = train_test_split(X, Y, train_size=0.7)\n",
    "    ts = Y_test.shape[0]\n",
    "    rdf = RandomForestClassifier(max_depth=10, n_estimators=20)\n",
    "    rdf.fit(X_train,Y_train)\n",
    "\n",
    "    # Predicting\n",
    "    rdf_pred = rdf.predict(X_test)\n",
    "\n",
    "    # Finding mispredicted samples\n",
    "    rdf_verror = np.asarray([int(rdf_pred[i] != Y_test[i]) for i in range(0,ts)])\n",
    "    rdf_error = np.sum(rdf_verror)\n",
    "    rdf_ccidx = np.where(rdf_verror == 0)\n",
    "    rdf_mcidx = np.where(rdf_verror == 1)\n",
    "\n",
    "    rdf_results.append(rdf_error/float(ts))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('miss-classified %age is ', 1.6516792071939805e-05)\n"
     ]
    }
   ],
   "source": [
    "print ('miss-classified %age is ', (sum(rdf_results)/len(rdf_results)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "99\n"
     ]
    }
   ],
   "source": [
    "# --------------------\n",
    "# Naive Bayes Classifier\n",
    "# --------------------\n",
    "nb_results = []\n",
    "for i in range(100):\n",
    "    clear_output()\n",
    "    print i\n",
    "    \n",
    "    X_train, X_test, Y_train, Y_test = train_test_split(X, Y, train_size=0.7)\n",
    "    ts = Y_test.shape[0]\n",
    "    nb = naive_bayes.GaussianNB()\n",
    "    nb.fit(X_train,Y_train)\n",
    "    \n",
    "    # Predicting\n",
    "    nb_pred = nb.predict(X_test)\n",
    "    \n",
    "    # Finding mispredicted samples\n",
    "    nb_verror = np.asarray([int(nb_pred[i] != Y_test[i]) for i in range(0,ts)])\n",
    "    nb_error = np.sum(nb_verror)\n",
    "    nb_ccidx = np.where(nb_verror == 0)\n",
    "    nb_mcidx = np.where(nb_verror == 1)\n",
    "\n",
    "    nb_results.append(nb_error/float(ts))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('misclassified %age for Decision Tree is ', 0.0, '%')\n",
      "('misclassified %age for bagging is ', 0.0, '%')\n",
      "('misclassified %age for adaboost is ', 0.0, '%')\n",
      "('misclassified %age for gradiant boosting is ', 0.0, '%')\n",
      "('misclassified %age for Random Forest is ', 0.0016516792071939804, '%')\n",
      "('misclassified %age for Navie Bayes is ', 6.4603596990273431, '%')\n"
     ]
    }
   ],
   "source": [
    "#print ('misclassified %age for SVM Linear is ', (sum(SVMLinear_results)/len(SVMLinear_results))*100, '%')\n",
    "#print ('misclassified %age for SVM Kernels is ', (sum(SVMKernel_results)/len(SVMKernel_results))*100, '%')\n",
    "print ('misclassified %age for Decision Tree is ', (sum(DTC_results)/len(DTC_results))*100, '%')\n",
    "print ('misclassified %age for bagging is ', (sum(bagb_results)/len(bagb_results))*100, '%')\n",
    "print ('misclassified %age for adaboost is ', (sum(adab_results)/len(adab_results))*100, '%')\n",
    "print ('misclassified %age for gradiant boosting is ', (sum(gradb_results)/len(gradb_results))*100, '%')\n",
    "print ('misclassified %age for Random Forest is ', (sum(rdf_results)/len(rdf_results))*100, '%')\n",
    "print ('misclassified %age for Navie Bayes is ', (sum(nb_results)/len(nb_results))*100, '%')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "max depth:  1  Error: 17.53 %\n",
      "max depth:  2  Error: 10.42 %\n",
      "max depth:  3  Error: 3.45 %\n",
      "max depth:  4  Error: 1.87 %\n",
      "max depth:  5  Error: 0.89 %\n",
      "max depth:  6  Error: 0.14 %\n",
      "max depth:  7  Error: 0.02 %\n",
      "max depth:  8  Error: 0.0 %\n",
      "max depth:  9  Error: 0.0 %\n",
      "max depth:  10  Error: 0.0 %\n",
      "max depth:  11  Error: 0.0 %\n",
      "max depth:  12  Error: 0.0 %\n",
      "max depth:  13  Error: 0.0 %\n",
      "max depth:  14  Error: 0.0 %\n",
      "max depth:  15  Error: 0.0 %\n",
      "max depth:  16  Error: 0.0 %\n",
      "max depth:  17  Error: 0.0 %\n",
      "max depth:  18  Error: 0.0 %\n",
      "max depth:  19  Error: 0.0 %\n"
     ]
    }
   ],
   "source": [
    "# --------------------\n",
    "# Random Forest Classifier\n",
    "# --------------------\n",
    "for k in np.arange(1,20,1):\n",
    "    rdf_results = []\n",
    "    for j in range(20):\n",
    "        \n",
    "        X_train, X_test, Y_train, Y_test = train_test_split(X, Y, train_size=0.6,  random_state=3)\n",
    "        ts = Y_test.shape[0]\n",
    "        rdf = RandomForestClassifier(max_depth=k, n_estimators=20)\n",
    "        rdf.fit(X_train,Y_train)\n",
    "\n",
    "        # Predicting\n",
    "        rdf_pred = rdf.predict(X_test)\n",
    "\n",
    "        # Finding mispredicted samples\n",
    "        rdf_verror = np.asarray([int(rdf_pred[i] != Y_test[i]) for i in range(0,ts)])\n",
    "        rdf_error = np.sum(rdf_verror)\n",
    "        rdf_ccidx = np.where(rdf_verror == 0)\n",
    "        rdf_mcidx = np.where(rdf_verror == 1)\n",
    "\n",
    "        rdf_results.append(rdf_error/float(ts))\n",
    "    print 'max depth: ',k, ' Error:', round((sum(rdf_results)/len(rdf_results))*100,2),'%'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "n_estimators:  1  Error: 0.6 %\n",
      "n_estimators:  2  Error: 0.34 %\n",
      "n_estimators:  3  Error: 0.15 %\n",
      "n_estimators:  4  Error: 0.05 %\n",
      "n_estimators:  5  Error: 0.02 %\n",
      "n_estimators:  6  Error: 0.02 %\n",
      "n_estimators:  7  Error: 0.01 %\n",
      "n_estimators:  8  Error: 0.01 %\n",
      "n_estimators:  9  Error: 0.01 %\n",
      "n_estimators:  10  Error: 0.0 %\n",
      "n_estimators:  11  Error: 0.0 %\n",
      "n_estimators:  12  Error: 0.0 %\n",
      "n_estimators:  13  Error: 0.0 %\n",
      "n_estimators:  14  Error: 0.0 %\n",
      "n_estimators:  15  Error: 0.0 %\n",
      "n_estimators:  16  Error: 0.0 %\n",
      "n_estimators:  17  Error: 0.0 %\n",
      "n_estimators:  18  Error: 0.0 %\n",
      "n_estimators:  19  Error: 0.0 %\n",
      "n_estimators:  20  Error: 0.0 %\n",
      "n_estimators:  21  Error: 0.0 %\n",
      "n_estimators:  22  Error: 0.0 %\n",
      "n_estimators:  23  Error: 0.0 %\n",
      "n_estimators:  24  Error: 0.0 %\n",
      "n_estimators:  25  Error: 0.0 %\n",
      "n_estimators:  26  Error: 0.0 %\n",
      "n_estimators:  27  Error: 0.0 %\n",
      "n_estimators:  28  Error: 0.0 %\n",
      "n_estimators:  29  Error: 0.0 %\n",
      "n_estimators:  30  Error: 0.0 %\n",
      "n_estimators:  31  Error: 0.0 %\n",
      "n_estimators:  32  Error: 0.0 %\n",
      "n_estimators:  33  Error: 0.0 %\n",
      "n_estimators:  34  Error: 0.0 %\n",
      "n_estimators:  35  Error: 0.0 %\n",
      "n_estimators:  36  Error: 0.0 %\n",
      "n_estimators:  37  Error: 0.0 %\n",
      "n_estimators:  38  Error: 0.0 %\n",
      "n_estimators:  39  Error: 0.0 %\n",
      "n_estimators:  40  Error: 0.0 %\n",
      "n_estimators:  41  Error: 0.0 %\n",
      "n_estimators:  42  Error: 0.0 %\n",
      "n_estimators:  43  Error: 0.0 %\n",
      "n_estimators:  44  Error: 0.0 %\n",
      "n_estimators:  45  Error: 0.0 %\n",
      "n_estimators:  46  Error: 0.0 %\n",
      "n_estimators:  47  Error: 0.0 %\n",
      "n_estimators:  48  Error: 0.0 %\n",
      "n_estimators:  49  Error: 0.0 %\n"
     ]
    }
   ],
   "source": [
    "# --------------------\n",
    "# Random Forest Classifier\n",
    "# --------------------\n",
    "for k in np.arange(1,50,1):\n",
    "    rdf_results = []\n",
    "    for j in range(20):\n",
    "        \n",
    "        X_train, X_test, Y_train, Y_test = train_test_split(X, Y, train_size=0.6,  random_state=3)\n",
    "        ts = Y_test.shape[0]\n",
    "        rdf = RandomForestClassifier(max_depth=10, n_estimators=k)\n",
    "        rdf.fit(X_train,Y_train)\n",
    "\n",
    "        # Predicting\n",
    "        rdf_pred = rdf.predict(X_test)\n",
    "\n",
    "        # Finding mispredicted samples\n",
    "        rdf_verror = np.asarray([int(rdf_pred[i] != Y_test[i]) for i in range(0,ts)])\n",
    "        rdf_error = np.sum(rdf_verror)\n",
    "        rdf_ccidx = np.where(rdf_verror == 0)\n",
    "        rdf_mcidx = np.where(rdf_verror == 1)\n",
    "\n",
    "        rdf_results.append(rdf_error/float(ts))\n",
    "    print 'n_estimators: ',k, ' Error:', round((sum(rdf_results)/len(rdf_results))*100,2),'%'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "99\n",
      "miss-classified %age is  1.6518004625e-06\n"
     ]
    }
   ],
   "source": [
    "# --------------------\n",
    "# Random Forest Classifier\n",
    "# --------------------\n",
    "rdf_results = []\n",
    "for j in range(100):\n",
    "    clear_output()\n",
    "    print j\n",
    "    X_train, X_test, Y_train, Y_test = train_test_split(X, Y, train_size=0.7, random_state=3)\n",
    "    ts = Y_test.shape[0]\n",
    "    rdf = RandomForestClassifier(max_depth=8, n_estimators=10)\n",
    "    rdf.fit(X_train,Y_train)\n",
    "\n",
    "    # Predicting\n",
    "    rdf_pred = rdf.predict(X_test)\n",
    "\n",
    "    # Finding mispredicted samples\n",
    "    rdf_verror = np.asarray([int(rdf_pred[i] != Y_test[i]) for i in range(0,ts)])\n",
    "    rdf_error = np.sum(rdf_verror)\n",
    "    rdf_ccidx = np.where(rdf_verror == 0)\n",
    "    rdf_mcidx = np.where(rdf_verror == 1)\n",
    "\n",
    "    rdf_results.append(rdf_error/float(ts))\n",
    "\n",
    "print 'miss-classified %age is ', (sum(rdf_results)/len(rdf_results))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "PUI2016_Python2",
   "language": "python",
   "name": "pui2016_python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}