Human_Activity_Recogn_using_ML_Models.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Project : \n",
    "## Human Activity Recognition : Predictions using ML Models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing necessary libraries\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Reading data from CSV file\n",
    "\n",
    "train = pd.read_csv('UCI_HAR_Dataset/csv_files/train.csv')\n",
    "test = pd.read_csv('UCI_HAR_Dataset/csv_files/test.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(7352, 564)\n",
      "(2947, 564)\n"
     ]
    }
   ],
   "source": [
    "# Checking the shape of train and test\n",
    "\n",
    "print(train.shape)\n",
    "print(test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>Activity_Name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\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",
       "      <th>1</th>\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",
       "      <th>2</th>\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",
       "      <th>3</th>\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",
       "      <th>4</th>\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",
       "   Activity_Name  \n",
       "0       Standing  \n",
       "1       Standing  \n",
       "2       Standing  \n",
       "3       Standing  \n",
       "4       Standing  \n",
       "\n",
       "[5 rows x 564 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Displaying first 5 rows of training data\n",
    "\n",
    "train.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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",
       "      <th>0</th>\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",
       "      <th>1</th>\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",
       "      <th>2</th>\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",
       "      <th>3</th>\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",
       "      <th>4</th>\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": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Displaying first 5 rows of testing data\n",
    "\n",
    "test.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Getting X_train and y_train from train data\n",
    "\n",
    "X_train = train.drop(['Subject', 'Activity', 'Activity_Name'], axis=1)\n",
    "y_train = train.Activity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Getting X_test and y_test from test data\n",
    "\n",
    "X_test = test.drop(['subject', 'Activity', 'ActivityName'], axis=1)\n",
    "y_test = test.Activity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_train and y_train : ((7352, 561),(7352,))\n",
      "X_test  and y_test  : ((2947, 561),(2947,))\n"
     ]
    }
   ],
   "source": [
    "# Displaying the shape of training and testing data\n",
    "\n",
    "print('X_train and y_train : ({},{})'.format(X_train.shape, y_train.shape))\n",
    "print('X_test  and y_test  : ({},{})'.format(X_test.shape, y_test.shape))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Let's use Linear discriminant analysis to find features that classifies the label well\n",
    "\n",
    "# Importing libraries\n",
    "\n",
    "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis as LDA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\discriminant_analysis.py:388: UserWarning: Variables are collinear.\n",
      "  warnings.warn(\"Variables are collinear.\")\n"
     ]
    }
   ],
   "source": [
    "lda = LDA()\n",
    "X_train = lda.fit_transform(X_train, y_train)\n",
    "X_test = lda.transform(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_train and y_train : ((7352, 5),(7352,))\n",
      "X_test  and y_test  : ((2947, 5),(2947,))\n"
     ]
    }
   ],
   "source": [
    "# Displaying the shape of training and testing data\n",
    "\n",
    "print('X_train and y_train : ({},{})'.format(X_train.shape, y_train.shape))\n",
    "print('X_test  and y_test  : ({},{})'.format(X_test.shape, y_test.shape))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Let's define some generic functions to create ML models"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Function to plot Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing necessary libraries\n",
    "\n",
    "import itertools\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Creating a function to print text in Bold and in given color\n",
    "\n",
    "from IPython.display import Markdown, display\n",
    "\n",
    "def printmd(string, color=None):\n",
    "    colorstr = \"<span style='color:{}'>{}</span>\".format(color, string)\n",
    "    display(Markdown(colorstr))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Function to plot Confusion Matrix\n",
    "\n",
    "def plot_confusion_matrix(cm, classes,\n",
    "                         normalize=False,\n",
    "                         title='Confusion Matrix',\n",
    "                         cmap = plt.cm.Blues):\n",
    "    if normalize:\n",
    "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
    "        \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",
    "\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",
    "\n",
    "    plt.tight_layout()\n",
    "    plt.ylabel('Actual label')\n",
    "    plt.xlabel('Predicted label')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Generic function to any model\n",
    "\n",
    "from datetime import datetime\n",
    "def perform_model(model, X_train, y_train, X_test, y_test, class_labels, cm_normalize=True, \\\n",
    "                 print_cm=True, cm_cmap=plt.cm.Greens):\n",
    "    \n",
    "    \n",
    "    # Let's create an empty dictionary to be returned by the function\n",
    "    results = dict()\n",
    "    \n",
    "    # Let's calculate & print the total training time\n",
    "    \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",
    "    printmd('Training_time(HH:MM:SS.ms) - {}'.format(results['training_time']), color='blue')\n",
    "        \n",
    "    # Let's calculate & print the test 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",
    "    printmd('testing time(HH:MM:SS:ms) - {}'.format(results['testing_time']), color='blue')\n",
    "    results['predicted'] = y_pred\n",
    "   \n",
    "    # Let's calculate the Accuracy of Model\n",
    "    \n",
    "    accuracy = metrics.accuracy_score(y_true=y_test, y_pred=y_pred)\n",
    "    results['accuracy'] = accuracy\n",
    "    printmd('**Accuracy:**', color='blue')\n",
    "    print('{}'.format(accuracy))\n",
    "        \n",
    "    # Let's get the Confusion Matrix\n",
    "    \n",
    "    cm = metrics.confusion_matrix(y_test, y_pred)\n",
    "    \n",
    "    # Plotting Confusion Matrix\n",
    "    \n",
    "    printmd('**Confusion Matrix:**', color='blue')\n",
    "    plt.figure(figsize=(8,8))\n",
    "    plt.grid(b=False)\n",
    "    plot_confusion_matrix(cm, classes=labels, title='Normalized confusion matrix', cmap=plt.cm.YlGn, )\n",
    "    plt.show()\n",
    "        \n",
    "    # Plotting Normalized Confusion Matrix\n",
    "    \n",
    "    printmd('**Normalized Confusion Matrix:**', color='blue')\n",
    "    plt.figure(figsize=(8,8))\n",
    "    plt.grid(b=False)\n",
    "    plot_confusion_matrix(cm, classes=class_labels, normalize=True, title='Normalized confusion matrix', cmap = cm_cmap)\n",
    "    plt.show()\n",
    "    \n",
    "    # PLotting classification report\n",
    "    \n",
    "    printmd('**Classifiction Report**', color='blue')\n",
    "    classification_report = metrics.classification_report(y_test, y_pred)\n",
    "    results['classification_report'] = classification_report\n",
    "    print(classification_report)\n",
    "    \n",
    "    # Adding the trained model to the results\n",
    "    \n",
    "    results['model'] = model\n",
    "    \n",
    "    return results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "def print_grid_search_attributes(model):\n",
    "    \n",
    "    # Let's print the best estimator that gave highest score\n",
    "    \n",
    "    printmd('**Best Estimator:**', color='blue')\n",
    "    print('{}\\n'.format(model.best_estimator_))\n",
    "\n",
    "\n",
    "    # Let's print the best parameters that gave best results\n",
    "    \n",
    "    printmd('**Best parameters:**', color='blue')\n",
    "    print('{}\\n'.format(model.best_params_))\n",
    "\n",
    "\n",
    "    #  Let's print the number of cross validation splits\n",
    "    \n",
    "    printmd('**Number of CrossValidation sets:**', color='blue')\n",
    "    print('{}\\n'.format(model.n_splits_))\n",
    "\n",
    "\n",
    "    # Let's print the Best score of the best estimator\n",
    "    \n",
    "    printmd('**Best Score:**', color='blue')\n",
    "    print('{}\\n'.format(model.best_score_))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Applying various Machine learning model with Grid-Search"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1. Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing necessary libraries\n",
    "\n",
    "from sklearn import linear_model\n",
    "from sklearn import metrics\n",
    "\n",
    "from sklearn.model_selection import GridSearchCV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Creating a list labels to be added to plots\n",
    "\n",
    "labels=['Laying', 'Sitting','Standing','Walking','Walking_Downstairs','Walking_Upstairs']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n",
      "  warnings.warn(CV_WARNING, FutureWarning)\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 3 folds for each of 10 candidates, totalling 30 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done  30 out of  30 | elapsed:    3.1s finished\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:433: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:460: 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"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>Training_time(HH:MM:SS.ms) - 0:00:03.267407</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>testing time(HH:MM:SS:ms) - 0:00:00.002174</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Accuracy:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9613165931455717\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Confusion Matrix:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Normalized Confusion Matrix:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Classifiction Report**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           1       0.97      0.99      0.98       496\n",
      "           2       0.97      0.96      0.97       471\n",
      "           3       1.00      0.98      0.99       420\n",
      "           4       0.94      0.87      0.91       491\n",
      "           5       0.89      0.96      0.93       532\n",
      "           6       1.00      1.00      1.00       537\n",
      "\n",
      "   micro avg       0.96      0.96      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"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Best Estimator:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LogisticRegression(C=30, class_weight=None, dual=False, fit_intercept=True,\n",
      "          intercept_scaling=1, max_iter=100, multi_class='warn',\n",
      "          n_jobs=None, penalty='l1', random_state=None, solver='warn',\n",
      "          tol=0.0001, verbose=0, warm_start=False)\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Best parameters:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'C': 30, 'penalty': 'l1'}\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Number of CrossValidation sets:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Best Score:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9846300326441785\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Let's define the parameters to be tuned\n",
    "\n",
    "parameters = {'C':[20, 25, 30, 35, 40], 'penalty':['l1', 'l2']}\n",
    "\n",
    "# Let's initiate the model\n",
    "\n",
    "log_reg = linear_model.LogisticRegression()\n",
    "log_reg_grid = GridSearchCV(log_reg, param_grid=parameters, 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)\n",
    "\n",
    "# Printing the best attributes of the model \n",
    "\n",
    "print_grid_search_attributes(log_reg_grid_results['model'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2. Support Vector Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing Necessary libraries\n",
    "\n",
    "from sklearn.svm import LinearSVC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "scrolled": false
   },
   "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:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n",
      "  warnings.warn(CV_WARNING, FutureWarning)\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   3 out of  18 | elapsed:    0.1s remaining:    0.7s\n",
      "[Parallel(n_jobs=-1)]: Done  18 out of  18 | elapsed:    0.5s finished\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\svm\\base.py:922: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.\n",
      "  \"the number of iterations.\", ConvergenceWarning)\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>Training_time(HH:MM:SS.ms) - 0:00:00.696371</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>testing time(HH:MM:SS:ms) - 0:00:00</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Accuracy:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9630132337970818\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Confusion Matrix:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Normalized Confusion Matrix:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Classifiction Report**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           1       0.97      0.99      0.98       496\n",
      "           2       0.98      0.97      0.97       471\n",
      "           3       1.00      0.99      0.99       420\n",
      "           4       0.95      0.87      0.91       491\n",
      "           5       0.89      0.96      0.93       532\n",
      "           6       1.00      1.00      1.00       537\n",
      "\n",
      "   micro avg       0.96      0.96      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"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Best Estimator:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LinearSVC(C=1, 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=0.0001,\n",
      "     verbose=0)\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Best parameters:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'C': 1}\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Number of CrossValidation sets:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Best Score:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.985038084874864\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Let's define the parameters to be tuned\n",
    "\n",
    "parameters = {'C':[0.25, 0.5, 1, 2, 4, 8]}\n",
    "\n",
    "# Let's initiate the model\n",
    "\n",
    "lin_svc = LinearSVC()\n",
    "lin_svc_grid = GridSearchCV(lin_svc, param_grid=parameters, verbose=1, n_jobs=-1)\n",
    "lin_svc_grid_results = perform_model(lin_svc_grid, X_train, y_train, X_test, y_test, class_labels=labels)\n",
    "\n",
    "# Printing the best attributes of the model \n",
    "\n",
    "print_grid_search_attributes(lin_svc_grid_results['model'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3. Kernel SVM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing Libraries\n",
    "\n",
    "from sklearn.svm import SVC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n",
      "  warnings.warn(CV_WARNING, FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>Training_time(HH:MM:SS.ms) - 0:00:49.726736</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>testing time(HH:MM:SS:ms) - 0:00:00.048000</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Accuracy:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9657278588394977\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Confusion Matrix:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Normalized Confusion Matrix:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Classifiction Report**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           1       0.98      0.99      0.98       496\n",
      "           2       0.97      0.98      0.98       471\n",
      "           3       1.00      0.98      0.99       420\n",
      "           4       0.95      0.89      0.92       491\n",
      "           5       0.90      0.96      0.93       532\n",
      "           6       1.00      1.00      1.00       537\n",
      "\n",
      "   micro avg       0.97      0.97      0.97      2947\n",
      "   macro avg       0.97      0.97      0.97      2947\n",
      "weighted avg       0.97      0.97      0.97      2947\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Best Estimator:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SVC(C=0.25, cache_size=200, class_weight=None, coef0=0.0,\n",
      "  decision_function_shape='ovr', degree=3, gamma=0.01, kernel='rbf',\n",
      "  max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
      "  tol=0.001, verbose=False)\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Best parameters:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'C': 0.25, 'gamma': 0.01}\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Number of CrossValidation sets:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Best Score:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.986126224156692\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Let's define the parameters to be tuned\n",
    "\n",
    "parameters = {'C':[0.125, 0.25, 0.5, 1], 'gamma':[0.01, 0.1, 1, 2]}\n",
    "\n",
    "# Let's initiate the model\n",
    "\n",
    "rbf_svc = SVC(kernel='rbf')\n",
    "rbf_svc_grid = GridSearchCV(rbf_svc, param_grid=parameters)\n",
    "rbf_svc_grid_results = perform_model(rbf_svc_grid, X_train, y_train, X_test, y_test, class_labels=labels)\n",
    "\n",
    "# Printing the best attributes of the model \n",
    "\n",
    "print_grid_search_attributes(rbf_svc_grid_results['model'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4. Decision Tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing libraries\n",
    "\n",
    "from sklearn.tree import DecisionTreeClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "scrolled": false
   },
   "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:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n",
      "  warnings.warn(CV_WARNING, FutureWarning)\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   3 out of  18 | elapsed:    0.0s remaining:    0.2s\n",
      "[Parallel(n_jobs=-1)]: Done  18 out of  18 | elapsed:    0.0s finished\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>Training_time(HH:MM:SS.ms) - 0:00:00.145372</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>testing time(HH:MM:SS:ms) - 0:00:00</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Accuracy:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9514760773668137\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Confusion Matrix:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Normalized Confusion Matrix:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Classifiction Report**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           1       0.92      0.99      0.96       496\n",
      "           2       0.98      0.95      0.96       471\n",
      "           3       1.00      0.94      0.97       420\n",
      "           4       0.94      0.87      0.90       491\n",
      "           5       0.89      0.95      0.92       532\n",
      "           6       1.00      1.00      1.00       537\n",
      "\n",
      "   micro avg       0.95      0.95      0.95      2947\n",
      "   macro avg       0.95      0.95      0.95      2947\n",
      "weighted avg       0.95      0.95      0.95      2947\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Best Estimator:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DecisionTreeClassifier(class_weight=None, criterion='gini', 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, presort=False, random_state=None,\n",
      "            splitter='best')\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Best parameters:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'max_depth': 5}\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Number of CrossValidation sets:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Best Score:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9831338411316648\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Let's define the parameters to be tuned\n",
    "\n",
    "parameters = {'max_depth':np.arange(4,10,1)}\n",
    "\n",
    "# Let's initiate the model\n",
    "\n",
    "dtree = DecisionTreeClassifier()\n",
    "dtree_grid = GridSearchCV(dtree, param_grid=parameters, verbose=1, n_jobs=-1)\n",
    "dtree_grid_results = perform_model(dtree_grid, X_train, y_train, X_test, y_test, class_labels=labels)\n",
    "\n",
    "# Printing the best attributes of the model \n",
    "\n",
    "print_grid_search_attributes(dtree_grid_results['model'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 5. Random Forest Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing libraries\n",
    "\n",
    "from sklearn.ensemble import RandomForestClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n",
      "  warnings.warn(CV_WARNING, FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>Training_time(HH:MM:SS.ms) - 0:00:21.379325</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>testing time(HH:MM:SS:ms) - 0:00:00.034036</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Accuracy:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9619952494061758\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Confusion Matrix:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Normalized Confusion Matrix:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Classifiction Report**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           1       0.97      0.99      0.98       496\n",
      "           2       0.98      0.97      0.98       471\n",
      "           3       1.00      0.98      0.99       420\n",
      "           4       0.96      0.86      0.91       491\n",
      "           5       0.88      0.97      0.92       532\n",
      "           6       1.00      1.00      1.00       537\n",
      "\n",
      "   micro avg       0.96      0.96      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"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Best Estimator:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
      "            max_depth=6, 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=110, n_jobs=None,\n",
      "            oob_score=False, random_state=None, verbose=0,\n",
      "            warm_start=False)\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Best parameters:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'max_depth': 6, 'n_estimators': 110}\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Number of CrossValidation sets:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Best Score:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9857181719260065\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Let's define the parameters to be tuned\n",
    "\n",
    "parameters = {'n_estimators': np.arange(10,201,20), 'max_depth':np.arange(4,15,2)}\n",
    "\n",
    "# Let's initiate the model\n",
    "\n",
    "rfc = RandomForestClassifier()\n",
    "rfc_grid = GridSearchCV(rfc, param_grid=parameters, n_jobs=-1)\n",
    "rfc_grid_results = perform_model(rfc_grid, X_train, y_train, X_test, y_test, class_labels=labels)\n",
    "\n",
    "# Printing the best attributes of the model \n",
    "\n",
    "print_grid_search_attributes(rfc_grid_results['model'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 6. Gradient Boosted Decision Tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# Importing Libraries\n",
    "\n",
    "from sklearn.ensemble import GradientBoostingClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n",
      "  warnings.warn(CV_WARNING, FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>Training_time(HH:MM:SS.ms) - 0:00:26.123057</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>testing time(HH:MM:SS:ms) - 0:00:00.033001</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Accuracy:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9541907024092298\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Confusion Matrix:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Normalized Confusion Matrix:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Classifiction Report**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           1       0.99      0.99      0.99       496\n",
      "           2       0.98      0.97      0.98       471\n",
      "           3       1.00      0.92      0.96       420\n",
      "           4       0.91      0.88      0.89       491\n",
      "           5       0.87      0.96      0.91       532\n",
      "           6       1.00      1.00      1.00       537\n",
      "\n",
      "   micro avg       0.95      0.95      0.95      2947\n",
      "   macro avg       0.96      0.95      0.95      2947\n",
      "weighted avg       0.96      0.95      0.95      2947\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Best Estimator:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "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=140,\n",
      "              n_iter_no_change=None, presort='auto', random_state=None,\n",
      "              subsample=1.0, tol=0.0001, validation_fraction=0.1,\n",
      "              verbose=0, warm_start=False)\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Best parameters:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'max_depth': 5, 'n_estimators': 140}\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Number of CrossValidation sets:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "<span style='color:blue'>**Best Score:**</span>"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9831338411316648\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Let's define the parameters to be tuned\n",
    "\n",
    "parameters = {'n_estimators': np.arange(120,150,10), 'max_depth':np.arange(3,7,1)}\n",
    "\n",
    "# Let's initiate the model\n",
    "\n",
    "gbdt = GradientBoostingClassifier()\n",
    "gbdt_grid = GridSearchCV(gbdt, param_grid=parameters, n_jobs=-1)\n",
    "gbdt_grid_results = perform_model(gbdt_grid, X_train, y_train, X_test, y_test, class_labels=labels)\n",
    "\n",
    "# Printing the best attributes of the model \n",
    "\n",
    "print_grid_search_attributes(gbdt_grid_results['model'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Let's compare all the models together"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "                     Accuracy     Error\n",
      "                     ----------   --------\n",
      "Logistic Regression : 96.13%      3.868%\n",
      "Linear SVC          : 96.3%       3.699% \n",
      "RBF SVM classifier  : 96.57%      3.427% \n",
      "Decision Tree       : 95.15%      4.852% \n",
      "Random Forest       : 96.2%       3.8% \n",
      "GradientBoosting    : 95.42%      4.581% \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(lin_svc_grid_results['accuracy'] * 100,\\\n",
    "                                                        100-(lin_svc_grid_results['accuracy'] * 100)))\n",
    "\n",
    "print('RBF SVM classifier  : {:.04}%      {:.04}% '.format(rbf_svc_grid_results['accuracy'] * 100,\\\n",
    "                                                          100-(rbf_svc_grid_results['accuracy'] * 100)))\n",
    "\n",
    "print('Decision Tree       : {:.04}%      {:.04}% '.format(dtree_grid_results['accuracy'] * 100,\\\n",
    "                                                        100-(dtree_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    : {:.04}%      {:.04}% '.format(gbdt_grid_results['accuracy'] * 100,\\\n",
    "                                                        100-(gbdt_grid_results['accuracy'] * 100)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Conclusion"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The above table shows that Logistic Regression, Linear SVC, RBF SVM classifier & Random Forest has highest Accuracy with lowest Error value. We can use any of these three models for future predictions"
   ]
  }
 ],
 "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": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}