Support Vector Machines for Energy-Efficiency Classification.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Support Vector Machines for Energy-Efficiency Classification\n",
    "<hr>\n",
    "### Energy-Efficiency Classification\n",
    "\n",
    "We would be studying Support Vector Machines by looking at the Energy Efficiency Dataset which contains assessment of the heating load and cooling load requirements of buildings (that is, energy efficiency) as a function of building parameters.  \n",
    "\n",
    "The creators of the dataset performed energy analysis using 12 different building shapes simulated in Ecotect. The buildings differed with respect to the glazing area, the glazing area distribution, and the orientation, amongst other parameters. They simulated various settings as functions of the afore-mentioned characteristics to obtain 768 building shapes. \n",
    "\n",
    "The dataset comprises 768 samples and 8 features, aiming to predict two real valued responses. It can also be used as a multi-class classification problem if the response is rounded to the nearest integer.\n",
    "\n",
    "You can download the dataset from : https://archive.ics.uci.edu/ml/datasets/Energy+efficiency\n",
    "\n",
    "The dataset contains eight attributes (or features, denoted by X1...X8) and two responses (or outcomes, denoted by y1 and y2). Our aim will be to use the eight features to classify the two responses y1 and y2 as high or low.\n",
    "\n",
    "Specifically:\n",
    "\n",
    "* X1 Relative Compactness\n",
    "* X2 Surface Area\n",
    "* X3 Wall Area\n",
    "* X4 Roof Area\n",
    "* X5 Overall Height\n",
    "* X6 Orientation\n",
    "* X7 Glazing Area\n",
    "* X8 Glazing Area Distribution\n",
    "* y1 Heating Load\n",
    "* y2 Cooling Load"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Save the downloaded `ENB2012_data.xlsx` in the current directory as your code. We then load it into a Pandas Dataframe using the read_excel() function. For this to work, you might have to run **`pip install xlrd`** if you don't already have the 'xlrd' module."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "# read .csv from provided dataset\n",
    "excel_filename=\"ENB2012_data.xlsx\"\n",
    "\n",
    "# df=pd.read_csv(csv_filename,index_col=0)\n",
    "df=pd.read_excel(excel_filename)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We take a look at our data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>X1</th>\n",
       "      <th>X2</th>\n",
       "      <th>X3</th>\n",
       "      <th>X4</th>\n",
       "      <th>X5</th>\n",
       "      <th>X6</th>\n",
       "      <th>X7</th>\n",
       "      <th>X8</th>\n",
       "      <th>Y1</th>\n",
       "      <th>Y2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.98</td>\n",
       "      <td>514.5</td>\n",
       "      <td>294.0</td>\n",
       "      <td>110.25</td>\n",
       "      <td>7.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>15.55</td>\n",
       "      <td>21.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.98</td>\n",
       "      <td>514.5</td>\n",
       "      <td>294.0</td>\n",
       "      <td>110.25</td>\n",
       "      <td>7.0</td>\n",
       "      <td>3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>15.55</td>\n",
       "      <td>21.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.98</td>\n",
       "      <td>514.5</td>\n",
       "      <td>294.0</td>\n",
       "      <td>110.25</td>\n",
       "      <td>7.0</td>\n",
       "      <td>4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>15.55</td>\n",
       "      <td>21.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.98</td>\n",
       "      <td>514.5</td>\n",
       "      <td>294.0</td>\n",
       "      <td>110.25</td>\n",
       "      <td>7.0</td>\n",
       "      <td>5</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>15.55</td>\n",
       "      <td>21.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.90</td>\n",
       "      <td>563.5</td>\n",
       "      <td>318.5</td>\n",
       "      <td>122.50</td>\n",
       "      <td>7.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>20.84</td>\n",
       "      <td>28.28</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     X1     X2     X3      X4   X5  X6   X7  X8     Y1     Y2\n",
       "0  0.98  514.5  294.0  110.25  7.0   2  0.0   0  15.55  21.33\n",
       "1  0.98  514.5  294.0  110.25  7.0   3  0.0   0  15.55  21.33\n",
       "2  0.98  514.5  294.0  110.25  7.0   4  0.0   0  15.55  21.33\n",
       "3  0.98  514.5  294.0  110.25  7.0   5  0.0   0  15.55  21.33\n",
       "4  0.90  563.5  318.5  122.50  7.0   2  0.0   0  20.84  28.28"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "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>X1</th>\n",
       "      <th>X2</th>\n",
       "      <th>X3</th>\n",
       "      <th>X4</th>\n",
       "      <th>X5</th>\n",
       "      <th>X6</th>\n",
       "      <th>X7</th>\n",
       "      <th>X8</th>\n",
       "      <th>Y1</th>\n",
       "      <th>Y2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>768.000000</td>\n",
       "      <td>768.000000</td>\n",
       "      <td>768.000000</td>\n",
       "      <td>768.000000</td>\n",
       "      <td>768.00000</td>\n",
       "      <td>768.000000</td>\n",
       "      <td>768.000000</td>\n",
       "      <td>768.00000</td>\n",
       "      <td>768.000000</td>\n",
       "      <td>768.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.764167</td>\n",
       "      <td>671.708333</td>\n",
       "      <td>318.500000</td>\n",
       "      <td>176.604167</td>\n",
       "      <td>5.25000</td>\n",
       "      <td>3.500000</td>\n",
       "      <td>0.234375</td>\n",
       "      <td>2.81250</td>\n",
       "      <td>22.307195</td>\n",
       "      <td>24.587760</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.105777</td>\n",
       "      <td>88.086116</td>\n",
       "      <td>43.626481</td>\n",
       "      <td>45.165950</td>\n",
       "      <td>1.75114</td>\n",
       "      <td>1.118763</td>\n",
       "      <td>0.133221</td>\n",
       "      <td>1.55096</td>\n",
       "      <td>10.090204</td>\n",
       "      <td>9.513306</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.620000</td>\n",
       "      <td>514.500000</td>\n",
       "      <td>245.000000</td>\n",
       "      <td>110.250000</td>\n",
       "      <td>3.50000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>6.010000</td>\n",
       "      <td>10.900000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.682500</td>\n",
       "      <td>606.375000</td>\n",
       "      <td>294.000000</td>\n",
       "      <td>140.875000</td>\n",
       "      <td>3.50000</td>\n",
       "      <td>2.750000</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>1.75000</td>\n",
       "      <td>12.992500</td>\n",
       "      <td>15.620000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.750000</td>\n",
       "      <td>673.750000</td>\n",
       "      <td>318.500000</td>\n",
       "      <td>183.750000</td>\n",
       "      <td>5.25000</td>\n",
       "      <td>3.500000</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>3.00000</td>\n",
       "      <td>18.950000</td>\n",
       "      <td>22.080000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.830000</td>\n",
       "      <td>741.125000</td>\n",
       "      <td>343.000000</td>\n",
       "      <td>220.500000</td>\n",
       "      <td>7.00000</td>\n",
       "      <td>4.250000</td>\n",
       "      <td>0.400000</td>\n",
       "      <td>4.00000</td>\n",
       "      <td>31.667500</td>\n",
       "      <td>33.132500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>0.980000</td>\n",
       "      <td>808.500000</td>\n",
       "      <td>416.500000</td>\n",
       "      <td>220.500000</td>\n",
       "      <td>7.00000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>0.400000</td>\n",
       "      <td>5.00000</td>\n",
       "      <td>43.100000</td>\n",
       "      <td>48.030000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               X1          X2          X3          X4         X5          X6  \\\n",
       "count  768.000000  768.000000  768.000000  768.000000  768.00000  768.000000   \n",
       "mean     0.764167  671.708333  318.500000  176.604167    5.25000    3.500000   \n",
       "std      0.105777   88.086116   43.626481   45.165950    1.75114    1.118763   \n",
       "min      0.620000  514.500000  245.000000  110.250000    3.50000    2.000000   \n",
       "25%      0.682500  606.375000  294.000000  140.875000    3.50000    2.750000   \n",
       "50%      0.750000  673.750000  318.500000  183.750000    5.25000    3.500000   \n",
       "75%      0.830000  741.125000  343.000000  220.500000    7.00000    4.250000   \n",
       "max      0.980000  808.500000  416.500000  220.500000    7.00000    5.000000   \n",
       "\n",
       "               X7         X8          Y1          Y2  \n",
       "count  768.000000  768.00000  768.000000  768.000000  \n",
       "mean     0.234375    2.81250   22.307195   24.587760  \n",
       "std      0.133221    1.55096   10.090204    9.513306  \n",
       "min      0.000000    0.00000    6.010000   10.900000  \n",
       "25%      0.100000    1.75000   12.992500   15.620000  \n",
       "50%      0.250000    3.00000   18.950000   22.080000  \n",
       "75%      0.400000    4.00000   31.667500   33.132500  \n",
       "max      0.400000    5.00000   43.100000   48.030000  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As you can see via the df.describe() method, the mean for our target variables Y1 and Y2 is approx 22 and 24 respectively. Since the data for y1 and y2 is continuous, we will use the following piece of code to convert it into either high or low.\n",
    "\n",
    "We do this by defining all those y1 and y2 values above or 20 are high and those below 20 are low. We then map 1 for high and 0 for low."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# handle Y1 and Y2 attrubte to binary [High Hl and CL; and Low HL and Cl]\n",
    "high = df.Y1 >= 20\n",
    "low = df.Y1 < 20\n",
    "df.loc[high,'Y1'] = 1\n",
    "df.loc[low,'Y1'] = 0\n",
    "\n",
    "high2 = df.Y2 >= 20\n",
    "low2 = df.Y2 < 20\n",
    "df.loc[high2,'Y2'] = 1\n",
    "df.loc[low2,'Y2'] = 0"
   ]
  },
  {
   "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>X1</th>\n",
       "      <th>X2</th>\n",
       "      <th>X3</th>\n",
       "      <th>X4</th>\n",
       "      <th>X5</th>\n",
       "      <th>X6</th>\n",
       "      <th>X7</th>\n",
       "      <th>X8</th>\n",
       "      <th>Y1</th>\n",
       "      <th>Y2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.98</td>\n",
       "      <td>514.5</td>\n",
       "      <td>294.0</td>\n",
       "      <td>110.25</td>\n",
       "      <td>7.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.98</td>\n",
       "      <td>514.5</td>\n",
       "      <td>294.0</td>\n",
       "      <td>110.25</td>\n",
       "      <td>7.0</td>\n",
       "      <td>3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.98</td>\n",
       "      <td>514.5</td>\n",
       "      <td>294.0</td>\n",
       "      <td>110.25</td>\n",
       "      <td>7.0</td>\n",
       "      <td>4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.98</td>\n",
       "      <td>514.5</td>\n",
       "      <td>294.0</td>\n",
       "      <td>110.25</td>\n",
       "      <td>7.0</td>\n",
       "      <td>5</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.90</td>\n",
       "      <td>563.5</td>\n",
       "      <td>318.5</td>\n",
       "      <td>122.50</td>\n",
       "      <td>7.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     X1     X2     X3      X4   X5  X6   X7  X8   Y1   Y2\n",
       "0  0.98  514.5  294.0  110.25  7.0   2  0.0   0  0.0  1.0\n",
       "1  0.98  514.5  294.0  110.25  7.0   3  0.0   0  0.0  1.0\n",
       "2  0.98  514.5  294.0  110.25  7.0   4  0.0   0  0.0  1.0\n",
       "3  0.98  514.5  294.0  110.25  7.0   5  0.0   0  0.0  1.0\n",
       "4  0.90  563.5  318.5  122.50  7.0   2  0.0   0  1.0  1.0"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we have converted the dataset into a binary classification task. We can observe that the data for y1 and y2 is either 0 or 1, i.e the Energy Efficiency is low if its 0 or high if its 1."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>X1</th>\n",
       "      <th>X2</th>\n",
       "      <th>X3</th>\n",
       "      <th>X4</th>\n",
       "      <th>X5</th>\n",
       "      <th>X6</th>\n",
       "      <th>X7</th>\n",
       "      <th>X8</th>\n",
       "      <th>Y1</th>\n",
       "      <th>Y2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>768.000000</td>\n",
       "      <td>768.000000</td>\n",
       "      <td>768.000000</td>\n",
       "      <td>768.000000</td>\n",
       "      <td>768.00000</td>\n",
       "      <td>768.000000</td>\n",
       "      <td>768.000000</td>\n",
       "      <td>768.00000</td>\n",
       "      <td>768.000000</td>\n",
       "      <td>768.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.764167</td>\n",
       "      <td>671.708333</td>\n",
       "      <td>318.500000</td>\n",
       "      <td>176.604167</td>\n",
       "      <td>5.25000</td>\n",
       "      <td>3.500000</td>\n",
       "      <td>0.234375</td>\n",
       "      <td>2.81250</td>\n",
       "      <td>0.483073</td>\n",
       "      <td>0.546875</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.105777</td>\n",
       "      <td>88.086116</td>\n",
       "      <td>43.626481</td>\n",
       "      <td>45.165950</td>\n",
       "      <td>1.75114</td>\n",
       "      <td>1.118763</td>\n",
       "      <td>0.133221</td>\n",
       "      <td>1.55096</td>\n",
       "      <td>0.500039</td>\n",
       "      <td>0.498122</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.620000</td>\n",
       "      <td>514.500000</td>\n",
       "      <td>245.000000</td>\n",
       "      <td>110.250000</td>\n",
       "      <td>3.50000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.682500</td>\n",
       "      <td>606.375000</td>\n",
       "      <td>294.000000</td>\n",
       "      <td>140.875000</td>\n",
       "      <td>3.50000</td>\n",
       "      <td>2.750000</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>1.75000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.750000</td>\n",
       "      <td>673.750000</td>\n",
       "      <td>318.500000</td>\n",
       "      <td>183.750000</td>\n",
       "      <td>5.25000</td>\n",
       "      <td>3.500000</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>3.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.830000</td>\n",
       "      <td>741.125000</td>\n",
       "      <td>343.000000</td>\n",
       "      <td>220.500000</td>\n",
       "      <td>7.00000</td>\n",
       "      <td>4.250000</td>\n",
       "      <td>0.400000</td>\n",
       "      <td>4.00000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>0.980000</td>\n",
       "      <td>808.500000</td>\n",
       "      <td>416.500000</td>\n",
       "      <td>220.500000</td>\n",
       "      <td>7.00000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>0.400000</td>\n",
       "      <td>5.00000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               X1          X2          X3          X4         X5          X6  \\\n",
       "count  768.000000  768.000000  768.000000  768.000000  768.00000  768.000000   \n",
       "mean     0.764167  671.708333  318.500000  176.604167    5.25000    3.500000   \n",
       "std      0.105777   88.086116   43.626481   45.165950    1.75114    1.118763   \n",
       "min      0.620000  514.500000  245.000000  110.250000    3.50000    2.000000   \n",
       "25%      0.682500  606.375000  294.000000  140.875000    3.50000    2.750000   \n",
       "50%      0.750000  673.750000  318.500000  183.750000    5.25000    3.500000   \n",
       "75%      0.830000  741.125000  343.000000  220.500000    7.00000    4.250000   \n",
       "max      0.980000  808.500000  416.500000  220.500000    7.00000    5.000000   \n",
       "\n",
       "               X7         X8          Y1          Y2  \n",
       "count  768.000000  768.00000  768.000000  768.000000  \n",
       "mean     0.234375    2.81250    0.483073    0.546875  \n",
       "std      0.133221    1.55096    0.500039    0.498122  \n",
       "min      0.000000    0.00000    0.000000    0.000000  \n",
       "25%      0.100000    1.75000    0.000000    0.000000  \n",
       "50%      0.250000    3.00000    0.000000    1.000000  \n",
       "75%      0.400000    4.00000    1.000000    1.000000  \n",
       "max      0.400000    5.00000    1.000000    1.000000  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We specify our features and target variables:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "features=list(df.columns[:-2])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X = df[features]\n",
    "y1 = df['Y1']\n",
    "y2 = df['Y2']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 1.,  0.])"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y2.unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((768, 8), (768,), (768,))"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.shape, y1.shape, y2.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To evaluate how well a trained model performs on unseen data, we will further split the dataset into separate training and test datasets. Splitting data into 60% training and 40% test data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\users\\hp\\appdata\\local\\programs\\python\\python35\\lib\\site-packages\\sklearn\\cross_validation.py:44: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.\n",
      "  \"This module will be removed in 0.20.\", DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "from sklearn import cross_validation, metrics\n",
    "from sklearn.cross_validation import train_test_split\n",
    "\n",
    "# split dataset to 60% training and 40% testing\n",
    "X_train, X_test, y_train, y_test = cross_validation.train_test_split(X,y1, test_size=0.4, random_state=0)\n",
    "X_train2, X_test2, y_train2, y_test2 = cross_validation.train_test_split(X,y2, test_size=0.4, random_state=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "### Feature importances with forests of trees\n",
    "\n",
    "This examples shows the use of forests of trees to evaluate the importance of features on an artificial classification task. The red bars are the feature importances of the forest, along with their inter-trees variability.\n",
    "\n",
    "For Classification into Y1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Feature ranking:\n",
      "1. feature 4 - X5 (0.350375) \n",
      "2. feature 3 - X4 (0.312105) \n",
      "3. feature 1 - X2 (0.139400) \n",
      "4. feature 0 - X1 (0.118307) \n",
      "5. feature 2 - X3 (0.029773) \n",
      "6. feature 6 - X7 (0.025721) \n",
      "7. feature 7 - X8 (0.021366) \n",
      "8. feature 5 - X6 (0.002953) \n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABHgAAANDCAYAAADW6KijAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAgML9f9Ya8MNOUos8AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xa1f4ef0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from sklearn.ensemble import ExtraTreesClassifier\n",
    "\n",
    "# Build a classification task using 3 informative features\n",
    "\n",
    "# Build a forest and compute the feature importances\n",
    "forest = ExtraTreesClassifier(n_estimators=250,\n",
    "                              random_state=0)\n",
    "\n",
    "forest.fit(X, y1)\n",
    "importances = forest.feature_importances_\n",
    "std = np.std([tree.feature_importances_ for tree in forest.estimators_],\n",
    "             axis=0)\n",
    "indices = np.argsort(importances)[::-1]\n",
    "\n",
    "# Print the feature ranking\n",
    "print(\"Feature ranking:\")\n",
    "\n",
    "for f in range(X.shape[1]):\n",
    "    print(\"%d. feature %d - %s (%f) \" % (f + 1, indices[f], features[indices[f]], importances[indices[f]]))\n",
    "\n",
    "# Plot the feature importances of the forest\n",
    "plt.figure(num=None, figsize=(14, 10), dpi=80, facecolor='w', edgecolor='k')\n",
    "plt.title(\"Feature importances\")\n",
    "plt.bar(range(X.shape[1]), importances[indices],\n",
    "       color=\"r\", yerr=std[indices], align=\"center\")\n",
    "plt.xticks(range(X.shape[1]), indices)\n",
    "plt.xlim([-1, X.shape[1]])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1. feature 4 - X5 (0.350375)\n",
      "2. feature 3 - X4 (0.312105)\n",
      "3. feature 1 - X2 (0.139400)\n",
      "4. feature 0 - X1 (0.118307)\n",
      "5. feature 2 - X3 (0.029773)\n"
     ]
    }
   ],
   "source": [
    "for f in range(5):\n",
    "    print(\"%d. feature %d - %s (%f)\" % (f + 1, indices[f], features[indices[f]] ,importances[indices[f]]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "best_features = []\n",
    "for i in indices[:5]:\n",
    "    best_features.append(features[i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAqcAAAINCAYAAAADYtaMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAAPYQAAD2EBqD+naQAAIABJREFUeJzt3XtY1HXe//HXDKAcAjxAYBqGtSpagUy5aquZUJSVeW3l\nRhFid63GmqbWbuZ6oDWzwy7ddkt0siyL27xrt22t3MK0bXXLQFszJE2RtKQmFSukED6/P/g568RB\nTIb5IM/Hdc1Vfuf7nXnDJ/DZd04OY4wRAAAAYAGnvwcAAAAAjiBOAQAAYA3iFAAAANYgTgEAAGAN\n4hQAAADWIE4BAABgDeIUAAAA1iBOAQAAYA3iFAAAANYgTgHAB3bt2iWn06lnn33W36MAQLtCnAI4\nYUuXLpXT6Wz0cvfdd/vsfl9//XXl5OT47PZPlMPh8PcIJ+TRRx/V0qVL/T0GgA4m0N8DADg5OBwO\n/eEPf9AZZ5zhtf3ss8/22X2+9tprysvL09y5c312Hz9V7969dejQIQUFBfl7lJ8sLy9P0dHRGj9+\nvL9HAdCBEKcAWs2ll16q5OTkNrs/Y4xPbreqqkqhoaEnfDudOnVqhWna3qFDhxQSEuLvMQB0UDys\nD6BNLVu2TOedd55CQ0PVvXt3paena/fu3V77vPvuuxo3bpx69+6t4OBgxcXFafr06aqurvbsM2HC\nBOXl5UmS5ykEAQEBkqQ1a9bI6XTqnXfe8brdxp4HmpWVpfDwcO3YsUOjR49WRESEMjIyPNe/9957\nuvTSS9WlSxeFhYVp5MiRWrdu3TG/zubu67PPPtMVV1yh8PBw9erVy/N1bN68WSkpKTrllFN0xhln\nqKCgwOs2jzx94h//+IcmTpyoqKgoRUZGavz48Tpw4ECDGfLy8nT22WcrODhYPXv21OTJk1VZWem1\nz8iRI3XuueequLhYI0aMUFhYmO6++27Fx8dry5Ytnu+l0+nUqFGjJEn79+/XHXfcoXPPPVfh4eGK\njIzU6NGj9e9//9vrtteuXSun06kVK1bo3nvv1emnn66QkBClpqbq008/bTDve++9p9GjR6tbt246\n5ZRTlJiYqEWLFnntU1paqmuuuUbdu3dXSEiIzj//fL366qte+xw+fFg5OTnq27evQkJCFBUVpeHD\nh6uwsPBYywbAApw5BdBqKisr9fXXX3tt6969u+ff7733Xs2ZM0fXXXedbrnlFn311VdatGiRLrzw\nQm3cuFERERGSpBUrVujQoUPKzs5W9+7d9f777+uRRx7Rnj17tHz5cknSpEmT9Pnnn+utt97S888/\n73UW1eFwtPj5ng6HQ4cPH1ZaWpqGDx+uP/7xj56zpqtXr9bo0aN13nnnad68eXI6nXr66ac1atQo\nvfvuuzrvvPOO6/vjcDhUV1enyy67TBdeeKEefPBBPf/887rtttsUFhamWbNmKSMjQ1dffbXy8/M1\nfvx4DRs2TL179/a6ncmTJ6tr167KyclRaWmp8vLyVF5errffftuzz7x583TPPffokksuUXZ2tme/\nDz74QP/85z89Ie9wOOR2uzV69Ghdd911yszMVExMjC666CJNnjxZ4eHh+v3vfy9jjGJiYiRJO3bs\n0F//+ldde+21io+PV0VFhR577DGNHDlSH3/8sWJjY73mXbhwoQICAnTnnXeqsrJS999/vzIyMrR+\n/XrPPm+++aauvPJKnXbaabr99tsVGxurkpISrVy5UlOmTJEkbdmyRb/4xS/Uq1cvzZw5U2FhYXrx\nxRc1duxYvfzyy7rqqqskSXPnztXChQv161//Wueff74OHjyoDz74QMXFxUpJSTmuNQPgBwYATtAz\nzzxjHA5Hg4vT6fTss2vXLhMYGGgWLlzodeyWLVtMUFCQue+++zzbqqurG9zHwoULTUBAgPnss888\n2yZPnux1H0esWbPGOJ1Os3btWq/tZWVlxuFwmKVLl3q2ZWVlGafTaWbNmtXgdvr27WtGjx7tta26\nutr06dPHpKWlNfXtOOZ93X///Z5tBw4cMKGhoSYgIMCsWLHCs720tNQ4HA6Tk5Pj2Xbk+zx48GBz\n+PBhz/YHH3zQOJ1O8+qrrxpjjPnqq69M586dzWWXXeY10+LFi43T6TTPPPOMZ9vIkSON0+k0Tzzx\nRIOv4eyzzzYXXXRRg+0//PBDg227du0ywcHBZv78+Z5ta9asMQ6HwwwcONBr3kWLFhmn02m2bNli\njDGmtrbWxMfHmz59+piDBw82uO0jUlJSTFJSkqmpqfHafsEFF5h+/fp5/pyUlGSuvPLKJm8HgN14\nWB9Aq3A4HHr00Uf11ltveS5vvvmm5/qXXnpJxhhde+21+vrrrz2XU089VT/72c+8zvp17tzZ8+9V\nVVX6+uuvNXToUNXV1Wnjxo0+mX/SpElef960aZO2bdum9PR0r3m/+eYbpaSkNHjKwPH4r//6L8+/\nR0ZGql+/fgoLC9M111zj2d63b1916dJFO3bsaHD8r3/9a8+ZT0m69dZbFRAQoNdee01S/VnImpoa\n3X777V7H3XLLLQoPD9fKlSu9tnfu3FlZWVktnv/oF3nV1dVp3759Cg0NVb9+/VRcXNxg/5tuuslr\n3uHDh8sY4/naNm7cqLKyMt1+++0KDw9v9D7379+vt99+W9dee63nDP2RyyWXXKJt27bpiy++kCR1\n6dJFW7Zs0fbt21v8NQGwBw/rA2g1559/fpMviNq+fbvq6up01llnNbjO4XB4vXjos88+0+zZs/Xq\nq69q//79Xvv9+DmTrSEwMFC9evXy2rZt2zZJUmZmZqPHOJ1OVVZWKjIy8rjuKzg42OupDlJ9oP74\n/o9sP/rrl+q/Bz/+HoaFhalHjx4qKyuTJJWXl0uqD9yjBQUFqU+fPtq1a5fX9p49eyowsOV/HRhj\n9PDDD+vRRx/Vzp07VVtb65ktKiqqwf6nn36615+7du0qSZ6v7dNPP5XD4dDAgQObvM/t27fLGKPZ\ns2fr97//fYPrHQ6HvvzyS/Xo0UP33HOPxo4dq759++rss8/WpZdeqhtvvFHnnHNOi79GAP5DnAJo\nE3V1dXI6nXrjjTfkdDZ80OaUU07x7JeamqoDBw5o5syZnrOKe/bs0fjx41VXV3fM+2rq+aZHIurH\njj5Te/S8kvTHP/5RiYmJjR53ZObjcfQZxJZsNz56R4KjHe8r8488d/jmm2/W/Pnz1a1bNzmdTk2d\nOrXR9WmNr+3I7d5xxx1KS0trdJ8j0T58+HB9+umneuWVV/T3v/9dTz31lHJzc/XYY4/ppptuavF9\nAvAP4hRAmzjzzDNljNEZZ5zR6NnTIzZv3qxt27bpueee0w033ODZ/tZbbzXYt6kI7dq1q4wxDV7B\nfuTMYkvnlaTw8HDPq9RtYIzRtm3bdOGFF3q2fffdd/riiy90+eWXS5LnBVSlpaVe7ztbU1OjnTt3\n6uKLL27RfTX1/X3ppZc0atQoPf74417bDxw4oOjo6OP5ciT957+Njz76qMnvdZ8+fSTVn/1tyXp0\n6dJF48eP1/jx41VVVaXhw4dr3rx5xCnQDvCcUwBt4pe//KWcTmeTn+i0b98+Sf85y/bjM3APP/xw\ng1gKCwuTJB08eNBre+/evRUQENDgeaF5eXktfhW/y+XSmWeeqYceekjfffddg+vdbneLbscXHn/8\ncR0+fNjz57y8PNXW1mr06NGSpNTUVAUFBTV4G6Ynn3xSBw8e1BVXXNGi+wkLC2v0LaoCAgIanPVc\nsWKF9uzZc7xfiiQpOTlZ8fHxevjhh5t82kZ0dLRGjhypxx57THv37m1w/dHrceS/pSNCQ0N11lln\n6fvvv/9J8wFoW5w5BdAqjvUQbZ8+fTR//nzdfffd2rlzp8aOHet5f9G//OUvmjhxoqZPn67+/fvr\nzDPP1IwZM7R7925FRETopZdeajSSXC6XjDG67bbblJaWpoCAAP3qV79SRESErr32Wk+cnXnmmfrb\n3/6mr776qsVfj8Ph0JNPPqnRo0dr4MCBmjBhgnr27Kk9e/bo7bffVmRkpF555ZXj+ya1kh9++EEp\nKSkaN26ctm7dqkcffVTDhw/3RGdUVJRmzpype+65R5deeqnGjBnj2W/w4MFeZ6Sb43K5lJ+fr3vv\nvVdnnXWWTj31VF100UW64oor9Ic//EE33XSThg0bps2bN+v555/3nG0+XkdeTDdmzBglJSVpwoQJ\n6tGjh7Zu3aqPP/5Yr7/+uiRp8eLFGj58uM455xzdcsst6tOnjyoqKrR+/Xrt2bPH82K5AQMGaOTI\nkXK5XOrWrZs2bNig//u///O8JRUAy/nlPQIAnFSeeeYZ43Q6TVFR0TH3/fOf/2xGjBhhwsPDTXh4\nuBkwYICZMmWK2bZtm2efrVu3mksuucRERESYU0891UyaNMls3rzZOJ1Or7dmqq2tNVOnTjUxMTEm\nICDA622l3G63ufbaa80pp5xiunfvbrKzs83HH3/c4DaysrJMREREk/N++OGH5pprrjHR0dEmJCTE\nxMfHm+uuu868/fbbzX6dZWVlLb6vkSNHmnPPPbfB9vj4eDNmzBjPn498n//xj3+YSZMmme7du5uI\niAiTmZlp9u/f3+D4vLw8M2DAANO5c2fTo0cPM3nyZFNZWdmi+zbGmIqKCnPllVeayMhI43Q6PW8r\n9f3335s777zT9OzZ04SFhZkRI0aY9957z1x00UVm1KhRnuOPvKXXSy+9dMzvjTHGrFu3zqSlpZnI\nyEgTHh5ukpKSTF5entc+O3fuNFlZWea0004znTt3NqeffroZM2aMefnllz37LFiwwAwZMsR069bN\nhIWFmQEDBpiFCxd6vZ0VAHs5jGmDZ9sDAE7Y0qVLddNNN2nDhg1t+jGxANCWfP6c08WLFys+Pl4h\nISEaMmSINmzY0Oz+Dz/8sPr376/Q0FDPRxbyPCEAAICOwadxunz5cs2YMUM5OTnauHGjEhMTlZaW\n1uQLCV544QXNnDlTOTk52rp1q5YsWaLly5dr1qxZvhwTANoNHuwCcLLzaZzm5uZq4sSJyszMVP/+\n/ZWfn6/Q0FAtWbKk0f3Xr1+vX/ziF/rVr36luLg4paamKj09Xe+//74vxwSAdqOl7zYAAO2Vz+K0\npqZGRUVFSklJ8WxzOBxKTU3V+vXrGz1m2LBhKioq8jz0v2PHDr322mue9+4DgI5s/Pjxqq2t5fmm\nAE5qPnsrKbfbrdraWsXExHhtj4mJUWlpaaPHpKeny+126xe/+IWMMaqtrdWkSZP0u9/9rtn7WbVq\nlc4444zj/pQTAAAA+N6hQ4dUVlamtLS0Rj/m+GhWvc/pmjVrtGDBAuXn52vw4MHavn27pkyZoh49\nejT6WcqStGrVKmVkZLTxpAAAADhey5YtO+Z7LfssTqOiohQQEKCKigqv7RUVFYqNjW30mDlz5igz\nM1MTJkyQJA0cOFDffvutJk6c2GScHvlovmXLlikhIaH1vgC0imnTpik3N9ffY6AJrI+9WBt7sTZ2\nY33sVFJSooyMDK+PVG6Kz+I0KChILpdLhYWFGjNmjKT6V5kWFhY2+SkdVVVVcjq9nwZ75M/GmEZf\nCHDkofyEhASeh2WhyMhI1sVirI+9WBt7sTZ2Y33s1pKnYPr0Yf3p06crKytLLpdLgwcPVm5urqqq\nqpSVlSVJyszMVK9evbRgwQJJ0pVXXqnc3FwlJSXp5z//ubZt26Y5c+ZozJgxvEIVAACgA/BpnI4b\nN05ut1tz5sxRRUWFkpKStGrVKkVHR0uSdu/ercDA/4wwe/ZsOZ1OzZ49W3v27FF0dLTGjBmj+fPn\n+3JMAAAAWMLnL4jKzs5WdnZ2o9etXr3a689HwnT27Nm+HgsAAAAW8vnHl6JjS09P9/cIaAbrYy/W\nxl6sjd1Yn/bPYdr5Z+EVFxfL5XKpqKiIJ0ADAABY6Hh6jTOnAAAAsAZxCgAAAGsQpwAAALAGcQoA\nAABrEKcAAACwBnEKAAAAaxCnAAAAsAZxCgAAAGsQpwAAALAGcQoAAABrEKcAAACwBnEKAAAAaxCn\nAAAAsAZxCgAAAGsQpwAAALAGcQoAAABrEKcAAACwBnEKAAAAaxCnAAAAsAZxCgAAAGsQpwAAALBG\noL8HwMmloKD+IknV1dKuXVLv3lJwcP229PT6CwAAQGOIU7Sqo+OzuFhyuepjNTnZv3MBAID2gYf1\nAQAAYA3iFAAAANYgTgEAAGAN4hQAAADWIE4BAABgDeIUAAAA1iBOAQAAYA3iFAAAANYgTgEAAGAN\n4hQAAADWIE4BAABgDeIUAAAA1iBOAQAAYA3iFAAAANYgTgEAAGAN4hQAAADWIE4BAABgjUB/DwCg\n7RQU1F8kqbpa2rVL6t1bCg6u35aeXn8BAMBfiFOgAzk6PouLJZerPlaTk/07FwAAR/CwPgAAAKxB\nnAIAAMAaxCkAAACsQZwCAADAGsQpAAAArEGcAgAAwBrEKQAAAKxBnAIAAMAaxCkAAACsQZwCAADA\nGsQpAAAArEGcAgAAwBrEKQAAAKxBnAIAAMAaxCkAAACsQZwCAADAGsQpAAAArEGcAgAAwBrEKQAA\nAKxBnAIAAMAaxCkAAACsQZwCAADAGsQpAAAArEGcAgAAwBrEKQAAAKxBnAIAAMAaxCkAAACsQZwC\nAADAGsQpAAAArEGcAgAAwBrEKQAAAKxBnAIAAMAaxCkAAACsQZwCAADAGsQpAAAArEGcAgAAwBrE\nKQAAAKxBnAIAAMAaPo/TxYsXKz4+XiEhIRoyZIg2bNjQ7P6VlZX6zW9+o9NOO03BwcHq37+/3njj\nDV+PCQAAAAsE+vLGly9frhkzZujxxx/X4MGDlZubq7S0NH3yySeKiopqsH9NTY1SU1MVGxurl19+\nWaeddpp27dqlLl26+HJMAAAAWMKncZqbm6uJEycqMzNTkpSfn6+VK1dqyZIl+u1vf9tg/6eeekoH\nDhzQv/71LwUEBEiS4uLifDkiAAAALOKzh/VrampUVFSklJQUzzaHw6HU1FStX7++0WNeffVVDR06\nVNnZ2YqNjdU555yj++67T3V1db4aEwAAABbx2ZlTt9ut2tpaxcTEeG2PiYlRaWlpo8fs2LFDq1ev\nVkZGhl5//XVt375dt956qw4fPqzZs2f7alQAAABYwqcP6x+vuro6xcTE6PHHH5fD4dCgQYO0e/du\nPfTQQ8eM02nTpikyMtJrW3p6utLT0305MgAAAI5SUFCggoICr22VlZUtPt5ncRoVFaWAgABVVFR4\nba+oqFBsbGyjx/To0UOdOnWSw+HwbEtISNDevXt1+PBhBQY2PW5ubq6Sk5NbZ3gAAAD8JI2dHCwu\nLpbL5WrR8T57zmlQUJBcLpcKCws924wxKiws1LBhwxo95oILLtD27du9tpWWlqpHjx7NhikAAABO\nDj59n9Pp06friSee0LPPPqutW7dq0qRJqqqqUlZWliQpMzNTd999t2f/W2+9Vfv27dOUKVO0bds2\nrVy5Uvfdd58mT57syzEBAABgCZ+ejhw3bpzcbrfmzJmjiooKJSUladWqVYqOjpYk7d692+uMaK9e\nvbRq1SpNmzZNiYmJ6tmzp6ZNm9bo204BAADg5OPzx8qzs7OVnZ3d6HWrV69usO3nP/+51q1b5+ux\nAAAAYCGff3wpAAAA0FLEKQAAAKxBnAIAAMAaxCkAAACsQZwCAADAGsQpAAAArEGcAgAAwBrEKQAA\nAKxBnAIAAMAaxCkAAACsQZwCAADAGsQpAAAArEGcAgAAwBrEKQAAAKxBnAIAAMAaxCkAAACsQZwC\nAADAGsQpAAAArEGcAgAAwBrEKQAAAKxBnAIAAMAaxCkAAACsQZwCAADAGsQpAAAArEGcAgAAwBrE\nKQAAAKxBnAIAAMAaxCkAAACsQZwCAADAGsQpAAAArEGcAgAAwBrEKQAAAKxBnAIAAMAaxCkAAACs\nQZwCAADAGsQpAAAArEGcAgAAwBrEKQAAAKxBnAIAAMAaxCkAAACsQZwCAADAGsQpAAAArEGcAgAA\nwBrEKQAAAKxBnAIAAMAaxCkAAACsQZwCAADAGsQpAAAArEGcAgAAwBrEKQAAAKxBnAIAAMAaxCkA\nAACsQZwCAADAGsQpAAAArEGcAgAAwBrEKQAAAKxBnAIAAMAaxCkAAACsQZwCAADAGsQpAAAArEGc\nAgAAwBrEKQAAAKxBnAIAAMAaxCkAAACsQZwCAADAGsQpAAAArEGcAgAAwBrEKQAAAKxBnAIAAMAa\nxCkAAACsQZwCAADAGsQpAAAArEGcAgAAwBrEKQAAAKxBnAIAAMAaxCkAAACsQZwCAADAGsQpAAAA\nrEGcAgAAwBrEKQAAAKxBnAIAAMAaxCkAAACs4fM4Xbx4seLj4xUSEqIhQ4Zow4YNLTruf//3f+V0\nOvXLX/7SxxMCAADAFj6N0+XLl2vGjBnKycnRxo0blZiYqLS0NLnd7maPKysr05133qkRI0b4cjwA\nAABYxqdxmpubq4kTJyozM1P9+/dXfn6+QkNDtWTJkiaPqaurU0ZGhu655x7Fx8f7cjwAAABYxmdx\nWlNTo6KiIqWkpHi2ORwOpaamav369U0el5OTo5iYGE2YMMFXowEAAMBSgb66YbfbrdraWsXExHht\nj4mJUWlpaaPHvPvuu3r66af14Ycf+mosAAAAWMxncXq8vv32W2VmZuqJJ55Q165dj/v4adOmKTIy\n0mtbenq60tPTW2tEAAAAHENBQYEKCgq8tlVWVrb4eJ/FaVRUlAICAlRRUeG1vaKiQrGxsQ32//TT\nT7Vr1y5deeWVMsZIqn/+qSR16tRJpaWlzT4HNTc3V8nJya34FQAAAOB4NXZysLi4WC6Xq0XH++w5\np0FBQXK5XCosLPRsM8aosLBQw4YNa7B/QkKCNm/erE2bNunDDz/Uhx9+qDFjxmjUqFH68MMPdfrp\np/tqVAAAAFjCpw/rT58+XVlZWXK5XBo8eLByc3NVVVWlrKwsSVJmZqZ69eqlBQsWqFOnThowYIDX\n8V26dJHD4VBCQoIvxwQAAIAlfBqn48aNk9vt1pw5c1RRUaGkpCStWrVK0dHRkqTdu3crMNCap70C\nAADAz3xehtnZ2crOzm70utWrVzd77NNPP+2LkQAAAGApn398KQAAANBSxCkAAACsQZwCAADAGsQp\nAAAArEGcAgAAwBq8jxMAWKCgoP4iSdXV0q5dUu/eUnBw/bb09PoLAJzsiFMAsMDR8VlcLLlc9bHK\npzID6Gh4WB8AAADWIE4BAABgDeIUAAAA1iBOAQAAYA3iFAAAANYgTgEAAGAN4hQAAADWIE4BAABg\nDeIUAAAA1iBOAQAAYA0+vrQdKy8vl9vt9vcYTSopCZGUoJKSEkmH/D1Os6KiohQXF+fvMQAA6PCI\n03aqvLxcCf36qaq62t+jNGOQpGJlZNwgaaO/h2lWaHCwSkpLCVQAAPyMOG2n3G63qqqrtUxSgr+H\naUKJpAzJ6hml/z9ndbXcbjdxCqCBgoL6iyRVV0u7dkm9e0vBwfXb0tPrLwBaB3HaziVISvb3EMfQ\nHmYEgKYcHZ/FxZLLVR+ryfxiA3yCF0QBAADAGsQpAAAArEGcAgAAwBrEKQAAAKxBnAIAAMAaxCkA\nAACsQZwCAADAGsQpAAAArEGcAgAAwBrEKQAAAKxBnAIAAMAaxCkAAACsQZwCAADAGoH+HgA4GZWX\nl8vtdvt7jGaVlIRISlBJSYmkQ/4ep0lRUVGKi4vz9xgAgDZCnAKtrLy8XAn9+qmqutrfoxzDIEnF\nysi4QdJGfw/TpNDgYJWUlhKoANBBEKdAK3O73aqqrtYySQn+HqYZJZIyJKvnLJGUUV0tt9tNnAJA\nB0GcAj6SICnZ30O0QHuZEwDQMfCCKAAAAFiDOAUAAIA1iFMAAABYgzgFAACANYhTAAAAWIM4BQAA\ngDWIUwAAAFiDOAUAAIA1iFMAAABYgzgFAACANYhTAAAAWIM4BQAAgDWIUwAAAFiDOAUAAIA1iFMA\nAABYgzgFAACANYhTAAAAWIM4BQAAgDWIUwAAAFiDOAUAAIA1iFMAAABYgzgFAACANYhTAAAAWIM4\nBQAAgDWIUwAAAFiDOAUAAIA1iFMAAABYgzgFAACANYhTAAAAWIM4BQAAgDWIUwAAAFiDOAUAAIA1\niFMAAABYgzgFAACANYhTAAAAWIM4BQAAgDWIUwAAAFiDOAUAAIA1iFMAAABYgzgFAACANYhTAAAA\nWIM4BQAAgDWIUwAAAFiDOAUAAIA1fB6nixcvVnx8vEJCQjRkyBBt2LChyX2ffPJJjRgxQt26dVO3\nbt108cUXN7s/AAAATi4+jdPly5drxowZysnJ0caNG5WYmKi0tDS53e5G91+7dq2uv/56rVmzRv/6\n1790+umn65JLLtEXX3zhyzEBAABgCZ/GaW5uriZOnKjMzEz1799f+fn5Cg0N1ZIlSxrd/7nnntOk\nSZN07rnnqm/fvnryySdVV1enwsJCX44JAAAAS/gsTmtqalRUVKSUlBTPNofDodTUVK1fv75Ft/Hd\nd9+ppqZG3bp189WYAAAAsIjP4tTtdqu2tlYxMTFe22NiYrR3794W3cbvfvc79ezZU6mpqb4YEQAA\nAJYJ9PcATVm4cKFefPFFrV27Vp06dTrm/tOmTVNkZKTXtvT0dKWnp/tqRAAAAPxIQUGBCgoKvLZV\nVla2+HifxWlUVJQCAgJUUVHhtb2iokKxsbHNHvvQQw/pgQceUGFhoQYOHNii+8vNzVVycvJPnhcA\nAAAnrrGTg8XFxXK5XC063mcP6wcFBcnlcnm9mMkYo8LCQg0bNqzJ4x544AHde++9WrVqlQYNGuSr\n8QAAAGAhnz6sP336dGVlZcnlcmnw4MHKzc1VVVWVsrKyJEmZmZnq1auXFixYIEm6//77NXfuXBUU\nFCguLs5z1vWUU05RWFiYL0cFAACABXwap+PGjZPb7dacOXNUUVGhpKQkrVq1StHR0ZKk3bt3KzDw\nPyPk5+erpqZG11xzjdftzJ07V3PmzPHlqAAAALCAz18QlZ2drezs7EavW716tdefd+7c6etxAAAA\nYDGff3zfclK6AAARYElEQVQpAAAA0FLEKQAAAKxBnAIAAMAaxCkAAACsQZwCAADAGsQpAAAArEGc\nAgAAwBrEKQAAAKxBnAIAAMAaxCkAAACsQZwCAADAGsQpAAAArEGcAgAAwBrEKQAAAKxBnAIAAMAa\nxCkAAACsQZwCAADAGsQpAAAArEGcAgAAwBrEKQAAAKxBnAIAAMAaxCkAAACsEejvAQCgrZWXl8vt\ndvt7jCaVlIRISlBJSYmkQ/4ep0lRUVGKi4vz9xgATjLEKYAOpby8XAn9+qmqutrfozRjkKRiZWTc\nIGmjv4dpUmhwsEpKSwlUAK2KOAXQobjdblVVV2uZpAR/D9OEEkkZkv0zVlfL7XYTpwBaFXEKoENK\nkJTs7yGOoT3MCACtjRdEAQAAwBrEKQAAAKxBnAIAAMAaxCkAAACsQZwCAADAGrxaHwBgDT4gofXw\nIQlor4hTAIAV+ICE1sWHJKC9Ik4BAFbgAxJaDx+SgPaMOAUAWKU9fPhAe5gRaK94QRQAAACsQZwC\nAADAGsQpAAAArEGcAgAAwBrEKQAAAKxBnAIAAMAaxCkAAACsQZwCAADAGsQpAAAArEGcAgAAwBp8\nfCkAAGi3CgrqL5JUXS3t2iX17i0FB9dvS0+vv6D9IE4BAEC7dXR8FhdLLld9rCYn+3cu/HQ8rA8A\nAABrEKcAAACwBnEKAAAAaxCnAAAAsAZxCgAAAGsQpwAAALAGcQoAAABrEKcAAACwBnEKAAAAaxCn\nAAAAsAZxCgAAAGsQpwAAALAGcQoAAABrEKcAAACwBnEKAAAAaxCnAAAAsAZxCgAAAGsQpwAAALAG\ncQoAAABrEKcAAACwBnEKAAAAaxCnAAAAsAZxCgAAAGsQpwAAALAGcQoAAABrEKcAAACwRqC/BwAA\nAPYrLy+X2+329xjNKikJkZSgkpISSYf8PU6ToqKiFBcX5+8xrEWcAgCAZpWXlyuhXz9VVVf7e5Rj\nGCSpWBkZN0ja6O9hmhQaHKyS0lICtQnEKQAAaJbb7VZVdbWWSUrw9zDNKJGUIVk9Z4mkjOpqud1u\n4rQJxCkAAGiRBEnJ/h6iBdrLnGgcL4gCAACANYhTAAAAWIM4BQAAgDWIUwAAAFiDOAUAAIA1fB6n\nixcvVnx8vEJCQjRkyBBt2LCh2f1XrFihhIQEhYSEKDExUa+//rqvRwQAAIAlfBqny5cv14wZM5ST\nk6ONGzcqMTFRaWlpTX7CxLp163T99dfrlltu0aZNm3TVVVdp7Nix+vjjj305JgAAACzh0zjNzc3V\nxIkTlZmZqf79+ys/P1+hoaFasmRJo/svWrRIl112maZPn65+/frpnnvuUXJysv7nf/7Hl2MCAADA\nEj6L05qaGhUVFSklJcWzzeFwKDU1VevXr2/0mPXr1ys1NdVrW1paWpP7AwAA4OTis0+Icrvdqq2t\nVUxMjNf2mJgYlZaWNnrM3r17G91/7969vhqz3Svx9wDNKPnRP23lq/nay9dt85y+nK09fN3tYcb2\ndtsnqj2sjcTvNZvntHk2W5w0H186bdo0RUZGem1LT09Xenq6nybyraioKIUGByujutrfoxxThr8H\naIHQ4GBFRUW1ym21p7WR7F+f1lwbqX2tD2tjL9vXRuL3ms1a+2fHNgUFBSooKPDaVllZ2eLjfRan\nUVFRCggIUEVFhdf2iooKxcbGNnpMbGzsce1/tNzcXCUnd5xP0o2Li1NJaWmTLy6zQUlJiDIypGXL\nnldCwiF/j9OsqKgoxcXFtcpttYe1kdrP+rTm2kjtY31YG9amNfB7zd71ae2fHds0dnKwuLhYLper\nRcf7LE6DgoLkcrlUWFioMWPGSJKMMSosLNSUKVMaPWbo0KENrn/zzTc1dOhQX43ZrsXFxbWL/7gT\nEhLUgf6/QVL7WRuJ9bEZa2Mv1sZuHXF9TiY+fVh/+vTpysrKksvl0uDBg5Wbm6uqqiplZWVJkjIz\nM9WrVy8tWLBAkjR16lSNHDlSf/rTn3T55ZeroKBARUVFeuKJJ3w5JgAAACzh0zgdN26c3G635syZ\no4qKCiUlJWnVqlWKjo6WJO3evVuBgf8ZYejQoXrhhRc0a9YszZo1Sz/72c/0yiuvaMCAAb4cEwAA\nAJbw+QuisrOzlZ2d3eh1q1evbrDt6quv1tVXX+3rsQAAAGAhn398KQAAANBSxCkAAACsQZwCAADA\nGsQpAAAArEGcAgAAwBrEKQAAAKxBnAIAAMAaxCkAAACsQZwCAADAGsQpAAAArOHzjy8FAADwlYKC\n+oskVVdLfftKd90lBQfXb0tPr7+g/SBOAQBAu0V8nnyIU7Qq/g8WAACcCOIUrYr4BAAAJ4IXRAEA\nAMAaxCkAAACsQZwCAADAGsQpAAAArEGcAgAAwBrEKQAAAKxBnAIAAMAaxCkAAACsQZwCAADAGsQp\nAAAArEGcAgAAwBrEKQAAAKxBnAIAAMAaxCkAAACsEejvAQC0nYKC+oskVVdLfftKd90lBQfXb0tP\nr78AAOAvxCnQgRCfAADb8bA+AAAArEGcAgAAwBrEKQAAAKxBnAIAAMAaxCkAAACsQZwCAADAGsQp\nAAAArEGcAgAAwBrEKQAAAKxBnAIAAMAafHwpAFigoKD+IknV1VLfvtJdd0nBwfXb+OhZAB0FcQoA\nFiA+AaAeD+sDAADAGsQpAAAArEGcAgAAwBrEKQAAAKzBC6IAAGgG76QAtC3iFACAZhCfQNviYX0A\nAABYgzgFAACANYhTAAAAWIM4BQAAgDWIUwAAAFiDOAUAAIA1iFMAAABYgzgFAACANYhTAAAAWIM4\nBQAAgDWIUwAAAFiDOAUAAIA1iFMAAABYgzgFAACANYhTAAAAWIM4BQAAgDWIUwAAAFiDOAUAAIA1\niFMAAABYgzgFAACANYhTAAAAWIM4BQAAgDWIUwAAAFiDOAUAAIA1iFMAAABYgzgFAACANYhTAAAA\nWIM4BQAAgDWIUwAAAFiDOAUAAIA1iFMAAABYgzgFAACANYhTAAAAWIM4hU8VFBT4ewQ0g/WxF2tj\nL9bGbqxP++ezON2/f79uuOEGRUZGqmvXrrr55pv13XffNbv/lClT1L9/f4WGhqp3796aOnWqDh48\n6KsR0Qb4JWE31sderI29WBu7sT7tn8/i9Prrr1dJSYkKCwu1cuVKvfPOO5o4cWKT+3/++ef64osv\n9Kc//UlbtmzR0qVL9cYbb+jmm2/21YgAAACwTKAvbnTr1q1atWqVioqKNGjQIEnSI488ossvv1wP\nPfSQYmNjGxwzcOBArVixwvPn+Ph43XvvvbrxxhtVV1cnp5NnIAAAAJzsfFJ869evV9euXT1hKkmp\nqalyOBx67733Wnw7Bw4cUEREBGEKAADQQfjkzOnevXt16qmnem0LCAhQt27dtHfv3hbdhtvt1vz5\n85t9KoAkHTp0SJJUUlLy04aFT1VWVqq4uNjfY6AJrI+9WBt7sTZ2Y33sdKTTjnRbc44rTmfOnKn7\n77+/yesdDkerROI333yjyy+/XGeffbbmzp3b7L5lZWWSpIyMjBO+X/iGy+Xy9whoButjL9bGXqyN\n3Vgfe5WVlemCCy5odp/jitM77rhDEyZMaHafPn36KDY2Vl9++aXX9traWu3bt6/R55se7dtvv1Va\nWpq6dOmil19+WQEBAc3un5aWpmXLlumMM85QSEhIy74QAAAAtJlDhw6prKxMaWlpx9zXYYwxrT3A\n1q1bNXDgQH3wwQee553+/e9/1+jRo7V79+4mA/Wbb75RWlqaQkJC9Nprr6lz586tPRoAAAAs5pM4\nlaTRo0fryy+/1KOPPqoffvhBN910kwYPHqznnntOUv1bR6WkpOi5557Teeedp2+++UYXX3yxqqur\n9ec//1mhoaGe24qOjuZFUQAAAB2AT14QJUkvvPCCJk+erNTUVDmdTl1zzTX67//+b8/1NTU1+uST\nT1RVVSVJKi4u1oYNGyRJZ511liTJGCOHw6GdO3cqLi7OV6MCAADAEj47cwoAAAAcLx4rBwAAgDWI\nU5yQuro6XXDBBbr66qu9th88eFBxcXGaPXu2JMnpdHpdAgIC9OKLL/pj5A6jpWtzxL59+9SrVy8F\nBATo4MGDbTlqh9OStfn3v/+t66+/XnFxcQoNDdXAgQO1aNEiP03csbT0Z2fq1Kk677zzFBwcrOTk\nZH+M2uG0ZG327dunyy67TD179lRwcLDi4uJ022236ZtvvvHT1DhuBjhBn3zyiQkLCzMvvPCCZ9uN\nN95okpKSzOHDh40xxjgcDvPss8+aL7/80lRUVJiKigrz/fff+2vkDqO5tampqfHad+zYsebyyy83\nTqfTVFZWtvWoHU5TazNo0CBTU1NjlixZYm6//XbzzjvvmJ07d5rnn3/ehIaGmsWLF/tx6o6jJb/X\npk6davLy8sz48ePNoEGD/DVqh3Os32v79+83+fn5pqioyJSXl5vVq1eb/v37mxtuuMGPU+N4EKdo\nFYsWLTLdunUze/fuNX/5y19M586dzebNmz3XOxwO88orr/hxwo7rWGtjjDF5eXnmoosuMqtXryZO\n21BL1uZov/nNb0xKSkobTtixtXR95s2bR5y2seP92Vm0aJGJi4trwwlxInhBFFpNSkqKnE6nNm/e\nrKlTp2rmzJme65xOp3r27Knq6mr16dNHkyZNOuYHOqD1NLc2H3/8sS6++GK9//772r59u0aNGqX9\n+/crIiLCjxN3HM2tzY/deOON+v7773lKTBtqyfrk5OTolVde4SMz21hLf3Y+//xz3XDDDYqLi9PS\npUvbeEr8JP6uY5w8tm7dahwOh0lMTDS1tbVe182fP9+sW7fObNq0yTzwwAMmODjYPPLII36atONp\nam2+//57k5iY6Hl4bM2aNZw5bWPN/dwc7Z///Kfp1KmTeeutt9pwOrRkfThz6h/HWpv09HQTGhpq\nHA6Hueqqq3gqWTvCC6LQap566imFhYVp586d2r17t9d1s2bN0tChQ5WYmKg777xTv/3tb/Xggw/6\nadKOp6m1ueuuuzRgwAClp6dLqn9v4aP/Cd9r7ufmiI8++khjx47VvHnzlJKS0sYTdmwtWR/4x7HW\n5uGHH9bGjRv117/+VZ9++qmmTZvmhynxk/i7jnFyOHJWZ82aNSY1NfWYz4tbuXKlcTqd5ocffmij\nCTuu5tYmKSnJBAYGei4BAQHG4XCYoKAgM2/ePD9O3TG05Odmy5YtJiYmxsyePdsPE3ZsLf29xpnT\ntne8f+e8++67xuFwmL1797bRhDgRxClOWFVVlenbt6+5/fbbjTHGlJWVmYiICJOfn9/kMfPnzzfd\nu3dvqxE7rGOtzY4dO8yWLVs8l6effto4nU7z3nvvma+++sqfo5/0WvJz89FHH5mYmBhz1113+WvM\nDut4fq8Rp23rp/yds3btWuN0Os2uXbvaakycAOIUJ2zKlCmmb9++5tChQ55tjz32mImIiDBlZWXm\n1VdfNU8++aT56KOPzPbt201eXp4JCwszOTk5fpy6Y2hqbcLDwxv9Jb1mzRrjcDh4zmkbONbabN68\n2URHR5vMzEyzd+9ez4X/aWgbLfnZ2b59u9m4caOZOHGi6d+/v9m0aZPZtGlTg7dpQ+s61t85r732\nmnn66afNRx99ZMrKyszf/vY3M2DAADNixAg/To3jQZzihKxdu9YEBQWZdevWNbju0ksvNampqeaN\nN94wgwYNMhERESY8PNwMGjTIPPHEE36YtmNpydr8GC+IahstWZu5c+cap9PZ4BIfH++HiTuWlv7s\njBw5stE14uyc77RkbdasWWOGDRtmunbtakJDQ02/fv3M3Xffze+1doS3kgIAAIA1eLU+AAAArEGc\nAgAAwBrEKQAAAKxBnAIAAMAaxCkAAACsQZwCAADAGsQpAAAArEGcAgAAwBrEKQAAAKxBnAIAAMAa\nxCkAAACsQZwCAADAGv8PSPmSC1tJqMgAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x870ac18>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the top 5 feature importances of the forest\n",
    "plt.figure(num=None, figsize=(8, 6), dpi=80, facecolor='w', edgecolor='k')\n",
    "plt.title(\"Feature importances\")\n",
    "plt.bar(range(5), importances[indices][:5], \n",
    "       color=\"r\",  yerr=std[indices][:5], align=\"center\")\n",
    "plt.xticks(range(5), best_features)\n",
    "plt.xlim([-1, 5])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For Classification into Y2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Feature ranking:\n",
      "1. feature 3 - X4 (0.303510) \n",
      "2. feature 4 - X5 (0.300616) \n",
      "3. feature 1 - X2 (0.140325) \n",
      "4. feature 0 - X1 (0.137824) \n",
      "5. feature 2 - X3 (0.057110) \n",
      "6. feature 6 - X7 (0.046795) \n",
      "7. feature 7 - X8 (0.010409) \n",
      "8. feature 5 - X6 (0.003411) \n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABHgAAANDCAYAAADW6KijAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xa7fbef0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from sklearn.ensemble import ExtraTreesClassifier\n",
    "\n",
    "# Build a classification task using 3 informative features\n",
    "\n",
    "# Build a forest and compute the feature importances\n",
    "forest = ExtraTreesClassifier(n_estimators=250,\n",
    "                              random_state=0)\n",
    "\n",
    "forest.fit(X, y2)\n",
    "importances = forest.feature_importances_\n",
    "std = np.std([tree.feature_importances_ for tree in forest.estimators_],\n",
    "             axis=0)\n",
    "indices = np.argsort(importances)[::-1]\n",
    "\n",
    "# Print the feature ranking\n",
    "print(\"Feature ranking:\")\n",
    "\n",
    "for f in range(X.shape[1]):\n",
    "    print(\"%d. feature %d - %s (%f) \" % (f + 1, indices[f], features[indices[f]], importances[indices[f]]))\n",
    "\n",
    "# Plot the feature importances of the forest\n",
    "plt.figure(num=None, figsize=(14, 10), dpi=80, facecolor='w', edgecolor='k')\n",
    "plt.title(\"Feature importances\")\n",
    "plt.bar(range(X.shape[1]), importances[indices],\n",
    "       color=\"r\", yerr=std[indices], align=\"center\")\n",
    "plt.xticks(range(X.shape[1]), indices)\n",
    "plt.xlim([-1, X.shape[1]])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1. feature 3 - X4 (0.303510)\n",
      "2. feature 4 - X5 (0.300616)\n",
      "3. feature 1 - X2 (0.140325)\n",
      "4. feature 0 - X1 (0.137824)\n",
      "5. feature 2 - X3 (0.057110)\n"
     ]
    }
   ],
   "source": [
    "for f in range(5):\n",
    "    print(\"%d. feature %d - %s (%f)\" % (f + 1, indices[f], features[indices[f]] ,importances[indices[f]]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAqcAAAINCAYAAAADYtaMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTgEAAGCN/w9n9kcpZNONlAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xa2246d8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "best_features = []\n",
    "for i in indices[:5]:\n",
    "    best_features.append(features[i])\n",
    "# Plot the top 5 feature importances of the forest\n",
    "plt.figure(num=None, figsize=(8, 6), dpi=80, facecolor='w', edgecolor='k')\n",
    "plt.title(\"Feature importances\")\n",
    "plt.bar(range(5), importances[indices][:5], \n",
    "       color=\"r\",  yerr=std[indices][:5], align=\"center\")\n",
    "plt.xticks(range(5), best_features)\n",
    "plt.xlim([-1, 5])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So for both Y1 and Y2, X4 and X5 are the most important features."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "___\n",
    "## Maximum margin classification with Support Vector Machines"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "Support Vector Machines(SVM) is a powerful and widely used learning algorithm. In other classification approaches we aim to minimize the classification error. Whereas in SVMs, our approach to acheive optimization is to maximize the margin. Here the margin is defined as the distance between the separating hyperplane also called as the decision boundary and the training samples that are closest to this hyperplane, which are the so-called support vectors. This can be understood from the following figure:\n",
    "\n",
    "<img src=\"images/svm.png\">"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Dealing with the nonlinearly separable case using slack variables\n",
    "\n",
    "Scikit-Learn implementation of SVMs provide us parameters to fine tune our algorithm implementions. These paramaters are explained according to Scikit-Learn documentation as follows:\n",
    "\n",
    "1. **The gamma parameter** : Intuitively, the gamma parameter defines how far the influence of a single training example reaches, with low values meaning 'far' and high values meaning 'close'. The gamma parameters can be seen as the inverse of the radius of influence of samples selected by the model as support vectors.\n",
    "\n",
    "2. **The C parameter** : The C parameter trades off misclassification of training examples against simplicity of the decision surface. A low C makes the decision surface smooth, while a high C aims at classifying all training examples correctly by giving the model freedom to select more samples as support vectors.\n",
    "\n",
    "You can better understand the use of these parameters in plotting the SVM classification from this [link](http://scikit-learn.org/stable/auto_examples/svm/plot_rbf_parameters.html).\n",
    "\n",
    "Essentially using the variable C, we can control the penalty for misclassification error. Large values of C account for large error penalties whereas smaller values for C means we are less strict about misclassification. We can then we use the parameter C to control the 'width' of the margin and therefore tune the bias-variance trade-off as illustrated in the following figure:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1UAAAGMCAYAAAAyUYgIAAABfGlDQ1BJQ0MgUHJvZmlsZQAAKJFj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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 21,
     "metadata": {
      "image/png": {
       "width": 600
      }
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(filename='./images/03_08.png', width=600) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "[[image source]](https://github.com/rasbt/python-machine-learning-book/tree/master/images/image_gallery)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This concept is related to regularization, as increasing the value of C increases the bias and lowers the variance of the model."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now that we learned the basic concepts behind the linear SVM, let's train a SVM model to classify :"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SVM\n",
      "Acurracy for Y1:  0.974025974026\n",
      "Acurracy for Y2:  0.970779220779\n",
      "time elapsed:  0.15600895881652832\n"
     ]
    }
   ],
   "source": [
    "from time import time\n",
    "from sklearn.svm import SVC\n",
    "\n",
    "t7=time()\n",
    "print (\"SVM\")\n",
    "svc = SVC(kernel='linear', C=1.0, random_state=0)\n",
    "svc2 = SVC(kernel='linear', C=1.0, random_state=0)\n",
    "\n",
    "clf_svc=svc.fit(X_train, y_train)\n",
    "clf_svc2=svc2.fit(X_train, y_train2)\n",
    "\n",
    "print (\"Acurracy for Y1: \", clf_svc.score(X_test,y_test))\n",
    "print (\"Acurracy for Y2: \", clf_svc2.score(X_test2,y_test2))\n",
    "\n",
    "t8=time()\n",
    "print (\"time elapsed: \", t8-t7)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### cross validation for SVM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cross result========\n",
      "For Y1\n",
      "[ 0.91612903  1.          1.          1.          1.        ]\n",
      "0.983225806452\n",
      "For Y2\n",
      "[ 1.          1.          0.92207792  0.92156863  0.92156863]\n",
      "0.953043035396\n",
      "time elapsed:  1.7180981636047363\n"
     ]
    }
   ],
   "source": [
    "tt7=time()\n",
    "print (\"cross result========\")\n",
    "scores = cross_validation.cross_val_score(svc, X,y1, cv=5)\n",
    "scores2 = cross_validation.cross_val_score(svc2, X,y2, cv=5)\n",
    "print (\"For Y1\") \n",
    "print (scores )\n",
    "print (scores.mean())\n",
    "print (\"For Y2\") \n",
    "print (scores2)\n",
    "print (scores2.mean())\n",
    "tt8=time()\n",
    "print (\"time elapsed: \", tt8-tt7)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Using the default implementation of SVM in Scikit-Learn, we achieved 98 percent accuracy for classifying Y1 i.e Heating load as high or low and 95 percent accuracy for classiying Y2 i.e Cooling load."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SVM\n",
      "Acurracy for Y1:  0.974025974026\n",
      "Acurracy for Y2:  0.74025974026\n",
      "time elapsed:  0.054003000259399414\n"
     ]
    }
   ],
   "source": [
    "from time import time\n",
    "from sklearn.svm import LinearSVC\n",
    "\n",
    "t7=time()\n",
    "print (\"SVM\")\n",
    "\n",
    "svc = LinearSVC(C=1.0, random_state=0)\n",
    "svc2 = LinearSVC(C=1.0, random_state=0)\n",
    "\n",
    "clf_svc=svc.fit(X_train, y_train)\n",
    "clf_svc2=svc2.fit(X_train, y_train2)\n",
    "\n",
    "print (\"Acurracy for Y1: \", clf_svc.score(X_test,y_test))\n",
    "print (\"Acurracy for Y2: \", clf_svc2.score(X_test2,y_test2))\n",
    "\n",
    "t8=time()\n",
    "print (\"time elapsed: \", t8-t7)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### cross validation for SVM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cross result========\n",
      "For Y1\n",
      "[ 0.91612903  0.72727273  1.          1.          0.76470588]\n",
      "0.881621528377\n",
      "For Y2\n",
      "[ 0.7012987   0.93506494  0.92207792  0.92156863  0.92156863]\n",
      "0.880315762669\n",
      "time elapsed:  0.3260188102722168\n"
     ]
    }
   ],
   "source": [
    "tt7=time()\n",
    "print (\"cross result========\")\n",
    "scores = cross_validation.cross_val_score(svc, X,y1, cv=5)\n",
    "scores2 = cross_validation.cross_val_score(svc2, X,y2, cv=5)\n",
    "print (\"For Y1\") \n",
    "print (scores )\n",
    "print (scores.mean())\n",
    "print (\"For Y2\") \n",
    "print (scores2)\n",
    "print (scores2.mean())\n",
    "tt8=time()\n",
    "print (\"time elapsed: \", tt8-tt7)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As you can see we achieved 88 percent accuracy for classifying both Y1 and Y2 using the LinearSVC variant of SVM implementation in Scikit-Learn.\n",
    "\n",
    "<hr>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Logistic regression versus SVM\n",
    "In practical classification tasks, linear logistic regression and linear SVMs often yield very similar results. Logistic regression tries to maximize the conditional probabilities of the training data, which makes it more prone to misclassifying abnormal data points or outliers whereas this is not the case with SVMs. The SVMs try to find a the decision boundary or support vectors that are farthest from the closest points in average. Logistic regression has the advantage over SVMs due to it's simpler implementation. Furthermore, logistic regression models can be easily updated, which is useful when working with streaming data.\n",
    "<hr>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## Solving non-linear problems using a kernel SVM\n",
    "\n",
    "Another reason why SVMs are highly popular among machine learning practitioners is that they can be used to solve nonlinear classification problems by what are called as kernels. Before we discuss the main concept behind kernel SVM, let's first define and create a sample dataset to see how such a nonlinear classification problem may look.\n",
    "Using the following code, we will create a simple dataset that has the form of\n",
    "an XOR gate using the logical_xor function from NumPy, where 100 samples\n",
    "will be assigned the class label 1 and 100 samples will be assigned the class\n",
    "label -1, respectively:\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAk4AAAGGCAYAAACNCg6xAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x9911e80>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "np.random.seed(0)\n",
    "X_xor = np.random.randn(200, 2)\n",
    "y_xor = np.logical_xor(X_xor[:, 0] > 0,\n",
    "                       X_xor[:, 1] > 0)\n",
    "y_xor = np.where(y_xor, 1, -1)\n",
    "\n",
    "plt.scatter(X_xor[y_xor == 1, 0],\n",
    "            X_xor[y_xor == 1, 1],\n",
    "            c='b', marker='x',\n",
    "            label='1')\n",
    "plt.scatter(X_xor[y_xor == -1, 0],\n",
    "            X_xor[y_xor == -1, 1],\n",
    "            c='r',\n",
    "            marker='s',\n",
    "            label='-1')\n",
    "\n",
    "plt.xlim([-3, 3])\n",
    "plt.ylim([-3, 3])\n",
    "plt.legend(loc='best')\n",
    "plt.tight_layout()\n",
    "# plt.savefig('./figures/xor.png', dpi=300)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After executing the code, we will have an XOR dataset with random noise,\n",
    "as shown."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As you might have noticed, the LinearSVM model that we used earlier would fail to classify samples as positive or negative using a simple linear hyperplane as the decision boundary.\n",
    "To do this, we use the kernel methods to deal with such linearly inseparable data. The basic idea behind kernel methods is to create nonlinear combinations of the original features to project them onto a higher dimensional space via a mapping function where it becomes linearly separable. As shown in the next figure, we can transform a two-dimensional dataset onto a new three-dimensional feature space where the classes become separable\n",
    "\n",
    "This allows us to separate the two classes shown in the plot via a linear hyperplane that becomes a nonlinear decision boundary if we project it back onto the original feature space:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "<img src=\"images/svmd.png\">"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Using the kernel trick to find separating hyperplanes in higher dimensional space\n",
    "\n",
    "Roughly speaking, the term kernel can be interpreted as a similarity function between a pair of samples. The minus sign inverts the distance measure into a similarity score and, due to the exponential term, the resulting similarity score will fall into a range between 1 (for exactly similar samples) and 0 (for very dissimilar samples).\n",
    "\n",
    "Let's try to train a kernel SVM that that can draw a nonlinear decision boundary separating the XOR data well. Here, we simply use the SVC class from scikit-learn that we imported earlier and replace the parameter kernel='linear' with kernel='rbf':"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from matplotlib.colors import ListedColormap\n",
    "import matplotlib.pyplot as plt\n",
    "import warnings\n",
    "\n",
    "\n",
    "def versiontuple(v):\n",
    "    return tuple(map(int, (v.split(\".\"))))\n",
    "\n",
    "\n",
    "def plot_decision_regions(X, y, classifier, test_idx=None, resolution=0.02):\n",
    "\n",
    "    # setup marker generator and color map\n",
    "    markers = ('s', 'x', 'o', '^', 'v')\n",
    "    colors = ('red', 'blue', 'lightgreen', 'gray', 'cyan')\n",
    "    cmap = ListedColormap(colors[:len(np.unique(y))])\n",
    "\n",
    "    # plot the decision surface\n",
    "    x1_min, x1_max = X[:, 0].min() - 1, X[:, 0].max() + 1\n",
    "    x2_min, x2_max = X[:, 1].min() - 1, X[:, 1].max() + 1\n",
    "    xx1, xx2 = np.meshgrid(np.arange(x1_min, x1_max, resolution),\n",
    "                           np.arange(x2_min, x2_max, resolution))\n",
    "    Z = classifier.predict(np.array([xx1.ravel(), xx2.ravel()]).T)\n",
    "    Z = Z.reshape(xx1.shape)\n",
    "    plt.contourf(xx1, xx2, Z, alpha=0.4, cmap=cmap)\n",
    "    plt.xlim(xx1.min(), xx1.max())\n",
    "    plt.ylim(xx2.min(), xx2.max())\n",
    "\n",
    "    for idx, cl in enumerate(np.unique(y)):\n",
    "        plt.scatter(x=X[y == cl, 0], y=X[y == cl, 1],\n",
    "                    alpha=0.8, c=cmap(idx),\n",
    "                    marker=markers[idx], label=cl)\n",
    "\n",
    "    # highlight test samples\n",
    "    if test_idx:\n",
    "        # plot all samples\n",
    "        if not versiontuple(np.__version__) >= versiontuple('1.9.0'):\n",
    "            X_test, y_test = X[list(test_idx), :], y[list(test_idx)]\n",
    "            warnings.warn('Please update to NumPy 1.9.0 or newer')\n",
    "        else:\n",
    "            X_test, y_test = X[test_idx, :], y[test_idx]\n",
    "\n",
    "        plt.scatter(X_test[:, 0],\n",
    "                    X_test[:, 1],\n",
    "                    c='',\n",
    "                    alpha=1.0,\n",
    "                    linewidths=1,\n",
    "                    marker='o',\n",
    "                    s=55, label='test set')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAk4AAAGGCAYAAACNCg6xAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n2SQAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x87081d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "svm = SVC(kernel='rbf', random_state=0, gamma=0.10, C=10.0)\n",
    "svm.fit(X_xor, y_xor)\n",
    "plot_decision_regions(X_xor, y_xor,\n",
    "                      classifier=svm)\n",
    "\n",
    "plt.legend(loc='upper left')\n",
    "plt.tight_layout()\n",
    "# plt.savefig('./figures/support_vector_machine_rbf_xor.png', dpi=300)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we can see in the resulting plot, the kernel SVM separates the XOR data relatively well. \n",
    "\n",
    "The gammaa parameter, which we set to gamma=0.1, can be understood as a cut-off parameter for the Gaussian sphere. If we increase the value for gammaa, we increase the influence or reach of the training samples, which leads to a softer decision boundary. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can use the rbf kernel to classify our data as well:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SVM\n",
      "Acurracy for Y1:  0.977272727273\n",
      "Acurracy for Y2:  0.967532467532\n",
      "time elapsed:  0.09100508689880371\n"
     ]
    }
   ],
   "source": [
    "from time import time\n",
    "from sklearn.svm import SVC\n",
    "\n",
    "t7=time()\n",
    "print (\"SVM\")\n",
    "svc = SVC(kernel='rbf', gamma=0.7, C=1.0, random_state=0)\n",
    "svc2 = SVC(kernel='rbf', gamma=0.7,  C=1.0, random_state=0)\n",
    "\n",
    "clf_svc=svc.fit(X_train, y_train)\n",
    "clf_svc2=svc2.fit(X_train, y_train2)\n",
    "\n",
    "print (\"Acurracy for Y1: \", clf_svc.score(X_test,y_test))\n",
    "print (\"Acurracy for Y2: \", clf_svc2.score(X_test2,y_test2))\n",
    "\n",
    "t8=time()\n",
    "print (\"time elapsed: \", t8-t7)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This gives an accuracy of 97 percent for Y1 (Heating load) and 96 percent for Y2 (Cooling load) which proves the effiency of Support Vector Machines. Actually, before the boom of Neural Networks and Deep Learning in general, SVMs enjoyed a lot of popularity amongs Machine Learning practitioners for a wide range of tasks. However, although support vector machines are powerful linear models that can be extended to nonlinear problems via the kernel trick, they have many parameters that have to be tuned in order to make good predictions. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "____"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}