Gesture-Phase-Detection.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Gesture-Phase-Detection\n",
    "---\n",
    "\n",
    "Dataset : https://archive.ics.uci.edu/ml/datasets/Gesture+Phase+Segmentation\n",
    "\n",
    "The dataset is composed by features extracted from 7 videos with people gesticulating, aiming at studying Gesture Phase Segmentation. \n",
    "Each video is represented by two files: a raw file, which contains the position of hands, wrists, head and spine of the user in each frame; and a processed file, which contains velocity and acceleration of hands and wrists. See the data set description for more information on the dataset.\n",
    "\n",
    "Raw files: 18 numeric attributes (double), a timestamp and a class attribute (nominal). \n",
    "Processed files: 32 numeric attributes (double) and a class attribute (nominal). \n",
    "A feature vector with up to 50 numeric attributes can be generated with the two files mentioned above."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.cross_validation import train_test_split\n",
    "from sklearn import cross_validation, metrics\n",
    "from sklearn import preprocessing\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# read .csv from provided dataset\n",
    "csv_filename1=\"a1_raw.csv\"\n",
    "csv_filename2=\"a1_va3.csv\"\n",
    "\n",
    "# df=pd.read_csv(csv_filename,index_col=0)\n",
    "df1=pd.read_csv(csv_filename1 , skiprows=[1,2,3,4])\n",
    "df2=pd.read_csv(csv_filename2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false,
    "scrolled": 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>lhx</th>\n",
       "      <th>lhy</th>\n",
       "      <th>lhz</th>\n",
       "      <th>rhx</th>\n",
       "      <th>rhy</th>\n",
       "      <th>rhz</th>\n",
       "      <th>hx</th>\n",
       "      <th>hy</th>\n",
       "      <th>hz</th>\n",
       "      <th>sx</th>\n",
       "      <th>sy</th>\n",
       "      <th>sz</th>\n",
       "      <th>lwx</th>\n",
       "      <th>lwy</th>\n",
       "      <th>lwz</th>\n",
       "      <th>rwx</th>\n",
       "      <th>rwy</th>\n",
       "      <th>rwz</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>phase</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5.003160</td>\n",
       "      <td>4.278530</td>\n",
       "      <td>1.542866</td>\n",
       "      <td>4.985812</td>\n",
       "      <td>4.182155</td>\n",
       "      <td>1.520330</td>\n",
       "      <td>5.037557</td>\n",
       "      <td>1.619226</td>\n",
       "      <td>1.778925</td>\n",
       "      <td>5.052367</td>\n",
       "      <td>4.225485</td>\n",
       "      <td>1.775536</td>\n",
       "      <td>4.983912</td>\n",
       "      <td>4.296833</td>\n",
       "      <td>1.569889</td>\n",
       "      <td>5.193762</td>\n",
       "      <td>4.335417</td>\n",
       "      <td>1.560144</td>\n",
       "      <td>5702167</td>\n",
       "      <td>Rest</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5.064488</td>\n",
       "      <td>4.290401</td>\n",
       "      <td>1.542146</td>\n",
       "      <td>4.955739</td>\n",
       "      <td>4.163175</td>\n",
       "      <td>1.511876</td>\n",
       "      <td>5.037724</td>\n",
       "      <td>1.618397</td>\n",
       "      <td>1.779722</td>\n",
       "      <td>5.045395</td>\n",
       "      <td>4.223284</td>\n",
       "      <td>1.777401</td>\n",
       "      <td>5.000410</td>\n",
       "      <td>4.301358</td>\n",
       "      <td>1.566544</td>\n",
       "      <td>5.164159</td>\n",
       "      <td>4.313107</td>\n",
       "      <td>1.552097</td>\n",
       "      <td>5702307</td>\n",
       "      <td>Rest</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5.067825</td>\n",
       "      <td>4.290883</td>\n",
       "      <td>1.542058</td>\n",
       "      <td>4.928284</td>\n",
       "      <td>4.157637</td>\n",
       "      <td>1.511306</td>\n",
       "      <td>5.038332</td>\n",
       "      <td>1.618043</td>\n",
       "      <td>1.780080</td>\n",
       "      <td>5.045374</td>\n",
       "      <td>4.223690</td>\n",
       "      <td>1.777571</td>\n",
       "      <td>5.001656</td>\n",
       "      <td>4.299812</td>\n",
       "      <td>1.566537</td>\n",
       "      <td>5.136817</td>\n",
       "      <td>4.307087</td>\n",
       "      <td>1.551576</td>\n",
       "      <td>5702338</td>\n",
       "      <td>Rest</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5.070332</td>\n",
       "      <td>4.290677</td>\n",
       "      <td>1.541985</td>\n",
       "      <td>4.916637</td>\n",
       "      <td>4.151067</td>\n",
       "      <td>1.510510</td>\n",
       "      <td>5.038742</td>\n",
       "      <td>1.618044</td>\n",
       "      <td>1.780114</td>\n",
       "      <td>5.045767</td>\n",
       "      <td>4.224827</td>\n",
       "      <td>1.777669</td>\n",
       "      <td>5.002672</td>\n",
       "      <td>4.298810</td>\n",
       "      <td>1.566489</td>\n",
       "      <td>5.125220</td>\n",
       "      <td>4.300282</td>\n",
       "      <td>1.550805</td>\n",
       "      <td>5702370</td>\n",
       "      <td>Rest</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.071611</td>\n",
       "      <td>4.290927</td>\n",
       "      <td>1.542046</td>\n",
       "      <td>4.906132</td>\n",
       "      <td>4.143034</td>\n",
       "      <td>1.509449</td>\n",
       "      <td>5.042224</td>\n",
       "      <td>1.618561</td>\n",
       "      <td>1.780209</td>\n",
       "      <td>5.047422</td>\n",
       "      <td>4.223671</td>\n",
       "      <td>1.778054</td>\n",
       "      <td>5.012298</td>\n",
       "      <td>4.298582</td>\n",
       "      <td>1.565061</td>\n",
       "      <td>5.114789</td>\n",
       "      <td>4.292008</td>\n",
       "      <td>1.549765</td>\n",
       "      <td>5702432</td>\n",
       "      <td>Rest</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        lhx       lhy       lhz       rhx       rhy       rhz        hx  \\\n",
       "0  5.003160  4.278530  1.542866  4.985812  4.182155  1.520330  5.037557   \n",
       "1  5.064488  4.290401  1.542146  4.955739  4.163175  1.511876  5.037724   \n",
       "2  5.067825  4.290883  1.542058  4.928284  4.157637  1.511306  5.038332   \n",
       "3  5.070332  4.290677  1.541985  4.916637  4.151067  1.510510  5.038742   \n",
       "4  5.071611  4.290927  1.542046  4.906132  4.143034  1.509449  5.042224   \n",
       "\n",
       "         hy        hz        sx        sy        sz       lwx       lwy  \\\n",
       "0  1.619226  1.778925  5.052367  4.225485  1.775536  4.983912  4.296833   \n",
       "1  1.618397  1.779722  5.045395  4.223284  1.777401  5.000410  4.301358   \n",
       "2  1.618043  1.780080  5.045374  4.223690  1.777571  5.001656  4.299812   \n",
       "3  1.618044  1.780114  5.045767  4.224827  1.777669  5.002672  4.298810   \n",
       "4  1.618561  1.780209  5.047422  4.223671  1.778054  5.012298  4.298582   \n",
       "\n",
       "        lwz       rwx       rwy       rwz  timestamp phase  \n",
       "0  1.569889  5.193762  4.335417  1.560144    5702167  Rest  \n",
       "1  1.566544  5.164159  4.313107  1.552097    5702307  Rest  \n",
       "2  1.566537  5.136817  4.307087  1.551576    5702338  Rest  \n",
       "3  1.566489  5.125220  4.300282  1.550805    5702370  Rest  \n",
       "4  1.565061  5.114789  4.292008  1.549765    5702432  Rest  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1743, 20)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['Rest', 'Preparation', 'Stroke', 'Hold', 'Retraction'], dtype=object)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1['phase'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import LabelEncoder\n",
    "le = LabelEncoder()\n",
    "df1['phase'] = le.fit_transform(df1['phase'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2, 1, 4, 0, 3], dtype=int64)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1['phase'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>...</th>\n",
       "      <th>24</th>\n",
       "      <th>25</th>\n",
       "      <th>26</th>\n",
       "      <th>27</th>\n",
       "      <th>28</th>\n",
       "      <th>29</th>\n",
       "      <th>30</th>\n",
       "      <th>31</th>\n",
       "      <th>32</th>\n",
       "      <th>Phase</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.005009</td>\n",
       "      <td>-0.000964</td>\n",
       "      <td>0.000573</td>\n",
       "      <td>0.008623</td>\n",
       "      <td>0.005667</td>\n",
       "      <td>0.001302</td>\n",
       "      <td>-0.000631</td>\n",
       "      <td>0.000130</td>\n",
       "      <td>-0.000048</td>\n",
       "      <td>0.007762</td>\n",
       "      <td>...</td>\n",
       "      <td>1.880800e-04</td>\n",
       "      <td>0.005133</td>\n",
       "      <td>0.010400</td>\n",
       "      <td>0.000646</td>\n",
       "      <td>0.007871</td>\n",
       "      <td>0.004631</td>\n",
       "      <td>0.000963</td>\n",
       "      <td>0.000092</td>\n",
       "      <td>0.000438</td>\n",
       "      <td>D</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.004905</td>\n",
       "      <td>0.001209</td>\n",
       "      <td>-0.000649</td>\n",
       "      <td>0.004737</td>\n",
       "      <td>0.003166</td>\n",
       "      <td>0.000819</td>\n",
       "      <td>-0.000572</td>\n",
       "      <td>-0.000015</td>\n",
       "      <td>0.000023</td>\n",
       "      <td>0.002706</td>\n",
       "      <td>...</td>\n",
       "      <td>-7.500000e-07</td>\n",
       "      <td>0.005093</td>\n",
       "      <td>0.005756</td>\n",
       "      <td>0.000573</td>\n",
       "      <td>0.003459</td>\n",
       "      <td>0.000730</td>\n",
       "      <td>0.000332</td>\n",
       "      <td>0.000012</td>\n",
       "      <td>0.000433</td>\n",
       "      <td>D</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.002393</td>\n",
       "      <td>-0.000216</td>\n",
       "      <td>0.000136</td>\n",
       "      <td>0.003028</td>\n",
       "      <td>0.001212</td>\n",
       "      <td>0.000336</td>\n",
       "      <td>-0.000449</td>\n",
       "      <td>0.000017</td>\n",
       "      <td>0.000047</td>\n",
       "      <td>0.002868</td>\n",
       "      <td>...</td>\n",
       "      <td>-3.920000e-05</td>\n",
       "      <td>0.002406</td>\n",
       "      <td>0.003279</td>\n",
       "      <td>0.000452</td>\n",
       "      <td>0.003261</td>\n",
       "      <td>0.002412</td>\n",
       "      <td>0.000852</td>\n",
       "      <td>0.000042</td>\n",
       "      <td>0.000202</td>\n",
       "      <td>D</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.001394</td>\n",
       "      <td>-0.000242</td>\n",
       "      <td>0.000056</td>\n",
       "      <td>0.001182</td>\n",
       "      <td>0.000575</td>\n",
       "      <td>0.000225</td>\n",
       "      <td>-0.000479</td>\n",
       "      <td>-0.000050</td>\n",
       "      <td>0.000104</td>\n",
       "      <td>0.001171</td>\n",
       "      <td>...</td>\n",
       "      <td>-3.184000e-05</td>\n",
       "      <td>0.001416</td>\n",
       "      <td>0.001334</td>\n",
       "      <td>0.000493</td>\n",
       "      <td>0.001358</td>\n",
       "      <td>0.000313</td>\n",
       "      <td>0.000611</td>\n",
       "      <td>0.000029</td>\n",
       "      <td>0.000596</td>\n",
       "      <td>D</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-0.000156</td>\n",
       "      <td>-0.000004</td>\n",
       "      <td>0.000023</td>\n",
       "      <td>0.001585</td>\n",
       "      <td>0.000630</td>\n",
       "      <td>0.000094</td>\n",
       "      <td>-0.000303</td>\n",
       "      <td>0.000097</td>\n",
       "      <td>0.000065</td>\n",
       "      <td>0.001579</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.015000e-05</td>\n",
       "      <td>0.000158</td>\n",
       "      <td>0.001709</td>\n",
       "      <td>0.000325</td>\n",
       "      <td>0.001713</td>\n",
       "      <td>0.000203</td>\n",
       "      <td>0.000069</td>\n",
       "      <td>0.000038</td>\n",
       "      <td>0.000069</td>\n",
       "      <td>D</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 33 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          1         2         3         4         5         6         7  \\\n",
       "0 -0.005009 -0.000964  0.000573  0.008623  0.005667  0.001302 -0.000631   \n",
       "1  0.004905  0.001209 -0.000649  0.004737  0.003166  0.000819 -0.000572   \n",
       "2 -0.002393 -0.000216  0.000136  0.003028  0.001212  0.000336 -0.000449   \n",
       "3 -0.001394 -0.000242  0.000056  0.001182  0.000575  0.000225 -0.000479   \n",
       "4 -0.000156 -0.000004  0.000023  0.001585  0.000630  0.000094 -0.000303   \n",
       "\n",
       "          8         9        10  ...              24        25        26  \\\n",
       "0  0.000130 -0.000048  0.007762  ...    1.880800e-04  0.005133  0.010400   \n",
       "1 -0.000015  0.000023  0.002706  ...   -7.500000e-07  0.005093  0.005756   \n",
       "2  0.000017  0.000047  0.002868  ...   -3.920000e-05  0.002406  0.003279   \n",
       "3 -0.000050  0.000104  0.001171  ...   -3.184000e-05  0.001416  0.001334   \n",
       "4  0.000097  0.000065  0.001579  ...   -2.015000e-05  0.000158  0.001709   \n",
       "\n",
       "         27        28        29        30        31        32  Phase  \n",
       "0  0.000646  0.007871  0.004631  0.000963  0.000092  0.000438      D  \n",
       "1  0.000573  0.003459  0.000730  0.000332  0.000012  0.000433      D  \n",
       "2  0.000452  0.003261  0.002412  0.000852  0.000042  0.000202      D  \n",
       "3  0.000493  0.001358  0.000313  0.000611  0.000029  0.000596      D  \n",
       "4  0.000325  0.001713  0.000203  0.000069  0.000038  0.000069      D  \n",
       "\n",
       "[5 rows x 33 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1743, 33)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['D', 'P', 'S', 'H', 'R'], dtype=object)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2['Phase'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['lhx', 'lhy', 'lhz', 'rhx', 'rhy', 'rhz', 'hx', 'hy', 'hz', 'sx', 'sy',\n",
       "       'sz', 'lwx', 'lwy', 'lwz', 'rwx', 'rwy', 'rwz', 'timestamp', 'phase'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['1', '2', '3', '4', '5', '6', '7', '8', '9', '10', '11', '12', '13',\n",
       "       '14', '15', '16', '17', '18', '19', '20', '21', '22', '23', '24', '25',\n",
       "       '26', '27', '28', '29', '30', '31', '32', 'Phase'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df2.rename(columns={'Phase': 'phase'}, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2, 1, 4, 0, 3], dtype=int64)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.phase.unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['D', 'P', 'S', 'H', 'R'], dtype=object)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.phase.unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "a = df2.phase == 'D'\n",
    "b = df2.phase == 'P'\n",
    "c = df2.phase == 'S'\n",
    "d = df2.phase == 'H'\n",
    "e = df2.phase == 'R'\n",
    "\n",
    "df2.loc[a,'phase'] = 'Rest'\n",
    "df2.loc[b,'phase'] = 'Preparation'\n",
    "df2.loc[c,'phase'] = 'Stroke'\n",
    "df2.loc[d,'phase'] = 'Hold'\n",
    "df2.loc[e,'phase'] = 'Retraction'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "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>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>...</th>\n",
       "      <th>24</th>\n",
       "      <th>25</th>\n",
       "      <th>26</th>\n",
       "      <th>27</th>\n",
       "      <th>28</th>\n",
       "      <th>29</th>\n",
       "      <th>30</th>\n",
       "      <th>31</th>\n",
       "      <th>32</th>\n",
       "      <th>phase</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.005009</td>\n",
       "      <td>-0.000964</td>\n",
       "      <td>0.000573</td>\n",
       "      <td>0.008623</td>\n",
       "      <td>0.005667</td>\n",
       "      <td>0.001302</td>\n",
       "      <td>-0.000631</td>\n",
       "      <td>0.000130</td>\n",
       "      <td>-0.000048</td>\n",
       "      <td>0.007762</td>\n",
       "      <td>...</td>\n",
       "      <td>1.880800e-04</td>\n",
       "      <td>0.005133</td>\n",
       "      <td>0.010400</td>\n",
       "      <td>0.000646</td>\n",
       "      <td>0.007871</td>\n",
       "      <td>0.004631</td>\n",
       "      <td>0.000963</td>\n",
       "      <td>0.000092</td>\n",
       "      <td>0.000438</td>\n",
       "      <td>Rest</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.004905</td>\n",
       "      <td>0.001209</td>\n",
       "      <td>-0.000649</td>\n",
       "      <td>0.004737</td>\n",
       "      <td>0.003166</td>\n",
       "      <td>0.000819</td>\n",
       "      <td>-0.000572</td>\n",
       "      <td>-0.000015</td>\n",
       "      <td>0.000023</td>\n",
       "      <td>0.002706</td>\n",
       "      <td>...</td>\n",
       "      <td>-7.500000e-07</td>\n",
       "      <td>0.005093</td>\n",
       "      <td>0.005756</td>\n",
       "      <td>0.000573</td>\n",
       "      <td>0.003459</td>\n",
       "      <td>0.000730</td>\n",
       "      <td>0.000332</td>\n",
       "      <td>0.000012</td>\n",
       "      <td>0.000433</td>\n",
       "      <td>Rest</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.002393</td>\n",
       "      <td>-0.000216</td>\n",
       "      <td>0.000136</td>\n",
       "      <td>0.003028</td>\n",
       "      <td>0.001212</td>\n",
       "      <td>0.000336</td>\n",
       "      <td>-0.000449</td>\n",
       "      <td>0.000017</td>\n",
       "      <td>0.000047</td>\n",
       "      <td>0.002868</td>\n",
       "      <td>...</td>\n",
       "      <td>-3.920000e-05</td>\n",
       "      <td>0.002406</td>\n",
       "      <td>0.003279</td>\n",
       "      <td>0.000452</td>\n",
       "      <td>0.003261</td>\n",
       "      <td>0.002412</td>\n",
       "      <td>0.000852</td>\n",
       "      <td>0.000042</td>\n",
       "      <td>0.000202</td>\n",
       "      <td>Rest</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 33 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          1         2         3         4         5         6         7  \\\n",
       "0 -0.005009 -0.000964  0.000573  0.008623  0.005667  0.001302 -0.000631   \n",
       "1  0.004905  0.001209 -0.000649  0.004737  0.003166  0.000819 -0.000572   \n",
       "2 -0.002393 -0.000216  0.000136  0.003028  0.001212  0.000336 -0.000449   \n",
       "\n",
       "          8         9        10  ...              24        25        26  \\\n",
       "0  0.000130 -0.000048  0.007762  ...    1.880800e-04  0.005133  0.010400   \n",
       "1 -0.000015  0.000023  0.002706  ...   -7.500000e-07  0.005093  0.005756   \n",
       "2  0.000017  0.000047  0.002868  ...   -3.920000e-05  0.002406  0.003279   \n",
       "\n",
       "         27        28        29        30        31        32  phase  \n",
       "0  0.000646  0.007871  0.004631  0.000963  0.000092  0.000438   Rest  \n",
       "1  0.000573  0.003459  0.000730  0.000332  0.000012  0.000433   Rest  \n",
       "2  0.000452  0.003261  0.002412  0.000852  0.000042  0.000202   Rest  \n",
       "\n",
       "[3 rows x 33 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['Rest', 'Preparation', 'Stroke', 'Hold', 'Retraction'], dtype=object)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.phase.unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import LabelEncoder\n",
    "le = LabelEncoder()\n",
    "df2['phase'] = le.fit_transform(df2['phase'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2, 1, 4, 0, 3], dtype=int64)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.phase.unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "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>lhx</th>\n",
       "      <th>lhy</th>\n",
       "      <th>lhz</th>\n",
       "      <th>rhx</th>\n",
       "      <th>rhy</th>\n",
       "      <th>rhz</th>\n",
       "      <th>hx</th>\n",
       "      <th>hy</th>\n",
       "      <th>hz</th>\n",
       "      <th>sx</th>\n",
       "      <th>sy</th>\n",
       "      <th>sz</th>\n",
       "      <th>lwx</th>\n",
       "      <th>lwy</th>\n",
       "      <th>lwz</th>\n",
       "      <th>rwx</th>\n",
       "      <th>rwy</th>\n",
       "      <th>rwz</th>\n",
       "      <th>timestamp</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>phase</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>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       lhx  lhy  lhz  rhx  rhy  rhz   hx   hy   hz   sx   sy   sz  lwx  lwy  \\\n",
       "phase                                                                         \n",
       "0       39   39   39   39   39   39   39   39   39   39   39   39   39   39   \n",
       "1      163  163  163  163  163  163  163  163  163  163  163  163  163  163   \n",
       "2      694  694  694  694  694  694  694  694  694  694  694  694  694  694   \n",
       "3      191  191  191  191  191  191  191  191  191  191  191  191  191  191   \n",
       "4      656  656  656  656  656  656  656  656  656  656  656  656  656  656   \n",
       "\n",
       "       lwz  rwx  rwy  rwz  timestamp  \n",
       "phase                                 \n",
       "0       39   39   39   39         39  \n",
       "1      163  163  163  163        163  \n",
       "2      694  694  694  694        694  \n",
       "3      191  191  191  191        191  \n",
       "4      656  656  656  656        656  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.groupby('phase').count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "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>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>...</th>\n",
       "      <th>23</th>\n",
       "      <th>24</th>\n",
       "      <th>25</th>\n",
       "      <th>26</th>\n",
       "      <th>27</th>\n",
       "      <th>28</th>\n",
       "      <th>29</th>\n",
       "      <th>30</th>\n",
       "      <th>31</th>\n",
       "      <th>32</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>phase</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>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>...</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "      <td>39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>...</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "      <td>163</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>...</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>...</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "      <td>191</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>...</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "      <td>656</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 32 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         1    2    3    4    5    6    7    8    9   10 ...    23   24   25  \\\n",
       "phase                                                   ...                   \n",
       "0       39   39   39   39   39   39   39   39   39   39 ...    39   39   39   \n",
       "1      163  163  163  163  163  163  163  163  163  163 ...   163  163  163   \n",
       "2      694  694  694  694  694  694  694  694  694  694 ...   694  694  694   \n",
       "3      191  191  191  191  191  191  191  191  191  191 ...   191  191  191   \n",
       "4      656  656  656  656  656  656  656  656  656  656 ...   656  656  656   \n",
       "\n",
       "        26   27   28   29   30   31   32  \n",
       "phase                                     \n",
       "0       39   39   39   39   39   39   39  \n",
       "1      163  163  163  163  163  163  163  \n",
       "2      694  694  694  694  694  694  694  \n",
       "3      191  191  191  191  191  191  191  \n",
       "4      656  656  656  656  656  656  656  \n",
       "\n",
       "[5 rows x 32 columns]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.groupby('phase').count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df1.sort('phase',inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df2.sort('phase',inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "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>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>...</th>\n",
       "      <th>24</th>\n",
       "      <th>25</th>\n",
       "      <th>26</th>\n",
       "      <th>27</th>\n",
       "      <th>28</th>\n",
       "      <th>29</th>\n",
       "      <th>30</th>\n",
       "      <th>31</th>\n",
       "      <th>32</th>\n",
       "      <th>phase</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1117</th>\n",
       "      <td>0.013652</td>\n",
       "      <td>-0.005854</td>\n",
       "      <td>-0.000167</td>\n",
       "      <td>0.007993</td>\n",
       "      <td>0.011719</td>\n",
       "      <td>-0.001821</td>\n",
       "      <td>0.011699</td>\n",
       "      <td>-0.006116</td>\n",
       "      <td>-0.000023</td>\n",
       "      <td>0.007367</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.000594</td>\n",
       "      <td>0.014855</td>\n",
       "      <td>0.014302</td>\n",
       "      <td>0.013201</td>\n",
       "      <td>0.012085</td>\n",
       "      <td>0.004621</td>\n",
       "      <td>0.012173</td>\n",
       "      <td>0.004385</td>\n",
       "      <td>0.009172</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1118</th>\n",
       "      <td>0.025679</td>\n",
       "      <td>0.029140</td>\n",
       "      <td>0.004168</td>\n",
       "      <td>0.034294</td>\n",
       "      <td>0.008579</td>\n",
       "      <td>0.001769</td>\n",
       "      <td>0.025674</td>\n",
       "      <td>0.029158</td>\n",
       "      <td>0.004169</td>\n",
       "      <td>0.017613</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000532</td>\n",
       "      <td>0.039063</td>\n",
       "      <td>0.035395</td>\n",
       "      <td>0.039073</td>\n",
       "      <td>0.018241</td>\n",
       "      <td>0.012018</td>\n",
       "      <td>0.008623</td>\n",
       "      <td>0.012314</td>\n",
       "      <td>0.005680</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1119</th>\n",
       "      <td>0.036604</td>\n",
       "      <td>0.044513</td>\n",
       "      <td>0.005200</td>\n",
       "      <td>0.032754</td>\n",
       "      <td>-0.018750</td>\n",
       "      <td>-0.000209</td>\n",
       "      <td>0.024677</td>\n",
       "      <td>0.029102</td>\n",
       "      <td>0.003881</td>\n",
       "      <td>0.026038</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.000077</td>\n",
       "      <td>0.057864</td>\n",
       "      <td>0.037742</td>\n",
       "      <td>0.038353</td>\n",
       "      <td>0.028690</td>\n",
       "      <td>0.006093</td>\n",
       "      <td>0.008853</td>\n",
       "      <td>0.000335</td>\n",
       "      <td>0.003600</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>371</th>\n",
       "      <td>-0.000753</td>\n",
       "      <td>0.010647</td>\n",
       "      <td>-0.004304</td>\n",
       "      <td>0.005471</td>\n",
       "      <td>0.010892</td>\n",
       "      <td>-0.005108</td>\n",
       "      <td>0.003455</td>\n",
       "      <td>0.007538</td>\n",
       "      <td>-0.003794</td>\n",
       "      <td>0.001323</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000180</td>\n",
       "      <td>0.011508</td>\n",
       "      <td>0.013216</td>\n",
       "      <td>0.009119</td>\n",
       "      <td>0.008922</td>\n",
       "      <td>0.005335</td>\n",
       "      <td>0.005696</td>\n",
       "      <td>0.004074</td>\n",
       "      <td>0.003962</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>871</th>\n",
       "      <td>-0.001645</td>\n",
       "      <td>0.001315</td>\n",
       "      <td>0.000080</td>\n",
       "      <td>0.010487</td>\n",
       "      <td>0.026691</td>\n",
       "      <td>-0.000251</td>\n",
       "      <td>-0.005443</td>\n",
       "      <td>-0.005227</td>\n",
       "      <td>0.000392</td>\n",
       "      <td>-0.007014</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.000067</td>\n",
       "      <td>0.002108</td>\n",
       "      <td>0.028678</td>\n",
       "      <td>0.007557</td>\n",
       "      <td>0.017659</td>\n",
       "      <td>0.000165</td>\n",
       "      <td>0.001954</td>\n",
       "      <td>0.004185</td>\n",
       "      <td>0.004051</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 33 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             1         2         3         4         5         6         7  \\\n",
       "1117  0.013652 -0.005854 -0.000167  0.007993  0.011719 -0.001821  0.011699   \n",
       "1118  0.025679  0.029140  0.004168  0.034294  0.008579  0.001769  0.025674   \n",
       "1119  0.036604  0.044513  0.005200  0.032754 -0.018750 -0.000209  0.024677   \n",
       "371  -0.000753  0.010647 -0.004304  0.005471  0.010892 -0.005108  0.003455   \n",
       "871  -0.001645  0.001315  0.000080  0.010487  0.026691 -0.000251 -0.005443   \n",
       "\n",
       "             8         9        10  ...          24        25        26  \\\n",
       "1117 -0.006116 -0.000023  0.007367  ...   -0.000594  0.014855  0.014302   \n",
       "1118  0.029158  0.004169  0.017613  ...    0.000532  0.039063  0.035395   \n",
       "1119  0.029102  0.003881  0.026038  ...   -0.000077  0.057864  0.037742   \n",
       "371   0.007538 -0.003794  0.001323  ...    0.000180  0.011508  0.013216   \n",
       "871  -0.005227  0.000392 -0.007014  ...   -0.000067  0.002108  0.028678   \n",
       "\n",
       "            27        28        29        30        31        32  phase  \n",
       "1117  0.013201  0.012085  0.004621  0.012173  0.004385  0.009172      4  \n",
       "1118  0.039073  0.018241  0.012018  0.008623  0.012314  0.005680      4  \n",
       "1119  0.038353  0.028690  0.006093  0.008853  0.000335  0.003600      4  \n",
       "371   0.009119  0.008922  0.005335  0.005696  0.004074  0.003962      4  \n",
       "871   0.007557  0.017659  0.000165  0.001954  0.004185  0.004051      4  \n",
       "\n",
       "[5 rows x 33 columns]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "left = pd.DataFrame({\n",
    "   ....:                      'key2': ['0', '2', '1', '3','0','1'],\n",
    "   ....:                      'A': ['A0', 'A1', 'A2', 'A3','A4','A5'],\n",
    "   ....:                      'B': ['B0', 'B1', 'B2', 'B3','B4','B5']})\n",
    "   ....: \n",
    "\n",
    "right = pd.DataFrame({\n",
    "   ....:                       'key2': ['0', '1', '2', '0', '1', '3'],\n",
    "   ....:                       'C': ['C0', 'C1', 'C2', 'C3', 'C4', 'C5'],\n",
    "   ....:                       'D': ['D0', 'D1', 'D2', 'D3', 'D4', 'D5']})\n",
    "   ....: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "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>A</th>\n",
       "      <th>B</th>\n",
       "      <th>key2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>A0</td>\n",
       "      <td>B0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>A1</td>\n",
       "      <td>B1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>A2</td>\n",
       "      <td>B2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>A3</td>\n",
       "      <td>B3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>A4</td>\n",
       "      <td>B4</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>A5</td>\n",
       "      <td>B5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    A   B key2\n",
       "0  A0  B0    0\n",
       "1  A1  B1    2\n",
       "2  A2  B2    1\n",
       "3  A3  B3    3\n",
       "4  A4  B4    0\n",
       "5  A5  B5    1"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "left"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "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>C</th>\n",
       "      <th>D</th>\n",
       "      <th>key2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>C0</td>\n",
       "      <td>D0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>C1</td>\n",
       "      <td>D1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>C2</td>\n",
       "      <td>D2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>C3</td>\n",
       "      <td>D3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>C4</td>\n",
       "      <td>D4</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>C5</td>\n",
       "      <td>D5</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    C   D key2\n",
       "0  C0  D0    0\n",
       "1  C1  D1    1\n",
       "2  C2  D2    2\n",
       "3  C3  D3    0\n",
       "4  C4  D4    1\n",
       "5  C5  D5    3"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "right"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "left.sort('key2',inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "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>A</th>\n",
       "      <th>B</th>\n",
       "      <th>key2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>A0</td>\n",
       "      <td>B0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>A4</td>\n",
       "      <td>B4</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>A2</td>\n",
       "      <td>B2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>A5</td>\n",
       "      <td>B5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>A1</td>\n",
       "      <td>B1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>A3</td>\n",
       "      <td>B3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    A   B key2\n",
       "0  A0  B0    0\n",
       "4  A4  B4    0\n",
       "2  A2  B2    1\n",
       "5  A5  B5    1\n",
       "1  A1  B1    2\n",
       "3  A3  B3    3"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "left"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "right.sort('key2',inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "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>C</th>\n",
       "      <th>D</th>\n",
       "      <th>key2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>C0</td>\n",
       "      <td>D0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>C3</td>\n",
       "      <td>D3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>C1</td>\n",
       "      <td>D1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>C4</td>\n",
       "      <td>D4</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>C2</td>\n",
       "      <td>D2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>C5</td>\n",
       "      <td>D5</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    C   D key2\n",
       "0  C0  D0    0\n",
       "3  C3  D3    0\n",
       "1  C1  D1    1\n",
       "4  C4  D4    1\n",
       "2  C2  D2    2\n",
       "5  C5  D5    3"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "right"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "result = pd.merge(left, right, on=['key2'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "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>A</th>\n",
       "      <th>B</th>\n",
       "      <th>key2</th>\n",
       "      <th>C</th>\n",
       "      <th>D</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>A0</td>\n",
       "      <td>B0</td>\n",
       "      <td>0</td>\n",
       "      <td>C0</td>\n",
       "      <td>D0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>A0</td>\n",
       "      <td>B0</td>\n",
       "      <td>0</td>\n",
       "      <td>C3</td>\n",
       "      <td>D3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>A4</td>\n",
       "      <td>B4</td>\n",
       "      <td>0</td>\n",
       "      <td>C0</td>\n",
       "      <td>D0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>A4</td>\n",
       "      <td>B4</td>\n",
       "      <td>0</td>\n",
       "      <td>C3</td>\n",
       "      <td>D3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>A2</td>\n",
       "      <td>B2</td>\n",
       "      <td>1</td>\n",
       "      <td>C1</td>\n",
       "      <td>D1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>A2</td>\n",
       "      <td>B2</td>\n",
       "      <td>1</td>\n",
       "      <td>C4</td>\n",
       "      <td>D4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>A5</td>\n",
       "      <td>B5</td>\n",
       "      <td>1</td>\n",
       "      <td>C1</td>\n",
       "      <td>D1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>A5</td>\n",
       "      <td>B5</td>\n",
       "      <td>1</td>\n",
       "      <td>C4</td>\n",
       "      <td>D4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>A1</td>\n",
       "      <td>B1</td>\n",
       "      <td>2</td>\n",
       "      <td>C2</td>\n",
       "      <td>D2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>A3</td>\n",
       "      <td>B3</td>\n",
       "      <td>3</td>\n",
       "      <td>C5</td>\n",
       "      <td>D5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    A   B key2   C   D\n",
       "0  A0  B0    0  C0  D0\n",
       "1  A0  B0    0  C3  D3\n",
       "2  A4  B4    0  C0  D0\n",
       "3  A4  B4    0  C3  D3\n",
       "4  A2  B2    1  C1  D1\n",
       "5  A2  B2    1  C4  D4\n",
       "6  A5  B5    1  C1  D1\n",
       "7  A5  B5    1  C4  D4\n",
       "8  A1  B1    2  C2  D2\n",
       "9  A3  B3    3  C5  D5"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "result2 = pd.merge(left, right, on=['key2'], how='right')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "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>A</th>\n",
       "      <th>B</th>\n",
       "      <th>key2</th>\n",
       "      <th>C</th>\n",
       "      <th>D</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>A0</td>\n",
       "      <td>B0</td>\n",
       "      <td>0</td>\n",
       "      <td>C0</td>\n",
       "      <td>D0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>A4</td>\n",
       "      <td>B4</td>\n",
       "      <td>0</td>\n",
       "      <td>C0</td>\n",
       "      <td>D0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>A0</td>\n",
       "      <td>B0</td>\n",
       "      <td>0</td>\n",
       "      <td>C3</td>\n",
       "      <td>D3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>A4</td>\n",
       "      <td>B4</td>\n",
       "      <td>0</td>\n",
       "      <td>C3</td>\n",
       "      <td>D3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>A2</td>\n",
       "      <td>B2</td>\n",
       "      <td>1</td>\n",
       "      <td>C1</td>\n",
       "      <td>D1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>A5</td>\n",
       "      <td>B5</td>\n",
       "      <td>1</td>\n",
       "      <td>C1</td>\n",
       "      <td>D1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>A2</td>\n",
       "      <td>B2</td>\n",
       "      <td>1</td>\n",
       "      <td>C4</td>\n",
       "      <td>D4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>A5</td>\n",
       "      <td>B5</td>\n",
       "      <td>1</td>\n",
       "      <td>C4</td>\n",
       "      <td>D4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>A1</td>\n",
       "      <td>B1</td>\n",
       "      <td>2</td>\n",
       "      <td>C2</td>\n",
       "      <td>D2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>A3</td>\n",
       "      <td>B3</td>\n",
       "      <td>3</td>\n",
       "      <td>C5</td>\n",
       "      <td>D5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    A   B key2   C   D\n",
       "0  A0  B0    0  C0  D0\n",
       "1  A4  B4    0  C0  D0\n",
       "2  A0  B0    0  C3  D3\n",
       "3  A4  B4    0  C3  D3\n",
       "4  A2  B2    1  C1  D1\n",
       "5  A5  B5    1  C1  D1\n",
       "6  A2  B2    1  C4  D4\n",
       "7  A5  B5    1  C4  D4\n",
       "8  A1  B1    2  C2  D2\n",
       "9  A3  B3    3  C5  D5"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df = pd.merge(df1, df2, on='phase')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "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>lhx</th>\n",
       "      <th>lhy</th>\n",
       "      <th>lhz</th>\n",
       "      <th>rhx</th>\n",
       "      <th>rhy</th>\n",
       "      <th>rhz</th>\n",
       "      <th>hx</th>\n",
       "      <th>hy</th>\n",
       "      <th>hz</th>\n",
       "      <th>sx</th>\n",
       "      <th>...</th>\n",
       "      <th>23</th>\n",
       "      <th>24</th>\n",
       "      <th>25</th>\n",
       "      <th>26</th>\n",
       "      <th>27</th>\n",
       "      <th>28</th>\n",
       "      <th>29</th>\n",
       "      <th>30</th>\n",
       "      <th>31</th>\n",
       "      <th>32</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.843293</td>\n",
       "      <td>3.074965</td>\n",
       "      <td>1.644773</td>\n",
       "      <td>8.228968</td>\n",
       "      <td>3.020815</td>\n",
       "      <td>1.570013</td>\n",
       "      <td>5.03905</td>\n",
       "      <td>1.756163</td>\n",
       "      <td>1.734586</td>\n",
       "      <td>5.049521</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.000049</td>\n",
       "      <td>0.000047</td>\n",
       "      <td>0.001516</td>\n",
       "      <td>0.001352</td>\n",
       "      <td>0.001703</td>\n",
       "      <td>0.001068</td>\n",
       "      <td>0.000416</td>\n",
       "      <td>0.000397</td>\n",
       "      <td>0.000240</td>\n",
       "      <td>0.000422</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.843293</td>\n",
       "      <td>3.074965</td>\n",
       "      <td>1.644773</td>\n",
       "      <td>8.228968</td>\n",
       "      <td>3.020815</td>\n",
       "      <td>1.570013</td>\n",
       "      <td>5.03905</td>\n",
       "      <td>1.756163</td>\n",
       "      <td>1.734586</td>\n",
       "      <td>5.049521</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.000273</td>\n",
       "      <td>0.000059</td>\n",
       "      <td>0.007427</td>\n",
       "      <td>0.017248</td>\n",
       "      <td>0.003709</td>\n",
       "      <td>0.012792</td>\n",
       "      <td>0.000039</td>\n",
       "      <td>0.000174</td>\n",
       "      <td>0.000440</td>\n",
       "      <td>0.000281</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.843293</td>\n",
       "      <td>3.074965</td>\n",
       "      <td>1.644773</td>\n",
       "      <td>8.228968</td>\n",
       "      <td>3.020815</td>\n",
       "      <td>1.570013</td>\n",
       "      <td>5.03905</td>\n",
       "      <td>1.756163</td>\n",
       "      <td>1.734586</td>\n",
       "      <td>5.049521</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.002584</td>\n",
       "      <td>-0.000602</td>\n",
       "      <td>0.007392</td>\n",
       "      <td>0.017023</td>\n",
       "      <td>0.002659</td>\n",
       "      <td>0.012178</td>\n",
       "      <td>0.002543</td>\n",
       "      <td>0.005297</td>\n",
       "      <td>0.001246</td>\n",
       "      <td>0.004187</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.843293</td>\n",
       "      <td>3.074965</td>\n",
       "      <td>1.644773</td>\n",
       "      <td>8.228968</td>\n",
       "      <td>3.020815</td>\n",
       "      <td>1.570013</td>\n",
       "      <td>5.03905</td>\n",
       "      <td>1.756163</td>\n",
       "      <td>1.734586</td>\n",
       "      <td>5.049521</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000008</td>\n",
       "      <td>0.000070</td>\n",
       "      <td>0.000688</td>\n",
       "      <td>0.001083</td>\n",
       "      <td>0.001302</td>\n",
       "      <td>0.000407</td>\n",
       "      <td>0.000281</td>\n",
       "      <td>0.000334</td>\n",
       "      <td>0.000434</td>\n",
       "      <td>0.000264</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.843293</td>\n",
       "      <td>3.074965</td>\n",
       "      <td>1.644773</td>\n",
       "      <td>8.228968</td>\n",
       "      <td>3.020815</td>\n",
       "      <td>1.570013</td>\n",
       "      <td>5.03905</td>\n",
       "      <td>1.756163</td>\n",
       "      <td>1.734586</td>\n",
       "      <td>5.049521</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.000092</td>\n",
       "      <td>-0.000110</td>\n",
       "      <td>0.001527</td>\n",
       "      <td>0.001989</td>\n",
       "      <td>0.001715</td>\n",
       "      <td>0.001108</td>\n",
       "      <td>0.000086</td>\n",
       "      <td>0.000280</td>\n",
       "      <td>0.000341</td>\n",
       "      <td>0.000221</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 52 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        lhx       lhy       lhz       rhx       rhy       rhz       hx  \\\n",
       "0  1.843293  3.074965  1.644773  8.228968  3.020815  1.570013  5.03905   \n",
       "1  1.843293  3.074965  1.644773  8.228968  3.020815  1.570013  5.03905   \n",
       "2  1.843293  3.074965  1.644773  8.228968  3.020815  1.570013  5.03905   \n",
       "3  1.843293  3.074965  1.644773  8.228968  3.020815  1.570013  5.03905   \n",
       "4  1.843293  3.074965  1.644773  8.228968  3.020815  1.570013  5.03905   \n",
       "\n",
       "         hy        hz        sx    ...           23        24        25  \\\n",
       "0  1.756163  1.734586  5.049521    ...    -0.000049  0.000047  0.001516   \n",
       "1  1.756163  1.734586  5.049521    ...    -0.000273  0.000059  0.007427   \n",
       "2  1.756163  1.734586  5.049521    ...    -0.002584 -0.000602  0.007392   \n",
       "3  1.756163  1.734586  5.049521    ...     0.000008  0.000070  0.000688   \n",
       "4  1.756163  1.734586  5.049521    ...    -0.000092 -0.000110  0.001527   \n",
       "\n",
       "         26        27        28        29        30        31        32  \n",
       "0  0.001352  0.001703  0.001068  0.000416  0.000397  0.000240  0.000422  \n",
       "1  0.017248  0.003709  0.012792  0.000039  0.000174  0.000440  0.000281  \n",
       "2  0.017023  0.002659  0.012178  0.002543  0.005297  0.001246  0.004187  \n",
       "3  0.001083  0.001302  0.000407  0.000281  0.000334  0.000434  0.000264  \n",
       "4  0.001989  0.001715  0.001108  0.000086  0.000280  0.000341  0.000221  \n",
       "\n",
       "[5 rows x 52 columns]"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['lhx', 'lhy', 'lhz', 'rhx', 'rhy', 'rhz', 'hx', 'hy', 'hz', 'sx', 'sy',\n",
       "       'sz', 'lwx', 'lwy', 'lwz', 'rwx', 'rwy', 'rwz', 'timestamp', 'phase',\n",
       "       '1', '2', '3', '4', '5', '6', '7', '8', '9', '10', '11', '12', '13',\n",
       "       '14', '15', '16', '17', '18', '19', '20', '21', '22', '23', '24', '25',\n",
       "       '26', '27', '28', '29', '30', '31', '32'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "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>lhx</th>\n",
       "      <th>lhy</th>\n",
       "      <th>lhz</th>\n",
       "      <th>rhx</th>\n",
       "      <th>rhy</th>\n",
       "      <th>rhz</th>\n",
       "      <th>hx</th>\n",
       "      <th>hy</th>\n",
       "      <th>hz</th>\n",
       "      <th>sx</th>\n",
       "      <th>...</th>\n",
       "      <th>23</th>\n",
       "      <th>24</th>\n",
       "      <th>25</th>\n",
       "      <th>26</th>\n",
       "      <th>27</th>\n",
       "      <th>28</th>\n",
       "      <th>29</th>\n",
       "      <th>30</th>\n",
       "      <th>31</th>\n",
       "      <th>32</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.843293</td>\n",
       "      <td>3.074965</td>\n",
       "      <td>1.644773</td>\n",
       "      <td>8.228968</td>\n",
       "      <td>3.020815</td>\n",
       "      <td>1.570013</td>\n",
       "      <td>5.03905</td>\n",
       "      <td>1.756163</td>\n",
       "      <td>1.734586</td>\n",
       "      <td>5.049521</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.000049</td>\n",
       "      <td>0.000047</td>\n",
       "      <td>0.001516</td>\n",
       "      <td>0.001352</td>\n",
       "      <td>0.001703</td>\n",
       "      <td>0.001068</td>\n",
       "      <td>0.000416</td>\n",
       "      <td>0.000397</td>\n",
       "      <td>0.00024</td>\n",
       "      <td>0.000422</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1 rows × 52 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        lhx       lhy       lhz       rhx       rhy       rhz       hx  \\\n",
       "0  1.843293  3.074965  1.644773  8.228968  3.020815  1.570013  5.03905   \n",
       "\n",
       "         hy        hz        sx    ...           23        24        25  \\\n",
       "0  1.756163  1.734586  5.049521    ...    -0.000049  0.000047  0.001516   \n",
       "\n",
       "         26        27        28        29        30       31        32  \n",
       "0  0.001352  0.001703  0.001068  0.000416  0.000397  0.00024  0.000422  \n",
       "\n",
       "[1 rows x 52 columns]"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[:1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((1743, 20), (1743, 33), (976543, 52))"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.shape,df2.shape,df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "df.drop('timestamp', axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "cols = list(df.columns)\n",
    "features = cols\n",
    "features.remove('phase')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "50"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((1743, 20), (1743, 33), (976543, 51))"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.shape,df2.shape,df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df1.drop('phase',axis=1,inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_1 = pd.concat([df1,df2],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_1.drop('timestamp' , axis=1, inplace=True )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1743, 51)"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_1.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "cols = list(df_1.columns)\n",
    "features = cols\n",
    "features.remove('phase')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "X = df_1[features]\n",
    "y = df_1['phase']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# split dataset to 60% training and 40% testing\n",
    "X_train, X_test, y_train, y_test = cross_validation.train_test_split(X, y, test_size=0.4, random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1045, 50) (1045,)\n"
     ]
    }
   ],
   "source": [
    "print (X_train.shape, y_train.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "# Unsupervised Learning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "50"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[ -3.24581191e-01  -2.92251490e-01   7.53364165e-03 ...,   3.24728085e-04\n",
      "    3.41055456e-04   3.57808090e-04]\n",
      " [ -3.61481275e-01  -3.43404679e-01  -9.08432680e-04 ...,  -5.07423576e-04\n",
      "    9.78392399e-05  -2.70505189e-04]\n",
      " [  2.37097518e-01  -5.92270728e-01  -1.24632609e-02 ...,   3.58624880e-04\n",
      "    3.90482340e-04   1.93933193e-04]\n",
      " ..., \n",
      " [  4.04553884e-05  -7.12943663e-05   1.15402630e-03 ...,  -4.86619567e-03\n",
      "    1.48671225e-02  -7.35656948e-03]\n",
      " [  1.29660243e-04   2.43834901e-05   9.59580183e-04 ...,   2.71273032e-03\n",
      "   -1.29078542e-02  -4.33512842e-03]\n",
      " [ -6.46273141e-05  -7.94059286e-05  -8.03733060e-04 ...,  -3.07743917e-05\n",
      "    5.64481537e-03   4.41337238e-04]]\n",
      "[  5.43602019e-01   1.94609502e-01   1.79945381e-01   6.06757941e-02\n",
      "   6.51627887e-03   3.58008210e-03   2.67915903e-03   2.33392917e-03\n",
      "   1.79531854e-03   1.48063948e-03   8.24068490e-04   6.16949441e-04\n",
      "   5.22681332e-04   3.02808424e-04   9.48140982e-05   8.22443484e-05\n",
      "   6.86532264e-05   5.74845805e-05   5.44861553e-05   3.78224060e-05\n",
      "   3.29970596e-05   2.22630456e-05   1.86548452e-05   1.34322604e-05\n",
      "   4.24253117e-06   3.74622695e-06   3.12518674e-06   2.80866166e-06\n",
      "   2.68473778e-06   2.38735187e-06   2.18231402e-06   1.95711839e-06\n",
      "   1.74097538e-06   1.57779695e-06   1.51902148e-06   1.15606438e-06\n",
      "   1.09387478e-06   4.78227553e-07   3.75018054e-07   3.06011531e-07\n",
      "   2.81524044e-07   2.41916791e-07   2.23069620e-07   1.76570338e-07\n",
      "   8.01344334e-08   5.95234827e-08   5.05723238e-08   2.85968368e-08\n",
      "   8.08258475e-09   6.73808980e-09]\n"
     ]
    }
   ],
   "source": [
    "# Apply PCA with the same number of dimensions as variables in the dataset\n",
    "from sklearn.decomposition import PCA\n",
    "pca = PCA(n_components=50)\n",
    "pca.fit(X)\n",
    "\n",
    "# Print the components and the amount of variance in the data contained in each dimension\n",
    "print(pca.components_)\n",
    "print(pca.explained_variance_ratio_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAh4AAAGHCAYAAAD/QltcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAAPYQAAD2EBqD+naQAAIABJREFUeJzs3XmYHFXZ/vHvPSEsSSYBQQKEJWyJQRFJRIiovEJkEQMq\nmxEFNxAFogFeULYogvoqEAEBd5AfGkRQCWtkUVFWTVg1IEvYQ9jDkABZ5vn9capJp9Mz093TXT3L\n/bmuvma66lTV06erq58+deqUIgIzMzOzPLQ0OwAzMzPrP5x4mJmZWW6ceJiZmVlunHiYmZlZbpx4\nmJmZWW6ceJiZmVlunHiYmZlZbpx4mJmZWW6ceJiZmVlunHjkRNKFkubWuOxjkn5V75gq3HbNcTdK\nM+ujt5D0LUntzY6jGpLeK+kWSa9JWibp3c2OqZzesP9J2kLSnyW9ktXlXs2OyaygXyUekg6W1N7B\nY5mk9zVw85E9atHejWW7qztxN0oz66PHkLSGpKmSPlRmdpDqqVeQtApwGbAW8HXgs8DjTYxnfFa3\nQ8vM7g3730XAO4HjSXX5r3KFJG1SchxcKulxSX+QtE2Z8qtJmiLp9iypeV3Sg5LOkbRlB9v4Qbbu\n6fV8gVY7SXtImtqs7a/SrA03UQAnAY+VmfdwvqFUbDS96EskB66PZBAwlbRP31wy7zvA93KPqHab\nAxsDX4yIC5odDPB+4GTgAuDVknk9ev+TtDqwA/CdiDivwsV+C1wDDADGAF8Fdpe0Q0Tcm613bWAm\nsC1wFfAb4DVSfXwKOARYvcy6PwXMBSZKGhwRC2t9bVY3HyW9x99uxsb7Y+IBcF1EzG52EJWKiCXN\njqEnkLR6RLzRV+tD0gCgpYrXp45mREQ7sLgugeVjePZ3QVOjWK6zuu3p+9+62d9q6nJ2RPy28ETS\nrcAM4CvZA+DXwDbAPhHxp+KFJZ0EnFa6UkkfBkYAOwN/Bj4J/L8q4rLG6HD/zkVE9JsHcDCwDBjb\nRblvZeU+XDL9Z8CbwNbZ851Iv3z2B74LzCP9ArgC2LBk2QuAR0umHQPcArwALCI1h+5TJp7HgF+V\nvI520q+yM4Hnsu3+AVi7zPJ7kH4Rv0b69XYVsFWZch8H7gdeB+7Nnq8Ud5nlrgQe6WDebcCdRc8/\nD9wIzAfeAP4NHNbBa54B7Ar8M4tpcgf1sRZwehZzG+mAew3w7pJ1Ft6v/YATgCez9d4AbF4mhu2z\n9byU1d09hRiKyowmnSJ4MVvXP4GJFeyLm2SxHAV8jdTatgR4NzAQOCXbH17Jtn0z8D9lll+W/S08\nTi7ah9tLtjmA1Nr3cFb3c0lfFqtWEO/W2b7wSPY65wG/BN5WUm4I8KNs3W9k7/Ofgfd0su4LyryW\nm7J5fy38X7LMhcDcDurzkKLXeCfw3jLLjwYuJX12FgEPAKdm86aWiWcZsHG5/S+btinw+2w/WEja\n7z/anf2vg7raFriWtI+3ZctuXzS/XOwdfn6L661k+qBs+nVFn4V24PxK4ixazy+A+7L/ry6sr4rl\nPwPckdXpS8DfgAklZb5KOm69ATwN/BgYVlLmr6Tjw9bZ/wuBh8iOt9l7c3vRvrBLyfLfyl5/Yb9Z\nQDpu/whYrZbPGcuPcTtmr/F10ufrs2XqYVi2rSeydT4EHAuo2s8A5T9vy4rmf4p07Hk1e533UnLc\n6+6jbivqDQ+WJx4fBtYuebytqNwqwCzgUWBwNm237A36ZlG5woHkHuAu0hfIadnOO6d4h6R84vEE\ncA7pF8XXSAerZcAeJeXmUj7xmAVcn33wfkD64ppesuxns3VenZU7Jtu5XyQ7kGbldgWWAndnsZwC\nvAzcVxp3mXr9TLaNcSXTN87inFI07Q7SF9bkLJ5rszJfKfOa/0v6cJ+WfZA+1EF9jMvKngZ8iXRQ\nf4J0oFqvzPs1K/swTiYdIF4DbivZ/kdIH9pHSU3uhwLTgJlFZd5ZVEfHZO/jX7K62LuLOiscJO4n\nHUT+N4tnQ9L++BTww2y7RwP/yeJ5d7b8oGxeOynx+XT2eFc2fypFB5Ns2oVZ+UuAw1h+ALq8gs/O\nUaQD9gnAF0kJ78Iy9fYb0gH0B6Qk8xjgT8CkTta9PenUUHtWx58mO/Bn9Vku8Vjh81RUn7OAB7Pt\nHk1KLB4HBhSVfTfpgPpctt0vkU5L3Z3N3zp7HcuAI4vqdo0O9r91gWdJSeK3SZ+f2aTP095F5Sre\n/zqop61IycZTwDezfebhrL63y8q8K1tvO3BxFvdeFeyHpYnHuwvryJ6fltXHjlUcb1clfQa/WXSc\nWAysW+HyhSTq79n+dwSpteS7RWW+lZW5jnQ8OYt0HLy95D3/S1ZvjwHfz8rel8WzP/AMcGL2fj+Z\nxT24TCz3kPbnr5BagNqBC2v5nGX70Zxs29/J1vnPbL8ZU1RujWy7z5GOy4dk61wGnFntZ4D0eZuZ\nLT8p20c+XXTca8/mH5Y9zgIuqfR9r+i9refKevqD5V/Y5R6LSsq+k3Sg/ykp23wq25lbisoUDiRP\nAIOKpu+bTT+iaFq5xKNcpnwvcH2ZHbRc4nFdSbkzsg9Sa/Z8cPYBOr+k3NtJX5g/KZp2V/YahxRN\n24UufjFl5VrJvmxKpv9v9iHasKPXnE27FniozGteRsmvmw7qY2CZMhtnMZ1Q5v26nxUPSkdm29oq\ne95CSjgeKdRlB6/7hqzeVimZ/g/ggS7qrHCQeJmVWw1UZp1DSa0MPy+atjZFrRwl5VdIPFj+RfKT\nknI/yF77Tl3EW+59O4CSL6Ps9Zxdw2ez8N58smR6tYnHc8DQoukTsxg/WjTtb6QkYUQn8RxNUStH\nF/vftKzs+KJpg7P955Eyr7HT/a+TmP6Y7dObFE1bj5RE/aVMXRzV2fpKyp6Y7U/rZnHOpiiBBi7P\nng/tap1F694nW2az7PkQ0o+yLn89k/r8LAV+30mZdUjH6GtKpn812+7BJfvRMmD/ommjste+hBVb\nBApfvgeVfJ7agT+UbOvH2XoLCX/FnzOWH+PeX/KaVjiWZu/Nq4V6LJr+XdLxfkQNn4FzKPlhUrQv\nv1zt57faR7+6qiUTpMxyQsljjxUKRfybtLMdQsr+3kbakct1Kvt1RCwqWvYy0pfERzsNJOLNwv+S\n1iSdMvg7MLbC1/Gzkml/JyUvm2TPdyUlTZdIWrvwyJa9g9Tyg6T1SOduL4yI14riu5H0S7vzQCLa\nSMnD/iWz9gduj4inisoWv+ahWTw3A5tJai1Zfm5E3FDB9t865y6pRdLbSAe4Bylfl7+KiGVFz/9O\n+rLfLHu+LTAS+FH22lYiaS1S/f0eGFZSv38GtpS0flexA5dFxEslryciYmm2HWXbWpXU/FnJvlHO\nR0nv+7SS6WeQXvuenS1c8r6tlr3OO7Jli2N6Bdi+wtfeCJdERHFn0BXeW0nrAB8EfhkRT9dpm3uQ\nTifeVpgQqQPlz4CRkrYqKd/V/rcSSS2kL8Q/RsRbV/tExLOkjqEfkDSkG6/h28DzpJabm0injo6N\niCuy+YWre8p+HjrwaeBfEfFoFutrpJbXAytY9hOkOjmlkzITSKclf1Qy/edZnKX79GsRcWnhSUT8\nl7S/zomI4qt+7sj+lr4fAZxbMu2cLM7CsX5Pqvuc/Scibi2K6QXScat42/uS9pEFJceZG0mt86VX\ntXX6GejCK8BgSbtVULZm/bVz6T+jss6lPySd79oOOD4iHuygXLmrYR4mfXl1SNLHSE3X7wFWK5pV\naY/5J0uev5z9XSv7uwVph/tLmWWD5Z3PColKudfxIOmLuCu/A/bOesHfLmkz0imQycWFJO1IOsjt\nQDpdUBzPMFY8sM2tYLtIEukSzK+QDpgDitb5QplFuqq3zbNl/93JZgt1+x3g1DLzg/TrcV4X4T9W\nbqKkg0nNy+8gHVwLHu1ifR0p/Bpa4T2OiPmSXmH5PlBWlvx8i9TKsW7RrML7VnAsqan5SUmzSH1k\nLoqIit7LOljhvY2IV9Lu8dZ7Wzj4dvbeVmsTUmtoqTlF84sT+K72v3LeTvq8/LeD7bQAGxVts1o/\nIyXR7aQvn3/Hip1oC19krax8lc9KJA0jfRmfI2nzolm3Ap+UtEVEdHYV4WZZLJ29nsI+u0KdRMQS\nSY+y8j79FCtbwMr7zKsl+0yx0pgfyeIcmT0vnF6u9HP2RJltvFyy7S1Jp/+eL1O2cJwp1tVnoDPn\nkfogXSPpGdKPqEsjYmYFy1asvyYeldqc9KZDeuPrRtIHSZ1Q/0r6wpxHavL7Aum8WyWWlZkmlvdY\nbiHtmJ8hdfIrtbTyiLt0JamJcH/SQbjQDH/ZW4GlZOQG0sFkCukDspj0K+DrrDyuzOsVbvsE0i+j\nX5CaJV8iffjPKrNOKF9vUF1P78J6Tye1iJVTyeXZK71GSZ8hnUr4A6mJ9jlSzMdT2a+WzkSNy/2e\nlCz+gHS++TVSHcykqI4j4veSbib9Yt2VdJ75OEmfqPHg1VG8AzqYXo/3ttF6YowPRcRNncx/IPu7\nNalDfFf2J/2YOpq0DxQLUqtH3pdydlTv3Xk/Oto/K/2cVbLtFlJfvv/rIKbSZLTm1xMRz0t6D6lP\n4x7Z4/OSfh0Rn+9q+Uo58ehA9iv6QlJGPA04QdJlUXIZWabcwDlbkA7QHfkk6Utnt0KzerbdL9Yc\ndFK8wz9C2tme7+KgUmi6Lfc6Rle00YhFkq4C9pN0NOnA8/esKbhgIumUwcTiZm5Ju1SyjU7sQ+oH\ncGjxxOz0VblfCV0p1Nu7SM3O5RRaHpZ0Ube12IfUN2Df4omSSpudq0kiHicdwLYktWIV1rkusCad\nDNaV1ePOwEkRcVrR9C3KlY+I+cBPgJ9kpzbuIiWHtSQeL5NasUp12kLTicL79q4uylVbt+U+J2OK\n5nfX86TThx1tp52VW1Lq6UpSh9bPUFni8WlS581yycVh2fzOEo9HSPvrVqR+b+UU6nU0RS2HkgaS\n9pnrK4izWluy4vu5BSnOQotezZ+zTjxC6ntXruW6Vh3u39n30dXZA0nnA4dK+k7htFl39cc+HpU6\nmvQL7xDSVQ23Audn/QdKHVR8flXSfsD6pGbmjiwjvflvJX+SRgJ7dzfwIjNJzaLHZyNDriD7Uiic\nJ74bOLi4n4Wkj5A++JX6HbAB6SqBbUi9uosVMvG39rusSfZzVWyjnGWUZPPZezCixvXNJh1Ivp7F\nt5KIeJ7UWvXlrI/MCgp1W6OVfrFI2h4YXzK50K9ozQrWeQ2pjr5eMv1o0n54dQXxlB4vplB0AMv6\n16ww0md2zvoZVjyVWI1HgHdk57QL29mGdAli1bJ4bga+IGmjTooWBrmqtG7fl71HhRgHk646mhsR\nXfaT6krWt+zPpNOZGxdtZziphfTvxf2z6i0ibiddOfIlSSsdoyStKumH2f8bkvod/C4i/lD6ILXm\nbSFpu042+SfSvnVy9iOwnBtIrcSTS6Z/idQn5aoqXmIlBBxeMm0yKc7rsufd+Zx15FJgvKRdVwpI\nGpaN/1OthdnyK3xeO/h+uy/7W+tneCX9scVDwEcljSkz79aImJvNOwW4ICKuAZD0OdKX8/mk0wjF\nXgL+IekCUi/zr5Gav37RSRxXk87hz5T0W9IASl8lXVpZyT0qOvowvjU9ItokfYU0fPJsSZeQfjlt\nTDq98Q+Wf2i/Sfqg3qJ0H4q1SZev3U/qjV6Ja0hN8KeTTuP8oWT+n0kHiqsk/ZR0vvhLpNNAK315\nV+Eq4KQs7ltJzcEHkr60qhYRkdXbDODu7H2dR+pvsVVEFDoiH07quHWfpJ+Tfk0PJyUII6isb0xH\nr+eTkv5E2k82A75M6pfw1nsREW9I+g9wgKSHSPvh/VnH6NLXdK+kX5N+uaxFurJje+AgUk/9v3VS\nH23Z6ZNjJa1KGithV9J57eL9sBV4StJlLD8d8xHgvaR9vRa/ypb9s6Rfkur3y6T9stxw5pWYTHrf\nZkv6GSnJ3JTU67/wns0ivbbvZp+bJcCMiCh3+u/7pC//6ySdTXofPkdqlflkjTGWcyKpQ+Utks4j\nJYSHkloRj63jdjpyEOnHzOVZ6+aNpC+wLUl94dYjXclW6Dx6ZQfruYYU+4Gky0dXEhGPSDqN9Jr/\nLukPpDGUtgOejogTIuIFSd8jJSfXkT6v7yCdur6TdEl0vW0q6QpSovH+7DVcHBH3ZXHX/DnrxA+B\nvUjHzQtJ++Zg0vfEJ0mfw5c6WrgDhf37HEkzSVe4/A74RZZ83ETqEzOS9D1wV0TU2n9oZY24VKan\nPlg+jkdHj4NIv+ruIDXdtZYsX7jsbd/s+U7Z8/1JHQy7GkDskZJpnyOdO11E+lI5iPLjLzxK6oVf\n+jrGlpQrxPOhkukfYvlAWAtJSdEvgW1LyhUGEFtEynL3Lhd3F3X8/7IYyg4UREp47srieIT0S+Bz\nlFy6mL3mKzpYR2l9rErqe/BUVv9/A95H+vDcWKZ+Si/Z3KTw/pdMH086wLxCajm6i5XHGxmZ1dHT\npEv7nsje/493UU+FbU7pYP5x2essDCy3Rwf70Pakg+zr2foKA4hNBZaWlG0hHcgLAws9Ruocu9Ll\nyGXiWZ/lA6W9BEwnJQHLSKdgIHWC/T6pxahQZ7OBQytYf9n3Jps3iZSQv046YE4orYvO6rM4xqJp\nY4pez0JS58+pJWWOz97PJcX7Z+n+V7Qf/I4VBxDbvZLX2NH+10E9bUP6LBcGELseeF81+1atZbPy\nq5Faum7PYniddAybBozMytxD15fg30Q6XrZ0Ue7gbP9fROoofhOwc0mZr5COn2+QWtfOoeSyX1IH\n+3vKrL/scSark7OKnk/NphUGEHuF5QOIlQ4MVtHnrJNt/4Wi41Y2bRDpO+bBrM7nk5Lnr7N8fI6K\nPwNZjD8iXcW0lOw7h5TIXJu9N6+TkvJzqXDslUofyjZmNZC0E2kn2TdSE6KZmfUxSjdUOxl4e5Rc\n/m7V6zF9PCQdLmmu0t0Ob+/i/F/h3Na5kp6R9IakByTtnle8ZmZmVr0e0cdD0gGkAVYOJTUbTyH1\nfRgVqTNYafmBpI5Fz5Kahp4hNTO9klvQZmZmVrUekXiQEo2fRsRFAJIOI/UF+ALp3H2pL5J6m+8Q\ny0cALDcQSx58rsrMzKxCTe/jkbVeLCLdJXBG0fQLSXcY/ESZZa5m+d1A9yZdqfFb4P+i/JDmZmZm\n1gP0hBaPdUijEJaOrDmfjgev2ow0oNHFpN7+W5Auc12F1HvYzMzMeqCekHjUooWUmBwaqcnmrmzQ\nmmPoIPHIBiDajXRp0xs5xWlmZtYXrE66bHxmRLzYnRX1hMTjBdI1xsNLpg8ndR4tZx6wOFY8TzQH\nWE/SKlE0BHmR3WjMgDJmZmb9xYGkrg01a3riEelOgrOAXUgjzxXuk7ILcHYHi93CyjdSGw3M6yDp\ngGws/4svvpgxY8oNWmqNMGXKFKZNK71DtDWS6zx/rvP8uc7zNWfOHD7zmc9AB3fUrkbTE4/MmcCF\nWQJSuJx2EOkmbUi6CHgqIo7Pyp8PHJ4NT3wOMIo05PePOtnGGwBjxoxh7NixjXgNVsawYcNc3zlz\nnefPdZ4/13nTdLurQo9IPCLi0uymWqeQTrHcTbpra+HOohtSdAv3iHhK0m6kYXrvIQ1XPY3yl96a\nmZlZD9EjEg+AiDgPOK+DeTuXmXYH6SY9ZmZm1kv0mCHTzczMrO9z4mENNWlSaR9gazTXef5c5/lz\nnfdeTR+5NC+SxgKzZs2a5Q5JZmZmVZg9ezbjxo0DGBcRs7uzLrd4mJmZWW6ceJiZmVlunHiYmZlZ\nbpx4mJmZWW6ceJiZmVlunHiYmZlZbpx4mJmZWW6ceJiZmVlunHiYmZlZbpx4mJmZWW6ceJiZmVlu\nnHiYmZlZbpx4mJmZWW6ceJiZmVlunHiYmZlZbpx4mJmZWW6ceJiZmVlu+l3i8bGPHcbkyVNpa2tr\ndihmZmb9Tr9LPObNO59zzx3P+PH7OPkwMzPLWb9LPEC0t+/OnDlTOPHEM5odjJmZWb/SDxOPpL19\nd2bMuKXZYZiZmfUr/TbxALFkySAiotmBmJmZ9Rv9OPEIBg5ciKRmB2JmZtZv9NvEo6XlOvba6wPN\nDsPMzKxfWaXZAeQvaGm5ljFjpnHqqZc3OxgzM7N+pd+1eKy//lc54og7uO22y2ltbW12OGZmZv1K\nv2vxuOqq8xk7dmyzwzAzM+uX+l2Lh5mZmTWPEw8zMzPLjRMPMzMzy40TDzMzM8uNEw8zMzPLjRMP\nMzMzy40TDzMzM8uNEw8zMzPLjRMPMzMzy02/SzyWLm12BGZmZv1Xv0s8Xn+92RGYmZn1Xz0m8ZB0\nuKS5kl6XdLuk7Tope7CkdknLsr/tkhZVsp2FC+sXs5mZmVWnRyQekg4AzgCmAtsC9wAzJa3TyWIL\ngPWKHptUsi0nHmZmZs3TIxIPYArw04i4KCIeAA4DFgFf6GSZiIjnI+K57PF8JRtaVFG7iJmZmTVC\n0xMPSQOBccCNhWkREcANwPhOFh0i6TFJT0j6k6StKtmeWzzMzMyap+mJB7AOMACYXzJ9PukUSjkP\nklpD9gIOJL2OWyVt0NXG3OJhZmbWPKs0O4BaRMTtwO2F55JuA+YAXyb1E+nQz38+hZtvHrbCtEmT\nJjFp0qQGRGpmZta7TJ8+nenTp68wbcGCBXVbf09IPF4AlgHDS6YPB56tZAURsVTSXcAWXZXde+9p\n/OAHY6sO0szMrD8o92N89uzZjBs3ri7rb/qplohYAswCdilMk6Ts+a2VrENSC7A1MK+rsu7jYWZm\n1jw9ocUD4EzgQkmzgDtJV7kMAi4EkHQR8FREHJ89P4l0quVhYE3gWGBj4Bddbch9PMzMzJqnRyQe\nEXFpNmbHKaRTLHcDuxVdIrshUDzY+VrAz0idT18mtZiMzy7F7ZQTDzMzs+bpEYkHQEScB5zXwbyd\nS54fBRxVy3Z8qsXMzKx5mt7HI29OPMzMzJrHiYeZmZnlpt8lHu7jYWZm1jz9LvFwi4eZmVnz9LvE\nwy0eZmZmzdPvEg+3eJiZmTVPv0s8Fi2CiGZHYWZm1j/1u8QjwqdbzMzMmqXfJR4AbW3NjsDMzKx/\ncuJhZmZmuemXicerrzY7AjMzs/6pXyYebvEwMzNrDiceZmZmlhsnHmZmZpabfpd4SE48zMzMmmWV\nWhaSNAD4ODAmm/RvYEZELKtXYI0yeLA7l5qZmTVL1YmHpC2Aq4ENgQezyd8EnpS0Z0Q8Usf46m7Q\nILd4mJmZNUstp1rOBh4FNoqIsRExFtgYmJvN69EGD3biYWZm1iy1nGrZCdghIl4qTIiIFyV9A7il\nbpE1iBMPMzOz5qmlxeNNoLXM9CHA4u6F03g+1WJmZtY8tSQeVwE/k7S9ltsB+Akwo77h1Z87l5qZ\nmTVPLYnHZOAR4DbgjexxC/Aw8LX6hdYYbvEwMzNrnqr7eETEK8DekrYE3pFNnhMRD9c1sgYZPBie\nfLLZUZiZmfVPNY3jARARDwEP1TGWXLhzqZmZWfNUlHhIOhM4KSIWZv93KCKOqktkDTJokPt4mJmZ\nNUulLR7bAgOL/u+13OJhZmbWPBUlHhHx4XL/90aDBsHixemx6qrNjsbMzKx/qfqqFkm/krTSOB6S\nBkv6VX3CapzBg9Nft3qYmZnlr5bLaQ8G1igzfQ3goO6F03hOPMzMzJqn4qtaJA0FlD1aJb1RNHsA\n8FHgufqGV3+DBqW/7mBqZmaWv2oup30FiOzx3zLzA5haj6AayS0eZmZmzVNN4vFhUmvHTcA+wEtF\n8xYDj0fEM3WMrSEKLR5OPMzMzPJXceIREX8DkLQp8GREtDcsqgZyi4eZmVnz1DJk+uMAkgYBGwOr\nlsy/tz6hNYZbPMzMzJqn6sRD0tuBC4A9OigyoFsRNdiAAR691MzMrFlquZz2R8CawPbA68DupEts\nHwL2ql9ojdPa6hYPMzOzZqjlJnE7A3tHxL8ktZM6lV4v6VXgm8DVdY2wAZx4mJmZNUctLR6DWT5e\nx8vA27P/7wPG1iOoRhs61ImHmZlZM9SSeDwIjM7+vwf4sqQRwGHAvHoF1khu8TAzM2uOWk61nAWs\nn/3/beA64EDSWB6fq09YjdXa6s6lZmZmzVB1i0dEXBwRF2b/zwI2AbYDNoqI39UaiKTDJc2V9Lqk\n2yVtV+Fyn5LULukPlW7LLR5mZmbNUVXiIWmgpEckjSlMi4hFETE7Il6oNQhJBwBnkIZc35Z0Cmem\npHW6WG4k8EPg5mq258TDzMysOapKPCJiCbB6A+KYAvw0Ii6KiAdI/UUWAV/oaAFJLcDFwMnA3Go2\n5s6lZmZmzVFL59JzgeMk1dI/ZCWSBgLjgBsL0yIigBuA8Z0sOhWYHxEXVLtNt3iYmZk1Ry3Jw3bA\nLsCuku4DFhbPjIhPVrm+dUijnc4vmT6f5VfPrEDSB4DPA9tUuS3AnUvNzMyapZbE4xXg8noHUilJ\nQ4CLgEMi4uVa1tHaCgsXQns7tNTS5mNmZmY1qeUmcZ+vcwwvAMuA4SXThwPPlim/OelKmislKZvW\nAiBpMTA6Ijrs8zFlyhQWLhwGwMc+BqusApMmTWLSpEndexVmZmZ9wPTp05k+ffoK0xYsWFC39St1\np2guSbcDd0TE17LnAp4Azo6IH5aUXRXYomQVpwFDgMnAQxGxtMw2xgKzZs2axfz5Y/noR+Gpp2DE\niAa8IDMzsz5k9uzZjBs3DmBcRMzuzrrq0kG0Ds4ELpQ0C7iTdJXLIOBCAEkXAU9FxPERsRj4T/HC\nkl4h9UmdU8nGWlvTX3cwNTMzy1ePSDwi4tJszI5TSKdY7gZ2i4jnsyIbAiu1YtSqkHi4g6mZmVm+\nekTiARAR5wHndTBv5y6WrarfiVs8zMzMmqNb13RIasRgYg3nxMPMzKw5qk48JLVIOknS08BrkjbL\npn9H0hfrHmEDDB2a/jrxMDMzy1ctLR4nku5CeyzpjrQF9wNfqkNMDbfaajBwoBMPMzOzvNWSeBwE\nHBoRvyGNv1FwD/COukSVA49eamZmlr9aEo8RwMMdrGtg98LJj+/XYmZmlr9aEo//AB8sM31f4K7u\nhZMfJx5n25HmAAAgAElEQVRmZmb5q+Vy2lOAX0saQUpcPilpNOkUzMfqGVwjDR3qxMPMzCxvVbd4\nRMQVwERgAunOtKcAY4CJEXF9fcNrHLd4mJmZ5a+mAcQi4u/AR+ocS65aW6GO97wxMzOzCtQyjsd2\nkrYvM317Se+tT1iN5xYPMzOz/NXSufRcYKMy00dk83oFJx5mZmb5qyXx2Aood0vcu7J5vYI7l5qZ\nmeWvlsTjTdIdZEutTx3vINtobvEwMzPLXy2Jx5+B70kaVpggaU3gu0Cvuqrl1VchotmRmJmZ9R+1\nXNVyDHAz8LikwoBh7wHmA5+tV2CN1toKy5bBG2/AGms0OxozM7P+oerEIyKelvRu4EBgG+B14AJg\nekQsqXN8DdPamv62tTnxMDMzy0ut43gsBH5W51hyNXRo+tvWBuuu29xYzMzM+ouaEg9JWwIfBtal\npJ9IRJxSh7garrjFw8zMzPJRdeIh6RDgfOAF4FmguHtmkIZQ7/EKicerrzY3DjMzs/6klhaPE4ET\nIuL/6h1MntziYWZmlr9aLqddC/h9vQPJmxMPMzOz/NWSePwe2LXegeRt8GCQnHiYmZnlqZZTLQ8D\n35G0A3AfsMIltBFxdj0Ca7SWFhgyxH08zMzM8lRL4nEo8BqwU/YoFkCvSDzAw6abmZnlrZYBxDZt\nRCDN4MTDzMwsX7X08egznHiYmZnlq9YBxDYE9gI2BlYtnhcRR9UhrlwMHerEw8zMLE+1DCC2CzAD\neBR4B3A/MBIQMLuewTVa4Q61ZmZmlo9aTrV8Dzg9IrYG3gD2ATYC/kYvG9/Dp1rMzMzyVUviMQa4\nKPt/KbBGRLwGnAwcV6/A8uDEw8zMLF+1JB4LWd6vYx6wedG8dbodUY7cx8PMzCxftXQuvR34ADAH\nuAY4Q9LWwCezeb2GWzzMzMzyVUvicRQwJPt/avb/AcBD2bxew51LzczM8lXLAGKPFv2/EDisrhHl\nqLUV3ngDli6FVWq6sNjMzMyq0e8HEAOfbjEzM8tLRb/zJb0EjIqIFyS9TLonS1kR8bZ6BddoQ4em\nv21tsNZazY3FzMysP6j0BMMUoNAu8PUGxZI7t3iYmZnlq6LEIyJ+DSBpFVJrx8yImN/IwPJQSDzc\nwdTMzCwfVfXxiIilwE+A1RsTTr7c4mFmZpavWjqX3glsW+9AmsGJh5mZWb5quYj0PNKgYRsCs0gj\nmb4lIu6tR2B5cOJhZmaWr1paPC4BNgXOBm4B7gbuKvpbE0mHS5or6XVJt0varpOyn5D0T0kvS3pN\n0l2SPlPtNgcOhNVXd+JhZmaWl1paPDatdxCSDgDOAA4lncqZAsyUNCoiXiizyIvAqcADwGJgInCB\npPkRcX012/bopWZmZvmpZeTSxxsQxxTgpxFxEYCkw4A9gS8APygTw80lk86WdDDpHjJVJx5u8TAz\nM8tHzQOFS9oK2Jjld6oFICJmVLmegcA44LtF6whJNwDjK1zHLsAo4G/VbBuceJiZmeWp6sRD0mbA\nH4GtSWN6KJtVGM10QJWrXCdbpnRckPnA6E7iGAo8DawGLAW+GhE3Vblthg514mFmZpaXWjqXngXM\nBdYFFgHvBD4E/Av4n7pF1rU2YBvgvcAJwDRJH6p2JW7xMDMzy08tp1rGAztn921pB9oj4h+Svkm6\n0qXaMT5eAJYBw0umDwee7WihiAigcKfce7NTP98ESvt/rGDKlCkMGzbsred33w1rrjkJmFRl2GZm\nZn3P9OnTmT59+grTFixYULf115J4DGD5fVteADYAHgQep5NTIx2JiCWSZgG7ADMAJCl7fnYVq2oh\nnXbp1LRp0xg7duxbzw85BO65p6qQzczM+qxJkyYxadKKP8Znz57NuHHj6rL+WhKP+0mnOOYCdwDH\nSlpMuhT20c4W7MSZwIVZAlK4nHYQcCGApIuApyLi+Oz5N0indh4hJRt7Ap8BDqt2wz7VYmZmlp9a\nEo9TgcHZ/ycDVwF/J42tcUAtQUTEpZLWAU4hnWK5G9gtIp7PimxI6kBaMBg4N5v+Omk8jwMj4rJq\nt+3OpWZmZvmpZRyPmUX/Pwy8Q9LbgJezfhc1iYjzSMOxl5u3c8nzk4CTat1WMbd4mJmZ5afqq1ok\nfUbS4OJpEfFSd5KOZiokHr0zejMzs96llstppwHzJf1W0kclVTtuR4/S2pqSjoULuy5rZmZm3VNL\n4rE+8CnSgGGXAvMknSvp/XWNLCe+Q62ZmVl+qk48ImJpRFwVEQeSBhGbAowE/iLpkTrH13BDh6a/\nTjzMzMwar+Z7tQBExCJJM4G1gE2AMXWJKkdu8TAzM8tPLadakDRI0oGSriHdL+XrpPu3vLOeweWh\nkHi8+mpz4zAzM+sParlJ3CXAx0j3abkU+E5E3FbvwPLiFg8zM7P81HKqZRmwPzAzIpbVOZ7cOfEw\nMzPLTy0DiB3YiECaZY01YMAAJx5mZmZ5qKmPR18iefRSMzOzvPT7xANS4uHOpWZmZo3nxAO3eJiZ\nmeXFiQdOPMzMzPJSUedSSUMrXWFE9LqTFkOHOvEwMzPLQ6VXtbxCujdLJXrdTePcx8PMzCwflSYe\nHy76fyTwfeBCoDBw2HjgYOCb9QosT62t8OyzzY7CzMys76so8YiIvxX+l3QycFRETC8qMkPSfcCh\nwK/rG2LjuY+HmZlZPmrpXDoe+FeZ6f8C3te9cJrDiYeZmVk+akk8ngQOKTP9S9m8XsedS83MzPJR\ny71apgCXS9oDuCOb9j5gS2CfegWWJ3cuNTMzy0fVLR4RcQ0wCrgSeFv2uBIYlc3rdVpbYckSePPN\nZkdiZmbWt9XS4kFEPAkcX+dYmqb4DrWrrdbcWMzMzPqymkYulfRBSRdLulXSiGzaZyV9oL7h5aM4\n8TAzM7PGqTrxkLQPMBN4HRgLFNoIhtFLW0GGZuOyOvEwMzNrrFpaPE4EDouIQ4AlRdNvISUivU6h\nxcMdTM3MzBqrlsRjNHBzmekLgDW7F05z+FSLmZlZPmpJPJ4Ftigz/QPAo90LpzmceJiZmeWjlsTj\n58BZkrYn3ThuA0kHAqcD59czuLwMGZL+OvEwMzNrrFoup/0+KWG5ERhEOu3yJnB6RJxTx9hyM2AA\nDB7sxMPMzKzRqk48IiKA0yT9kHTKZQjwn4h4rd7B5cmjl5qZmTVeTQOIAUTEYuA/dYylqXyjODMz\ns8arOvGQNBj4BrALsC4l/UQiYrP6hJYvJx5mZmaNV0uLxy+AnYD/B8wjdTDt9Zx4mJmZNV4ticce\nwJ4RcUu9g2mmoUOdeJiZmTVaLZfTvgy8VO9Ams2dS83MzBqvlsTjJOAUSYPqHUwz+VSLmZlZ49Vy\nquVoYHNgvqTHWPF+LUREr71fixMPMzOzxqol8fhT3aPoAZx4mJmZNV4tA4h9uxGBNJs7l5qZmTVe\nLX08+qTWVli4EJYta3YkZmZmfVdFLR6SXgJGRcQLkl6mk7E7IuJt9QouT4U71L72Ggwb1txYzMzM\n+qpKT7VMAQonIr7eiEAkHQ4cA6wH3AMcGRH/7KDsl4CDgHdlk2YBx3dUvhKFxKOtzYmHmZlZo1SU\neETEr8v9Xy+SDgDOAA4F7iQlOjMljYqIF8osshPwW+BW4A3SEO5/lrRVRMyrJYbixMPMzMwao1t9\nPCStLmlo8aPGVU0BfhoRF0XEA8BhwCLgC+UKR8RnI+InEXFvRPwX+BLptexS4/YZmkXuxMPMzKxx\nqk48JA2W9GNJzwELSSOZFj+qXd9AYBxwY2FaRARwAzC+wtUMBgbSjRFVCy0eCxb0iVvPmJmZ9Ui1\ntHj8ANgZ+ArwJqm1YSrwDKnfRbXWAQYA80umzyf196jE/wFPk5KVqrW1tfG9700FJvCpT32cTTed\nwOTJU2lz84eZmVld1TKA2ETgoIj4q6QLgL9HxMOSHgcOBH5T1wi7IOkbwP7AThGxuKvyU6ZMYVhR\n79GlS5dy331P8swzPwS+xUsviZdeCs49dyY33bQPt912Oa2F5hAzM7M+bvr06UyfPn2FaQsWLKjb\n+pXOalSxgPQasFVEPCHpKeCTEXGnpE2B+yJiSJXrG0jqz7FPRMwomn4hMCwiPtHJsscAxwO7RMRd\nXWxnLDBr1qxZjB27fFT3yZOncu6542lv332lZVparuWII+7grLO+Vc1LMjMz61Nmz57NuHHjAMZF\nxOzurKuWUy2PAptm/z9Aam2A1BLySrUri4glpMth3+oYKknZ81s7Wk7SscAJwG5dJR2dufLKW2hv\n363svPb23Zkx45ZaV21mZmYlakk8LgC2yf7/PnC4pDeAacAPa4zjTOAQSQdJegfwE2AQcCGApIsk\nfbdQWNJxwCmkq16ekDQ8ewyuZqMRwZIlgwF1UEIsWTKIaluFzMzMrLxa7tUyrej/G7JEYRzwcETc\nW0sQEXGppHVIycRw4G5SS8bzWZENgaVFixxGuorlspJVfTtbR0UkMXDgQtJArOWSj2DgwIWkBhgz\nMzPrrlo6l64gIh4HHq/Des4Dzutg3s4lzzctV64WEyfuyLnnzizbxwOuY8cdP1CvTZmZmfV7ld6r\nZXKlK4yIs2sPJ3+nnXYMN920D3PmRJZ8CAhaWq5j4MBp/OlPl3PFFbD33suXiQi3gpiZmdWgmnu1\nVCKAXpV4tLa2ctttl3PiiWcwY8aZLFkyiIEDF7HXXjvyjW9czhFHtPLxj8MJJ7Tx6qunc+WVt7Bk\nyWAGDlzIxIk7ctppx/hyWzMzswpVfTltb9XR5bSlSlsz2tvhpJPa+O539wGOAnZjeavITMaMOdNj\nfZiZWZ/W7Mtp36JMd9bR05S+nJYWaGs7HekooHAqBkC0t+/OnDlTOPHEM/IO08zMrFeqKfGQ9EVJ\n95PuDPuGpPuzW9X3SVdeeQsRHuvDzMysu6q+qkXSKaRzDucAt2WTxwPTJG0cESfXMb6mq2asjz7W\n+GNmZlZ3tVxO+xXgkIgoHsh9hqR7SclIn0o8PNaHmZlZ/dRyqmUg8K8y02dRh3FBeqKJE3ekpWVm\n2XktLdex114e68PMzKwStSQe/4/U6lHqUHK+M21eTjvtGMaMOZOWlmtJLR+Qrmq5ljFjpnHqqUc3\nMzwzM7Neo9arWr6YdSj9Rfa4DzgEaJd0ZuFRxzibqjDWxxFH3MHIkbsyYMDeDB26K0cccYcvpTUz\nM6tCLadG3gUUruHdPPv7QvZ4V1G5PjVASGtrK2ed9S3OOgv23DN1JD3rrGZHZWZm1rvUcpO4Dzci\nkN5k9Ghx1VXNjsLMzKz3qfpUi6S3dzJv6+6F0zuMHg2PPgqLFzc7EjMzs96llj4e90nas3SipGOA\nO7sfUs83ejQsWwZz5zY7EjMzs96llsTjTOBySedLWkPSCEk3AscCn65veD3TqFHp74MPNjcOMzOz\n3qbqxCMifkAaqfSDwL3Z403g3RHxx/qG1zOtvz4MGeLEw8zMrFq1Xk77MHA/MBIYCvwuIp6tV1A9\nnZROt/z3v82OxMzMrHeppXPpjqRWji2Bd5MGEztH0u8krVXn+HqsUaPc4mFmZlatWlo8bgJ+B+wQ\nEXMi4hfAtsDGwH31DK4nGz3aiYeZmVm1ahlAbNeI+FvxhIh4JGsJOaE+YfV8o0bBc8/BK6/Amms2\nOxozM7PeoZbOpX/rYHp7RHyn+yH1DqNHp7/u52FmZla5ihMPSddIGlb0/BuS1ix6vrak/9Q7wJ7K\nl9SamZlVr5oWj92A1YqeHw+8rej5KsDoegTVGwwZAhts4BYPMzOzalSTeKiL5/2OO5iamZlVp9Zx\nPAwnHmZmZtWqJvEIVr7VfenzfmXUKHjoIWhvb3YkZmZmvUM1l9MKuFDSm9nz1YGfSFqYPV+t/GJ9\n1+jR8Prr8NRTsPHGzY7GzMys56sm8fh1yfOLy5S5qBux9DrFl9Q68TAzM+taxYlHRHy+kYH0Rpts\nAgMHpn4eEyY0OxozM7Oez51Lu2GVVWCLLdzB1MzMrFJOPLrJd6k1MzOrnBOPbvJdas3MzCrnxKOb\nRo+Gxx9PV7eYmZlZ55x4dNOoURABjzzS7EjMzMx6Pice3VS4pNanW8zMzLrmxKOb1lkH1lrLiYeZ\nmVklnHh0k5ROt/jKFjMzs6458agD3yzOzMysMk486qCQeES/vmWemZlZ15x41MGoUfDyy/Dii82O\nxMzMrGfrMYmHpMMlzZX0uqTbJW3XSdmtJF2WlW+XNDnPWEv5yhYzM7PK9IjEQ9IBwBnAVGBb4B5g\npqR1OlhkEPAIcBwwL5cgO7HFFqmTqRMPMzOzzvWIxAOYAvw0Ii6KiAeAw4BFwBfKFY6If0XEcRFx\nKbA4xzjLWmMN2HhjX9liZmbWlaYnHpIGAuOAGwvTIiKAG4DxzYqrWr6yxczMrGtNTzyAdYABwPyS\n6fOB9fIPpza+S62ZmVnXekLi0SeMGgUPPwzLljU7EjMzs55rlWYHALwALAOGl0wfDjxb741NmTKF\nYcOGrTBt0qRJTJo0qVvrHT0aFi+Gxx6DzTfv1qrMzMyaZvr06UyfPn2FaQsWLKjb+pueeETEEkmz\ngF2AGQCSlD0/u97bmzZtGmPHjq33ahk1Kv3973+deJiZWe9V7sf47NmzGTduXF3W31NOtZwJHCLp\nIEnvAH5CumT2QgBJF0n6bqGwpIGStpH0HmBVYET2vGlf+RttlK5ucQdTMzOzjjW9xQMgIi7Nxuw4\nhXSK5W5gt4h4PiuyIbC0aJENgLuAwiDlx2SPvwE75xJ0iZYW2HJLJx5mZmad6RGJB0BEnAec18G8\nnUueP07Paa15i+9Sa2Zm1rke9+Xdm3ksDzMzs8458aij0aPh6afhtdeaHYmZmVnP5MSjjgpXtjz0\nUHPjMDMz66mceNSR71JrZmbWOScedbTmmrDuuk48zMzMOuLEo858ZYuZmVnHnHjUma9sMTMz65gT\njzor3KU2ouuyZmZm/Y0TjzobNQra2uDZut/ezszMrPdz4lFnvrLFzMysY0486myzzWDAAHjwQZ9r\nMTMzK+XEo47a2to45pipSBM45piPs+mmE5g8eSptbW3NDs3MzKxHcOJRJ21tbYwfvw/nnjuepUuv\n57XXruCxx67n3HPHM378Pk4+zMzMcOJRNyeccDpz5hxFe/vugLKpor19d+bMmcKJJ57RzPDMzMx6\nBCcedXLllbfQ3r5b2Xnt7bszY8YtOUdkZmbW8zjxqIOIYMmSwSxv6SglliwZRHhwDzMz6+eceNSB\nJAYOXAh0lFgEAwcuROooMTEzM+sfnHjUycSJO9LSMrPsvJaW69hrrw/kHJGZmVnP48SjTk477RjG\njDmTlpZrWd7yEbS0XMuYMdM49dSjmxmemZlZj+DEo05aW1u57bbLOeKIOxg5cldGjNibAQN25V3v\nuoPbbruc1tbWZodoZmbWdKs0O4C+pLW1lbPO+hZnnZU6nB51lLjkEhg0qNmRmZmZ9Qxu8WgQSey3\nX7pZ3C2+ktbMzAxw4tFQO+wAI0bA73/f7EjMzMx6BiceDdTSAvvuC5dfDu3tzY7GzMys+Zx4NNj+\n+8O8eT7dYmZmBk48Gs6nW8zMzJZz4tFgPt1iZma2nBOPHOy3HzzzDNx6a7MjMTMzay4nHjkYPz6d\nbrn00mZHYmZm1lxOPHLQ0gL77OPTLWZmZk48cuLTLWZmZk48cvP+98MGG/jqFjMz69+ceOSkcLrl\nsst8usXMzPovJx45Kpxuue22ZkdiZmbWHE48crTjjrD++j7dYmZm/ZcTjxz5dIuZmfV3Tjxytt9+\n8PTTcPvtzY7EzMwsf048crbjjrDeej7dYmZm/ZMTj5wNGLDi6ZaIaHZIZmZmuXHi0QR77tnGU09N\nZcSICWy00cfZdNMJTJ48lba2tmaHZmZm1lBOPHLW1tbG//7vPsB4nn32ep5++goee+x6zj13POPH\n71M2+aikVaTSlhO3sJiZWTP1mMRD0uGS5kp6XdLtkrbrovx+kuZk5e+RtEdesXbHCSeczpw5RwG7\nA8qmivb23ZkzZwonnngGkBKUyZOnsummHbeKVFKmmnIF9Ux0fvvb39ZlXXknX715e65z13l/2J7r\nPP/t1U1ENP0BHAC8ARwEvAP4KfASsE4H5d8PLAGOAkYDpwBvAlt1so2xQMyaNSuaaeTIXQLaA6LM\noz1GjpwQr776arzznR+JlpZri8q2R0vLtfHOd34kXn311YrKRERV5Y488uQYOXKXGDFirxg5cpc4\n8siT35pfaZnScquvPrzmdTUqpr6+Pde567w/bM91ns/2CmbNmhVAAGOju9/53V1BPR7A7cBZRc8F\nPAUc20H5S4AZJdNuA87rZBtNTzza29tjxIi9Okg60mPgwL1i9OiTQrq27PyWlmti8uSpceSRJ2fJ\nRMdlIqKico1NdCbWtK68k6++tT3Xueu8P2zPdd7o7RXrU4kHMDBrvdirZPqFwB87WOZxYHLJtG8B\nd3WynaYnHhFdt3gMGbJLrLpq52UGDpwQq63WeZk115wQp54asdZanZdbb70Jse++nScnX/nK1Pjq\nV2tNdCauVK6SZKheZcrH1Ne35zp3nfeH7bnOG729Yn0t8VgfaAe2L5n+f8BtHSzzJnBAybSvAPM6\n2U6PSDy6esOPPPLkLltFBg2aGGus0XmZlpa9Yp11loXUeTnYK6Dz5AQmdFlm1VUnxPvfH7H66qXl\nJq5QrrV1Qgwd2nXStOaanZdZe+0JsfbanZdZd90JceWVEcOHd518rbdefcrcckt0WW799bte1/rr\nT4j11++6zO23R5lyE2taV1dlNtigsjL//Ge5mGpbVyVlZs3qCdub2LDt9Z46cJ33le2NHDlhpe+u\neiYeq9TUMaR3Wh1gzpw5TQ1iv/0mcPXV/8vcuQ8R8X7SWaVAupWRI3/Dfvv9kMsuuxaYxfLOp8WC\nYcOeBeD11zsuM3z4PK666m4+9rF5zJvXcbl11pnHsmVr8PLLd3UY85AhbwLitdc6LtPSspg11/wX\n7e1LgOJyC4DZbz1buPBNJJWUWdErr3Rd5sUXU0ydlXnuucVMnPgvUoNax+WefbbrdVVWZjE77tj1\n9ubN63pdlZVZzA47lNveinVer+0980wlZRaz3XZd10Fl66pse+PG9YTtrVjn9dteb6qDvLfnOm/U\n9hYuXMysWbOy43BS9N25eocLVkgR0d11dC8AaSCwCNgnImYUTb8QGBYRnyizzOPAGRFxdtG0bwF7\nR8S2HWzn08Bv6hu9mZlZv3JgRHR9SVEnmt7iERFLJM0CdgFmACilWbsAZ3ew2G1l5n8km96RmcCB\nwGOkK2jMzMysMqsDI0nfpd3S9BYPAEn7kzqTHgbcCUwB9gXeERHPS7oIeCoijs/Kjwf+CnwTuBqY\nBHyDdO7pP7m/ADMzM6tI01s8ACLiUknrkMbjGA7cDewWEc9nRTYElhaVvy07dXJa9niIdJrFSYeZ\nmVkP1iNaPMzMzKx/6DFDppuZmVnf58TDzMzMctMvEo9qb0BnlZP0QUkzJD0tqV3SXmXKnCLpGUmL\nJF0vaYtmxNpXSPqmpDslvSppvqQ/ShpVUmY1SedKekFSm6TLJK3brJh7O0mHZTejXJA9bpW0e9F8\n13cDSfpGdnw5s2ia67zOJE3N6rn48Z+i+XWp8z6feEg6ADgDmApsC9wDzMw6s1r3DSZ1Bv4qaVS7\nFUg6DjgCOBR4H7CQVP+r5hlkH/NB4Bxge2AC6bYDf5a0RlGZHwF7AvsAHwI2AC7POc6+5EngONII\nyOOAm4ArJI3J5ru+GyT7oXgo6dhdzHXeGPeTLvJYL3t8oGhefeq8u0Of9vQHVd6Azo9u1XU7K99z\n5xlgStHzocDrwP7NjrevPIB1srr/QFEdvwl8oqjM6KzM+5odb195AC8Cn3d9N7SOhwAPAjsDfwHO\nzKa7zhtT31OB2R3Mq1ud9+kWj2xU1HHAjYVpkWrrBmB8s+LqLyRtSsqYi+v/VeAOXP/1tCapteml\n7Pk40qXyxfX+IPAErvduk9Qi6VPAINKgha7vxjkXuDIibiqZ/l5c542yZXbq/BFJF0vaKJtet/28\nR4zj0UDrAAOA+SXT55MyNWus9UhfiOXqf738w+l7slF+fwT8I5aPY7MesDhL8oq53rtB0rtIicbq\nQBvpl98DkrbF9V13WXL3HlKSUWo4rvNGuB34HKmVaX3SXd9vzvb9uh1X+nriYdbXnQdsxYrnYa0x\nHgC2AYaRRla+SNKHmhtS3yRpQ1JCPSEiljQ7nv4iIoqHQ79f0p3A48D+1PFWI336VAvwArCMlB0X\nGw48m384/c6zpD41rv8GkPRj4KPA/0TEM0WzngVWlTS0ZBHXezdExNKIeDQi7oqIE0idHb+G67sR\nxgFvB2ZLWiJpCbAT8DVJi0m/sldznTdWRCwA/gtsQR338z6deGSZcuEGdMAKN6C7tVlx9RcRMZe0\nQxbX/1DS1Riu/27Iko69gQ9HxBMls2eRbjFQXO+jgY3p/EaKVp0WYDVc341wA7A16VTLNtnjX8DF\nRf8vwXXeUJKGAJuTLhKo237eH061nAlcmN0Bt3ADukGkm9JZN0kaTMqGlU3aTNI2wEsR8SSpufRE\nSQ+T7gz8HdJVRVc0Idw+QdJ5pBsj7gUslFRoUVoQEW9ExKuSfgmcKellUn+Es4FbIuLO5kTdu0n6\nLnAtqSNdK+lO1zsBu7q+6y8iFgIr3HtL0kLgxYiYkz13ndeZpB8CV5JOr4wAvk1KNi6p537e5xOP\n6PoGdNY97yVd5hbZ44xs+q+BL0TEDyQNAn5Kuvri78AeEbG4GcH2EYeR6vqvJdM/D1yU/T+FdJrx\nMtKv8uuAw3OKry9al7RPrw8sAO4lJR2Fqy1c341XOk6Q67z+NgR+C6wNPA/8A9ghIl7M5telzn2T\nODMzM8tNn+7jYWZmZj2LEw8zMzPLjRMPMzMzy40TDzMzM8uNEw8zMzPLjRMPMzMzy40TDzMzM8uN\nE8cDOJgAAARsSURBVA8zMzPLjRMPMzMzy40TD7N+RtJwSedIekTSG5IelzRD0s7Njq2nkHSBpD80\nOw6zvqjP36vFzJaTtAnpzsAvAUcD9wMDgd2BHwNbNS86M+sP3OJh1r+cT7rJ03YR8aeIeDgi5kTE\nNGAHAEkbSbpCUpukBZJ+J2ndwgokTZV0l6TPZ60lbZJ+LKlF0rGS5kmaL+n44g1Lapd0mKRrJC3K\nWlz2KSnzLkk3ZvNfkPTT7A7IhfkXSPqjpKMlPZOV+bGkAUVlVpV0uqSnJL0m6TZJOxXNP1jSy5J2\nlfSfLP5rC3f5lTQVOBjYO4t5maQP1fVdMOvHnHiY9ROS1gJ2A34cEW+Uzs9uey1gBulOwh8EJgCb\nAZeUFN+c1EqyG/Ap4EvA1cAGwIeA44BTJW1XstwpwO+BdwO/AS6RNDqLbxAwE3gRGAfsm23/nJJ1\nfDiL6X+Ag4DPZY+Cc4Htgf2BrbPtXStp86Iyg0gtPgdmr3Nj4PRs3unApaQ7bw4n3ZH21tL6MrPa\n+FSLWf+xBSDgwU7KTADeCYyMiGcAJB0E/FvSuIiYlZUT8PmIWAQ8IOkvwKiI2COb/5Ck40hJwj+L\n1n9pRFyQ/X+ypI8ARwJHkJKA1YCDssRojqQjgCslHRcRz2fLvQQcEenW2v+VdDWwC/BLSRuTkpCN\nIuLZrPyZkvYAPg+cmE1bBfhyRPz/9u3m1aYoDuP490kppRTFQMrAOxEhycQ/IJQyMDBUt4xNJP+C\nzIzcYqgYomTgpZu63aSriBsDkWKAK2QZrH3q2Hm5J+dug/P91Kmz1t6ttfZoP639WzPNM54HTgOU\nUj4mmQUW9s0paUgMHtLoyBzu2QC87IUOgFLKdJL3wEagFzxmmtDR8xr41hrrNbC81Xe/1b4HbOub\ne6q1G3OHujO7HuiFgEdN6Oh5BWxp/m8BFlADSf/zLgTe9rU/9UJH3xjttUqaBwYPaXQ8AQr1BX/1\nH8f62mqX3/TNx+fcP82zmBqAdgDfW/d9+MsYcwlmkv6RNR7SiCilvKPWUIwlWdS+nmQJMA2sSrKy\nr38Ttebj0RCWsecX7enm/zSwrbW2fdRi2D99Huo3Sd3xWFFKedb6vRlgnV+acSQNmcFDGi1j1Bfq\nRJLDSdYk2ZDkJHC3lHKTesT2UpLtSXYDF4FbpZTJIcx/pDkNszbJWWAX9Rgv1GLTz8DFJJuT7AfO\nAeNzrbUopTwBLgPjSQ4lWZ1kd5JTTZ3HXM0AW5OsS7IsibvD0pAYPKQRUkp5Tv0McYt6euMhcJ1a\nBHqiue0A8A643Vx7Sj25MvB0v+g704w1BRwDjpZSHjdrm6WeklkKTFBPltygFp8O4jgwTn2+x8AV\nYCfwYoAxLlB3WR4Ab4C9A65B0m/k5xotSZofSb4DB0sp1/73WiT9P+54SJKkzhg8JHXF7VVJfmqR\nJEndccdDkiR1xuAhSZI6Y/CQJEmdMXhIkqTOGDwkSVJnDB6SJKkzBg9JktQZg4ckSeqMwUOSJHXm\nB8vcmgZOhzULAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x9f2a160>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "plt.plot(list(pca.explained_variance_ratio_),'-o')\n",
    "plt.title('Explained variance ratio as function of PCA components')\n",
    "plt.ylabel('Explained variance ratio')\n",
    "plt.xlabel('Component')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[ 4.97920203  1.99254767]\n",
      " [-1.05764711 -0.24918246]\n",
      " [-1.0436656  -0.27278704]\n",
      " [-0.54357051 -0.36484743]\n",
      " [-0.55093989 -0.38420189]\n",
      " [-0.55136204 -0.39983171]\n",
      " [ 4.89826696  1.9399872 ]\n",
      " [ 4.90657144  1.94518439]\n",
      " [ 4.9156538   1.95073956]\n",
      " [ 4.92688358  1.95743474]]\n"
     ]
    }
   ],
   "source": [
    "# First we reduce the data to two dimensions using PCA to capture variation\n",
    "pca = PCA(n_components=2)\n",
    "reduced_data = pca.fit_transform(X)\n",
    "print(reduced_data[:10])  # print upto 10 elements"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Import clustering modules\n",
    "from sklearn.cluster import KMeans\n",
    "from sklearn.mixture import GMM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "KMeans(algorithm='auto', copy_x=True, init='k-means++', max_iter=300,\n",
      "    n_clusters=5, n_init=10, n_jobs=1, precompute_distances='auto',\n",
      "    random_state=None, tol=0.0001, verbose=0)\n"
     ]
    }
   ],
   "source": [
    "kmeans = KMeans(n_clusters=5)\n",
    "clusters = kmeans.fit(reduced_data)\n",
    "print(clusters)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Plot the decision boundary by building a mesh grid to populate a graph.\n",
    "x_min, x_max = reduced_data[:, 0].min() - 1, reduced_data[:, 0].max() + 1\n",
    "y_min, y_max = reduced_data[:, 1].min() - 1, reduced_data[:, 1].max() + 1\n",
    "hx = (x_max-x_min)/1000.\n",
    "hy = (y_max-y_min)/1000.\n",
    "xx, yy = np.meshgrid(np.arange(x_min, x_max, hx), np.arange(y_min, y_max, hy))\n",
    "\n",
    "# Obtain labels for each point in mesh. Use last trained model.\n",
    "Z = clusters.predict(np.c_[xx.ravel(), yy.ravel()])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "*** K MEANS CENTROIDS ***\n",
      "[[-1.17899217  0.10923204]\n",
      " [ 0.27548729 -0.38245498]\n",
      " [ 3.756748    0.90595973]\n",
      " [ 1.39096176 -1.96519098]\n",
      " [ 0.75118636  1.29505563]]\n",
      "*** CENTROIDS TRANSFERED TO ORIGINAL SPACE ***\n",
      "[[  5.07538292e+00   4.41675804e+00   1.48300647e+00   4.89805784e+00\n",
      "    4.42388104e+00   1.47298994e+00   5.11735459e+00   1.65395594e+00\n",
      "    1.75891510e+00   5.11532176e+00   4.27820796e+00   1.74234987e+00\n",
      "    4.87798635e+00   4.39801313e+00   1.52204810e+00   5.11281319e+00\n",
      "    4.41659689e+00   1.50415344e+00  -5.40056281e-04  -4.95952441e-04\n",
      "    1.44278537e-05   9.41674920e-04  -1.35628216e-03  -5.30549161e-05\n",
      "   -3.36744289e-04  -4.49121329e-04   6.24981635e-06   7.16538288e-04\n",
      "   -1.19945479e-03  -7.60296260e-05   1.61708693e-05   8.25263536e-05\n",
      "   -6.08851488e-06  -5.56774559e-05   9.80624431e-05  -2.71805994e-05\n",
      "    3.50391876e-05   8.09840737e-05  -3.19512796e-06  -4.32899449e-05\n",
      "    9.60457482e-05  -2.31884663e-05   6.89822610e-03   8.87193291e-03\n",
      "    5.54773746e-03   6.76588644e-03   1.65099577e-03   2.09064605e-03\n",
      "    1.34773538e-03   1.49376116e-03]\n",
      " [  4.78102189e+00   4.16053187e+00   1.49441066e+00   5.50074921e+00\n",
      "    3.52701459e+00   1.48623279e+00   5.05655324e+00   1.65862936e+00\n",
      "    1.75389730e+00   5.09000923e+00   4.26055097e+00   1.74287503e+00\n",
      "    4.63071699e+00   4.14504217e+00   1.52282229e+00   5.61406502e+00\n",
      "    3.60672090e+00   1.50528539e+00  -2.29132707e-04  -4.11373351e-04\n",
      "   -4.03684693e-05  -3.41441798e-04   1.24565835e-03   1.39963561e-04\n",
      "   -1.18818735e-04  -4.04977103e-04  -4.94248913e-05  -1.93037411e-04\n",
      "    1.09204403e-03   1.50114997e-04   1.42745014e-05   6.71861930e-05\n",
      "    1.92156325e-06  -3.47303087e-05  -8.60677459e-05   2.48593520e-05\n",
      "    1.83542030e-05   5.15247467e-05   2.41614191e-06  -2.47780547e-05\n",
      "   -1.04840613e-04   2.52700796e-05   9.59185803e-03   1.32621783e-02\n",
      "    8.28195621e-03   1.08917138e-02   2.10923930e-03   2.81244997e-03\n",
      "    1.79568725e-03   2.14718957e-03]\n",
      " [  3.18533235e+00   2.70068060e+00   1.51946680e+00   7.06738949e+00\n",
      "    2.76322387e+00   1.51883638e+00   5.04868835e+00   1.69280837e+00\n",
      "    1.73204626e+00   5.06855381e+00   4.22837301e+00   1.74226038e+00\n",
      "    3.24120591e+00   2.83158943e+00   1.51399715e+00   6.97957185e+00\n",
      "    2.87339666e+00   1.52325939e+00   3.33804012e-03   3.10702353e-03\n",
      "    1.62356900e-04  -3.39428443e-03   9.19754390e-04   2.69057051e-05\n",
      "    2.53180975e-03   2.46599099e-03   2.22584285e-04  -2.57895704e-03\n",
      "    5.55667517e-04   7.48502470e-05  -8.49234522e-05  -2.37471298e-04\n",
      "    4.58835118e-05   2.01223183e-04  -1.01230998e-04   2.06735427e-05\n",
      "   -8.76506093e-05  -2.58024720e-04   3.60118245e-05   1.86209802e-04\n",
      "   -1.85442476e-04   2.50613353e-05   2.21797955e-02   2.05765979e-02\n",
      "    2.02969083e-02   1.95150932e-02   3.57096803e-03   3.28914110e-03\n",
      "    3.10904773e-03   3.04428995e-03]\n",
      " [  4.99108929e+00   4.37805174e+00   1.50425206e+00   5.90226891e+00\n",
      "    2.16290098e+00   1.49594540e+00   4.94259891e+00   1.65096849e+00\n",
      "    1.75486187e+00   5.05145990e+00   4.24207796e+00   1.74419313e+00\n",
      "    4.83119005e+00   4.29727175e+00   1.52863826e+00   5.91741446e+00\n",
      "    2.39624872e+00   1.49868821e+00  -1.37127778e-03  -1.96819956e-03\n",
      "   -2.45678823e-04  -1.33442705e-03   6.44622192e-03   5.69222502e-04\n",
      "   -9.92902408e-04  -1.72351521e-03  -2.90320797e-04  -7.88461817e-04\n",
      "    5.79406958e-03   6.25071836e-04   5.91137710e-05   1.86459898e-04\n",
      "   -4.05904685e-06  -1.09540502e-04  -4.35398517e-04   1.27732222e-04\n",
      "    3.78699319e-05   1.45835463e-04  -3.14239223e-06  -9.20968717e-05\n",
      "   -4.54632752e-04   1.19259079e-04   8.65447031e-03   1.81909751e-02\n",
      "    7.70342653e-03   1.46680533e-02   2.28220353e-03   3.97779342e-03\n",
      "    2.02127222e-03   2.97445365e-03]\n",
      " [  4.02023025e+00   3.44544311e+00   1.49647050e+00   5.79041452e+00\n",
      "    4.26485748e+00   1.49124042e+00   5.13932083e+00   1.67730375e+00\n",
      "    1.74491779e+00   5.11090912e+00   4.26229537e+00   1.74165114e+00\n",
      "    3.95502367e+00   3.53437635e+00   1.51511289e+00   5.90161893e+00\n",
      "    4.24047319e+00   1.51703837e+00   1.97763401e-03   2.08897226e-03\n",
      "    1.90680800e-04  -7.47474831e-04  -2.79031040e-03  -2.25712922e-04\n",
      "    1.54005503e-03   1.67152957e-03   2.34569198e-04  -6.46275984e-04\n",
      "   -2.64833004e-03  -2.35668684e-04  -5.69321928e-05  -1.37632478e-04\n",
      "    2.30270405e-05   1.10682843e-04   1.71038523e-04  -5.41225320e-05\n",
      "   -3.63703968e-05  -1.36359800e-04   1.92884112e-05   1.05568392e-04\n",
      "    1.27894712e-04  -4.55253926e-05   1.50519728e-02   1.23166580e-02\n",
      "    1.32556356e-02   1.13077260e-02   2.53123989e-03   2.11571438e-03\n",
      "    2.12205338e-03   1.86362322e-03]]\n"
     ]
    }
   ],
   "source": [
    "# Find the centroids for KMeans or the cluster means for GMM \n",
    "\n",
    "centroids = kmeans.cluster_centers_\n",
    "print('*** K MEANS CENTROIDS ***')\n",
    "print(centroids)\n",
    "\n",
    "# TRANSFORM DATA BACK TO ORIGINAL SPACE FOR ANSWERING 7\n",
    "print('*** CENTROIDS TRANSFERED TO ORIGINAL SPACE ***')\n",
    "print(pca.inverse_transform(centroids))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeQAAAFzCAYAAADvzXoCAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nEB1AQSaEEEJ0wP8DiLfbYulNl7IAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xbdb1f98>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Put the result into a color plot\n",
    "Z = Z.reshape(xx.shape)\n",
    "plt.figure(1)\n",
    "plt.clf()\n",
    "plt.imshow(Z, interpolation='nearest',\n",
    "           extent=(xx.min(), xx.max(), yy.min(), yy.max()),\n",
    "           cmap=plt.cm.Paired,\n",
    "           aspect='auto', origin='lower')\n",
    "\n",
    "plt.plot(reduced_data[:, 0], reduced_data[:, 1], 'k.', markersize=2)\n",
    "plt.scatter(centroids[:, 0], centroids[:, 1],\n",
    "            marker='x', s=169, linewidths=3,\n",
    "            color='w', zorder=10)\n",
    "plt.title('Clustering on the seeds dataset (PCA-reduced data)\\n'\n",
    "          'Centroids are marked with white cross')\n",
    "plt.xlim(x_min, x_max)\n",
    "plt.ylim(y_min, y_max)\n",
    "plt.xticks(())\n",
    "plt.yticks(())\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Applying agglomerative clustering via scikit-learn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cluster labels: [3 2 2 ..., 2 3 0]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.cluster import AgglomerativeClustering\n",
    "\n",
    "ac = AgglomerativeClustering(n_clusters=5, affinity='euclidean', linkage='complete')\n",
    "labels = ac.fit_predict(X)\n",
    "print('Cluster labels: %s' % labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "___"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X = df_1[features]\n",
    "y = df_1['phase']\n",
    "# split dataset to 60% training and 40% testing\n",
    "X_train, X_test, y_train, y_test = cross_validation.train_test_split(X, y, test_size=0.4, random_state=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### K Means"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "ename": "SyntaxError",
     "evalue": "Missing parentheses in call to 'print' (<ipython-input-70-ad1ef4095e7d>, line 4)",
     "output_type": "error",
     "traceback": [
      "\u001b[0;36m  File \u001b[0;32m\"<ipython-input-70-ad1ef4095e7d>\"\u001b[0;36m, line \u001b[0;32m4\u001b[0m\n\u001b[0;31m    print clf.labels_.shape\u001b[0m\n\u001b[0m            ^\u001b[0m\n\u001b[0;31mSyntaxError\u001b[0m\u001b[0;31m:\u001b[0m Missing parentheses in call to 'print'\n"
     ]
    }
   ],
   "source": [
    "from sklearn import cluster\n",
    "clf = cluster.KMeans(init='k-means++', n_clusters=5, random_state=5)\n",
    "clf.fit(X_train)\n",
    "print clf.labels_.shape\n",
    "print clf.labels_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Predict clusters on testing data\n",
    "y_pred = clf.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn import metrics\n",
    "print \"Addjusted rand score:{:.2}\".format(metrics.adjusted_rand_score(y_test, y_pred))\n",
    "print \"Homogeneity score:{:.2} \".format(metrics.homogeneity_score(y_test, y_pred)) \n",
    "print \"Completeness score: {:.2} \".format(metrics.completeness_score(y_test, y_pred))\n",
    "print \"Confusion matrix\"\n",
    "print metrics.confusion_matrix(y_test, y_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Affinity Propogation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Affinity propagation\n",
    "aff = cluster.AffinityPropagation()\n",
    "aff.fit(X_train)\n",
    "print aff.cluster_centers_indices_.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "y_pred = aff.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn import metrics\n",
    "print \"Addjusted rand score:{:.2}\".format(metrics.adjusted_rand_score(y_test, y_pred))\n",
    "print \"Homogeneity score:{:.2} \".format(metrics.homogeneity_score(y_test, y_pred)) \n",
    "print \"Completeness score: {:.2} \".format(metrics.completeness_score(y_test, y_pred))\n",
    "print \"Confusion matrix\"\n",
    "print metrics.confusion_matrix(y_test, y_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## MeanShift"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "ms = cluster.MeanShift()\n",
    "ms.fit(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "y_pred = ms.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn import metrics\n",
    "print \"Addjusted rand score:{:.2}\".format(metrics.adjusted_rand_score(y_test, y_pred))\n",
    "print \"Homogeneity score:{:.2} \".format(metrics.homogeneity_score(y_test, y_pred)) \n",
    "print \"Completeness score: {:.2} \".format(metrics.completeness_score(y_test, y_pred))\n",
    "print \"Confusion matrix\"\n",
    "print metrics.confusion_matrix(y_test, y_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Mixture of Guassian Models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn import mixture\n",
    "\n",
    "# Define a heldout dataset to estimate covariance type\n",
    "X_train_heldout, X_test_heldout, y_train_heldout, y_test_heldout = train_test_split(\n",
    "        X_train, y_train,test_size=0.25, random_state=42)\n",
    "for covariance_type in ['spherical','tied','diag','full']:\n",
    "    gm=mixture.GMM(n_components=100, covariance_type=covariance_type, random_state=42, n_init=5)\n",
    "    gm.fit(X_train_heldout)\n",
    "    y_pred=gm.predict(X_test_heldout)\n",
    "    print \"Adjusted rand score for covariance={}:{:.2}\".format(covariance_type, \n",
    "                                                               metrics.adjusted_rand_score(y_test_heldout, y_pred))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "___"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "pca = PCA(n_components=2)\n",
    "X = pca.fit_transform(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "c = []\n",
    "from matplotlib.pyplot import cm \n",
    "n=6\n",
    "color=iter(cm.rainbow(np.linspace(0,1,n)))\n",
    "for i in range(n):\n",
    "    c.append(next(color))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAxYAAAJOCAYAAAAqFJGJAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nEBERERGRoKmwEBERERGRoP1/RXsIviFOmM0AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xe5bcd68>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "n = 5\n",
    "f, (ax1, ax2) = plt.subplots(1, 2, figsize=(8,6))\n",
    "\n",
    "km = KMeans(n_clusters= n , random_state=0)\n",
    "y_km = km.fit_predict(X)\n",
    "\n",
    "for i in range(n):\n",
    "    ax1.scatter(X[y_km==i,0], X[y_km==i,1], c=c[i], marker='o', s=40, label='cluster{}'.format(i))\n",
    "ax1.set_title('K-means clustering')\n",
    "\n",
    "ac = AgglomerativeClustering(n_clusters=n, affinity='euclidean', linkage='complete')\n",
    "y_ac = ac.fit_predict(X)\n",
    "for i in range(n):\n",
    "    ax2.scatter(X[y_ac==i,0], X[y_ac==i,1], c=c[i], marker='o', s=40, label='cluster{}'.format(i))\n",
    "ax2.set_title('Agglomerative clustering')\n",
    "\n",
    "# Put a legend below current axis\n",
    "plt.legend()\n",
    "    \n",
    "plt.tight_layout()\n",
    "#plt.savefig('./figures/kmeans_and_ac.png', dpi=300)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "____"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Classification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import os\n",
    "from sklearn.tree import DecisionTreeClassifier, export_graphviz\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.cross_validation import train_test_split\n",
    "from sklearn import cross_validation, metrics\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.naive_bayes import BernoulliNB\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.svm import SVC\n",
    "from time import time\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.metrics import roc_auc_score , classification_report\n",
    "from sklearn.grid_search import GridSearchCV\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.metrics import precision_score, recall_score, accuracy_score, classification_report\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X = df_1[features]\n",
    "y = df_1['phase']\n",
    "# split dataset to 60% training and 40% testing\n",
    "X_train, X_test, y_train, y_test = cross_validation.train_test_split(X, y, test_size=0.4, random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1045, 50) (1045,) (698, 50) (698,)\n"
     ]
    }
   ],
   "source": [
    "print (X_train.shape, y_train.shape,X_test.shape, y_test.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Decision Tree accuracy and time elapsed caculation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DecisionTree\n",
      "Acurracy:  0.823782234957\n",
      "time elapsed:  0.15800905227661133\n"
     ]
    }
   ],
   "source": [
    "t0=time()\n",
    "print (\"DecisionTree\")\n",
    "\n",
    "dt = DecisionTreeClassifier(min_samples_split=20,random_state=99)\n",
    "# dt = DecisionTreeClassifier(min_samples_split=20,max_depth=5,random_state=99)\n",
    "\n",
    "clf_dt=dt.fit(X_train,y_train)\n",
    "\n",
    "print (\"Acurracy: \", clf_dt.score(X_test,y_test))\n",
    "t1=time()\n",
    "print (\"time elapsed: \", t1-t0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cross result========\n",
      "[ 0.49287749  0.7277937   0.64756447  0.68678161  0.5433526 ]\n",
      "0.619673973884\n",
      "time elapsed:  1.3700783252716064\n"
     ]
    }
   ],
   "source": [
    "tt0=time()\n",
    "print (\"cross result========\")\n",
    "scores = cross_validation.cross_val_score(dt, X,y, cv=5)\n",
    "print (scores)\n",
    "print (scores.mean())\n",
    "tt1=time()\n",
    "print (\"time elapsed: \", tt1-tt0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Random Forest accuracy and time elapsed caculation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RandomForest\n",
      "Acurracy:  0.893982808023\n",
      "time elapsed:  0.8110466003417969\n"
     ]
    }
   ],
   "source": [
    "t2=time()\n",
    "print (\"RandomForest\")\n",
    "rf = RandomForestClassifier(n_estimators=100,n_jobs=-1)\n",
    "clf_rf = rf.fit(X_train,y_train)\n",
    "print (\"Acurracy: \", clf_rf.score(X_test,y_test))\n",
    "t3=time()\n",
    "print (\"time elapsed: \", t3-t2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cross result========\n",
      "[ 0.74928775  0.79083095  0.79369628  0.79022989  0.70520231]\n",
      "0.765849433423\n",
      "time elapsed:  4.255243301391602\n"
     ]
    }
   ],
   "source": [
    "tt2=time()\n",
    "print (\"cross result========\")\n",
    "scores = cross_validation.cross_val_score(rf, X,y, cv=5)\n",
    "print (scores)\n",
    "print (scores.mean())\n",
    "tt3=time()\n",
    "print (\"time elapsed: \", tt3-tt2)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Naive Bayes accuracy and time elapsed caculation\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NaiveBayes\n",
      "Acurracy:  0.385386819484\n",
      "time elapsed:  0.01000070571899414\n"
     ]
    }
   ],
   "source": [
    "t4=time()\n",
    "print (\"NaiveBayes\")\n",
    "nb = BernoulliNB()\n",
    "clf_nb=nb.fit(X_train,y_train)\n",
    "print (\"Acurracy: \", clf_nb.score(X_test,y_test))\n",
    "t5=time()\n",
    "print (\"time elapsed: \", t5-t4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cross result========\n",
      "[ 0.39031339  0.37249284  0.40974212  0.37356322  0.43641618]\n",
      "0.396505550139\n",
      "time elapsed:  0.19201111793518066\n"
     ]
    }
   ],
   "source": [
    "tt4=time()\n",
    "print (\"cross result========\")\n",
    "scores = cross_validation.cross_val_score(nb, X,y, cv=5)\n",
    "print (scores)\n",
    "print (scores.mean())\n",
    "tt5=time()\n",
    "print (\"time elapsed: \", tt5-tt4)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# KNN accuracy and time elapsed caculation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "KNN\n",
      "Acurracy:  0.843839541547\n",
      "time elapsed:  0.059003353118896484\n"
     ]
    }
   ],
   "source": [
    "t6=time()\n",
    "print (\"KNN\")\n",
    "# knn = KNeighborsClassifier(n_neighbors=3)\n",
    "knn = KNeighborsClassifier()\n",
    "clf_knn=knn.fit(X_train, y_train)\n",
    "print (\"Acurracy: \", clf_knn.score(X_test,y_test) )\n",
    "t7=time()\n",
    "print (\"time elapsed: \", t7-t6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cross result========\n",
      "[ 0.56410256  0.63896848  0.70487106  0.73563218  0.54913295]\n",
      "0.638541447507\n",
      "time elapsed:  0.23401331901550293\n"
     ]
    }
   ],
   "source": [
    "tt6=time()\n",
    "print (\"cross result========\")\n",
    "scores = cross_validation.cross_val_score(knn, X,y, cv=5)\n",
    "print (scores)\n",
    "print (scores.mean())\n",
    "tt7=time()\n",
    "print (\"time elapsed: \", tt7-tt6)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# SVM accuracy and time elapsed caculation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SVM\n",
      "Acurracy:  0.753581661891\n",
      "time elapsed:  0.2240126132965088\n"
     ]
    }
   ],
   "source": [
    "t7=time()\n",
    "print (\"SVM\")\n",
    "svc = SVC()\n",
    "clf_svc=svc.fit(X_train, y_train)\n",
    "print (\"Acurracy: \", clf_svc.score(X_test,y_test) )\n",
    "t8=time()\n",
    "print (\"time elapsed: \", t8-t7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cross result========\n",
      "[ 0.73504274  0.7277937   0.77363897  0.75287356  0.6300578 ]\n",
      "0.723881353297\n",
      "time elapsed:  33.91093945503235\n"
     ]
    }
   ],
   "source": [
    "tt7=time()\n",
    "print (\"cross result========\")\n",
    "scores = cross_validation.cross_val_score(svc, X,y, cv=5)\n",
    "print (scores)\n",
    "print (scores.mean())\n",
    "tt8=time()\n",
    "print (\"time elapsed: \", tt7-tt6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 3 folds for each of 4 candidates, totalling 12 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done  10 out of  12 | elapsed:    9.7s remaining:    1.9s\n",
      "[Parallel(n_jobs=-1)]: Done  12 out of  12 | elapsed:    9.9s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best score: 0.779\n",
      "Best parameters set:\n",
      "\tC: 10\n",
      "\tkernel: 'linear'\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "          0       1.00      0.22      0.36        18\n",
      "          1       0.32      0.19      0.24        57\n",
      "          2       0.84      0.98      0.91       280\n",
      "          3       0.76      0.27      0.40        82\n",
      "          4       0.79      0.92      0.85       261\n",
      "\n",
      "avg / total       0.77      0.79      0.76       698\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.svm import SVC\n",
    "from sklearn.cross_validation import cross_val_score\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn import grid_search\n",
    "\n",
    "svc = SVC()\n",
    "\n",
    "parameters = {'kernel':('linear', 'rbf'), 'C':[1, 10]}\n",
    "\n",
    "grid = grid_search.GridSearchCV(svc, parameters, n_jobs=-1, verbose=1, scoring='accuracy')\n",
    "\n",
    "\n",
    "grid.fit(X_train, y_train)\n",
    "\n",
    "print ('Best score: %0.3f' % grid.best_score_)\n",
    "\n",
    "print ('Best parameters set:')\n",
    "best_parameters = grid.best_estimator_.get_params()\n",
    "\n",
    "for param_name in sorted(parameters.keys()):\n",
    "    print ('\\t%s: %r' % (param_name, best_parameters[param_name]))\n",
    "    \n",
    "predictions = grid.predict(X_test)\n",
    "print (classification_report(y_test, predictions))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 3 folds for each of 30 candidates, totalling 90 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done  34 tasks      | elapsed:   10.6s\n",
      "[Parallel(n_jobs=-1)]: Done  90 out of  90 | elapsed:   11.8s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best score: 0.879\n",
      "Best parameters set:\n",
      "\tclf__C: 30\n",
      "\tclf__gamma: 1\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "          0       0.89      0.89      0.89        18\n",
      "          1       0.55      0.63      0.59        57\n",
      "          2       0.94      0.96      0.95       280\n",
      "          3       0.85      0.77      0.81        82\n",
      "          4       0.93      0.91      0.92       261\n",
      "\n",
      "avg / total       0.89      0.89      0.89       698\n",
      "\n"
     ]
    }
   ],
   "source": [
    "pipeline = Pipeline([\n",
    "    ('clf', SVC(kernel='rbf', gamma=0.01, C=100))\n",
    "])\n",
    "\n",
    "parameters = {\n",
    "    'clf__gamma': (0.01, 0.03, 0.1, 0.3, 1),\n",
    "    'clf__C': (0.1, 0.3, 1, 3, 10, 30),\n",
    "}\n",
    "\n",
    "grid_search = GridSearchCV(pipeline, parameters, n_jobs=-1, verbose=1, scoring='accuracy')\n",
    "\n",
    "grid_search.fit(X_train, y_train)\n",
    "\n",
    "print ('Best score: %0.3f' % grid_search.best_score_)\n",
    "\n",
    "print ('Best parameters set:')\n",
    "best_parameters = grid_search.best_estimator_.get_params()\n",
    "\n",
    "for param_name in sorted(parameters.keys()):\n",
    "    print ('\\t%s: %r' % (param_name, best_parameters[param_name]))\n",
    "    \n",
    "predictions = grid_search.predict(X_test)\n",
    "print (classification_report(y_test, predictions))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Ensemble Learning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.base import BaseEstimator\n",
    "from sklearn.base import ClassifierMixin\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.externals import six\n",
    "from sklearn.base import clone\n",
    "from sklearn.pipeline import _name_estimators\n",
    "import numpy as np\n",
    "import operator\n",
    "\n",
    "\n",
    "class MajorityVoteClassifier(BaseEstimator, \n",
    "                             ClassifierMixin):\n",
    "    \"\"\" A majority vote ensemble classifier\n",
    "\n",
    "    Parameters\n",
    "    ----------\n",
    "    classifiers : array-like, shape = [n_classifiers]\n",
    "      Different classifiers for the ensemble\n",
    "\n",
    "    vote : str, {'classlabel', 'probability'} (default='label')\n",
    "      If 'classlabel' the prediction is based on the argmax of\n",
    "        class labels. Else if 'probability', the argmax of\n",
    "        the sum of probabilities is used to predict the class label\n",
    "        (recommended for calibrated classifiers).\n",
    "\n",
    "    weights : array-like, shape = [n_classifiers], optional (default=None)\n",
    "      If a list of `int` or `float` values are provided, the classifiers\n",
    "      are weighted by importance; Uses uniform weights if `weights=None`.\n",
    "\n",
    "    \"\"\"\n",
    "    def __init__(self, classifiers, vote='classlabel', weights=None):\n",
    "\n",
    "        self.classifiers = classifiers\n",
    "        self.named_classifiers = {key: value for key, value\n",
    "                                  in _name_estimators(classifiers)}\n",
    "        self.vote = vote\n",
    "        self.weights = weights\n",
    "\n",
    "    def fit(self, X, y):\n",
    "        \"\"\" Fit classifiers.\n",
    "\n",
    "        Parameters\n",
    "        ----------\n",
    "        X : {array-like, sparse matrix}, shape = [n_samples, n_features]\n",
    "            Matrix of training samples.\n",
    "\n",
    "        y : array-like, shape = [n_samples]\n",
    "            Vector of target class labels.\n",
    "\n",
    "        Returns\n",
    "        -------\n",
    "        self : object\n",
    "\n",
    "        \"\"\"\n",
    "        if self.vote not in ('probability', 'classlabel'):\n",
    "            raise ValueError(\"vote must be 'probability' or 'classlabel'\"\n",
    "                             \"; got (vote=%r)\"\n",
    "                             % self.vote)\n",
    "\n",
    "        if self.weights and len(self.weights) != len(self.classifiers):\n",
    "            raise ValueError('Number of classifiers and weights must be equal'\n",
    "                             '; got %d weights, %d classifiers'\n",
    "                             % (len(self.weights), len(self.classifiers)))\n",
    "\n",
    "        # Use LabelEncoder to ensure class labels start with 0, which\n",
    "        # is important for np.argmax call in self.predict\n",
    "        self.lablenc_ = LabelEncoder()\n",
    "        self.lablenc_.fit(y)\n",
    "        self.classes_ = self.lablenc_.classes_\n",
    "        self.classifiers_ = []\n",
    "        for clf in self.classifiers:\n",
    "            fitted_clf = clone(clf).fit(X, self.lablenc_.transform(y))\n",
    "            self.classifiers_.append(fitted_clf)\n",
    "        return self\n",
    "\n",
    "    def predict(self, X):\n",
    "        \"\"\" Predict class labels for X.\n",
    "\n",
    "        Parameters\n",
    "        ----------\n",
    "        X : {array-like, sparse matrix}, shape = [n_samples, n_features]\n",
    "            Matrix of training samples.\n",
    "\n",
    "        Returns\n",
    "        ----------\n",
    "        maj_vote : array-like, shape = [n_samples]\n",
    "            Predicted class labels.\n",
    "            \n",
    "        \"\"\"\n",
    "        if self.vote == 'probability':\n",
    "            maj_vote = np.argmax(self.predict_proba(X), axis=1)\n",
    "        else:  # 'classlabel' vote\n",
    "\n",
    "            #  Collect results from clf.predict calls\n",
    "            predictions = np.asarray([clf.predict(X)\n",
    "                                      for clf in self.classifiers_]).T\n",
    "\n",
    "            maj_vote = np.apply_along_axis(\n",
    "                                      lambda x:\n",
    "                                      np.argmax(np.bincount(x,\n",
    "                                                weights=self.weights)),\n",
    "                                      axis=1,\n",
    "                                      arr=predictions)\n",
    "        maj_vote = self.lablenc_.inverse_transform(maj_vote)\n",
    "        return maj_vote\n",
    "\n",
    "    def predict_proba(self, X):\n",
    "        \"\"\" Predict class probabilities for X.\n",
    "\n",
    "        Parameters\n",
    "        ----------\n",
    "        X : {array-like, sparse matrix}, shape = [n_samples, n_features]\n",
    "            Training vectors, where n_samples is the number of samples and\n",
    "            n_features is the number of features.\n",
    "\n",
    "        Returns\n",
    "        ----------\n",
    "        avg_proba : array-like, shape = [n_samples, n_classes]\n",
    "            Weighted average probability for each class per sample.\n",
    "\n",
    "        \"\"\"\n",
    "        probas = np.asarray([clf.predict_proba(X)\n",
    "                             for clf in self.classifiers_])\n",
    "        avg_proba = np.average(probas, axis=0, weights=self.weights)\n",
    "        return avg_proba\n",
    "\n",
    "    def get_params(self, deep=True):\n",
    "        \"\"\" Get classifier parameter names for GridSearch\"\"\"\n",
    "        if not deep:\n",
    "            return super(MajorityVoteClassifier, self).get_params(deep=False)\n",
    "        else:\n",
    "            out = self.named_classifiers.copy()\n",
    "            for name, step in six.iteritems(self.named_classifiers):\n",
    "                for key, value in six.iteritems(step.get_params(deep=True)):\n",
    "                    out['%s__%s' % (name, key)] = value\n",
    "            return out"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10-fold cross validation:\n",
      "\n",
      "Accuracy: 0.74 (+/- 0.02) [Logistic Regression]\n",
      "Accuracy: 0.66 (+/- 0.02) [Decision Tree]\n",
      "Accuracy: 0.86 (+/- 0.03) [KNN]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.cross_validation import cross_val_score\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.neighbors import KNeighborsClassifier \n",
    "from sklearn.pipeline import Pipeline\n",
    "import numpy as np\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "clf1 = LogisticRegression(penalty='l2', \n",
    "                          C=0.001, \n",
    "                          random_state=0)\n",
    "\n",
    "clf2 = DecisionTreeClassifier(max_depth=1, \n",
    "                              criterion='entropy', \n",
    "                              random_state=0)\n",
    "\n",
    "clf3 = KNeighborsClassifier(n_neighbors=1, \n",
    "                            p=2, \n",
    "                            metric='minkowski')\n",
    "\n",
    "pipe1 = Pipeline([['sc', StandardScaler()],\n",
    "                  ['clf', clf1]])\n",
    "pipe3 = Pipeline([['sc', StandardScaler()],\n",
    "                  ['clf', clf3]])\n",
    "\n",
    "clf_labels = ['Logistic Regression', 'Decision Tree', 'KNN']\n",
    "\n",
    "print('10-fold cross validation:\\n')\n",
    "for clf, label in zip([pipe1, clf2, pipe3], clf_labels):\n",
    "    scores = cross_val_score(estimator=clf, \n",
    "                             X=X_train, \n",
    "                             y=y_train, \n",
    "                             cv=10, \n",
    "                             scoring='accuracy')\n",
    "    print(\"Accuracy: %0.2f (+/- %0.2f) [%s]\" \n",
    "               % (scores.mean(), scores.std(), label))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.74 (+/- 0.02) [Logistic Regression]\n",
      "Accuracy: 0.66 (+/- 0.02) [Decision Tree]\n",
      "Accuracy: 0.86 (+/- 0.03) [KNN]\n",
      "Accuracy: 0.76 (+/- 0.02) [Majority Voting]\n"
     ]
    }
   ],
   "source": [
    "# Majority Rule (hard) Voting\n",
    "\n",
    "mv_clf = MajorityVoteClassifier(\n",
    "                classifiers=[pipe1, clf2, pipe3])\n",
    "\n",
    "clf_labels += ['Majority Voting']\n",
    "all_clf = [pipe1, clf2, pipe3, mv_clf]\n",
    "\n",
    "for clf, label in zip(all_clf, clf_labels):\n",
    "    scores = cross_val_score(estimator=clf, \n",
    "                             X=X_train, \n",
    "                             y=y_train, \n",
    "                             cv=10, \n",
    "                             scoring='accuracy')\n",
    "    print(\"Accuracy: %0.2f (+/- %0.2f) [%s]\" \n",
    "               % (scores.mean(), scores.std(), label))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'decisiontreeclassifier': DecisionTreeClassifier(class_weight=None, criterion='entropy', max_depth=1,\n",
       "             max_features=None, max_leaf_nodes=None,\n",
       "             min_impurity_split=1e-07, min_samples_leaf=1,\n",
       "             min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
       "             presort=False, random_state=0, splitter='best'),\n",
       " 'decisiontreeclassifier__class_weight': None,\n",
       " 'decisiontreeclassifier__criterion': 'entropy',\n",
       " 'decisiontreeclassifier__max_depth': 1,\n",
       " 'decisiontreeclassifier__max_features': None,\n",
       " 'decisiontreeclassifier__max_leaf_nodes': None,\n",
       " 'decisiontreeclassifier__min_impurity_split': 1e-07,\n",
       " 'decisiontreeclassifier__min_samples_leaf': 1,\n",
       " 'decisiontreeclassifier__min_samples_split': 2,\n",
       " 'decisiontreeclassifier__min_weight_fraction_leaf': 0.0,\n",
       " 'decisiontreeclassifier__presort': False,\n",
       " 'decisiontreeclassifier__random_state': 0,\n",
       " 'decisiontreeclassifier__splitter': 'best',\n",
       " 'pipeline-1': Pipeline(steps=[['sc', StandardScaler(copy=True, with_mean=True, with_std=True)], ['clf', LogisticRegression(C=0.001, class_weight=None, dual=False, fit_intercept=True,\n",
       "           intercept_scaling=1, max_iter=100, multi_class='ovr', n_jobs=1,\n",
       "           penalty='l2', random_state=0, solver='liblinear', tol=0.0001,\n",
       "           verbose=0, warm_start=False)]]),\n",
       " 'pipeline-1__clf': LogisticRegression(C=0.001, class_weight=None, dual=False, fit_intercept=True,\n",
       "           intercept_scaling=1, max_iter=100, multi_class='ovr', n_jobs=1,\n",
       "           penalty='l2', random_state=0, solver='liblinear', tol=0.0001,\n",
       "           verbose=0, warm_start=False),\n",
       " 'pipeline-1__clf__C': 0.001,\n",
       " 'pipeline-1__clf__class_weight': None,\n",
       " 'pipeline-1__clf__dual': False,\n",
       " 'pipeline-1__clf__fit_intercept': True,\n",
       " 'pipeline-1__clf__intercept_scaling': 1,\n",
       " 'pipeline-1__clf__max_iter': 100,\n",
       " 'pipeline-1__clf__multi_class': 'ovr',\n",
       " 'pipeline-1__clf__n_jobs': 1,\n",
       " 'pipeline-1__clf__penalty': 'l2',\n",
       " 'pipeline-1__clf__random_state': 0,\n",
       " 'pipeline-1__clf__solver': 'liblinear',\n",
       " 'pipeline-1__clf__tol': 0.0001,\n",
       " 'pipeline-1__clf__verbose': 0,\n",
       " 'pipeline-1__clf__warm_start': False,\n",
       " 'pipeline-1__sc': StandardScaler(copy=True, with_mean=True, with_std=True),\n",
       " 'pipeline-1__sc__copy': True,\n",
       " 'pipeline-1__sc__with_mean': True,\n",
       " 'pipeline-1__sc__with_std': True,\n",
       " 'pipeline-1__steps': [['sc',\n",
       "   StandardScaler(copy=True, with_mean=True, with_std=True)],\n",
       "  ['clf',\n",
       "   LogisticRegression(C=0.001, class_weight=None, dual=False, fit_intercept=True,\n",
       "             intercept_scaling=1, max_iter=100, multi_class='ovr', n_jobs=1,\n",
       "             penalty='l2', random_state=0, solver='liblinear', tol=0.0001,\n",
       "             verbose=0, warm_start=False)]],\n",
       " 'pipeline-2': Pipeline(steps=[['sc', StandardScaler(copy=True, with_mean=True, with_std=True)], ['clf', KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski',\n",
       "            metric_params=None, n_jobs=1, n_neighbors=1, p=2,\n",
       "            weights='uniform')]]),\n",
       " 'pipeline-2__clf': KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski',\n",
       "            metric_params=None, n_jobs=1, n_neighbors=1, p=2,\n",
       "            weights='uniform'),\n",
       " 'pipeline-2__clf__algorithm': 'auto',\n",
       " 'pipeline-2__clf__leaf_size': 30,\n",
       " 'pipeline-2__clf__metric': 'minkowski',\n",
       " 'pipeline-2__clf__metric_params': None,\n",
       " 'pipeline-2__clf__n_jobs': 1,\n",
       " 'pipeline-2__clf__n_neighbors': 1,\n",
       " 'pipeline-2__clf__p': 2,\n",
       " 'pipeline-2__clf__weights': 'uniform',\n",
       " 'pipeline-2__sc': StandardScaler(copy=True, with_mean=True, with_std=True),\n",
       " 'pipeline-2__sc__copy': True,\n",
       " 'pipeline-2__sc__with_mean': True,\n",
       " 'pipeline-2__sc__with_std': True,\n",
       " 'pipeline-2__steps': [['sc',\n",
       "   StandardScaler(copy=True, with_mean=True, with_std=True)],\n",
       "  ['clf',\n",
       "   KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski',\n",
       "              metric_params=None, n_jobs=1, n_neighbors=1, p=2,\n",
       "              weights='uniform')]]}"
      ]
     },
     "execution_count": 100,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mv_clf.get_params()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.757+/-0.01 {'pipeline-1__clf__C': 0.001, 'decisiontreeclassifier__max_depth': 1}\n",
      "0.802+/-0.01 {'pipeline-1__clf__C': 0.1, 'decisiontreeclassifier__max_depth': 1}\n",
      "0.836+/-0.01 {'pipeline-1__clf__C': 100.0, 'decisiontreeclassifier__max_depth': 1}\n",
      "0.782+/-0.01 {'pipeline-1__clf__C': 0.001, 'decisiontreeclassifier__max_depth': 2}\n",
      "0.815+/-0.01 {'pipeline-1__clf__C': 0.1, 'decisiontreeclassifier__max_depth': 2}\n",
      "0.845+/-0.01 {'pipeline-1__clf__C': 100.0, 'decisiontreeclassifier__max_depth': 2}\n"
     ]
    }
   ],
   "source": [
    "from sklearn.grid_search import GridSearchCV\n",
    "\n",
    "params = {'decisiontreeclassifier__max_depth': [1, 2],\n",
    "          'pipeline-1__clf__C': [0.001, 0.1, 100.0]}\n",
    "\n",
    "grid = GridSearchCV(estimator=mv_clf, \n",
    "                    param_grid=params, \n",
    "                    cv=10, \n",
    "                    scoring='accuracy')\n",
    "grid.fit(X_train, y_train)\n",
    "\n",
    "for params, mean_score, scores in grid.grid_scores_:\n",
    "    print(\"%0.3f+/-%0.2f %r\"\n",
    "            % (mean_score, scores.std() / 2, params))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best parameters: {'pipeline-1__clf__C': 100.0, 'decisiontreeclassifier__max_depth': 2}\n",
      "Accuracy: 0.84\n"
     ]
    }
   ],
   "source": [
    "print('Best parameters: %s' % grid.best_params_)\n",
    "print('Accuracy: %.2f' % grid.best_score_)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Bagging -- Building an ensemble of classifiers from bootstrap samples"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Decision tree train/test accuracies 1.000/0.832\n",
      "Bagging train/test accuracies 1.000/0.903\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import BaggingClassifier\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "\n",
    "tree = DecisionTreeClassifier(criterion='entropy', \n",
    "                              max_depth=None)\n",
    "\n",
    "bag = BaggingClassifier(base_estimator=tree,\n",
    "                        n_estimators=500, \n",
    "                        max_samples=1.0, \n",
    "                        max_features=1.0, \n",
    "                        bootstrap=True, \n",
    "                        bootstrap_features=False, \n",
    "                        n_jobs=1, \n",
    "                        random_state=1)\n",
    "\n",
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "tree = tree.fit(X_train, y_train)\n",
    "y_train_pred = tree.predict(X_train)\n",
    "y_test_pred = tree.predict(X_test)\n",
    "\n",
    "tree_train = accuracy_score(y_train, y_train_pred)\n",
    "tree_test = accuracy_score(y_test, y_test_pred)\n",
    "print('Decision tree train/test accuracies %.3f/%.3f'\n",
    "      % (tree_train, tree_test))\n",
    "\n",
    "bag = bag.fit(X_train, y_train)\n",
    "y_train_pred = bag.predict(X_train)\n",
    "y_test_pred = bag.predict(X_test)\n",
    "\n",
    "bag_train = accuracy_score(y_train, y_train_pred) \n",
    "bag_test = accuracy_score(y_test, y_test_pred) \n",
    "print('Bagging train/test accuracies %.3f/%.3f'\n",
    "      % (bag_train, bag_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Leveraging weak learners via adaptive boosting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Decision tree train/test accuracies 0.665/0.672\n",
      "AdaBoost train/test accuracies 0.772/0.734\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import AdaBoostClassifier\n",
    "\n",
    "tree = DecisionTreeClassifier(criterion='entropy', \n",
    "                              max_depth=1)\n",
    "\n",
    "ada = AdaBoostClassifier(base_estimator=tree,\n",
    "                         n_estimators=500, \n",
    "                         learning_rate=0.1,\n",
    "                         random_state=0)\n",
    "\n",
    "tree = tree.fit(X_train, y_train)\n",
    "y_train_pred = tree.predict(X_train)\n",
    "y_test_pred = tree.predict(X_test)\n",
    "\n",
    "tree_train = accuracy_score(y_train, y_train_pred)\n",
    "tree_test = accuracy_score(y_test, y_test_pred)\n",
    "print('Decision tree train/test accuracies %.3f/%.3f'\n",
    "      % (tree_train, tree_test))\n",
    "\n",
    "ada = ada.fit(X_train, y_train)\n",
    "y_train_pred = ada.predict(X_train)\n",
    "y_test_pred = ada.predict(X_test)\n",
    "\n",
    "ada_train = accuracy_score(y_train, y_train_pred) \n",
    "ada_test = accuracy_score(y_test, y_test_pred) \n",
    "print('AdaBoost train/test accuracies %.3f/%.3f'\n",
    "      % (ada_train, ada_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "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
}