Skip to content

Instantly share code, notes, and snippets.

@pronojitsaha

pronojitsaha/AnalyticsVIdhya Hackathon 3.X

Created September 17, 2015 17:05
Show Gist options
  • Save pronojitsaha/61a6e3755c41441bd3c5 to your computer and use it in GitHub Desktop.
Save pronojitsaha/61a6e3755c41441bd3c5 to your computer and use it in GitHub Desktop.
Download ZIP
Raw
AnalyticsVIdhya Hackathon 3.X
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#AnalyticsVidhya Hackathon3.0X "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#[Predict customer worth for Happy Customer Bank ](http://datahack.analyticsvidhya.com/contest/datahackathon-3x)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#1. Background "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### The Problem"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Happy Customer Bank is a mid-sized private bank which deals in all kinds of loans. They have presence across all major cities in India and focus on lending products. They have a digital arm which sources customers from the internet.\n",
"\n",
"Digital arms of banks today face challenges with lead conversion, they source leads through mediums like search, display, email campaigns and via affiliate partners. Here Happy Customer Bank faces same challenge of low conversion ratio. They have given a problem to identify the customers segments having higher conversion ratio for a specific loan product so that they can specifically target these customers, here they have provided a partial data set for salaried customers only from the last 3 months. They also capture basic details about customers like gender, DOB, existing EMI, employer Name, Loan Amount Required, Monthly Income, City, Interaction data and many others."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### The Data"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We have train and test data set, train data set has both input and output variable(s). Need to predict probability of disbursal for test data set.\n",
"\n",
"The data was provided in two files as follows: \n",
"1. train - Training set. 3 months of data, in no particular order. 87,020 rows. It has 31 features.\n",
"\n",
"2. test - Test set. 3 months of data. 37,717 rows. It has 29 features. \n",
"\n",
"3. sampleSubmission.csv - Sample submission file in the correct format.\n",
"\n",
"**Input Variables:**\n",
"\n",
"ID -\tUnique ID (can not be used for predictions) \n",
"Gender-\tSex \n",
"City -\tCurrent City \n",
"Monthly_Income - Monthly Income in rupees \n",
"DOB -\tDate of Birth \n",
"Lead_Creation_Date -\tLead Created on date \n",
"Loan_Amount_Applied -\tLoan Amount Requested (INR) \n",
"Loan_Tenure_Applied -\tLoan Tenure Requested (in years) \n",
"Existing_EMI -\tEMI of Existing Loans (INR) \n",
"Employer_Name - Employer Name \n",
"Salary_Account- Salary account with Bank \n",
"Mobile_Verified - Mobile Verified (Y/N) \n",
"Var5- Continuous classified variable \n",
"Var1- Categorical variable with multiple levels \n",
"Loan_Amount_Submitted- Loan Amount Revised and Selected after seeing Eligibility \n",
"Loan_Tenure_Submitted-\tLoan Tenure Revised and Selected after seeing Eligibility (Years) \n",
"Interest_Rate-\tInterest Rate of Submitted Loan Amount \n",
"Processing_Fee- Processing Fee of Submitted Loan Amount (INR) \n",
"EMI_Loan_Submitted- EMI of Submitted Loan Amount (INR) \n",
"Filled_Form- Filled Application form post quote \n",
"Device_Type- Device from which application was made (Browser/ Mobile) \n",
"Var2- Categorical Variable with multiple Levels \n",
"Source-\tCategorical Variable with multiple Levels \n",
"Var4-\tCategorical Variable with multiple Levels \n",
"\n",
"\n",
"**Output Variables:**\n",
"\n",
"LoggedIn-\tApplication Logged (Variable for understanding the problem – cannot be used in prediction) \n",
"Disbursed-\tLoan Disbursed (Target Variable) "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Evaluation Metric"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Evaluation metrics of this challenge is ROC_AUC. To read more detail about ROC_AUC refer this article “[Model Evaluation Metrics](http://www.analyticsvidhya.com/blog/2015/01/model-perform-part-2/)”."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# 2. Data Exploration"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"#Import Libraries and Data\n",
"\n",
"from sklearn.ensemble import RandomForestClassifier\n",
"from sklearn.ensemble import GradientBoostingClassifier\n",
"import pandas as pd\n",
"import numpy as np\n",
"from sklearn import preprocessing\n",
"from sklearn import ensemble\n",
"from sklearn.metrics import roc_curve, auc\n",
"import pandas as pd\n",
"from ggplot import *\n",
"\n",
"from datetime import datetime, timedelta\n",
"from sklearn.grid_search import GridSearchCV\n",
"import xgboost as xgb\n",
"import cPickle\n",
"import matplotlib.pyplot as plt\n",
"%matplotlib inline"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>ID</th>\n",
" <th>Gender</th>\n",
" <th>City</th>\n",
" <th>Monthly_Income</th>\n",
" <th>DOB</th>\n",
" <th>Lead_Creation_Date</th>\n",
" <th>Loan_Amount_Applied</th>\n",
" <th>Loan_Tenure_Applied</th>\n",
" <th>Existing_EMI</th>\n",
" <th>Employer_Name</th>\n",
" <th>...</th>\n",
" <th>Interest_Rate</th>\n",
" <th>Processing_Fee</th>\n",
" <th>EMI_Loan_Submitted</th>\n",
" <th>Filled_Form</th>\n",
" <th>Device_Type</th>\n",
" <th>Var2</th>\n",
" <th>Source</th>\n",
" <th>Var4</th>\n",
" <th>LoggedIn</th>\n",
" <th>Disbursed</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>ID000002C20</td>\n",
" <td>Female</td>\n",
" <td>Delhi</td>\n",
" <td>20000</td>\n",
" <td>1978-05-23</td>\n",
" <td>2015-05-15</td>\n",
" <td>300000</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>CYBOSOL</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>G</td>\n",
" <td>S122</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>ID000004E40</td>\n",
" <td>Male</td>\n",
" <td>Mumbai</td>\n",
" <td>35000</td>\n",
" <td>1985-10-07</td>\n",
" <td>2015-05-04</td>\n",
" <td>200000</td>\n",
" <td>2</td>\n",
" <td>0</td>\n",
" <td>TATA CONSULTANCY SERVICES LTD (TCS)</td>\n",
" <td>...</td>\n",
" <td>13.25</td>\n",
" <td>NaN</td>\n",
" <td>6762.9</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>G</td>\n",
" <td>S122</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>ID000007H20</td>\n",
" <td>Male</td>\n",
" <td>Panchkula</td>\n",
" <td>22500</td>\n",
" <td>1981-10-10</td>\n",
" <td>2015-05-19</td>\n",
" <td>600000</td>\n",
" <td>4</td>\n",
" <td>0</td>\n",
" <td>ALCHEMIST HOSPITALS LTD</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S143</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>ID000008I30</td>\n",
" <td>Male</td>\n",
" <td>Saharsa</td>\n",
" <td>35000</td>\n",
" <td>1987-11-30</td>\n",
" <td>2015-05-09</td>\n",
" <td>1000000</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>BIHAR GOVERNMENT</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S143</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>ID000009J40</td>\n",
" <td>Male</td>\n",
" <td>Bengaluru</td>\n",
" <td>100000</td>\n",
" <td>1984-02-17</td>\n",
" <td>2015-05-20</td>\n",
" <td>500000</td>\n",
" <td>2</td>\n",
" <td>25000</td>\n",
" <td>GLOBAL EDGE SOFTWARE</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S134</td>\n",
" <td>3</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>5 rows × 26 columns</p>\n",
"</div>"
],
"text/plain": [
" ID Gender City Monthly_Income DOB \\\n",
"0 ID000002C20 Female Delhi 20000 1978-05-23 \n",
"1 ID000004E40 Male Mumbai 35000 1985-10-07 \n",
"2 ID000007H20 Male Panchkula 22500 1981-10-10 \n",
"3 ID000008I30 Male Saharsa 35000 1987-11-30 \n",
"4 ID000009J40 Male Bengaluru 100000 1984-02-17 \n",
"\n",
" Lead_Creation_Date Loan_Amount_Applied Loan_Tenure_Applied Existing_EMI \\\n",
"0 2015-05-15 300000 5 0 \n",
"1 2015-05-04 200000 2 0 \n",
"2 2015-05-19 600000 4 0 \n",
"3 2015-05-09 1000000 5 0 \n",
"4 2015-05-20 500000 2 25000 \n",
"\n",
" Employer_Name ... Interest_Rate Processing_Fee \\\n",
"0 CYBOSOL ... NaN NaN \n",
"1 TATA CONSULTANCY SERVICES LTD (TCS) ... 13.25 NaN \n",
"2 ALCHEMIST HOSPITALS LTD ... NaN NaN \n",
"3 BIHAR GOVERNMENT ... NaN NaN \n",
"4 GLOBAL EDGE SOFTWARE ... NaN NaN \n",
"\n",
" EMI_Loan_Submitted Filled_Form Device_Type Var2 Source Var4 LoggedIn \\\n",
"0 NaN N Web-browser G S122 1 0 \n",
"1 6762.9 N Web-browser G S122 3 0 \n",
"2 NaN N Web-browser B S143 1 0 \n",
"3 NaN N Web-browser B S143 3 0 \n",
"4 NaN N Web-browser B S134 3 1 \n",
"\n",
" Disbursed \n",
"0 0 \n",
"1 0 \n",
"2 0 \n",
"3 0 \n",
"4 0 \n",
"\n",
"[5 rows x 26 columns]"
]
},
"execution_count": 2,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"train=pd.read_csv('Train.csv', parse_dates = [4,5])\n",
"train.head()"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Monthly_Income</th>\n",
" <th>Loan_Amount_Applied</th>\n",
" <th>Loan_Tenure_Applied</th>\n",
" <th>Existing_EMI</th>\n",
" <th>Var5</th>\n",
" <th>Loan_Amount_Submitted</th>\n",
" <th>Loan_Tenure_Submitted</th>\n",
" <th>Interest_Rate</th>\n",
" <th>Processing_Fee</th>\n",
" <th>EMI_Loan_Submitted</th>\n",
" <th>Var4</th>\n",
" <th>LoggedIn</th>\n",
" <th>Disbursed</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>count</th>\n",
" <td>8.702000e+04</td>\n",
" <td>86949.000000</td>\n",
" <td>86949.000000</td>\n",
" <td>86949.000000</td>\n",
" <td>87020.000000</td>\n",
" <td>52407.000000</td>\n",
" <td>52407.000000</td>\n",
" <td>27726.000000</td>\n",
" <td>27420.000000</td>\n",
" <td>27726.000000</td>\n",
" <td>87020.000000</td>\n",
" <td>87020.000000</td>\n",
" <td>87020.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>mean</th>\n",
" <td>5.884997e+04</td>\n",
" <td>230250.699928</td>\n",
" <td>2.131399</td>\n",
" <td>3696.227824</td>\n",
" <td>4.961503</td>\n",
" <td>395010.590188</td>\n",
" <td>3.891369</td>\n",
" <td>19.197474</td>\n",
" <td>5131.150839</td>\n",
" <td>10999.528377</td>\n",
" <td>2.949805</td>\n",
" <td>0.029350</td>\n",
" <td>0.014629</td>\n",
" </tr>\n",
" <tr>\n",
" <th>std</th>\n",
" <td>2.177511e+06</td>\n",
" <td>354206.759468</td>\n",
" <td>2.014193</td>\n",
" <td>39810.211920</td>\n",
" <td>5.670385</td>\n",
" <td>308248.136255</td>\n",
" <td>1.165359</td>\n",
" <td>5.834213</td>\n",
" <td>4725.837644</td>\n",
" <td>7512.323050</td>\n",
" <td>1.697720</td>\n",
" <td>0.168785</td>\n",
" <td>0.120062</td>\n",
" </tr>\n",
" <tr>\n",
" <th>min</th>\n",
" <td>0.000000e+00</td>\n",
" <td>0.000000</td>\n",
" <td>0.000000</td>\n",
" <td>0.000000</td>\n",
" <td>0.000000</td>\n",
" <td>50000.000000</td>\n",
" <td>1.000000</td>\n",
" <td>11.990000</td>\n",
" <td>200.000000</td>\n",
" <td>1176.410000</td>\n",
" <td>0.000000</td>\n",
" <td>0.000000</td>\n",
" <td>0.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>25%</th>\n",
" <td>1.650000e+04</td>\n",
" <td>0.000000</td>\n",
" <td>0.000000</td>\n",
" <td>0.000000</td>\n",
" <td>0.000000</td>\n",
" <td>200000.000000</td>\n",
" <td>3.000000</td>\n",
" <td>15.250000</td>\n",
" <td>2000.000000</td>\n",
" <td>6491.600000</td>\n",
" <td>1.000000</td>\n",
" <td>0.000000</td>\n",
" <td>0.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>50%</th>\n",
" <td>2.500000e+04</td>\n",
" <td>100000.000000</td>\n",
" <td>2.000000</td>\n",
" <td>0.000000</td>\n",
" <td>2.000000</td>\n",
" <td>300000.000000</td>\n",
" <td>4.000000</td>\n",
" <td>18.000000</td>\n",
" <td>4000.000000</td>\n",
" <td>9392.970000</td>\n",
" <td>3.000000</td>\n",
" <td>0.000000</td>\n",
" <td>0.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>75%</th>\n",
" <td>4.000000e+04</td>\n",
" <td>300000.000000</td>\n",
" <td>4.000000</td>\n",
" <td>3500.000000</td>\n",
" <td>11.000000</td>\n",
" <td>500000.000000</td>\n",
" <td>5.000000</td>\n",
" <td>20.000000</td>\n",
" <td>6250.000000</td>\n",
" <td>12919.040000</td>\n",
" <td>5.000000</td>\n",
" <td>0.000000</td>\n",
" <td>0.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>max</th>\n",
" <td>4.445544e+08</td>\n",
" <td>10000000.000000</td>\n",
" <td>10.000000</td>\n",
" <td>10000000.000000</td>\n",
" <td>18.000000</td>\n",
" <td>3000000.000000</td>\n",
" <td>6.000000</td>\n",
" <td>37.000000</td>\n",
" <td>50000.000000</td>\n",
" <td>144748.280000</td>\n",
" <td>7.000000</td>\n",
" <td>1.000000</td>\n",
" <td>1.000000</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Monthly_Income Loan_Amount_Applied Loan_Tenure_Applied \\\n",
"count 8.702000e+04 86949.000000 86949.000000 \n",
"mean 5.884997e+04 230250.699928 2.131399 \n",
"std 2.177511e+06 354206.759468 2.014193 \n",
"min 0.000000e+00 0.000000 0.000000 \n",
"25% 1.650000e+04 0.000000 0.000000 \n",
"50% 2.500000e+04 100000.000000 2.000000 \n",
"75% 4.000000e+04 300000.000000 4.000000 \n",
"max 4.445544e+08 10000000.000000 10.000000 \n",
"\n",
" Existing_EMI Var5 Loan_Amount_Submitted \\\n",
"count 86949.000000 87020.000000 52407.000000 \n",
"mean 3696.227824 4.961503 395010.590188 \n",
"std 39810.211920 5.670385 308248.136255 \n",
"min 0.000000 0.000000 50000.000000 \n",
"25% 0.000000 0.000000 200000.000000 \n",
"50% 0.000000 2.000000 300000.000000 \n",
"75% 3500.000000 11.000000 500000.000000 \n",
"max 10000000.000000 18.000000 3000000.000000 \n",
"\n",
" Loan_Tenure_Submitted Interest_Rate Processing_Fee \\\n",
"count 52407.000000 27726.000000 27420.000000 \n",
"mean 3.891369 19.197474 5131.150839 \n",
"std 1.165359 5.834213 4725.837644 \n",
"min 1.000000 11.990000 200.000000 \n",
"25% 3.000000 15.250000 2000.000000 \n",
"50% 4.000000 18.000000 4000.000000 \n",
"75% 5.000000 20.000000 6250.000000 \n",
"max 6.000000 37.000000 50000.000000 \n",
"\n",
" EMI_Loan_Submitted Var4 LoggedIn Disbursed \n",
"count 27726.000000 87020.000000 87020.000000 87020.000000 \n",
"mean 10999.528377 2.949805 0.029350 0.014629 \n",
"std 7512.323050 1.697720 0.168785 0.120062 \n",
"min 1176.410000 0.000000 0.000000 0.000000 \n",
"25% 6491.600000 1.000000 0.000000 0.000000 \n",
"50% 9392.970000 3.000000 0.000000 0.000000 \n",
"75% 12919.040000 5.000000 0.000000 0.000000 \n",
"max 144748.280000 7.000000 1.000000 1.000000 "
]
},
"execution_count": 3,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"train.describe()"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"Index([u'ID', u'Gender', u'City', u'Monthly_Income', u'DOB',\n",
" u'Lead_Creation_Date', u'Loan_Amount_Applied', u'Loan_Tenure_Applied',\n",
" u'Existing_EMI', u'Employer_Name', u'Salary_Account',\n",
" u'Mobile_Verified', u'Var5', u'Var1', u'Loan_Amount_Submitted',\n",
" u'Loan_Tenure_Submitted', u'Interest_Rate', u'Processing_Fee',\n",
" u'EMI_Loan_Submitted', u'Filled_Form', u'Device_Type', u'Var2',\n",
" u'Source', u'Var4', u'LoggedIn', u'Disbursed'],\n",
" dtype='object')"
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"train.columns"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"ID False\n",
"Gender False\n",
"City True\n",
"Monthly_Income False\n",
"DOB False\n",
"Lead_Creation_Date False\n",
"Loan_Amount_Applied True\n",
"Loan_Tenure_Applied True\n",
"Existing_EMI True\n",
"Employer_Name True\n",
"Salary_Account True\n",
"Mobile_Verified False\n",
"Var5 False\n",
"Var1 False\n",
"Loan_Amount_Submitted True\n",
"Loan_Tenure_Submitted True\n",
"Interest_Rate True\n",
"Processing_Fee True\n",
"EMI_Loan_Submitted True\n",
"Filled_Form False\n",
"Device_Type False\n",
"Var2 False\n",
"Source False\n",
"Var4 False\n",
"LoggedIn False\n",
"Disbursed False\n",
"dtype: bool"
]
},
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"pd.isnull(train).any() #check for features having missing values"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>ID</th>\n",
" <th>Gender</th>\n",
" <th>City</th>\n",
" <th>Monthly_Income</th>\n",
" <th>DOB</th>\n",
" <th>Lead_Creation_Date</th>\n",
" <th>Loan_Amount_Applied</th>\n",
" <th>Loan_Tenure_Applied</th>\n",
" <th>Existing_EMI</th>\n",
" <th>Employer_Name</th>\n",
" <th>...</th>\n",
" <th>Loan_Amount_Submitted</th>\n",
" <th>Loan_Tenure_Submitted</th>\n",
" <th>Interest_Rate</th>\n",
" <th>Processing_Fee</th>\n",
" <th>EMI_Loan_Submitted</th>\n",
" <th>Filled_Form</th>\n",
" <th>Device_Type</th>\n",
" <th>Var2</th>\n",
" <th>Source</th>\n",
" <th>Var4</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>ID000026A10</td>\n",
" <td>Male</td>\n",
" <td>Dehradun</td>\n",
" <td>21500</td>\n",
" <td>1987-04-03</td>\n",
" <td>2015-05-05</td>\n",
" <td>100000</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>APTARA INC</td>\n",
" <td>...</td>\n",
" <td>100000</td>\n",
" <td>3</td>\n",
" <td>20</td>\n",
" <td>1000</td>\n",
" <td>2649.39</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S122</td>\n",
" <td>3</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>ID000054C40</td>\n",
" <td>Male</td>\n",
" <td>Mumbai</td>\n",
" <td>42000</td>\n",
" <td>1980-05-12</td>\n",
" <td>2015-05-01</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>ATUL LTD</td>\n",
" <td>...</td>\n",
" <td>690000</td>\n",
" <td>5</td>\n",
" <td>24</td>\n",
" <td>13800</td>\n",
" <td>19849.90</td>\n",
" <td>Y</td>\n",
" <td>Mobile</td>\n",
" <td>C</td>\n",
" <td>S133</td>\n",
" <td>5</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>ID000066O10</td>\n",
" <td>Female</td>\n",
" <td>Jaipur</td>\n",
" <td>10000</td>\n",
" <td>1989-09-19</td>\n",
" <td>2015-05-01</td>\n",
" <td>300000</td>\n",
" <td>2</td>\n",
" <td>0</td>\n",
" <td>SHAREKHAN PVT LTD</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S133</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>ID000110G00</td>\n",
" <td>Female</td>\n",
" <td>Chennai</td>\n",
" <td>14650</td>\n",
" <td>1991-08-15</td>\n",
" <td>2015-05-01</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>MAERSK GLOBAL SERVICE CENTRES</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Mobile</td>\n",
" <td>C</td>\n",
" <td>S133</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>ID000113J30</td>\n",
" <td>Male</td>\n",
" <td>Chennai</td>\n",
" <td>23400</td>\n",
" <td>1987-07-22</td>\n",
" <td>2015-05-01</td>\n",
" <td>100000</td>\n",
" <td>1</td>\n",
" <td>5000</td>\n",
" <td>SCHAWK</td>\n",
" <td>...</td>\n",
" <td>100000</td>\n",
" <td>2</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S143</td>\n",
" <td>1</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>5 rows × 24 columns</p>\n",
"</div>"
],
"text/plain": [
" ID Gender City Monthly_Income DOB \\\n",
"0 ID000026A10 Male Dehradun 21500 1987-04-03 \n",
"1 ID000054C40 Male Mumbai 42000 1980-05-12 \n",
"2 ID000066O10 Female Jaipur 10000 1989-09-19 \n",
"3 ID000110G00 Female Chennai 14650 1991-08-15 \n",
"4 ID000113J30 Male Chennai 23400 1987-07-22 \n",
"\n",
" Lead_Creation_Date Loan_Amount_Applied Loan_Tenure_Applied Existing_EMI \\\n",
"0 2015-05-05 100000 3 0 \n",
"1 2015-05-01 0 0 0 \n",
"2 2015-05-01 300000 2 0 \n",
"3 2015-05-01 0 0 0 \n",
"4 2015-05-01 100000 1 5000 \n",
"\n",
" Employer_Name ... Loan_Amount_Submitted \\\n",
"0 APTARA INC ... 100000 \n",
"1 ATUL LTD ... 690000 \n",
"2 SHAREKHAN PVT LTD ... NaN \n",
"3 MAERSK GLOBAL SERVICE CENTRES ... NaN \n",
"4 SCHAWK ... 100000 \n",
"\n",
" Loan_Tenure_Submitted Interest_Rate Processing_Fee EMI_Loan_Submitted \\\n",
"0 3 20 1000 2649.39 \n",
"1 5 24 13800 19849.90 \n",
"2 NaN NaN NaN NaN \n",
"3 NaN NaN NaN NaN \n",
"4 2 NaN NaN NaN \n",
"\n",
" Filled_Form Device_Type Var2 Source Var4 \n",
"0 N Web-browser B S122 3 \n",
"1 Y Mobile C S133 5 \n",
"2 N Web-browser B S133 1 \n",
"3 N Mobile C S133 1 \n",
"4 N Web-browser B S143 1 \n",
"\n",
"[5 rows x 24 columns]"
]
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"test=pd.read_csv('Test.csv', parse_dates = [4,5]) #read in the test data set\n",
"test.head()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# 3. Data Munging"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#####Outliers\n",
"\n",
"A bit of due-diligence reveals that there are only 24 data points whose monthly income is more than Rs 1,00,00,000. I drilled down further and found there are 84 data points only with monthly income more than Rs 10,00,000. Now what's absurd is that their loan amount is less than their monthly income. So I decided to treat these as outliers (albeit a crude way) and replaced them by NaN for each of these 84 data points. "
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>ID</th>\n",
" <th>Gender</th>\n",
" <th>City</th>\n",
" <th>Monthly_Income</th>\n",
" <th>DOB</th>\n",
" <th>Lead_Creation_Date</th>\n",
" <th>Loan_Amount_Applied</th>\n",
" <th>Loan_Tenure_Applied</th>\n",
" <th>Existing_EMI</th>\n",
" <th>Employer_Name</th>\n",
" <th>...</th>\n",
" <th>Interest_Rate</th>\n",
" <th>Processing_Fee</th>\n",
" <th>EMI_Loan_Submitted</th>\n",
" <th>Filled_Form</th>\n",
" <th>Device_Type</th>\n",
" <th>Var2</th>\n",
" <th>Source</th>\n",
" <th>Var4</th>\n",
" <th>LoggedIn</th>\n",
" <th>Disbursed</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>2506</th>\n",
" <td>ID003596I10</td>\n",
" <td>Female</td>\n",
" <td>Gurgaon</td>\n",
" <td>3000000</td>\n",
" <td>1972-06-06</td>\n",
" <td>2015-05-04</td>\n",
" <td>1500000</td>\n",
" <td>5</td>\n",
" <td>30000</td>\n",
" <td>MAKE MYTRIP INDIA PVT LTD (MAKEMYTRIP.COM)</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>E</td>\n",
" <td>S133</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3757</th>\n",
" <td>ID005397P20</td>\n",
" <td>Male</td>\n",
" <td>Pune</td>\n",
" <td>1200000</td>\n",
" <td>1981-12-19</td>\n",
" <td>2015-05-05</td>\n",
" <td>300000</td>\n",
" <td>2</td>\n",
" <td>28000</td>\n",
" <td>ZS ASSOCIATES INDIA PVT LTD</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S134</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4316</th>\n",
" <td>ID006170I00</td>\n",
" <td>Male</td>\n",
" <td>Baramulla</td>\n",
" <td>54954545</td>\n",
" <td>1986-06-28</td>\n",
" <td>2015-05-06</td>\n",
" <td>1000000</td>\n",
" <td>5</td>\n",
" <td>150000</td>\n",
" <td>514949848</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S159</td>\n",
" <td>2</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4320</th>\n",
" <td>ID006174M40</td>\n",
" <td>Male</td>\n",
" <td>Bengaluru</td>\n",
" <td>1200000</td>\n",
" <td>1982-02-02</td>\n",
" <td>2015-05-06</td>\n",
" <td>300000</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>HEALTHFORE TECHNOLOGIES LTD</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S122</td>\n",
" <td>2</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>5625</th>\n",
" <td>ID008020M00</td>\n",
" <td>Male</td>\n",
" <td>Delhi</td>\n",
" <td>2000000</td>\n",
" <td>1985-01-01</td>\n",
" <td>2015-05-07</td>\n",
" <td>0</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>SPREADTRUM COMMUNICATION INDIA PVT. LTD</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S127</td>\n",
" <td>2</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>5760</th>\n",
" <td>ID008211V10</td>\n",
" <td>Male</td>\n",
" <td>Gautam Buddha Nagar</td>\n",
" <td>1800000</td>\n",
" <td>1976-07-01</td>\n",
" <td>2015-05-07</td>\n",
" <td>400000</td>\n",
" <td>3</td>\n",
" <td>13700</td>\n",
" <td>HONDA CARS INDIA LTD</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S143</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>5928</th>\n",
" <td>ID008452C20</td>\n",
" <td>Male</td>\n",
" <td>Mumbai</td>\n",
" <td>10000000</td>\n",
" <td>1982-12-06</td>\n",
" <td>2015-05-07</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>ABC</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Mobile</td>\n",
" <td>F</td>\n",
" <td>S122</td>\n",
" <td>2</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>6616</th>\n",
" <td>ID009435X00</td>\n",
" <td>Male</td>\n",
" <td>Bengaluru</td>\n",
" <td>1100000</td>\n",
" <td>1984-09-21</td>\n",
" <td>2015-05-08</td>\n",
" <td>2500000</td>\n",
" <td>5</td>\n",
" <td>55000</td>\n",
" <td>BSR AND COMPANY</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S133</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>8673</th>\n",
" <td>ID012321X10</td>\n",
" <td>Male</td>\n",
" <td>Hyderabad</td>\n",
" <td>10000000</td>\n",
" <td>1986-11-09</td>\n",
" <td>2015-05-11</td>\n",
" <td>200000</td>\n",
" <td>3</td>\n",
" <td>50000</td>\n",
" <td>DEEP HOSPITAL</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S122</td>\n",
" <td>2</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>8933</th>\n",
" <td>ID012688A30</td>\n",
" <td>Male</td>\n",
" <td>Bengaluru</td>\n",
" <td>1200000</td>\n",
" <td>1987-08-02</td>\n",
" <td>2015-05-11</td>\n",
" <td>1000000</td>\n",
" <td>5</td>\n",
" <td>25000</td>\n",
" <td>ACCENTURE SERVICES PVT LTD</td>\n",
" <td>...</td>\n",
" <td>13.99</td>\n",
" <td>2500</td>\n",
" <td>23263.07</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S122</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>10960</th>\n",
" <td>ID015572Y20</td>\n",
" <td>Male</td>\n",
" <td>Chennai</td>\n",
" <td>2301000</td>\n",
" <td>1979-02-14</td>\n",
" <td>2015-05-13</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>FLEXIBLE STEEL LACING COMPANY P LTD</td>\n",
" <td>...</td>\n",
" <td>16.00</td>\n",
" <td>12000</td>\n",
" <td>42188.44</td>\n",
" <td>Y</td>\n",
" <td>Mobile</td>\n",
" <td>C</td>\n",
" <td>S133</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>11047</th>\n",
" <td>ID015699V40</td>\n",
" <td>Male</td>\n",
" <td>Sibsagar</td>\n",
" <td>13000000</td>\n",
" <td>2061-06-01</td>\n",
" <td>2015-05-13</td>\n",
" <td>1000000</td>\n",
" <td>5</td>\n",
" <td>30000</td>\n",
" <td>OIL &amp; NATURAL GAS CORP LTD (ONGC)</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S133</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>11299</th>\n",
" <td>ID016080M00</td>\n",
" <td>Male</td>\n",
" <td>Hyderabad</td>\n",
" <td>2000000</td>\n",
" <td>1978-07-23</td>\n",
" <td>2015-05-14</td>\n",
" <td>500000</td>\n",
" <td>1</td>\n",
" <td>50000</td>\n",
" <td>MOBILEIRON INDIA SOFTWARE PVT LTD</td>\n",
" <td>...</td>\n",
" <td>14.85</td>\n",
" <td>10000</td>\n",
" <td>45093.77</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S133</td>\n",
" <td>4</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>13777</th>\n",
" <td>ID019643N30</td>\n",
" <td>Male</td>\n",
" <td>Ahmedabad</td>\n",
" <td>12000000</td>\n",
" <td>1967-11-11</td>\n",
" <td>2015-05-16</td>\n",
" <td>1100000</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>CJFJF</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S133</td>\n",
" <td>2</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>14643</th>\n",
" <td>ID020903Z30</td>\n",
" <td>Male</td>\n",
" <td>Mumbai</td>\n",
" <td>12000000</td>\n",
" <td>1992-05-08</td>\n",
" <td>2015-05-17</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>TREEG SECURITY PRVT LTD</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Mobile</td>\n",
" <td>C</td>\n",
" <td>S133</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>15872</th>\n",
" <td>ID022660O00</td>\n",
" <td>Male</td>\n",
" <td>Bagalkote</td>\n",
" <td>1100000</td>\n",
" <td>1975-07-10</td>\n",
" <td>2015-05-19</td>\n",
" <td>2000000</td>\n",
" <td>4</td>\n",
" <td>38000</td>\n",
" <td>SRV POLYTECHNIC</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S133</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>17693</th>\n",
" <td>ID025342S20</td>\n",
" <td>Female</td>\n",
" <td>Bengaluru</td>\n",
" <td>5040000</td>\n",
" <td>1980-04-22</td>\n",
" <td>2015-05-21</td>\n",
" <td>500000</td>\n",
" <td>5</td>\n",
" <td>12200</td>\n",
" <td>FIRST ADVANTAGE OFFSHORE SERVICES PVT LTD (ZAP...</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S133</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>18553</th>\n",
" <td>ID026631H10</td>\n",
" <td>Male</td>\n",
" <td>Mumbai</td>\n",
" <td>20000000</td>\n",
" <td>1976-06-14</td>\n",
" <td>2015-05-22</td>\n",
" <td>0</td>\n",
" <td>5</td>\n",
" <td>10000</td>\n",
" <td>CHANNEL TECHNOLOGIES PVT LTD</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S133</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>22492</th>\n",
" <td>ID032214A40</td>\n",
" <td>Male</td>\n",
" <td>Bengaluru</td>\n",
" <td>1600000</td>\n",
" <td>1978-05-21</td>\n",
" <td>2015-05-26</td>\n",
" <td>100000</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>NARAYANA SINGH</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S122</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>24356</th>\n",
" <td>ID034909R40</td>\n",
" <td>Male</td>\n",
" <td>Delhi</td>\n",
" <td>3636366</td>\n",
" <td>1988-02-02</td>\n",
" <td>2015-05-28</td>\n",
" <td>300000</td>\n",
" <td>2</td>\n",
" <td>0</td>\n",
" <td>GOOD EARTH ECO FUTURES PVT LTD</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S122</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>25533</th>\n",
" <td>ID036559D40</td>\n",
" <td>Male</td>\n",
" <td>Delhi</td>\n",
" <td>120100132</td>\n",
" <td>1975-03-13</td>\n",
" <td>2015-05-29</td>\n",
" <td>500000</td>\n",
" <td>0</td>\n",
" <td>5454365</td>\n",
" <td>54543543535</td>\n",
" <td>...</td>\n",
" <td>15.50</td>\n",
" <td>5000</td>\n",
" <td>12026.60</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S153</td>\n",
" <td>4</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>25853</th>\n",
" <td>ID037036M10</td>\n",
" <td>Male</td>\n",
" <td>Delhi</td>\n",
" <td>1800000</td>\n",
" <td>2058-03-31</td>\n",
" <td>2015-05-30</td>\n",
" <td>500000</td>\n",
" <td>5</td>\n",
" <td>12000</td>\n",
" <td>DELHI UNIVERSITY</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S133</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>26017</th>\n",
" <td>ID037284A40</td>\n",
" <td>Male</td>\n",
" <td>Pune</td>\n",
" <td>2500000</td>\n",
" <td>1988-07-31</td>\n",
" <td>2015-05-30</td>\n",
" <td>2500000</td>\n",
" <td>4</td>\n",
" <td>0</td>\n",
" <td>VIIKING VENTURES PVT LTD</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S162</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>26103</th>\n",
" <td>ID037390C00</td>\n",
" <td>Female</td>\n",
" <td>Visakhapatnam</td>\n",
" <td>20000000</td>\n",
" <td>1986-01-31</td>\n",
" <td>2015-05-30</td>\n",
" <td>200000</td>\n",
" <td>2</td>\n",
" <td>0</td>\n",
" <td>VARA RATNAMALA ILLA</td>\n",
" <td>...</td>\n",
" <td>20.00</td>\n",
" <td>2000</td>\n",
" <td>10179.16</td>\n",
" <td>Y</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S159</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>26494</th>\n",
" <td>ID037940G00</td>\n",
" <td>Male</td>\n",
" <td>Mumbai</td>\n",
" <td>3600000</td>\n",
" <td>1980-01-10</td>\n",
" <td>2015-05-31</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>TATA</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Mobile</td>\n",
" <td>C</td>\n",
" <td>S133</td>\n",
" <td>2</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>26739</th>\n",
" <td>ID038307J20</td>\n",
" <td>Male</td>\n",
" <td>Mumbai</td>\n",
" <td>5300020</td>\n",
" <td>1972-09-02</td>\n",
" <td>2015-05-31</td>\n",
" <td>100000</td>\n",
" <td>3</td>\n",
" <td>20000</td>\n",
" <td>JET AIRWAYS INDIA LTD</td>\n",
" <td>...</td>\n",
" <td>15.50</td>\n",
" <td>1000</td>\n",
" <td>3491.07</td>\n",
" <td>Y</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S133</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>28026</th>\n",
" <td>ID040122E20</td>\n",
" <td>Male</td>\n",
" <td>Delhi</td>\n",
" <td>100000000</td>\n",
" <td>1967-02-01</td>\n",
" <td>2015-06-02</td>\n",
" <td>100000</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>DASFASD</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S157</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>28104</th>\n",
" <td>ID040241T10</td>\n",
" <td>Male</td>\n",
" <td>Delhi</td>\n",
" <td>10000000</td>\n",
" <td>1965-01-01</td>\n",
" <td>2015-06-02</td>\n",
" <td>100000</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>KDLFAJOIDSJ</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S157</td>\n",
" <td>2</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>28118</th>\n",
" <td>ID040262O20</td>\n",
" <td>Male</td>\n",
" <td>Delhi</td>\n",
" <td>100000000</td>\n",
" <td>1965-01-01</td>\n",
" <td>2015-06-02</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>UYUHKJHKJ</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S157</td>\n",
" <td>2</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>28369</th>\n",
" <td>ID040642E20</td>\n",
" <td>Female</td>\n",
" <td>Pune</td>\n",
" <td>8595656</td>\n",
" <td>1988-06-13</td>\n",
" <td>2015-06-02</td>\n",
" <td>100000</td>\n",
" <td>1</td>\n",
" <td>56566</td>\n",
" <td>FGJHGJHG</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S157</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>...</th>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>49780</th>\n",
" <td>ID071345B00</td>\n",
" <td>Male</td>\n",
" <td>Mumbai</td>\n",
" <td>20000000</td>\n",
" <td>1979-10-12</td>\n",
" <td>2015-06-26</td>\n",
" <td>1000000</td>\n",
" <td>0</td>\n",
" <td>200000</td>\n",
" <td>NILESH TELI</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S133</td>\n",
" <td>2</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>50200</th>\n",
" <td>ID071956O10</td>\n",
" <td>Male</td>\n",
" <td>Gurgaon</td>\n",
" <td>50000000</td>\n",
" <td>1976-02-14</td>\n",
" <td>2015-06-26</td>\n",
" <td>500000</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>CHOICE CLOTHING CO PVT LTD</td>\n",
" <td>...</td>\n",
" <td>14.85</td>\n",
" <td>10000</td>\n",
" <td>11855.63</td>\n",
" <td>Y</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S127</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>50324</th>\n",
" <td>ID072143T30</td>\n",
" <td>Male</td>\n",
" <td>Delhi</td>\n",
" <td>4500000</td>\n",
" <td>1978-03-04</td>\n",
" <td>2015-06-26</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>JKBANK</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Mobile</td>\n",
" <td>C</td>\n",
" <td>S133</td>\n",
" <td>2</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>51208</th>\n",
" <td>ID073464O40</td>\n",
" <td>Male</td>\n",
" <td>Delhi</td>\n",
" <td>1350000</td>\n",
" <td>1988-10-10</td>\n",
" <td>2015-06-27</td>\n",
" <td>300000</td>\n",
" <td>5</td>\n",
" <td>8000</td>\n",
" <td>NAGARRO SOFTWARE PVT LTD</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S133</td>\n",
" <td>2</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>51643</th>\n",
" <td>ID074088O30</td>\n",
" <td>Male</td>\n",
" <td>Kolkata</td>\n",
" <td>1200000</td>\n",
" <td>1983-12-21</td>\n",
" <td>2015-06-28</td>\n",
" <td>200000</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>IBM GLOBAL SERVICES INDIA LTD</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>B</td>\n",
" <td>S133</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>52987</th>\n",
" <td>ID076036M10</td>\n",
" <td>Male</td>\n",
" <td>Bengaluru</td>\n",
" <td>5000000</td>\n",
" <td>1987-11-18</td>\n",
" <td>2015-06-29</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>ACCENTURE SERVICES PVT LTD</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Mobile</td>\n",
" <td>C</td>\n",
" <td>S122</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>53141</th>\n",
" <td>ID076246O10</td>\n",
" <td>Male</td>\n",
" <td>Delhi</td>\n",
" <td>444554443</td>\n",
" <td>1988-11-11</td>\n",
" <td>2015-06-30</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>GOOGLE</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Mobile</td>\n",
" <td>C</td>\n",
" <td>S122</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>56379</th>\n",
" <td>ID080915D00</td>\n",
" <td>Male</td>\n",
" <td>Delhi</td>\n",
" <td>1100000</td>\n",
" <td>1991-11-22</td>\n",
" <td>2015-07-03</td>\n",
" <td>650000</td>\n",
" <td>4</td>\n",
" <td>15000</td>\n",
" <td>SK</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>G</td>\n",
" <td>S122</td>\n",
" <td>2</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>58255</th>\n",
" <td>ID083578O30</td>\n",
" <td>Male</td>\n",
" <td>Hoshiarpur</td>\n",
" <td>2000000</td>\n",
" <td>1992-05-21</td>\n",
" <td>2015-07-03</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>200000</td>\n",
" <td>IZHAR NABI ZOUR</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>G</td>\n",
" <td>S122</td>\n",
" <td>2</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>58918</th>\n",
" <td>ID084523X30</td>\n",
" <td>Female</td>\n",
" <td>Aurangabad</td>\n",
" <td>1500000</td>\n",
" <td>1980-02-01</td>\n",
" <td>2015-07-04</td>\n",
" <td>1000000</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>LUPIN</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>G</td>\n",
" <td>S122</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>63345</th>\n",
" <td>ID090880K00</td>\n",
" <td>Male</td>\n",
" <td>Mumbai</td>\n",
" <td>1350000</td>\n",
" <td>1981-06-12</td>\n",
" <td>2015-07-09</td>\n",
" <td>300000</td>\n",
" <td>4</td>\n",
" <td>25000</td>\n",
" <td>ICICI BANK LTD</td>\n",
" <td>...</td>\n",
" <td>16.00</td>\n",
" <td>6000</td>\n",
" <td>8502.08</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>G</td>\n",
" <td>S122</td>\n",
" <td>4</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>68002</th>\n",
" <td>ID097622S20</td>\n",
" <td>Male</td>\n",
" <td>Delhi</td>\n",
" <td>100000000</td>\n",
" <td>1986-09-09</td>\n",
" <td>2015-07-15</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>11500</td>\n",
" <td>GHK DEVELOPMENT CONSULTANTS INDIA PVT LTD</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>G</td>\n",
" <td>S122</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>69856</th>\n",
" <td>ID100287F20</td>\n",
" <td>Male</td>\n",
" <td>Bengaluru</td>\n",
" <td>1300000</td>\n",
" <td>1981-09-26</td>\n",
" <td>2015-07-16</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>EMULEX COMMUNICATIONS PVT LTD</td>\n",
" <td>...</td>\n",
" <td>11.99</td>\n",
" <td>10000</td>\n",
" <td>26328.93</td>\n",
" <td>Y</td>\n",
" <td>Mobile</td>\n",
" <td>G</td>\n",
" <td>S122</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>70972</th>\n",
" <td>ID101907N20</td>\n",
" <td>Male</td>\n",
" <td>Hisar</td>\n",
" <td>1800000</td>\n",
" <td>1990-03-04</td>\n",
" <td>2015-07-18</td>\n",
" <td>200000</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>RAMRUP</td>\n",
" <td>...</td>\n",
" <td>22.00</td>\n",
" <td>4000</td>\n",
" <td>6301.22</td>\n",
" <td>Y</td>\n",
" <td>Web-browser</td>\n",
" <td>G</td>\n",
" <td>S122</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>72114</th>\n",
" <td>ID103535D00</td>\n",
" <td>Male</td>\n",
" <td>Kolkata</td>\n",
" <td>2632400</td>\n",
" <td>1985-08-03</td>\n",
" <td>2015-07-19</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>INDUSIND BANK LTD</td>\n",
" <td>...</td>\n",
" <td>11.99</td>\n",
" <td>999</td>\n",
" <td>39493.39</td>\n",
" <td>Y</td>\n",
" <td>Mobile</td>\n",
" <td>G</td>\n",
" <td>S122</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>73352</th>\n",
" <td>ID105286M10</td>\n",
" <td>Female</td>\n",
" <td>Kolkata</td>\n",
" <td>5300000</td>\n",
" <td>1982-10-07</td>\n",
" <td>2015-07-20</td>\n",
" <td>200000</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>DANIELI INDIA LTD</td>\n",
" <td>...</td>\n",
" <td>15.25</td>\n",
" <td>4000</td>\n",
" <td>4784.27</td>\n",
" <td>Y</td>\n",
" <td>Web-browser</td>\n",
" <td>G</td>\n",
" <td>S122</td>\n",
" <td>5</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>73394</th>\n",
" <td>ID105346U10</td>\n",
" <td>Male</td>\n",
" <td>Bengaluru</td>\n",
" <td>4684000</td>\n",
" <td>1988-07-15</td>\n",
" <td>2015-07-20</td>\n",
" <td>1000000</td>\n",
" <td>5</td>\n",
" <td>17536</td>\n",
" <td>OCWEN FINANCIAL SOLUTIONS PVT LTD</td>\n",
" <td>...</td>\n",
" <td>13.99</td>\n",
" <td>5000</td>\n",
" <td>23263.07</td>\n",
" <td>Y</td>\n",
" <td>Web-browser</td>\n",
" <td>G</td>\n",
" <td>S122</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>73938</th>\n",
" <td>ID106123R30</td>\n",
" <td>Male</td>\n",
" <td>Bengaluru</td>\n",
" <td>1960000</td>\n",
" <td>1987-12-11</td>\n",
" <td>2015-07-21</td>\n",
" <td>500000</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>MICROSOFT INDIA R AND D PVT LTD</td>\n",
" <td>...</td>\n",
" <td>13.50</td>\n",
" <td>6250</td>\n",
" <td>16967.64</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>G</td>\n",
" <td>S122</td>\n",
" <td>4</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>78225</th>\n",
" <td>ID112183T30</td>\n",
" <td>Male</td>\n",
" <td>Durgapur</td>\n",
" <td>6500000</td>\n",
" <td>1979-05-05</td>\n",
" <td>2015-07-23</td>\n",
" <td>1000000</td>\n",
" <td>5</td>\n",
" <td>150000</td>\n",
" <td>IOB</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>G</td>\n",
" <td>S122</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>78578</th>\n",
" <td>ID112691H10</td>\n",
" <td>Male</td>\n",
" <td>Bengaluru</td>\n",
" <td>1200000</td>\n",
" <td>1979-06-19</td>\n",
" <td>2015-07-24</td>\n",
" <td>300000</td>\n",
" <td>2</td>\n",
" <td>8000</td>\n",
" <td>COGNIZANT INDIA PVT LTD</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>G</td>\n",
" <td>S122</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>79089</th>\n",
" <td>ID113457T20</td>\n",
" <td>Male</td>\n",
" <td>Mumbai</td>\n",
" <td>34000000</td>\n",
" <td>1970-01-01</td>\n",
" <td>2015-07-24</td>\n",
" <td>1000000</td>\n",
" <td>5</td>\n",
" <td>10000000</td>\n",
" <td>JAISEN JAYANT GAIKWAD</td>\n",
" <td>...</td>\n",
" <td>13.99</td>\n",
" <td>5000</td>\n",
" <td>23263.07</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>G</td>\n",
" <td>S122</td>\n",
" <td>4</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>79260</th>\n",
" <td>ID113692U20</td>\n",
" <td>Male</td>\n",
" <td>Thane</td>\n",
" <td>13500000</td>\n",
" <td>1974-07-12</td>\n",
" <td>2015-07-25</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>ACG INSPECTION SYSTEMS PVT LTD</td>\n",
" <td>...</td>\n",
" <td>11.99</td>\n",
" <td>15000</td>\n",
" <td>33359.09</td>\n",
" <td>Y</td>\n",
" <td>Mobile</td>\n",
" <td>G</td>\n",
" <td>S122</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>82681</th>\n",
" <td>ID118606U10</td>\n",
" <td>Male</td>\n",
" <td>Hyderabad</td>\n",
" <td>1200000</td>\n",
" <td>1982-05-05</td>\n",
" <td>2015-07-28</td>\n",
" <td>500000</td>\n",
" <td>2</td>\n",
" <td>0</td>\n",
" <td>JONES LANG LASALLE</td>\n",
" <td>...</td>\n",
" <td>15.25</td>\n",
" <td>10000</td>\n",
" <td>24302.76</td>\n",
" <td>Y</td>\n",
" <td>Web-browser</td>\n",
" <td>G</td>\n",
" <td>S122</td>\n",
" <td>5</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>83459</th>\n",
" <td>ID119712I20</td>\n",
" <td>Male</td>\n",
" <td>Hyderabad</td>\n",
" <td>1833333</td>\n",
" <td>1985-01-31</td>\n",
" <td>2015-07-29</td>\n",
" <td>500000</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>KANTAR OPERATIONS / TNS INDIA PVT LTD</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>G</td>\n",
" <td>S122</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>84923</th>\n",
" <td>ID121796M10</td>\n",
" <td>Male</td>\n",
" <td>Gurgaon</td>\n",
" <td>100000000</td>\n",
" <td>1991-07-24</td>\n",
" <td>2015-07-30</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>30000</td>\n",
" <td>RAHEESH KHAN</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>G</td>\n",
" <td>S122</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>85387</th>\n",
" <td>ID122441H10</td>\n",
" <td>Male</td>\n",
" <td>Agartala</td>\n",
" <td>5000000</td>\n",
" <td>1969-07-30</td>\n",
" <td>2015-07-30</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>TECHNOLOGY</td>\n",
" <td>...</td>\n",
" <td>15.50</td>\n",
" <td>15000</td>\n",
" <td>36079.79</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>G</td>\n",
" <td>S122</td>\n",
" <td>4</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>85512</th>\n",
" <td>ID122622G20</td>\n",
" <td>Male</td>\n",
" <td>Delhi</td>\n",
" <td>5000000</td>\n",
" <td>1982-07-30</td>\n",
" <td>2015-07-30</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>KELLTON TECH</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>G</td>\n",
" <td>S122</td>\n",
" <td>7</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>85888</th>\n",
" <td>ID123157V20</td>\n",
" <td>Male</td>\n",
" <td>Mumbai</td>\n",
" <td>1800000</td>\n",
" <td>1990-04-05</td>\n",
" <td>2015-07-31</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>ANGLO EASTERN SHIP MANAGEMENT INDIA PVT LTD</td>\n",
" <td>...</td>\n",
" <td>11.99</td>\n",
" <td>10000</td>\n",
" <td>26328.93</td>\n",
" <td>Y</td>\n",
" <td>Mobile</td>\n",
" <td>G</td>\n",
" <td>S122</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>86413</th>\n",
" <td>ID123955N00</td>\n",
" <td>Male</td>\n",
" <td>Gwalior</td>\n",
" <td>5000009</td>\n",
" <td>1987-06-27</td>\n",
" <td>2015-07-31</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>GOOD EARTH ECO FUTURES P...</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>G</td>\n",
" <td>S122</td>\n",
" <td>7</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>86726</th>\n",
" <td>ID124397N20</td>\n",
" <td>Male</td>\n",
" <td>Nashik</td>\n",
" <td>3500000</td>\n",
" <td>1986-12-30</td>\n",
" <td>2015-07-31</td>\n",
" <td>300000</td>\n",
" <td>5</td>\n",
" <td>14700</td>\n",
" <td>PRIME FOCUS LTD</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>N</td>\n",
" <td>Web-browser</td>\n",
" <td>G</td>\n",
" <td>S122</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>89 rows × 26 columns</p>\n",
"</div>"
],
"text/plain": [
" ID Gender City Monthly_Income DOB \\\n",
"2506 ID003596I10 Female Gurgaon 3000000 1972-06-06 \n",
"3757 ID005397P20 Male Pune 1200000 1981-12-19 \n",
"4316 ID006170I00 Male Baramulla 54954545 1986-06-28 \n",
"4320 ID006174M40 Male Bengaluru 1200000 1982-02-02 \n",
"5625 ID008020M00 Male Delhi 2000000 1985-01-01 \n",
"5760 ID008211V10 Male Gautam Buddha Nagar 1800000 1976-07-01 \n",
"5928 ID008452C20 Male Mumbai 10000000 1982-12-06 \n",
"6616 ID009435X00 Male Bengaluru 1100000 1984-09-21 \n",
"8673 ID012321X10 Male Hyderabad 10000000 1986-11-09 \n",
"8933 ID012688A30 Male Bengaluru 1200000 1987-08-02 \n",
"10960 ID015572Y20 Male Chennai 2301000 1979-02-14 \n",
"11047 ID015699V40 Male Sibsagar 13000000 2061-06-01 \n",
"11299 ID016080M00 Male Hyderabad 2000000 1978-07-23 \n",
"13777 ID019643N30 Male Ahmedabad 12000000 1967-11-11 \n",
"14643 ID020903Z30 Male Mumbai 12000000 1992-05-08 \n",
"15872 ID022660O00 Male Bagalkote 1100000 1975-07-10 \n",
"17693 ID025342S20 Female Bengaluru 5040000 1980-04-22 \n",
"18553 ID026631H10 Male Mumbai 20000000 1976-06-14 \n",
"22492 ID032214A40 Male Bengaluru 1600000 1978-05-21 \n",
"24356 ID034909R40 Male Delhi 3636366 1988-02-02 \n",
"25533 ID036559D40 Male Delhi 120100132 1975-03-13 \n",
"25853 ID037036M10 Male Delhi 1800000 2058-03-31 \n",
"26017 ID037284A40 Male Pune 2500000 1988-07-31 \n",
"26103 ID037390C00 Female Visakhapatnam 20000000 1986-01-31 \n",
"26494 ID037940G00 Male Mumbai 3600000 1980-01-10 \n",
"26739 ID038307J20 Male Mumbai 5300020 1972-09-02 \n",
"28026 ID040122E20 Male Delhi 100000000 1967-02-01 \n",
"28104 ID040241T10 Male Delhi 10000000 1965-01-01 \n",
"28118 ID040262O20 Male Delhi 100000000 1965-01-01 \n",
"28369 ID040642E20 Female Pune 8595656 1988-06-13 \n",
"... ... ... ... ... ... \n",
"49780 ID071345B00 Male Mumbai 20000000 1979-10-12 \n",
"50200 ID071956O10 Male Gurgaon 50000000 1976-02-14 \n",
"50324 ID072143T30 Male Delhi 4500000 1978-03-04 \n",
"51208 ID073464O40 Male Delhi 1350000 1988-10-10 \n",
"51643 ID074088O30 Male Kolkata 1200000 1983-12-21 \n",
"52987 ID076036M10 Male Bengaluru 5000000 1987-11-18 \n",
"53141 ID076246O10 Male Delhi 444554443 1988-11-11 \n",
"56379 ID080915D00 Male Delhi 1100000 1991-11-22 \n",
"58255 ID083578O30 Male Hoshiarpur 2000000 1992-05-21 \n",
"58918 ID084523X30 Female Aurangabad 1500000 1980-02-01 \n",
"63345 ID090880K00 Male Mumbai 1350000 1981-06-12 \n",
"68002 ID097622S20 Male Delhi 100000000 1986-09-09 \n",
"69856 ID100287F20 Male Bengaluru 1300000 1981-09-26 \n",
"70972 ID101907N20 Male Hisar 1800000 1990-03-04 \n",
"72114 ID103535D00 Male Kolkata 2632400 1985-08-03 \n",
"73352 ID105286M10 Female Kolkata 5300000 1982-10-07 \n",
"73394 ID105346U10 Male Bengaluru 4684000 1988-07-15 \n",
"73938 ID106123R30 Male Bengaluru 1960000 1987-12-11 \n",
"78225 ID112183T30 Male Durgapur 6500000 1979-05-05 \n",
"78578 ID112691H10 Male Bengaluru 1200000 1979-06-19 \n",
"79089 ID113457T20 Male Mumbai 34000000 1970-01-01 \n",
"79260 ID113692U20 Male Thane 13500000 1974-07-12 \n",
"82681 ID118606U10 Male Hyderabad 1200000 1982-05-05 \n",
"83459 ID119712I20 Male Hyderabad 1833333 1985-01-31 \n",
"84923 ID121796M10 Male Gurgaon 100000000 1991-07-24 \n",
"85387 ID122441H10 Male Agartala 5000000 1969-07-30 \n",
"85512 ID122622G20 Male Delhi 5000000 1982-07-30 \n",
"85888 ID123157V20 Male Mumbai 1800000 1990-04-05 \n",
"86413 ID123955N00 Male Gwalior 5000009 1987-06-27 \n",
"86726 ID124397N20 Male Nashik 3500000 1986-12-30 \n",
"\n",
" Lead_Creation_Date Loan_Amount_Applied Loan_Tenure_Applied \\\n",
"2506 2015-05-04 1500000 5 \n",
"3757 2015-05-05 300000 2 \n",
"4316 2015-05-06 1000000 5 \n",
"4320 2015-05-06 300000 0 \n",
"5625 2015-05-07 0 5 \n",
"5760 2015-05-07 400000 3 \n",
"5928 2015-05-07 0 0 \n",
"6616 2015-05-08 2500000 5 \n",
"8673 2015-05-11 200000 3 \n",
"8933 2015-05-11 1000000 5 \n",
"10960 2015-05-13 0 0 \n",
"11047 2015-05-13 1000000 5 \n",
"11299 2015-05-14 500000 1 \n",
"13777 2015-05-16 1100000 5 \n",
"14643 2015-05-17 0 0 \n",
"15872 2015-05-19 2000000 4 \n",
"17693 2015-05-21 500000 5 \n",
"18553 2015-05-22 0 5 \n",
"22492 2015-05-26 100000 1 \n",
"24356 2015-05-28 300000 2 \n",
"25533 2015-05-29 500000 0 \n",
"25853 2015-05-30 500000 5 \n",
"26017 2015-05-30 2500000 4 \n",
"26103 2015-05-30 200000 2 \n",
"26494 2015-05-31 0 0 \n",
"26739 2015-05-31 100000 3 \n",
"28026 2015-06-02 100000 1 \n",
"28104 2015-06-02 100000 1 \n",
"28118 2015-06-02 0 1 \n",
"28369 2015-06-02 100000 1 \n",
"... ... ... ... \n",
"49780 2015-06-26 1000000 0 \n",
"50200 2015-06-26 500000 5 \n",
"50324 2015-06-26 0 0 \n",
"51208 2015-06-27 300000 5 \n",
"51643 2015-06-28 200000 3 \n",
"52987 2015-06-29 0 0 \n",
"53141 2015-06-30 0 0 \n",
"56379 2015-07-03 650000 4 \n",
"58255 2015-07-03 0 1 \n",
"58918 2015-07-04 1000000 5 \n",
"63345 2015-07-09 300000 4 \n",
"68002 2015-07-15 0 0 \n",
"69856 2015-07-16 0 0 \n",
"70972 2015-07-18 200000 0 \n",
"72114 2015-07-19 0 0 \n",
"73352 2015-07-20 200000 5 \n",
"73394 2015-07-20 1000000 5 \n",
"73938 2015-07-21 500000 3 \n",
"78225 2015-07-23 1000000 5 \n",
"78578 2015-07-24 300000 2 \n",
"79089 2015-07-24 1000000 5 \n",
"79260 2015-07-25 0 0 \n",
"82681 2015-07-28 500000 2 \n",
"83459 2015-07-29 500000 3 \n",
"84923 2015-07-30 0 0 \n",
"85387 2015-07-30 0 0 \n",
"85512 2015-07-30 0 0 \n",
"85888 2015-07-31 0 0 \n",
"86413 2015-07-31 0 0 \n",
"86726 2015-07-31 300000 5 \n",
"\n",
" Existing_EMI Employer_Name \\\n",
"2506 30000 MAKE MYTRIP INDIA PVT LTD (MAKEMYTRIP.COM) \n",
"3757 28000 ZS ASSOCIATES INDIA PVT LTD \n",
"4316 150000 514949848 \n",
"4320 0 HEALTHFORE TECHNOLOGIES LTD \n",
"5625 0 SPREADTRUM COMMUNICATION INDIA PVT. LTD \n",
"5760 13700 HONDA CARS INDIA LTD \n",
"5928 0 ABC \n",
"6616 55000 BSR AND COMPANY \n",
"8673 50000 DEEP HOSPITAL \n",
"8933 25000 ACCENTURE SERVICES PVT LTD \n",
"10960 0 FLEXIBLE STEEL LACING COMPANY P LTD \n",
"11047 30000 OIL & NATURAL GAS CORP LTD (ONGC) \n",
"11299 50000 MOBILEIRON INDIA SOFTWARE PVT LTD \n",
"13777 0 CJFJF \n",
"14643 0 TREEG SECURITY PRVT LTD \n",
"15872 38000 SRV POLYTECHNIC \n",
"17693 12200 FIRST ADVANTAGE OFFSHORE SERVICES PVT LTD (ZAP... \n",
"18553 10000 CHANNEL TECHNOLOGIES PVT LTD \n",
"22492 0 NARAYANA SINGH \n",
"24356 0 GOOD EARTH ECO FUTURES PVT LTD \n",
"25533 5454365 54543543535 \n",
"25853 12000 DELHI UNIVERSITY \n",
"26017 0 VIIKING VENTURES PVT LTD \n",
"26103 0 VARA RATNAMALA ILLA \n",
"26494 0 TATA \n",
"26739 20000 JET AIRWAYS INDIA LTD \n",
"28026 0 DASFASD \n",
"28104 0 KDLFAJOIDSJ \n",
"28118 0 UYUHKJHKJ \n",
"28369 56566 FGJHGJHG \n",
"... ... ... \n",
"49780 200000 NILESH TELI \n",
"50200 0 CHOICE CLOTHING CO PVT LTD \n",
"50324 0 JKBANK \n",
"51208 8000 NAGARRO SOFTWARE PVT LTD \n",
"51643 0 IBM GLOBAL SERVICES INDIA LTD \n",
"52987 0 ACCENTURE SERVICES PVT LTD \n",
"53141 0 GOOGLE \n",
"56379 15000 SK \n",
"58255 200000 IZHAR NABI ZOUR \n",
"58918 0 LUPIN \n",
"63345 25000 ICICI BANK LTD \n",
"68002 11500 GHK DEVELOPMENT CONSULTANTS INDIA PVT LTD \n",
"69856 0 EMULEX COMMUNICATIONS PVT LTD \n",
"70972 0 RAMRUP \n",
"72114 0 INDUSIND BANK LTD \n",
"73352 0 DANIELI INDIA LTD \n",
"73394 17536 OCWEN FINANCIAL SOLUTIONS PVT LTD \n",
"73938 0 MICROSOFT INDIA R AND D PVT LTD \n",
"78225 150000 IOB \n",
"78578 8000 COGNIZANT INDIA PVT LTD \n",
"79089 10000000 JAISEN JAYANT GAIKWAD \n",
"79260 0 ACG INSPECTION SYSTEMS PVT LTD \n",
"82681 0 JONES LANG LASALLE \n",
"83459 0 KANTAR OPERATIONS / TNS INDIA PVT LTD \n",
"84923 30000 RAHEESH KHAN \n",
"85387 0 TECHNOLOGY \n",
"85512 0 KELLTON TECH \n",
"85888 0 ANGLO EASTERN SHIP MANAGEMENT INDIA PVT LTD \n",
"86413 0 GOOD EARTH ECO FUTURES P... \n",
"86726 14700 PRIME FOCUS LTD \n",
"\n",
" ... Interest_Rate Processing_Fee EMI_Loan_Submitted Filled_Form \\\n",
"2506 ... NaN NaN NaN N \n",
"3757 ... NaN NaN NaN N \n",
"4316 ... NaN NaN NaN N \n",
"4320 ... NaN NaN NaN N \n",
"5625 ... NaN NaN NaN N \n",
"5760 ... NaN NaN NaN N \n",
"5928 ... NaN NaN NaN N \n",
"6616 ... NaN NaN NaN N \n",
"8673 ... NaN NaN NaN N \n",
"8933 ... 13.99 2500 23263.07 N \n",
"10960 ... 16.00 12000 42188.44 Y \n",
"11047 ... NaN NaN NaN N \n",
"11299 ... 14.85 10000 45093.77 N \n",
"13777 ... NaN NaN NaN N \n",
"14643 ... NaN NaN NaN N \n",
"15872 ... NaN NaN NaN N \n",
"17693 ... NaN NaN NaN N \n",
"18553 ... NaN NaN NaN N \n",
"22492 ... NaN NaN NaN N \n",
"24356 ... NaN NaN NaN N \n",
"25533 ... 15.50 5000 12026.60 N \n",
"25853 ... NaN NaN NaN N \n",
"26017 ... NaN NaN NaN N \n",
"26103 ... 20.00 2000 10179.16 Y \n",
"26494 ... NaN NaN NaN N \n",
"26739 ... 15.50 1000 3491.07 Y \n",
"28026 ... NaN NaN NaN N \n",
"28104 ... NaN NaN NaN N \n",
"28118 ... NaN NaN NaN N \n",
"28369 ... NaN NaN NaN N \n",
"... ... ... ... ... ... \n",
"49780 ... NaN NaN NaN N \n",
"50200 ... 14.85 10000 11855.63 Y \n",
"50324 ... NaN NaN NaN N \n",
"51208 ... NaN NaN NaN N \n",
"51643 ... NaN NaN NaN N \n",
"52987 ... NaN NaN NaN N \n",
"53141 ... NaN NaN NaN N \n",
"56379 ... NaN NaN NaN N \n",
"58255 ... NaN NaN NaN N \n",
"58918 ... NaN NaN NaN N \n",
"63345 ... 16.00 6000 8502.08 N \n",
"68002 ... NaN NaN NaN N \n",
"69856 ... 11.99 10000 26328.93 Y \n",
"70972 ... 22.00 4000 6301.22 Y \n",
"72114 ... 11.99 999 39493.39 Y \n",
"73352 ... 15.25 4000 4784.27 Y \n",
"73394 ... 13.99 5000 23263.07 Y \n",
"73938 ... 13.50 6250 16967.64 N \n",
"78225 ... NaN NaN NaN N \n",
"78578 ... NaN NaN NaN N \n",
"79089 ... 13.99 5000 23263.07 N \n",
"79260 ... 11.99 15000 33359.09 Y \n",
"82681 ... 15.25 10000 24302.76 Y \n",
"83459 ... NaN NaN NaN N \n",
"84923 ... NaN NaN NaN N \n",
"85387 ... 15.50 15000 36079.79 N \n",
"85512 ... NaN NaN NaN N \n",
"85888 ... 11.99 10000 26328.93 Y \n",
"86413 ... NaN NaN NaN N \n",
"86726 ... NaN NaN NaN N \n",
"\n",
" Device_Type Var2 Source Var4 LoggedIn Disbursed \n",
"2506 Web-browser E S133 3 0 0 \n",
"3757 Web-browser B S134 3 0 0 \n",
"4316 Web-browser B S159 2 0 0 \n",
"4320 Web-browser B S122 2 0 0 \n",
"5625 Web-browser B S127 2 0 0 \n",
"5760 Web-browser B S143 3 0 0 \n",
"5928 Mobile F S122 2 0 0 \n",
"6616 Web-browser B S133 3 0 0 \n",
"8673 Web-browser B S122 2 0 0 \n",
"8933 Web-browser B S122 3 0 0 \n",
"10960 Mobile C S133 5 0 0 \n",
"11047 Web-browser B S133 3 0 0 \n",
"11299 Web-browser B S133 4 0 0 \n",
"13777 Web-browser B S133 2 0 0 \n",
"14643 Mobile C S133 3 0 0 \n",
"15872 Web-browser B S133 3 0 0 \n",
"17693 Web-browser B S133 3 0 0 \n",
"18553 Web-browser B S133 3 0 0 \n",
"22492 Web-browser B S122 3 0 0 \n",
"24356 Web-browser B S122 3 0 0 \n",
"25533 Web-browser B S153 4 0 0 \n",
"25853 Web-browser B S133 3 0 0 \n",
"26017 Web-browser B S162 3 0 0 \n",
"26103 Web-browser B S159 5 0 0 \n",
"26494 Mobile C S133 2 0 0 \n",
"26739 Web-browser B S133 5 0 0 \n",
"28026 Web-browser B S157 3 0 0 \n",
"28104 Web-browser B S157 2 0 0 \n",
"28118 Web-browser B S157 2 0 0 \n",
"28369 Web-browser B S157 3 0 0 \n",
"... ... ... ... ... ... ... \n",
"49780 Web-browser B S133 2 0 0 \n",
"50200 Web-browser B S127 5 0 0 \n",
"50324 Mobile C S133 2 0 0 \n",
"51208 Web-browser B S133 2 0 0 \n",
"51643 Web-browser B S133 3 0 0 \n",
"52987 Mobile C S122 3 0 0 \n",
"53141 Mobile C S122 3 0 0 \n",
"56379 Web-browser G S122 2 0 0 \n",
"58255 Web-browser G S122 2 0 0 \n",
"58918 Web-browser G S122 3 0 0 \n",
"63345 Web-browser G S122 4 0 0 \n",
"68002 Web-browser G S122 3 0 0 \n",
"69856 Mobile G S122 5 0 0 \n",
"70972 Web-browser G S122 5 0 0 \n",
"72114 Mobile G S122 5 0 0 \n",
"73352 Web-browser G S122 5 1 0 \n",
"73394 Web-browser G S122 5 0 0 \n",
"73938 Web-browser G S122 4 0 0 \n",
"78225 Web-browser G S122 3 0 0 \n",
"78578 Web-browser G S122 3 0 0 \n",
"79089 Web-browser G S122 4 0 0 \n",
"79260 Mobile G S122 5 0 0 \n",
"82681 Web-browser G S122 5 1 0 \n",
"83459 Web-browser G S122 3 0 0 \n",
"84923 Web-browser G S122 3 0 0 \n",
"85387 Web-browser G S122 4 0 0 \n",
"85512 Web-browser G S122 7 0 0 \n",
"85888 Mobile G S122 5 0 0 \n",
"86413 Web-browser G S122 7 0 0 \n",
"86726 Web-browser G S122 3 0 0 \n",
"\n",
"[89 rows x 26 columns]"
]
},
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"train.ix[train.ix[:,'Monthly_Income'] > 1000000,:]"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"train.ix[train.ix[:,'Monthly_Income'] > 1000000,'Monthly_Income'] = float('nan')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"##### Feature Engineering\n",
"Extracting date and month from 'Lead_Creation_Date' feature. "
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"(array([5, 6, 7]),\n",
" array([15, 4, 19, 9, 20, 1, 2, 3, 13, 5, 8, 24, 12, 7, 10, 6, 11,\n",
" 26, 17, 21, 18, 14, 16, 27, 25, 23, 22, 30, 28, 31, 29]))"
]
},
"execution_count": 9,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"train['month'] = [date.month for date in train['Lead_Creation_Date']]\n",
"train['day'] = [date.day for date in train['Lead_Creation_Date']]\n",
"train['month'].unique(), train['day'].unique()"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": true
},
"source": [
"Extracting age form 'DOB' feature"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"array([ 37, 30, 34, 28, 31, 33, 40, 43, 26, 39, 25, 32, 42,\n",
" 24, 21, 27, 36, -49, 23, 35, 20, 41, 29, -35, 38, 45,\n",
" 47, -44, 22, 44, 19, 49, -46, -43, 50, -48, -45, -47, 48,\n",
" 46, -41, -40, -37, -39, 18, -42, -38, -36, -33, -27, 0, -26,\n",
" -34, -17, -32, -31, -28, -23, -14, -30, -21])"
]
},
"execution_count": 10,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"train['Age'] = [(date.today().year - row.year) for row in train['DOB']]\n",
"train['Age'].unique()"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"train.ix[train['Age'] < 0,'Age'] = 100 + train.ix[train['Age'] < 0,'Age']"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"array([37, 30, 34, 28, 31, 33, 40, 43, 26, 39, 25, 32, 42, 24, 21, 27, 36,\n",
" 51, 23, 35, 20, 41, 29, 65, 38, 45, 47, 56, 22, 44, 19, 49, 54, 57,\n",
" 50, 52, 55, 53, 48, 46, 59, 60, 63, 61, 18, 58, 62, 64, 67, 73, 74,\n",
" 66, 83, 68, 69, 72, 77, 86, 70, 79])"
]
},
"execution_count": 12,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"train.ix[train['Age'] == 0,'Age'] = 50 + train.ix[train['Age'] == 0,'Age']\n",
"train['Age'].unique()"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"#similarly for test"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"(array([5, 6, 7]),\n",
" array([ 5, 1, 2, 3, 4, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17,\n",
" 18, 19, 20, 28, 21, 23, 22, 29, 24, 25, 26, 27, 30, 31]))"
]
},
"execution_count": 14,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"test['month'] = [date.month for date in test['Lead_Creation_Date']]\n",
"test['day'] = [date.day for date in test['Lead_Creation_Date']]\n",
"test['month'].unique(), test['day'].unique()"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"array([ 28, 35, 26, 24, 29, 43, 25, 32, 30, 20, 36, 27, 40,\n",
" 31, 23, -44, 33, 22, 46, 39, 34, -48, 37, 41, 21, 44,\n",
" -38, -41, 45, -42, 38, 47, -45, 49, 50, -36, 19, 48, -46,\n",
" -49, 42, -47, -43, 18, -39, -37, -40, -35, -28, 0, -32, -17,\n",
" -34, -27])"
]
},
"execution_count": 15,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"test['Age'] = [(date.today().year - row.year) for row in test['DOB']]\n",
"test['Age'].unique()"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"test.ix[test['Age'] < 0,'Age'] = 100 + test.ix[test['Age'] < 0,'Age']"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"array([28, 35, 26, 24, 29, 43, 25, 32, 30, 20, 36, 27, 40, 31, 23, 56, 33,\n",
" 22, 46, 39, 34, 52, 37, 41, 21, 44, 62, 59, 45, 58, 38, 47, 55, 49,\n",
" 50, 64, 19, 48, 54, 51, 42, 53, 57, 18, 61, 63, 60, 65, 72, 68, 83,\n",
" 66, 73])"
]
},
"execution_count": 17,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"test.ix[test['Age'] == 0,'Age'] = 50 + test.ix[test['Age'] == 0,'Age']\n",
"test['Age'].unique()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"In banking loan disbursal there are two key ratios. They are self-explanatory.\n",
" 1. Loan to Income Ratio\n",
" 2. Fixed Obligation to Income Ratio\n",
"We calculate them as follows."
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"train['emi2income'] = train['EMI_Loan_Submitted']/train['Monthly_Income']\n",
"train['FOIR'] = (train['Existing_EMI']+train['EMI_Loan_Submitted'])/train['Monthly_Income']"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"test['emi2income'] = test['EMI_Loan_Submitted']/test['Monthly_Income']\n",
"test['FOIR'] = (test['Existing_EMI']+test['EMI_Loan_Submitted'])/test['Monthly_Income']"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"# Encoding the categorical features"
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"categorical_vars = ['Gender', 'City', 'Salary_Account', 'Employer_Name', 'Mobile_Verified', 'Var1', 'Filled_Form', 'Device_Type', 'Var2', 'Source', 'Var4']"
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"['Female' 'Male']\n",
"['Delhi' 'Mumbai' 'Panchkula' 'Saharsa' 'Bengaluru' 'Sindhudurg' 'Kochi'\n",
" 'Surat' 'Pune' 'Bhubaneswar' 'Howrah' 'Chennai' 'Ludhiana' 'Lucknow'\n",
" 'Bardhaman' 'Indore' 'Hyderabad' 'Udaipur' 'Faridabad' 'Angul' 'Kolkata'\n",
" 'Lakhisarai' 'Visakhapatnam' 'Patna' 'Gautam Buddha Nagar' 'Chandigarh'\n",
" 'Cuddalore' 'Ghaziabad' 'Meerut' 'Ahmedabad' 'Vijayawada' 'Rourkela'\n",
" 'Dibrugarh' 'Madurai' 'Haridwar' 'Panipat' 'Vadodara' 'Gurgaon' 'Dehradun'\n",
" 'Coimbatore' 'Hooghly' 'Bastar' 'Jaipur' 'Valsad' 'Bharuch'\n",
" 'Dakshina Kannada' 'Buxar' 'Banka' 'Rupnagar' 'VISNAGAR' 'Nagpur'\n",
" 'Anantapur' 'Thiruvalla' 'Kutch' 'Prakasam' 'Koppal' 'Amravati' 'Guwahati'\n",
" 'Bhopal' 'Shimoga' 'Bijnor' 'Bhilwara' 'Palwal' 'Bathinda' 'Bhiwadi'\n",
" 'Jammu' 'Kottayam' 'Gwalior' 'Agra' 'Satara' 'Varanasi' 'Kaithal'\n",
" 'Asansol' 'Panaji' 'Jodhpur' 'Alwar' 'Ambala' 'Mysore' 'Amareli'\n",
" 'Hamirpur' 'Morvi' 'Noida' 'Midnapore East' 'Beawar' 'Dindigul' 'Sitapur'\n",
" 'Neemuch' 'Jalandhar' 'Raipur' 'Kanpur Nagar' 'Bulandshahr' 'Ganganagar'\n",
" 'Vellore' 'Aurangabad' 'Ambur' 'Amritsar' 'Birbhum' 'Warangal' 'Thane'\n",
" 'Cachar' 'Ramanathapuram' 'Rewari' 'Tanuku' 'North 24 Parganas' 'Daman'\n",
" 'Firozpur' 'Hubli' 'Bolangir' 'Mehsana' 'Pondicherry' 'Vapi' 'Ghazipur'\n",
" 'Patiala' 'Ponda' 'Baleswar' 'Kanchipuram' 'Pathankot' 'Yamuna Nagar'\n",
" 'Kohima' 'Sidhi' 'Hapur' 'Rajkot' 'Sonitpur' 'Tirunelveli' 'Deoghar'\n",
" 'Jagatsinghpur' 'Jhabua' 'Ajmer' 'Ranchi' 'Nashik' 'Moradabad' 'South Goa'\n",
" 'Kapurthala' 'ADIPUR' 'Durgapur' 'Patan' 'Viluppuram' 'Rishikesh'\n",
" 'Dhanbad' 'Rewa' 'Gulbarga' 'Burdwan' 'Gandhidham' 'Guntur' 'Khammam'\n",
" 'Rudrapur' 'Kupwara' 'Rohtak' 'Kanpur' 'Faizabad' 'Sonipat' 'Gorkakhpur'\n",
" 'Sambalpur' 'West Godavari' 'East Khasi Hills' 'Karnal' 'Jharsuguda'\n",
" 'Wayanad' 'Ernakulam' 'Kharagpur' 'Navi Mumbai' 'Muzaffarpur' 'Amreli'\n",
" 'Rangareddy' 'Akola' 'Kodad' 'Margao' 'Salem' 'Bhandara' 'Mahasamund'\n",
" 'Aligarh' 'Ratnagiri' 'Kota' 'East Singhbhum' 'Erode' 'Purulia'\n",
" 'Kendujhar' 'Almora' 'Morena' 'Krishna' 'Chhindwara' 'Chhatarpur' 'Rohtas'\n",
" 'Pilibhit' 'Shivpuri' 'Jamnagar' 'Ambikapur' 'Karimnagar' 'Silvassa'\n",
" 'Bhavnagar' 'Shilong' 'Cuttack' 'Dadra & nagar Haveli' 'Kurnool' 'Sangli'\n",
" 'Gandhinagar' 'Mangalore' 'Barabanki' 'Ambedkar Nagar' 'Belgaum'\n",
" 'Gorakhpur' 'Bhadrak' 'Dakshin Dinajpur' 'Alleppey' 'Nabha' 'Krishnagiri'\n",
" 'Vizianagaram' 'Chamarajanagar' 'Kolhapur' 'Secunderabad' 'Tiruchirapalli'\n",
" 'Namakkal' 'Jalpaiguri' 'Kamrup Rural' 'Alappuzha' 'Shahpura' 'Tezpur'\n",
" 'Mathura' 'Bellary' 'Gonda' 'Jamshedpur' 'Kurukshetra' 'Solan' 'Thrissur'\n",
" 'Guna' 'Hisar' 'Haldwani' 'Sonbhadra' 'Sibsagar' 'Sagar' 'Anand'\n",
" 'Surendranagar' 'Udupi and Uttara Kannada' 'Nawanshahr' 'Kozhikode'\n",
" 'Mohali' 'Hospet' 'Kumbakonam' 'Rayagada' 'Raigarh' 'Navsari' 'Tumkur'\n",
" 'Bettiah' 'Ratlam' 'Harda' 'Tiruppur' 'Satna' 'Jhunjhunu' 'Gaya'\n",
" 'Nalgonda' 'Purnia' 'Rajahmundry' 'Siliguri' 'Hosur' 'VALLABH VIDYANAGAR'\n",
" 'Jabalpur' 'Durg' 'Mokokchung' 'Ahmednagar' 'Thiruvananthapuram' 'Deoria'\n",
" 'Sant Kabir Nagar' 'Mayurbhanj' 'Chandrapur' 'Karad' 'Koraput' 'Allahabad'\n",
" 'Nanded' 'Dahod' 'Khurdha' 'Mandya' 'Bomdila' 'Thoothukudi' 'Kadapa'\n",
" 'Perambalur' 'Jhajjar' 'Andman & Nicobar' 'Bidar' 'Mahbubnagar' 'Banswara'\n",
" 'Chittoor' 'Dholpur' 'Nilgiris' 'VIRPUR' 'Srikakulam' 'Rajpura'\n",
" 'South 24 Parganas' 'Papum Pare' 'Barpeta' 'Badaun' 'Sirsa' 'Baramulla'\n",
" 'Jalgaon' 'Pollachi' 'Nellore' 'Chittorgarh' 'Kannur' 'Yavatmal'\n",
" 'Behrampur' 'Kushinagar' 'Kashipur' 'Agartala' 'Dhamtari' 'Shimla'\n",
" 'Midnapore West' 'Theni' 'KAMREJ' 'Ujjain' 'Buldhana' 'Jalore' 'Thanjavur'\n",
" 'Ongole' 'East Godavari' 'Baddi' 'Saharanpur' 'Karur' 'Kanyakumari'\n",
" 'Damoh' 'Etah' 'Bilaspur' 'Bokaro' 'Medak' 'Aluva' 'Tadipatri' 'Latur'\n",
" 'Bahadurgarh' 'Ganjam' 'Hassan' 'Sehore' 'Tirupati' 'Hanumangarh' 'Sikar'\n",
" 'Bhilai' 'Barmer' 'Margoa' 'Bijapur' 'GANDEVI' 'Darrang' 'Solapur'\n",
" 'Pratapgarh' 'Pudukkottai' 'Palakkad' 'Karauli' 'Barwani' 'Roorkee'\n",
" 'Siddharthnagar' 'Magadh' 'Malappuram' 'Fatehabad' 'Samastipur' 'Kheda'\n",
" 'Nadia' 'Moga' 'GODHRA' 'Srinagar' 'Kalka' 'Bhiwani' 'Modasa' 'Dharmapuri'\n",
" 'Ichalkaranji' 'Farrukhabad' 'Nagapattinam' 'Davanagere' 'Mau' 'Dhenkanal'\n",
" 'SURENDERNAGAR' 'MANDVI' 'Ankleshwar' 'Fatehgarh Sahib' 'Hoshiarpur'\n",
" 'Sabarkantha' 'Nayagarh' 'Rae Bareli' 'Sirohi' 'Nalanda' 'Raigad' 'Siwan'\n",
" 'Kollam' 'Korba' 'Mandsaur' 'Pauri Garhwal' 'Banaskantha' 'Bhojpur'\n",
" 'UMBERGAON' 'Washim' 'Adilabad' 'Banaskhantha' 'Dindori' 'Malda'\n",
" 'Balaghat' 'Wardha' 'Kakinada' 'Tinsukia' 'Dhule' 'Jajapur'\n",
" 'Jyotiba Phule Nagar' 'Halol' 'Sonepur' 'Bhagalpur' 'Kendrapara'\n",
" 'Motihari' 'Bharatpur' 'Janigir - Champa' 'Begusarai' 'Narsinghpur'\n",
" 'Darjeeling' 'Nizamabad' 'BARDOLI' 'Churu' 'Keonjhar' 'Bikaner' 'Azamgarh'\n",
" 'Nagercoil' 'Jhansi' 'Dausa' 'Mundra' 'Goa' 'Kangra' 'Proddattur'\n",
" 'Uttar Dinajpur' 'Ballia' 'Kanpur Dehat' 'Veraval' 'Poonch'\n",
" 'Muzaffarnagar' 'Vyara' 'Bandipore' 'Unnao' 'Dimapur' 'Nandurbar' 'Dhar'\n",
" 'Bankura' 'Sultanpur' 'Betul' 'Gangtok' 'Rajsamand' 'LUNAWADA'\n",
" 'Murshidabad' 'Bareilly' 'sri ganganagar' 'Pathanamthitta' 'Hajipur'\n",
" 'Bargarh' 'Anuppur' 'Nagaon' 'Malegaon' 'Khargone' 'Virudhunagar'\n",
" 'DHANGARDHA' 'Pontashaib' 'Kaushambi' 'Cooch Behar' 'Haveri' 'Sitamarhi'\n",
" 'Pakur' 'Anantnag' 'Ariyalur' 'Kolar' 'North Cachar Hills' 'PALANPUR'\n",
" 'Idukki' 'Banda' 'West Singhbhum' 'Faridkot' 'Ramban' 'Daman & Diu'\n",
" 'Ashoknagar' 'Tiruvannamalai' 'Burhanpur' 'Chikkaballapur' 'Sanga Reddy'\n",
" 'Dharwad' 'Bagalkote' 'Travancore' 'Porbandar' 'Gurdaspur' 'Deogarh'\n",
" 'Katihar' 'Chidambaram' 'Mahendragarh' 'Chamoli' 'Shajapur' 'Kadi' 'Pali'\n",
" 'Dumka' 'Kodagu' 'Baran' 'Bagpat' 'Kishtwar' 'Fatehpur' 'Madhubani'\n",
" 'Silchar' 'Darbhanga' 'Nagaur' 'Mirzapur' 'Imphal West' 'Puri' 'Jind'\n",
" 'Gondia' 'Parbhani' 'Lunglei' 'Surguja' 'Upper Subansiri' 'Saran' nan\n",
" 'Baramati' 'Nainital' 'Udham Singh Nagar' 'Junagadh' 'Anjaw' 'Marigaon'\n",
" 'Bhind' 'Dewas' 'Panchmahal' 'Udalguri' 'Chitrakoot' 'Chapra' 'Araria'\n",
" 'Jaunpur' 'Tandur' 'Basti' 'Seraikela Kharsawan' 'Osmanabad' 'Sangrur'\n",
" 'Nadiad' 'Imphal East' 'Sivagangai' 'Chikkamagaluru' 'Tiruvallur' 'UDWADA'\n",
" 'Sangamner' 'Kathua' 'Mandla' 'Lakhimpur Kheri' 'Gadag' 'Jagadalpur'\n",
" 'Fazilka' 'MAHUVA' 'Bahraich' 'Koriya' 'Malout' 'Yadgir' 'Godda'\n",
" 'Jharkhand' 'Mancherial' 'Lohardaga' 'Kishanganj' 'Champawat' 'Karnataka'\n",
" 'Jalna' 'Bhuj' 'Samba' 'Madhepura' 'Phagwara' 'Garhwa' 'Jhalawar'\n",
" 'Damanjodi' 'Etawah' 'Gajapati' 'Shahdol' 'Munger' 'Churachandpur'\n",
" 'Kailashahar' 'Katni' 'Lakhimpur' 'Rajnandgaon' 'Rajgarh' 'VIJAPUR'\n",
" 'Kasaragod' 'Aizawl' 'Jorhat' 'Sawai Madhopur' 'Tiruvarur' 'Kandhamal'\n",
" 'Hingoli' 'Lalitpur' 'Barnala' 'Narmada' 'Sundargarh' 'Raichur' 'Itanagar'\n",
" 'Bundi' 'Dhubri' 'Auraiya' 'Panch Mahals' 'Una' 'Hoshangabad'\n",
" 'Maharajganj' 'Firozabad' 'Ropar' 'Jaisalmer' 'SAYAN' 'Chitradurga' 'Beed'\n",
" 'BHILAD' 'Gadchiroli' 'Sheopur' 'Hardoi' 'Mewat' 'Nabarangpur' 'Chandauli'\n",
" 'Chinnamiram' 'Ramgarh' 'Kargil' 'Palamu' 'Mansa' 'Chatra' 'NALIYA'\n",
" 'Khanna' 'Suryapet' 'Jagdalpur' 'Mandi' 'Jashpur' 'Balasore'\n",
" 'Gandhi Nagar' 'Mahabub Nagar' 'Golaghat' 'Khagaria' 'Bongaigaon'\n",
" 'Vidisha' 'AHMEDB' 'Khandwa' 'Siddipet' 'ANJAR' 'Kamrup Metropolitian'\n",
" 'Sheikhpura' 'Seoni' 'KHAMBHAT' 'Sirmaur' 'Bageshwar' 'Kabri Anglong'\n",
" 'Pithoragarh' 'KALOL' 'Nalbari' 'CHOTILA' 'IDAR' 'Umaria' 'Goalpara'\n",
" 'Chandel' 'Tirur' 'Udhampur' 'Dantewada' 'Jamtara' 'Hailakandi' 'Koderma'\n",
" 'Kannauj' 'BHACHAU' 'Jamui' 'Muktsar' 'Abohar' 'West Garo Hills' 'DEESA'\n",
" 'Boudh' 'Datia' 'Gumla' 'Surendra Nagar' 'Rampur' 'BILIMORA' 'Ri-Bhoi'\n",
" 'Gopal Ganj' 'CHIKHLI (GUJ.)' 'Hazaribagh' 'Gadwal' 'Kokrajhar'\n",
" 'Shahjahanpur' 'Giridih' 'Dungarpur' 'Raisen' 'Namchi' 'SOMNATH JUNAGADHA'\n",
" 'Tarn Taran' 'Mainpuri' 'Thoubal' 'SILVASA' 'Reasi' 'Nawadah' 'AMALSAD'\n",
" 'Jalaun' 'Jaintia Hills' 'Dhalai' 'Himatnagar' 'Tonk' 'Malabar' 'Pulwama'\n",
" 'BAJWA' 'DHANDHUKA' 'Janjgir-Champa' 'DWARKA' 'RADHANPUR' 'Doda'\n",
" 'Narayanpur' 'KAPADWANJ' 'Baksa' 'Tamenglong' 'DHORAJI' 'Siruguppa'\n",
" 'Lakshadweep' 'Lohit']\n",
"['HDFC Bank' 'ICICI Bank' 'State Bank of India' 'HSBC' 'Yes Bank' nan\n",
" 'Kotak Bank' 'Indian Overseas Bank' 'Bank of Maharasthra' 'Axis Bank'\n",
" 'Central Bank of India' 'Standard Chartered Bank' 'Andhra Bank'\n",
" 'Bank of India' 'IndusInd Bank' 'Corporation bank' 'UCO Bank'\n",
" 'The Ratnakar Bank Ltd' 'Citibank' 'Karur Vysya Bank'\n",
" 'Punjab National Bank' 'Lakshmi Vilas bank' 'Syndicate Bank'\n",
" 'Allahabad Bank' 'Bank of Baroda' 'Canara Bank'\n",
" 'Oriental Bank of Commerce' 'Vijaya Bank' 'State Bank of Hyderabad'\n",
" 'IDBI Bank' 'State Bank of Patiala' 'Union Bank of India' 'ING Vysya'\n",
" 'Federal Bank' 'Dena Bank' 'Punjab & Sind bank' 'J&K Bank' 'Deutsche Bank'\n",
" 'Tamil Nadu Mercantile Bank' 'Indian Bank' 'United Bank of India'\n",
" 'Abhyuday Co-op Bank Ltd' 'State Bank of Bikaner & Jaipur' 'Saraswat Bank'\n",
" 'State Bank of Travancore' 'Karnataka Bank' 'South Indian Bank'\n",
" 'State Bank of Mysore' 'Bank of Rajasthan' 'State Bank of Indore'\n",
" 'Dhanalakshmi Bank Ltd' 'Catholic Syrian Bank' 'India Bulls'\n",
" 'Kerala Gramin Bank' 'Firstrand Bank Limited' 'GIC Housing Finance Ltd'\n",
" 'B N P Paribas' 'Industrial And Commercial Bank Of China Limited']\n",
"['CYBOSOL' 'TATA CONSULTANCY SERVICES LTD (TCS)' 'ALCHEMIST HOSPITALS LTD'\n",
" ..., 'UTTAM VALUE STEEL LTD,WARDHA' 'MAYO COLLEGE'\n",
" 'BANGALORE INSTITUTE OF TECHNOLOGY']\n",
"['N' 'Y']\n",
"['HBXX' 'HBXA' 'HAXM' 'HAXB' 'HBXC' 'HBXD' 'HBXH' 'HAXA' 'HBXB' 'HAYT'\n",
" 'HCXD' 'HVYS' 'HAVC' 'HCXG' 'HAZD' 'HCYS' 'HCXF' 'HAXC' 'HAXF']\n",
"['N' 'Y']\n",
"['Web-browser' 'Mobile']\n",
"['G' 'B' 'C' 'E' 'F' 'D' 'A']\n",
"['S122' 'S143' 'S134' 'S133' 'S159' 'S151' 'S137' 'S127' 'S144' 'S123'\n",
" 'S156' 'S153' 'S124' 'S161' 'S139' 'S154' 'S157' 'S138' 'S162' 'S141'\n",
" 'S158' 'S125' 'S129' 'S136' 'S130' 'S155' 'S160' 'S150' 'S135' 'S140']\n",
"[1 3 5 4 2 7 6 0]\n"
]
}
],
"source": [
"for var in categorical_vars:\n",
" print train[var].unique()"
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"number = preprocessing.LabelEncoder()\n",
"for var in categorical_vars:\n",
" train[var] = number.fit_transform(train[var].astype('str'))\n",
" test[var] = number.fit_transform(test[var].astype('str'))"
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"(array([0, 1]), array([37172, 49848]))\n",
"(array([0, 1]), array([16167, 21550]))\n",
"(array([ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12,\n",
" 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25,\n",
" 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38,\n",
" 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51,\n",
" 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64,\n",
" 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77,\n",
" 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90,\n",
" 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103,\n",
" 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116,\n",
" 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129,\n",
" 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142,\n",
" 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155,\n",
" 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168,\n",
" 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181,\n",
" 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194,\n",
" 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207,\n",
" 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 220,\n",
" 221, 222, 223, 224, 225, 226, 227, 228, 229, 230, 231, 232, 233,\n",
" 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246,\n",
" 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259,\n",
" 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272,\n",
" 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 285,\n",
" 286, 287, 288, 289, 290, 291, 292, 293, 294, 295, 296, 297, 298,\n",
" 299, 300, 301, 302, 303, 304, 305, 306, 307, 308, 309, 310, 311,\n",
" 312, 313, 314, 315, 316, 317, 318, 319, 320, 321, 322, 323, 324,\n",
" 325, 326, 327, 328, 329, 330, 331, 332, 333, 334, 335, 336, 337,\n",
" 338, 339, 340, 341, 342, 343, 344, 345, 346, 347, 348, 349, 350,\n",
" 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363,\n",
" 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376,\n",
" 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389,\n",
" 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402,\n",
" 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415,\n",
" 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428,\n",
" 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441,\n",
" 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454,\n",
" 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467,\n",
" 468, 469, 470, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480,\n",
" 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493,\n",
" 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505, 506,\n",
" 507, 508, 509, 510, 511, 512, 513, 514, 515, 516, 517, 518, 519,\n",
" 520, 521, 522, 523, 524, 525, 526, 527, 528, 529, 530, 531, 532,\n",
" 533, 534, 535, 536, 537, 538, 539, 540, 541, 542, 543, 544, 545,\n",
" 546, 547, 548, 549, 550, 551, 552, 553, 554, 555, 556, 557, 558,\n",
" 559, 560, 561, 562, 563, 564, 565, 566, 567, 568, 569, 570, 571,\n",
" 572, 573, 574, 575, 576, 577, 578, 579, 580, 581, 582, 583, 584,\n",
" 585, 586, 587, 588, 589, 590, 591, 592, 593, 594, 595, 596, 597,\n",
" 598, 599, 600, 601, 602, 603, 604, 605, 606, 607, 608, 609, 610,\n",
" 611, 612, 613, 614, 615, 616, 617, 618, 619, 620, 621, 622, 623,\n",
" 624, 625, 626, 627, 628, 629, 630, 631, 632, 633, 634, 635, 636,\n",
" 637, 638, 639, 640, 641, 642, 643, 644, 645, 646, 647, 648, 649,\n",
" 650, 651, 652, 653, 654, 655, 656, 657, 658, 659, 660, 661, 662,\n",
" 663, 664, 665, 666, 667, 668, 669, 670, 671, 672, 673, 674, 675,\n",
" 676, 677, 678, 679, 680, 681, 682, 683, 684, 685, 686, 687, 688,\n",
" 689, 690, 691, 692, 693, 694, 695, 696, 697]), array([ 7, 9, 2, 6, 6, 34, 44, 219, 1788,\n",
" 54, 4, 43, 18, 34, 20, 126, 9, 5,\n",
" 18, 86, 8, 58, 4, 2, 8, 31, 7,\n",
" 63, 42, 34, 2, 10, 50, 1, 18, 5,\n",
" 2, 12, 53, 3, 3, 237, 10, 2, 5,\n",
" 3, 3, 2, 4, 36, 12, 1, 13, 21,\n",
" 3, 1, 4, 14, 11, 7, 4, 3, 2,\n",
" 1, 4, 17, 6, 12, 11, 4, 7, 32,\n",
" 37, 6, 10, 4, 5, 8, 3, 5, 48,\n",
" 1, 9, 9, 18, 79, 38, 10824, 12, 4,\n",
" 13, 16, 6, 26, 68, 22, 39, 26, 3,\n",
" 45, 19, 7, 513, 277, 8, 12, 15, 15,\n",
" 29, 36, 12, 20, 8, 2, 5, 1, 24,\n",
" 9, 3, 17, 6, 3, 1, 1, 3, 4,\n",
" 2, 1, 2, 1, 870, 27, 2, 2, 6916,\n",
" 4, 16, 6, 11, 4, 1, 4, 3, 52,\n",
" 15, 2, 7, 1147, 11, 27, 84, 3, 1,\n",
" 3, 1, 1, 3, 3, 3, 10, 19, 3,\n",
" 2, 1, 1, 5, 14, 2, 2, 7, 10,\n",
" 314, 12527, 2, 10, 6, 24, 1, 3, 40,\n",
" 6, 10, 35, 14, 10, 3, 31, 24, 2,\n",
" 47, 4, 1, 4, 1, 23, 43, 39, 10,\n",
" 10, 128, 75, 6, 5, 10, 447, 5, 6,\n",
" 21, 8, 4, 1, 9, 5, 1, 7, 8,\n",
" 3, 1, 4, 23, 39, 79, 18, 8, 11,\n",
" 2, 338, 10, 560, 29, 1, 38, 4, 2,\n",
" 2, 9, 10, 1, 18, 17, 24, 2, 3,\n",
" 146, 8, 1212, 156, 97, 2, 7, 22, 6,\n",
" 6, 8, 32, 4, 3, 34, 12, 8, 1,\n",
" 1, 3, 34, 151, 9, 19, 8, 58, 144,\n",
" 47, 7272, 1, 20, 12, 1, 13, 734, 5,\n",
" 60, 5, 12, 8, 2, 1331, 8, 23, 91,\n",
" 2, 50, 21, 7, 27, 58, 56, 103, 3,\n",
" 2, 7, 2, 1, 6, 9, 32, 6, 11,\n",
" 3, 13, 23, 14, 108, 8, 12, 3, 6,\n",
" 1, 1, 1, 1, 48, 6, 4, 12, 40,\n",
" 9, 4, 5, 31, 1, 7, 1, 24, 188,\n",
" 7, 86, 18, 9, 21, 5, 1, 46, 36,\n",
" 53, 30, 6, 14, 4, 3, 5, 6, 9,\n",
" 9, 8, 1, 38, 7, 6, 18, 5, 8,\n",
" 6, 1, 2, 492, 3, 4, 2, 6, 2,\n",
" 12, 94, 2888, 26, 5, 16, 14, 2, 54,\n",
" 49, 56, 31, 15, 11, 1, 61, 18, 8,\n",
" 24, 1, 2, 4, 1, 1, 1, 14, 2,\n",
" 1, 580, 255, 4, 4, 6, 1, 7, 375,\n",
" 1, 12, 4, 4, 16, 9, 1, 1, 18,\n",
" 21, 5, 3, 7, 6, 1, 4, 8, 68,\n",
" 5, 18, 19, 2, 24, 6, 8, 12, 217,\n",
" 11, 6, 43, 27, 6, 1, 17, 91, 1,\n",
" 28, 4, 6, 8, 1, 10795, 5, 1, 16,\n",
" 12, 33, 166, 2, 2, 1, 41, 17, 4,\n",
" 20, 10, 28, 594, 6, 5, 1, 28, 30,\n",
" 2, 23, 5, 1, 2, 1, 233, 90, 20,\n",
" 3, 7, 7, 4, 98, 16, 17, 373, 149,\n",
" 1, 36, 12, 11, 7, 32, 5, 10, 62,\n",
" 73, 2, 33, 3, 26, 3, 6, 7, 9,\n",
" 11, 41, 461, 3, 3, 5, 2, 2, 19,\n",
" 30, 126, 7, 1, 9, 16, 2, 6, 10,\n",
" 1, 5207, 11, 5, 10, 1, 11, 16, 35,\n",
" 28, 289, 1, 47, 3, 104, 7, 6, 14,\n",
" 10, 2, 7, 1, 110, 34, 8, 21, 8,\n",
" 2, 12, 71, 2, 5, 36, 3, 20, 12,\n",
" 54, 25, 10, 1, 1, 1, 4, 3, 18,\n",
" 20, 4, 98, 11, 3, 24, 11, 3, 48,\n",
" 11, 3, 2, 41, 14, 1, 167, 16, 1,\n",
" 3, 6, 2, 1, 6, 1, 1, 7, 30,\n",
" 18, 6, 5, 2, 2, 16, 31, 11, 59,\n",
" 34, 5, 2, 5, 31, 1, 4, 2, 10,\n",
" 4, 14, 59, 14, 2, 57, 6, 42, 26,\n",
" 21, 15, 8, 9, 802, 1, 6, 2, 6,\n",
" 5, 1, 2, 9, 2, 5, 905, 35, 15,\n",
" 12, 185, 38, 2, 79, 7, 141, 55, 59,\n",
" 115, 3, 7, 16, 10, 1, 3, 23, 1,\n",
" 6, 122, 3, 8, 5, 25, 32, 1, 4,\n",
" 13, 1, 7, 2, 3, 1, 6, 624, 29,\n",
" 69, 161, 75, 4, 4, 265, 37, 20, 764,\n",
" 31, 11, 58, 33, 3, 8, 6, 45, 14,\n",
" 5, 28, 15, 1003, 8]))\n",
"(array([ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12,\n",
" 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25,\n",
" 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38,\n",
" 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51,\n",
" 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64,\n",
" 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77,\n",
" 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90,\n",
" 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103,\n",
" 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116,\n",
" 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129,\n",
" 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142,\n",
" 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155,\n",
" 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168,\n",
" 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181,\n",
" 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194,\n",
" 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207,\n",
" 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 220,\n",
" 221, 222, 223, 224, 225, 226, 227, 228, 229, 230, 231, 232, 233,\n",
" 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246,\n",
" 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259,\n",
" 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272,\n",
" 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 285,\n",
" 286, 287, 288, 289, 290, 291, 292, 293, 294, 295, 296, 297, 298,\n",
" 299, 300, 301, 302, 303, 304, 305, 306, 307, 308, 309, 310, 311,\n",
" 312, 313, 314, 315, 316, 317, 318, 319, 320, 321, 322, 323, 324,\n",
" 325, 326, 327, 328, 329, 330, 331, 332, 333, 334, 335, 336, 337,\n",
" 338, 339, 340, 341, 342, 343, 344, 345, 346, 347, 348, 349, 350,\n",
" 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363,\n",
" 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376,\n",
" 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389,\n",
" 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402,\n",
" 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415,\n",
" 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428,\n",
" 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441,\n",
" 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454,\n",
" 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467,\n",
" 468, 469, 470, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480,\n",
" 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493,\n",
" 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505, 506,\n",
" 507, 508, 509, 510, 511, 512, 513, 514, 515, 516, 517, 518, 519,\n",
" 520, 521, 522, 523, 524, 525, 526, 527, 528, 529, 530, 531, 532,\n",
" 533, 534, 535, 536, 537, 538, 539, 540, 541, 542, 543, 544, 545,\n",
" 546, 547, 548, 549, 550, 551, 552, 553, 554, 555, 556, 557, 558,\n",
" 559, 560, 561, 562, 563, 564, 565, 566, 567, 568, 569, 570, 571,\n",
" 572, 573, 574, 575, 576, 577, 578, 579, 580, 581, 582, 583, 584,\n",
" 585, 586, 587, 588, 589, 590, 591, 592, 593, 594, 595, 596, 597,\n",
" 598, 599, 600, 601, 602, 603, 604, 605, 606, 607, 608, 609, 610]), array([ 1, 1, 1, 1, 3, 9, 7, 132, 740, 25, 3,\n",
" 16, 9, 21, 11, 72, 6, 1, 9, 33, 7, 20,\n",
" 3, 2, 5, 14, 2, 30, 15, 16, 5, 25, 10,\n",
" 3, 2, 2, 23, 1, 99, 2, 3, 2, 18, 2,\n",
" 4, 15, 1, 2, 11, 10, 2, 1, 1, 1, 1,\n",
" 6, 8, 5, 3, 16, 18, 4, 6, 3, 1, 1,\n",
" 13, 2, 1, 2, 8, 36, 20, 4698, 4, 1, 2,\n",
" 5, 7, 2, 8, 47, 5, 13, 14, 23, 10, 3,\n",
" 222, 130, 5, 4, 7, 5, 13, 11, 9, 7, 3,\n",
" 3, 1, 2, 7, 6, 4, 5, 2, 3, 1, 1,\n",
" 2, 396, 12, 2, 1, 2979, 1, 7, 6, 8, 3,\n",
" 1, 20, 9, 1, 2, 512, 4, 12, 34, 1, 2,\n",
" 3, 3, 7, 7, 1, 3, 4, 5, 1, 1, 7,\n",
" 10, 130, 5409, 4, 3, 3, 2, 1, 9, 3, 3,\n",
" 15, 3, 1, 10, 4, 1, 15, 1, 1, 2, 11,\n",
" 23, 16, 3, 5, 60, 41, 2, 3, 204, 2, 13,\n",
" 2, 1, 3, 3, 4, 1, 2, 3, 9, 13, 32,\n",
" 2, 2, 1, 147, 5, 1, 235, 7, 3, 10, 2,\n",
" 2, 4, 2, 2, 8, 6, 16, 3, 55, 4, 531,\n",
" 75, 37, 5, 1, 16, 2, 2, 11, 5, 21, 4,\n",
" 3, 4, 3, 19, 50, 9, 3, 21, 75, 16, 3138,\n",
" 1, 5, 7, 2, 2, 317, 4, 20, 3, 6, 3,\n",
" 1, 561, 4, 8, 36, 2, 16, 5, 9, 20, 22,\n",
" 41, 1, 1, 2, 1, 2, 2, 7, 1, 5, 1,\n",
" 8, 11, 9, 44, 6, 6, 1, 2, 1, 3, 2,\n",
" 13, 4, 2, 8, 21, 4, 2, 1, 1, 14, 5,\n",
" 10, 77, 31, 1, 5, 4, 3, 1, 26, 17, 18,\n",
" 11, 1, 4, 2, 2, 4, 4, 3, 7, 3, 20,\n",
" 4, 3, 14, 4, 1, 3, 1, 200, 2, 1, 2,\n",
" 37, 1394, 10, 3, 6, 6, 1, 24, 27, 28, 13,\n",
" 11, 1, 3, 19, 8, 2, 6, 1, 2, 1, 1,\n",
" 17, 1, 1, 233, 110, 1, 1, 159, 8, 2, 1,\n",
" 3, 4, 6, 1, 13, 8, 3, 1, 7, 4, 1,\n",
" 4, 4, 31, 3, 8, 9, 14, 1, 1, 7, 71,\n",
" 5, 4, 22, 9, 1, 2, 5, 54, 3, 15, 1,\n",
" 1, 2, 1, 4630, 6, 2, 10, 5, 12, 57, 1,\n",
" 1, 21, 9, 3, 7, 1, 10, 285, 1, 6, 17,\n",
" 14, 9, 4, 1, 1, 93, 38, 13, 2, 5, 4,\n",
" 3, 53, 5, 6, 176, 68, 10, 4, 6, 2, 18,\n",
" 1, 5, 31, 37, 21, 3, 16, 1, 1, 3, 2,\n",
" 7, 6, 12, 214, 2, 2, 3, 3, 10, 9, 57,\n",
" 2, 2, 1, 6, 1, 1, 6, 2220, 4, 2, 3,\n",
" 2, 5, 6, 14, 15, 141, 27, 1, 52, 3, 2,\n",
" 1, 2, 1, 3, 4, 3, 29, 12, 4, 7, 2,\n",
" 1, 6, 39, 19, 1, 4, 6, 21, 1, 16, 1,\n",
" 1, 3, 5, 7, 2, 1, 44, 1, 2, 15, 5,\n",
" 2, 16, 6, 12, 14, 62, 5, 3, 3, 1, 1,\n",
" 1, 3, 12, 9, 4, 4, 1, 1, 5, 13, 5,\n",
" 29, 19, 4, 4, 1, 10, 1, 4, 3, 11, 28,\n",
" 2, 1, 29, 23, 14, 13, 2, 3, 1, 1, 347,\n",
" 1, 2, 1, 3, 1, 1, 3, 401, 16, 17, 8,\n",
" 80, 30, 33, 4, 46, 33, 31, 52, 3, 4, 10,\n",
" 4, 2, 6, 49, 4, 1, 14, 7, 6, 7, 1,\n",
" 5, 1, 2, 269, 11, 30, 70, 27, 3, 4, 108,\n",
" 15, 18, 316, 22, 4, 2, 28, 10, 1, 3, 15,\n",
" 12, 1, 11, 3, 398, 4]))\n",
"(array([ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16,\n",
" 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33,\n",
" 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50,\n",
" 51, 52, 53, 54, 55, 56, 57]), array([ 108, 238, 485, 8783, 8, 1126, 1170, 406, 5,\n",
" 990, 14, 445, 2376, 649, 182, 125, 42, 253,\n",
" 7, 8, 17695, 328, 13636, 1550, 678, 11, 555,\n",
" 612, 503, 2, 59, 200, 326, 4, 2067, 50,\n",
" 524, 66, 1201, 195, 160, 995, 331, 597, 11843,\n",
" 18, 255, 177, 227, 415, 71, 83, 237, 951,\n",
" 183, 252, 779, 11764]))\n",
"(array([ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16,\n",
" 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33,\n",
" 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50,\n",
" 51, 52, 53, 54, 55, 56, 57]), array([ 53, 1, 107, 221, 3807, 7, 549, 543, 170, 3, 395,\n",
" 13, 203, 1022, 299, 86, 51, 24, 124, 4, 2, 7485,\n",
" 146, 5911, 663, 318, 10, 218, 289, 208, 1, 19, 79,\n",
" 109, 888, 19, 237, 18, 546, 70, 63, 439, 117, 257,\n",
" 5267, 14, 130, 86, 106, 199, 32, 30, 107, 379, 93,\n",
" 102, 341, 5037]))\n",
"(array([ 0, 1, 2, ..., 43565, 43566, 43567]), array([ 1, 1, 1, ..., 21, 1, 1]))\n",
"(array([ 0, 1, 2, ..., 22519, 22520, 22521]), array([1, 1, 1, ..., 6, 1, 1]))\n",
"(array([0, 1]), array([30539, 56481]))\n",
"(array([0, 1]), array([13270, 24447]))\n",
"(array([ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16,\n",
" 17, 18]), array([ 384, 2909, 2011, 1536, 15, 268, 508, 109, 2123,\n",
" 4479, 9010, 1964, 970, 59294, 237, 722, 78, 217, 186]))\n",
"(array([ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16,\n",
" 17, 18]), array([ 186, 1305, 868, 635, 7, 118, 202, 52, 919,\n",
" 2023, 3942, 854, 417, 25607, 111, 268, 36, 101, 66]))\n",
"(array([0, 1]), array([67530, 19490]))\n",
"(array([0, 1]), array([29210, 8507]))\n",
"(array([0, 1]), array([22704, 64316]))\n",
"(array([0, 1]), array([ 9928, 27789]))\n",
"(array([0, 1, 2, 3, 4, 5, 6]), array([ 5, 37280, 14210, 634, 1315, 544, 33032]))\n",
"(array([0, 1, 2, 3, 4, 5, 6]), array([ 4, 16201, 6156, 284, 540, 226, 14306]))\n",
"(array([ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16,\n",
" 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29]), array([38567, 73, 24, 1, 1931, 3, 1, 29885, 1301,\n",
" 2, 3, 1724, 3, 3, 1, 57, 4332, 299,\n",
" 10, 720, 494, 1, 4, 308, 650, 208, 5599,\n",
" 11, 769, 36]))\n",
"(array([ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16,\n",
" 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27]), array([16682, 39, 19, 1, 873, 1, 1, 1, 13015,\n",
" 599, 2, 726, 2, 1, 26, 1, 1808, 148,\n",
" 9, 298, 211, 1, 124, 279, 86, 2400, 340, 24]))\n",
"(array([0, 1, 2, 3, 4, 5, 6, 7]), array([ 2546, 23906, 5931, 25260, 6577, 20266, 232, 2302]))\n",
"(array([0, 1, 2, 3, 4, 5, 6, 7]), array([ 1018, 10410, 2550, 11020, 2834, 8826, 97, 962]))\n"
]
}
],
"source": [
"for var in categorical_vars:\n",
" print np.unique(train[var], return_counts= True)\n",
" print np.unique(test[var], return_counts= True)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"##### Dropping the less relevant categories. "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Since we have 87000 data points, so any category which occurs less than 0.1% i.e. around 87 times in the data set, are merged into one category. For simplification we take this occurance number as 100. "
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"for var in categorical_vars:\n",
" val_count = train[var].value_counts()\n",
" values = val_count[val_count<100].index\n",
" for v in values:\n",
" train.ix[train[var] == v,var] = 999 #encoding all the less relevant categories to a placeholder number"
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"#similarly for test\n",
"for var in categorical_vars:\n",
" val_count = test[var].value_counts()\n",
" values = val_count[val_count<100].index\n",
" for v in values:\n",
" test.ix[test[var] == v,var] = 999"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"##### Understanding the structure of the numerical features. "
]
},
{
"cell_type": "code",
"execution_count": 27,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.axes._subplots.AxesSubplot at 0x10a4c9390>"
]
},
"execution_count": 27,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAZUAAAEACAYAAAB78OvLAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nrkJggg==\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x10a4c1550>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"train['Monthly_Income'].hist()"
]
},
{
"cell_type": "code",
"execution_count": 28,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.axes._subplots.AxesSubplot at 0x10a16eb90>"
]
},
"execution_count": 28,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAZEAAAEACAYAAAByG0uxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nZtba/wN8PXzpS0YRMwAAAABJRU5ErkJggg==\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x10a176ad0>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"train['EMI_Loan_Submitted'].hist()"
]
},
{
"cell_type": "code",
"execution_count": 29,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.axes._subplots.AxesSubplot at 0x10abb2b90>"
]
},
"execution_count": 29,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAYYAAAENCAYAAAAL98L+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n1G4AAAAASUVORK5CYII=\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x10d107590>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"train['Existing_EMI'].hist()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"As we can see the numerical features are highly skewed. TO negate its effect in the analysis, we use transformations, in our case we take the log of such features. "
]
},
{
"cell_type": "code",
"execution_count": 31,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"numeric_vars = ['Monthly_Income', 'Loan_Amount_Applied', 'Loan_Amount_Submitted', 'EMI_Loan_Submitted', 'Existing_EMI']"
]
},
{
"cell_type": "code",
"execution_count": 32,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"for var in numeric_vars:\n",
" train[var] = np.log(train[var]+1)\n",
" test[var] = np.log(test[var]+1)"
]
},
{
"cell_type": "code",
"execution_count": 33,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.axes._subplots.AxesSubplot at 0x10a76b390>"
]
},
"execution_count": 33,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEACAYAAABcXmojAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x10a747b10>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"train['Monthly_Income'].hist()"
]
},
{
"cell_type": "code",
"execution_count": 34,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.axes._subplots.AxesSubplot at 0x10acbc690>"
]
},
"execution_count": 34,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAYQAAAEACAYAAACznAEdAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCTAgSJIaBgRJEgD/H9+1EdIoM0QbAAAAAElFTkSuQmCC\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x10acc5c90>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"train['EMI_Loan_Submitted'].hist()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# 4. Data Analysis "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"##### Create Hold-Out set"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"I like to keep 20% of the data as hold out set and the remaining for training and cross-validation. The tain set has data points from 3 months - may, june & july. Digging deeper, there are 16222 data points after 17.07.15. We keep them as our hold out set and the remaining as our training set for model building. "
]
},
{
"cell_type": "code",
"execution_count": 37,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"16222"
]
},
"execution_count": 37,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"sum(train['Lead_Creation_Date'] > datetime(2015,7,17))"
]
},
{
"cell_type": "code",
"execution_count": 38,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"((70798, 31), (16222, 31), (37717, 29))"
]
},
"execution_count": 38,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"train_X = train.ix[train['Lead_Creation_Date'] <= datetime(2015,7,17),]\n",
"hold_out = train.ix[train['Lead_Creation_Date'] > datetime(2015,7,17),]\n",
"test_X = test.copy()\n",
"train_X.shape, hold_out.shape, test_X.shape"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We fill the missing values with a simple placeholder value -999. No particular reason why I chose this number, I just wanted all the missing values to be treated same. "
]
},
{
"cell_type": "code",
"execution_count": 39,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"train_X = train_X.fillna(-999)\n",
"test_X = test_X.fillna(-999)\n",
"hold_out = hold_out.fillna(-999)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"##### Checking for some insights from the categorical features"
]
},
{
"cell_type": "code",
"execution_count": 40,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Gender Disbursed\n",
"0 0 36854\n",
" 1 318\n",
"1 0 48893\n",
" 1 955\n",
"Name: Disbursed, dtype: int64\n",
"Device_Type Disbursed\n",
"0 0 22440\n",
" 1 264\n",
"1 0 63307\n",
" 1 1009\n",
"Name: Disbursed, dtype: int64\n",
"Mobile_Verified Disbursed\n",
"0 0 30312\n",
" 1 227\n",
"1 0 55435\n",
" 1 1046\n",
"Name: Disbursed, dtype: int64\n",
"Filled_Form Disbursed\n",
"0 0 66704\n",
" 1 826\n",
"1 0 19043\n",
" 1 447\n",
"Name: Disbursed, dtype: int64\n",
"Var2 Disbursed\n",
"1 0 36652\n",
" 1 628\n",
"2 0 14034\n",
" 1 176\n",
"3 0 634\n",
"4 0 1294\n",
" 1 21\n",
"5 0 538\n",
" 1 6\n",
"6 0 32590\n",
" 1 442\n",
"999 0 5\n",
"Name: Disbursed, dtype: int64\n",
"Source Disbursed\n",
"0 0 37993\n",
" 1 574\n",
"4 0 1917\n",
" 1 14\n",
"7 0 29399\n",
" 1 486\n",
"8 0 1254\n",
" 1 47\n",
"11 0 1723\n",
" 1 1\n",
"16 0 4240\n",
" 1 92\n",
"17 0 297\n",
" 1 2\n",
"19 0 716\n",
" 1 4\n",
"20 0 493\n",
" 1 1\n",
"23 0 308\n",
"24 0 646\n",
" 1 4\n",
"25 0 207\n",
" 1 1\n",
"26 0 5556\n",
" 1 43\n",
"28 0 767\n",
" 1 2\n",
"999 0 231\n",
" 1 2\n",
"Name: Disbursed, dtype: int64\n",
"Var4 Disbursed\n",
"0 0 2545\n",
" 1 1\n",
"1 0 23760\n",
" 1 146\n",
"2 0 5784\n",
" 1 147\n",
"3 0 24852\n",
" 1 408\n",
"4 0 6457\n",
" 1 120\n",
"5 0 19819\n",
" 1 447\n",
"6 0 231\n",
" 1 1\n",
"7 0 2299\n",
" 1 3\n",
"Name: Disbursed, dtype: int64\n"
]
}
],
"source": [
"for var in ['Gender', 'Device_Type', 'Mobile_Verified', 'Filled_Form', 'Var2', 'Source', 'Var4']:\n",
" print train.groupby([var,'Disbursed'])['Disbursed'].count()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"One particular thing which stands out above, is that when either of Mobile_Verified and Device_type feature is 1 then there is a high probability that the loan will be disbursed. Since the number of positive loan disbursal is high in case of 'Mobile_verified' being 1 (1046 cases) than 'Device_type' being 1 (1009 cases), so we now proceed with dividing the train data by 'Mobile_verified' for our further analysis."
]
},
{
"cell_type": "code",
"execution_count": 41,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"(24719, 31)"
]
},
"execution_count": 41,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"trainX_mobN = train_X.ix[train_X['Mobile_Verified']==0,:]\n",
"trainX_mobN.shape"
]
},
{
"cell_type": "code",
"execution_count": 42,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"(46079, 31)"
]
},
"execution_count": 42,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"trainX_mobY = train_X.ix[train_X['Mobile_Verified']==1,:]\n",
"trainX_mobY.shape"
]
},
{
"cell_type": "code",
"execution_count": 43,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"(24719,)"
]
},
"execution_count": 43,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"trainY_mobN = train_X.ix[train_X['Mobile_Verified']==0,'Disbursed']\n",
"trainY_mobN.shape"
]
},
{
"cell_type": "code",
"execution_count": 44,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"(46079,)"
]
},
"execution_count": 44,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"trainY_mobY = train_X.ix[train_X['Mobile_Verified']==1,'Disbursed']\n",
"trainY_mobY.shape"
]
},
{
"cell_type": "code",
"execution_count": 45,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"(13270, 29)"
]
},
"execution_count": 45,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"test_mobN = test_X.ix[test_X['Mobile_Verified']==0,:]\n",
"test_mobN.shape"
]
},
{
"cell_type": "code",
"execution_count": 46,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"(24447, 29)"
]
},
"execution_count": 46,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"test_mobY = test_X.ix[test_X['Mobile_Verified']==1,:]\n",
"test_mobY.shape"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We drop the fields not required. I dropped 'DOB' & 'Lead_Creation_Date' as I had extracted the day, month and age values out of it, so I believe they dont add any more value to the analysis. "
]
},
{
"cell_type": "code",
"execution_count": 48,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"((24719, 26), (46079, 26), (13270, 26), (24447, 26))"
]
},
"execution_count": 48,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"trainX_mobN = trainX_mobN.drop(['ID', 'LoggedIn','DOB','Lead_Creation_Date', 'Disbursed'], 1)\n",
"trainX_mobY = trainX_mobY.drop(['ID', 'LoggedIn','DOB','Lead_Creation_Date', 'Disbursed'], 1)\n",
"test_mobN = test_mobN.drop(['ID','DOB','Lead_Creation_Date'], 1)\n",
"test_mobY = test_mobY.drop(['ID','DOB','Lead_Creation_Date'], 1)\n",
"trainX_mobN.shape, trainX_mobY.shape, test_mobN.shape, test_mobY.shape"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We divide the hold out set by 'MObile_Verified' as well. "
]
},
{
"cell_type": "code",
"execution_count": 49,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"(5820, 31)"
]
},
"execution_count": 49,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"holdoutX_mobN = hold_out.ix[hold_out['Mobile_Verified']==0,:]\n",
"holdoutX_mobN.shape"
]
},
{
"cell_type": "code",
"execution_count": 50,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"(10402, 31)"
]
},
"execution_count": 50,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"holdoutX_mobY = hold_out.ix[hold_out['Mobile_Verified']==1,:]\n",
"holdoutX_mobY.shape"
]
},
{
"cell_type": "code",
"execution_count": 51,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"(5820,)"
]
},
"execution_count": 51,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"holdoutY_mobN = hold_out.ix[hold_out['Mobile_Verified']==0,'Disbursed']\n",
"holdoutY_mobN.shape"
]
},
{
"cell_type": "code",
"execution_count": 52,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"(10402,)"
]
},
"execution_count": 52,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"holdoutY_mobY = hold_out.ix[hold_out['Mobile_Verified']==1,'Disbursed']\n",
"holdoutY_mobY.shape"
]
},
{
"cell_type": "code",
"execution_count": 53,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"holdoutX_mobN = holdoutX_mobN.drop(['ID', 'LoggedIn','DOB','Lead_Creation_Date', 'Disbursed'], 1)\n",
"holdoutX_mobY = holdoutX_mobY.drop(['ID', 'LoggedIn','DOB','Lead_Creation_Date', 'Disbursed'], 1)"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": true
},
"source": [
"### Building the model"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The problem at hand is one of classification with probabilities. Now we can employ logistic regression, naive bayes, support vector machines, random forest classifier, etc. However, increasingly in the recent past one tool which has outdone others, especially in such competitions is XGBOOST. \n",
"\n",
"[XGBOOST](https://github.com/dmlc/xgboost) is an optimized general purpose gradient boosting library. The library is parallelized, and also provides an optimized distributed version.\n",
"\n",
"It implements machine learning algorithms under the Gradient Boosting framework, including Generalized Linear Model (GLM) and Gradient Boosted Decision Trees (GBDT). XGBoost can also be distributed and scale to Terascale data. \n",
"\n",
"We shall use XGBOOST for our analysis. We first create a weight array which gives more weight to the data points which have loan disbursed. This will take care of the high skewness of our data as there are roughly only 10% loan disbursals in our data set. "
]
},
{
"cell_type": "code",
"execution_count": 54,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"trainX_mobN_weight = [2 if data == 1 else 1 for data in trainY_mobN]"
]
},
{
"cell_type": "code",
"execution_count": 55,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"trainX_mobY_weight = [2 if data == 1 else 1 for data in trainY_mobY]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We then set the various parameters, details of which can be found [here](https://github.com/dmlc/xgboost/blob/master/doc/parameter.md). You can also check out the documentation on tips for how to avoid overfitting and tune the model [here](https://xgboost.readthedocs.org/en/latest/param_tuning.html). "
]
},
{
"cell_type": "code",
"execution_count": 59,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"params = {}\n",
"params[\"objective\"] = \"binary:logistic\"\n",
"params[\"max_depth\"] = 5\n",
"params[\"eta\"] = 0.01 #higher is more conservative [0,1], if reduced then increase num_rounds\n",
"params[\"eval_metric \"] = 'auc'\n",
"params[\"seed\"] = 0\n",
"params[\"silent\"] = 1\n",
"plst = list(params.items())\n",
"num_rounds = 10000\n",
"\n",
"xgtrain = xgb.DMatrix(trainX_mobN, label=trainY_mobN, weight= trainX_mobN_weight)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Among the other parameters, eta (the learning rate), num_rounds of training and max_depth are the most important ones. To find the optimal values of these, we can use xgb.cv and run it for 10,000 iterations by setting eta at 0.01 and try this for variou max_depth to finally find the number of trees where overfitting occurs. "
]
},
{
"cell_type": "code",
"execution_count": 60,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"[0]\tcv-test-auc:0.835790+0.026039\tcv-train-auc:0.865063+0.033036\n",
"[1]\tcv-test-auc:0.837902+0.027286\tcv-train-auc:0.866750+0.034308\n",
"[2]\tcv-test-auc:0.852885+0.007698\tcv-train-auc:0.881495+0.011006\n",
"[3]\tcv-test-auc:0.852425+0.007253\tcv-train-auc:0.882929+0.011493\n",
"[4]\tcv-test-auc:0.863046+0.017608\tcv-train-auc:0.892005+0.008745\n",
"[5]\tcv-test-auc:0.863345+0.017537\tcv-train-auc:0.892090+0.008792\n",
"[6]\tcv-test-auc:0.863298+0.017534\tcv-train-auc:0.892097+0.008793\n",
"[7]\tcv-test-auc:0.863197+0.017513\tcv-train-auc:0.892283+0.008844\n",
"[8]\tcv-test-auc:0.863125+0.017542\tcv-train-auc:0.892462+0.008948\n",
"[9]\tcv-test-auc:0.863112+0.017564\tcv-train-auc:0.892387+0.008761\n",
"[10]\tcv-test-auc:0.862802+0.017560\tcv-train-auc:0.892802+0.008910\n",
"[11]\tcv-test-auc:0.862759+0.017563\tcv-train-auc:0.892804+0.008913\n",
"[12]\tcv-test-auc:0.862586+0.017742\tcv-train-auc:0.893417+0.007858\n",
"[13]\tcv-test-auc:0.862693+0.017733\tcv-train-auc:0.893807+0.008040\n",
"[14]\tcv-test-auc:0.863630+0.019047\tcv-train-auc:0.897081+0.011546\n",
"[15]\tcv-test-auc:0.863415+0.018918\tcv-train-auc:0.897184+0.011751\n",
"[16]\tcv-test-auc:0.863998+0.018679\tcv-train-auc:0.897907+0.011921\n",
"[17]\tcv-test-auc:0.863989+0.018571\tcv-train-auc:0.897958+0.011905\n",
"[18]\tcv-test-auc:0.863922+0.018640\tcv-train-auc:0.897926+0.011819\n",
"[19]\tcv-test-auc:0.864024+0.018537\tcv-train-auc:0.898462+0.011825\n",
"[20]\tcv-test-auc:0.864221+0.018142\tcv-train-auc:0.901214+0.010781\n",
"[21]\tcv-test-auc:0.864968+0.018707\tcv-train-auc:0.901229+0.010716\n",
"[22]\tcv-test-auc:0.865237+0.018730\tcv-train-auc:0.901326+0.010576\n",
"[23]\tcv-test-auc:0.865027+0.018692\tcv-train-auc:0.901915+0.010653\n",
"[24]\tcv-test-auc:0.865432+0.018563\tcv-train-auc:0.902635+0.010585\n",
"[25]\tcv-test-auc:0.865526+0.018410\tcv-train-auc:0.903042+0.010439\n",
"[26]\tcv-test-auc:0.865277+0.018825\tcv-train-auc:0.903190+0.010881\n",
"[27]\tcv-test-auc:0.865018+0.018580\tcv-train-auc:0.903787+0.011050\n",
"[28]\tcv-test-auc:0.865338+0.018449\tcv-train-auc:0.904115+0.011183\n",
"[29]\tcv-test-auc:0.875415+0.012663\tcv-train-auc:0.910892+0.005225\n",
"[30]\tcv-test-auc:0.875419+0.012513\tcv-train-auc:0.911706+0.005606\n",
"[31]\tcv-test-auc:0.875592+0.012364\tcv-train-auc:0.911897+0.005794\n",
"[32]\tcv-test-auc:0.875771+0.012372\tcv-train-auc:0.912021+0.005568\n",
"[33]\tcv-test-auc:0.875540+0.011960\tcv-train-auc:0.912512+0.006123\n",
"[34]\tcv-test-auc:0.875532+0.011873\tcv-train-auc:0.912672+0.005939\n",
"[35]\tcv-test-auc:0.875499+0.011823\tcv-train-auc:0.912904+0.005952\n",
"[36]\tcv-test-auc:0.875396+0.011761\tcv-train-auc:0.913000+0.005940\n",
"[37]\tcv-test-auc:0.875217+0.011736\tcv-train-auc:0.913447+0.005723\n",
"[38]\tcv-test-auc:0.875164+0.011582\tcv-train-auc:0.913664+0.005857\n",
"[39]\tcv-test-auc:0.875090+0.011883\tcv-train-auc:0.913623+0.005664\n",
"[40]\tcv-test-auc:0.877217+0.010618\tcv-train-auc:0.915911+0.004651\n",
"[41]\tcv-test-auc:0.877279+0.010716\tcv-train-auc:0.916044+0.004447\n",
"[42]\tcv-test-auc:0.877188+0.010813\tcv-train-auc:0.916310+0.004509\n",
"[43]\tcv-test-auc:0.877118+0.010841\tcv-train-auc:0.916561+0.004608\n",
"[44]\tcv-test-auc:0.877108+0.010703\tcv-train-auc:0.916938+0.004514\n",
"[45]\tcv-test-auc:0.877128+0.010625\tcv-train-auc:0.917190+0.004742\n",
"[46]\tcv-test-auc:0.877040+0.010619\tcv-train-auc:0.917397+0.004770\n",
"[47]\tcv-test-auc:0.877182+0.010537\tcv-train-auc:0.917627+0.004785\n",
"[48]\tcv-test-auc:0.877786+0.011251\tcv-train-auc:0.918322+0.005342\n",
"[49]\tcv-test-auc:0.877781+0.011233\tcv-train-auc:0.918478+0.005364\n",
"[50]\tcv-test-auc:0.877774+0.011258\tcv-train-auc:0.918599+0.005326\n",
"[51]\tcv-test-auc:0.877777+0.011223\tcv-train-auc:0.918736+0.005397\n",
"[52]\tcv-test-auc:0.877691+0.011343\tcv-train-auc:0.918767+0.005309\n",
"[53]\tcv-test-auc:0.877739+0.011498\tcv-train-auc:0.919153+0.005284\n",
"[54]\tcv-test-auc:0.877686+0.011645\tcv-train-auc:0.919330+0.005097\n",
"[55]\tcv-test-auc:0.877637+0.011537\tcv-train-auc:0.919415+0.005154\n",
"[56]\tcv-test-auc:0.877807+0.011551\tcv-train-auc:0.919477+0.005087\n",
"[57]\tcv-test-auc:0.878130+0.011988\tcv-train-auc:0.919799+0.005280\n",
"[58]\tcv-test-auc:0.877995+0.011727\tcv-train-auc:0.920047+0.005371\n",
"[59]\tcv-test-auc:0.878245+0.011975\tcv-train-auc:0.920177+0.005352\n",
"[60]\tcv-test-auc:0.878369+0.012043\tcv-train-auc:0.920384+0.005557\n",
"[61]\tcv-test-auc:0.878410+0.011901\tcv-train-auc:0.920464+0.005547\n",
"[62]\tcv-test-auc:0.878383+0.011775\tcv-train-auc:0.920733+0.005672\n",
"[63]\tcv-test-auc:0.878304+0.011726\tcv-train-auc:0.920998+0.005960\n",
"[64]\tcv-test-auc:0.878280+0.011382\tcv-train-auc:0.921398+0.006269\n",
"[65]\tcv-test-auc:0.878091+0.011503\tcv-train-auc:0.921804+0.006012\n",
"[66]\tcv-test-auc:0.877997+0.011388\tcv-train-auc:0.922036+0.006204\n",
"[67]\tcv-test-auc:0.883202+0.011663\tcv-train-auc:0.924453+0.006030\n",
"[68]\tcv-test-auc:0.883275+0.011627\tcv-train-auc:0.924623+0.006100\n",
"[69]\tcv-test-auc:0.883324+0.011556\tcv-train-auc:0.924856+0.006189\n",
"[70]\tcv-test-auc:0.883374+0.011604\tcv-train-auc:0.925157+0.006343\n",
"[71]\tcv-test-auc:0.883425+0.011613\tcv-train-auc:0.925482+0.006529\n",
"[72]\tcv-test-auc:0.883434+0.011553\tcv-train-auc:0.925767+0.006636\n",
"[73]\tcv-test-auc:0.883221+0.011721\tcv-train-auc:0.925955+0.006509\n",
"[74]\tcv-test-auc:0.883133+0.011718\tcv-train-auc:0.926091+0.006536\n",
"[75]\tcv-test-auc:0.883208+0.011707\tcv-train-auc:0.926193+0.006591\n",
"[76]\tcv-test-auc:0.883108+0.011589\tcv-train-auc:0.926297+0.006546\n",
"[77]\tcv-test-auc:0.883239+0.011657\tcv-train-auc:0.926374+0.006616\n",
"[78]\tcv-test-auc:0.883147+0.011680\tcv-train-auc:0.926525+0.006648\n",
"[79]\tcv-test-auc:0.882998+0.011592\tcv-train-auc:0.926550+0.006685\n",
"[80]\tcv-test-auc:0.882888+0.011679\tcv-train-auc:0.926863+0.006227\n",
"[81]\tcv-test-auc:0.882853+0.011527\tcv-train-auc:0.927087+0.006093\n",
"[82]\tcv-test-auc:0.882791+0.011482\tcv-train-auc:0.927149+0.006045\n",
"[83]\tcv-test-auc:0.882729+0.011388\tcv-train-auc:0.927354+0.006096\n",
"[84]\tcv-test-auc:0.882393+0.011244\tcv-train-auc:0.927463+0.006036\n",
"[85]\tcv-test-auc:0.882476+0.011415\tcv-train-auc:0.927663+0.006059\n",
"[86]\tcv-test-auc:0.882314+0.011277\tcv-train-auc:0.927899+0.006093\n",
"[87]\tcv-test-auc:0.882333+0.011333\tcv-train-auc:0.928068+0.006114\n",
"[88]\tcv-test-auc:0.882214+0.011394\tcv-train-auc:0.928318+0.006046\n",
"[89]\tcv-test-auc:0.882142+0.011455\tcv-train-auc:0.928446+0.006025\n",
"[90]\tcv-test-auc:0.882131+0.011465\tcv-train-auc:0.928582+0.006002\n",
"[91]\tcv-test-auc:0.882045+0.011375\tcv-train-auc:0.928657+0.005975\n",
"[92]\tcv-test-auc:0.883499+0.009045\tcv-train-auc:0.929798+0.004351\n",
"[93]\tcv-test-auc:0.883611+0.008984\tcv-train-auc:0.929928+0.004337\n",
"[94]\tcv-test-auc:0.883448+0.009098\tcv-train-auc:0.930085+0.004275\n",
"[95]\tcv-test-auc:0.883287+0.009079\tcv-train-auc:0.930142+0.004289\n",
"[96]\tcv-test-auc:0.883119+0.008994\tcv-train-auc:0.930206+0.004326\n",
"[97]\tcv-test-auc:0.883665+0.008194\tcv-train-auc:0.931210+0.002820\n",
"[98]\tcv-test-auc:0.883669+0.008226\tcv-train-auc:0.931346+0.002877\n",
"[99]\tcv-test-auc:0.883354+0.008544\tcv-train-auc:0.931496+0.002738\n",
"[100]\tcv-test-auc:0.883660+0.008621\tcv-train-auc:0.931848+0.002777\n",
"[101]\tcv-test-auc:0.883605+0.008651\tcv-train-auc:0.931916+0.002750\n",
"[102]\tcv-test-auc:0.883656+0.008609\tcv-train-auc:0.932036+0.002736\n",
"[103]\tcv-test-auc:0.883736+0.008590\tcv-train-auc:0.932313+0.002907\n",
"[104]\tcv-test-auc:0.883713+0.008591\tcv-train-auc:0.932379+0.002823\n",
"[105]\tcv-test-auc:0.883826+0.008520\tcv-train-auc:0.932602+0.002887\n",
"[106]\tcv-test-auc:0.883693+0.008552\tcv-train-auc:0.932658+0.002839\n",
"[107]\tcv-test-auc:0.883605+0.008598\tcv-train-auc:0.932746+0.002865\n",
"[108]\tcv-test-auc:0.883626+0.008484\tcv-train-auc:0.932955+0.002917\n",
"[109]\tcv-test-auc:0.883574+0.008442\tcv-train-auc:0.933182+0.002992\n",
"[110]\tcv-test-auc:0.883672+0.008577\tcv-train-auc:0.933466+0.002999\n",
"[111]\tcv-test-auc:0.883446+0.008644\tcv-train-auc:0.933748+0.002715\n",
"[112]\tcv-test-auc:0.883501+0.008504\tcv-train-auc:0.933899+0.002636\n",
"[113]\tcv-test-auc:0.883328+0.008446\tcv-train-auc:0.934005+0.002732\n",
"[114]\tcv-test-auc:0.883474+0.008498\tcv-train-auc:0.934150+0.002731\n",
"[115]\tcv-test-auc:0.883402+0.008512\tcv-train-auc:0.934274+0.002768\n",
"[116]\tcv-test-auc:0.883430+0.008538\tcv-train-auc:0.934925+0.003569\n",
"[117]\tcv-test-auc:0.883593+0.008594\tcv-train-auc:0.935082+0.003543\n",
"[118]\tcv-test-auc:0.883420+0.008356\tcv-train-auc:0.935219+0.003572\n",
"[119]\tcv-test-auc:0.883392+0.008414\tcv-train-auc:0.935354+0.003509\n",
"[120]\tcv-test-auc:0.883307+0.008340\tcv-train-auc:0.935506+0.003584\n",
"[121]\tcv-test-auc:0.883357+0.008340\tcv-train-auc:0.935807+0.003876\n",
"[122]\tcv-test-auc:0.883386+0.008346\tcv-train-auc:0.936010+0.003889\n",
"[123]\tcv-test-auc:0.883365+0.008386\tcv-train-auc:0.936090+0.003931\n",
"[124]\tcv-test-auc:0.883487+0.008008\tcv-train-auc:0.936294+0.003828\n",
"[125]\tcv-test-auc:0.883509+0.007880\tcv-train-auc:0.936418+0.003872\n",
"[126]\tcv-test-auc:0.883394+0.007796\tcv-train-auc:0.936646+0.003767\n",
"[127]\tcv-test-auc:0.883351+0.007752\tcv-train-auc:0.936771+0.003771\n",
"[128]\tcv-test-auc:0.883177+0.007772\tcv-train-auc:0.936906+0.003736\n",
"[129]\tcv-test-auc:0.883205+0.007647\tcv-train-auc:0.937013+0.003684\n",
"[130]\tcv-test-auc:0.883165+0.007649\tcv-train-auc:0.937056+0.003720\n",
"[131]\tcv-test-auc:0.883205+0.007656\tcv-train-auc:0.937268+0.003677\n",
"[132]\tcv-test-auc:0.883146+0.007726\tcv-train-auc:0.937367+0.003607\n",
"[133]\tcv-test-auc:0.883083+0.007676\tcv-train-auc:0.937562+0.003561\n",
"[134]\tcv-test-auc:0.883434+0.007200\tcv-train-auc:0.937880+0.003286\n",
"[135]\tcv-test-auc:0.883436+0.007186\tcv-train-auc:0.938121+0.003167\n",
"[136]\tcv-test-auc:0.883441+0.007195\tcv-train-auc:0.938201+0.003112\n",
"[137]\tcv-test-auc:0.883449+0.007057\tcv-train-auc:0.938216+0.003087\n",
"[138]\tcv-test-auc:0.883398+0.007126\tcv-train-auc:0.938469+0.002906\n",
"[139]\tcv-test-auc:0.883466+0.007163\tcv-train-auc:0.938617+0.002942\n",
"[140]\tcv-test-auc:0.883576+0.006928\tcv-train-auc:0.938792+0.002840\n",
"[141]\tcv-test-auc:0.883696+0.006984\tcv-train-auc:0.939482+0.003100\n",
"[142]\tcv-test-auc:0.883780+0.007016\tcv-train-auc:0.939602+0.003127\n",
"[143]\tcv-test-auc:0.883760+0.006930\tcv-train-auc:0.939817+0.003095\n",
"[144]\tcv-test-auc:0.883687+0.006938\tcv-train-auc:0.939925+0.003083\n",
"[145]\tcv-test-auc:0.883731+0.006761\tcv-train-auc:0.940093+0.002996\n",
"[146]\tcv-test-auc:0.883865+0.006569\tcv-train-auc:0.940255+0.002944\n",
"[147]\tcv-test-auc:0.883734+0.006515\tcv-train-auc:0.940242+0.002990\n",
"[148]\tcv-test-auc:0.883890+0.006583\tcv-train-auc:0.940340+0.002983\n",
"[149]\tcv-test-auc:0.884054+0.006497\tcv-train-auc:0.940449+0.002971\n",
"[150]\tcv-test-auc:0.883916+0.006608\tcv-train-auc:0.940541+0.002908\n",
"[151]\tcv-test-auc:0.883820+0.006570\tcv-train-auc:0.940765+0.002856\n",
"[152]\tcv-test-auc:0.883855+0.006578\tcv-train-auc:0.940831+0.002842\n",
"[153]\tcv-test-auc:0.883779+0.006477\tcv-train-auc:0.940947+0.002828\n",
"[154]\tcv-test-auc:0.884005+0.006639\tcv-train-auc:0.941027+0.002813\n",
"[155]\tcv-test-auc:0.884085+0.006607\tcv-train-auc:0.941077+0.002767\n",
"[156]\tcv-test-auc:0.883978+0.006602\tcv-train-auc:0.941288+0.002956\n",
"[157]\tcv-test-auc:0.884024+0.006562\tcv-train-auc:0.941351+0.002964\n",
"[158]\tcv-test-auc:0.884094+0.006587\tcv-train-auc:0.941409+0.002945\n",
"[159]\tcv-test-auc:0.884176+0.006971\tcv-train-auc:0.941657+0.003173\n",
"[160]\tcv-test-auc:0.884207+0.006961\tcv-train-auc:0.941685+0.003138\n",
"[161]\tcv-test-auc:0.887696+0.002020\tcv-train-auc:0.943742+0.002113\n",
"[162]\tcv-test-auc:0.887750+0.002024\tcv-train-auc:0.943757+0.002051\n",
"[163]\tcv-test-auc:0.887752+0.002044\tcv-train-auc:0.943837+0.002011\n",
"[164]\tcv-test-auc:0.888064+0.002164\tcv-train-auc:0.944062+0.002061\n",
"[165]\tcv-test-auc:0.888285+0.001884\tcv-train-auc:0.944743+0.002097\n",
"[166]\tcv-test-auc:0.888320+0.001946\tcv-train-auc:0.944796+0.001999\n",
"[167]\tcv-test-auc:0.888242+0.001975\tcv-train-auc:0.944910+0.002002\n",
"[168]\tcv-test-auc:0.888242+0.002020\tcv-train-auc:0.945049+0.002062\n",
"[169]\tcv-test-auc:0.888143+0.002008\tcv-train-auc:0.945339+0.002034\n",
"[170]\tcv-test-auc:0.888126+0.002080\tcv-train-auc:0.945471+0.002058\n",
"[171]\tcv-test-auc:0.888004+0.002198\tcv-train-auc:0.945684+0.002125\n",
"[172]\tcv-test-auc:0.888083+0.002305\tcv-train-auc:0.945834+0.002136\n",
"[173]\tcv-test-auc:0.888062+0.002341\tcv-train-auc:0.945991+0.002140\n",
"[174]\tcv-test-auc:0.888108+0.002368\tcv-train-auc:0.946089+0.002180\n",
"[175]\tcv-test-auc:0.888328+0.002605\tcv-train-auc:0.946352+0.002191\n",
"[176]\tcv-test-auc:0.888384+0.002661\tcv-train-auc:0.946416+0.002159\n",
"[177]\tcv-test-auc:0.888497+0.002770\tcv-train-auc:0.946717+0.002050\n",
"[178]\tcv-test-auc:0.888495+0.002883\tcv-train-auc:0.946831+0.002034\n",
"[179]\tcv-test-auc:0.888668+0.002979\tcv-train-auc:0.947004+0.002065\n",
"[180]\tcv-test-auc:0.888729+0.003005\tcv-train-auc:0.947084+0.002105\n",
"[181]\tcv-test-auc:0.888746+0.002989\tcv-train-auc:0.947201+0.002128\n",
"[182]\tcv-test-auc:0.888680+0.003146\tcv-train-auc:0.947293+0.002169\n",
"[183]\tcv-test-auc:0.888765+0.003068\tcv-train-auc:0.947456+0.002064\n",
"[184]\tcv-test-auc:0.888766+0.003112\tcv-train-auc:0.947567+0.002115\n",
"[185]\tcv-test-auc:0.888563+0.003070\tcv-train-auc:0.947697+0.002019\n",
"[186]\tcv-test-auc:0.888563+0.003081\tcv-train-auc:0.947724+0.002003\n",
"[187]\tcv-test-auc:0.888557+0.003152\tcv-train-auc:0.947799+0.002040\n",
"[188]\tcv-test-auc:0.888694+0.003044\tcv-train-auc:0.948088+0.001922\n",
"[189]\tcv-test-auc:0.889121+0.003571\tcv-train-auc:0.948370+0.001782\n",
"[190]\tcv-test-auc:0.888900+0.003369\tcv-train-auc:0.948429+0.001790\n",
"[191]\tcv-test-auc:0.888952+0.003583\tcv-train-auc:0.948702+0.001725\n",
"[192]\tcv-test-auc:0.888883+0.003576\tcv-train-auc:0.948842+0.001780\n",
"[193]\tcv-test-auc:0.888877+0.003752\tcv-train-auc:0.949028+0.001764\n",
"[194]\tcv-test-auc:0.888639+0.003623\tcv-train-auc:0.949148+0.001806\n",
"[195]\tcv-test-auc:0.888728+0.003704\tcv-train-auc:0.949393+0.001801\n",
"[196]\tcv-test-auc:0.888736+0.003746\tcv-train-auc:0.949522+0.001843\n",
"[197]\tcv-test-auc:0.888717+0.003769\tcv-train-auc:0.949663+0.001830\n",
"[198]\tcv-test-auc:0.888659+0.003782\tcv-train-auc:0.949802+0.001877\n",
"[199]\tcv-test-auc:0.888772+0.003791\tcv-train-auc:0.949888+0.001905\n",
"[200]\tcv-test-auc:0.888736+0.003832\tcv-train-auc:0.950008+0.001845\n",
"[201]\tcv-test-auc:0.888818+0.003726\tcv-train-auc:0.950143+0.001925\n",
"[202]\tcv-test-auc:0.888817+0.003725\tcv-train-auc:0.950244+0.001939\n",
"[203]\tcv-test-auc:0.888785+0.003746\tcv-train-auc:0.950340+0.001966\n",
"[204]\tcv-test-auc:0.888735+0.003718\tcv-train-auc:0.950443+0.001960\n",
"[205]\tcv-test-auc:0.888757+0.003618\tcv-train-auc:0.950539+0.002030\n",
"[206]\tcv-test-auc:0.888770+0.003651\tcv-train-auc:0.950630+0.002059\n",
"[207]\tcv-test-auc:0.888682+0.003736\tcv-train-auc:0.950840+0.002085\n",
"[208]\tcv-test-auc:0.888708+0.003696\tcv-train-auc:0.950966+0.002031\n",
"[209]\tcv-test-auc:0.888667+0.003618\tcv-train-auc:0.951056+0.002046\n",
"[210]\tcv-test-auc:0.888610+0.003656\tcv-train-auc:0.951145+0.002116\n",
"[211]\tcv-test-auc:0.888610+0.003637\tcv-train-auc:0.951241+0.002140\n",
"[212]\tcv-test-auc:0.888722+0.003527\tcv-train-auc:0.951333+0.002179\n",
"[213]\tcv-test-auc:0.888675+0.003532\tcv-train-auc:0.951391+0.002171\n",
"[214]\tcv-test-auc:0.888571+0.003599\tcv-train-auc:0.951462+0.002234\n",
"[215]\tcv-test-auc:0.888650+0.003600\tcv-train-auc:0.951573+0.002296\n",
"[216]\tcv-test-auc:0.888632+0.003607\tcv-train-auc:0.951658+0.002323\n",
"[217]\tcv-test-auc:0.888597+0.003593\tcv-train-auc:0.951720+0.002370\n",
"[218]\tcv-test-auc:0.888503+0.003562\tcv-train-auc:0.951800+0.002396\n",
"[219]\tcv-test-auc:0.888520+0.003549\tcv-train-auc:0.951880+0.002436\n",
"[220]\tcv-test-auc:0.888426+0.003584\tcv-train-auc:0.951967+0.002489\n",
"[221]\tcv-test-auc:0.888389+0.003669\tcv-train-auc:0.952018+0.002521\n",
"[222]\tcv-test-auc:0.888328+0.003758\tcv-train-auc:0.952134+0.002499\n",
"[223]\tcv-test-auc:0.888292+0.003569\tcv-train-auc:0.952209+0.002511\n",
"[224]\tcv-test-auc:0.888266+0.003657\tcv-train-auc:0.952287+0.002519\n",
"[225]\tcv-test-auc:0.888257+0.003652\tcv-train-auc:0.952333+0.002562\n",
"[226]\tcv-test-auc:0.888365+0.003736\tcv-train-auc:0.952446+0.002582\n",
"[227]\tcv-test-auc:0.888305+0.003626\tcv-train-auc:0.952493+0.002606\n",
"[228]\tcv-test-auc:0.888344+0.003610\tcv-train-auc:0.952564+0.002593\n",
"[229]\tcv-test-auc:0.888301+0.003655\tcv-train-auc:0.952700+0.002661\n",
"[230]\tcv-test-auc:0.888418+0.003550\tcv-train-auc:0.952824+0.002623\n",
"[231]\tcv-test-auc:0.888474+0.003521\tcv-train-auc:0.952892+0.002646\n",
"[232]\tcv-test-auc:0.888473+0.003486\tcv-train-auc:0.952937+0.002666\n",
"[233]\tcv-test-auc:0.888468+0.003431\tcv-train-auc:0.953053+0.002628\n",
"[234]\tcv-test-auc:0.888534+0.003401\tcv-train-auc:0.953150+0.002639\n",
"[235]\tcv-test-auc:0.888572+0.003566\tcv-train-auc:0.953270+0.002516\n",
"[236]\tcv-test-auc:0.888653+0.003581\tcv-train-auc:0.953345+0.002479\n",
"[237]\tcv-test-auc:0.888613+0.003570\tcv-train-auc:0.953484+0.002527\n",
"[238]\tcv-test-auc:0.888737+0.003604\tcv-train-auc:0.953569+0.002507\n",
"[239]\tcv-test-auc:0.888808+0.003866\tcv-train-auc:0.953737+0.002514\n",
"[240]\tcv-test-auc:0.888806+0.003799\tcv-train-auc:0.953881+0.002538\n",
"[241]\tcv-test-auc:0.888808+0.003813\tcv-train-auc:0.954010+0.002572\n",
"[242]\tcv-test-auc:0.888851+0.003903\tcv-train-auc:0.954232+0.002547\n",
"[243]\tcv-test-auc:0.888916+0.003804\tcv-train-auc:0.954332+0.002587\n",
"[244]\tcv-test-auc:0.888940+0.003941\tcv-train-auc:0.954433+0.002599\n",
"[245]\tcv-test-auc:0.888987+0.003923\tcv-train-auc:0.954557+0.002664\n",
"[246]\tcv-test-auc:0.888927+0.003935\tcv-train-auc:0.954827+0.002678\n",
"[247]\tcv-test-auc:0.888964+0.003887\tcv-train-auc:0.954994+0.002643\n",
"[248]\tcv-test-auc:0.888967+0.004003\tcv-train-auc:0.955011+0.002646\n",
"[249]\tcv-test-auc:0.888894+0.003985\tcv-train-auc:0.955205+0.002677\n",
"[250]\tcv-test-auc:0.888958+0.004088\tcv-train-auc:0.955315+0.002708\n",
"[251]\tcv-test-auc:0.888902+0.004125\tcv-train-auc:0.955395+0.002717\n",
"[252]\tcv-test-auc:0.888853+0.004140\tcv-train-auc:0.955527+0.002730\n",
"[253]\tcv-test-auc:0.888875+0.004080\tcv-train-auc:0.955661+0.002793\n",
"[254]\tcv-test-auc:0.888799+0.004118\tcv-train-auc:0.955767+0.002821\n",
"[255]\tcv-test-auc:0.888698+0.004170\tcv-train-auc:0.955937+0.002663\n",
"[256]\tcv-test-auc:0.888657+0.004191\tcv-train-auc:0.956000+0.002685\n",
"[257]\tcv-test-auc:0.888640+0.004221\tcv-train-auc:0.956100+0.002737\n",
"[258]\tcv-test-auc:0.888599+0.004254\tcv-train-auc:0.956282+0.002712\n",
"[259]\tcv-test-auc:0.888686+0.004211\tcv-train-auc:0.956461+0.002852\n",
"[260]\tcv-test-auc:0.888488+0.004499\tcv-train-auc:0.956573+0.002896\n",
"[261]\tcv-test-auc:0.888490+0.004521\tcv-train-auc:0.956674+0.002924\n",
"[262]\tcv-test-auc:0.888582+0.004688\tcv-train-auc:0.956754+0.002911\n",
"[263]\tcv-test-auc:0.888516+0.004746\tcv-train-auc:0.956859+0.002872\n",
"[264]\tcv-test-auc:0.888456+0.004735\tcv-train-auc:0.956973+0.002903\n",
"[265]\tcv-test-auc:0.888391+0.004705\tcv-train-auc:0.957036+0.002927\n",
"[266]\tcv-test-auc:0.888403+0.004717\tcv-train-auc:0.957177+0.002917\n",
"[267]\tcv-test-auc:0.888419+0.004690\tcv-train-auc:0.957264+0.002944\n",
"[268]\tcv-test-auc:0.888440+0.004781\tcv-train-auc:0.957328+0.002982\n",
"[269]\tcv-test-auc:0.888347+0.004662\tcv-train-auc:0.957422+0.003003\n",
"[270]\tcv-test-auc:0.888243+0.004669\tcv-train-auc:0.957503+0.003027\n",
"[271]\tcv-test-auc:0.888278+0.004693\tcv-train-auc:0.957571+0.003025\n",
"[272]\tcv-test-auc:0.888324+0.004818\tcv-train-auc:0.957645+0.003057\n",
"[273]\tcv-test-auc:0.888300+0.004899\tcv-train-auc:0.957722+0.003064\n",
"[274]\tcv-test-auc:0.888379+0.004833\tcv-train-auc:0.957803+0.003083\n",
"[275]\tcv-test-auc:0.888321+0.004917\tcv-train-auc:0.957883+0.003066\n",
"[276]\tcv-test-auc:0.888365+0.004831\tcv-train-auc:0.957953+0.003095\n",
"[277]\tcv-test-auc:0.888405+0.004859\tcv-train-auc:0.958111+0.003038\n",
"[278]\tcv-test-auc:0.888354+0.004863\tcv-train-auc:0.958177+0.003030\n",
"[279]\tcv-test-auc:0.888388+0.004857\tcv-train-auc:0.958305+0.002989\n",
"[280]\tcv-test-auc:0.888385+0.004830\tcv-train-auc:0.958362+0.002990\n",
"[281]\tcv-test-auc:0.888369+0.004902\tcv-train-auc:0.958450+0.003007\n",
"[282]\tcv-test-auc:0.888219+0.004902\tcv-train-auc:0.958540+0.003027\n",
"[283]\tcv-test-auc:0.888224+0.004959\tcv-train-auc:0.958664+0.003055\n",
"[284]\tcv-test-auc:0.888136+0.004947\tcv-train-auc:0.958734+0.003048\n",
"[285]\tcv-test-auc:0.888202+0.004897\tcv-train-auc:0.958838+0.003038\n",
"[286]\tcv-test-auc:0.887939+0.005032\tcv-train-auc:0.958936+0.003045\n",
"[287]\tcv-test-auc:0.887954+0.005010\tcv-train-auc:0.959032+0.003065\n",
"[288]\tcv-test-auc:0.887973+0.005078\tcv-train-auc:0.959209+0.002977\n",
"[289]\tcv-test-auc:0.887886+0.005117\tcv-train-auc:0.959392+0.002933\n",
"[290]\tcv-test-auc:0.887938+0.005142\tcv-train-auc:0.959547+0.002895\n",
"[291]\tcv-test-auc:0.887971+0.005200\tcv-train-auc:0.959689+0.002904\n",
"[292]\tcv-test-auc:0.887872+0.005236\tcv-train-auc:0.959802+0.002891\n",
"[293]\tcv-test-auc:0.887891+0.005279\tcv-train-auc:0.959920+0.002887\n",
"[294]\tcv-test-auc:0.887817+0.005313\tcv-train-auc:0.960034+0.002853\n",
"[295]\tcv-test-auc:0.887804+0.005327\tcv-train-auc:0.960121+0.002835\n",
"[296]\tcv-test-auc:0.887736+0.005310\tcv-train-auc:0.960256+0.002846\n",
"[297]\tcv-test-auc:0.887706+0.005316\tcv-train-auc:0.960366+0.002863\n",
"[298]\tcv-test-auc:0.887754+0.005429\tcv-train-auc:0.960474+0.002847\n",
"[299]\tcv-test-auc:0.887762+0.005441\tcv-train-auc:0.960577+0.002871\n",
"[300]\tcv-test-auc:0.887803+0.005472\tcv-train-auc:0.960688+0.002879\n",
"[301]\tcv-test-auc:0.887742+0.005439\tcv-train-auc:0.960802+0.002878\n",
"[302]\tcv-test-auc:0.887827+0.005296\tcv-train-auc:0.960901+0.002883\n",
"[303]\tcv-test-auc:0.887840+0.005277\tcv-train-auc:0.961032+0.002891\n",
"[304]\tcv-test-auc:0.887868+0.005290\tcv-train-auc:0.961157+0.002885\n",
"[305]\tcv-test-auc:0.887862+0.005286\tcv-train-auc:0.961231+0.002865\n",
"[306]\tcv-test-auc:0.887862+0.005274\tcv-train-auc:0.961389+0.002871\n",
"[307]\tcv-test-auc:0.887852+0.005262\tcv-train-auc:0.961473+0.002882\n",
"[308]\tcv-test-auc:0.887766+0.005248\tcv-train-auc:0.961563+0.002891\n",
"[309]\tcv-test-auc:0.887824+0.005135\tcv-train-auc:0.961686+0.002943\n",
"[310]\tcv-test-auc:0.887743+0.005160\tcv-train-auc:0.961765+0.002955\n",
"[311]\tcv-test-auc:0.887607+0.005261\tcv-train-auc:0.961870+0.002974\n",
"[312]\tcv-test-auc:0.887672+0.005220\tcv-train-auc:0.961985+0.002937\n",
"[313]\tcv-test-auc:0.887648+0.005199\tcv-train-auc:0.962128+0.002896\n",
"[314]\tcv-test-auc:0.887619+0.005200\tcv-train-auc:0.962258+0.002865\n",
"[315]\tcv-test-auc:0.887660+0.005180\tcv-train-auc:0.962359+0.002866\n",
"[316]\tcv-test-auc:0.887618+0.005182\tcv-train-auc:0.962465+0.002882\n",
"[317]\tcv-test-auc:0.887530+0.005193\tcv-train-auc:0.962568+0.002894\n",
"[318]\tcv-test-auc:0.887454+0.005195\tcv-train-auc:0.962671+0.002920\n",
"[319]\tcv-test-auc:0.887519+0.005222\tcv-train-auc:0.962783+0.002902\n",
"[320]\tcv-test-auc:0.887427+0.005324\tcv-train-auc:0.962936+0.002890\n",
"[321]\tcv-test-auc:0.887413+0.005297\tcv-train-auc:0.963048+0.002874\n",
"[322]\tcv-test-auc:0.887345+0.005390\tcv-train-auc:0.963161+0.002865\n",
"[323]\tcv-test-auc:0.887323+0.005371\tcv-train-auc:0.963269+0.002865\n",
"[324]\tcv-test-auc:0.887216+0.005545\tcv-train-auc:0.963370+0.002856\n",
"[325]\tcv-test-auc:0.887167+0.005553\tcv-train-auc:0.963466+0.002842\n",
"[326]\tcv-test-auc:0.887174+0.005639\tcv-train-auc:0.963569+0.002847\n",
"[327]\tcv-test-auc:0.887132+0.005568\tcv-train-auc:0.963663+0.002837\n",
"[328]\tcv-test-auc:0.887087+0.005565\tcv-train-auc:0.963772+0.002812\n",
"[329]\tcv-test-auc:0.887033+0.005609\tcv-train-auc:0.963878+0.002802\n",
"[330]\tcv-test-auc:0.887012+0.005590\tcv-train-auc:0.963976+0.002793\n",
"[331]\tcv-test-auc:0.886997+0.005591\tcv-train-auc:0.964083+0.002779\n",
"[332]\tcv-test-auc:0.887005+0.005640\tcv-train-auc:0.964191+0.002774\n",
"[333]\tcv-test-auc:0.886988+0.005622\tcv-train-auc:0.964275+0.002762\n",
"[334]\tcv-test-auc:0.886956+0.005689\tcv-train-auc:0.964406+0.002705\n",
"[335]\tcv-test-auc:0.886946+0.005630\tcv-train-auc:0.964493+0.002725\n",
"[336]\tcv-test-auc:0.886941+0.005684\tcv-train-auc:0.964613+0.002747\n",
"[337]\tcv-test-auc:0.886892+0.005677\tcv-train-auc:0.964713+0.002678\n",
"[338]\tcv-test-auc:0.886847+0.005709\tcv-train-auc:0.964796+0.002685\n",
"[339]\tcv-test-auc:0.886876+0.005726\tcv-train-auc:0.964877+0.002685\n",
"[340]\tcv-test-auc:0.886760+0.005770\tcv-train-auc:0.964993+0.002649\n",
"[341]\tcv-test-auc:0.886629+0.005841\tcv-train-auc:0.965080+0.002658\n",
"[342]\tcv-test-auc:0.886547+0.005859\tcv-train-auc:0.965198+0.002635\n",
"[343]\tcv-test-auc:0.886472+0.005775\tcv-train-auc:0.965283+0.002634\n",
"[344]\tcv-test-auc:0.886467+0.005720\tcv-train-auc:0.965365+0.002633\n",
"[345]\tcv-test-auc:0.886513+0.005715\tcv-train-auc:0.965456+0.002633\n",
"[346]\tcv-test-auc:0.886437+0.005697\tcv-train-auc:0.965548+0.002637\n",
"[347]\tcv-test-auc:0.886386+0.005610\tcv-train-auc:0.965642+0.002673\n",
"[348]\tcv-test-auc:0.886416+0.005508\tcv-train-auc:0.965760+0.002669\n",
"[349]\tcv-test-auc:0.886560+0.005426\tcv-train-auc:0.965871+0.002672\n",
"[350]\tcv-test-auc:0.886509+0.005370\tcv-train-auc:0.965982+0.002679\n",
"[351]\tcv-test-auc:0.886579+0.005285\tcv-train-auc:0.966067+0.002670\n",
"[352]\tcv-test-auc:0.886573+0.005263\tcv-train-auc:0.966154+0.002674\n",
"[353]\tcv-test-auc:0.886680+0.005208\tcv-train-auc:0.966278+0.002702\n",
"[354]\tcv-test-auc:0.886708+0.005233\tcv-train-auc:0.966415+0.002669\n",
"[355]\tcv-test-auc:0.886779+0.005204\tcv-train-auc:0.966531+0.002620\n",
"[356]\tcv-test-auc:0.886927+0.005246\tcv-train-auc:0.966649+0.002571\n",
"[357]\tcv-test-auc:0.887074+0.005232\tcv-train-auc:0.966733+0.002551\n",
"[358]\tcv-test-auc:0.887217+0.005202\tcv-train-auc:0.966837+0.002556\n",
"[359]\tcv-test-auc:0.887257+0.005170\tcv-train-auc:0.966932+0.002576\n",
"[360]\tcv-test-auc:0.887342+0.005193\tcv-train-auc:0.967039+0.002572\n",
"[361]\tcv-test-auc:0.887371+0.005190\tcv-train-auc:0.967129+0.002582\n",
"[362]\tcv-test-auc:0.887445+0.005203\tcv-train-auc:0.967218+0.002582\n",
"[363]\tcv-test-auc:0.887516+0.005207\tcv-train-auc:0.967308+0.002595\n",
"[364]\tcv-test-auc:0.887502+0.005198\tcv-train-auc:0.967375+0.002603\n",
"[365]\tcv-test-auc:0.887567+0.005172\tcv-train-auc:0.967473+0.002587\n",
"[366]\tcv-test-auc:0.887633+0.005164\tcv-train-auc:0.967541+0.002589\n",
"[367]\tcv-test-auc:0.887649+0.005094\tcv-train-auc:0.967604+0.002590\n",
"[368]\tcv-test-auc:0.887725+0.005086\tcv-train-auc:0.967673+0.002593\n",
"[369]\tcv-test-auc:0.887782+0.005137\tcv-train-auc:0.967726+0.002596\n",
"[370]\tcv-test-auc:0.887813+0.005076\tcv-train-auc:0.967806+0.002618\n",
"[371]\tcv-test-auc:0.887879+0.005097\tcv-train-auc:0.967904+0.002556\n",
"[372]\tcv-test-auc:0.887988+0.005159\tcv-train-auc:0.967983+0.002572\n",
"[373]\tcv-test-auc:0.888035+0.005131\tcv-train-auc:0.968031+0.002572\n",
"[374]\tcv-test-auc:0.888130+0.005182\tcv-train-auc:0.968100+0.002568\n",
"[375]\tcv-test-auc:0.888157+0.005133\tcv-train-auc:0.968159+0.002567\n",
"[376]\tcv-test-auc:0.888200+0.005097\tcv-train-auc:0.968216+0.002567\n",
"[377]\tcv-test-auc:0.888296+0.005111\tcv-train-auc:0.968309+0.002525\n",
"[378]\tcv-test-auc:0.888387+0.005091\tcv-train-auc:0.968379+0.002537\n",
"[379]\tcv-test-auc:0.888382+0.005069\tcv-train-auc:0.968421+0.002542\n",
"[380]\tcv-test-auc:0.888401+0.005064\tcv-train-auc:0.968489+0.002563\n",
"[381]\tcv-test-auc:0.888419+0.005071\tcv-train-auc:0.968597+0.002518\n",
"[382]\tcv-test-auc:0.888454+0.005021\tcv-train-auc:0.968647+0.002519\n",
"[383]\tcv-test-auc:0.888467+0.004984\tcv-train-auc:0.968728+0.002497\n",
"[384]\tcv-test-auc:0.888453+0.004918\tcv-train-auc:0.968784+0.002519\n",
"[385]\tcv-test-auc:0.888432+0.004914\tcv-train-auc:0.968866+0.002469\n",
"[386]\tcv-test-auc:0.888514+0.004888\tcv-train-auc:0.968902+0.002465\n",
"[387]\tcv-test-auc:0.888579+0.004864\tcv-train-auc:0.968974+0.002474\n",
"[388]\tcv-test-auc:0.888525+0.004849\tcv-train-auc:0.969037+0.002497\n",
"[389]\tcv-test-auc:0.888496+0.004855\tcv-train-auc:0.969084+0.002493\n",
"[390]\tcv-test-auc:0.888532+0.004892\tcv-train-auc:0.969173+0.002465\n",
"[391]\tcv-test-auc:0.888482+0.004944\tcv-train-auc:0.969233+0.002479\n",
"[392]\tcv-test-auc:0.888477+0.004879\tcv-train-auc:0.969273+0.002489\n",
"[393]\tcv-test-auc:0.888565+0.004897\tcv-train-auc:0.969330+0.002488\n",
"[394]\tcv-test-auc:0.888561+0.004863\tcv-train-auc:0.969404+0.002454\n",
"[395]\tcv-test-auc:0.888590+0.004889\tcv-train-auc:0.969465+0.002461\n",
"[396]\tcv-test-auc:0.888610+0.004879\tcv-train-auc:0.969518+0.002469\n",
"[397]\tcv-test-auc:0.888583+0.004935\tcv-train-auc:0.969578+0.002468\n",
"[398]\tcv-test-auc:0.888606+0.004902\tcv-train-auc:0.969626+0.002464\n",
"[399]\tcv-test-auc:0.888648+0.004943\tcv-train-auc:0.969695+0.002432\n",
"[400]\tcv-test-auc:0.888673+0.004931\tcv-train-auc:0.969739+0.002432\n",
"[401]\tcv-test-auc:0.888682+0.004953\tcv-train-auc:0.969784+0.002439\n",
"[402]\tcv-test-auc:0.888707+0.004962\tcv-train-auc:0.969825+0.002437\n",
"[403]\tcv-test-auc:0.888774+0.005008\tcv-train-auc:0.969928+0.002375\n",
"[404]\tcv-test-auc:0.888787+0.004988\tcv-train-auc:0.969968+0.002372\n",
"[405]\tcv-test-auc:0.888756+0.004996\tcv-train-auc:0.970003+0.002393\n",
"[406]\tcv-test-auc:0.888804+0.004974\tcv-train-auc:0.970044+0.002387\n",
"[407]\tcv-test-auc:0.888832+0.004931\tcv-train-auc:0.970093+0.002397\n",
"[408]\tcv-test-auc:0.888834+0.004972\tcv-train-auc:0.970179+0.002339\n",
"[409]\tcv-test-auc:0.888891+0.005009\tcv-train-auc:0.970219+0.002345\n",
"[410]\tcv-test-auc:0.888913+0.004995\tcv-train-auc:0.970263+0.002349\n",
"[411]\tcv-test-auc:0.888944+0.004988\tcv-train-auc:0.970306+0.002358\n",
"[412]\tcv-test-auc:0.888968+0.005059\tcv-train-auc:0.970337+0.002363\n",
"[413]\tcv-test-auc:0.889005+0.005029\tcv-train-auc:0.970378+0.002373\n",
"[414]\tcv-test-auc:0.889031+0.005060\tcv-train-auc:0.970417+0.002378\n",
"[415]\tcv-test-auc:0.889043+0.005048\tcv-train-auc:0.970457+0.002380\n",
"[416]\tcv-test-auc:0.889071+0.005056\tcv-train-auc:0.970512+0.002356\n",
"[417]\tcv-test-auc:0.889163+0.005001\tcv-train-auc:0.970568+0.002379\n",
"[418]\tcv-test-auc:0.889165+0.005053\tcv-train-auc:0.970603+0.002383\n",
"[419]\tcv-test-auc:0.889186+0.005044\tcv-train-auc:0.970677+0.002381\n",
"[420]\tcv-test-auc:0.889230+0.005012\tcv-train-auc:0.970711+0.002390\n",
"[421]\tcv-test-auc:0.889258+0.005031\tcv-train-auc:0.970749+0.002401\n",
"[422]\tcv-test-auc:0.889254+0.004975\tcv-train-auc:0.970784+0.002413\n",
"[423]\tcv-test-auc:0.889301+0.004978\tcv-train-auc:0.970839+0.002399\n",
"[424]\tcv-test-auc:0.889337+0.004974\tcv-train-auc:0.970876+0.002407\n",
"[425]\tcv-test-auc:0.889319+0.005017\tcv-train-auc:0.970925+0.002421\n",
"[426]\tcv-test-auc:0.889370+0.005019\tcv-train-auc:0.970974+0.002415\n",
"[427]\tcv-test-auc:0.889377+0.005053\tcv-train-auc:0.971018+0.002426\n",
"[428]\tcv-test-auc:0.889414+0.005084\tcv-train-auc:0.971055+0.002429\n",
"[429]\tcv-test-auc:0.889434+0.005061\tcv-train-auc:0.971095+0.002418\n",
"[430]\tcv-test-auc:0.889436+0.005110\tcv-train-auc:0.971128+0.002427\n",
"[431]\tcv-test-auc:0.889416+0.005146\tcv-train-auc:0.971164+0.002429\n",
"[432]\tcv-test-auc:0.889451+0.005183\tcv-train-auc:0.971208+0.002401\n",
"[433]\tcv-test-auc:0.889451+0.005200\tcv-train-auc:0.971258+0.002410\n",
"[434]\tcv-test-auc:0.889493+0.005215\tcv-train-auc:0.971295+0.002421\n",
"[435]\tcv-test-auc:0.889504+0.005260\tcv-train-auc:0.971330+0.002429\n",
"[436]\tcv-test-auc:0.889508+0.005271\tcv-train-auc:0.971356+0.002437\n",
"[437]\tcv-test-auc:0.889550+0.005293\tcv-train-auc:0.971416+0.002459\n",
"[438]\tcv-test-auc:0.889521+0.005283\tcv-train-auc:0.971454+0.002465\n",
"[439]\tcv-test-auc:0.889555+0.005311\tcv-train-auc:0.971487+0.002491\n",
"[440]\tcv-test-auc:0.889556+0.005263\tcv-train-auc:0.971515+0.002490\n",
"[441]\tcv-test-auc:0.889556+0.005235\tcv-train-auc:0.971562+0.002513\n",
"[442]\tcv-test-auc:0.889707+0.005281\tcv-train-auc:0.971613+0.002532\n",
"[443]\tcv-test-auc:0.889726+0.005298\tcv-train-auc:0.971635+0.002535\n",
"[444]\tcv-test-auc:0.889681+0.005303\tcv-train-auc:0.971679+0.002549\n",
"[445]\tcv-test-auc:0.889812+0.005353\tcv-train-auc:0.971714+0.002555\n",
"[446]\tcv-test-auc:0.889856+0.005354\tcv-train-auc:0.971733+0.002558\n",
"[447]\tcv-test-auc:0.889871+0.005341\tcv-train-auc:0.971776+0.002580\n",
"[448]\tcv-test-auc:0.889921+0.005386\tcv-train-auc:0.971808+0.002609\n",
"[449]\tcv-test-auc:0.889930+0.005370\tcv-train-auc:0.971852+0.002613\n",
"[450]\tcv-test-auc:0.889918+0.005362\tcv-train-auc:0.971899+0.002632\n",
"[451]\tcv-test-auc:0.889951+0.005299\tcv-train-auc:0.971950+0.002652\n",
"[452]\tcv-test-auc:0.890000+0.005282\tcv-train-auc:0.971972+0.002650\n",
"[453]\tcv-test-auc:0.890022+0.005271\tcv-train-auc:0.972029+0.002678\n",
"[454]\tcv-test-auc:0.890035+0.005271\tcv-train-auc:0.972044+0.002671\n",
"[455]\tcv-test-auc:0.889997+0.005288\tcv-train-auc:0.972083+0.002676\n",
"[456]\tcv-test-auc:0.890048+0.005312\tcv-train-auc:0.972114+0.002704\n",
"[457]\tcv-test-auc:0.890093+0.005325\tcv-train-auc:0.972153+0.002710\n",
"[458]\tcv-test-auc:0.890112+0.005330\tcv-train-auc:0.972184+0.002724\n",
"[459]\tcv-test-auc:0.890117+0.005371\tcv-train-auc:0.972244+0.002740\n",
"[460]\tcv-test-auc:0.890162+0.005370\tcv-train-auc:0.972267+0.002752\n",
"[461]\tcv-test-auc:0.890182+0.005373\tcv-train-auc:0.972308+0.002757\n",
"[462]\tcv-test-auc:0.890181+0.005378\tcv-train-auc:0.972319+0.002774\n",
"[463]\tcv-test-auc:0.890227+0.005395\tcv-train-auc:0.972387+0.002801\n",
"[464]\tcv-test-auc:0.890250+0.005391\tcv-train-auc:0.972401+0.002798\n",
"[465]\tcv-test-auc:0.890228+0.005388\tcv-train-auc:0.972445+0.002812\n",
"[466]\tcv-test-auc:0.890269+0.005469\tcv-train-auc:0.972499+0.002813\n",
"[467]\tcv-test-auc:0.890301+0.005485\tcv-train-auc:0.972526+0.002786\n",
"[468]\tcv-test-auc:0.890297+0.005501\tcv-train-auc:0.972569+0.002792\n",
"[469]\tcv-test-auc:0.890307+0.005517\tcv-train-auc:0.972621+0.002806\n",
"[470]\tcv-test-auc:0.890365+0.005504\tcv-train-auc:0.972659+0.002814\n",
"[471]\tcv-test-auc:0.890368+0.005501\tcv-train-auc:0.972701+0.002852\n",
"[472]\tcv-test-auc:0.890391+0.005473\tcv-train-auc:0.972736+0.002851\n",
"[473]\tcv-test-auc:0.890404+0.005548\tcv-train-auc:0.972768+0.002836\n",
"[474]\tcv-test-auc:0.890422+0.005562\tcv-train-auc:0.972808+0.002867\n",
"[475]\tcv-test-auc:0.890423+0.005601\tcv-train-auc:0.972833+0.002878\n",
"[476]\tcv-test-auc:0.890469+0.005612\tcv-train-auc:0.972860+0.002875\n",
"[477]\tcv-test-auc:0.890542+0.005627\tcv-train-auc:0.972874+0.002894\n",
"[478]\tcv-test-auc:0.890504+0.005675\tcv-train-auc:0.972925+0.002908\n",
"[479]\tcv-test-auc:0.890498+0.005723\tcv-train-auc:0.972992+0.002907\n",
"[480]\tcv-test-auc:0.890527+0.005739\tcv-train-auc:0.973015+0.002917\n",
"[481]\tcv-test-auc:0.890532+0.005790\tcv-train-auc:0.973037+0.002941\n",
"[482]\tcv-test-auc:0.890568+0.005803\tcv-train-auc:0.973063+0.002941\n",
"[483]\tcv-test-auc:0.890568+0.005829\tcv-train-auc:0.973111+0.002971\n",
"[484]\tcv-test-auc:0.890605+0.005830\tcv-train-auc:0.973141+0.002947\n",
"[485]\tcv-test-auc:0.890657+0.005894\tcv-train-auc:0.973167+0.002940\n",
"[486]\tcv-test-auc:0.890667+0.005954\tcv-train-auc:0.973211+0.002951\n",
"[487]\tcv-test-auc:0.890707+0.005963\tcv-train-auc:0.973243+0.002956\n",
"[488]\tcv-test-auc:0.890702+0.005971\tcv-train-auc:0.973273+0.002943\n",
"[489]\tcv-test-auc:0.890744+0.005961\tcv-train-auc:0.973308+0.002949\n",
"[490]\tcv-test-auc:0.890764+0.005949\tcv-train-auc:0.973344+0.002961\n",
"[491]\tcv-test-auc:0.890778+0.005970\tcv-train-auc:0.973360+0.002972\n",
"[492]\tcv-test-auc:0.890766+0.005977\tcv-train-auc:0.973428+0.002941\n",
"[493]\tcv-test-auc:0.890774+0.006021\tcv-train-auc:0.973457+0.002953\n",
"[494]\tcv-test-auc:0.890797+0.006021\tcv-train-auc:0.973495+0.002955\n",
"[495]\tcv-test-auc:0.890798+0.006035\tcv-train-auc:0.973530+0.002940\n",
"[496]\tcv-test-auc:0.890835+0.006031\tcv-train-auc:0.973599+0.002939\n",
"[497]\tcv-test-auc:0.890872+0.006082\tcv-train-auc:0.973619+0.002942\n",
"[498]\tcv-test-auc:0.890862+0.006109\tcv-train-auc:0.973660+0.002949\n",
"[499]\tcv-test-auc:0.890869+0.006145\tcv-train-auc:0.973705+0.002899\n",
"[500]\tcv-test-auc:0.890843+0.006155\tcv-train-auc:0.973755+0.002918\n",
"[501]\tcv-test-auc:0.890872+0.006152\tcv-train-auc:0.973788+0.002911\n",
"[502]\tcv-test-auc:0.890887+0.006136\tcv-train-auc:0.973851+0.002906\n",
"[503]\tcv-test-auc:0.890908+0.006133\tcv-train-auc:0.973877+0.002909\n",
"[504]\tcv-test-auc:0.890962+0.006158\tcv-train-auc:0.973902+0.002915\n",
"[505]\tcv-test-auc:0.891023+0.006152\tcv-train-auc:0.973927+0.002923\n",
"[506]\tcv-test-auc:0.891003+0.006179\tcv-train-auc:0.973972+0.002902\n",
"[507]\tcv-test-auc:0.891023+0.006173\tcv-train-auc:0.974024+0.002909\n",
"[508]\tcv-test-auc:0.891042+0.006209\tcv-train-auc:0.974077+0.002873\n",
"[509]\tcv-test-auc:0.891051+0.006230\tcv-train-auc:0.974110+0.002896\n",
"[510]\tcv-test-auc:0.891061+0.006222\tcv-train-auc:0.974139+0.002902\n",
"[511]\tcv-test-auc:0.891079+0.006268\tcv-train-auc:0.974189+0.002898\n",
"[512]\tcv-test-auc:0.891097+0.006245\tcv-train-auc:0.974224+0.002873\n",
"[513]\tcv-test-auc:0.891150+0.006231\tcv-train-auc:0.974267+0.002886\n",
"[514]\tcv-test-auc:0.891233+0.006192\tcv-train-auc:0.974331+0.002832\n",
"[515]\tcv-test-auc:0.891258+0.006236\tcv-train-auc:0.974377+0.002834\n",
"[516]\tcv-test-auc:0.891255+0.006226\tcv-train-auc:0.974418+0.002867\n",
"[517]\tcv-test-auc:0.891278+0.006239\tcv-train-auc:0.974440+0.002868\n",
"[518]\tcv-test-auc:0.891262+0.006231\tcv-train-auc:0.974498+0.002858\n",
"[519]\tcv-test-auc:0.891326+0.006235\tcv-train-auc:0.974567+0.002835\n",
"[520]\tcv-test-auc:0.891337+0.006248\tcv-train-auc:0.974593+0.002830\n",
"[521]\tcv-test-auc:0.891345+0.006232\tcv-train-auc:0.974647+0.002880\n",
"[522]\tcv-test-auc:0.891393+0.006260\tcv-train-auc:0.974679+0.002891\n",
"[523]\tcv-test-auc:0.891502+0.006236\tcv-train-auc:0.974734+0.002860\n",
"[524]\tcv-test-auc:0.891504+0.006220\tcv-train-auc:0.974787+0.002884\n",
"[525]\tcv-test-auc:0.891492+0.006250\tcv-train-auc:0.974798+0.002902\n",
"[526]\tcv-test-auc:0.891446+0.006256\tcv-train-auc:0.974842+0.002876\n",
"[527]\tcv-test-auc:0.891471+0.006273\tcv-train-auc:0.974861+0.002881\n",
"[528]\tcv-test-auc:0.891506+0.006281\tcv-train-auc:0.974885+0.002906\n",
"[529]\tcv-test-auc:0.891443+0.006267\tcv-train-auc:0.974905+0.002927\n",
"[530]\tcv-test-auc:0.891482+0.006255\tcv-train-auc:0.974974+0.002888\n",
"[531]\tcv-test-auc:0.891538+0.006281\tcv-train-auc:0.975007+0.002894\n",
"[532]\tcv-test-auc:0.891494+0.006238\tcv-train-auc:0.975019+0.002901\n",
"[533]\tcv-test-auc:0.891522+0.006201\tcv-train-auc:0.975082+0.002855\n",
"[534]\tcv-test-auc:0.891572+0.006180\tcv-train-auc:0.975105+0.002873\n",
"[535]\tcv-test-auc:0.891557+0.006195\tcv-train-auc:0.975143+0.002888\n",
"[536]\tcv-test-auc:0.891562+0.006178\tcv-train-auc:0.975191+0.002844\n",
"[537]\tcv-test-auc:0.891574+0.006121\tcv-train-auc:0.975219+0.002872\n",
"[538]\tcv-test-auc:0.891548+0.006148\tcv-train-auc:0.975257+0.002860\n",
"[539]\tcv-test-auc:0.891545+0.006174\tcv-train-auc:0.975303+0.002834\n",
"[540]\tcv-test-auc:0.891532+0.006158\tcv-train-auc:0.975334+0.002852\n",
"[541]\tcv-test-auc:0.891498+0.006145\tcv-train-auc:0.975364+0.002866\n",
"[542]\tcv-test-auc:0.891552+0.006103\tcv-train-auc:0.975402+0.002833\n",
"[543]\tcv-test-auc:0.891577+0.006137\tcv-train-auc:0.975439+0.002826\n",
"[544]\tcv-test-auc:0.891581+0.006089\tcv-train-auc:0.975443+0.002826\n",
"[545]\tcv-test-auc:0.891588+0.006076\tcv-train-auc:0.975471+0.002812\n",
"[546]\tcv-test-auc:0.891616+0.006036\tcv-train-auc:0.975547+0.002777\n",
"[547]\tcv-test-auc:0.891655+0.006068\tcv-train-auc:0.975618+0.002808\n",
"[548]\tcv-test-auc:0.891682+0.006071\tcv-train-auc:0.975646+0.002815\n",
"[549]\tcv-test-auc:0.891677+0.006059\tcv-train-auc:0.975689+0.002822\n",
"[550]\tcv-test-auc:0.891721+0.006048\tcv-train-auc:0.975732+0.002774\n",
"[551]\tcv-test-auc:0.891755+0.006032\tcv-train-auc:0.975764+0.002785\n",
"[552]\tcv-test-auc:0.891724+0.006014\tcv-train-auc:0.975809+0.002790\n",
"[553]\tcv-test-auc:0.891701+0.005963\tcv-train-auc:0.975827+0.002796\n",
"[554]\tcv-test-auc:0.891727+0.005921\tcv-train-auc:0.975874+0.002761\n",
"[555]\tcv-test-auc:0.891706+0.005936\tcv-train-auc:0.975909+0.002756\n",
"[556]\tcv-test-auc:0.891734+0.005924\tcv-train-auc:0.975931+0.002765\n",
"[557]\tcv-test-auc:0.891748+0.005927\tcv-train-auc:0.975966+0.002755\n",
"[558]\tcv-test-auc:0.891778+0.005927\tcv-train-auc:0.975999+0.002772\n",
"[559]\tcv-test-auc:0.891773+0.005877\tcv-train-auc:0.976031+0.002782\n",
"[560]\tcv-test-auc:0.891783+0.005857\tcv-train-auc:0.976107+0.002748\n",
"[561]\tcv-test-auc:0.891791+0.005877\tcv-train-auc:0.976128+0.002756\n",
"[562]\tcv-test-auc:0.891779+0.005849\tcv-train-auc:0.976158+0.002758\n",
"[563]\tcv-test-auc:0.891789+0.005802\tcv-train-auc:0.976179+0.002760\n",
"[564]\tcv-test-auc:0.891815+0.005811\tcv-train-auc:0.976225+0.002728\n",
"[565]\tcv-test-auc:0.891785+0.005760\tcv-train-auc:0.976256+0.002733\n",
"[566]\tcv-test-auc:0.891790+0.005763\tcv-train-auc:0.976296+0.002756\n",
"[567]\tcv-test-auc:0.891763+0.005750\tcv-train-auc:0.976337+0.002772\n",
"[568]\tcv-test-auc:0.891797+0.005759\tcv-train-auc:0.976353+0.002790\n",
"[569]\tcv-test-auc:0.891810+0.005723\tcv-train-auc:0.976437+0.002768\n",
"[570]\tcv-test-auc:0.891840+0.005681\tcv-train-auc:0.976449+0.002775\n",
"[571]\tcv-test-auc:0.891852+0.005650\tcv-train-auc:0.976493+0.002783\n",
"[572]\tcv-test-auc:0.891878+0.005657\tcv-train-auc:0.976508+0.002776\n",
"[573]\tcv-test-auc:0.891871+0.005627\tcv-train-auc:0.976529+0.002793\n",
"[574]\tcv-test-auc:0.891874+0.005612\tcv-train-auc:0.976555+0.002808\n",
"[575]\tcv-test-auc:0.891860+0.005603\tcv-train-auc:0.976598+0.002816\n",
"[576]\tcv-test-auc:0.891880+0.005617\tcv-train-auc:0.976637+0.002824\n",
"[577]\tcv-test-auc:0.891877+0.005611\tcv-train-auc:0.976661+0.002835\n",
"[578]\tcv-test-auc:0.891910+0.005600\tcv-train-auc:0.976684+0.002851\n",
"[579]\tcv-test-auc:0.891939+0.005581\tcv-train-auc:0.976700+0.002846\n",
"[580]\tcv-test-auc:0.891953+0.005585\tcv-train-auc:0.976754+0.002807\n",
"[581]\tcv-test-auc:0.891888+0.005512\tcv-train-auc:0.976785+0.002822\n",
"[582]\tcv-test-auc:0.891902+0.005467\tcv-train-auc:0.976818+0.002835\n",
"[583]\tcv-test-auc:0.891885+0.005495\tcv-train-auc:0.976848+0.002836\n",
"[584]\tcv-test-auc:0.891909+0.005496\tcv-train-auc:0.976871+0.002821\n",
"[585]\tcv-test-auc:0.891919+0.005496\tcv-train-auc:0.976900+0.002826\n",
"[586]\tcv-test-auc:0.891935+0.005459\tcv-train-auc:0.976915+0.002834\n",
"[587]\tcv-test-auc:0.891950+0.005476\tcv-train-auc:0.976966+0.002857\n",
"[588]\tcv-test-auc:0.891896+0.005410\tcv-train-auc:0.977010+0.002871\n",
"[589]\tcv-test-auc:0.891903+0.005364\tcv-train-auc:0.977046+0.002831\n",
"[590]\tcv-test-auc:0.891939+0.005343\tcv-train-auc:0.977078+0.002844\n",
"[591]\tcv-test-auc:0.891939+0.005356\tcv-train-auc:0.977105+0.002850\n",
"[592]\tcv-test-auc:0.891953+0.005338\tcv-train-auc:0.977127+0.002861\n",
"[593]\tcv-test-auc:0.891955+0.005347\tcv-train-auc:0.977174+0.002826\n",
"[594]\tcv-test-auc:0.891950+0.005360\tcv-train-auc:0.977183+0.002820\n",
"[595]\tcv-test-auc:0.891931+0.005339\tcv-train-auc:0.977214+0.002830\n",
"[596]\tcv-test-auc:0.891944+0.005359\tcv-train-auc:0.977231+0.002840\n",
"[597]\tcv-test-auc:0.891928+0.005352\tcv-train-auc:0.977272+0.002854\n",
"[598]\tcv-test-auc:0.891958+0.005293\tcv-train-auc:0.977329+0.002818\n",
"[599]\tcv-test-auc:0.891972+0.005313\tcv-train-auc:0.977374+0.002841\n",
"[600]\tcv-test-auc:0.892011+0.005280\tcv-train-auc:0.977441+0.002782\n",
"[601]\tcv-test-auc:0.892033+0.005266\tcv-train-auc:0.977473+0.002786\n",
"[602]\tcv-test-auc:0.891999+0.005283\tcv-train-auc:0.977508+0.002787\n",
"[603]\tcv-test-auc:0.892000+0.005224\tcv-train-auc:0.977526+0.002796\n",
"[604]\tcv-test-auc:0.891989+0.005228\tcv-train-auc:0.977537+0.002803\n",
"[605]\tcv-test-auc:0.891981+0.005229\tcv-train-auc:0.977581+0.002803\n",
"[606]\tcv-test-auc:0.892003+0.005258\tcv-train-auc:0.977614+0.002815\n",
"[607]\tcv-test-auc:0.892037+0.005210\tcv-train-auc:0.977683+0.002790\n",
"[608]\tcv-test-auc:0.892007+0.005198\tcv-train-auc:0.977687+0.002793\n",
"[609]\tcv-test-auc:0.892032+0.005181\tcv-train-auc:0.977708+0.002785\n",
"[610]\tcv-test-auc:0.892107+0.005202\tcv-train-auc:0.977750+0.002797\n",
"[611]\tcv-test-auc:0.892093+0.005216\tcv-train-auc:0.977763+0.002804\n",
"[612]\tcv-test-auc:0.892095+0.005254\tcv-train-auc:0.977790+0.002811\n",
"[613]\tcv-test-auc:0.892135+0.005208\tcv-train-auc:0.977822+0.002781\n",
"[614]\tcv-test-auc:0.892058+0.005181\tcv-train-auc:0.977858+0.002795\n",
"[615]\tcv-test-auc:0.892109+0.005266\tcv-train-auc:0.977877+0.002803\n",
"[616]\tcv-test-auc:0.892123+0.005260\tcv-train-auc:0.977935+0.002783\n",
"[617]\tcv-test-auc:0.892123+0.005238\tcv-train-auc:0.977990+0.002726\n",
"[618]\tcv-test-auc:0.892127+0.005246\tcv-train-auc:0.978029+0.002696\n",
"[619]\tcv-test-auc:0.892157+0.005222\tcv-train-auc:0.978048+0.002700\n",
"[620]\tcv-test-auc:0.892209+0.005266\tcv-train-auc:0.978064+0.002706\n",
"[621]\tcv-test-auc:0.892163+0.005245\tcv-train-auc:0.978133+0.002664\n",
"[622]\tcv-test-auc:0.892154+0.005253\tcv-train-auc:0.978142+0.002679\n",
"[623]\tcv-test-auc:0.892140+0.005244\tcv-train-auc:0.978191+0.002636\n",
"[624]\tcv-test-auc:0.892195+0.005286\tcv-train-auc:0.978216+0.002653\n",
"[625]\tcv-test-auc:0.892165+0.005323\tcv-train-auc:0.978282+0.002632\n",
"[626]\tcv-test-auc:0.892223+0.005317\tcv-train-auc:0.978315+0.002611\n",
"[627]\tcv-test-auc:0.892208+0.005294\tcv-train-auc:0.978342+0.002629\n",
"[628]\tcv-test-auc:0.892219+0.005361\tcv-train-auc:0.978402+0.002608\n",
"[629]\tcv-test-auc:0.892245+0.005394\tcv-train-auc:0.978435+0.002617\n",
"[630]\tcv-test-auc:0.892240+0.005402\tcv-train-auc:0.978475+0.002601\n",
"[631]\tcv-test-auc:0.892228+0.005395\tcv-train-auc:0.978500+0.002600\n",
"[632]\tcv-test-auc:0.892226+0.005410\tcv-train-auc:0.978520+0.002617\n",
"[633]\tcv-test-auc:0.892225+0.005404\tcv-train-auc:0.978561+0.002581\n",
"[634]\tcv-test-auc:0.892227+0.005381\tcv-train-auc:0.978580+0.002593\n",
"[635]\tcv-test-auc:0.892176+0.005387\tcv-train-auc:0.978655+0.002536\n",
"[636]\tcv-test-auc:0.892174+0.005428\tcv-train-auc:0.978697+0.002550\n",
"[637]\tcv-test-auc:0.892152+0.005445\tcv-train-auc:0.978726+0.002558\n",
"[638]\tcv-test-auc:0.892112+0.005459\tcv-train-auc:0.978749+0.002561\n",
"[639]\tcv-test-auc:0.892152+0.005467\tcv-train-auc:0.978813+0.002529\n",
"[640]\tcv-test-auc:0.892122+0.005445\tcv-train-auc:0.978882+0.002521\n",
"[641]\tcv-test-auc:0.892114+0.005489\tcv-train-auc:0.978914+0.002538\n",
"[642]\tcv-test-auc:0.892134+0.005502\tcv-train-auc:0.978934+0.002546\n",
"[643]\tcv-test-auc:0.892099+0.005478\tcv-train-auc:0.978976+0.002562\n",
"[644]\tcv-test-auc:0.892076+0.005541\tcv-train-auc:0.979043+0.002518\n",
"[645]\tcv-test-auc:0.892112+0.005485\tcv-train-auc:0.979070+0.002495\n",
"[646]\tcv-test-auc:0.892118+0.005496\tcv-train-auc:0.979099+0.002508\n",
"[647]\tcv-test-auc:0.892074+0.005498\tcv-train-auc:0.979129+0.002516\n",
"[648]\tcv-test-auc:0.892053+0.005517\tcv-train-auc:0.979140+0.002526\n",
"[649]\tcv-test-auc:0.892085+0.005541\tcv-train-auc:0.979218+0.002489\n",
"[650]\tcv-test-auc:0.892103+0.005586\tcv-train-auc:0.979289+0.002461\n",
"[651]\tcv-test-auc:0.892094+0.005562\tcv-train-auc:0.979353+0.002494\n",
"[652]\tcv-test-auc:0.892084+0.005585\tcv-train-auc:0.979401+0.002471\n",
"[653]\tcv-test-auc:0.892091+0.005573\tcv-train-auc:0.979432+0.002478\n",
"[654]\tcv-test-auc:0.892090+0.005579\tcv-train-auc:0.979446+0.002489\n",
"[655]\tcv-test-auc:0.892077+0.005529\tcv-train-auc:0.979469+0.002508\n",
"[656]\tcv-test-auc:0.892058+0.005536\tcv-train-auc:0.979492+0.002516\n",
"[657]\tcv-test-auc:0.892057+0.005589\tcv-train-auc:0.979538+0.002484\n",
"[658]\tcv-test-auc:0.892104+0.005594\tcv-train-auc:0.979565+0.002494\n",
"[659]\tcv-test-auc:0.892110+0.005548\tcv-train-auc:0.979595+0.002503\n",
"[660]\tcv-test-auc:0.892077+0.005584\tcv-train-auc:0.979636+0.002458\n",
"[661]\tcv-test-auc:0.892058+0.005603\tcv-train-auc:0.979655+0.002463\n",
"[662]\tcv-test-auc:0.892049+0.005598\tcv-train-auc:0.979687+0.002470\n",
"[663]\tcv-test-auc:0.892013+0.005599\tcv-train-auc:0.979699+0.002472\n",
"[664]\tcv-test-auc:0.892005+0.005588\tcv-train-auc:0.979742+0.002416\n",
"[665]\tcv-test-auc:0.892006+0.005591\tcv-train-auc:0.979776+0.002423\n",
"[666]\tcv-test-auc:0.892012+0.005612\tcv-train-auc:0.979808+0.002436\n",
"[667]\tcv-test-auc:0.892033+0.005606\tcv-train-auc:0.979827+0.002450\n",
"[668]\tcv-test-auc:0.892005+0.005623\tcv-train-auc:0.979844+0.002464\n",
"[669]\tcv-test-auc:0.891992+0.005590\tcv-train-auc:0.979868+0.002463\n",
"[670]\tcv-test-auc:0.891970+0.005616\tcv-train-auc:0.979894+0.002461\n",
"[671]\tcv-test-auc:0.892002+0.005585\tcv-train-auc:0.979949+0.002406\n",
"[672]\tcv-test-auc:0.891950+0.005579\tcv-train-auc:0.980010+0.002358\n",
"[673]\tcv-test-auc:0.891944+0.005650\tcv-train-auc:0.980058+0.002332\n",
"[674]\tcv-test-auc:0.891932+0.005651\tcv-train-auc:0.980096+0.002306\n",
"[675]\tcv-test-auc:0.891938+0.005635\tcv-train-auc:0.980139+0.002318\n",
"[676]\tcv-test-auc:0.891946+0.005633\tcv-train-auc:0.980166+0.002291\n",
"[677]\tcv-test-auc:0.891948+0.005608\tcv-train-auc:0.980195+0.002305\n",
"[678]\tcv-test-auc:0.891958+0.005618\tcv-train-auc:0.980215+0.002314\n",
"[679]\tcv-test-auc:0.891968+0.005635\tcv-train-auc:0.980225+0.002309\n",
"[680]\tcv-test-auc:0.891938+0.005648\tcv-train-auc:0.980248+0.002322\n",
"[681]\tcv-test-auc:0.891902+0.005676\tcv-train-auc:0.980278+0.002340\n",
"[682]\tcv-test-auc:0.891882+0.005720\tcv-train-auc:0.980335+0.002289\n",
"[683]\tcv-test-auc:0.891888+0.005714\tcv-train-auc:0.980352+0.002306\n",
"[684]\tcv-test-auc:0.891870+0.005698\tcv-train-auc:0.980376+0.002314\n",
"[685]\tcv-test-auc:0.891875+0.005702\tcv-train-auc:0.980403+0.002323\n",
"[686]\tcv-test-auc:0.891870+0.005720\tcv-train-auc:0.980453+0.002281\n",
"[687]\tcv-test-auc:0.891876+0.005714\tcv-train-auc:0.980477+0.002299\n",
"[688]\tcv-test-auc:0.891909+0.005735\tcv-train-auc:0.980530+0.002259\n",
"[689]\tcv-test-auc:0.891900+0.005734\tcv-train-auc:0.980578+0.002241\n",
"[690]\tcv-test-auc:0.891878+0.005768\tcv-train-auc:0.980614+0.002247\n",
"[691]\tcv-test-auc:0.891877+0.005765\tcv-train-auc:0.980641+0.002259\n",
"[692]\tcv-test-auc:0.891854+0.005775\tcv-train-auc:0.980676+0.002213\n",
"[693]\tcv-test-auc:0.891872+0.005790\tcv-train-auc:0.980683+0.002215\n",
"[694]\tcv-test-auc:0.891860+0.005779\tcv-train-auc:0.980726+0.002170\n",
"[695]\tcv-test-auc:0.891876+0.005758\tcv-train-auc:0.980785+0.002136\n",
"[696]\tcv-test-auc:0.891836+0.005770\tcv-train-auc:0.980826+0.002111\n",
"[697]\tcv-test-auc:0.891870+0.005775\tcv-train-auc:0.980881+0.002103\n",
"[698]\tcv-test-auc:0.891901+0.005816\tcv-train-auc:0.980907+0.002115\n",
"[699]\tcv-test-auc:0.891852+0.005811\tcv-train-auc:0.980936+0.002132\n",
"[700]\tcv-test-auc:0.891837+0.005801\tcv-train-auc:0.980967+0.002149\n",
"[701]\tcv-test-auc:0.891866+0.005816\tcv-train-auc:0.981004+0.002168\n",
"[702]\tcv-test-auc:0.891889+0.005817\tcv-train-auc:0.981012+0.002172\n",
"[703]\tcv-test-auc:0.891910+0.005826\tcv-train-auc:0.981055+0.002138\n",
"[704]\tcv-test-auc:0.891914+0.005824\tcv-train-auc:0.981073+0.002148\n",
"[705]\tcv-test-auc:0.891876+0.005843\tcv-train-auc:0.981112+0.002115\n",
"[706]\tcv-test-auc:0.891878+0.005879\tcv-train-auc:0.981145+0.002135\n",
"[707]\tcv-test-auc:0.891849+0.005870\tcv-train-auc:0.981183+0.002111\n",
"[708]\tcv-test-auc:0.891824+0.005869\tcv-train-auc:0.981220+0.002137\n",
"[709]\tcv-test-auc:0.891822+0.005852\tcv-train-auc:0.981233+0.002144\n",
"[710]\tcv-test-auc:0.891815+0.005879\tcv-train-auc:0.981245+0.002150\n",
"[711]\tcv-test-auc:0.891836+0.005878\tcv-train-auc:0.981255+0.002153\n",
"[712]\tcv-test-auc:0.891803+0.005893\tcv-train-auc:0.981297+0.002181\n",
"[713]\tcv-test-auc:0.891810+0.005861\tcv-train-auc:0.981329+0.002155\n",
"[714]\tcv-test-auc:0.891835+0.005924\tcv-train-auc:0.981379+0.002133\n",
"[715]\tcv-test-auc:0.891829+0.005958\tcv-train-auc:0.981412+0.002136\n",
"[716]\tcv-test-auc:0.891818+0.005964\tcv-train-auc:0.981434+0.002151\n",
"[717]\tcv-test-auc:0.891822+0.005970\tcv-train-auc:0.981487+0.002138\n",
"[718]\tcv-test-auc:0.891814+0.005923\tcv-train-auc:0.981527+0.002148\n",
"[719]\tcv-test-auc:0.891866+0.006006\tcv-train-auc:0.981563+0.002150\n",
"[720]\tcv-test-auc:0.891837+0.006020\tcv-train-auc:0.981605+0.002164\n",
"[721]\tcv-test-auc:0.891881+0.006060\tcv-train-auc:0.981642+0.002177\n",
"[722]\tcv-test-auc:0.891900+0.006076\tcv-train-auc:0.981685+0.002189\n",
"[723]\tcv-test-auc:0.891895+0.006059\tcv-train-auc:0.981694+0.002186\n",
"[724]\tcv-test-auc:0.891887+0.006077\tcv-train-auc:0.981719+0.002155\n",
"[725]\tcv-test-auc:0.891880+0.006093\tcv-train-auc:0.981765+0.002172\n",
"[726]\tcv-test-auc:0.891901+0.006095\tcv-train-auc:0.981791+0.002180\n",
"[727]\tcv-test-auc:0.891867+0.006066\tcv-train-auc:0.981842+0.002185\n",
"[728]\tcv-test-auc:0.891889+0.006074\tcv-train-auc:0.981870+0.002198\n",
"[729]\tcv-test-auc:0.891908+0.006109\tcv-train-auc:0.981900+0.002211\n",
"[730]\tcv-test-auc:0.891914+0.006149\tcv-train-auc:0.981945+0.002230\n",
"[731]\tcv-test-auc:0.891950+0.006163\tcv-train-auc:0.981991+0.002224\n",
"[732]\tcv-test-auc:0.891962+0.006170\tcv-train-auc:0.982045+0.002213\n",
"[733]\tcv-test-auc:0.891973+0.006172\tcv-train-auc:0.982101+0.002197\n",
"[734]\tcv-test-auc:0.891972+0.006182\tcv-train-auc:0.982154+0.002166\n",
"[735]\tcv-test-auc:0.891928+0.006187\tcv-train-auc:0.982190+0.002183\n",
"[736]\tcv-test-auc:0.891966+0.006204\tcv-train-auc:0.982219+0.002195\n",
"[737]\tcv-test-auc:0.891993+0.006208\tcv-train-auc:0.982253+0.002208\n",
"[738]\tcv-test-auc:0.891980+0.006227\tcv-train-auc:0.982270+0.002205\n",
"[739]\tcv-test-auc:0.891970+0.006248\tcv-train-auc:0.982335+0.002188\n",
"[740]\tcv-test-auc:0.892006+0.006197\tcv-train-auc:0.982368+0.002174\n",
"[741]\tcv-test-auc:0.892043+0.006221\tcv-train-auc:0.982390+0.002154\n",
"[742]\tcv-test-auc:0.891981+0.006212\tcv-train-auc:0.982409+0.002172\n",
"[743]\tcv-test-auc:0.892010+0.006207\tcv-train-auc:0.982449+0.002135\n",
"[744]\tcv-test-auc:0.892022+0.006175\tcv-train-auc:0.982470+0.002140\n",
"[745]\tcv-test-auc:0.892026+0.006163\tcv-train-auc:0.982485+0.002139\n",
"[746]\tcv-test-auc:0.892029+0.006196\tcv-train-auc:0.982517+0.002161\n",
"[747]\tcv-test-auc:0.892008+0.006187\tcv-train-auc:0.982528+0.002154\n",
"[748]\tcv-test-auc:0.892091+0.006225\tcv-train-auc:0.982568+0.002164\n",
"[749]\tcv-test-auc:0.892082+0.006240\tcv-train-auc:0.982577+0.002166\n",
"[750]\tcv-test-auc:0.892087+0.006251\tcv-train-auc:0.982612+0.002168\n",
"[751]\tcv-test-auc:0.892078+0.006243\tcv-train-auc:0.982673+0.002152\n",
"[752]\tcv-test-auc:0.892067+0.006255\tcv-train-auc:0.982693+0.002151\n",
"[753]\tcv-test-auc:0.892081+0.006295\tcv-train-auc:0.982723+0.002160\n",
"[754]\tcv-test-auc:0.892102+0.006314\tcv-train-auc:0.982783+0.002155\n",
"[755]\tcv-test-auc:0.892167+0.006299\tcv-train-auc:0.982814+0.002141\n",
"[756]\tcv-test-auc:0.893227+0.007833\tcv-train-auc:0.982845+0.002149\n",
"[757]\tcv-test-auc:0.893229+0.007860\tcv-train-auc:0.982857+0.002146\n",
"[758]\tcv-test-auc:0.893188+0.007853\tcv-train-auc:0.982877+0.002158\n",
"[759]\tcv-test-auc:0.893203+0.007855\tcv-train-auc:0.982915+0.002165\n",
"[760]\tcv-test-auc:0.893061+0.007645\tcv-train-auc:0.982978+0.002148\n",
"[761]\tcv-test-auc:0.893062+0.007633\tcv-train-auc:0.982987+0.002160\n",
"[762]\tcv-test-auc:0.893056+0.007641\tcv-train-auc:0.983037+0.002171\n",
"[763]\tcv-test-auc:0.893066+0.007669\tcv-train-auc:0.983058+0.002175\n",
"[764]\tcv-test-auc:0.893106+0.007699\tcv-train-auc:0.983105+0.002189\n",
"[765]\tcv-test-auc:0.893119+0.007712\tcv-train-auc:0.983117+0.002191\n",
"[766]\tcv-test-auc:0.893105+0.007679\tcv-train-auc:0.983163+0.002186\n",
"[767]\tcv-test-auc:0.893102+0.007702\tcv-train-auc:0.983189+0.002188\n",
"[768]\tcv-test-auc:0.893061+0.007726\tcv-train-auc:0.983229+0.002218\n",
"[769]\tcv-test-auc:0.893074+0.007739\tcv-train-auc:0.983252+0.002195\n",
"[770]\tcv-test-auc:0.893049+0.007690\tcv-train-auc:0.983274+0.002196\n",
"[771]\tcv-test-auc:0.893087+0.007702\tcv-train-auc:0.983334+0.002185\n",
"[772]\tcv-test-auc:0.893046+0.007691\tcv-train-auc:0.983344+0.002185\n",
"[773]\tcv-test-auc:0.893004+0.007724\tcv-train-auc:0.983375+0.002197\n",
"[774]\tcv-test-auc:0.892961+0.007705\tcv-train-auc:0.983387+0.002194\n",
"[775]\tcv-test-auc:0.893093+0.007896\tcv-train-auc:0.983443+0.002191\n",
"[776]\tcv-test-auc:0.893107+0.007905\tcv-train-auc:0.983467+0.002200\n",
"[777]\tcv-test-auc:0.893543+0.008558\tcv-train-auc:0.983505+0.002219\n",
"[778]\tcv-test-auc:0.893556+0.008515\tcv-train-auc:0.983531+0.002185\n",
"[779]\tcv-test-auc:0.893650+0.008451\tcv-train-auc:0.983580+0.002182\n",
"[780]\tcv-test-auc:0.893641+0.008464\tcv-train-auc:0.983605+0.002182\n",
"[781]\tcv-test-auc:0.893634+0.008498\tcv-train-auc:0.983629+0.002201\n",
"[782]\tcv-test-auc:0.893625+0.008483\tcv-train-auc:0.983672+0.002192\n",
"[783]\tcv-test-auc:0.893632+0.008493\tcv-train-auc:0.983719+0.002210\n",
"[784]\tcv-test-auc:0.893583+0.008483\tcv-train-auc:0.983744+0.002219\n",
"[785]\tcv-test-auc:0.893575+0.008463\tcv-train-auc:0.983765+0.002194\n",
"[786]\tcv-test-auc:0.893572+0.008455\tcv-train-auc:0.983812+0.002208\n",
"[787]\tcv-test-auc:0.893576+0.008451\tcv-train-auc:0.983845+0.002223\n",
"[788]\tcv-test-auc:0.893576+0.008487\tcv-train-auc:0.983859+0.002227\n",
"[789]\tcv-test-auc:0.893576+0.008474\tcv-train-auc:0.983893+0.002233\n",
"[790]\tcv-test-auc:0.893578+0.008480\tcv-train-auc:0.983930+0.002249\n",
"[791]\tcv-test-auc:0.893571+0.008487\tcv-train-auc:0.983977+0.002276\n",
"[792]\tcv-test-auc:0.893568+0.008469\tcv-train-auc:0.984000+0.002268\n",
"[793]\tcv-test-auc:0.893595+0.008460\tcv-train-auc:0.984072+0.002244\n",
"[794]\tcv-test-auc:0.893577+0.008453\tcv-train-auc:0.984104+0.002247\n",
"[795]\tcv-test-auc:0.893548+0.008459\tcv-train-auc:0.984124+0.002254\n",
"[796]\tcv-test-auc:0.893577+0.008472\tcv-train-auc:0.984183+0.002282\n",
"[797]\tcv-test-auc:0.893533+0.008451\tcv-train-auc:0.984196+0.002278\n",
"[798]\tcv-test-auc:0.893517+0.008452\tcv-train-auc:0.984210+0.002276\n",
"[799]\tcv-test-auc:0.893528+0.008431\tcv-train-auc:0.984234+0.002295\n",
"[800]\tcv-test-auc:0.893499+0.008449\tcv-train-auc:0.984270+0.002303\n",
"[801]\tcv-test-auc:0.893510+0.008485\tcv-train-auc:0.984325+0.002276\n",
"[802]\tcv-test-auc:0.893523+0.008490\tcv-train-auc:0.984346+0.002280\n",
"[803]\tcv-test-auc:0.893512+0.008482\tcv-train-auc:0.984412+0.002310\n",
"[804]\tcv-test-auc:0.893524+0.008480\tcv-train-auc:0.984436+0.002315\n",
"[805]\tcv-test-auc:0.893534+0.008438\tcv-train-auc:0.984453+0.002317\n",
"[806]\tcv-test-auc:0.893497+0.008497\tcv-train-auc:0.984478+0.002322\n",
"[807]\tcv-test-auc:0.893532+0.008540\tcv-train-auc:0.984499+0.002329\n",
"[808]\tcv-test-auc:0.893526+0.008607\tcv-train-auc:0.984526+0.002339\n",
"[809]\tcv-test-auc:0.893519+0.008634\tcv-train-auc:0.984576+0.002301\n",
"[810]\tcv-test-auc:0.893524+0.008650\tcv-train-auc:0.984593+0.002286\n",
"[811]\tcv-test-auc:0.893558+0.008694\tcv-train-auc:0.984648+0.002306\n",
"[812]\tcv-test-auc:0.893587+0.008682\tcv-train-auc:0.984676+0.002315\n",
"[813]\tcv-test-auc:0.893583+0.008710\tcv-train-auc:0.984716+0.002327\n",
"[814]\tcv-test-auc:0.893564+0.008734\tcv-train-auc:0.984749+0.002323\n",
"[815]\tcv-test-auc:0.893577+0.008798\tcv-train-auc:0.984797+0.002345\n",
"[816]\tcv-test-auc:0.893595+0.008766\tcv-train-auc:0.984821+0.002353\n",
"[817]\tcv-test-auc:0.893577+0.008769\tcv-train-auc:0.984835+0.002354\n",
"[818]\tcv-test-auc:0.893599+0.008789\tcv-train-auc:0.984851+0.002353\n",
"[819]\tcv-test-auc:0.893609+0.008760\tcv-train-auc:0.984894+0.002364\n",
"[820]\tcv-test-auc:0.893673+0.008916\tcv-train-auc:0.984942+0.002362\n",
"[821]\tcv-test-auc:0.893696+0.008916\tcv-train-auc:0.984968+0.002365\n",
"[822]\tcv-test-auc:0.893693+0.008949\tcv-train-auc:0.984986+0.002369\n",
"[823]\tcv-test-auc:0.893729+0.008937\tcv-train-auc:0.985018+0.002358\n",
"[824]\tcv-test-auc:0.893677+0.008934\tcv-train-auc:0.985037+0.002362\n",
"[825]\tcv-test-auc:0.893662+0.008957\tcv-train-auc:0.985055+0.002362\n",
"[826]\tcv-test-auc:0.893656+0.008992\tcv-train-auc:0.985100+0.002343\n",
"[827]\tcv-test-auc:0.893654+0.008986\tcv-train-auc:0.985136+0.002332\n",
"[828]\tcv-test-auc:0.893675+0.009031\tcv-train-auc:0.985173+0.002350\n",
"[829]\tcv-test-auc:0.893679+0.009062\tcv-train-auc:0.985220+0.002356\n",
"[830]\tcv-test-auc:0.893657+0.009050\tcv-train-auc:0.985239+0.002358\n",
"[831]\tcv-test-auc:0.893654+0.009040\tcv-train-auc:0.985276+0.002346\n",
"[832]\tcv-test-auc:0.893648+0.009028\tcv-train-auc:0.985288+0.002350\n",
"[833]\tcv-test-auc:0.893698+0.009015\tcv-train-auc:0.985324+0.002350\n",
"[834]\tcv-test-auc:0.893696+0.009020\tcv-train-auc:0.985350+0.002335\n",
"[835]\tcv-test-auc:0.893666+0.009007\tcv-train-auc:0.985385+0.002332\n",
"[836]\tcv-test-auc:0.893695+0.009021\tcv-train-auc:0.985419+0.002338\n",
"[837]\tcv-test-auc:0.893681+0.009017\tcv-train-auc:0.985434+0.002350\n",
"[838]\tcv-test-auc:0.893648+0.008987\tcv-train-auc:0.985471+0.002365\n",
"[839]\tcv-test-auc:0.893606+0.008997\tcv-train-auc:0.985495+0.002374\n",
"[840]\tcv-test-auc:0.893630+0.009026\tcv-train-auc:0.985526+0.002384\n",
"[841]\tcv-test-auc:0.893635+0.009035\tcv-train-auc:0.985538+0.002388\n",
"[842]\tcv-test-auc:0.893649+0.009064\tcv-train-auc:0.985569+0.002379\n",
"[843]\tcv-test-auc:0.893634+0.009064\tcv-train-auc:0.985611+0.002377\n",
"[844]\tcv-test-auc:0.893670+0.009119\tcv-train-auc:0.985649+0.002386\n",
"[845]\tcv-test-auc:0.893668+0.009068\tcv-train-auc:0.985694+0.002386\n",
"[846]\tcv-test-auc:0.893687+0.009066\tcv-train-auc:0.985725+0.002407\n",
"[847]\tcv-test-auc:0.893651+0.009067\tcv-train-auc:0.985742+0.002408\n",
"[848]\tcv-test-auc:0.893624+0.009077\tcv-train-auc:0.985778+0.002402\n",
"[849]\tcv-test-auc:0.893623+0.009059\tcv-train-auc:0.985824+0.002408\n",
"[850]\tcv-test-auc:0.893645+0.009070\tcv-train-auc:0.985883+0.002422\n",
"[851]\tcv-test-auc:0.893622+0.009080\tcv-train-auc:0.985928+0.002419\n",
"[852]\tcv-test-auc:0.893636+0.009096\tcv-train-auc:0.985943+0.002420\n",
"[853]\tcv-test-auc:0.893650+0.009087\tcv-train-auc:0.986004+0.002457\n",
"[854]\tcv-test-auc:0.893690+0.009036\tcv-train-auc:0.986042+0.002444\n",
"[855]\tcv-test-auc:0.893689+0.008993\tcv-train-auc:0.986064+0.002423\n",
"[856]\tcv-test-auc:0.893694+0.008991\tcv-train-auc:0.986088+0.002431\n",
"[857]\tcv-test-auc:0.893715+0.008978\tcv-train-auc:0.986142+0.002398\n",
"[858]\tcv-test-auc:0.893729+0.008989\tcv-train-auc:0.986165+0.002406\n",
"[859]\tcv-test-auc:0.893753+0.008976\tcv-train-auc:0.986196+0.002401\n",
"[860]\tcv-test-auc:0.893731+0.008963\tcv-train-auc:0.986215+0.002402\n",
"[861]\tcv-test-auc:0.893729+0.008969\tcv-train-auc:0.986244+0.002394\n",
"[862]\tcv-test-auc:0.893742+0.008950\tcv-train-auc:0.986292+0.002384\n",
"[863]\tcv-test-auc:0.893713+0.008937\tcv-train-auc:0.986315+0.002390\n",
"[864]\tcv-test-auc:0.893764+0.008951\tcv-train-auc:0.986360+0.002362\n",
"[865]\tcv-test-auc:0.893768+0.008949\tcv-train-auc:0.986389+0.002369\n",
"[866]\tcv-test-auc:0.893783+0.008939\tcv-train-auc:0.986436+0.002385\n",
"[867]\tcv-test-auc:0.893807+0.008928\tcv-train-auc:0.986472+0.002380\n",
"[868]\tcv-test-auc:0.893832+0.008915\tcv-train-auc:0.986509+0.002376\n",
"[869]\tcv-test-auc:0.893817+0.008932\tcv-train-auc:0.986531+0.002388\n",
"[870]\tcv-test-auc:0.893872+0.008958\tcv-train-auc:0.986558+0.002403\n",
"[871]\tcv-test-auc:0.893859+0.008933\tcv-train-auc:0.986571+0.002414\n",
"[872]\tcv-test-auc:0.893870+0.008923\tcv-train-auc:0.986597+0.002389\n",
"[873]\tcv-test-auc:0.893886+0.008930\tcv-train-auc:0.986627+0.002407\n",
"[874]\tcv-test-auc:0.893902+0.008912\tcv-train-auc:0.986660+0.002408\n",
"[875]\tcv-test-auc:0.893889+0.008890\tcv-train-auc:0.986709+0.002405\n",
"[876]\tcv-test-auc:0.893881+0.008869\tcv-train-auc:0.986789+0.002374\n",
"[877]\tcv-test-auc:0.893910+0.008888\tcv-train-auc:0.986812+0.002387\n",
"[878]\tcv-test-auc:0.893912+0.008890\tcv-train-auc:0.986841+0.002398\n",
"[879]\tcv-test-auc:0.893935+0.008891\tcv-train-auc:0.986873+0.002408\n",
"[880]\tcv-test-auc:0.893903+0.008922\tcv-train-auc:0.986926+0.002353\n",
"[881]\tcv-test-auc:0.893903+0.008908\tcv-train-auc:0.986976+0.002359\n",
"[882]\tcv-test-auc:0.893927+0.008900\tcv-train-auc:0.986993+0.002349\n",
"[883]\tcv-test-auc:0.893942+0.008921\tcv-train-auc:0.987065+0.002336\n",
"[884]\tcv-test-auc:0.893963+0.008937\tcv-train-auc:0.987092+0.002347\n",
"[885]\tcv-test-auc:0.893937+0.008928\tcv-train-auc:0.987117+0.002359\n",
"[886]\tcv-test-auc:0.893966+0.008916\tcv-train-auc:0.987155+0.002325\n",
"[887]\tcv-test-auc:0.893936+0.008916\tcv-train-auc:0.987183+0.002329\n",
"[888]\tcv-test-auc:0.893973+0.008903\tcv-train-auc:0.987229+0.002324\n",
"[889]\tcv-test-auc:0.893974+0.008869\tcv-train-auc:0.987286+0.002300\n",
"[890]\tcv-test-auc:0.893964+0.008867\tcv-train-auc:0.987313+0.002304\n",
"[891]\tcv-test-auc:0.893928+0.008914\tcv-train-auc:0.987375+0.002266\n",
"[892]\tcv-test-auc:0.893926+0.008925\tcv-train-auc:0.987427+0.002250\n",
"[893]\tcv-test-auc:0.893985+0.008937\tcv-train-auc:0.987487+0.002253\n",
"[894]\tcv-test-auc:0.894009+0.008910\tcv-train-auc:0.987532+0.002277\n",
"[895]\tcv-test-auc:0.893985+0.008895\tcv-train-auc:0.987553+0.002288\n",
"[896]\tcv-test-auc:0.894024+0.008941\tcv-train-auc:0.987592+0.002267\n",
"[897]\tcv-test-auc:0.894067+0.008900\tcv-train-auc:0.987635+0.002261\n",
"[898]\tcv-test-auc:0.894044+0.008894\tcv-train-auc:0.987658+0.002266\n",
"[899]\tcv-test-auc:0.894056+0.008910\tcv-train-auc:0.987669+0.002260\n",
"[900]\tcv-test-auc:0.894031+0.008907\tcv-train-auc:0.987687+0.002265\n",
"[901]\tcv-test-auc:0.893985+0.008899\tcv-train-auc:0.987725+0.002228\n",
"[902]\tcv-test-auc:0.894031+0.008885\tcv-train-auc:0.987748+0.002224\n",
"[903]\tcv-test-auc:0.894048+0.008868\tcv-train-auc:0.987757+0.002230\n",
"[904]\tcv-test-auc:0.894076+0.008910\tcv-train-auc:0.987787+0.002241\n",
"[905]\tcv-test-auc:0.894083+0.008905\tcv-train-auc:0.987830+0.002210\n",
"[906]\tcv-test-auc:0.894135+0.008843\tcv-train-auc:0.987866+0.002204\n",
"[907]\tcv-test-auc:0.894124+0.008898\tcv-train-auc:0.987934+0.002185\n",
"[908]\tcv-test-auc:0.894130+0.008900\tcv-train-auc:0.987974+0.002189\n",
"[909]\tcv-test-auc:0.894109+0.008902\tcv-train-auc:0.988003+0.002193\n",
"[910]\tcv-test-auc:0.894159+0.008908\tcv-train-auc:0.988034+0.002206\n",
"[911]\tcv-test-auc:0.894158+0.008874\tcv-train-auc:0.988107+0.002183\n",
"[912]\tcv-test-auc:0.894204+0.008900\tcv-train-auc:0.988146+0.002155\n",
"[913]\tcv-test-auc:0.894181+0.008898\tcv-train-auc:0.988165+0.002163\n",
"[914]\tcv-test-auc:0.894182+0.008884\tcv-train-auc:0.988209+0.002155\n",
"[915]\tcv-test-auc:0.894171+0.008912\tcv-train-auc:0.988226+0.002151\n",
"[916]\tcv-test-auc:0.894189+0.008934\tcv-train-auc:0.988268+0.002129\n",
"[917]\tcv-test-auc:0.894173+0.008907\tcv-train-auc:0.988274+0.002131\n",
"[918]\tcv-test-auc:0.894181+0.008919\tcv-train-auc:0.988289+0.002138\n",
"[919]\tcv-test-auc:0.894208+0.008920\tcv-train-auc:0.988352+0.002127\n",
"[920]\tcv-test-auc:0.894206+0.008922\tcv-train-auc:0.988384+0.002121\n",
"[921]\tcv-test-auc:0.894217+0.008922\tcv-train-auc:0.988446+0.002083\n",
"[922]\tcv-test-auc:0.894209+0.008923\tcv-train-auc:0.988464+0.002092\n",
"[923]\tcv-test-auc:0.894250+0.008890\tcv-train-auc:0.988511+0.002096\n",
"[924]\tcv-test-auc:0.894278+0.008926\tcv-train-auc:0.988527+0.002102\n",
"[925]\tcv-test-auc:0.894267+0.008929\tcv-train-auc:0.988561+0.002111\n",
"[926]\tcv-test-auc:0.894242+0.008936\tcv-train-auc:0.988573+0.002120\n",
"[927]\tcv-test-auc:0.894269+0.008934\tcv-train-auc:0.988593+0.002128\n",
"[928]\tcv-test-auc:0.894267+0.008950\tcv-train-auc:0.988621+0.002143\n",
"[929]\tcv-test-auc:0.894284+0.008943\tcv-train-auc:0.988647+0.002155\n",
"[930]\tcv-test-auc:0.894288+0.008961\tcv-train-auc:0.988668+0.002154\n",
"[931]\tcv-test-auc:0.894261+0.008978\tcv-train-auc:0.988724+0.002111\n",
"[932]\tcv-test-auc:0.894254+0.008981\tcv-train-auc:0.988738+0.002110\n",
"[933]\tcv-test-auc:0.894241+0.008979\tcv-train-auc:0.988747+0.002117\n",
"[934]\tcv-test-auc:0.894232+0.008972\tcv-train-auc:0.988797+0.002106\n",
"[935]\tcv-test-auc:0.894229+0.009013\tcv-train-auc:0.988838+0.002073\n",
"[936]\tcv-test-auc:0.894237+0.009013\tcv-train-auc:0.988861+0.002081\n",
"[937]\tcv-test-auc:0.894193+0.008986\tcv-train-auc:0.988880+0.002090\n",
"[938]\tcv-test-auc:0.894178+0.008990\tcv-train-auc:0.988916+0.002107\n",
"[939]\tcv-test-auc:0.894166+0.008990\tcv-train-auc:0.988941+0.002109\n",
"[940]\tcv-test-auc:0.894209+0.009009\tcv-train-auc:0.988972+0.002096\n",
"[941]\tcv-test-auc:0.894205+0.009022\tcv-train-auc:0.988995+0.002092\n",
"[942]\tcv-test-auc:0.894237+0.009020\tcv-train-auc:0.989015+0.002100\n",
"[943]\tcv-test-auc:0.894228+0.009033\tcv-train-auc:0.989047+0.002115\n",
"[944]\tcv-test-auc:0.894206+0.009039\tcv-train-auc:0.989112+0.002096\n",
"[945]\tcv-test-auc:0.894201+0.009030\tcv-train-auc:0.989149+0.002052\n",
"[946]\tcv-test-auc:0.894238+0.009029\tcv-train-auc:0.989177+0.002043\n",
"[947]\tcv-test-auc:0.894244+0.009001\tcv-train-auc:0.989204+0.002058\n",
"[948]\tcv-test-auc:0.894253+0.009016\tcv-train-auc:0.989224+0.002058\n",
"[949]\tcv-test-auc:0.894234+0.008990\tcv-train-auc:0.989277+0.002017\n",
"[950]\tcv-test-auc:0.894238+0.009015\tcv-train-auc:0.989310+0.002030\n",
"[951]\tcv-test-auc:0.894294+0.008989\tcv-train-auc:0.989353+0.002021\n",
"[952]\tcv-test-auc:0.894281+0.009002\tcv-train-auc:0.989366+0.002018\n",
"[953]\tcv-test-auc:0.894275+0.008996\tcv-train-auc:0.989378+0.002017\n",
"[954]\tcv-test-auc:0.894250+0.008986\tcv-train-auc:0.989419+0.002011\n",
"[955]\tcv-test-auc:0.894265+0.008980\tcv-train-auc:0.989445+0.002026\n",
"[956]\tcv-test-auc:0.894269+0.008975\tcv-train-auc:0.989470+0.002019\n",
"[957]\tcv-test-auc:0.894290+0.008992\tcv-train-auc:0.989505+0.002021\n",
"[958]\tcv-test-auc:0.894251+0.009010\tcv-train-auc:0.989540+0.002001\n",
"[959]\tcv-test-auc:0.894243+0.008985\tcv-train-auc:0.989561+0.002014\n",
"[960]\tcv-test-auc:0.894239+0.008984\tcv-train-auc:0.989575+0.002015\n",
"[961]\tcv-test-auc:0.894240+0.008997\tcv-train-auc:0.989594+0.002011\n",
"[962]\tcv-test-auc:0.894214+0.008975\tcv-train-auc:0.989640+0.001998\n",
"[963]\tcv-test-auc:0.894229+0.008956\tcv-train-auc:0.989666+0.002014\n",
"[964]\tcv-test-auc:0.894237+0.008962\tcv-train-auc:0.989680+0.002013\n",
"[965]\tcv-test-auc:0.894228+0.008919\tcv-train-auc:0.989689+0.002017\n",
"[966]\tcv-test-auc:0.894236+0.008953\tcv-train-auc:0.989709+0.002032\n",
"[967]\tcv-test-auc:0.894205+0.008942\tcv-train-auc:0.989741+0.002041\n",
"[968]\tcv-test-auc:0.894211+0.008952\tcv-train-auc:0.989776+0.002004\n",
"[969]\tcv-test-auc:0.894212+0.008943\tcv-train-auc:0.989796+0.002010\n",
"[970]\tcv-test-auc:0.894253+0.008983\tcv-train-auc:0.989814+0.002018\n",
"[971]\tcv-test-auc:0.894266+0.008940\tcv-train-auc:0.989835+0.002025\n",
"[972]\tcv-test-auc:0.894282+0.008963\tcv-train-auc:0.989845+0.002033\n",
"[973]\tcv-test-auc:0.894271+0.008962\tcv-train-auc:0.989867+0.002047\n",
"[974]\tcv-test-auc:0.894240+0.008929\tcv-train-auc:0.989894+0.002057\n",
"[975]\tcv-test-auc:0.894247+0.008917\tcv-train-auc:0.989942+0.002006\n",
"[976]\tcv-test-auc:0.894304+0.008930\tcv-train-auc:0.989976+0.001982\n",
"[977]\tcv-test-auc:0.894330+0.008930\tcv-train-auc:0.990019+0.001943\n",
"[978]\tcv-test-auc:0.894298+0.008879\tcv-train-auc:0.990040+0.001937\n",
"[979]\tcv-test-auc:0.894286+0.008909\tcv-train-auc:0.990056+0.001934\n",
"[980]\tcv-test-auc:0.894274+0.008918\tcv-train-auc:0.990103+0.001909\n",
"[981]\tcv-test-auc:0.894295+0.008918\tcv-train-auc:0.990133+0.001901\n",
"[982]\tcv-test-auc:0.894303+0.008987\tcv-train-auc:0.990169+0.001896\n",
"[983]\tcv-test-auc:0.894336+0.008975\tcv-train-auc:0.990177+0.001900\n",
"[984]\tcv-test-auc:0.894318+0.008960\tcv-train-auc:0.990186+0.001897\n",
"[985]\tcv-test-auc:0.894332+0.008944\tcv-train-auc:0.990208+0.001902\n",
"[986]\tcv-test-auc:0.894335+0.008928\tcv-train-auc:0.990244+0.001883\n",
"[987]\tcv-test-auc:0.894335+0.008935\tcv-train-auc:0.990283+0.001875\n",
"[988]\tcv-test-auc:0.894311+0.008930\tcv-train-auc:0.990317+0.001877\n",
"[989]\tcv-test-auc:0.894314+0.008964\tcv-train-auc:0.990348+0.001879\n",
"[990]\tcv-test-auc:0.894297+0.008975\tcv-train-auc:0.990388+0.001862\n",
"[991]\tcv-test-auc:0.894253+0.008972\tcv-train-auc:0.990439+0.001875\n",
"[992]\tcv-test-auc:0.894265+0.008950\tcv-train-auc:0.990456+0.001879\n",
"[993]\tcv-test-auc:0.894249+0.008954\tcv-train-auc:0.990470+0.001884\n",
"[994]\tcv-test-auc:0.894264+0.008958\tcv-train-auc:0.990499+0.001891\n",
"[995]\tcv-test-auc:0.894256+0.008932\tcv-train-auc:0.990512+0.001899\n",
"[996]\tcv-test-auc:0.894270+0.008972\tcv-train-auc:0.990542+0.001889\n",
"[997]\tcv-test-auc:0.894263+0.008945\tcv-train-auc:0.990568+0.001867\n",
"[998]\tcv-test-auc:0.894273+0.008982\tcv-train-auc:0.990595+0.001867\n",
"[999]\tcv-test-auc:0.894320+0.008974\tcv-train-auc:0.990612+0.001860\n",
"[1000]\tcv-test-auc:0.894275+0.008968\tcv-train-auc:0.990623+0.001856\n",
"[1001]\tcv-test-auc:0.894269+0.008980\tcv-train-auc:0.990644+0.001862\n",
"[1002]\tcv-test-auc:0.894312+0.008992\tcv-train-auc:0.990665+0.001873\n",
"[1003]\tcv-test-auc:0.894301+0.009001\tcv-train-auc:0.990684+0.001881\n",
"[1004]\tcv-test-auc:0.894303+0.008972\tcv-train-auc:0.990706+0.001858\n",
"[1005]\tcv-test-auc:0.894276+0.008972\tcv-train-auc:0.990722+0.001864\n",
"[1006]\tcv-test-auc:0.894263+0.008958\tcv-train-auc:0.990750+0.001862\n",
"[1007]\tcv-test-auc:0.894248+0.008956\tcv-train-auc:0.990776+0.001868\n",
"[1008]\tcv-test-auc:0.894232+0.008950\tcv-train-auc:0.990812+0.001835\n",
"[1009]\tcv-test-auc:0.894254+0.008969\tcv-train-auc:0.990815+0.001836\n",
"[1010]\tcv-test-auc:0.894256+0.008976\tcv-train-auc:0.990826+0.001841\n",
"[1011]\tcv-test-auc:0.894253+0.008975\tcv-train-auc:0.990852+0.001827\n",
"[1012]\tcv-test-auc:0.894228+0.008959\tcv-train-auc:0.990867+0.001836\n",
"[1013]\tcv-test-auc:0.894207+0.008955\tcv-train-auc:0.990883+0.001835\n",
"[1014]\tcv-test-auc:0.894210+0.008921\tcv-train-auc:0.990897+0.001845\n",
"[1015]\tcv-test-auc:0.894224+0.008863\tcv-train-auc:0.990932+0.001826\n",
"[1016]\tcv-test-auc:0.894222+0.008848\tcv-train-auc:0.990958+0.001823\n",
"[1017]\tcv-test-auc:0.894197+0.008883\tcv-train-auc:0.990972+0.001822\n",
"[1018]\tcv-test-auc:0.894209+0.008852\tcv-train-auc:0.990984+0.001830\n",
"[1019]\tcv-test-auc:0.894197+0.008860\tcv-train-auc:0.991000+0.001826\n",
"[1020]\tcv-test-auc:0.894191+0.008863\tcv-train-auc:0.991023+0.001844\n",
"[1021]\tcv-test-auc:0.894203+0.008786\tcv-train-auc:0.991066+0.001814\n",
"[1022]\tcv-test-auc:0.894190+0.008812\tcv-train-auc:0.991088+0.001822\n",
"[1023]\tcv-test-auc:0.894183+0.008819\tcv-train-auc:0.991105+0.001827\n",
"[1024]\tcv-test-auc:0.894170+0.008804\tcv-train-auc:0.991115+0.001830\n",
"[1025]\tcv-test-auc:0.894158+0.008788\tcv-train-auc:0.991134+0.001819\n",
"[1026]\tcv-test-auc:0.894171+0.008772\tcv-train-auc:0.991167+0.001779\n",
"[1027]\tcv-test-auc:0.894204+0.008751\tcv-train-auc:0.991207+0.001763\n",
"[1028]\tcv-test-auc:0.894196+0.008756\tcv-train-auc:0.991229+0.001769\n",
"[1029]\tcv-test-auc:0.894174+0.008778\tcv-train-auc:0.991264+0.001773\n",
"[1030]\tcv-test-auc:0.894183+0.008784\tcv-train-auc:0.991279+0.001753\n",
"[1031]\tcv-test-auc:0.894157+0.008787\tcv-train-auc:0.991295+0.001752\n",
"[1032]\tcv-test-auc:0.894187+0.008769\tcv-train-auc:0.991336+0.001740\n",
"[1033]\tcv-test-auc:0.894170+0.008759\tcv-train-auc:0.991345+0.001743\n",
"[1034]\tcv-test-auc:0.894180+0.008753\tcv-train-auc:0.991350+0.001742\n",
"[1035]\tcv-test-auc:0.894189+0.008733\tcv-train-auc:0.991387+0.001741\n",
"[1036]\tcv-test-auc:0.894160+0.008716\tcv-train-auc:0.991394+0.001744\n",
"[1037]\tcv-test-auc:0.894127+0.008679\tcv-train-auc:0.991409+0.001740\n",
"[1038]\tcv-test-auc:0.894159+0.008663\tcv-train-auc:0.991429+0.001749\n",
"[1039]\tcv-test-auc:0.894186+0.008672\tcv-train-auc:0.991435+0.001750\n",
"[1040]\tcv-test-auc:0.894173+0.008660\tcv-train-auc:0.991456+0.001756\n",
"[1041]\tcv-test-auc:0.894168+0.008677\tcv-train-auc:0.991488+0.001725\n",
"[1042]\tcv-test-auc:0.894159+0.008694\tcv-train-auc:0.991501+0.001728\n",
"[1043]\tcv-test-auc:0.894154+0.008675\tcv-train-auc:0.991507+0.001729\n",
"[1044]\tcv-test-auc:0.894145+0.008664\tcv-train-auc:0.991520+0.001724\n",
"[1045]\tcv-test-auc:0.894206+0.008656\tcv-train-auc:0.991529+0.001730\n",
"[1046]\tcv-test-auc:0.894215+0.008669\tcv-train-auc:0.991542+0.001733\n",
"[1047]\tcv-test-auc:0.894214+0.008642\tcv-train-auc:0.991569+0.001709\n",
"[1048]\tcv-test-auc:0.894205+0.008680\tcv-train-auc:0.991580+0.001714\n",
"[1049]\tcv-test-auc:0.894202+0.008705\tcv-train-auc:0.991596+0.001728\n",
"[1050]\tcv-test-auc:0.894188+0.008700\tcv-train-auc:0.991615+0.001739\n",
"[1051]\tcv-test-auc:0.894164+0.008723\tcv-train-auc:0.991631+0.001743\n",
"[1052]\tcv-test-auc:0.894178+0.008735\tcv-train-auc:0.991657+0.001713\n",
"[1053]\tcv-test-auc:0.894149+0.008738\tcv-train-auc:0.991692+0.001683\n",
"[1054]\tcv-test-auc:0.894162+0.008698\tcv-train-auc:0.991713+0.001662\n",
"[1055]\tcv-test-auc:0.894182+0.008668\tcv-train-auc:0.991727+0.001663\n",
"[1056]\tcv-test-auc:0.894208+0.008696\tcv-train-auc:0.991744+0.001660\n",
"[1057]\tcv-test-auc:0.894222+0.008701\tcv-train-auc:0.991769+0.001655\n",
"[1058]\tcv-test-auc:0.894198+0.008689\tcv-train-auc:0.991784+0.001658\n",
"[1059]\tcv-test-auc:0.894211+0.008683\tcv-train-auc:0.991785+0.001658\n",
"[1060]\tcv-test-auc:0.894208+0.008681\tcv-train-auc:0.991795+0.001649\n",
"[1061]\tcv-test-auc:0.894203+0.008710\tcv-train-auc:0.991822+0.001650\n",
"[1062]\tcv-test-auc:0.894215+0.008713\tcv-train-auc:0.991831+0.001645\n",
"[1063]\tcv-test-auc:0.894253+0.008708\tcv-train-auc:0.991856+0.001626\n",
"[1064]\tcv-test-auc:0.894226+0.008722\tcv-train-auc:0.991871+0.001630\n",
"[1065]\tcv-test-auc:0.894229+0.008683\tcv-train-auc:0.991894+0.001641\n",
"[1066]\tcv-test-auc:0.894229+0.008735\tcv-train-auc:0.991919+0.001614\n",
"[1067]\tcv-test-auc:0.894239+0.008695\tcv-train-auc:0.991954+0.001593\n",
"[1068]\tcv-test-auc:0.894261+0.008704\tcv-train-auc:0.991957+0.001596\n",
"[1069]\tcv-test-auc:0.894231+0.008680\tcv-train-auc:0.991970+0.001595\n",
"[1070]\tcv-test-auc:0.894208+0.008723\tcv-train-auc:0.991996+0.001565\n",
"[1071]\tcv-test-auc:0.894254+0.008728\tcv-train-auc:0.992016+0.001557\n",
"[1072]\tcv-test-auc:0.894253+0.008719\tcv-train-auc:0.992031+0.001562\n",
"[1073]\tcv-test-auc:0.894255+0.008704\tcv-train-auc:0.992053+0.001548\n",
"[1074]\tcv-test-auc:0.894250+0.008712\tcv-train-auc:0.992061+0.001544\n",
"[1075]\tcv-test-auc:0.894222+0.008748\tcv-train-auc:0.992088+0.001514\n",
"[1076]\tcv-test-auc:0.894201+0.008751\tcv-train-auc:0.992107+0.001521\n",
"[1077]\tcv-test-auc:0.894207+0.008764\tcv-train-auc:0.992125+0.001532\n",
"[1078]\tcv-test-auc:0.894213+0.008738\tcv-train-auc:0.992128+0.001532\n",
"[1079]\tcv-test-auc:0.894224+0.008755\tcv-train-auc:0.992152+0.001511\n",
"[1080]\tcv-test-auc:0.894253+0.008789\tcv-train-auc:0.992171+0.001518\n",
"[1081]\tcv-test-auc:0.894262+0.008801\tcv-train-auc:0.992185+0.001508\n",
"[1082]\tcv-test-auc:0.894295+0.008812\tcv-train-auc:0.992218+0.001490\n",
"[1083]\tcv-test-auc:0.894321+0.008824\tcv-train-auc:0.992228+0.001495\n",
"[1084]\tcv-test-auc:0.894305+0.008824\tcv-train-auc:0.992251+0.001501\n",
"[1085]\tcv-test-auc:0.894327+0.008846\tcv-train-auc:0.992254+0.001501\n",
"[1086]\tcv-test-auc:0.894285+0.008878\tcv-train-auc:0.992274+0.001502\n",
"[1087]\tcv-test-auc:0.894272+0.008895\tcv-train-auc:0.992304+0.001499\n",
"[1088]\tcv-test-auc:0.894300+0.008927\tcv-train-auc:0.992321+0.001507\n",
"[1089]\tcv-test-auc:0.894324+0.008913\tcv-train-auc:0.992351+0.001489\n",
"[1090]\tcv-test-auc:0.894352+0.008903\tcv-train-auc:0.992378+0.001480\n",
"[1091]\tcv-test-auc:0.894347+0.008940\tcv-train-auc:0.992403+0.001449\n",
"[1092]\tcv-test-auc:0.894332+0.008996\tcv-train-auc:0.992429+0.001445\n",
"[1093]\tcv-test-auc:0.894359+0.008991\tcv-train-auc:0.992436+0.001450\n",
"[1094]\tcv-test-auc:0.894353+0.009021\tcv-train-auc:0.992466+0.001436\n",
"[1095]\tcv-test-auc:0.894361+0.009020\tcv-train-auc:0.992486+0.001442\n",
"[1096]\tcv-test-auc:0.894286+0.009029\tcv-train-auc:0.992498+0.001442\n",
"[1097]\tcv-test-auc:0.894296+0.009066\tcv-train-auc:0.992517+0.001418\n",
"[1098]\tcv-test-auc:0.894273+0.009060\tcv-train-auc:0.992534+0.001413\n",
"[1099]\tcv-test-auc:0.894261+0.009060\tcv-train-auc:0.992568+0.001393\n",
"[1100]\tcv-test-auc:0.894258+0.009086\tcv-train-auc:0.992592+0.001397\n",
"[1101]\tcv-test-auc:0.894288+0.009078\tcv-train-auc:0.992638+0.001364\n",
"[1102]\tcv-test-auc:0.894298+0.009119\tcv-train-auc:0.992663+0.001348\n",
"[1103]\tcv-test-auc:0.894309+0.009129\tcv-train-auc:0.992683+0.001349\n",
"[1104]\tcv-test-auc:0.894331+0.009127\tcv-train-auc:0.992702+0.001336\n",
"[1105]\tcv-test-auc:0.894378+0.009108\tcv-train-auc:0.992760+0.001310\n",
"[1106]\tcv-test-auc:0.894330+0.009138\tcv-train-auc:0.992777+0.001313\n",
"[1107]\tcv-test-auc:0.894363+0.009144\tcv-train-auc:0.992806+0.001297\n",
"[1108]\tcv-test-auc:0.894363+0.009153\tcv-train-auc:0.992811+0.001296\n",
"[1109]\tcv-test-auc:0.894349+0.009155\tcv-train-auc:0.992827+0.001293\n",
"[1110]\tcv-test-auc:0.894365+0.009130\tcv-train-auc:0.992859+0.001253\n",
"[1111]\tcv-test-auc:0.894348+0.009159\tcv-train-auc:0.992882+0.001260\n",
"[1112]\tcv-test-auc:0.894352+0.009159\tcv-train-auc:0.992887+0.001259\n",
"[1113]\tcv-test-auc:0.894374+0.009169\tcv-train-auc:0.992897+0.001263\n",
"[1114]\tcv-test-auc:0.894412+0.009165\tcv-train-auc:0.992923+0.001245\n",
"[1115]\tcv-test-auc:0.894385+0.009134\tcv-train-auc:0.992935+0.001244\n",
"[1116]\tcv-test-auc:0.894415+0.009138\tcv-train-auc:0.992990+0.001210\n",
"[1117]\tcv-test-auc:0.894417+0.009171\tcv-train-auc:0.993005+0.001219\n",
"[1118]\tcv-test-auc:0.894448+0.009160\tcv-train-auc:0.993027+0.001204\n",
"[1119]\tcv-test-auc:0.894406+0.009130\tcv-train-auc:0.993042+0.001206\n",
"[1120]\tcv-test-auc:0.894423+0.009123\tcv-train-auc:0.993065+0.001170\n",
"[1121]\tcv-test-auc:0.894443+0.009145\tcv-train-auc:0.993076+0.001176\n",
"[1122]\tcv-test-auc:0.894406+0.009125\tcv-train-auc:0.993097+0.001182\n",
"[1123]\tcv-test-auc:0.894420+0.009162\tcv-train-auc:0.993105+0.001189\n",
"[1124]\tcv-test-auc:0.894429+0.009170\tcv-train-auc:0.993109+0.001188\n",
"[1125]\tcv-test-auc:0.894451+0.009177\tcv-train-auc:0.993151+0.001143\n",
"[1126]\tcv-test-auc:0.894429+0.009164\tcv-train-auc:0.993164+0.001143\n",
"[1127]\tcv-test-auc:0.894422+0.009176\tcv-train-auc:0.993185+0.001151\n",
"[1128]\tcv-test-auc:0.894424+0.009178\tcv-train-auc:0.993210+0.001141\n",
"[1129]\tcv-test-auc:0.894441+0.009218\tcv-train-auc:0.993236+0.001119\n",
"[1130]\tcv-test-auc:0.894437+0.009207\tcv-train-auc:0.993244+0.001120\n",
"[1131]\tcv-test-auc:0.894420+0.009193\tcv-train-auc:0.993249+0.001124\n",
"[1132]\tcv-test-auc:0.894424+0.009197\tcv-train-auc:0.993268+0.001127\n",
"[1133]\tcv-test-auc:0.894435+0.009201\tcv-train-auc:0.993295+0.001116\n",
"[1134]\tcv-test-auc:0.894450+0.009232\tcv-train-auc:0.993314+0.001103\n",
"[1135]\tcv-test-auc:0.894479+0.009259\tcv-train-auc:0.993330+0.001106\n",
"[1136]\tcv-test-auc:0.894478+0.009262\tcv-train-auc:0.993351+0.001099\n",
"[1137]\tcv-test-auc:0.894479+0.009199\tcv-train-auc:0.993379+0.001093\n",
"[1138]\tcv-test-auc:0.894483+0.009211\tcv-train-auc:0.993389+0.001093\n",
"[1139]\tcv-test-auc:0.894480+0.009207\tcv-train-auc:0.993414+0.001069\n",
"[1140]\tcv-test-auc:0.894496+0.009216\tcv-train-auc:0.993422+0.001070\n",
"[1141]\tcv-test-auc:0.894489+0.009205\tcv-train-auc:0.993443+0.001072\n",
"[1142]\tcv-test-auc:0.894498+0.009237\tcv-train-auc:0.993461+0.001062\n",
"[1143]\tcv-test-auc:0.894498+0.009209\tcv-train-auc:0.993474+0.001068\n",
"[1144]\tcv-test-auc:0.894475+0.009180\tcv-train-auc:0.993498+0.001057\n",
"[1145]\tcv-test-auc:0.894451+0.009163\tcv-train-auc:0.993520+0.001049\n",
"[1146]\tcv-test-auc:0.894432+0.009187\tcv-train-auc:0.993549+0.001036\n",
"[1147]\tcv-test-auc:0.894445+0.009173\tcv-train-auc:0.993576+0.001035\n",
"[1148]\tcv-test-auc:0.894420+0.009113\tcv-train-auc:0.993590+0.001030\n",
"[1149]\tcv-test-auc:0.894423+0.009140\tcv-train-auc:0.993597+0.001028\n",
"[1150]\tcv-test-auc:0.894419+0.009125\tcv-train-auc:0.993618+0.001007\n",
"[1151]\tcv-test-auc:0.894382+0.009105\tcv-train-auc:0.993625+0.001009\n",
"[1152]\tcv-test-auc:0.894365+0.009113\tcv-train-auc:0.993637+0.001009\n",
"[1153]\tcv-test-auc:0.894360+0.009119\tcv-train-auc:0.993644+0.001008\n",
"[1154]\tcv-test-auc:0.894378+0.009101\tcv-train-auc:0.993665+0.000995\n",
"[1155]\tcv-test-auc:0.894388+0.009095\tcv-train-auc:0.993674+0.001000\n",
"[1156]\tcv-test-auc:0.894402+0.009057\tcv-train-auc:0.993696+0.000987\n",
"[1157]\tcv-test-auc:0.894389+0.009027\tcv-train-auc:0.993701+0.000986\n",
"[1158]\tcv-test-auc:0.894355+0.009006\tcv-train-auc:0.993715+0.000977\n",
"[1159]\tcv-test-auc:0.894380+0.009010\tcv-train-auc:0.993733+0.000988\n",
"[1160]\tcv-test-auc:0.894332+0.009063\tcv-train-auc:0.993756+0.000988\n",
"[1161]\tcv-test-auc:0.894356+0.009056\tcv-train-auc:0.993770+0.000987\n",
"[1162]\tcv-test-auc:0.894400+0.009011\tcv-train-auc:0.993801+0.000945\n",
"[1163]\tcv-test-auc:0.894433+0.009009\tcv-train-auc:0.993809+0.000952\n",
"[1164]\tcv-test-auc:0.894405+0.009015\tcv-train-auc:0.993816+0.000950\n",
"[1165]\tcv-test-auc:0.894393+0.008994\tcv-train-auc:0.993833+0.000954\n",
"[1166]\tcv-test-auc:0.894378+0.008974\tcv-train-auc:0.993843+0.000958\n",
"[1167]\tcv-test-auc:0.894350+0.008947\tcv-train-auc:0.993862+0.000955\n",
"[1168]\tcv-test-auc:0.894361+0.008963\tcv-train-auc:0.993871+0.000948\n",
"[1169]\tcv-test-auc:0.894310+0.008970\tcv-train-auc:0.993881+0.000942\n",
"[1170]\tcv-test-auc:0.894321+0.008950\tcv-train-auc:0.993904+0.000924\n",
"[1171]\tcv-test-auc:0.894306+0.008941\tcv-train-auc:0.993914+0.000927\n",
"[1172]\tcv-test-auc:0.894277+0.008885\tcv-train-auc:0.993930+0.000929\n",
"[1173]\tcv-test-auc:0.894282+0.008879\tcv-train-auc:0.993939+0.000926\n",
"[1174]\tcv-test-auc:0.894290+0.008890\tcv-train-auc:0.993951+0.000926\n",
"[1175]\tcv-test-auc:0.894234+0.008868\tcv-train-auc:0.993969+0.000918\n",
"[1176]\tcv-test-auc:0.894244+0.008873\tcv-train-auc:0.993993+0.000928\n",
"[1177]\tcv-test-auc:0.894250+0.008800\tcv-train-auc:0.994037+0.000891\n",
"[1178]\tcv-test-auc:0.894227+0.008821\tcv-train-auc:0.994048+0.000897\n",
"[1179]\tcv-test-auc:0.894222+0.008852\tcv-train-auc:0.994066+0.000876\n",
"[1180]\tcv-test-auc:0.894196+0.008830\tcv-train-auc:0.994077+0.000881\n",
"[1181]\tcv-test-auc:0.894211+0.008802\tcv-train-auc:0.994109+0.000869\n",
"[1182]\tcv-test-auc:0.894221+0.008872\tcv-train-auc:0.994127+0.000854\n",
"[1183]\tcv-test-auc:0.894224+0.008860\tcv-train-auc:0.994142+0.000850\n",
"[1184]\tcv-test-auc:0.894213+0.008829\tcv-train-auc:0.994150+0.000851\n",
"[1185]\tcv-test-auc:0.894188+0.008821\tcv-train-auc:0.994161+0.000853\n",
"[1186]\tcv-test-auc:0.894201+0.008832\tcv-train-auc:0.994168+0.000852\n",
"[1187]\tcv-test-auc:0.894189+0.008850\tcv-train-auc:0.994175+0.000849\n",
"[1188]\tcv-test-auc:0.894152+0.008836\tcv-train-auc:0.994189+0.000850\n",
"[1189]\tcv-test-auc:0.894197+0.008864\tcv-train-auc:0.994200+0.000847\n",
"[1190]\tcv-test-auc:0.894170+0.008834\tcv-train-auc:0.994214+0.000828\n",
"[1191]\tcv-test-auc:0.894151+0.008831\tcv-train-auc:0.994232+0.000829\n",
"[1192]\tcv-test-auc:0.894151+0.008816\tcv-train-auc:0.994245+0.000826\n",
"[1193]\tcv-test-auc:0.894165+0.008801\tcv-train-auc:0.994260+0.000821\n",
"[1194]\tcv-test-auc:0.894156+0.008797\tcv-train-auc:0.994267+0.000822\n",
"[1195]\tcv-test-auc:0.894149+0.008779\tcv-train-auc:0.994280+0.000825\n",
"[1196]\tcv-test-auc:0.894137+0.008790\tcv-train-auc:0.994293+0.000827\n",
"[1197]\tcv-test-auc:0.894140+0.008793\tcv-train-auc:0.994312+0.000819\n",
"[1198]\tcv-test-auc:0.894121+0.008792\tcv-train-auc:0.994318+0.000818\n",
"[1199]\tcv-test-auc:0.894116+0.008802\tcv-train-auc:0.994326+0.000821\n",
"[1200]\tcv-test-auc:0.894136+0.008786\tcv-train-auc:0.994348+0.000820\n",
"[1201]\tcv-test-auc:0.894146+0.008803\tcv-train-auc:0.994359+0.000817\n",
"[1202]\tcv-test-auc:0.894127+0.008832\tcv-train-auc:0.994384+0.000804\n",
"[1203]\tcv-test-auc:0.894123+0.008844\tcv-train-auc:0.994389+0.000805\n",
"[1204]\tcv-test-auc:0.894119+0.008804\tcv-train-auc:0.994398+0.000798\n",
"[1205]\tcv-test-auc:0.894088+0.008825\tcv-train-auc:0.994422+0.000793\n",
"[1206]\tcv-test-auc:0.894064+0.008825\tcv-train-auc:0.994428+0.000793\n",
"[1207]\tcv-test-auc:0.894069+0.008808\tcv-train-auc:0.994447+0.000778\n",
"[1208]\tcv-test-auc:0.894060+0.008756\tcv-train-auc:0.994457+0.000777\n",
"[1209]\tcv-test-auc:0.894033+0.008796\tcv-train-auc:0.994480+0.000766\n",
"[1210]\tcv-test-auc:0.894015+0.008781\tcv-train-auc:0.994490+0.000772\n",
"[1211]\tcv-test-auc:0.894003+0.008799\tcv-train-auc:0.994498+0.000767\n",
"[1212]\tcv-test-auc:0.894040+0.008769\tcv-train-auc:0.994510+0.000768\n",
"[1213]\tcv-test-auc:0.894038+0.008771\tcv-train-auc:0.994516+0.000772\n",
"[1214]\tcv-test-auc:0.894028+0.008798\tcv-train-auc:0.994525+0.000769\n",
"[1215]\tcv-test-auc:0.894022+0.008799\tcv-train-auc:0.994537+0.000769\n",
"[1216]\tcv-test-auc:0.894037+0.008792\tcv-train-auc:0.994561+0.000764\n",
"[1217]\tcv-test-auc:0.894021+0.008779\tcv-train-auc:0.994570+0.000759\n",
"[1218]\tcv-test-auc:0.894075+0.008728\tcv-train-auc:0.994601+0.000753\n",
"[1219]\tcv-test-auc:0.894049+0.008729\tcv-train-auc:0.994618+0.000755\n",
"[1220]\tcv-test-auc:0.894028+0.008766\tcv-train-auc:0.994632+0.000743\n",
"[1221]\tcv-test-auc:0.894026+0.008765\tcv-train-auc:0.994649+0.000749\n",
"[1222]\tcv-test-auc:0.893981+0.008794\tcv-train-auc:0.994661+0.000747\n",
"[1223]\tcv-test-auc:0.893981+0.008796\tcv-train-auc:0.994680+0.000742\n",
"[1224]\tcv-test-auc:0.893982+0.008805\tcv-train-auc:0.994697+0.000724\n",
"[1225]\tcv-test-auc:0.893956+0.008775\tcv-train-auc:0.994721+0.000711\n",
"[1226]\tcv-test-auc:0.893979+0.008812\tcv-train-auc:0.994729+0.000713\n",
"[1227]\tcv-test-auc:0.893961+0.008783\tcv-train-auc:0.994746+0.000713\n",
"[1228]\tcv-test-auc:0.893925+0.008822\tcv-train-auc:0.994760+0.000706\n",
"[1229]\tcv-test-auc:0.893928+0.008823\tcv-train-auc:0.994772+0.000708\n",
"[1230]\tcv-test-auc:0.893964+0.008782\tcv-train-auc:0.994792+0.000697\n",
"[1231]\tcv-test-auc:0.893980+0.008741\tcv-train-auc:0.994806+0.000686\n",
"[1232]\tcv-test-auc:0.893997+0.008747\tcv-train-auc:0.994820+0.000688\n",
"[1233]\tcv-test-auc:0.893978+0.008749\tcv-train-auc:0.994832+0.000694\n",
"[1234]\tcv-test-auc:0.894031+0.008789\tcv-train-auc:0.994851+0.000703\n",
"[1235]\tcv-test-auc:0.894061+0.008738\tcv-train-auc:0.994863+0.000693\n",
"[1236]\tcv-test-auc:0.894065+0.008701\tcv-train-auc:0.994878+0.000684\n",
"[1237]\tcv-test-auc:0.894102+0.008672\tcv-train-auc:0.994897+0.000672\n",
"[1238]\tcv-test-auc:0.894095+0.008685\tcv-train-auc:0.994907+0.000670\n",
"[1239]\tcv-test-auc:0.894104+0.008711\tcv-train-auc:0.994924+0.000667\n",
"[1240]\tcv-test-auc:0.894127+0.008720\tcv-train-auc:0.994945+0.000657\n",
"[1241]\tcv-test-auc:0.894121+0.008731\tcv-train-auc:0.994954+0.000650\n",
"[1242]\tcv-test-auc:0.894135+0.008732\tcv-train-auc:0.994957+0.000649\n",
"[1243]\tcv-test-auc:0.894127+0.008773\tcv-train-auc:0.994966+0.000642\n",
"[1244]\tcv-test-auc:0.894136+0.008797\tcv-train-auc:0.994987+0.000632\n",
"[1245]\tcv-test-auc:0.894143+0.008832\tcv-train-auc:0.995000+0.000628\n",
"[1246]\tcv-test-auc:0.894157+0.008844\tcv-train-auc:0.995003+0.000629\n",
"[1247]\tcv-test-auc:0.894197+0.008855\tcv-train-auc:0.995008+0.000633\n",
"[1248]\tcv-test-auc:0.894151+0.008818\tcv-train-auc:0.995017+0.000626\n",
"[1249]\tcv-test-auc:0.894149+0.008787\tcv-train-auc:0.995031+0.000628\n",
"[1250]\tcv-test-auc:0.894202+0.008831\tcv-train-auc:0.995041+0.000631\n",
"[1251]\tcv-test-auc:0.894200+0.008801\tcv-train-auc:0.995052+0.000626\n",
"[1252]\tcv-test-auc:0.894200+0.008809\tcv-train-auc:0.995057+0.000622\n",
"[1253]\tcv-test-auc:0.894205+0.008803\tcv-train-auc:0.995073+0.000627\n",
"[1254]\tcv-test-auc:0.894225+0.008811\tcv-train-auc:0.995081+0.000624\n",
"[1255]\tcv-test-auc:0.894220+0.008811\tcv-train-auc:0.995091+0.000626\n",
"[1256]\tcv-test-auc:0.894220+0.008814\tcv-train-auc:0.995106+0.000629\n",
"[1257]\tcv-test-auc:0.894190+0.008836\tcv-train-auc:0.995122+0.000622\n",
"[1258]\tcv-test-auc:0.894222+0.008865\tcv-train-auc:0.995128+0.000623\n",
"[1259]\tcv-test-auc:0.894254+0.008843\tcv-train-auc:0.995145+0.000616\n",
"[1260]\tcv-test-auc:0.894266+0.008828\tcv-train-auc:0.995165+0.000617\n",
"[1261]\tcv-test-auc:0.894275+0.008776\tcv-train-auc:0.995183+0.000602\n",
"[1262]\tcv-test-auc:0.894312+0.008718\tcv-train-auc:0.995197+0.000600\n",
"[1263]\tcv-test-auc:0.894329+0.008752\tcv-train-auc:0.995201+0.000600\n",
"[1264]\tcv-test-auc:0.894327+0.008745\tcv-train-auc:0.995213+0.000602\n",
"[1265]\tcv-test-auc:0.894281+0.008731\tcv-train-auc:0.995233+0.000581\n",
"[1266]\tcv-test-auc:0.894269+0.008724\tcv-train-auc:0.995243+0.000577\n",
"[1267]\tcv-test-auc:0.894250+0.008740\tcv-train-auc:0.995253+0.000586\n",
"[1268]\tcv-test-auc:0.894269+0.008738\tcv-train-auc:0.995264+0.000584\n",
"[1269]\tcv-test-auc:0.894328+0.008740\tcv-train-auc:0.995280+0.000580\n",
"[1270]\tcv-test-auc:0.894300+0.008744\tcv-train-auc:0.995285+0.000578\n",
"[1271]\tcv-test-auc:0.894313+0.008739\tcv-train-auc:0.995297+0.000578\n",
"[1272]\tcv-test-auc:0.894331+0.008705\tcv-train-auc:0.995304+0.000573\n",
"[1273]\tcv-test-auc:0.894331+0.008687\tcv-train-auc:0.995317+0.000556\n",
"[1274]\tcv-test-auc:0.894326+0.008668\tcv-train-auc:0.995325+0.000555\n",
"[1275]\tcv-test-auc:0.894340+0.008664\tcv-train-auc:0.995333+0.000555\n",
"[1276]\tcv-test-auc:0.894299+0.008671\tcv-train-auc:0.995346+0.000547\n",
"[1277]\tcv-test-auc:0.894357+0.008719\tcv-train-auc:0.995359+0.000561\n",
"[1278]\tcv-test-auc:0.894403+0.008689\tcv-train-auc:0.995370+0.000555\n",
"[1279]\tcv-test-auc:0.894413+0.008700\tcv-train-auc:0.995379+0.000550\n",
"[1280]\tcv-test-auc:0.894395+0.008704\tcv-train-auc:0.995389+0.000539\n",
"[1281]\tcv-test-auc:0.894359+0.008728\tcv-train-auc:0.995402+0.000542\n",
"[1282]\tcv-test-auc:0.894348+0.008704\tcv-train-auc:0.995404+0.000540\n",
"[1283]\tcv-test-auc:0.894336+0.008775\tcv-train-auc:0.995412+0.000540\n",
"[1284]\tcv-test-auc:0.894304+0.008718\tcv-train-auc:0.995433+0.000535\n",
"[1285]\tcv-test-auc:0.894330+0.008660\tcv-train-auc:0.995439+0.000533\n",
"[1286]\tcv-test-auc:0.894325+0.008672\tcv-train-auc:0.995440+0.000534\n",
"[1287]\tcv-test-auc:0.894338+0.008648\tcv-train-auc:0.995448+0.000534\n",
"[1288]\tcv-test-auc:0.894331+0.008678\tcv-train-auc:0.995463+0.000542\n",
"[1289]\tcv-test-auc:0.894319+0.008699\tcv-train-auc:0.995471+0.000541\n",
"[1290]\tcv-test-auc:0.894315+0.008705\tcv-train-auc:0.995488+0.000531\n",
"[1291]\tcv-test-auc:0.894337+0.008708\tcv-train-auc:0.995512+0.000519\n",
"[1292]\tcv-test-auc:0.894364+0.008719\tcv-train-auc:0.995520+0.000522\n",
"[1293]\tcv-test-auc:0.894387+0.008709\tcv-train-auc:0.995532+0.000517\n",
"[1294]\tcv-test-auc:0.894404+0.008705\tcv-train-auc:0.995545+0.000521\n",
"[1295]\tcv-test-auc:0.894374+0.008721\tcv-train-auc:0.995559+0.000520\n",
"[1296]\tcv-test-auc:0.894357+0.008718\tcv-train-auc:0.995567+0.000521\n",
"[1297]\tcv-test-auc:0.894360+0.008720\tcv-train-auc:0.995575+0.000515\n",
"[1298]\tcv-test-auc:0.894355+0.008731\tcv-train-auc:0.995582+0.000514\n",
"[1299]\tcv-test-auc:0.894373+0.008715\tcv-train-auc:0.995591+0.000513\n",
"[1300]\tcv-test-auc:0.894349+0.008722\tcv-train-auc:0.995603+0.000519\n",
"[1301]\tcv-test-auc:0.894340+0.008718\tcv-train-auc:0.995607+0.000517\n",
"[1302]\tcv-test-auc:0.894376+0.008745\tcv-train-auc:0.995615+0.000506\n",
"[1303]\tcv-test-auc:0.894371+0.008756\tcv-train-auc:0.995633+0.000507\n",
"[1304]\tcv-test-auc:0.894379+0.008737\tcv-train-auc:0.995638+0.000510\n",
"[1305]\tcv-test-auc:0.894315+0.008733\tcv-train-auc:0.995651+0.000512\n",
"[1306]\tcv-test-auc:0.894378+0.008770\tcv-train-auc:0.995663+0.000515\n",
"[1307]\tcv-test-auc:0.894387+0.008782\tcv-train-auc:0.995676+0.000504\n",
"[1308]\tcv-test-auc:0.894404+0.008814\tcv-train-auc:0.995683+0.000506\n",
"[1309]\tcv-test-auc:0.894421+0.008803\tcv-train-auc:0.995692+0.000508\n",
"[1310]\tcv-test-auc:0.894394+0.008820\tcv-train-auc:0.995696+0.000504\n",
"[1311]\tcv-test-auc:0.894415+0.008811\tcv-train-auc:0.995704+0.000495\n",
"[1312]\tcv-test-auc:0.894392+0.008848\tcv-train-auc:0.995721+0.000499\n",
"[1313]\tcv-test-auc:0.894364+0.008836\tcv-train-auc:0.995726+0.000499\n",
"[1314]\tcv-test-auc:0.894385+0.008822\tcv-train-auc:0.995741+0.000498\n",
"[1315]\tcv-test-auc:0.894397+0.008854\tcv-train-auc:0.995746+0.000499\n",
"[1316]\tcv-test-auc:0.894416+0.008855\tcv-train-auc:0.995748+0.000500\n",
"[1317]\tcv-test-auc:0.894442+0.008760\tcv-train-auc:0.995754+0.000496\n",
"[1318]\tcv-test-auc:0.894462+0.008751\tcv-train-auc:0.995762+0.000485\n",
"[1319]\tcv-test-auc:0.894466+0.008759\tcv-train-auc:0.995768+0.000480\n",
"[1320]\tcv-test-auc:0.894454+0.008784\tcv-train-auc:0.995775+0.000477\n",
"[1321]\tcv-test-auc:0.894452+0.008747\tcv-train-auc:0.995778+0.000480\n",
"[1322]\tcv-test-auc:0.894453+0.008738\tcv-train-auc:0.995780+0.000481\n",
"[1323]\tcv-test-auc:0.894485+0.008726\tcv-train-auc:0.995788+0.000480\n",
"[1324]\tcv-test-auc:0.894478+0.008711\tcv-train-auc:0.995800+0.000475\n",
"[1325]\tcv-test-auc:0.894483+0.008737\tcv-train-auc:0.995809+0.000477\n",
"[1326]\tcv-test-auc:0.894475+0.008722\tcv-train-auc:0.995823+0.000474\n",
"[1327]\tcv-test-auc:0.894497+0.008756\tcv-train-auc:0.995835+0.000475\n",
"[1328]\tcv-test-auc:0.894470+0.008770\tcv-train-auc:0.995846+0.000475\n",
"[1329]\tcv-test-auc:0.894446+0.008784\tcv-train-auc:0.995856+0.000471\n",
"[1330]\tcv-test-auc:0.894486+0.008794\tcv-train-auc:0.995873+0.000464\n",
"[1331]\tcv-test-auc:0.894490+0.008822\tcv-train-auc:0.995891+0.000460\n",
"[1332]\tcv-test-auc:0.894500+0.008842\tcv-train-auc:0.995896+0.000461\n",
"[1333]\tcv-test-auc:0.894466+0.008846\tcv-train-auc:0.995902+0.000460\n"
]
},
{
"ename": "KeyboardInterrupt",
"evalue": "",
"output_type": "error",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[0;31mKeyboardInterrupt\u001b[0m Traceback (most recent call last)",
"\u001b[0;32m<ipython-input-60-333eeb3fab1f>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mxgb\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcv\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mparams\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mxgtrain\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnum_rounds\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnfold\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m4\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmetrics\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m{\u001b[0m\u001b[0;34m'auc'\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
"\u001b[0;32m/Users/pksaha/anaconda/lib/python2.7/site-packages/xgboost/training.pyc\u001b[0m in \u001b[0;36mcv\u001b[0;34m(params, dtrain, num_boost_round, nfold, metrics, obj, feval, fpreproc, show_stdv, seed)\u001b[0m\n\u001b[1;32m 245\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnum_boost_round\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 246\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mfold\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mcvfolds\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 247\u001b[0;31m \u001b[0mfold\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mupdate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mobj\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 248\u001b[0m \u001b[0mres\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0maggcv\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0meval\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeval\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mf\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mcvfolds\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mshow_stdv\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 249\u001b[0m \u001b[0msys\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstderr\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwrite\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mres\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;34m'\\n'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Users/pksaha/anaconda/lib/python2.7/site-packages/xgboost/training.pyc\u001b[0m in \u001b[0;36mupdate\u001b[0;34m(self, iteration, fobj)\u001b[0m\n\u001b[1;32m 150\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mupdate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0miteration\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfobj\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 151\u001b[0m \u001b[0;34m\"\"\"\"Update the boosters for one iteration\"\"\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 152\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbst\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mupdate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdtrain\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0miteration\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfobj\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 153\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 154\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0meval\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0miteration\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeval\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Users/pksaha/anaconda/lib/python2.7/site-packages/xgboost/core.pyc\u001b[0m in \u001b[0;36mupdate\u001b[0;34m(self, dtrain, iteration, fobj)\u001b[0m\n\u001b[1;32m 521\u001b[0m \u001b[0;32mraise\u001b[0m \u001b[0mTypeError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'invalid training matrix: {}'\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mformat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdtrain\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__name__\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 522\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mfobj\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 523\u001b[0;31m \u001b[0m_check_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_LIB\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mXGBoosterUpdateOneIter\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mhandle\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0miteration\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtrain\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mhandle\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 524\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 525\u001b[0m \u001b[0mpred\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpredict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdtrain\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;31mKeyboardInterrupt\u001b[0m: "
]
}
],
"source": [
"xgb.cv(params, xgtrain, num_rounds, nfold=4, metrics={'auc'})"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"As we can see from above, the test set auc decreases after around 285 rounds while the train set auc keeps increasing, clearly indicating overfitting at this point. We now go ahead with fine tuning other parameters and then finally building the model as below. "
]
},
{
"cell_type": "code",
"execution_count": 133,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"params = {}\n",
"params[\"objective\"] = \"binary:logistic\"\n",
"#To avoid overfitting: The first way is to directly control model complexity\n",
"params[\"min_child_weight\"] = 1 #The larger, the more conservative the algorithm will be.\n",
"params[\"max_depth\"] = 5\n",
"#params[\"gamma\"] = 0 #The larger, the more conservative the algorithm will be.\n",
"params[\"eta\"] = 0.03 #higher is more conservative [0,1], if reduced then increase num_rounds\n",
"#The second way is to add randomness to make training robust to noise\n",
"params[\"subsample\"] = 0.9\n",
"params[\"colsample_bytree\"] = 0.8\n",
"\n",
"#Handle Imbalanced Dataset\n",
"#If you care only about the ranking order (AUC) of your prediction\n",
"params[\"scale_pos_weight\"] = 0.8 #ratio of labels in target variable\n",
"params[\"eval_metric \"] = 'auc'\n",
"#If you care about predicting the right probability\n",
"params[\"max_delta_step\"]= 8 #should be high for skewed data\n",
"\n",
"params[\"seed\"] = 0\n",
"params[\"silent\"] = 1\n",
"params[\"nthread\"] = 4\n",
"plst = list(params.items())\n",
"num_rounds = 300"
]
},
{
"cell_type": "code",
"execution_count": 134,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"xgtrain = xgb.DMatrix(trainX_mobN, label=trainY_mobN, weight= trainX_mobN_weight)\n",
"xgtest = xgb.DMatrix(holdoutX_mobN)\n",
"model_mobN = xgb.train(plst, xgtrain, num_rounds)\n",
"pred_test_y_xgb_mobN = model_mobN.predict(xgtest)"
]
},
{
"cell_type": "code",
"execution_count": 135,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.axes._subplots.AxesSubplot at 0x11a2759d0>"
]
},
"execution_count": 135,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAX8AAAEZCAYAAAB/6SUgAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x11a360fd0>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"xgb.plot_importance(model_mobN)"
]
},
{
"cell_type": "code",
"execution_count": 136,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"holdoutX_mobN['predicted'] = pred_test_y_xgb_mobN"
]
},
{
"cell_type": "code",
"execution_count": 137,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"xgtrain = xgb.DMatrix(trainX_mobY, label=trainY_mobY, weight= trainX_mobY_weight)\n",
"xgtest = xgb.DMatrix(holdoutX_mobY)\n",
"model_mobY = xgb.train(plst, xgtrain, num_rounds)\n",
"pred_test_y_xgb_mobY = model_mobY.predict(xgtest)"
]
},
{
"cell_type": "code",
"execution_count": 138,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.axes._subplots.AxesSubplot at 0x119e29a50>"
]
},
"execution_count": 138,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAX8AAAEZCAYAAAB/6SUgAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nLnaL+wAAAABJRU5ErkJggg==\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x118f3d710>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"xgb.plot_importance(model_mobY)"
]
},
{
"cell_type": "code",
"execution_count": 139,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"holdoutX_mobY['predicted'] = pred_test_y_xgb_mobY"
]
},
{
"cell_type": "code",
"execution_count": 145,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"#Save model\n",
"with open('model_mobY', 'wb') as f:\n",
" cPickle.dump(model_mobY, f)\n",
"with open('model_mobN', 'wb') as f:\n",
" cPickle.dump(model_mobN, f)"
]
},
{
"cell_type": "code",
"execution_count": 140,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"holdout_full = pd.concat([holdoutX_mobN,holdoutX_mobY], axis = 0)"
]
},
{
"cell_type": "code",
"execution_count": 141,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"holdout_full = holdout_full.sort_index()"
]
},
{
"cell_type": "code",
"execution_count": 142,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>ID</th>\n",
" <th>Gender</th>\n",
" <th>City</th>\n",
" <th>Monthly_Income</th>\n",
" <th>DOB</th>\n",
" <th>Lead_Creation_Date</th>\n",
" <th>Loan_Amount_Applied</th>\n",
" <th>Loan_Tenure_Applied</th>\n",
" <th>Existing_EMI</th>\n",
" <th>Employer_Name</th>\n",
" <th>...</th>\n",
" <th>Var2</th>\n",
" <th>Source</th>\n",
" <th>Var4</th>\n",
" <th>LoggedIn</th>\n",
" <th>Disbursed</th>\n",
" <th>month</th>\n",
" <th>day</th>\n",
" <th>Age</th>\n",
" <th>emi2income</th>\n",
" <th>FOIR</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>55785</th>\n",
" <td>ID080049V40</td>\n",
" <td>1</td>\n",
" <td>172</td>\n",
" <td>10.308986</td>\n",
" <td>1977-05-03</td>\n",
" <td>2015-07-30</td>\n",
" <td>0.000000</td>\n",
" <td>0</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>30</td>\n",
" <td>38</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>55970</th>\n",
" <td>ID080335V00</td>\n",
" <td>1</td>\n",
" <td>625</td>\n",
" <td>10.184938</td>\n",
" <td>1992-06-19</td>\n",
" <td>2015-07-26</td>\n",
" <td>13.122365</td>\n",
" <td>5</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>26</td>\n",
" <td>23</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>58253</th>\n",
" <td>ID083576M10</td>\n",
" <td>1</td>\n",
" <td>172</td>\n",
" <td>10.373522</td>\n",
" <td>1984-05-26</td>\n",
" <td>2015-07-30</td>\n",
" <td>12.611541</td>\n",
" <td>5</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>30</td>\n",
" <td>31</td>\n",
" <td>0.218091</td>\n",
" <td>0.218091</td>\n",
" </tr>\n",
" <tr>\n",
" <th>58283</th>\n",
" <td>ID083617B20</td>\n",
" <td>0</td>\n",
" <td>999</td>\n",
" <td>10.819798</td>\n",
" <td>1984-10-18</td>\n",
" <td>2015-07-21</td>\n",
" <td>13.122365</td>\n",
" <td>5</td>\n",
" <td>10.064798</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>21</td>\n",
" <td>31</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>61302</th>\n",
" <td>ID087951T10</td>\n",
" <td>0</td>\n",
" <td>371</td>\n",
" <td>10.126671</td>\n",
" <td>1977-06-18</td>\n",
" <td>2015-07-30</td>\n",
" <td>0.000000</td>\n",
" <td>0</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>30</td>\n",
" <td>38</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>61484</th>\n",
" <td>ID088187V20</td>\n",
" <td>1</td>\n",
" <td>172</td>\n",
" <td>10.341775</td>\n",
" <td>1990-06-20</td>\n",
" <td>2015-07-29</td>\n",
" <td>0.000000</td>\n",
" <td>0</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>29</td>\n",
" <td>25</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>61661</th>\n",
" <td>ID088430E00</td>\n",
" <td>1</td>\n",
" <td>371</td>\n",
" <td>10.373522</td>\n",
" <td>1986-07-09</td>\n",
" <td>2015-07-20</td>\n",
" <td>0.000000</td>\n",
" <td>0</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>5</td>\n",
" <td>1</td>\n",
" <td>1</td>\n",
" <td>7</td>\n",
" <td>20</td>\n",
" <td>29</td>\n",
" <td>0.443974</td>\n",
" <td>0.443974</td>\n",
" </tr>\n",
" <tr>\n",
" <th>62930</th>\n",
" <td>ID090277F20</td>\n",
" <td>0</td>\n",
" <td>523</td>\n",
" <td>10.757924</td>\n",
" <td>1984-06-08</td>\n",
" <td>2015-07-20</td>\n",
" <td>12.206078</td>\n",
" <td>5</td>\n",
" <td>10.203629</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>20</td>\n",
" <td>31</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>63039</th>\n",
" <td>ID090415N00</td>\n",
" <td>1</td>\n",
" <td>523</td>\n",
" <td>10.308986</td>\n",
" <td>1982-01-27</td>\n",
" <td>2015-07-22</td>\n",
" <td>12.611541</td>\n",
" <td>3</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>22</td>\n",
" <td>33</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>63245</th>\n",
" <td>ID090728O30</td>\n",
" <td>1</td>\n",
" <td>446</td>\n",
" <td>10.545368</td>\n",
" <td>1979-10-07</td>\n",
" <td>2015-07-21</td>\n",
" <td>11.512935</td>\n",
" <td>2</td>\n",
" <td>6.908755</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>5</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>21</td>\n",
" <td>36</td>\n",
" <td>0.126337</td>\n",
" <td>0.152653</td>\n",
" </tr>\n",
" <tr>\n",
" <th>63920</th>\n",
" <td>ID091690O00</td>\n",
" <td>1</td>\n",
" <td>172</td>\n",
" <td>10.404293</td>\n",
" <td>1975-11-23</td>\n",
" <td>2015-07-20</td>\n",
" <td>11.512935</td>\n",
" <td>2</td>\n",
" <td>7.550135</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>20</td>\n",
" <td>40</td>\n",
" <td>0.145479</td>\n",
" <td>0.203055</td>\n",
" </tr>\n",
" <tr>\n",
" <th>64068</th>\n",
" <td>ID091921L10</td>\n",
" <td>0</td>\n",
" <td>134</td>\n",
" <td>10.239996</td>\n",
" <td>1988-03-09</td>\n",
" <td>2015-07-23</td>\n",
" <td>11.512935</td>\n",
" <td>2</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>23</td>\n",
" <td>27</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>64511</th>\n",
" <td>ID092566G10</td>\n",
" <td>0</td>\n",
" <td>172</td>\n",
" <td>10.518700</td>\n",
" <td>1984-04-30</td>\n",
" <td>2015-07-19</td>\n",
" <td>12.206078</td>\n",
" <td>5</td>\n",
" <td>9.546884</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>19</td>\n",
" <td>31</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>64784</th>\n",
" <td>ID092958I30</td>\n",
" <td>0</td>\n",
" <td>446</td>\n",
" <td>10.308986</td>\n",
" <td>1996-06-08</td>\n",
" <td>2015-07-28</td>\n",
" <td>13.122365</td>\n",
" <td>5</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>28</td>\n",
" <td>19</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>64929</th>\n",
" <td>ID093172O20</td>\n",
" <td>0</td>\n",
" <td>673</td>\n",
" <td>10.463132</td>\n",
" <td>1981-12-09</td>\n",
" <td>2015-07-19</td>\n",
" <td>11.512935</td>\n",
" <td>5</td>\n",
" <td>9.903538</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>19</td>\n",
" <td>34</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>64990</th>\n",
" <td>ID093270I00</td>\n",
" <td>1</td>\n",
" <td>8</td>\n",
" <td>10.341775</td>\n",
" <td>1988-05-24</td>\n",
" <td>2015-07-19</td>\n",
" <td>12.611541</td>\n",
" <td>1</td>\n",
" <td>9.047939</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>19</td>\n",
" <td>27</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>65139</th>\n",
" <td>ID093490U00</td>\n",
" <td>1</td>\n",
" <td>523</td>\n",
" <td>10.341775</td>\n",
" <td>1987-05-16</td>\n",
" <td>2015-07-21</td>\n",
" <td>13.815512</td>\n",
" <td>5</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>4</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>21</td>\n",
" <td>28</td>\n",
" <td>0.352698</td>\n",
" <td>0.352698</td>\n",
" </tr>\n",
" <tr>\n",
" <th>65308</th>\n",
" <td>ID093734E40</td>\n",
" <td>1</td>\n",
" <td>446</td>\n",
" <td>11.141876</td>\n",
" <td>1986-04-28</td>\n",
" <td>2015-07-21</td>\n",
" <td>14.220976</td>\n",
" <td>5</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>21</td>\n",
" <td>29</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>65633</th>\n",
" <td>ID094187P20</td>\n",
" <td>0</td>\n",
" <td>523</td>\n",
" <td>10.457401</td>\n",
" <td>1982-10-20</td>\n",
" <td>2015-07-22</td>\n",
" <td>13.122365</td>\n",
" <td>3</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>22</td>\n",
" <td>33</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>65849</th>\n",
" <td>ID094496M10</td>\n",
" <td>1</td>\n",
" <td>523</td>\n",
" <td>10.373522</td>\n",
" <td>1981-08-14</td>\n",
" <td>2015-07-22</td>\n",
" <td>12.206078</td>\n",
" <td>5</td>\n",
" <td>9.268704</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>5</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>22</td>\n",
" <td>34</td>\n",
" <td>0.145394</td>\n",
" <td>0.476644</td>\n",
" </tr>\n",
" <tr>\n",
" <th>65930</th>\n",
" <td>ID094623J30</td>\n",
" <td>1</td>\n",
" <td>172</td>\n",
" <td>10.502461</td>\n",
" <td>1987-05-01</td>\n",
" <td>2015-07-21</td>\n",
" <td>12.611541</td>\n",
" <td>2</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>21</td>\n",
" <td>28</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>65945</th>\n",
" <td>ID094641B10</td>\n",
" <td>0</td>\n",
" <td>523</td>\n",
" <td>9.998843</td>\n",
" <td>1989-04-14</td>\n",
" <td>2015-07-21</td>\n",
" <td>0.000000</td>\n",
" <td>0</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>2</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>21</td>\n",
" <td>26</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>66083</th>\n",
" <td>ID094832K20</td>\n",
" <td>0</td>\n",
" <td>130</td>\n",
" <td>10.021315</td>\n",
" <td>1986-09-10</td>\n",
" <td>2015-07-25</td>\n",
" <td>13.122365</td>\n",
" <td>5</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>25</td>\n",
" <td>29</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>66108</th>\n",
" <td>ID094866S10</td>\n",
" <td>1</td>\n",
" <td>999</td>\n",
" <td>10.950824</td>\n",
" <td>1985-07-01</td>\n",
" <td>2015-07-18</td>\n",
" <td>12.206078</td>\n",
" <td>5</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>18</td>\n",
" <td>30</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>66145</th>\n",
" <td>ID094930E00</td>\n",
" <td>0</td>\n",
" <td>446</td>\n",
" <td>9.903538</td>\n",
" <td>1989-10-18</td>\n",
" <td>2015-07-22</td>\n",
" <td>0.000000</td>\n",
" <td>0</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>22</td>\n",
" <td>26</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>66287</th>\n",
" <td>ID095145L00</td>\n",
" <td>0</td>\n",
" <td>446</td>\n",
" <td>9.852247</td>\n",
" <td>1984-11-12</td>\n",
" <td>2015-07-19</td>\n",
" <td>11.512935</td>\n",
" <td>4</td>\n",
" <td>8.160804</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>19</td>\n",
" <td>31</td>\n",
" <td>0.277412</td>\n",
" <td>0.461623</td>\n",
" </tr>\n",
" <tr>\n",
" <th>66436</th>\n",
" <td>ID095355N00</td>\n",
" <td>1</td>\n",
" <td>371</td>\n",
" <td>10.621352</td>\n",
" <td>1984-08-01</td>\n",
" <td>2015-07-20</td>\n",
" <td>12.206078</td>\n",
" <td>2</td>\n",
" <td>8.673000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>20</td>\n",
" <td>31</td>\n",
" <td>0.234186</td>\n",
" <td>0.376674</td>\n",
" </tr>\n",
" <tr>\n",
" <th>67001</th>\n",
" <td>ID096183J30</td>\n",
" <td>0</td>\n",
" <td>87</td>\n",
" <td>11.507923</td>\n",
" <td>2050-01-01</td>\n",
" <td>2015-07-21</td>\n",
" <td>13.122365</td>\n",
" <td>5</td>\n",
" <td>10.463132</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>1</td>\n",
" <td>1</td>\n",
" <td>7</td>\n",
" <td>21</td>\n",
" <td>65</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>67171</th>\n",
" <td>ID096423P30</td>\n",
" <td>0</td>\n",
" <td>172</td>\n",
" <td>11.018646</td>\n",
" <td>1987-11-22</td>\n",
" <td>2015-07-27</td>\n",
" <td>12.611541</td>\n",
" <td>3</td>\n",
" <td>9.903538</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>1</td>\n",
" <td>1</td>\n",
" <td>7</td>\n",
" <td>27</td>\n",
" <td>28</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>67505</th>\n",
" <td>ID096902A20</td>\n",
" <td>1</td>\n",
" <td>134</td>\n",
" <td>10.463132</td>\n",
" <td>1987-05-11</td>\n",
" <td>2015-07-20</td>\n",
" <td>12.206078</td>\n",
" <td>5</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>20</td>\n",
" <td>28</td>\n",
" <td>0.132932</td>\n",
" <td>0.132932</td>\n",
" </tr>\n",
" <tr>\n",
" <th>...</th>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>86990</th>\n",
" <td>ID124780G00</td>\n",
" <td>1</td>\n",
" <td>363</td>\n",
" <td>11.289794</td>\n",
" <td>1973-05-16</td>\n",
" <td>2015-07-31</td>\n",
" <td>0.000000</td>\n",
" <td>0</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>42</td>\n",
" <td>0.503238</td>\n",
" <td>0.503238</td>\n",
" </tr>\n",
" <tr>\n",
" <th>86991</th>\n",
" <td>ID124782I20</td>\n",
" <td>1</td>\n",
" <td>363</td>\n",
" <td>11.289794</td>\n",
" <td>1973-05-16</td>\n",
" <td>2015-07-31</td>\n",
" <td>0.000000</td>\n",
" <td>0</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>42</td>\n",
" <td>0.576927</td>\n",
" <td>0.576927</td>\n",
" </tr>\n",
" <tr>\n",
" <th>86992</th>\n",
" <td>ID124783J30</td>\n",
" <td>0</td>\n",
" <td>172</td>\n",
" <td>10.126671</td>\n",
" <td>1992-07-19</td>\n",
" <td>2015-07-31</td>\n",
" <td>11.289794</td>\n",
" <td>1</td>\n",
" <td>0.000000</td>\n",
" <td>16163</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>23</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>86993</th>\n",
" <td>ID124785L00</td>\n",
" <td>0</td>\n",
" <td>446</td>\n",
" <td>10.463132</td>\n",
" <td>1981-12-01</td>\n",
" <td>2015-07-31</td>\n",
" <td>0.000000</td>\n",
" <td>0</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>34</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>86994</th>\n",
" <td>ID124786M10</td>\n",
" <td>1</td>\n",
" <td>295</td>\n",
" <td>9.615872</td>\n",
" <td>1984-09-12</td>\n",
" <td>2015-07-31</td>\n",
" <td>12.206078</td>\n",
" <td>3</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>4</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>31</td>\n",
" <td>0.587418</td>\n",
" <td>0.587418</td>\n",
" </tr>\n",
" <tr>\n",
" <th>86995</th>\n",
" <td>ID124787N20</td>\n",
" <td>1</td>\n",
" <td>999</td>\n",
" <td>10.308986</td>\n",
" <td>1978-07-27</td>\n",
" <td>2015-07-31</td>\n",
" <td>0.000000</td>\n",
" <td>0</td>\n",
" <td>9.210440</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>37</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>86996</th>\n",
" <td>ID124788O30</td>\n",
" <td>1</td>\n",
" <td>523</td>\n",
" <td>10.819798</td>\n",
" <td>1986-05-30</td>\n",
" <td>2015-07-31</td>\n",
" <td>13.122365</td>\n",
" <td>2</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>29</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>86997</th>\n",
" <td>ID124789P40</td>\n",
" <td>1</td>\n",
" <td>172</td>\n",
" <td>10.373522</td>\n",
" <td>1986-10-21</td>\n",
" <td>2015-07-31</td>\n",
" <td>-999.000000</td>\n",
" <td>-999</td>\n",
" <td>-999.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>29</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>86998</th>\n",
" <td>ID124790Q00</td>\n",
" <td>1</td>\n",
" <td>8</td>\n",
" <td>10.714440</td>\n",
" <td>2056-08-31</td>\n",
" <td>2015-07-31</td>\n",
" <td>0.000000</td>\n",
" <td>0</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>59</td>\n",
" <td>0.499586</td>\n",
" <td>0.499586</td>\n",
" </tr>\n",
" <tr>\n",
" <th>86999</th>\n",
" <td>ID124791R10</td>\n",
" <td>0</td>\n",
" <td>999</td>\n",
" <td>9.472782</td>\n",
" <td>1990-06-05</td>\n",
" <td>2015-07-31</td>\n",
" <td>12.206078</td>\n",
" <td>2</td>\n",
" <td>8.517393</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>25</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>87000</th>\n",
" <td>ID124792S20</td>\n",
" <td>0</td>\n",
" <td>999</td>\n",
" <td>10.849822</td>\n",
" <td>2062-11-01</td>\n",
" <td>2015-07-31</td>\n",
" <td>12.611541</td>\n",
" <td>5</td>\n",
" <td>10.071076</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>53</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>87001</th>\n",
" <td>ID124793T30</td>\n",
" <td>1</td>\n",
" <td>147</td>\n",
" <td>10.878066</td>\n",
" <td>1987-06-29</td>\n",
" <td>2015-07-31</td>\n",
" <td>0.000000</td>\n",
" <td>0</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>28</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>87002</th>\n",
" <td>ID124794U40</td>\n",
" <td>0</td>\n",
" <td>999</td>\n",
" <td>9.472782</td>\n",
" <td>1990-06-05</td>\n",
" <td>2015-07-31</td>\n",
" <td>11.512935</td>\n",
" <td>2</td>\n",
" <td>8.517393</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>25</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>87003</th>\n",
" <td>ID124795V00</td>\n",
" <td>1</td>\n",
" <td>536</td>\n",
" <td>10.596660</td>\n",
" <td>2060-01-01</td>\n",
" <td>2015-07-31</td>\n",
" <td>13.458837</td>\n",
" <td>0</td>\n",
" <td>9.042040</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>55</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>87004</th>\n",
" <td>ID124796W10</td>\n",
" <td>1</td>\n",
" <td>41</td>\n",
" <td>10.373522</td>\n",
" <td>1978-05-22</td>\n",
" <td>2015-07-31</td>\n",
" <td>13.815512</td>\n",
" <td>5</td>\n",
" <td>0.000000</td>\n",
" <td>98</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>37</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>87005</th>\n",
" <td>ID124798Y30</td>\n",
" <td>1</td>\n",
" <td>999</td>\n",
" <td>11.661354</td>\n",
" <td>1969-02-21</td>\n",
" <td>2015-07-31</td>\n",
" <td>13.815512</td>\n",
" <td>5</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>46</td>\n",
" <td>0.312743</td>\n",
" <td>0.312743</td>\n",
" </tr>\n",
" <tr>\n",
" <th>87006</th>\n",
" <td>ID124799Z40</td>\n",
" <td>0</td>\n",
" <td>87</td>\n",
" <td>9.210440</td>\n",
" <td>1989-05-01</td>\n",
" <td>2015-07-31</td>\n",
" <td>11.512935</td>\n",
" <td>2</td>\n",
" <td>8.160804</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>26</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>87007</th>\n",
" <td>ID124802C20</td>\n",
" <td>1</td>\n",
" <td>271</td>\n",
" <td>9.615872</td>\n",
" <td>1988-05-02</td>\n",
" <td>2015-07-31</td>\n",
" <td>12.206078</td>\n",
" <td>3</td>\n",
" <td>6.216606</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>27</td>\n",
" <td>0.548179</td>\n",
" <td>0.581513</td>\n",
" </tr>\n",
" <tr>\n",
" <th>87008</th>\n",
" <td>ID124803D30</td>\n",
" <td>0</td>\n",
" <td>523</td>\n",
" <td>10.463132</td>\n",
" <td>1981-03-31</td>\n",
" <td>2015-07-31</td>\n",
" <td>13.815512</td>\n",
" <td>5</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>34</td>\n",
" <td>0.594617</td>\n",
" <td>0.594617</td>\n",
" </tr>\n",
" <tr>\n",
" <th>87009</th>\n",
" <td>ID124804E40</td>\n",
" <td>1</td>\n",
" <td>172</td>\n",
" <td>11.798112</td>\n",
" <td>1985-08-14</td>\n",
" <td>2015-07-31</td>\n",
" <td>12.206078</td>\n",
" <td>0</td>\n",
" <td>10.434145</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>4</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>30</td>\n",
" <td>0.041085</td>\n",
" <td>0.296724</td>\n",
" </tr>\n",
" <tr>\n",
" <th>87010</th>\n",
" <td>ID124806G10</td>\n",
" <td>1</td>\n",
" <td>462</td>\n",
" <td>10.239996</td>\n",
" <td>1973-06-10</td>\n",
" <td>2015-07-31</td>\n",
" <td>0.000000</td>\n",
" <td>0</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>42</td>\n",
" <td>0.495621</td>\n",
" <td>0.495621</td>\n",
" </tr>\n",
" <tr>\n",
" <th>87011</th>\n",
" <td>ID124808I30</td>\n",
" <td>1</td>\n",
" <td>87</td>\n",
" <td>9.615872</td>\n",
" <td>1990-06-01</td>\n",
" <td>2015-07-31</td>\n",
" <td>0.000000</td>\n",
" <td>0</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>25</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>87012</th>\n",
" <td>ID124810K00</td>\n",
" <td>1</td>\n",
" <td>87</td>\n",
" <td>10.736418</td>\n",
" <td>1985-01-02</td>\n",
" <td>2015-07-31</td>\n",
" <td>12.611541</td>\n",
" <td>3</td>\n",
" <td>0.000000</td>\n",
" <td>7977</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>4</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>30</td>\n",
" <td>0.219743</td>\n",
" <td>0.219743</td>\n",
" </tr>\n",
" <tr>\n",
" <th>87013</th>\n",
" <td>ID124811L10</td>\n",
" <td>1</td>\n",
" <td>582</td>\n",
" <td>10.085851</td>\n",
" <td>1990-01-01</td>\n",
" <td>2015-07-31</td>\n",
" <td>12.611541</td>\n",
" <td>3</td>\n",
" <td>0.000000</td>\n",
" <td>16802</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>25</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>87014</th>\n",
" <td>ID124812M20</td>\n",
" <td>0</td>\n",
" <td>523</td>\n",
" <td>10.799596</td>\n",
" <td>1982-05-31</td>\n",
" <td>2015-07-31</td>\n",
" <td>12.899222</td>\n",
" <td>5</td>\n",
" <td>0.000000</td>\n",
" <td>17199</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>33</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>87015</th>\n",
" <td>ID124813N30</td>\n",
" <td>0</td>\n",
" <td>999</td>\n",
" <td>11.183059</td>\n",
" <td>1969-11-27</td>\n",
" <td>2015-07-31</td>\n",
" <td>13.815512</td>\n",
" <td>5</td>\n",
" <td>9.581973</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>46</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>87016</th>\n",
" <td>ID124814O40</td>\n",
" <td>0</td>\n",
" <td>363</td>\n",
" <td>9.680406</td>\n",
" <td>1990-12-01</td>\n",
" <td>2015-07-31</td>\n",
" <td>0.000000</td>\n",
" <td>0</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>25</td>\n",
" <td>0.589110</td>\n",
" <td>0.589110</td>\n",
" </tr>\n",
" <tr>\n",
" <th>87017</th>\n",
" <td>ID124816Q10</td>\n",
" <td>1</td>\n",
" <td>87</td>\n",
" <td>11.678448</td>\n",
" <td>1972-01-28</td>\n",
" <td>2015-07-31</td>\n",
" <td>0.000000</td>\n",
" <td>0</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>43</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>87018</th>\n",
" <td>ID124818S30</td>\n",
" <td>1</td>\n",
" <td>87</td>\n",
" <td>11.502178</td>\n",
" <td>1977-04-27</td>\n",
" <td>2015-07-31</td>\n",
" <td>13.592368</td>\n",
" <td>5</td>\n",
" <td>9.522300</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>38</td>\n",
" <td>-999.000000</td>\n",
" <td>-999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>87019</th>\n",
" <td>ID124821V10</td>\n",
" <td>1</td>\n",
" <td>446</td>\n",
" <td>10.652566</td>\n",
" <td>1988-10-31</td>\n",
" <td>2015-07-31</td>\n",
" <td>0.000000</td>\n",
" <td>0</td>\n",
" <td>0.000000</td>\n",
" <td>999</td>\n",
" <td>...</td>\n",
" <td>6</td>\n",
" <td>0</td>\n",
" <td>4</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>7</td>\n",
" <td>31</td>\n",
" <td>27</td>\n",
" <td>0.445669</td>\n",
" <td>0.445669</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>16222 rows × 31 columns</p>\n",
"</div>"
],
"text/plain": [
" ID Gender City Monthly_Income DOB \\\n",
"55785 ID080049V40 1 172 10.308986 1977-05-03 \n",
"55970 ID080335V00 1 625 10.184938 1992-06-19 \n",
"58253 ID083576M10 1 172 10.373522 1984-05-26 \n",
"58283 ID083617B20 0 999 10.819798 1984-10-18 \n",
"61302 ID087951T10 0 371 10.126671 1977-06-18 \n",
"61484 ID088187V20 1 172 10.341775 1990-06-20 \n",
"61661 ID088430E00 1 371 10.373522 1986-07-09 \n",
"62930 ID090277F20 0 523 10.757924 1984-06-08 \n",
"63039 ID090415N00 1 523 10.308986 1982-01-27 \n",
"63245 ID090728O30 1 446 10.545368 1979-10-07 \n",
"63920 ID091690O00 1 172 10.404293 1975-11-23 \n",
"64068 ID091921L10 0 134 10.239996 1988-03-09 \n",
"64511 ID092566G10 0 172 10.518700 1984-04-30 \n",
"64784 ID092958I30 0 446 10.308986 1996-06-08 \n",
"64929 ID093172O20 0 673 10.463132 1981-12-09 \n",
"64990 ID093270I00 1 8 10.341775 1988-05-24 \n",
"65139 ID093490U00 1 523 10.341775 1987-05-16 \n",
"65308 ID093734E40 1 446 11.141876 1986-04-28 \n",
"65633 ID094187P20 0 523 10.457401 1982-10-20 \n",
"65849 ID094496M10 1 523 10.373522 1981-08-14 \n",
"65930 ID094623J30 1 172 10.502461 1987-05-01 \n",
"65945 ID094641B10 0 523 9.998843 1989-04-14 \n",
"66083 ID094832K20 0 130 10.021315 1986-09-10 \n",
"66108 ID094866S10 1 999 10.950824 1985-07-01 \n",
"66145 ID094930E00 0 446 9.903538 1989-10-18 \n",
"66287 ID095145L00 0 446 9.852247 1984-11-12 \n",
"66436 ID095355N00 1 371 10.621352 1984-08-01 \n",
"67001 ID096183J30 0 87 11.507923 2050-01-01 \n",
"67171 ID096423P30 0 172 11.018646 1987-11-22 \n",
"67505 ID096902A20 1 134 10.463132 1987-05-11 \n",
"... ... ... ... ... ... \n",
"86990 ID124780G00 1 363 11.289794 1973-05-16 \n",
"86991 ID124782I20 1 363 11.289794 1973-05-16 \n",
"86992 ID124783J30 0 172 10.126671 1992-07-19 \n",
"86993 ID124785L00 0 446 10.463132 1981-12-01 \n",
"86994 ID124786M10 1 295 9.615872 1984-09-12 \n",
"86995 ID124787N20 1 999 10.308986 1978-07-27 \n",
"86996 ID124788O30 1 523 10.819798 1986-05-30 \n",
"86997 ID124789P40 1 172 10.373522 1986-10-21 \n",
"86998 ID124790Q00 1 8 10.714440 2056-08-31 \n",
"86999 ID124791R10 0 999 9.472782 1990-06-05 \n",
"87000 ID124792S20 0 999 10.849822 2062-11-01 \n",
"87001 ID124793T30 1 147 10.878066 1987-06-29 \n",
"87002 ID124794U40 0 999 9.472782 1990-06-05 \n",
"87003 ID124795V00 1 536 10.596660 2060-01-01 \n",
"87004 ID124796W10 1 41 10.373522 1978-05-22 \n",
"87005 ID124798Y30 1 999 11.661354 1969-02-21 \n",
"87006 ID124799Z40 0 87 9.210440 1989-05-01 \n",
"87007 ID124802C20 1 271 9.615872 1988-05-02 \n",
"87008 ID124803D30 0 523 10.463132 1981-03-31 \n",
"87009 ID124804E40 1 172 11.798112 1985-08-14 \n",
"87010 ID124806G10 1 462 10.239996 1973-06-10 \n",
"87011 ID124808I30 1 87 9.615872 1990-06-01 \n",
"87012 ID124810K00 1 87 10.736418 1985-01-02 \n",
"87013 ID124811L10 1 582 10.085851 1990-01-01 \n",
"87014 ID124812M20 0 523 10.799596 1982-05-31 \n",
"87015 ID124813N30 0 999 11.183059 1969-11-27 \n",
"87016 ID124814O40 0 363 9.680406 1990-12-01 \n",
"87017 ID124816Q10 1 87 11.678448 1972-01-28 \n",
"87018 ID124818S30 1 87 11.502178 1977-04-27 \n",
"87019 ID124821V10 1 446 10.652566 1988-10-31 \n",
"\n",
" Lead_Creation_Date Loan_Amount_Applied Loan_Tenure_Applied \\\n",
"55785 2015-07-30 0.000000 0 \n",
"55970 2015-07-26 13.122365 5 \n",
"58253 2015-07-30 12.611541 5 \n",
"58283 2015-07-21 13.122365 5 \n",
"61302 2015-07-30 0.000000 0 \n",
"61484 2015-07-29 0.000000 0 \n",
"61661 2015-07-20 0.000000 0 \n",
"62930 2015-07-20 12.206078 5 \n",
"63039 2015-07-22 12.611541 3 \n",
"63245 2015-07-21 11.512935 2 \n",
"63920 2015-07-20 11.512935 2 \n",
"64068 2015-07-23 11.512935 2 \n",
"64511 2015-07-19 12.206078 5 \n",
"64784 2015-07-28 13.122365 5 \n",
"64929 2015-07-19 11.512935 5 \n",
"64990 2015-07-19 12.611541 1 \n",
"65139 2015-07-21 13.815512 5 \n",
"65308 2015-07-21 14.220976 5 \n",
"65633 2015-07-22 13.122365 3 \n",
"65849 2015-07-22 12.206078 5 \n",
"65930 2015-07-21 12.611541 2 \n",
"65945 2015-07-21 0.000000 0 \n",
"66083 2015-07-25 13.122365 5 \n",
"66108 2015-07-18 12.206078 5 \n",
"66145 2015-07-22 0.000000 0 \n",
"66287 2015-07-19 11.512935 4 \n",
"66436 2015-07-20 12.206078 2 \n",
"67001 2015-07-21 13.122365 5 \n",
"67171 2015-07-27 12.611541 3 \n",
"67505 2015-07-20 12.206078 5 \n",
"... ... ... ... \n",
"86990 2015-07-31 0.000000 0 \n",
"86991 2015-07-31 0.000000 0 \n",
"86992 2015-07-31 11.289794 1 \n",
"86993 2015-07-31 0.000000 0 \n",
"86994 2015-07-31 12.206078 3 \n",
"86995 2015-07-31 0.000000 0 \n",
"86996 2015-07-31 13.122365 2 \n",
"86997 2015-07-31 -999.000000 -999 \n",
"86998 2015-07-31 0.000000 0 \n",
"86999 2015-07-31 12.206078 2 \n",
"87000 2015-07-31 12.611541 5 \n",
"87001 2015-07-31 0.000000 0 \n",
"87002 2015-07-31 11.512935 2 \n",
"87003 2015-07-31 13.458837 0 \n",
"87004 2015-07-31 13.815512 5 \n",
"87005 2015-07-31 13.815512 5 \n",
"87006 2015-07-31 11.512935 2 \n",
"87007 2015-07-31 12.206078 3 \n",
"87008 2015-07-31 13.815512 5 \n",
"87009 2015-07-31 12.206078 0 \n",
"87010 2015-07-31 0.000000 0 \n",
"87011 2015-07-31 0.000000 0 \n",
"87012 2015-07-31 12.611541 3 \n",
"87013 2015-07-31 12.611541 3 \n",
"87014 2015-07-31 12.899222 5 \n",
"87015 2015-07-31 13.815512 5 \n",
"87016 2015-07-31 0.000000 0 \n",
"87017 2015-07-31 0.000000 0 \n",
"87018 2015-07-31 13.592368 5 \n",
"87019 2015-07-31 0.000000 0 \n",
"\n",
" Existing_EMI Employer_Name ... Var2 Source Var4 LoggedIn \\\n",
"55785 0.000000 999 ... 6 0 1 0 \n",
"55970 0.000000 999 ... 6 0 3 0 \n",
"58253 0.000000 999 ... 6 0 5 0 \n",
"58283 10.064798 999 ... 6 0 1 0 \n",
"61302 0.000000 999 ... 6 0 1 0 \n",
"61484 0.000000 999 ... 6 0 3 0 \n",
"61661 0.000000 999 ... 6 0 5 1 \n",
"62930 10.203629 999 ... 6 0 1 0 \n",
"63039 0.000000 999 ... 6 0 3 0 \n",
"63245 6.908755 999 ... 6 0 5 1 \n",
"63920 7.550135 999 ... 6 0 5 0 \n",
"64068 0.000000 999 ... 6 0 1 0 \n",
"64511 9.546884 999 ... 6 0 1 0 \n",
"64784 0.000000 999 ... 6 0 1 0 \n",
"64929 9.903538 999 ... 6 0 1 0 \n",
"64990 9.047939 999 ... 6 0 3 0 \n",
"65139 0.000000 999 ... 6 0 4 0 \n",
"65308 0.000000 999 ... 6 0 1 0 \n",
"65633 0.000000 999 ... 6 0 3 0 \n",
"65849 9.268704 999 ... 6 0 5 1 \n",
"65930 0.000000 999 ... 6 0 3 0 \n",
"65945 0.000000 999 ... 6 0 2 0 \n",
"66083 0.000000 999 ... 6 0 1 0 \n",
"66108 0.000000 999 ... 6 0 3 0 \n",
"66145 0.000000 999 ... 6 0 3 0 \n",
"66287 8.160804 999 ... 6 0 1 0 \n",
"66436 8.673000 999 ... 6 0 5 0 \n",
"67001 10.463132 999 ... 6 0 1 1 \n",
"67171 9.903538 999 ... 6 0 1 1 \n",
"67505 0.000000 999 ... 6 0 5 0 \n",
"... ... ... ... ... ... ... ... \n",
"86990 0.000000 999 ... 6 0 5 0 \n",
"86991 0.000000 999 ... 6 0 5 0 \n",
"86992 0.000000 16163 ... 6 0 7 0 \n",
"86993 0.000000 999 ... 6 0 3 0 \n",
"86994 0.000000 999 ... 6 0 4 0 \n",
"86995 9.210440 999 ... 6 0 3 0 \n",
"86996 0.000000 999 ... 6 0 3 0 \n",
"86997 -999.000000 999 ... 6 0 1 0 \n",
"86998 0.000000 999 ... 6 0 5 0 \n",
"86999 8.517393 999 ... 6 0 1 0 \n",
"87000 10.071076 999 ... 6 0 1 0 \n",
"87001 0.000000 999 ... 6 0 3 0 \n",
"87002 8.517393 999 ... 6 0 1 0 \n",
"87003 9.042040 999 ... 6 0 3 0 \n",
"87004 0.000000 98 ... 6 0 7 0 \n",
"87005 0.000000 999 ... 6 0 3 0 \n",
"87006 8.160804 999 ... 6 0 1 0 \n",
"87007 6.216606 999 ... 6 0 5 0 \n",
"87008 0.000000 999 ... 6 0 5 0 \n",
"87009 10.434145 999 ... 6 0 4 0 \n",
"87010 0.000000 999 ... 6 0 5 0 \n",
"87011 0.000000 999 ... 6 0 3 0 \n",
"87012 0.000000 7977 ... 6 0 4 0 \n",
"87013 0.000000 16802 ... 6 0 3 0 \n",
"87014 0.000000 17199 ... 6 0 3 0 \n",
"87015 9.581973 999 ... 6 0 3 0 \n",
"87016 0.000000 999 ... 6 0 5 0 \n",
"87017 0.000000 999 ... 6 0 3 0 \n",
"87018 9.522300 999 ... 6 0 3 0 \n",
"87019 0.000000 999 ... 6 0 4 0 \n",
"\n",
" Disbursed month day Age emi2income FOIR \n",
"55785 0 7 30 38 -999.000000 -999.000000 \n",
"55970 0 7 26 23 -999.000000 -999.000000 \n",
"58253 0 7 30 31 0.218091 0.218091 \n",
"58283 0 7 21 31 -999.000000 -999.000000 \n",
"61302 0 7 30 38 -999.000000 -999.000000 \n",
"61484 0 7 29 25 -999.000000 -999.000000 \n",
"61661 1 7 20 29 0.443974 0.443974 \n",
"62930 0 7 20 31 -999.000000 -999.000000 \n",
"63039 0 7 22 33 -999.000000 -999.000000 \n",
"63245 0 7 21 36 0.126337 0.152653 \n",
"63920 0 7 20 40 0.145479 0.203055 \n",
"64068 0 7 23 27 -999.000000 -999.000000 \n",
"64511 0 7 19 31 -999.000000 -999.000000 \n",
"64784 0 7 28 19 -999.000000 -999.000000 \n",
"64929 0 7 19 34 -999.000000 -999.000000 \n",
"64990 0 7 19 27 -999.000000 -999.000000 \n",
"65139 0 7 21 28 0.352698 0.352698 \n",
"65308 0 7 21 29 -999.000000 -999.000000 \n",
"65633 0 7 22 33 -999.000000 -999.000000 \n",
"65849 0 7 22 34 0.145394 0.476644 \n",
"65930 0 7 21 28 -999.000000 -999.000000 \n",
"65945 0 7 21 26 -999.000000 -999.000000 \n",
"66083 0 7 25 29 -999.000000 -999.000000 \n",
"66108 0 7 18 30 -999.000000 -999.000000 \n",
"66145 0 7 22 26 -999.000000 -999.000000 \n",
"66287 0 7 19 31 0.277412 0.461623 \n",
"66436 0 7 20 31 0.234186 0.376674 \n",
"67001 1 7 21 65 -999.000000 -999.000000 \n",
"67171 1 7 27 28 -999.000000 -999.000000 \n",
"67505 0 7 20 28 0.132932 0.132932 \n",
"... ... ... ... ... ... ... \n",
"86990 0 7 31 42 0.503238 0.503238 \n",
"86991 0 7 31 42 0.576927 0.576927 \n",
"86992 0 7 31 23 -999.000000 -999.000000 \n",
"86993 0 7 31 34 -999.000000 -999.000000 \n",
"86994 0 7 31 31 0.587418 0.587418 \n",
"86995 0 7 31 37 -999.000000 -999.000000 \n",
"86996 0 7 31 29 -999.000000 -999.000000 \n",
"86997 0 7 31 29 -999.000000 -999.000000 \n",
"86998 0 7 31 59 0.499586 0.499586 \n",
"86999 0 7 31 25 -999.000000 -999.000000 \n",
"87000 0 7 31 53 -999.000000 -999.000000 \n",
"87001 0 7 31 28 -999.000000 -999.000000 \n",
"87002 0 7 31 25 -999.000000 -999.000000 \n",
"87003 0 7 31 55 -999.000000 -999.000000 \n",
"87004 0 7 31 37 -999.000000 -999.000000 \n",
"87005 0 7 31 46 0.312743 0.312743 \n",
"87006 0 7 31 26 -999.000000 -999.000000 \n",
"87007 0 7 31 27 0.548179 0.581513 \n",
"87008 0 7 31 34 0.594617 0.594617 \n",
"87009 0 7 31 30 0.041085 0.296724 \n",
"87010 0 7 31 42 0.495621 0.495621 \n",
"87011 0 7 31 25 -999.000000 -999.000000 \n",
"87012 0 7 31 30 0.219743 0.219743 \n",
"87013 0 7 31 25 -999.000000 -999.000000 \n",
"87014 0 7 31 33 -999.000000 -999.000000 \n",
"87015 0 7 31 46 -999.000000 -999.000000 \n",
"87016 0 7 31 25 0.589110 0.589110 \n",
"87017 0 7 31 43 -999.000000 -999.000000 \n",
"87018 0 7 31 38 -999.000000 -999.000000 \n",
"87019 0 7 31 27 0.445669 0.445669 \n",
"\n",
"[16222 rows x 31 columns]"
]
},
"execution_count": 142,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"hold_out"
]
},
{
"cell_type": "code",
"execution_count": 143,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAqIAAAHzCAYAAAD7O388AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nLp+9yXOhAAAAAElFTkSuQmCC\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x11aac8890>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/plain": [
"<ggplot: (296405173)>"
]
},
"execution_count": 143,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"%matplotlib inline\n",
"fpr, tpr, _ = roc_curve(hold_out['Disbursed'], holdout_full['predicted'])\n",
"df = pd.DataFrame(dict(fpr=fpr, tpr=tpr))\n",
"ggplot(df, aes(x='fpr', y='tpr')) + geom_line() + geom_abline(linetype='dashed')"
]
},
{
"cell_type": "code",
"execution_count": 144,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"0.846528928043\n"
]
}
],
"source": [
"roc_auc = auc(fpr, tpr)\n",
"print roc_auc"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Our model gave a fairly decent hold_out set score. We now go ahead and train the model on the entire train data set. "
]
},
{
"cell_type": "code",
"execution_count": 156,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"trainXfull_mobN = pd.concat([trainX_mobN, holdoutX_mobN.ix[:,:-1]], axis = 0)\n",
"trainXfull_mobN = trainXfull_mobN.sort_index()"
]
},
{
"cell_type": "code",
"execution_count": 160,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"trainYfull_mobN = pd.concat([trainY_mobN, holdoutY_mobN], axis = 0)\n",
"trainYfull_mobN = trainYfull_mobN.sort_index()"
]
},
{
"cell_type": "code",
"execution_count": 161,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"trainXfull_mobY = pd.concat([trainX_mobY, holdoutX_mobY.ix[:,:-1]], axis = 0)\n",
"trainXfull_mobY = trainXfull_mobY.sort_index()\n",
"trainYfull_mobY = pd.concat([trainY_mobY, holdoutY_mobY], axis = 0)\n",
"trainYfull_mobY = trainYfull_mobY.sort_index()"
]
},
{
"cell_type": "code",
"execution_count": 162,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"#calculate the weihts for the entire train data set\n",
"trainXfull_mobN_weight = [2 if data == 1 else 1 for data in trainYfull_mobN]\n",
"trainXfull_mobY_weight = [2 if data == 1 else 1 for data in trainYfull_mobY]"
]
},
{
"cell_type": "code",
"execution_count": 163,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"#train on the full data set\n",
"xgtrain_full_mobN = xgb.DMatrix(trainXfull_mobN, label=trainYfull_mobN, weight= trainXfull_mobN_weight)\n",
"xgtest = xgb.DMatrix(test_mobN)\n",
"modelfull_mobN = xgb.train(plst, xgtrain_full_mobN, num_rounds)\n",
"finalpred_mobN = modelfull_mobN.predict(xgtest)"
]
},
{
"cell_type": "code",
"execution_count": 164,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"xgtrain_full_mobY = xgb.DMatrix(trainXfull_mobY, label=trainYfull_mobY, weight= trainXfull_mobY_weight)\n",
"xgtest = xgb.DMatrix(test_mobY)\n",
"modelfull_mobY = xgb.train(plst, xgtrain_full_mobY, num_rounds)\n",
"finalpred_mobY = modelfull_mobY.predict(xgtest)"
]
},
{
"cell_type": "code",
"execution_count": 165,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"test_mobY['Disbursed'] = finalpred_mobY\n",
"test_mobN['Disbursed'] = finalpred_mobN"
]
},
{
"cell_type": "code",
"execution_count": 166,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"test_full = pd.concat([test_mobY,test_mobN], axis = 0)"
]
},
{
"cell_type": "code",
"execution_count": 167,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"test_full = test_full.sort_index()"
]
},
{
"cell_type": "code",
"execution_count": 168,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"test_full['ID'] = test['ID']"
]
},
{
"cell_type": "code",
"execution_count": 169,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"test_full.to_csv('submission.csv', columns=['ID','Disbursed'], index=False)"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": true
},
"source": [
"The above submission scored 82.85 on the private LB, good enough for a top 10 spot in the weekend hackathon and a top 20 in the weeklong. "
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": true
},
"source": [
"### Scope of Improvement\n",
"1. As there was a gap of almost two points between hold out set AUC score and private LB AUC score, so chosing a better hold out set for better evaluation of the model is of top most priority\n",
"2. Better tuning of Hyperparameters \n",
"3. Ensembling of model to achieve higher score"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 2",
"language": "python",
"name": "python2"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 2
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython2",
"version": "2.7.10"
}
},
"nbformat": 4,
"nbformat_minor": 0
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment