Higgsbason_HoTe3_C7305c_2383_batch49.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 257,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import os\n",
    "\n",
    "#for splitting \n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "#for preprocessing,metrics evaluation\n",
    "from sklearn import preprocessing, metrics, cross_validation\n",
    "from sklearn.metrics import confusion_matrix, accuracy_score, recall_score, classification_report, cohen_kappa_score\n",
    "\n",
    "#for scaling\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "#models \n",
    "#svm\n",
    "from sklearn import svm\n",
    "#Randomforest\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.ensemble import BaggingClassifier\n",
    "#Decision Tree\n",
    "from sklearn.tree import DecisionTreeClassifier \n",
    "\n",
    "#Gridsearch\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "#Visualization\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 258,
   "metadata": {},
   "outputs": [],
   "source": [
    "#change working directory\n",
    "os.chdir(\"D:/HoTes/CSC7305c_ML_Hot_2383_Batch49/Datasets\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 259,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Read train and test datasets from csv\n",
    "hb_data= pd.read_csv(\"CSE7302c_train.csv\", sep=',')\n",
    "hb_data_test= pd.read_csv(\"CSE7302c_test.csv\", sep=',', na_values=\"?\",  keep_default_na=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 260,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(68636, 30)"
      ]
     },
     "execution_count": 260,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Checking train data Dimensions\n",
    "hb_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 261,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 68636 entries, 0 to 68635\n",
      "Data columns (total 30 columns):\n",
      "class                       68636 non-null int64\n",
      "lepton_pT                   68636 non-null float64\n",
      "lepton_eta                  68636 non-null float64\n",
      "lepton_phi                  68636 non-null float64\n",
      "missing_energy_magnitude    68636 non-null float64\n",
      "missing_energy_phi          68636 non-null float64\n",
      "jet1pt                      68636 non-null float64\n",
      "jet1eta                     68636 non-null float64\n",
      "jet1phi                     68636 non-null float64\n",
      "jet1b.tag                   68636 non-null float64\n",
      "jet2pt                      68636 non-null float64\n",
      "jet2eta                     68636 non-null float64\n",
      "jet2phi                     68636 non-null float64\n",
      "jet2b.tag                   68636 non-null float64\n",
      "jet3pt                      68636 non-null float64\n",
      "jet3eta                     68636 non-null float64\n",
      "jet3phi                     68636 non-null float64\n",
      "jet3b.tag                   68636 non-null float64\n",
      "jet4pt                      68636 non-null float64\n",
      "jet4eta                     68636 non-null float64\n",
      "jet4phi                     68636 non-null float64\n",
      "jet4b.tag                   68636 non-null float64\n",
      "m_jj                        68636 non-null float64\n",
      "m_jjj                       68636 non-null float64\n",
      "m_lv                        68636 non-null float64\n",
      "m_jlv                       68636 non-null float64\n",
      "m_bb                        68636 non-null float64\n",
      "m_wbb                       68636 non-null float64\n",
      "m_wwbb                      68636 non-null float64\n",
      "id                          68636 non-null int64\n",
      "dtypes: float64(28), int64(2)\n",
      "memory usage: 15.7 MB\n"
     ]
    }
   ],
   "source": [
    "#checking train data datatypes\n",
    "hb_data.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 262,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(29414, 30)"
      ]
     },
     "execution_count": 262,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#check test data dimensions\n",
    "hb_data_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 263,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 29414 entries, 0 to 29413\n",
      "Data columns (total 30 columns):\n",
      "class                       29414 non-null int64\n",
      "lepton_pT                   29414 non-null float64\n",
      "lepton_eta                  29414 non-null float64\n",
      "lepton_phi                  29414 non-null float64\n",
      "missing_energy_magnitude    29414 non-null float64\n",
      "missing_energy_phi          29414 non-null float64\n",
      "jet1pt                      29414 non-null float64\n",
      "jet1eta                     29414 non-null float64\n",
      "jet1phi                     29414 non-null float64\n",
      "jet1b.tag                   29414 non-null float64\n",
      "jet2pt                      29414 non-null float64\n",
      "jet2eta                     29414 non-null float64\n",
      "jet2phi                     29414 non-null float64\n",
      "jet2b.tag                   29414 non-null float64\n",
      "jet3pt                      29414 non-null float64\n",
      "jet3eta                     29414 non-null float64\n",
      "jet3phi                     29414 non-null float64\n",
      "jet3b.tag                   29414 non-null float64\n",
      "jet4pt                      29414 non-null float64\n",
      "jet4eta                     29414 non-null float64\n",
      "jet4phi                     29413 non-null float64\n",
      "jet4b.tag                   29413 non-null float64\n",
      "m_jj                        29413 non-null float64\n",
      "m_jjj                       29413 non-null float64\n",
      "m_lv                        29413 non-null float64\n",
      "m_jlv                       29413 non-null float64\n",
      "m_bb                        29413 non-null float64\n",
      "m_wbb                       29413 non-null float64\n",
      "m_wwbb                      29413 non-null float64\n",
      "id                          29414 non-null int64\n",
      "dtypes: float64(28), int64(2)\n",
      "memory usage: 6.7 MB\n"
     ]
    }
   ],
   "source": [
    "#check test data dimensions\n",
    "hb_data_test.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 264,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "#changing object to float datatyes in test\n",
    "test_cols= [\"jet4phi\",\"jet4b.tag\",\"m_jj\",\"m_jjj\",\"m_lv\",\"m_jlv\",\"m_bb\",\"m_wbb\",\"m_wwbb\"]\n",
    "hb_data_test[test_cols] = hb_data_test[test_cols].astype('float64')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 265,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 29414 entries, 0 to 29413\n",
      "Data columns (total 30 columns):\n",
      "class                       29414 non-null int64\n",
      "lepton_pT                   29414 non-null float64\n",
      "lepton_eta                  29414 non-null float64\n",
      "lepton_phi                  29414 non-null float64\n",
      "missing_energy_magnitude    29414 non-null float64\n",
      "missing_energy_phi          29414 non-null float64\n",
      "jet1pt                      29414 non-null float64\n",
      "jet1eta                     29414 non-null float64\n",
      "jet1phi                     29414 non-null float64\n",
      "jet1b.tag                   29414 non-null float64\n",
      "jet2pt                      29414 non-null float64\n",
      "jet2eta                     29414 non-null float64\n",
      "jet2phi                     29414 non-null float64\n",
      "jet2b.tag                   29414 non-null float64\n",
      "jet3pt                      29414 non-null float64\n",
      "jet3eta                     29414 non-null float64\n",
      "jet3phi                     29414 non-null float64\n",
      "jet3b.tag                   29414 non-null float64\n",
      "jet4pt                      29414 non-null float64\n",
      "jet4eta                     29414 non-null float64\n",
      "jet4phi                     29413 non-null float64\n",
      "jet4b.tag                   29413 non-null float64\n",
      "m_jj                        29413 non-null float64\n",
      "m_jjj                       29413 non-null float64\n",
      "m_lv                        29413 non-null float64\n",
      "m_jlv                       29413 non-null float64\n",
      "m_bb                        29413 non-null float64\n",
      "m_wbb                       29413 non-null float64\n",
      "m_wwbb                      29413 non-null float64\n",
      "id                          29414 non-null int64\n",
      "dtypes: float64(28), int64(2)\n",
      "memory usage: 6.7 MB\n"
     ]
    }
   ],
   "source": [
    "hb_data_test.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 266,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train target modified datatype: category\n",
      "test target modified datatype: category\n"
     ]
    }
   ],
   "source": [
    "#convert target columns into category in train and test\n",
    "hb_data['class'] = hb_data['class'].astype('category')\n",
    "hb_data_test['class'] = hb_data_test['class'].astype('category')\n",
    "print(\"train target modified datatype:\",hb_data['class'].dtypes)\n",
    "print(\"test target modified datatype:\",hb_data_test['class'].dtypes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 267,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train modified dimensions: (68636, 29)\n",
      "test modified dimensions: (29414, 29)\n"
     ]
    }
   ],
   "source": [
    "#Dopping id columns\n",
    "hb_data = hb_data.drop('id',1)\n",
    "hb_data_test = hb_data_test.drop('id',1)\n",
    "print(\"train modified dimensions:\",hb_data.shape)\n",
    "print(\"test modified dimensions:\",hb_data_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 268,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "na values in Train \n",
      " class                       0\n",
      "lepton_pT                   0\n",
      "lepton_eta                  0\n",
      "lepton_phi                  0\n",
      "missing_energy_magnitude    0\n",
      "missing_energy_phi          0\n",
      "jet1pt                      0\n",
      "jet1eta                     0\n",
      "jet1phi                     0\n",
      "jet1b.tag                   0\n",
      "jet2pt                      0\n",
      "jet2eta                     0\n",
      "jet2phi                     0\n",
      "jet2b.tag                   0\n",
      "jet3pt                      0\n",
      "jet3eta                     0\n",
      "jet3phi                     0\n",
      "jet3b.tag                   0\n",
      "jet4pt                      0\n",
      "jet4eta                     0\n",
      "jet4phi                     0\n",
      "jet4b.tag                   0\n",
      "m_jj                        0\n",
      "m_jjj                       0\n",
      "m_lv                        0\n",
      "m_jlv                       0\n",
      "m_bb                        0\n",
      "m_wbb                       0\n",
      "m_wwbb                      0\n",
      "dtype: int64\n",
      "na values in Test \n",
      " class                       0\n",
      "lepton_pT                   0\n",
      "lepton_eta                  0\n",
      "lepton_phi                  0\n",
      "missing_energy_magnitude    0\n",
      "missing_energy_phi          0\n",
      "jet1pt                      0\n",
      "jet1eta                     0\n",
      "jet1phi                     0\n",
      "jet1b.tag                   0\n",
      "jet2pt                      0\n",
      "jet2eta                     0\n",
      "jet2phi                     0\n",
      "jet2b.tag                   0\n",
      "jet3pt                      0\n",
      "jet3eta                     0\n",
      "jet3phi                     0\n",
      "jet3b.tag                   0\n",
      "jet4pt                      0\n",
      "jet4eta                     0\n",
      "jet4phi                     1\n",
      "jet4b.tag                   1\n",
      "m_jj                        1\n",
      "m_jjj                       1\n",
      "m_lv                        1\n",
      "m_jlv                       1\n",
      "m_bb                        1\n",
      "m_wbb                       1\n",
      "m_wwbb                      1\n",
      "dtype: int64\n"
     ]
    }
   ],
   "source": [
    "#checking null/na values\n",
    "print(\"na values in Train \\n\", hb_data.isna().sum() )\n",
    "print(\"na values in Test \\n\", hb_data_test.isna().sum() )\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 269,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "#dropping na from test as they exists in only one row\n",
    "hb_data_test.dropna(inplace=True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 270,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(29413, 29)"
      ]
     },
     "execution_count": 270,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hb_data_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 271,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "na values in Train \n",
      " class                       0\n",
      "lepton_pT                   0\n",
      "lepton_eta                  0\n",
      "lepton_phi                  0\n",
      "missing_energy_magnitude    0\n",
      "missing_energy_phi          0\n",
      "jet1pt                      0\n",
      "jet1eta                     0\n",
      "jet1phi                     0\n",
      "jet1b.tag                   0\n",
      "jet2pt                      0\n",
      "jet2eta                     0\n",
      "jet2phi                     0\n",
      "jet2b.tag                   0\n",
      "jet3pt                      0\n",
      "jet3eta                     0\n",
      "jet3phi                     0\n",
      "jet3b.tag                   0\n",
      "jet4pt                      0\n",
      "jet4eta                     0\n",
      "jet4phi                     0\n",
      "jet4b.tag                   0\n",
      "m_jj                        0\n",
      "m_jjj                       0\n",
      "m_lv                        0\n",
      "m_jlv                       0\n",
      "m_bb                        0\n",
      "m_wbb                       0\n",
      "m_wwbb                      0\n",
      "dtype: int64\n",
      "na values in Test \n",
      " class                       0\n",
      "lepton_pT                   0\n",
      "lepton_eta                  0\n",
      "lepton_phi                  0\n",
      "missing_energy_magnitude    0\n",
      "missing_energy_phi          0\n",
      "jet1pt                      0\n",
      "jet1eta                     0\n",
      "jet1phi                     0\n",
      "jet1b.tag                   0\n",
      "jet2pt                      0\n",
      "jet2eta                     0\n",
      "jet2phi                     0\n",
      "jet2b.tag                   0\n",
      "jet3pt                      0\n",
      "jet3eta                     0\n",
      "jet3phi                     0\n",
      "jet3b.tag                   0\n",
      "jet4pt                      0\n",
      "jet4eta                     0\n",
      "jet4phi                     0\n",
      "jet4b.tag                   0\n",
      "m_jj                        0\n",
      "m_jjj                       0\n",
      "m_lv                        0\n",
      "m_jlv                       0\n",
      "m_bb                        0\n",
      "m_wbb                       0\n",
      "m_wwbb                      0\n",
      "dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print(\"na values in Train \\n\", hb_data.isna().sum() )\n",
    "print(\"na values in Test \\n\", hb_data_test.isna().sum() )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 272,
   "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>lepton_pT</th>\n",
       "      <th>lepton_eta</th>\n",
       "      <th>lepton_phi</th>\n",
       "      <th>missing_energy_magnitude</th>\n",
       "      <th>missing_energy_phi</th>\n",
       "      <th>jet1pt</th>\n",
       "      <th>jet1eta</th>\n",
       "      <th>jet1phi</th>\n",
       "      <th>jet1b.tag</th>\n",
       "      <th>jet2pt</th>\n",
       "      <th>...</th>\n",
       "      <th>jet4eta</th>\n",
       "      <th>jet4phi</th>\n",
       "      <th>jet4b.tag</th>\n",
       "      <th>m_jj</th>\n",
       "      <th>m_jjj</th>\n",
       "      <th>m_lv</th>\n",
       "      <th>m_jlv</th>\n",
       "      <th>m_bb</th>\n",
       "      <th>m_wbb</th>\n",
       "      <th>m_wwbb</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>68636.000000</td>\n",
       "      <td>68636.000000</td>\n",
       "      <td>68636.000000</td>\n",
       "      <td>68636.000000</td>\n",
       "      <td>68636.000000</td>\n",
       "      <td>68636.000000</td>\n",
       "      <td>68636.000000</td>\n",
       "      <td>68636.000000</td>\n",
       "      <td>68636.000000</td>\n",
       "      <td>68636.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>68636.000000</td>\n",
       "      <td>68636.000000</td>\n",
       "      <td>68636.000000</td>\n",
       "      <td>68636.000000</td>\n",
       "      <td>68636.000000</td>\n",
       "      <td>68636.000000</td>\n",
       "      <td>68636.000000</td>\n",
       "      <td>68636.000000</td>\n",
       "      <td>68636.000000</td>\n",
       "      <td>68636.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.989390</td>\n",
       "      <td>-0.004752</td>\n",
       "      <td>-0.000865</td>\n",
       "      <td>0.998634</td>\n",
       "      <td>-0.004615</td>\n",
       "      <td>0.988293</td>\n",
       "      <td>-0.004521</td>\n",
       "      <td>0.001059</td>\n",
       "      <td>0.997492</td>\n",
       "      <td>0.991622</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.005873</td>\n",
       "      <td>-0.001390</td>\n",
       "      <td>0.989418</td>\n",
       "      <td>1.031975</td>\n",
       "      <td>1.022349</td>\n",
       "      <td>1.050091</td>\n",
       "      <td>1.010198</td>\n",
       "      <td>0.972071</td>\n",
       "      <td>1.031742</td>\n",
       "      <td>0.959118</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.562599</td>\n",
       "      <td>1.005846</td>\n",
       "      <td>1.006407</td>\n",
       "      <td>0.597851</td>\n",
       "      <td>1.006065</td>\n",
       "      <td>0.473853</td>\n",
       "      <td>1.010098</td>\n",
       "      <td>1.007629</td>\n",
       "      <td>1.026956</td>\n",
       "      <td>0.498138</td>\n",
       "      <td>...</td>\n",
       "      <td>1.009085</td>\n",
       "      <td>1.007525</td>\n",
       "      <td>1.395692</td>\n",
       "      <td>0.648727</td>\n",
       "      <td>0.369765</td>\n",
       "      <td>0.166135</td>\n",
       "      <td>0.399576</td>\n",
       "      <td>0.521868</td>\n",
       "      <td>0.363594</td>\n",
       "      <td>0.313324</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.274697</td>\n",
       "      <td>-2.434976</td>\n",
       "      <td>-1.742508</td>\n",
       "      <td>0.001283</td>\n",
       "      <td>-1.743944</td>\n",
       "      <td>0.140434</td>\n",
       "      <td>-2.968735</td>\n",
       "      <td>-1.741237</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.189737</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.497265</td>\n",
       "      <td>-1.742691</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.110875</td>\n",
       "      <td>0.310499</td>\n",
       "      <td>0.133012</td>\n",
       "      <td>0.295983</td>\n",
       "      <td>0.048125</td>\n",
       "      <td>0.362572</td>\n",
       "      <td>0.403166</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.590204</td>\n",
       "      <td>-0.744166</td>\n",
       "      <td>-0.869156</td>\n",
       "      <td>0.578363</td>\n",
       "      <td>-0.877512</td>\n",
       "      <td>0.676977</td>\n",
       "      <td>-0.696157</td>\n",
       "      <td>-0.864770</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.656964</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.721686</td>\n",
       "      <td>-0.874392</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.791868</td>\n",
       "      <td>0.846218</td>\n",
       "      <td>0.985762</td>\n",
       "      <td>0.766271</td>\n",
       "      <td>0.671995</td>\n",
       "      <td>0.819659</td>\n",
       "      <td>0.770090</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.853738</td>\n",
       "      <td>-0.003950</td>\n",
       "      <td>0.003191</td>\n",
       "      <td>0.892574</td>\n",
       "      <td>-0.008968</td>\n",
       "      <td>0.892804</td>\n",
       "      <td>-0.002006</td>\n",
       "      <td>-0.006040</td>\n",
       "      <td>1.086538</td>\n",
       "      <td>0.889382</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.007956</td>\n",
       "      <td>-0.004703</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.896056</td>\n",
       "      <td>0.950510</td>\n",
       "      <td>0.989726</td>\n",
       "      <td>0.916662</td>\n",
       "      <td>0.873269</td>\n",
       "      <td>0.947334</td>\n",
       "      <td>0.871402</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1.234030</td>\n",
       "      <td>0.734318</td>\n",
       "      <td>0.869884</td>\n",
       "      <td>1.294878</td>\n",
       "      <td>0.867539</td>\n",
       "      <td>1.167717</td>\n",
       "      <td>0.689175</td>\n",
       "      <td>0.873302</td>\n",
       "      <td>2.173076</td>\n",
       "      <td>1.200490</td>\n",
       "      <td>...</td>\n",
       "      <td>0.709938</td>\n",
       "      <td>0.874380</td>\n",
       "      <td>3.101961</td>\n",
       "      <td>1.026683</td>\n",
       "      <td>1.083862</td>\n",
       "      <td>1.020199</td>\n",
       "      <td>1.141150</td>\n",
       "      <td>1.138059</td>\n",
       "      <td>1.138514</td>\n",
       "      <td>1.056399</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>7.805887</td>\n",
       "      <td>2.433894</td>\n",
       "      <td>1.743236</td>\n",
       "      <td>7.998711</td>\n",
       "      <td>1.743229</td>\n",
       "      <td>6.534160</td>\n",
       "      <td>2.964723</td>\n",
       "      <td>1.741454</td>\n",
       "      <td>2.173076</td>\n",
       "      <td>6.498635</td>\n",
       "      <td>...</td>\n",
       "      <td>2.498009</td>\n",
       "      <td>1.743372</td>\n",
       "      <td>3.101961</td>\n",
       "      <td>16.131908</td>\n",
       "      <td>8.940148</td>\n",
       "      <td>4.565248</td>\n",
       "      <td>7.442589</td>\n",
       "      <td>8.407887</td>\n",
       "      <td>7.318191</td>\n",
       "      <td>6.015647</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 28 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          lepton_pT    lepton_eta    lepton_phi  missing_energy_magnitude  \\\n",
       "count  68636.000000  68636.000000  68636.000000              68636.000000   \n",
       "mean       0.989390     -0.004752     -0.000865                  0.998634   \n",
       "std        0.562599      1.005846      1.006407                  0.597851   \n",
       "min        0.274697     -2.434976     -1.742508                  0.001283   \n",
       "25%        0.590204     -0.744166     -0.869156                  0.578363   \n",
       "50%        0.853738     -0.003950      0.003191                  0.892574   \n",
       "75%        1.234030      0.734318      0.869884                  1.294878   \n",
       "max        7.805887      2.433894      1.743236                  7.998711   \n",
       "\n",
       "       missing_energy_phi        jet1pt       jet1eta       jet1phi  \\\n",
       "count        68636.000000  68636.000000  68636.000000  68636.000000   \n",
       "mean            -0.004615      0.988293     -0.004521      0.001059   \n",
       "std              1.006065      0.473853      1.010098      1.007629   \n",
       "min             -1.743944      0.140434     -2.968735     -1.741237   \n",
       "25%             -0.877512      0.676977     -0.696157     -0.864770   \n",
       "50%             -0.008968      0.892804     -0.002006     -0.006040   \n",
       "75%              0.867539      1.167717      0.689175      0.873302   \n",
       "max              1.743229      6.534160      2.964723      1.741454   \n",
       "\n",
       "          jet1b.tag        jet2pt      ...            jet4eta       jet4phi  \\\n",
       "count  68636.000000  68636.000000      ...       68636.000000  68636.000000   \n",
       "mean       0.997492      0.991622      ...          -0.005873     -0.001390   \n",
       "std        1.026956      0.498138      ...           1.009085      1.007525   \n",
       "min        0.000000      0.189737      ...          -2.497265     -1.742691   \n",
       "25%        0.000000      0.656964      ...          -0.721686     -0.874392   \n",
       "50%        1.086538      0.889382      ...          -0.007956     -0.004703   \n",
       "75%        2.173076      1.200490      ...           0.709938      0.874380   \n",
       "max        2.173076      6.498635      ...           2.498009      1.743372   \n",
       "\n",
       "          jet4b.tag          m_jj         m_jjj          m_lv         m_jlv  \\\n",
       "count  68636.000000  68636.000000  68636.000000  68636.000000  68636.000000   \n",
       "mean       0.989418      1.031975      1.022349      1.050091      1.010198   \n",
       "std        1.395692      0.648727      0.369765      0.166135      0.399576   \n",
       "min        0.000000      0.110875      0.310499      0.133012      0.295983   \n",
       "25%        0.000000      0.791868      0.846218      0.985762      0.766271   \n",
       "50%        0.000000      0.896056      0.950510      0.989726      0.916662   \n",
       "75%        3.101961      1.026683      1.083862      1.020199      1.141150   \n",
       "max        3.101961     16.131908      8.940148      4.565248      7.442589   \n",
       "\n",
       "               m_bb         m_wbb        m_wwbb  \n",
       "count  68636.000000  68636.000000  68636.000000  \n",
       "mean       0.972071      1.031742      0.959118  \n",
       "std        0.521868      0.363594      0.313324  \n",
       "min        0.048125      0.362572      0.403166  \n",
       "25%        0.671995      0.819659      0.770090  \n",
       "50%        0.873269      0.947334      0.871402  \n",
       "75%        1.138059      1.138514      1.056399  \n",
       "max        8.407887      7.318191      6.015647  \n",
       "\n",
       "[8 rows x 28 columns]"
      ]
     },
     "execution_count": 272,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hb_data.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 273,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Visualization\n",
    "#heatmap for multicollinearity check\n",
    "float_columns= hb_data.select_dtypes(include='float64').columns\n",
    "cor=hb_data[float_columns].corr()\n",
    "plt.figure(figsize = (10,10))\n",
    "sns.heatmap(cor)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 274,
   "metadata": {},
   "outputs": [],
   "source": [
    "#by checking multi collinearity dropping high collinearity variables\n",
    "#m_jjj and m_wwbb are causing high multicollinearity\n",
    "#collinear_columns= ['m_jjj', 'm_wwbb','m_jlv', 'm_bb']\n",
    "#hb_data= hb_data.drop(collinear_columns, axis=1)\n",
    "#hb_data_test = hb_data_test.drop(collinear_columns, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 275,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.000000    33700\n",
       "2.173076    22387\n",
       "2.173076     5688\n",
       "1.086538     5413\n",
       "1.086538     1448\n",
       "Name: jet1b.tag, dtype: int64"
      ]
     },
     "execution_count": 275,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hb_data[\"jet1b.tag\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 276,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.000000    34289\n",
       "2.214872    22294\n",
       "2.214872     5660\n",
       "1.107436     5135\n",
       "1.107436     1258\n",
       "Name: jet2b.tag, dtype: int64"
      ]
     },
     "execution_count": 276,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hb_data[\"jet2b.tag\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 277,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.000000    39089\n",
       "2.548224    19626\n",
       "2.548224     4906\n",
       "1.274112     3952\n",
       "1.274112     1063\n",
       "Name: jet3b.tag, dtype: int64"
      ]
     },
     "execution_count": 277,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hb_data[\"jet3b.tag\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 278,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.000000    44714\n",
       "3.101961    15800\n",
       "3.101961     4063\n",
       "1.550981     3179\n",
       "1.550981      880\n",
       "Name: jet4b.tag, dtype: int64"
      ]
     },
     "execution_count": 278,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hb_data[\"jet4b.tag\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 279,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.000000    19044\n",
       "3.101961     8605\n",
       "1.550981     1764\n",
       "Name: jet4b.tag, dtype: int64"
      ]
     },
     "execution_count": 279,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hb_data_test[\"jet4b.tag\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 280,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaUAAAEyCAYAAABasc9VAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAEdVJREFUeJzt3XGMHOV9xvHnyfnSHgnhaDi18QF12kZO41I4ONFElqJA2hjSAFagKkhNS5TKUpW2IEWuoEKljVqFylKatKpSuYSSFkKowLFoS+KmApQmrd2csRMDxhKgAD5TcQQMoTkFY37949Z3ttn1zrIzO7+9+X6kU/beHdY/vZnZ5+add95xRAgAgAzeVHcBAAAcQSgBANIglAAAaRBKAIA0CCUAQBqEEgAgDUIJAJAGoQQASINQAgCksaKKDz3ttNNi1apVVXw0AGAI7dy587mImOi2XSWhtGrVKs3MzFTx0QCAIWT7ySLbMXwHAEiDUAIApEEoAQDSIJQAAGkQSgCANAglAEAahBIAIA1CCQCQRiU3zwJNt3XXrDZt26cDB+e1cnxMG9et1vqpybrLAtIjlICSbd01q+u37NH8ocOSpNmD87p+yx5JIpiALhi+A0q2adu+xUA6Yv7QYW3atq+mioDhQSgBJTtwcL6ndgBLCCWgZCvHx3pqB7CEUAJKtnHdao2NjhzTNjY6oo3rVtdUETA8mOgAlOzIZAZm3wG9I5SACqyfmiSEgDeA4TsAQBqEEgAgDUIJAJAGoQQASIOJDgCAY9S5diOhBABYVPfajQzfAQAW1b12I6EEAFhU99qNhBIAYFHdazcSSgCARXWv3chEBwDAorrXbiSUAADHqHPtxkLDd7bHbd9l+1Hbe22/r+rCAADNU/RM6fOSvh4RV9h+s6STKqwJANBQXUPJ9tskvV/S1ZIUEa9IeqXasgAATVRk+O7nJM1J+gfbu2zfbPstx29ke4PtGdszc3NzpRcKAFj+ioTSCknnSvpCRExJ+j9J1x2/UURsjojpiJiemJgouUwAQBMUCaX9kvZHxI7W73dpIaQAAChV11CKiP+V9LTtI3dOfVDSI5VWBQBopKKz7/5A0u2tmXdPSPp4dSUBAJqqUChFxG5J0xXXAgBoONa+AwCkQSgBANIglAAAaRBKAIA0CCUAQBqEEgAgDUIJAJAGoQQASINQAgCkQSgBANIglAAAaRBKAIA0CCUAQBqEEgAgDUIJAJAGoQQASINQAgCkUfRx6AO3ddesNm3bpwMH57VyfEwb163W+qnJussCAFQoZSht3TWr67fs0fyhw5Kk2YPzun7LHkkimABgGUs5fLdp277FQDpi/tBhbdq2r6aKAACDkDKUDhyc76kdALA8pAylleNjPbUDAJaHlKG0cd1qjY2OHNM2NjqijetW11QRAGAQUk50ODKZgdl3ANAsKUNJWggmQggABq/OW3LShhIAYPDqviUn5TUlAEA96r4lh1ACACyq+5actMN3LDMEAIO3cnxMs20CaFC35KQ8Uzoypjl7cF6hpTHNrbtm6y4NAJa1C9490VN72QqFku3v295je7ftmaqLqntMEwCa6v5H53pqL1svw3cXRMRzlVVylLrHNAGgqer+/k05fMcyQwBQj7q/f4uGUkj6d9s7bW9ot4HtDbZnbM/MzfV3mscyQwBQj43rVmt0xMe0jY54YN+/RUNpbUScK+liSZ+0/f7jN4iIzRExHRHTExP9XRBbPzWpy8+b1IgXOmbE1uXnscIDAAxEdPm9QoVCKSIOtP73WUlflXR+lUVt3TWru3fO6nAs9MThCN29c5bZdwBQsU3b9unQa8em0KHXIs/Ns7bfYvvkI68lfUjSQ1UWxew7DLutu2a19qb79M7r/k1rb7qPP6gwNOqe6FBk9t1PS/qqF4bSVkj6ckR8vcqi6u4UoB91rx0G9CP9zbMR8UREnN36WRMRf1F1UXXP/gD6wZk+hlndE81STgmvu1OAfnCmj2G2fmpSn/noWZocH5MlTY6P6TMfPavZj67gIX8YZnUPfwD9qvN5dilDSeIhfxheG9etPuaaksSZPlBU2lAChhVn+hh2PHkWWGY408ewqnv2aMqJDgCAetQ9e5QzJaACPKQSw6ru2aOcKQEl4yGVGGZ13ydKKAElq3v4A+hH3feJMnwHlKzu4Q+gH3XPHiWUgJJx8yyGXZ2zR9MO37HKMoZV3cMfwDBLeaZU9zx5oB91D38AwyxlKJ3oQjEHNoYBN88Cb0zK4TsuFANAM6UMpbrnyQMA6pEylLhQDAD1qXOiWcpQWj81qXPPPOWYtnPPPIUxegCoWN0rkqQMpRu27tG3H3/+mLZvP/68bti6p6aKAKAZ6l6RJGUo3bHj6Z7aAQDlqHuiWcop4YcjemoHsmGVcAyrulckSXmm9Cb31g5kUveYPNCPC9490VN72VKG0msdTog6tQOZ1D0mD/Tj/kfnemovW8pQAoZZu6GPE7UDmdS9/xJKQMlG3H6cuVM7kEmn3XRQuy+hBJSMiToYZp1200HtvilDabRDVZ3agUwmO8xS6tQOYEnKr/lDr/XWDmTCMlnAG5fyPqURu+1QB2PyGAY8Twl44wqHku0RSTOSZiPiI9WVxJg8hh/PUwLemF7OlK6RtFfS2yqqZdH42KgOzh9q2w4Mgxu27tEdO57W4QiN2LrqV87Qn68/q+6ygK5OPWlUL/zo9d+/p540mO/fQteUbJ8u6dcl3VxtOUf+vd7agUxu2LpHt21/avHM/nCEbtv+FAsKYyjceMma162e8yYvtA9C0YkOn5P0R5I6TjWwvcH2jO2Zubn+7vxtl9Inagcy+fKOp3pqBzKZefL5162e81ostA9C11Cy/RFJz0bEzhNtFxGbI2I6IqYnJgazRhKQEctkYZjdvr39H0+d2stW5ExpraRLbX9f0lckXWj7tkqrAgDUotPfToP6m6prKEXE9RFxekSsknSlpPsi4rcqrwwYUtz8DbxxHCZAyVaMtD+sOrUDWNLTzbMR8YCkByqpBFgm5jssPdKpHcCSlH+6dSoqZbEAsIx0uh90UPeJpvyer/tCGwA01ZqVJ/fUXjZCCSgZN39jmP3X4+3vR+rUXraUoQQMs7qfRwP0o+6TAkIJAJAGoQQASINQAgCkQSgBJVv78z/VUzuAJYQSULLfmD6zp3YASwgloGR/es/DPbUDWEIoASVr99TkE7UDWEIoAQDSIJQAAGkQSgCANAglAEAahBIAIA1CCQCQBqEEAEiDUAJK1umxSTxOCeiOUAJKNjba/rDq1A5gCUcJULL5Q6/11A5gCaEElGzl+FhP7QCWEEpAyVa9vX34dGoHsIRQAkq2/YkXemoHsIRQAkp2OKKndgBLCCUAQBqEEgAgDUIJAJAGoQQASINQAgCk0TWUbP+k7f+x/V3bD9v+s0EUBgBonhUFtvmxpAsj4mXbo5K+ZftrEbG94toAAA3TNZQiIiS93Pp1tPXDDRcAgNIVuqZke8T2bknPSvpGROxos80G2zO2Z+bm5squEwDQAIVCKSIOR8Q5kk6XdL7tX2qzzeaImI6I6YmJibLrBAA0QE+z7yLioKQHJF1USTUAgEYrMvtuwvZ46/WYpF+V9GjVhQEAmqfI7Lt3SPqS7REthNg/R8S/VlsWAKCJisy++56kqQHUAgBoOFZ0AACkQSgBANIglAAAaRBKAIA0CCUAQBqEEgAgDUIJAJAGoQQASINQAgCkQSgBANIglAAAaRBKAIA0CCUAQBqEEgAgDUIJAJAGoQQASINQAgCkQSgBANIglAAAaRBKAIA0CCUAQBqEEgAgDUIJAJAGoQQASINQAgCkQSgBANIglAAAaRBKAIA0CCUAQBpdQ8n2Gbbvt73X9sO2rxlEYQCA5llRYJtXJX0qIh60fbKknba/ERGPVFwbAKBhup4pRcQzEfFg6/UPJe2VNFl1YQCA5unpmpLtVZKmJO1o894G2zO2Z+bm5sqpDgDQKIVDyfZbJd0t6dqIeOn49yNic0RMR8T0xMREmTUCABqiUCjZHtVCIN0eEVuqLQkA0FRFZt9Z0hcl7Y2Iz1ZfEgCgqYqcKa2V9DFJF9re3fr5cMV1AQAaqOuU8Ij4liQPoBYAQMOxogMAIA1CCQCQBqEEAEiDUAIApEEoAQDSIJQAAGkQSgCANAglAEAahBIAIA1CCQCQBqEEAEiDUAIApEEoAQDSIJQAAGkQSgCANAglAEAahBIAIA1CCQCQBqEEAEiDUAIApEEoAQDSIJQAAGkQSgCANAglAEAahBIAIA1CCQCQBqEEAEiDUAIApEEoAQDSIJQAAGl0DSXbt9h+1vZDgygIANBcRc6UbpV0UcV1AADQPZQi4puSnh9ALQCAhivtmpLtDbZnbM/Mzc2V9bEAgAYpLZQiYnNETEfE9MTERFkfCwBoEGbfAQDSIJQAAGkUmRJ+h6T/lrTa9n7bn6i+LABAE63otkFEXDWIQgAAYPgOAJAGoQQASINQAgCkQSgBANIglAAAaRBKAIA0CCUAQBqEEgAgDUIJAJAGoQQASINQAgCkQSgBANIglAAAaRBKAIA0CCUAQBqEEgAgDUIJAJAGoQQASINQAgCkQSgBANIglAAAaRBKAIA0CCUAQBqEEgAgDUIJAJAGoQSUrNNBxcEGdMdxApQsemwHMjn1pNGe2stGKAElWzk+1lM7kMmNl6zR6IiPaRsdsW68ZM1A/n1CCSjZxnWrNTY6ckzb2OiINq5bXVNFQHHrpya16YqzNTk+JkuaHB/TpivO1vqpyYH8+yuKbGT7IkmflzQi6eaIuKnKoj73m+fo2jt3t20Hsjty8G7atk8HDs5r5fiYNq5bPbCDGujX+qnJ2vbXrqFke0TS30r6NUn7JX3H9j0R8UhVRXFQY9jVeVADw6zImdL5kh6LiCckyfZXJF0mqbJQkjioAaCJilxTmpT09FG/72+1HcP2Btsztmfm5ubKqg8A0CBFQslt2l43uzUiNkfEdERMT0xM9F8ZAKBxioTSfklnHPX76ZIOVFMOAKDJioTSdyS9y/Y7bb9Z0pWS7qm2LABAE3Wd6BARr9r+fUnbtDAl/JaIeLjyygAAjVPoPqWIuFfSvRXXAgBoOFZ0AACkQSgBANJwRPlrF9uek/RkSR93mqTnSvqspqIP+0cf9o8+7N8w9+HPRkTX+4UqCaUy2Z6JiOm66xhm9GH/6MP+0Yf9a0IfMnwHAEiDUAIApDEMobS57gKWAfqwf/Rh/+jD/i37Pkx/TQkA0BzDcKYEAGgIQgkAkEaKULJ9ke19th+zfV2b93/C9p2t93fYXjX4KnMr0IdX256zvbv187t11JmZ7VtsP2v7oQ7v2/Zft/r4e7bPHXSN2RXoww/YfvGo/fBPBl1jdrbPsH2/7b22H7Z9TZttlu2+WHsoHfW49YslvUfSVbbfc9xmn5D0QkT8gqS/kvSXg60yt4J9KEl3RsQ5rZ+bB1rkcLhV0kUneP9iSe9q/WyQ9IUB1DRsbtWJ+1CS/vOo/fDTA6hp2Lwq6VMR8YuS3ivpk22O52W7L9YeSjrqcesR8YqkI49bP9plkr7Uen2XpA/abvfwwaYq0ofoIiK+Ken5E2xymaR/jAXbJY3bfsdgqhsOBfoQXUTEMxHxYOv1DyXt1euf9r1s98UMoVTkceuL20TEq5JelPT2gVQ3HAo9sl7S5a1T/btsn9HmfZxY0X7Gib3P9ndtf832mrqLyax1qWJK0o7j3lq2+2KGUCryuPVCj2RvsCL98y+SVkXEL0v6Dy2deaI49sP+PaiFNdDOlvQ3krbWXE9att8q6W5J10bES8e/3eY/WRb7YoZQKvK49cVtbK+QdIoYIjha1z6MiB9ExI9bv/69pPMGVNtyUmRfxQlExEsR8XLr9b2SRm2fVnNZ6dge1UIg3R4RW9pssmz3xQyhVORx6/dI+p3W6ysk3Rfc9Xu0rn143HjzpVoYp0Zv7pH0262ZT++V9GJEPFN3UcPE9s8cuR5s+3wtfAf9oN6qcmn1zxcl7Y2Iz3bYbNnui4WePFulTo9bt/1pSTMRcY8W/g/6J9uPaeEM6cr6Ks6nYB/+oe1LtTCz53lJV9dWcFK275D0AUmn2d4v6UZJo5IUEX+nhacvf1jSY5J+JOnj9VSaV4E+vELS79l+VdK8pCv5A/N11kr6mKQ9tne32v5Y0pnS8t8XWWYIAJBGhuE7AAAkEUoAgEQIJQBAGoQSACANQgkAkAahBABIg1ACAKTx/yp2FFp2EKGlAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 504x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (7,5))\n",
    "\n",
    "plt.scatter(x = hb_data['jet1b.tag'],y = hb_data.jet1pt)\n",
    "#plt.title('Years at company and Age')\n",
    "#plt.xlabel('YearsAtCompany')\n",
    "#plt.ylabel('Age')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 281,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAa4AAAEyCAYAAACiUTSiAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAADglJREFUeJzt3H2MZXddx/HP191WiTQ2sRMe2q3bhIZYEalOCoQ/RCixgKEomLSJPChmo5EIgRCqTTCgJhgSTFQiWaUBTeUhBUqhJaWNGCSRyrRsa9eCNiihhYThoS2EBtj26x8zrfswszPLPdyZ38zrldxk7rlnf+eX397d95x7z73V3QGAUfzYVk8AAE6FcAEwFOECYCjCBcBQhAuAoQgXAEMRLgCGIlwADEW4ABjK3q046FlnndX79+/fikMDsE3deuutX+/uhY3225Jw7d+/P0tLS1txaAC2qar60mb281IhAEMRLgCGIlwADEW4ABiKcAEwFOECYCgzh6uqfqKq/r2qbq+qw1X15ikmBgBrmeJzXN9L8pzu/k5VnZbk01X18e7+zARjA8AxZg5Xd3eS76zePW311rOOu5Hzrrj+mINUkv956wt/1IcFYItN8h5XVe2pqkNJvpbkpu6+ZY19DlTVUlUtLS8vz3S846OVrJTyvCuun2lcALa/ScLV3Q9199OSnJPkoqp6yhr7HOzuxe5eXFjY8KuoTn68U9wOwM4x6VWF3X1fkn9JcsmU4wLAI6a4qnChqs5c/fkxSS5O8vlZxwWAtUxxVeETkrynqvZkJYQf6O6PTTAuAJxgiqsK70hy4QRzAYAN+eYMAIYiXAAMRbgAGIpwATAU4QJgKMIFwFCEC4ChCBcAQxEuAIYiXAAMRbgAGIpwATAU4QJgKMIFwFCEC4ChCBcAQxEuAIYiXAAMRbgAGIpwATAU4QJgKMIFwFCEC4ChCBcAQxEuAIYiXAAMRbgAGIpwATAU4QJgKMIFwFCEC4ChCBcAQxEuAIYyc7iqal9VfbKq7qqqw1X1mikmBgBr2TvBGEeSvL67b6uqM5LcWlU3dfd/TjA2ABxj5jOu7v5qd9+2+vO3k9yV5OxZxwWAtUz6HldV7U9yYZJb1njsQFUtVdXS8vLylIcFYBeZLFxV9dgkH0zy2u5+4PjHu/tgdy929+LCwsJUhwVgl5kkXFV1WlaidXV3f2iKMQFgLVNcVVhJ3pXkru5+++xTAoD1TXHG9awkL0vynKo6tHp7wQTjAsAJZr4cvrs/naQmmAsAbMg3ZwAwFOECYChTfHMGALvMtZ+7N2+78Qv5yn0P5olnPiZv+NUn58UXzue7J4QLgFNy7efuzes+cCgP98r9e+97MK/7wKEkmUu8hAu2wFb+tgqzeuMH73g0Wo94uFe2CxfsQNd+7t684Zrb84OHVv7l33vfg3nDNbcnmc9vqzCr7x15+JS2T83FGTBnb/7o4Uej9YgfPNR580cPb9GMYCzCBXP2re/+4JS2A8cSLgCGIlwADEW4ABiKcAEwFOECYCjCBcBQhAuAoQgXAEMRLgCGIlwADEW4ABiKcAEwFOECYCjCBcBQhAuAoQgXAEMRLgCGIlwADEW4ABiKcAEwFOECYCjCBcBQhAuAoQgXAEOZJFxVdVVVfa2q7pxiPABYz1RnXO9OcslEYwHAuiYJV3d/Ksk3pxgLAE5mbu9xVdWBqlqqqqXl5eV5HRaAHWZu4erug9292N2LCwsL8zosADuMqwoBGIpwATCUqS6Hf2+Sf0vy5Kq6p6peNcW4AHC8vVMM0t2XTzEOAGzES4UADEW4ABiKcAEwFOECYCjCBcBQhAuAoQgXAEMRLgCGIlwADEW4ABiKcAEwFOECYCjCBcBQhAuAoQgXAEMRLgCGIlwADEW4ABiKcAEwFOECYCjCBcBQhAuAoQgXAEMRLgCGIlwADEW4ABiKcAEwFOECYCjCBcBQhAuAoQgXAEMRLgCGIlwADGWScFXVJVX1haq6u6qumGJMAFjLzOGqqj1J3pHk+UkuSHJ5VV0w67gAsJYpzrguSnJ3d3+xu7+f5H1JLp1gXAA4wRThOjvJl4+6f8/qtmNU1YGqWqqqpeXl5QkOC8BuNEW4ao1tfcKG7oPdvdjdiwsLCxMcFoDdaIpw3ZNk31H3z0nylQnGBYATTBGuzyY5v6rOq6rTk1yW5LoJxgWAE+yddYDuPlJVr05yY5I9Sa7q7sMzzwwA1jBzuJKku29IcsMUYwHAyfjmDACGIlwADEW4ABiKcAEwFOECYCjCBcBQhAuAoQgXAEMRLgCGIlwADEW4ABiKcAEwFOECYCjCBcBQhAuAoQgXAEMRLgCGIlwADEW4ABiKcAEwFOECYCjCBcBQhAuAoQgXAEMRLgCGIlwADEW4ABiKcAEwFOECYCjCBcBQhAuAoQgXAEOZKVxV9ZtVdbiqHq6qxakmBQDrmfWM684kv5HkUxPMBQA2tHeWP9zddyVJVU0zGwDYwNze46qqA1W1VFVLy8vL8zosADvMhmdcVXVzksev8dCV3f2RzR6ouw8mOZgki4uLvekZAsBRNgxXd188j4kAwGa4HB6Aocx6OfyvV9U9SZ6Z5PqqunGaaQHA2ma9qvDDST480VwAYENeKgRgKMIFwFCEC4ChCBcAQxEuAIYiXAAMRbgAGIpwATAU4QJgKMIFwFCEC4ChCBcAQxEuAIYiXAAMRbgAGIpwATAU4QJgKMIFwFCEC4ChCBcAQxEuAIYiXAAMRbgAGIpwATAU4QJgKMIFc7an6pS2A8cSLpizy5++75S2w3bzW88495S2T23vXI4CPOrPXvzzSZL33vLlPNSdPVW5/On7Ht0O291WP4eru+dyoKMtLi720tLSD/3n919x/bqP/e9bX/hDjwvA1qmqW7t7caP9vFQIwFCEC4ChCBcAQ5kpXFX1tqr6fFXdUVUfrqozp5oYAKxl1jOum5I8pbufmuS/kvzR7FMCgPXNFK7u/kR3H1m9+5kk58w+JQBY35Tvcf1Oko9POB4AnGDDDyBX1c1JHr/GQ1d290dW97kyyZEkV59knANJDiTJuefO59PVAOw8G4aruy8+2eNV9Yokv5bkuX2STzN398EkB5OVDyCf4jwBIMmMX/lUVZckeWOSX+7u704zJQBY36zvcf1NkjOS3FRVh6rqnRPMCQDWNdMZV3c/aaqJAMBm+OYMAIYiXAAMRbgAGIpwATAU4QJgKMIFwFCEC4ChCBcAQxEuAIYiXAAMRbgAGIpwATAU4QJgKMIFwFCEC4ChCBcAQxEuAIYiXAAMRbgAGIpwATAU4QJgKMIFwFCEC4ChCBcAQxEuAIYiXAAMRbgAGIpwATAU4QJgKMIFwFCEC4ChCBcAQxEuAIYyU7iq6k+r6o6qOlRVn6iqJ041MQBYy6xnXG/r7qd299OSfCzJmyaYEwCsa6ZwdfcDR939ySQ923QA4OT2zjpAVf15kpcnuT/Jr5xkvwNJDiTJueeeO+thAdilNjzjqqqbq+rONW6XJkl3X9nd+5JcneTV643T3Qe7e7G7FxcWFmaa9OPOOP2UtgOwc2wYru6+uLufssbtI8ft+k9JXvKjmeaxbrnyeSdE6nFnnJ5brnzePA4PwBaa6aXCqjq/u/979e6Lknx+9iltjkgB7E6zvsf11qp6cpKHk3wpye/NPiUAWN9M4eruubw0CACP8M0ZAAxFuAAYinABMBThAmAowgXAUIQLgKFU9/y/F7eqlrPyua8pnJXk6xONtVtZw9lZw9lZw9mNvoY/090bfifgloRrSlW11N2LWz2PkVnD2VnD2VnD2e2WNfRSIQBDES4AhrITwnVwqyewA1jD2VnD2VnD2e2KNRz+PS4AdpedcMYFwC4iXAAMZZhwVdUlVfWFqrq7qq5Y4/Efr6r3rz5+S1Xtn/8st7dNrOErq2q5qg6t3n53K+a5XVXVVVX1taq6c53Hq6r+anV976iqX5z3HLe7Tazhs6vq/qOeg2+a9xy3u6raV1WfrKq7qupwVb1mjX129HNxiHBV1Z4k70jy/CQXJLm8qi44brdXJflWdz8pyV8m+Yv5znJ72+QaJsn7u/tpq7e/n+skt793J7nkJI8/P8n5q7cDSf52DnMazbtz8jVMkn896jn4ljnMaTRHkry+u382yTOS/MEa/5Z39HNxiHAluSjJ3d39xe7+fpL3Jbn0uH0uTfKe1Z+vSfLcqqo5znG728wachLd/akk3zzJLpcm+Yde8ZkkZ1bVE+YzuzFsYg3ZQHd/tbtvW/3520nuSnL2cbvt6OfiKOE6O8mXj7p/T078i3p0n+4+kuT+JD89l9mNYTNrmCQvWX1p4Zqq2jefqe0Ym11jTu6ZVXV7VX28qn5uqyezna2+JXJhkluOe2hHPxdHCddaZ07HX8e/mX12s82sz0eT7O/upya5Of9/BsvmeA7O7rasfF/dLyT56yTXbvF8tq2qemySDyZ5bXc/cPzDa/yRHfNcHCVc9yQ5+rf/c5J8Zb19qmpvkp+KlySOtuEadvc3uvt7q3f/LskvzWluO8VmnqecRHc/0N3fWf35hiSnVdVZWzytbaeqTstKtK7u7g+tscuOfi6OEq7PJjm/qs6rqtOTXJbkuuP2uS7JK1Z/fmmSf26frj7ahmt43GvgL8rKa+ds3nVJXr56Rdczktzf3V/d6kmNpKoe/8h701V1UVb+j/rG1s5qe1ldn3cluau7377Objv6ubh3qyewGd19pKpeneTGJHuSXNXdh6vqLUmWuvu6rPxF/mNV3Z2VM63Ltm7G288m1/APq+pFWblq6ZtJXrllE96Gquq9SZ6d5KyquifJnyQ5LUm6+51JbkjygiR3J/lukt/empluX5tYw5cm+f2qOpLkwSSX+QX0BM9K8rIk/1FVh1a3/XGSc5Pd8Vz0lU8ADGWUlwoBIIlwATAY4QJgKMIFwFCEC4ChCBcAQxEuAIbyf597Vx4QeFcFAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 504x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (7,5))\n",
    "\n",
    "plt.scatter(x = hb_data['jet1b.tag'],y = hb_data.jet1eta)\n",
    "#plt.title('Years at company and Age')\n",
    "#plt.xlabel('YearsAtCompany')\n",
    "#plt.ylabel('Age')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 282,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (7,5))\n",
    "\n",
    "plt.scatter(x = hb_data['jet1b.tag'],y = hb_data.jet1phi)\n",
    "#plt.title('Years at company and Age')\n",
    "#plt.xlabel('YearsAtCompany')\n",
    "#plt.ylabel('Age')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 283,
   "metadata": {},
   "outputs": [],
   "source": [
    "#sns.pairplot(hb_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 302,
   "metadata": {},
   "outputs": [],
   "source": [
    "#dropping tag variables\n",
    "tag_cols= ['jet1b.tag','jet2b.tag','jet3b.tag', 'jet4b.tag']\n",
    "hb_data= hb_data.drop(tag_cols,1)\n",
    "hb_data_test= hb_data_test.drop(tag_cols,1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 303,
   "metadata": {},
   "outputs": [
    {
     "ename": "KeyError",
     "evalue": "'class'",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
      "\u001b[1;32mC:\\ProgramData\\Anaconda3\\lib\\site-packages\\pandas\\core\\indexes\\base.py\u001b[0m in \u001b[0;36mget_loc\u001b[1;34m(self, key, method, tolerance)\u001b[0m\n\u001b[0;32m   3062\u001b[0m             \u001b[1;32mtry\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 3063\u001b[1;33m                 \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_engine\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget_loc\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   3064\u001b[0m             \u001b[1;32mexcept\u001b[0m \u001b[0mKeyError\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mpandas\\_libs\\index.pyx\u001b[0m in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[1;34m()\u001b[0m\n",
      "\u001b[1;32mpandas\\_libs\\index.pyx\u001b[0m in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[1;34m()\u001b[0m\n",
      "\u001b[1;32mpandas\\_libs\\hashtable_class_helper.pxi\u001b[0m in \u001b[0;36mpandas._libs.hashtable.PyObjectHashTable.get_item\u001b[1;34m()\u001b[0m\n",
      "\u001b[1;32mpandas\\_libs\\hashtable_class_helper.pxi\u001b[0m in \u001b[0;36mpandas._libs.hashtable.PyObjectHashTable.get_item\u001b[1;34m()\u001b[0m\n",
      "\u001b[1;31mKeyError\u001b[0m: 'class'",
      "\nDuring handling of the above exception, another exception occurred:\n",
      "\u001b[1;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-303-990393f1f906>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m()\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mtrain_target\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mhb_data\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;34m'class'\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mcopy\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      2\u001b[0m \u001b[0mtest_target\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mhb_data_test\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;34m'class'\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mcopy\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      3\u001b[0m \u001b[0mhb_data\u001b[0m\u001b[1;33m=\u001b[0m \u001b[0mhb_data\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mdrop\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'class'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0maxis\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      4\u001b[0m \u001b[0mhb_data_test\u001b[0m\u001b[1;33m=\u001b[0m \u001b[0mhb_data_test\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mdrop\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'class'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0maxis\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mC:\\ProgramData\\Anaconda3\\lib\\site-packages\\pandas\\core\\frame.py\u001b[0m in \u001b[0;36m__getitem__\u001b[1;34m(self, key)\u001b[0m\n\u001b[0;32m   2683\u001b[0m             \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_getitem_multilevel\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2684\u001b[0m         \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 2685\u001b[1;33m             \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_getitem_column\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   2686\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2687\u001b[0m     \u001b[1;32mdef\u001b[0m \u001b[0m_getitem_column\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mkey\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mC:\\ProgramData\\Anaconda3\\lib\\site-packages\\pandas\\core\\frame.py\u001b[0m in \u001b[0;36m_getitem_column\u001b[1;34m(self, key)\u001b[0m\n\u001b[0;32m   2690\u001b[0m         \u001b[1;31m# get column\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2691\u001b[0m         \u001b[1;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mcolumns\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mis_unique\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 2692\u001b[1;33m             \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_get_item_cache\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   2693\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2694\u001b[0m         \u001b[1;31m# duplicate columns & possible reduce dimensionality\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mC:\\ProgramData\\Anaconda3\\lib\\site-packages\\pandas\\core\\generic.py\u001b[0m in \u001b[0;36m_get_item_cache\u001b[1;34m(self, item)\u001b[0m\n\u001b[0;32m   2484\u001b[0m         \u001b[0mres\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mcache\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mitem\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2485\u001b[0m         \u001b[1;32mif\u001b[0m \u001b[0mres\u001b[0m \u001b[1;32mis\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 2486\u001b[1;33m             \u001b[0mvalues\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_data\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mitem\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   2487\u001b[0m             \u001b[0mres\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_box_item_values\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mitem\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mvalues\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2488\u001b[0m             \u001b[0mcache\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mitem\u001b[0m\u001b[1;33m]\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mres\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mC:\\ProgramData\\Anaconda3\\lib\\site-packages\\pandas\\core\\internals.py\u001b[0m in \u001b[0;36mget\u001b[1;34m(self, item, fastpath)\u001b[0m\n\u001b[0;32m   4113\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   4114\u001b[0m             \u001b[1;32mif\u001b[0m \u001b[1;32mnot\u001b[0m \u001b[0misna\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mitem\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 4115\u001b[1;33m                 \u001b[0mloc\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mitems\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget_loc\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mitem\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   4116\u001b[0m             \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   4117\u001b[0m                 \u001b[0mindexer\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0marange\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mlen\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mitems\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0misna\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mitems\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mC:\\ProgramData\\Anaconda3\\lib\\site-packages\\pandas\\core\\indexes\\base.py\u001b[0m in \u001b[0;36mget_loc\u001b[1;34m(self, key, method, tolerance)\u001b[0m\n\u001b[0;32m   3063\u001b[0m                 \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_engine\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget_loc\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   3064\u001b[0m             \u001b[1;32mexcept\u001b[0m \u001b[0mKeyError\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 3065\u001b[1;33m                 \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_engine\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget_loc\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_maybe_cast_indexer\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   3066\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   3067\u001b[0m         \u001b[0mindexer\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget_indexer\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mkey\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmethod\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mmethod\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mtolerance\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mtolerance\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mpandas\\_libs\\index.pyx\u001b[0m in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[1;34m()\u001b[0m\n",
      "\u001b[1;32mpandas\\_libs\\index.pyx\u001b[0m in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[1;34m()\u001b[0m\n",
      "\u001b[1;32mpandas\\_libs\\hashtable_class_helper.pxi\u001b[0m in \u001b[0;36mpandas._libs.hashtable.PyObjectHashTable.get_item\u001b[1;34m()\u001b[0m\n",
      "\u001b[1;32mpandas\\_libs\\hashtable_class_helper.pxi\u001b[0m in \u001b[0;36mpandas._libs.hashtable.PyObjectHashTable.get_item\u001b[1;34m()\u001b[0m\n",
      "\u001b[1;31mKeyError\u001b[0m: 'class'"
     ]
    }
   ],
   "source": [
    "train_target=hb_data['class'].copy()\n",
    "test_target=hb_data_test['class'].copy()\n",
    "hb_data= hb_data.drop('class', axis=1)\n",
    "hb_data_test= hb_data_test.drop('class', axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 304,
   "metadata": {},
   "outputs": [],
   "source": [
    "hb_data3= hb_data.copy()\n",
    "hb_data_test3= hb_data_test.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "'''#label encoding\n",
    "lab_enc = preprocessing.LabelEncoder()\n",
    "tar = lab_enc.fit_transform(y_train)\n",
    "tar_val = lab_enc.fit_transform(y_val)'''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 305,
   "metadata": {},
   "outputs": [],
   "source": [
    "#train test splitting\n",
    "X_train, X_val, y_train, y_val = train_test_split(hb_data3, train_target, test_size = 0.25, random_state = 42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 306,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:5: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  \"\"\"\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\pandas\\core\\frame.py:3141: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  self.loc._setitem_with_indexer((slice(None), indexer), value)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\pandas\\core\\frame.py:3113: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  self._setitem_array(key, value)\n"
     ]
    }
   ],
   "source": [
    "#standardizing data\n",
    "float_columns= hb_data3.select_dtypes(include='float64').columns\n",
    "standardizer = StandardScaler()\n",
    "standardizer.fit(X_train[float_columns])\n",
    "X_train[float_columns] = standardizer.transform(X_train[float_columns])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 307,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:4: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  after removing the cwd from sys.path.\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\pandas\\core\\frame.py:3141: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  self.loc._setitem_with_indexer((slice(None), indexer), value)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\pandas\\core\\frame.py:3113: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  self._setitem_array(key, value)\n"
     ]
    }
   ],
   "source": [
    "#standardizing data\n",
    "standardizer = StandardScaler()\n",
    "standardizer.fit(X_val[float_columns])\n",
    "X_val[float_columns] = standardizer.transform(X_val[float_columns])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 308,
   "metadata": {},
   "outputs": [],
   "source": [
    "standardizer = StandardScaler()\n",
    "standardizer.fit(hb_data_test3[float_columns])\n",
    "hb_data_test3[float_columns] = standardizer.transform(hb_data_test3[float_columns])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 309,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Dummification\n",
    "X_train= pd.get_dummies(X_train )\n",
    "X_val= pd.get_dummies(X_val)\n",
    "hb_data_test3 = pd.get_dummies(hb_data_test3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 310,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Modified Train dimensions (51477, 24)\n",
      "Modified Validation dimensions (17159, 24)\n",
      "Modified Test dimensions (29413, 24)\n"
     ]
    }
   ],
   "source": [
    "print(\"Modified Train dimensions\",X_train.shape)\n",
    "print(\"Modified Validation dimensions\",X_val.shape)\n",
    "print(\"Modified Test dimensions\",hb_data_test3.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 311,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "lepton_pT                   float64\n",
       "lepton_eta                  float64\n",
       "lepton_phi                  float64\n",
       "missing_energy_magnitude    float64\n",
       "missing_energy_phi          float64\n",
       "jet1pt                      float64\n",
       "jet1eta                     float64\n",
       "jet1phi                     float64\n",
       "jet2pt                      float64\n",
       "jet2eta                     float64\n",
       "jet2phi                     float64\n",
       "jet3pt                      float64\n",
       "jet3eta                     float64\n",
       "jet3phi                     float64\n",
       "jet4pt                      float64\n",
       "jet4eta                     float64\n",
       "jet4phi                     float64\n",
       "m_jj                        float64\n",
       "m_jjj                       float64\n",
       "m_lv                        float64\n",
       "m_jlv                       float64\n",
       "m_bb                        float64\n",
       "m_wbb                       float64\n",
       "m_wwbb                      float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 311,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 312,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "lepton_pT                   float64\n",
       "lepton_eta                  float64\n",
       "lepton_phi                  float64\n",
       "missing_energy_magnitude    float64\n",
       "missing_energy_phi          float64\n",
       "jet1pt                      float64\n",
       "jet1eta                     float64\n",
       "jet1phi                     float64\n",
       "jet2pt                      float64\n",
       "jet2eta                     float64\n",
       "jet2phi                     float64\n",
       "jet3pt                      float64\n",
       "jet3eta                     float64\n",
       "jet3phi                     float64\n",
       "jet4pt                      float64\n",
       "jet4eta                     float64\n",
       "jet4phi                     float64\n",
       "m_jj                        float64\n",
       "m_jjj                       float64\n",
       "m_lv                        float64\n",
       "m_jlv                       float64\n",
       "m_bb                        float64\n",
       "m_wbb                       float64\n",
       "m_wwbb                      float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 312,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hb_data_test3.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Model buiding\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### SVM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 324,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.6369446548944189\n",
      "Accuracy: 0.6363424441983798\n"
     ]
    }
   ],
   "source": [
    "from sklearn.svm import SVC # \"Support Vector Classifier\" \n",
    "clf_svm = SVC(kernel='linear', gamma=5) \n",
    "  \n",
    "# fitting x samples and y classes \n",
    "clf_svm.fit(X_train, y_train)\n",
    "predicted_svm = clf_svm.predict(X_train)\n",
    "preds_val_svm=clf_svm.predict(X_val)\n",
    "print ('Accuracy:', accuracy_score(y_train, predicted_svm))\n",
    "print ('Accuracy:', accuracy_score(y_val, preds_val_svm))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 325,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test Accuracy: 0.6363172746744636\n"
     ]
    }
   ],
   "source": [
    "preds_test_svm=clf_svm.predict(hb_data_test3)\n",
    "print ('Test Accuracy:', accuracy_score(test_target, preds_test_svm))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 326,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Classifion Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "          0       0.69      0.42      0.52     24286\n",
      "          1       0.62      0.83      0.71     27191\n",
      "\n",
      "avg / total       0.65      0.64      0.62     51477\n",
      "\n",
      "Val Classifion Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "          0       0.68      0.42      0.52      8071\n",
      "          1       0.62      0.83      0.71      9088\n",
      "\n",
      "avg / total       0.65      0.64      0.62     17159\n",
      "\n",
      "Test Classifion Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "          0       0.68      0.43      0.53     13866\n",
      "          1       0.62      0.82      0.70     15547\n",
      "\n",
      "avg / total       0.65      0.64      0.62     29413\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print ('Train Classifion Report:\\n', classification_report(y_train, predicted_svm))\n",
    "print ('Val Classifion Report:\\n', classification_report(y_val, preds_val_svm))\n",
    "print ('Test Classifion Report:\\n', classification_report(test_target, preds_test_svm))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 327,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Kappa: 0.25644613251850124\n",
      "Val kappa: 0.2542385309813725\n",
      "Test Kappa: 0.2553838354948952\n"
     ]
    }
   ],
   "source": [
    "print ('Train Kappa:', cohen_kappa_score(y_train, predicted_svm))\n",
    "print ('Val kappa:', cohen_kappa_score(y_val, preds_val_svm))\n",
    "print ('Test Kappa:', cohen_kappa_score(test_target, preds_test_svm))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Decison Trees"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 481,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf_entropy = DecisionTreeClassifier( \n",
    "            criterion = \"entropy\", random_state = 100, \n",
    "            max_depth = 10, min_samples_leaf = 8) \n",
    "clf_gini = DecisionTreeClassifier(criterion = \"gini\", \n",
    "            random_state = 100,max_depth=10, min_samples_leaf=8) \n",
    "  \n",
    "# Performing training \n",
    "clf_entropy.fit(X_train, y_train)\n",
    "clf_gini.fit(X_train, y_train) \n",
    "\n",
    "predicted_DT_gini = clf_gini.predict(X_train)\n",
    "predicted_DT_entropy = clf_entropy.predict(X_train)\n",
    "\n",
    "preds1_gini=clf_gini.predict(X_val)\n",
    "preds1_entropy=clf_entropy.predict(X_val)\n",
    "\n",
    "preds_test_gini=clf_gini.predict(hb_data_test3)\n",
    "preds_test_entropy = clf_entropy.predict(hb_data_test3)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 469,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy train gini: 0.7426617712764925\n",
      "Accuracy val gini: 0.6761466285914097\n",
      "Accuracy train entropy: 0.7391650640091691\n",
      "Accuracy val entropy: 0.6762049070458651\n"
     ]
    }
   ],
   "source": [
    "print ('Accuracy train gini:', accuracy_score(y_train, predicted_DT_gini))\n",
    "print ('Accuracy val gini:', accuracy_score(y_val, preds1_gini))\n",
    "\n",
    "print ('Accuracy train entropy:', accuracy_score(y_train, predicted_DT_entropy))\n",
    "print ('Accuracy val entropy:', accuracy_score(y_val, preds1_entropy))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 470,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy test gini: 0.682113351239248\n",
      "Accuracy test entropy: 0.6824533369598477\n"
     ]
    }
   ],
   "source": [
    "print ('Accuracy test gini:', accuracy_score(test_target, preds_test_gini))\n",
    "print ('Accuracy test entropy:', accuracy_score(test_target, preds_test_entropy))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 471,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Kappa train gini: 0.48327119605857116\n",
      "Kappa val gini: 0.3493739648048684\n",
      "Kappa train entropy: 0.4772004843312231\n",
      "Kappa val entropy: 0.3507210721532541\n",
      "Test Kappa_gini: 0.36190941076572314\n",
      "Test Kappa_entropy: 0.36366404461146085\n"
     ]
    }
   ],
   "source": [
    "print ('Kappa train gini:', cohen_kappa_score(y_train, predicted_DT_gini))\n",
    "print ('Kappa val gini:', cohen_kappa_score(y_val, preds1_gini))\n",
    "\n",
    "print ('Kappa train entropy:', cohen_kappa_score(y_train, predicted_DT_entropy))\n",
    "print ('Kappa val entropy:', cohen_kappa_score(y_val, preds1_entropy))\n",
    "\n",
    "print ('Test Kappa_gini:', cohen_kappa_score(test_target, preds_test_gini))\n",
    "print ('Test Kappa_entropy:', cohen_kappa_score(test_target, preds_test_entropy))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 474,
   "metadata": {
    "scrolled": false
   },
   "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>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.2822</td>\n",
       "      <td>m_bb</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.1426</td>\n",
       "      <td>m_wwbb</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.0944</td>\n",
       "      <td>m_wbb</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.0806</td>\n",
       "      <td>jet1pt</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0732</td>\n",
       "      <td>m_jlv</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.0721</td>\n",
       "      <td>m_jjj</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.0447</td>\n",
       "      <td>lepton_pT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.0315</td>\n",
       "      <td>m_jj</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.0295</td>\n",
       "      <td>missing_energy_magnitude</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.0193</td>\n",
       "      <td>jet2pt</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.0164</td>\n",
       "      <td>jet3pt</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0.0138</td>\n",
       "      <td>jet4pt</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0.0126</td>\n",
       "      <td>m_lv</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.0105</td>\n",
       "      <td>lepton_eta</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.0098</td>\n",
       "      <td>lepton_phi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0.0094</td>\n",
       "      <td>jet2eta</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0.0091</td>\n",
       "      <td>jet1eta</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>0.0076</td>\n",
       "      <td>jet4eta</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>0.0075</td>\n",
       "      <td>jet2phi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>0.0074</td>\n",
       "      <td>jet3phi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>0.0071</td>\n",
       "      <td>jet3eta</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>0.0066</td>\n",
       "      <td>jet4phi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>0.0066</td>\n",
       "      <td>jet1phi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>0.0054</td>\n",
       "      <td>missing_energy_phi</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         0                         1\n",
       "0   0.2822                      m_bb\n",
       "1   0.1426                    m_wwbb\n",
       "2   0.0944                     m_wbb\n",
       "3   0.0806                    jet1pt\n",
       "4   0.0732                     m_jlv\n",
       "5   0.0721                     m_jjj\n",
       "6   0.0447                 lepton_pT\n",
       "7   0.0315                      m_jj\n",
       "8   0.0295  missing_energy_magnitude\n",
       "9   0.0193                    jet2pt\n",
       "10  0.0164                    jet3pt\n",
       "11  0.0138                    jet4pt\n",
       "12  0.0126                      m_lv\n",
       "13  0.0105                lepton_eta\n",
       "14  0.0098                lepton_phi\n",
       "15  0.0094                   jet2eta\n",
       "16  0.0091                   jet1eta\n",
       "17  0.0076                   jet4eta\n",
       "18  0.0075                   jet2phi\n",
       "19  0.0074                   jet3phi\n",
       "20  0.0071                   jet3eta\n",
       "21  0.0066                   jet4phi\n",
       "22  0.0066                   jet1phi\n",
       "23  0.0054        missing_energy_phi"
      ]
     },
     "execution_count": 474,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "impf_DT=sorted(zip(map(lambda x: round(x, 4), clf_entropy.feature_importances_), X_train.columns), \n",
    "             reverse=True)\n",
    "impf_DT = pd.DataFrame(impf_entropy, index=None)\n",
    "impf_DT\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 494,
   "metadata": {},
   "outputs": [],
   "source": [
    "impf1_DT = impf_DT[1]\n",
    "impf1_DT=impf1_DT[0:7]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 503,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf_entropy1 = DecisionTreeClassifier( \n",
    "            criterion = \"entropy\", random_state = 10, \n",
    "            max_depth = 10, min_samples_leaf = 8) \n",
    "clf_gini1 = DecisionTreeClassifier(criterion = \"gini\", \n",
    "            random_state = 10,max_depth=10, min_samples_leaf=8) \n",
    "  \n",
    "# Performing training \n",
    "clf_entropy1.fit(X_train[impf1_DT], y_train)\n",
    "clf_gini1.fit(X_train[impf1_DT], y_train) \n",
    "\n",
    "predicted_DT_gini1 = clf_gini1.predict(X_train[impf1_DT])\n",
    "predicted_DT_entropy1 = clf_entropy1.predict(X_train[impf1_DT])\n",
    "\n",
    "preds1_gini1=clf_gini1.predict(X_val[impf1_DT])\n",
    "preds1_entropy1=clf_entropy1.predict(X_val[impf1_DT])\n",
    "\n",
    "preds_test_gini1=clf_gini1.predict(hb_data_test3[impf1_DT])\n",
    "preds_test_entropy1 = clf_entropy1.predict(hb_data_test3[impf1_DT])\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 504,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy train gini: 0.7351049983487771\n",
      "Accuracy val gini: 0.6771956407716068\n",
      "Accuracy train entropy: 0.7301901820230394\n",
      "Accuracy val entropy: 0.6782446529518037\n"
     ]
    }
   ],
   "source": [
    "print ('Accuracy train gini:', accuracy_score(y_train, predicted_DT_gini1))\n",
    "print ('Accuracy val gini:', accuracy_score(y_val, preds1_gini1))\n",
    "\n",
    "print ('Accuracy train entropy:', accuracy_score(y_train, predicted_DT_entropy1))\n",
    "print ('Accuracy val entropy:', accuracy_score(y_val, preds1_entropy1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 505,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy test gini: 0.682147349811308\n",
      "Accuracy test entropy: 0.6862951756026247\n"
     ]
    }
   ],
   "source": [
    "print ('Accuracy test gini:', accuracy_score(test_target, preds_test_gini1))\n",
    "print ('Accuracy test entropy:', accuracy_score(test_target, preds_test_entropy1))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 506,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Kappa train gini: 0.46808450283638936\n",
      "Kappa val gini: 0.35185925102410476\n",
      "Kappa train entropy: 0.4580161873415587\n",
      "Kappa val entropy: 0.35368759563756436\n",
      "Test Kappa_gini: 0.36197019020651255\n",
      "Test Kappa_entropy: 0.369858691551472\n"
     ]
    }
   ],
   "source": [
    "print ('Kappa train gini:', cohen_kappa_score(y_train, predicted_DT_gini1))\n",
    "print ('Kappa val gini:', cohen_kappa_score(y_val, preds1_gini1))\n",
    "\n",
    "print ('Kappa train entropy:', cohen_kappa_score(y_train, predicted_DT_entropy1))\n",
    "print ('Kappa val entropy:', cohen_kappa_score(y_val, preds1_entropy1))\n",
    "\n",
    "print ('Test Kappa_gini:', cohen_kappa_score(test_target, preds_test_gini1))\n",
    "print ('Test Kappa_entropy:', cohen_kappa_score(test_target, preds_test_entropy1))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Random Forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 313,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(bootstrap=True, class_weight=None, criterion='entropy',\n",
       "            max_depth=5, max_features=18, 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=60, n_jobs=10,\n",
       "            oob_score=False, random_state=1, verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 313,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Random Forest\n",
    "np.random.seed(12345)\n",
    "clf = RandomForestClassifier(n_estimators=60, criterion='entropy',n_jobs=10, \n",
    "                             random_state=1,max_depth=5, max_features=18,bootstrap=True)\n",
    "\n",
    "# Train the Classifier to take the training features and learn how they relate\n",
    "# to the training y (the species)\n",
    "clf.fit(X_train, y_train)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 314,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Accuracy: 0.688365677875556\n",
      "Val Accuracy: 0.6831983215805116\n"
     ]
    }
   ],
   "source": [
    "predicted_rf = clf.predict(X_train)\n",
    "preds1=clf.predict(X_val)\n",
    "print ('Train Accuracy:', accuracy_score(y_train, predicted_rf))\n",
    "print ('Val Accuracy:', accuracy_score(y_val, preds1))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 315,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test Accuracy: 0.6802774283480094\n"
     ]
    }
   ],
   "source": [
    "preds_test=clf.predict(hb_data_test3)\n",
    "print ('Test Accuracy:', accuracy_score(test_target, preds_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 316,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Classifion Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "          0       0.67      0.68      0.67     24286\n",
      "          1       0.71      0.70      0.70     27191\n",
      "\n",
      "avg / total       0.69      0.69      0.69     51477\n",
      "\n",
      "Val Classifion Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "          0       0.66      0.67      0.67      8071\n",
      "          1       0.70      0.69      0.70      9088\n",
      "\n",
      "avg / total       0.68      0.68      0.68     17159\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print ('Train Classifion Report:\\n', classification_report(y_train, predicted_rf))\n",
    "print ('Val Classifion Report:\\n', classification_report(y_val, preds1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 317,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test Classifion Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "          0       0.66      0.67      0.67     13866\n",
      "          1       0.70      0.69      0.69     15547\n",
      "\n",
      "avg / total       0.68      0.68      0.68     29413\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print ('Test Classifion Report:\\n', classification_report(test_target, preds_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 318,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Kappa: 0.3753281752042569\n",
      "Val kappa: 0.3649291187893111\n"
     ]
    }
   ],
   "source": [
    "print ('Train Kappa:', cohen_kappa_score(y_train, predicted_rf))\n",
    "print ('Val kappa:', cohen_kappa_score(y_val, preds1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 319,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test Kappa: 0.359333647712449\n"
     ]
    }
   ],
   "source": [
    "print ('Test Kappa:', cohen_kappa_score(test_target, preds_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 329,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.4409</td>\n",
       "      <td>m_bb</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.1662</td>\n",
       "      <td>m_wwbb</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.1364</td>\n",
       "      <td>m_wbb</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.0780</td>\n",
       "      <td>m_jjj</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0695</td>\n",
       "      <td>jet1pt</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.0555</td>\n",
       "      <td>m_jlv</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.0225</td>\n",
       "      <td>missing_energy_magnitude</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.0105</td>\n",
       "      <td>lepton_pT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.0104</td>\n",
       "      <td>jet2pt</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.0056</td>\n",
       "      <td>m_jj</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.0011</td>\n",
       "      <td>jet4pt</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0.0006</td>\n",
       "      <td>m_lv</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0.0004</td>\n",
       "      <td>jet1eta</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.0003</td>\n",
       "      <td>jet4eta</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.0003</td>\n",
       "      <td>jet3pt</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0.0003</td>\n",
       "      <td>jet3eta</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0.0003</td>\n",
       "      <td>jet2phi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>0.0003</td>\n",
       "      <td>jet2eta</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>0.0003</td>\n",
       "      <td>jet1phi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>0.0002</td>\n",
       "      <td>missing_energy_phi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>0.0002</td>\n",
       "      <td>lepton_phi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>0.0001</td>\n",
       "      <td>lepton_eta</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>0.0001</td>\n",
       "      <td>jet4phi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>0.0001</td>\n",
       "      <td>jet3phi</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         0                         1\n",
       "0   0.4409                      m_bb\n",
       "1   0.1662                    m_wwbb\n",
       "2   0.1364                     m_wbb\n",
       "3   0.0780                     m_jjj\n",
       "4   0.0695                    jet1pt\n",
       "5   0.0555                     m_jlv\n",
       "6   0.0225  missing_energy_magnitude\n",
       "7   0.0105                 lepton_pT\n",
       "8   0.0104                    jet2pt\n",
       "9   0.0056                      m_jj\n",
       "10  0.0011                    jet4pt\n",
       "11  0.0006                      m_lv\n",
       "12  0.0004                   jet1eta\n",
       "13  0.0003                   jet4eta\n",
       "14  0.0003                    jet3pt\n",
       "15  0.0003                   jet3eta\n",
       "16  0.0003                   jet2phi\n",
       "17  0.0003                   jet2eta\n",
       "18  0.0003                   jet1phi\n",
       "19  0.0002        missing_energy_phi\n",
       "20  0.0002                lepton_phi\n",
       "21  0.0001                lepton_eta\n",
       "22  0.0001                   jet4phi\n",
       "23  0.0001                   jet3phi"
      ]
     },
     "execution_count": 329,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "impf=sorted(zip(map(lambda x: round(x, 4), clf.feature_importances_), X_train.columns), \n",
    "             reverse=True)\n",
    "impf = pd.DataFrame(impf, index=None)\n",
    "impf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 431,
   "metadata": {},
   "outputs": [],
   "source": [
    "impf1 = impf[1]\n",
    "impf1=impf1[0:18]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 448,
   "metadata": {},
   "outputs": [],
   "source": [
    "np.random.seed(11111)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 456,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
       "            max_depth=10, max_features=12, max_leaf_nodes=None,\n",
       "            min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "            min_samples_leaf=4, min_samples_split=2,\n",
       "            min_weight_fraction_leaf=0.0, n_estimators=100, n_jobs=10,\n",
       "            oob_score=True, random_state=3, verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 456,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create a random forest Classifier. By convention, clf means 'Classifier'\n",
    "clf1 = RandomForestClassifier(n_estimators=100, criterion='gini',n_jobs=10, \n",
    "                             random_state=3,max_depth=10, max_features=12,bootstrap=True, \n",
    "                             min_samples_leaf=4,oob_score=True)\n",
    "\n",
    "# Train the Classifier to take the training features and learn how they relate\n",
    "# to the training y (the species)\n",
    "\n",
    "clf1.fit(X_train[impf1], y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 457,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Accuracy: 0.688365677875556\n",
      "Val Accuracy: 0.6831983215805116\n"
     ]
    }
   ],
   "source": [
    "predicted_rf1 = clf1.predict(X_train[impf1])\n",
    "preds2=clf1.predict(X_val[impf1])\n",
    "print ('Train Accuracy:', accuracy_score(y_train, predicted_rf))\n",
    "print ('Val Accuracy:', accuracy_score(y_val, preds1))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 458,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test Accuracy: 0.710264168904906\n"
     ]
    }
   ],
   "source": [
    "preds_test1=clf1.predict(hb_data_test3[impf1])\n",
    "print ('Test Accuracy:', accuracy_score(test_target, preds_test1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 459,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Classifion Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "          0       0.78      0.77      0.77     24286\n",
      "          1       0.79      0.81      0.80     27191\n",
      "\n",
      "avg / total       0.79      0.79      0.79     51477\n",
      "\n",
      "Val Classifion Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "          0       0.70      0.69      0.69      8071\n",
      "          1       0.73      0.73      0.73      9088\n",
      "\n",
      "avg / total       0.71      0.71      0.71     17159\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print ('Train Classifion Report:\\n', classification_report(y_train, predicted_rf1))\n",
    "print ('Val Classifion Report:\\n', classification_report(y_val, preds2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 460,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test Classifion Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "          0       0.69      0.69      0.69     13866\n",
      "          1       0.73      0.73      0.73     15547\n",
      "\n",
      "avg / total       0.71      0.71      0.71     29413\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print ('Test Classifion Report:\\n', classification_report(test_target, preds_test1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 461,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Kappa: 0.5757861597372083\n",
      "Val kappa: 0.42644260585849403\n"
     ]
    }
   ],
   "source": [
    "print ('Train Kappa:', cohen_kappa_score(y_train, predicted_rf1))\n",
    "print ('Val kappa:', cohen_kappa_score(y_val, preds2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 462,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test Kappa: 0.4186702015643772\n"
     ]
    }
   ],
   "source": [
    "print ('Test Kappa:', cohen_kappa_score(test_target, preds_test1))"
   ]
  }
 ],
 "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}