Human Activity Recognition Part 2 - ML Models.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "IMPORT TRAINING AND TESTING SETS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "train = pd.read_csv('csv_files/train.csv')\n",
    "test = pd.read_csv('csv_files/test.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((7352, 564), (2947, 564))"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.shape, test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tBodyAccmeanX</th>\n",
       "      <th>tBodyAccmeanY</th>\n",
       "      <th>tBodyAccmeanZ</th>\n",
       "      <th>tBodyAccstdX</th>\n",
       "      <th>tBodyAccstdY</th>\n",
       "      <th>tBodyAccstdZ</th>\n",
       "      <th>tBodyAccmadX</th>\n",
       "      <th>tBodyAccmadY</th>\n",
       "      <th>tBodyAccmadZ</th>\n",
       "      <th>tBodyAccmaxX</th>\n",
       "      <th>...</th>\n",
       "      <th>angletBodyAccMeangravity</th>\n",
       "      <th>angletBodyAccJerkMeangravityMean</th>\n",
       "      <th>angletBodyGyroMeangravityMean</th>\n",
       "      <th>angletBodyGyroJerkMeangravityMean</th>\n",
       "      <th>angleXgravityMean</th>\n",
       "      <th>angleYgravityMean</th>\n",
       "      <th>angleZgravityMean</th>\n",
       "      <th>subject</th>\n",
       "      <th>Activity</th>\n",
       "      <th>ActivityName</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.288585</td>\n",
       "      <td>-0.020294</td>\n",
       "      <td>-0.132905</td>\n",
       "      <td>-0.995279</td>\n",
       "      <td>-0.983111</td>\n",
       "      <td>-0.913526</td>\n",
       "      <td>-0.995112</td>\n",
       "      <td>-0.983185</td>\n",
       "      <td>-0.923527</td>\n",
       "      <td>-0.934724</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.112754</td>\n",
       "      <td>0.030400</td>\n",
       "      <td>-0.464761</td>\n",
       "      <td>-0.018446</td>\n",
       "      <td>-0.841247</td>\n",
       "      <td>0.179941</td>\n",
       "      <td>-0.058627</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>STANDING</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.278419</td>\n",
       "      <td>-0.016411</td>\n",
       "      <td>-0.123520</td>\n",
       "      <td>-0.998245</td>\n",
       "      <td>-0.975300</td>\n",
       "      <td>-0.960322</td>\n",
       "      <td>-0.998807</td>\n",
       "      <td>-0.974914</td>\n",
       "      <td>-0.957686</td>\n",
       "      <td>-0.943068</td>\n",
       "      <td>...</td>\n",
       "      <td>0.053477</td>\n",
       "      <td>-0.007435</td>\n",
       "      <td>-0.732626</td>\n",
       "      <td>0.703511</td>\n",
       "      <td>-0.844788</td>\n",
       "      <td>0.180289</td>\n",
       "      <td>-0.054317</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>STANDING</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.279653</td>\n",
       "      <td>-0.019467</td>\n",
       "      <td>-0.113462</td>\n",
       "      <td>-0.995380</td>\n",
       "      <td>-0.967187</td>\n",
       "      <td>-0.978944</td>\n",
       "      <td>-0.996520</td>\n",
       "      <td>-0.963668</td>\n",
       "      <td>-0.977469</td>\n",
       "      <td>-0.938692</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.118559</td>\n",
       "      <td>0.177899</td>\n",
       "      <td>0.100699</td>\n",
       "      <td>0.808529</td>\n",
       "      <td>-0.848933</td>\n",
       "      <td>0.180637</td>\n",
       "      <td>-0.049118</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>STANDING</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.279174</td>\n",
       "      <td>-0.026201</td>\n",
       "      <td>-0.123283</td>\n",
       "      <td>-0.996091</td>\n",
       "      <td>-0.983403</td>\n",
       "      <td>-0.990675</td>\n",
       "      <td>-0.997099</td>\n",
       "      <td>-0.982750</td>\n",
       "      <td>-0.989302</td>\n",
       "      <td>-0.938692</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.036788</td>\n",
       "      <td>-0.012892</td>\n",
       "      <td>0.640011</td>\n",
       "      <td>-0.485366</td>\n",
       "      <td>-0.848649</td>\n",
       "      <td>0.181935</td>\n",
       "      <td>-0.047663</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>STANDING</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.276629</td>\n",
       "      <td>-0.016570</td>\n",
       "      <td>-0.115362</td>\n",
       "      <td>-0.998139</td>\n",
       "      <td>-0.980817</td>\n",
       "      <td>-0.990482</td>\n",
       "      <td>-0.998321</td>\n",
       "      <td>-0.979672</td>\n",
       "      <td>-0.990441</td>\n",
       "      <td>-0.942469</td>\n",
       "      <td>...</td>\n",
       "      <td>0.123320</td>\n",
       "      <td>0.122542</td>\n",
       "      <td>0.693578</td>\n",
       "      <td>-0.615971</td>\n",
       "      <td>-0.847865</td>\n",
       "      <td>0.185151</td>\n",
       "      <td>-0.043892</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>STANDING</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 564 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   tBodyAccmeanX  tBodyAccmeanY  tBodyAccmeanZ  tBodyAccstdX  tBodyAccstdY  \\\n",
       "0       0.288585      -0.020294      -0.132905     -0.995279     -0.983111   \n",
       "1       0.278419      -0.016411      -0.123520     -0.998245     -0.975300   \n",
       "2       0.279653      -0.019467      -0.113462     -0.995380     -0.967187   \n",
       "3       0.279174      -0.026201      -0.123283     -0.996091     -0.983403   \n",
       "4       0.276629      -0.016570      -0.115362     -0.998139     -0.980817   \n",
       "\n",
       "   tBodyAccstdZ  tBodyAccmadX  tBodyAccmadY  tBodyAccmadZ  tBodyAccmaxX  ...  \\\n",
       "0     -0.913526     -0.995112     -0.983185     -0.923527     -0.934724  ...   \n",
       "1     -0.960322     -0.998807     -0.974914     -0.957686     -0.943068  ...   \n",
       "2     -0.978944     -0.996520     -0.963668     -0.977469     -0.938692  ...   \n",
       "3     -0.990675     -0.997099     -0.982750     -0.989302     -0.938692  ...   \n",
       "4     -0.990482     -0.998321     -0.979672     -0.990441     -0.942469  ...   \n",
       "\n",
       "   angletBodyAccMeangravity  angletBodyAccJerkMeangravityMean  \\\n",
       "0                 -0.112754                          0.030400   \n",
       "1                  0.053477                         -0.007435   \n",
       "2                 -0.118559                          0.177899   \n",
       "3                 -0.036788                         -0.012892   \n",
       "4                  0.123320                          0.122542   \n",
       "\n",
       "   angletBodyGyroMeangravityMean  angletBodyGyroJerkMeangravityMean  \\\n",
       "0                      -0.464761                          -0.018446   \n",
       "1                      -0.732626                           0.703511   \n",
       "2                       0.100699                           0.808529   \n",
       "3                       0.640011                          -0.485366   \n",
       "4                       0.693578                          -0.615971   \n",
       "\n",
       "   angleXgravityMean  angleYgravityMean  angleZgravityMean  subject  Activity  \\\n",
       "0          -0.841247           0.179941          -0.058627        1         5   \n",
       "1          -0.844788           0.180289          -0.054317        1         5   \n",
       "2          -0.848933           0.180637          -0.049118        1         5   \n",
       "3          -0.848649           0.181935          -0.047663        1         5   \n",
       "4          -0.847865           0.185151          -0.043892        1         5   \n",
       "\n",
       "   ActivityName  \n",
       "0      STANDING  \n",
       "1      STANDING  \n",
       "2      STANDING  \n",
       "3      STANDING  \n",
       "4      STANDING  \n",
       "\n",
       "[5 rows x 564 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tBodyAccmeanX</th>\n",
       "      <th>tBodyAccmeanY</th>\n",
       "      <th>tBodyAccmeanZ</th>\n",
       "      <th>tBodyAccstdX</th>\n",
       "      <th>tBodyAccstdY</th>\n",
       "      <th>tBodyAccstdZ</th>\n",
       "      <th>tBodyAccmadX</th>\n",
       "      <th>tBodyAccmadY</th>\n",
       "      <th>tBodyAccmadZ</th>\n",
       "      <th>tBodyAccmaxX</th>\n",
       "      <th>...</th>\n",
       "      <th>angletBodyAccMeangravity</th>\n",
       "      <th>angletBodyAccJerkMeangravityMean</th>\n",
       "      <th>angletBodyGyroMeangravityMean</th>\n",
       "      <th>angletBodyGyroJerkMeangravityMean</th>\n",
       "      <th>angleXgravityMean</th>\n",
       "      <th>angleYgravityMean</th>\n",
       "      <th>angleZgravityMean</th>\n",
       "      <th>subject</th>\n",
       "      <th>Activity</th>\n",
       "      <th>ActivityName</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.257178</td>\n",
       "      <td>-0.023285</td>\n",
       "      <td>-0.014654</td>\n",
       "      <td>-0.938404</td>\n",
       "      <td>-0.920091</td>\n",
       "      <td>-0.667683</td>\n",
       "      <td>-0.952501</td>\n",
       "      <td>-0.925249</td>\n",
       "      <td>-0.674302</td>\n",
       "      <td>-0.894088</td>\n",
       "      <td>...</td>\n",
       "      <td>0.006462</td>\n",
       "      <td>0.162920</td>\n",
       "      <td>-0.825886</td>\n",
       "      <td>0.271151</td>\n",
       "      <td>-0.720009</td>\n",
       "      <td>0.276801</td>\n",
       "      <td>-0.057978</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>STANDING</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.286027</td>\n",
       "      <td>-0.013163</td>\n",
       "      <td>-0.119083</td>\n",
       "      <td>-0.975415</td>\n",
       "      <td>-0.967458</td>\n",
       "      <td>-0.944958</td>\n",
       "      <td>-0.986799</td>\n",
       "      <td>-0.968401</td>\n",
       "      <td>-0.945823</td>\n",
       "      <td>-0.894088</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.083495</td>\n",
       "      <td>0.017500</td>\n",
       "      <td>-0.434375</td>\n",
       "      <td>0.920593</td>\n",
       "      <td>-0.698091</td>\n",
       "      <td>0.281343</td>\n",
       "      <td>-0.083898</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>STANDING</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.275485</td>\n",
       "      <td>-0.026050</td>\n",
       "      <td>-0.118152</td>\n",
       "      <td>-0.993819</td>\n",
       "      <td>-0.969926</td>\n",
       "      <td>-0.962748</td>\n",
       "      <td>-0.994403</td>\n",
       "      <td>-0.970735</td>\n",
       "      <td>-0.963483</td>\n",
       "      <td>-0.939260</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.034956</td>\n",
       "      <td>0.202302</td>\n",
       "      <td>0.064103</td>\n",
       "      <td>0.145068</td>\n",
       "      <td>-0.702771</td>\n",
       "      <td>0.280083</td>\n",
       "      <td>-0.079346</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>STANDING</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.270298</td>\n",
       "      <td>-0.032614</td>\n",
       "      <td>-0.117520</td>\n",
       "      <td>-0.994743</td>\n",
       "      <td>-0.973268</td>\n",
       "      <td>-0.967091</td>\n",
       "      <td>-0.995274</td>\n",
       "      <td>-0.974471</td>\n",
       "      <td>-0.968897</td>\n",
       "      <td>-0.938610</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.017067</td>\n",
       "      <td>0.154438</td>\n",
       "      <td>0.340134</td>\n",
       "      <td>0.296407</td>\n",
       "      <td>-0.698954</td>\n",
       "      <td>0.284114</td>\n",
       "      <td>-0.077108</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>STANDING</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.274833</td>\n",
       "      <td>-0.027848</td>\n",
       "      <td>-0.129527</td>\n",
       "      <td>-0.993852</td>\n",
       "      <td>-0.967445</td>\n",
       "      <td>-0.978295</td>\n",
       "      <td>-0.994111</td>\n",
       "      <td>-0.965953</td>\n",
       "      <td>-0.977346</td>\n",
       "      <td>-0.938610</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.002223</td>\n",
       "      <td>-0.040046</td>\n",
       "      <td>0.736715</td>\n",
       "      <td>-0.118545</td>\n",
       "      <td>-0.692245</td>\n",
       "      <td>0.290722</td>\n",
       "      <td>-0.073857</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>STANDING</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 564 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   tBodyAccmeanX  tBodyAccmeanY  tBodyAccmeanZ  tBodyAccstdX  tBodyAccstdY  \\\n",
       "0       0.257178      -0.023285      -0.014654     -0.938404     -0.920091   \n",
       "1       0.286027      -0.013163      -0.119083     -0.975415     -0.967458   \n",
       "2       0.275485      -0.026050      -0.118152     -0.993819     -0.969926   \n",
       "3       0.270298      -0.032614      -0.117520     -0.994743     -0.973268   \n",
       "4       0.274833      -0.027848      -0.129527     -0.993852     -0.967445   \n",
       "\n",
       "   tBodyAccstdZ  tBodyAccmadX  tBodyAccmadY  tBodyAccmadZ  tBodyAccmaxX  ...  \\\n",
       "0     -0.667683     -0.952501     -0.925249     -0.674302     -0.894088  ...   \n",
       "1     -0.944958     -0.986799     -0.968401     -0.945823     -0.894088  ...   \n",
       "2     -0.962748     -0.994403     -0.970735     -0.963483     -0.939260  ...   \n",
       "3     -0.967091     -0.995274     -0.974471     -0.968897     -0.938610  ...   \n",
       "4     -0.978295     -0.994111     -0.965953     -0.977346     -0.938610  ...   \n",
       "\n",
       "   angletBodyAccMeangravity  angletBodyAccJerkMeangravityMean  \\\n",
       "0                  0.006462                          0.162920   \n",
       "1                 -0.083495                          0.017500   \n",
       "2                 -0.034956                          0.202302   \n",
       "3                 -0.017067                          0.154438   \n",
       "4                 -0.002223                         -0.040046   \n",
       "\n",
       "   angletBodyGyroMeangravityMean  angletBodyGyroJerkMeangravityMean  \\\n",
       "0                      -0.825886                           0.271151   \n",
       "1                      -0.434375                           0.920593   \n",
       "2                       0.064103                           0.145068   \n",
       "3                       0.340134                           0.296407   \n",
       "4                       0.736715                          -0.118545   \n",
       "\n",
       "   angleXgravityMean  angleYgravityMean  angleZgravityMean  subject  Activity  \\\n",
       "0          -0.720009           0.276801          -0.057978        2         5   \n",
       "1          -0.698091           0.281343          -0.083898        2         5   \n",
       "2          -0.702771           0.280083          -0.079346        2         5   \n",
       "3          -0.698954           0.284114          -0.077108        2         5   \n",
       "4          -0.692245           0.290722          -0.073857        2         5   \n",
       "\n",
       "   ActivityName  \n",
       "0      STANDING  \n",
       "1      STANDING  \n",
       "2      STANDING  \n",
       "3      STANDING  \n",
       "4      STANDING  \n",
       "\n",
       "[5 rows x 564 columns]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# get x_train and y_train from train\n",
    "X_train = train.drop(['subject', 'Activity', 'ActivityName'], axis = 1)\n",
    "y_train = train.ActivityName"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "#get x_test and y_test from test\n",
    "X_test = test.drop(['subject', 'Activity', 'ActivityName'], axis = 1)\n",
    "y_test = test.ActivityName"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((7352, 561), (7352,))"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape, y_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((2947, 561), (2947,))"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test.shape, y_test.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "PLOT CONFUSION MATRIX"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "labels = ['LAYING', 'SITTING','STANDING','WALKING','WALKING_DOWNSTAIRS','WALKING_UPSTAIRS']\n",
    "\n",
    "# function to plot confusion matrix\n",
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "def plot_confusion_matrix(cm, classes, normalize=True, title='Normalized Confusion matrix',cmap=plt.cm.Blues):\n",
    "    if normalize:\n",
    "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
    "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
    "    plt.title(title)\n",
    "    plt.colorbar()\n",
    "    tick_marks = np.arange(len(classes))\n",
    "    plt.xticks(tick_marks, classes, rotation=90)\n",
    "    plt.yticks(tick_marks, classes)\n",
    "    fmt = '.2f' if normalize else 'd'\n",
    "    thresh = cm.max() / 2.\n",
    "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
    "        plt.text(j, i, format(cm[i, j], fmt),\n",
    "                 horizontalalignment=\"center\",\n",
    "                 color=\"white\" if cm[i, j] > thresh else \"black\")\n",
    "    plt.tight_layout()\n",
    "    plt.ylabel('True label')\n",
    "    plt.xlabel('Predicted label')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "# generic function to run any model specified\n",
    "from datetime import datetime\n",
    "def perform_model(model, X_train, y_train, X_test, y_test, class_labels, cm_normalize=True):\n",
    "    results = dict() # to store results at various phases\n",
    "    train_start_time = datetime.now()\n",
    "    model.fit(X_train, y_train)\n",
    "    train_end_time = datetime.now()\n",
    "    results['training_time'] = train_end_time - train_start_time\n",
    "    print(\"Training time :\", train_end_time - train_start_time)\n",
    "    \n",
    "    test_start_time = datetime.now()\n",
    "    y_pred = model.predict(X_test)\n",
    "    test_end_time = datetime.now()\n",
    "    results['testing_time'] = test_end_time - test_start_time\n",
    "    results['predicted'] = y_pred\n",
    "    print('testing time :', test_start_time - test_end_time)\n",
    "    \n",
    "    # calculate overall accuracy of model\n",
    "    accuracy = metrics.accuracy_score(y_true = y_test, y_pred = y_pred)\n",
    "    # store accuracy in results\n",
    "    results['accuracy'] = accuracy\n",
    "    print(\"ACCURACY : \", accuracy)\n",
    "    \n",
    "    # confusion matrix\n",
    "    cm = metrics.confusion_matrix(y_test, y_pred)\n",
    "    results['confusion_matrix'] = cm\n",
    "    print(\"CONFUSION MATRIX : \")\n",
    "    print(cm)\n",
    "    \n",
    "    # plot normalized confusion matrix\n",
    "    plt.figure(figsize = (8,8))\n",
    "    plt.grid(b = False)\n",
    "    plot_confusion_matrix(cm,classes=class_labels)\n",
    "    plt.show()\n",
    "    \n",
    "    # get classification report\n",
    "    print(\"Classification report :\")\n",
    "    classification_report = metrics.classification_report(y_test, y_pred)\n",
    "    results['classification_report'] = classification_report\n",
    "    print(classification_report)\n",
    "    \n",
    "    # add the trained model to the results\n",
    "    results['model'] = model\n",
    "    \n",
    "    return results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "METHOD TO PRINT GRIDSEARCH ATTRIBUTES"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "def print_grid_search_attributes(model):\n",
    "    # estimator that gave highest score among all the estimators formed in grid search\n",
    "    print(\"Best estimator :\", model.best_estimator_)\n",
    "    \n",
    "    # parameters that gave the best results while performing grid search\n",
    "    print(\"Best parameters :\", model.best_params_)\n",
    "    \n",
    "    # number of cross validation splits\n",
    "    print(\"No of cross validation splits :\", model.n_splits_)\n",
    "    \n",
    "    # avg cross validated score of the best estimator, from the grid search\n",
    "    print(\"Avg cross validate scores of best estimators :\", model.best_score_)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### LOGISTIC REGRESSION"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import linear_model\n",
    "from sklearn import metrics\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "import itertools"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 3 folds for each of 24 candidates, totalling 72 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  42 tasks      | elapsed:  7.1min\n",
      "[Parallel(n_jobs=-1)]: Done  72 out of  72 | elapsed: 14.3min finished\n",
      "C:\\ANACONDAUPDATED\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:469: FutureWarning: Default multi_class will be changed to 'auto' in 0.22. Specify the multi_class option to silence this warning.\n",
      "  \"this warning.\", FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training time : 0:14:30.937607\n",
      "testing time : -1 day, 23:59:59.984617\n",
      "ACCURACY :  0.9626739056667798\n",
      "CONFUSION MATRIX : \n",
      "[[537   0   0   0   0   0]\n",
      " [  1 428  58   0   0   4]\n",
      " [  0  12 519   1   0   0]\n",
      " [  0   0   0 495   1   0]\n",
      " [  0   0   0   3 409   8]\n",
      " [  0   0   0  22   0 449]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification report :\n",
      "                    precision    recall  f1-score   support\n",
      "\n",
      "            LAYING       1.00      1.00      1.00       537\n",
      "           SITTING       0.97      0.87      0.92       491\n",
      "          STANDING       0.90      0.98      0.94       532\n",
      "           WALKING       0.95      1.00      0.97       496\n",
      "WALKING_DOWNSTAIRS       1.00      0.97      0.99       420\n",
      "  WALKING_UPSTAIRS       0.97      0.95      0.96       471\n",
      "\n",
      "          accuracy                           0.96      2947\n",
      "         macro avg       0.97      0.96      0.96      2947\n",
      "      weighted avg       0.96      0.96      0.96      2947\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# start grid search\n",
    "parameters = {'C':[0.01, 0.1, 1, 10, 20, 30], 'penalty':['l2', 'l1'], 'solver':['liblinear', 'saga']}\n",
    "log_reg = linear_model.LogisticRegression()\n",
    "log_reg_grid = GridSearchCV(log_reg, param_grid = parameters, cv=3, verbose = 1, n_jobs = -1)\n",
    "log_reg_grid_results = perform_model(log_reg_grid, X_train, y_train, X_test, y_test, class_labels=labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best estimator : LogisticRegression(C=30, class_weight=None, dual=False, fit_intercept=True,\n",
      "                   intercept_scaling=1, l1_ratio=None, max_iter=100,\n",
      "                   multi_class='warn', n_jobs=None, penalty='l2',\n",
      "                   random_state=None, solver='warn', tol=0.0001, verbose=0,\n",
      "                   warm_start=False)\n",
      "Best parameters : {'C': 30, 'penalty': 'l2'}\n",
      "No of cross validation splits : 3\n",
      "Avg cross validate scores of best estimators : 0.9461371055495104\n"
     ]
    }
   ],
   "source": [
    "# observe the attributes of the model \n",
    "print_grid_search_attributes(log_reg_grid_results['model'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### LINEAR SVC WITH GRID SEARCH"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.svm import LinearSVC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 3 folds for each of 6 candidates, totalling 18 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ANACONDAUPDATED\\lib\\site-packages\\sklearn\\model_selection\\_split.py:1978: FutureWarning: The default value of cv will change from 3 to 5 in version 0.22. Specify it explicitly to silence this warning.\n",
      "  warnings.warn(CV_WARNING, FutureWarning)\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  18 out of  18 | elapsed:  1.3min finished\n",
      "C:\\ANACONDAUPDATED\\lib\\site-packages\\sklearn\\svm\\base.py:929: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.\n",
      "  \"the number of iterations.\", ConvergenceWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training time : 0:01:24.788407\n",
      "testing time : 0:00:00\n",
      "ACCURACY :  0.9555480149304377\n",
      "CONFUSION MATRIX : \n",
      "[[537   0   0   0   0   0]\n",
      " [  2 385  99   0   0   5]\n",
      " [  0   2 529   1   0   0]\n",
      " [  0   0   0 495   0   1]\n",
      " [  0   0   0   2 413   5]\n",
      " [  0   0   0  13   1 457]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification report :\n",
      "                    precision    recall  f1-score   support\n",
      "\n",
      "            LAYING       1.00      1.00      1.00       537\n",
      "           SITTING       0.99      0.78      0.88       491\n",
      "          STANDING       0.84      0.99      0.91       532\n",
      "           WALKING       0.97      1.00      0.98       496\n",
      "WALKING_DOWNSTAIRS       1.00      0.98      0.99       420\n",
      "  WALKING_UPSTAIRS       0.98      0.97      0.97       471\n",
      "\n",
      "          accuracy                           0.96      2947\n",
      "         macro avg       0.96      0.96      0.96      2947\n",
      "      weighted avg       0.96      0.96      0.95      2947\n",
      "\n"
     ]
    }
   ],
   "source": [
    "parameters = {'C':[0.125, 0.5, 1, 2, 8, 16]}\n",
    "lr_svc = LinearSVC(tol=0.00005)\n",
    "lr_svc_grid = GridSearchCV(lr_svc, param_grid=parameters, n_jobs=-1, verbose=1)\n",
    "lr_svc_grid_results = perform_model(lr_svc_grid, X_train, y_train, X_test, y_test, class_labels=labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best estimator : LinearSVC(C=8, class_weight=None, dual=True, fit_intercept=True,\n",
      "          intercept_scaling=1, loss='squared_hinge', max_iter=1000,\n",
      "          multi_class='ovr', penalty='l2', random_state=None, tol=5e-05,\n",
      "          verbose=0)\n",
      "Best parameters : {'C': 8}\n",
      "No of cross validation splits : 3\n",
      "Avg cross validate scores of best estimators : 0.9468171926006529\n"
     ]
    }
   ],
   "source": [
    "print_grid_search_attributes(lr_svc_grid_results['model'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### KERNEL SVM WITH GRID SEARCH"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ANACONDAUPDATED\\lib\\site-packages\\sklearn\\model_selection\\_split.py:1978: FutureWarning: The default value of cv will change from 3 to 5 in version 0.22. Specify it explicitly to silence this warning.\n",
      "  warnings.warn(CV_WARNING, FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training time : 0:05:06.479497\n",
      "testing time : -1 day, 23:59:56.235748\n",
      "ACCURACY :  0.9626739056667798\n",
      "CONFUSION MATRIX : \n",
      "[[537   0   0   0   0   0]\n",
      " [  0 441  48   0   0   2]\n",
      " [  0  12 520   0   0   0]\n",
      " [  0   0   0 489   2   5]\n",
      " [  0   0   0   4 397  19]\n",
      " [  0   0   0  17   1 453]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification report :\n",
      "                    precision    recall  f1-score   support\n",
      "\n",
      "            LAYING       1.00      1.00      1.00       537\n",
      "           SITTING       0.97      0.90      0.93       491\n",
      "          STANDING       0.92      0.98      0.95       532\n",
      "           WALKING       0.96      0.99      0.97       496\n",
      "WALKING_DOWNSTAIRS       0.99      0.95      0.97       420\n",
      "  WALKING_UPSTAIRS       0.95      0.96      0.95       471\n",
      "\n",
      "          accuracy                           0.96      2947\n",
      "         macro avg       0.96      0.96      0.96      2947\n",
      "      weighted avg       0.96      0.96      0.96      2947\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.svm import SVC\n",
    "parameters = {'C':[2,8,16],\\\n",
    "              'gamma': [ 0.0078125, 0.125, 2]}\n",
    "rbf_svm = SVC(kernel='rbf')\n",
    "rbf_svm_grid = GridSearchCV(rbf_svm,param_grid=parameters, n_jobs=-1)\n",
    "rbf_svm_grid_results = perform_model(rbf_svm_grid, X_train, y_train, X_test, y_test, class_labels=labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best estimator : SVC(C=16, cache_size=200, class_weight=None, coef0=0.0,\n",
      "    decision_function_shape='ovr', degree=3, gamma=0.0078125, kernel='rbf',\n",
      "    max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
      "    tol=0.001, verbose=False)\n",
      "Best parameters : {'C': 16, 'gamma': 0.0078125}\n",
      "No of cross validation splits : 3\n",
      "Avg cross validate scores of best estimators : 0.9440968443960827\n"
     ]
    }
   ],
   "source": [
    "print_grid_search_attributes(rbf_svm_grid_results['model'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### DECISION TREES WITH GRID SEARCH CV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ANACONDAUPDATED\\lib\\site-packages\\sklearn\\model_selection\\_split.py:1978: FutureWarning: The default value of cv will change from 3 to 5 in version 0.22. Specify it explicitly to silence this warning.\n",
      "  warnings.warn(CV_WARNING, FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training time : 0:00:13.682457\n",
      "testing time : 0:00:00\n",
      "ACCURACY :  0.8629114353579912\n",
      "CONFUSION MATRIX : \n",
      "[[537   0   0   0   0   0]\n",
      " [  0 385 106   0   0   0]\n",
      " [  0  93 439   0   0   0]\n",
      " [  0   0   0 472  16   8]\n",
      " [  0   0   0  18 341  61]\n",
      " [  0   0   0  78  24 369]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification report :\n",
      "                    precision    recall  f1-score   support\n",
      "\n",
      "            LAYING       1.00      1.00      1.00       537\n",
      "           SITTING       0.81      0.78      0.79       491\n",
      "          STANDING       0.81      0.83      0.82       532\n",
      "           WALKING       0.83      0.95      0.89       496\n",
      "WALKING_DOWNSTAIRS       0.90      0.81      0.85       420\n",
      "  WALKING_UPSTAIRS       0.84      0.78      0.81       471\n",
      "\n",
      "          accuracy                           0.86      2947\n",
      "         macro avg       0.86      0.86      0.86      2947\n",
      "      weighted avg       0.86      0.86      0.86      2947\n",
      "\n",
      "Best estimator : DecisionTreeClassifier(class_weight=None, criterion='gini', max_depth=7,\n",
      "                       max_features=None, max_leaf_nodes=None,\n",
      "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
      "                       min_samples_leaf=1, min_samples_split=2,\n",
      "                       min_weight_fraction_leaf=0.0, presort=False,\n",
      "                       random_state=None, splitter='best')\n",
      "Best parameters : {'max_depth': 7}\n",
      "No of cross validation splits : 3\n",
      "Avg cross validate scores of best estimators : 0.8354189336235038\n"
     ]
    }
   ],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "parameters = {'max_depth':np.arange(3,10,2)}\n",
    "dt = DecisionTreeClassifier()\n",
    "dt_grid = GridSearchCV(dt,param_grid=parameters, n_jobs=-1)\n",
    "dt_grid_results = perform_model(dt_grid, X_train, y_train, X_test, y_test, class_labels=labels)\n",
    "print_grid_search_attributes(dt_grid_results['model'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### RANDOM FOREST CLASSIFIER WITH GRID SEARCH"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ANACONDAUPDATED\\lib\\site-packages\\sklearn\\model_selection\\_split.py:1978: FutureWarning: The default value of cv will change from 3 to 5 in version 0.22. Specify it explicitly to silence this warning.\n",
      "  warnings.warn(CV_WARNING, FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training time : 0:07:39.977187\n",
      "testing time : -1 day, 23:59:59.920996\n",
      "ACCURACY :  0.9097387173396675\n",
      "CONFUSION MATRIX : \n",
      "[[537   0   0   0   0   0]\n",
      " [  0 419  72   0   0   0]\n",
      " [  0  54 478   0   0   0]\n",
      " [  0   0   0 484  10   2]\n",
      " [  0   0   0  37 338  45]\n",
      " [  0   0   0  40   6 425]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification report :\n",
      "                    precision    recall  f1-score   support\n",
      "\n",
      "            LAYING       1.00      1.00      1.00       537\n",
      "           SITTING       0.89      0.85      0.87       491\n",
      "          STANDING       0.87      0.90      0.88       532\n",
      "           WALKING       0.86      0.98      0.92       496\n",
      "WALKING_DOWNSTAIRS       0.95      0.80      0.87       420\n",
      "  WALKING_UPSTAIRS       0.90      0.90      0.90       471\n",
      "\n",
      "          accuracy                           0.91      2947\n",
      "         macro avg       0.91      0.91      0.91      2947\n",
      "      weighted avg       0.91      0.91      0.91      2947\n",
      "\n",
      "Best estimator : RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
      "                       max_depth=7, max_features='auto', max_leaf_nodes=None,\n",
      "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
      "                       min_samples_leaf=1, min_samples_split=2,\n",
      "                       min_weight_fraction_leaf=0.0, n_estimators=90,\n",
      "                       n_jobs=None, oob_score=False, random_state=None,\n",
      "                       verbose=0, warm_start=False)\n",
      "Best parameters : {'max_depth': 7, 'n_estimators': 90}\n",
      "No of cross validation splits : 3\n",
      "Avg cross validate scores of best estimators : 0.9156692056583242\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "params = {'n_estimators': np.arange(10,201,20), 'max_depth':np.arange(3,15,2)}\n",
    "rfc = RandomForestClassifier()\n",
    "rfc_grid = GridSearchCV(rfc, param_grid=params, n_jobs=-1)\n",
    "rfc_grid_results = perform_model(rfc_grid, X_train, y_train, X_test, y_test, class_labels=labels)\n",
    "print_grid_search_attributes(rfc_grid_results['model'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### GRADIENT BOOSTED DECISION TREES WITH GRID SEARCH"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ANACONDAUPDATED\\lib\\site-packages\\sklearn\\model_selection\\_split.py:1978: FutureWarning: The default value of cv will change from 3 to 5 in version 0.22. Specify it explicitly to silence this warning.\n",
      "  warnings.warn(CV_WARNING, FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training time : 4:39:55.174200\n",
      "testing time : -1 day, 23:59:59.790438\n",
      "ACCURACY :  0.9243298269426535\n",
      "CONFUSION MATRIX : \n",
      "[[537   0   0   0   0   0]\n",
      " [  0 402  87   0   0   2]\n",
      " [  0  37 495   0   0   0]\n",
      " [  0   0   0 482   7   7]\n",
      " [  0   0   0   7 371  42]\n",
      " [  0   1   0  28   5 437]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification report :\n",
      "                    precision    recall  f1-score   support\n",
      "\n",
      "            LAYING       1.00      1.00      1.00       537\n",
      "           SITTING       0.91      0.82      0.86       491\n",
      "          STANDING       0.85      0.93      0.89       532\n",
      "           WALKING       0.93      0.97      0.95       496\n",
      "WALKING_DOWNSTAIRS       0.97      0.88      0.92       420\n",
      "  WALKING_UPSTAIRS       0.90      0.93      0.91       471\n",
      "\n",
      "          accuracy                           0.92      2947\n",
      "         macro avg       0.93      0.92      0.92      2947\n",
      "      weighted avg       0.93      0.92      0.92      2947\n",
      "\n",
      "Best estimator : GradientBoostingClassifier(criterion='friedman_mse', init=None,\n",
      "                           learning_rate=0.1, loss='deviance', max_depth=5,\n",
      "                           max_features=None, max_leaf_nodes=None,\n",
      "                           min_impurity_decrease=0.0, min_impurity_split=None,\n",
      "                           min_samples_leaf=1, min_samples_split=2,\n",
      "                           min_weight_fraction_leaf=0.0, n_estimators=160,\n",
      "                           n_iter_no_change=None, presort='auto',\n",
      "                           random_state=None, subsample=1.0, tol=0.0001,\n",
      "                           validation_fraction=0.1, verbose=0,\n",
      "                           warm_start=False)\n",
      "Best parameters : {'max_depth': 5, 'n_estimators': 160}\n",
      "No of cross validation splits : 3\n",
      "Avg cross validate scores of best estimators : 0.9053318824809575\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import GradientBoostingClassifier\n",
    "param_grid = {'max_depth': np.arange(5,8,1), \\\n",
    "             'n_estimators':np.arange(130,170,10)}\n",
    "gbdt = GradientBoostingClassifier()\n",
    "gbdt_grid = GridSearchCV(gbdt, param_grid=param_grid, n_jobs=-1)\n",
    "gbdt_grid_results = perform_model(gbdt_grid, X_train, y_train, X_test, y_test, class_labels=labels)\n",
    "print_grid_search_attributes(gbdt_grid_results['model'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### COMPARING ALL MODELS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "                     Accuracy     Error\n",
      "                     ----------   --------\n",
      "Logistic Regression : 96.27%       3.733%\n",
      "Linear SVC          : 95.55%       4.445% \n",
      "rbf SVM classifier  : 96.27%      3.733% \n",
      "DecisionTree        : 86.29%      13.71% \n",
      "Random Forest       : 90.97%      9.026% \n",
      "GradientBoosting DT : 92.43%      7.567% \n"
     ]
    }
   ],
   "source": [
    "print('\\n                     Accuracy     Error')\n",
    "print('                     ----------   --------')\n",
    "print('Logistic Regression : {:.04}%       {:.04}%'.format(log_reg_grid_results['accuracy'] * 100,\\\n",
    "                                                  100-(log_reg_grid_results['accuracy'] * 100)))\n",
    "\n",
    "print('Linear SVC          : {:.04}%       {:.04}% '.format(lr_svc_grid_results['accuracy'] * 100,\\\n",
    "                                                        100-(lr_svc_grid_results['accuracy'] * 100)))\n",
    "\n",
    "print('rbf SVM classifier  : {:.04}%      {:.04}% '.format(rbf_svm_grid_results['accuracy'] * 100,\\\n",
    "                                                          100-(rbf_svm_grid_results['accuracy'] * 100)))\n",
    "\n",
    "print('DecisionTree        : {:.04}%      {:.04}% '.format(dt_grid_results['accuracy'] * 100,\\\n",
    "                                                        100-(dt_grid_results['accuracy'] * 100)))\n",
    "\n",
    "print('Random Forest       : {:.04}%      {:.04}% '.format(rfc_grid_results['accuracy'] * 100,\\\n",
    "                                                           100-(rfc_grid_results['accuracy'] * 100)))\n",
    "print('GradientBoosting DT : {:.04}%      {:.04}% '.format(gbdt_grid_results['accuracy'] * 100,\\\n",
    "                                                        100-(gbdt_grid_results['accuracy'] * 100)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can choose Logistic Regression, Linear SVC or RBF SVM classifier as the final model.\n",
    "Tree based models give less accurate results."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "# save all the models \n",
    "import pickle\n",
    "pickle.dump(log_reg_grid_results['model'] ,open('models/logistic_regression_model.pkl', 'wb'))\n",
    "pickle.dump(lr_svc_grid_results['model'] ,open('models/linear_svc_model.pkl', 'wb'))\n",
    "pickle.dump(rbf_svm_grid_results['model'] ,open('models/svm_rbf_model.pkl', 'wb'))\n",
    "pickle.dump(dt_grid_results['model'] ,open('models/dec_tree_model.pkl', 'wb'))\n",
    "pickle.dump(rfc_grid_results['model'] ,open('models/rand_forest_model.pkl', 'wb'))\n",
    "pickle.dump(gbdt_grid_results['model'] ,open('grad_boosted_dec_tree_model.pkl', 'wb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "# end of notebook 2"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}