Data Science Portfolio

Details of my Data Science projects

Home About GitHub project Currently v2.1.0

© 2024. All rights reserved.

Credit EDA

13 Jul 2024

Problem Statement

This case study aims to identify patterns which indicate if a client has difficulty paying their instalments which may be used for taking actions such as denying the loan, reducing the amount of loan, lending (to risky applicants) at a higher interest rate, etc. This will ensure that the consumers capable of repaying the loan are not rejected. Identification of such applicants using EDA is the aim of this case study.

In other words, the company wants to understand the driving factors (or driver variables) behind loan default, i.e. the variables which are strong indicators of default.  The company can utilise this knowledge for its portfolio and risk assessment.

Business Understanding

The loan providing companies find it hard to give loans to the people due to their insufficient or non-existent credit history. Because of that, some consumers use it to their advantage by becoming a defaulter. Consumer finance company which specialises in lending various types of loans to urban customers. Te task is to use EDA to analyse the patterns present in the data. This will ensure that the applicants capable of repaying the loan are not rejected.

When the company receives a loan application, the company has to decide for loan approval based on the applicant’s profile. Two types of risks are associated with the bank’s decision:

If the applicant is likely to repay the loan, then not approving the loan results in a loss of business to the company

If the applicant is not likely to repay the loan, i.e. he/she is likely to default, then approving the loan may lead to a financial loss for the company.

The data given below contains the information about the loan application at the time of applying for the loan. It contains two types of scenarios:

The client with payment difficulties: he/she had late payment more than X days on at least one of the first Y instalments of the loan in our sample,

All other cases: All other cases when the payment is paid on time.

When a client applies for a loan, there are four types of decisions that could be taken by the client/company):

Approved: The Company has approved loan Application

Cancelled: The client cancelled the application sometime during approval. Either the client changed her/his mind about the loan or in some cases due to a higher risk of the client, he received worse pricing which he did not want.

Refused: The company had rejected the loan (because the client does not meet their requirements etc.).

Unused offer:  Loan has been cancelled by the client but at different stages of the process.

In this case study, you will use EDA to understand how consumer attributes and loan attributes influence the tendency to default.

Steps to perform EDA:

Data Cleaning

• Column Renaming
• Missing Value Imputation
• Outlier Analysis
• Incorrect values replacement

Data Imbalance Data Visualization

• Univariate Analysis
• Bivariate Analysis
• Multivaratie Analysis

Credit EDA

• This file comprises of all the analysis of the given dataset.
• Note: Some code cells, specifically those that generates Graphs/Charts may run for a longer time.
• Coded using VScode, Ubuntu 22-64bit, 16GB RAM, 4.5GHZ

# import all libraries numpy, pandas, matplotlib, seaborn. 

import numpy as np, pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns; sns.set_theme(color_codes=True)

import warnings
warnings.filterwarnings('ignore')

from IPython.core.interactiveshell import InteractiveShell
InteractiveShell.ast_node_interactivity = 'all'
# %matplotlib inline

# Set custom display properties in pandas
pd.set_option("display.max_rows", 900)
pd.set_option("display.max_columns", 900) 
pd.set_option('display.float_format', lambda x: '%.3f' % x)

# read 'application_data.csv' and store the dataframe as 'curr_appl_data'
curr_appl_data = pd.read_csv("application_data.csv", sep=",", header=0)

# read 'previous_application.csv' and store the dataframe as 'prev_appl_data'
prev_appl_data = pd.read_csv("previous_application.csv", sep=",", header=0)

Custom Functions

# classify_feature_dtype function is used to distinguish numerical columns either as categorical/discrete or non categorical and return as dict
def classify_feature_dtype(df,cols):
    
    d_categories = {'int_cat': [], "float_ts":[] }
    for col in cols:
        if (len(df[col].unique()) < 10):
            d_categories['int_cat'].append(col)
        else:
            d_categories['float_ts'].append(col)
    return d_categories

# show_stats function is used to print all statistical information for a given set of columns
def show_stats(df, cols):
    for col in list(cols):
        print("Total Nulls: {0},\nMode: {1}".format(df[col].isna().sum(), df[col].mode()[0]))
        if len(df[col].unique()) < 50:
            print("\nUnique: {0}\n".format(df[col].unique()))
        if (df[col].dtype == int) or (df[col].dtype == float):
            print("Median   : {0}, \nVariance: {1}, \n\nDescribe: {2} \n".format(df[col].median(), df[col].var(), df[col].describe()))
        print("ValueCounts: {0} \n\n\n".format((df[col].value_counts(normalize=True) * 100).head(5)))
        print("------------------------------------------------------------------")

# check_cols_null_pct - to return the percentage of null values in each columns in a dataframe
def check_cols_null_pct(df):
    df_non_na = df.count() / len(df)  # Ratio of non null values
    df_na_pct = (1 - df_non_na) * 100 # Find the Percentage of null values
    return df_na_pct.sort_values(ascending=False) # Sort the resulting values in descending order

# univariate_plots function generates charts based on the data type of the cols, as part of the univariate analysis 
# it takes dataframe, columns, train data 0,1, and feature type as args.
def univariate_plots(df, cols, t0=None, t1=None, ftype=None):
    
    for col in cols:
        #generate plots and graphs for category type. (generates piechart, countplot, boxplot / if training data is provided it generates bar chart instead)
        if ftype == "category":
            if (t0 is not None) and (t1 is not None):
                fig, axs = plt.subplots(1, 3, figsize=(20, 6))
            else:
                fig, axs = plt.subplots(1, 2, figsize=(20, 6))
            
            col_idx = 0
            
            axs[col_idx].pie(x=df[col].value_counts().head(5), labels=df[col].value_counts().head(5).index, autopct="%1.1f%%", 
                    radius=1, textprops={"fontsize": 10, "color": "Black"}, startangle=90, rotatelabels=False)
            axs[col_idx].set_title("Pichart of {0}".format(col), y=1); plt.xticks(rotation=45); plt.ylabel("Percentage")
            fig.subplots_adjust(wspace=0.5, hspace=0.3)
            
            col_idx += 1

            sns.countplot(data=df, y=col, order=df[col].value_counts().index, palette="viridis",  ax=axs[col_idx])
            axs[col_idx].set_title("Countplot of {0}".format(col)); plt.xticks(rotation=45); plt.xlabel(col); plt.ylabel("Count")
            fig.subplots_adjust(wspace=0.5, hspace=0.3)

            col_idx += 1
            
            if (t0 is not None) and (t1 is not None):
                dff = pd.DataFrame()
                dff['target_0_nodffty']= ((t0[col].value_counts())/len(t0))
                dff['target_1_paydfty']= ((t1[col].value_counts())/len(t1))
                dff.plot.bar(ax=axs[col_idx])
                axs[col_idx].set_title('Plotting data for target in terms of percentage'); plt.xticks(rotation=45); plt.xlabel(col)
                fig.subplots_adjust(wspace=0.5, hspace=0.3)
                
            # else:
            #     sns.boxplot(data=df, y=col, palette="viridis", flierprops=dict( marker="o", markersize=6, markerfacecolor="red", markeredgecolor="black", ),
            #             medianprops=dict(linestyle="-", linewidth=3, color="#FF9900"), whiskerprops=dict(linestyle="-", linewidth=2, color="black"),
            #             capprops=dict(linestyle="-", linewidth=2, color="black"), ax=axs[col_idx])
            #     axs[col_idx].set_title("Boxplot of {0}".format(col)); plt.xticks(rotation=45); plt.xlabel(col)
            #     fig.subplots_adjust(wspace=0.5, hspace=0.3)

            plt.suptitle("Univariate analysis of {0}".format(col), fontsize=12, y=0.95)
            plt.tight_layout()
            plt.subplots_adjust(top=0.85)
            plt.show()
            plt.clf()

        #generate plots and graphs for numerical types. (generates boxplot, histplot, kdeplot, scatterplot)
        elif ftype == "non_categorical":        
            fig, axs = plt.subplots(1, 4, figsize=(20, 6))
            col_idx = 0
            sns.boxplot(data=df, y=col, palette="viridis", 
                        flierprops=dict( marker="o", markersize=6, markerfacecolor="red", markeredgecolor="black"),
                        medianprops=dict(linestyle="-", linewidth=3, color="#FF9900"), whiskerprops=dict(linestyle="-", linewidth=2, color="black"),
                        capprops=dict(linestyle="-", linewidth=2, color="black"), ax=axs[col_idx])
            axs[col_idx].set_title("Boxplot of {0}".format(col)); plt.xticks(rotation=45); plt.xlabel(col)
            fig.subplots_adjust(wspace=0.5, hspace=0.3)
            col_idx += 1

            axs[col_idx].hist(data=df, x=col, label=col)
            axs[col_idx].set_title("Histogram of {0}".format(col)); plt.xticks(rotation=45); plt.xlabel(col)
            fig.subplots_adjust(wspace=0.5, hspace=0.3)
            col_idx += 1

            sns.kdeplot(df[col], shade=True, ax=axs[col_idx])
            axs[col_idx].set_title("KDE plot of {0}".format(col)); plt.xticks(rotation=45); plt.xlabel(col)
            fig.subplots_adjust(wspace=0.5, hspace=0.3)
            col_idx += 1

            sns.scatterplot(df[col], ax=axs[col_idx])
            axs[col_idx].set_title("Scatterplot of {0}".format(col)); plt.xticks(rotation=45); plt.xlabel(col)
            fig.subplots_adjust(wspace=0.5, hspace=0.3)

            plt.suptitle("Univariate analysis of {0}".format(col), fontsize=12, y=0.95)
            plt.tight_layout()
            plt.subplots_adjust(top=0.85)
            plt.show()
            plt.clf()

# bivariate_plots function generates charts based on the data type of the cols, as part of the bivariate analysis 
# it takes dataframe, columns, hue, train data 0,1, as args.
def bivariate_plots(df, col, hue, train0, train1):
    fig = plt.figure(figsize=(20,10))

    ax1 = plt.subplot(221)
    df[col].value_counts().plot.pie(autopct = "%1.1f%%", radius=1.4, textprops={"fontsize": 10, "color": "Black"}, startangle=30, rotatelabels=True, ax=ax1)
    plt.figtext(0.25, 0.55, 'Pie Chart for : '+ col, ha='center')
    # plt.title('Plotting data for the column: '+ col, loc='right', fontdict={'verticalalignment': 'baseline', 'horizontalalignment': 'right'})
    # fig.subplots_adjust(wspace=0.5, hspace=0.3)

    ax2 = plt.subplot(222)
    df = pd.DataFrame()
    df['0']= ((train0[col].value_counts())/len(train0))
    df['1']= ((train1[col].value_counts())/len(train1))
    df.plot.bar(ax=ax2,)
    plt.title('Plotting data for target in terms of total count')
    plt.xticks(rotation=45)
    fig.subplots_adjust(wspace=0.5, hspace=0.3)

    ax3 = plt.subplot(223)
    sns.countplot(x=col, hue=hue, data=train0, ax = ax3)
    plt.xticks(rotation=45)
    plt.title('Plotting data for Target=0 in terms of percentage')
    fig.subplots_adjust(wspace=0.5, hspace=0.3)

    ax4 = plt.subplot(224)
    sns.countplot(x=col, hue=hue, data=train1, ax = ax4)
    plt.xticks(rotation=45)
    plt.title('Plotting data for Target=1 in terms of percentage')
    fig.subplots_adjust(wspace=0.5, hspace=0.3)
    
    plt.tight_layout()
    
    plt.show()
    plt.clf()

Data Cleaning - Current Application Dataset

check_cols_null_pct(curr_appl_data)

COMMONAREA_MEDI                69.872
COMMONAREA_AVG                 69.872
COMMONAREA_MODE                69.872
NONLIVINGAPARTMENTS_MODE       69.433
NONLIVINGAPARTMENTS_AVG        69.433
NONLIVINGAPARTMENTS_MEDI       69.433
FONDKAPREMONT_MODE             68.386
LIVINGAPARTMENTS_MODE          68.355
LIVINGAPARTMENTS_AVG           68.355
LIVINGAPARTMENTS_MEDI          68.355
FLOORSMIN_AVG                  67.849
FLOORSMIN_MODE                 67.849
FLOORSMIN_MEDI                 67.849
YEARS_BUILD_MEDI               66.498
YEARS_BUILD_MODE               66.498
YEARS_BUILD_AVG                66.498
OWN_CAR_AGE                    65.991
LANDAREA_MEDI                  59.377
LANDAREA_MODE                  59.377
LANDAREA_AVG                   59.377
BASEMENTAREA_MEDI              58.516
BASEMENTAREA_AVG               58.516
BASEMENTAREA_MODE              58.516
EXT_SOURCE_1                   56.381
NONLIVINGAREA_MODE             55.179
NONLIVINGAREA_AVG              55.179
NONLIVINGAREA_MEDI             55.179
ELEVATORS_MEDI                 53.296
ELEVATORS_AVG                  53.296
ELEVATORS_MODE                 53.296
WALLSMATERIAL_MODE             50.841
APARTMENTS_MEDI                50.750
APARTMENTS_AVG                 50.750
APARTMENTS_MODE                50.750
ENTRANCES_MEDI                 50.349
ENTRANCES_AVG                  50.349
ENTRANCES_MODE                 50.349
LIVINGAREA_AVG                 50.193
LIVINGAREA_MODE                50.193
LIVINGAREA_MEDI                50.193
HOUSETYPE_MODE                 50.176
FLOORSMAX_MODE                 49.761
FLOORSMAX_MEDI                 49.761
FLOORSMAX_AVG                  49.761
YEARS_BEGINEXPLUATATION_MODE   48.781
YEARS_BEGINEXPLUATATION_MEDI   48.781
YEARS_BEGINEXPLUATATION_AVG    48.781
TOTALAREA_MODE                 48.269
EMERGENCYSTATE_MODE            47.398
OCCUPATION_TYPE                31.346
EXT_SOURCE_3                   19.825
AMT_REQ_CREDIT_BUREAU_HOUR     13.502
AMT_REQ_CREDIT_BUREAU_DAY      13.502
AMT_REQ_CREDIT_BUREAU_WEEK     13.502
AMT_REQ_CREDIT_BUREAU_MON      13.502
AMT_REQ_CREDIT_BUREAU_QRT      13.502
AMT_REQ_CREDIT_BUREAU_YEAR     13.502
NAME_TYPE_SUITE                 0.420
OBS_30_CNT_SOCIAL_CIRCLE        0.332
DEF_30_CNT_SOCIAL_CIRCLE        0.332
OBS_60_CNT_SOCIAL_CIRCLE        0.332
DEF_60_CNT_SOCIAL_CIRCLE        0.332
EXT_SOURCE_2                    0.215
AMT_GOODS_PRICE                 0.090
AMT_ANNUITY                     0.004
CNT_FAM_MEMBERS                 0.001
DAYS_LAST_PHONE_CHANGE          0.000
CNT_CHILDREN                    0.000
FLAG_DOCUMENT_8                 0.000
NAME_CONTRACT_TYPE              0.000
CODE_GENDER                     0.000
FLAG_OWN_CAR                    0.000
FLAG_DOCUMENT_2                 0.000
FLAG_DOCUMENT_3                 0.000
FLAG_DOCUMENT_4                 0.000
FLAG_DOCUMENT_5                 0.000
FLAG_DOCUMENT_6                 0.000
FLAG_DOCUMENT_7                 0.000
FLAG_DOCUMENT_9                 0.000
FLAG_DOCUMENT_21                0.000
FLAG_DOCUMENT_10                0.000
FLAG_DOCUMENT_11                0.000
FLAG_OWN_REALTY                 0.000
FLAG_DOCUMENT_13                0.000
FLAG_DOCUMENT_14                0.000
FLAG_DOCUMENT_15                0.000
FLAG_DOCUMENT_16                0.000
FLAG_DOCUMENT_17                0.000
FLAG_DOCUMENT_18                0.000
FLAG_DOCUMENT_19                0.000
FLAG_DOCUMENT_20                0.000
FLAG_DOCUMENT_12                0.000
AMT_CREDIT                      0.000
AMT_INCOME_TOTAL                0.000
FLAG_PHONE                      0.000
LIVE_CITY_NOT_WORK_CITY         0.000
REG_CITY_NOT_WORK_CITY          0.000
TARGET                          0.000
REG_CITY_NOT_LIVE_CITY          0.000
LIVE_REGION_NOT_WORK_REGION     0.000
REG_REGION_NOT_WORK_REGION      0.000
REG_REGION_NOT_LIVE_REGION      0.000
HOUR_APPR_PROCESS_START         0.000
WEEKDAY_APPR_PROCESS_START      0.000
REGION_RATING_CLIENT_W_CITY     0.000
REGION_RATING_CLIENT            0.000
FLAG_EMAIL                      0.000
FLAG_CONT_MOBILE                0.000
ORGANIZATION_TYPE               0.000
FLAG_WORK_PHONE                 0.000
FLAG_EMP_PHONE                  0.000
FLAG_MOBIL                      0.000
DAYS_ID_PUBLISH                 0.000
DAYS_REGISTRATION               0.000
DAYS_EMPLOYED                   0.000
DAYS_BIRTH                      0.000
REGION_POPULATION_RELATIVE      0.000
NAME_HOUSING_TYPE               0.000
NAME_FAMILY_STATUS              0.000
NAME_EDUCATION_TYPE             0.000
NAME_INCOME_TYPE                0.000
SK_ID_CURR                      0.000
dtype: float64

• The above outputs list contains many (~40+) columns that have null percentage greater than or equal to 50%.

# Checks the percentage of null values in each columns of a dataframe
curr_appl_na_pct = check_cols_null_pct(curr_appl_data)
curr_appl_data1 = curr_appl_data.loc[:, (curr_appl_na_pct < 50)]
curr_appl_data1.columns

Index(['SK_ID_CURR', 'TARGET', 'NAME_CONTRACT_TYPE', 'CODE_GENDER',
       'FLAG_OWN_CAR', 'FLAG_OWN_REALTY', 'CNT_CHILDREN', 'AMT_INCOME_TOTAL',
       'AMT_CREDIT', 'AMT_ANNUITY', 'AMT_GOODS_PRICE', 'NAME_TYPE_SUITE',
       'NAME_INCOME_TYPE', 'NAME_EDUCATION_TYPE', 'NAME_FAMILY_STATUS',
       'NAME_HOUSING_TYPE', 'REGION_POPULATION_RELATIVE', 'DAYS_BIRTH',
       'DAYS_EMPLOYED', 'DAYS_REGISTRATION', 'DAYS_ID_PUBLISH', 'FLAG_MOBIL',
       'FLAG_EMP_PHONE', 'FLAG_WORK_PHONE', 'FLAG_CONT_MOBILE', 'FLAG_PHONE',
       'FLAG_EMAIL', 'OCCUPATION_TYPE', 'CNT_FAM_MEMBERS',
       'REGION_RATING_CLIENT', 'REGION_RATING_CLIENT_W_CITY',
       'WEEKDAY_APPR_PROCESS_START', 'HOUR_APPR_PROCESS_START',
       'REG_REGION_NOT_LIVE_REGION', 'REG_REGION_NOT_WORK_REGION',
       'LIVE_REGION_NOT_WORK_REGION', 'REG_CITY_NOT_LIVE_CITY',
       'REG_CITY_NOT_WORK_CITY', 'LIVE_CITY_NOT_WORK_CITY',
       'ORGANIZATION_TYPE', 'EXT_SOURCE_2', 'EXT_SOURCE_3',
       'YEARS_BEGINEXPLUATATION_AVG', 'FLOORSMAX_AVG',
       'YEARS_BEGINEXPLUATATION_MODE', 'FLOORSMAX_MODE',
       'YEARS_BEGINEXPLUATATION_MEDI', 'FLOORSMAX_MEDI', 'TOTALAREA_MODE',
       'EMERGENCYSTATE_MODE', 'OBS_30_CNT_SOCIAL_CIRCLE',
       'DEF_30_CNT_SOCIAL_CIRCLE', 'OBS_60_CNT_SOCIAL_CIRCLE',
       'DEF_60_CNT_SOCIAL_CIRCLE', 'DAYS_LAST_PHONE_CHANGE', 'FLAG_DOCUMENT_2',
       'FLAG_DOCUMENT_3', 'FLAG_DOCUMENT_4', 'FLAG_DOCUMENT_5',
       'FLAG_DOCUMENT_6', 'FLAG_DOCUMENT_7', 'FLAG_DOCUMENT_8',
       'FLAG_DOCUMENT_9', 'FLAG_DOCUMENT_10', 'FLAG_DOCUMENT_11',
       'FLAG_DOCUMENT_12', 'FLAG_DOCUMENT_13', 'FLAG_DOCUMENT_14',
       'FLAG_DOCUMENT_15', 'FLAG_DOCUMENT_16', 'FLAG_DOCUMENT_17',
       'FLAG_DOCUMENT_18', 'FLAG_DOCUMENT_19', 'FLAG_DOCUMENT_20',
       'FLAG_DOCUMENT_21', 'AMT_REQ_CREDIT_BUREAU_HOUR',
       'AMT_REQ_CREDIT_BUREAU_DAY', 'AMT_REQ_CREDIT_BUREAU_WEEK',
       'AMT_REQ_CREDIT_BUREAU_MON', 'AMT_REQ_CREDIT_BUREAU_QRT',
       'AMT_REQ_CREDIT_BUREAU_YEAR'],
      dtype='object')

• Since there are many columns that have null values and are more than 50%it is generally not recommended to impute it,
• Therefore we drop all columns greater than 50%

#helper function to print all statistical information for given columns in df
show_stats(curr_appl_data1, curr_appl_data1.columns) 

Total Nulls: 0,
Mode: 100002
Median   : 278202.0, 
Variance: 10565820148.159698, 

Describe: count   307511.000
mean    278180.519
std     102790.175
min     100002.000
25%     189145.500
50%     278202.000
75%     367142.500
max     456255.000
Name: SK_ID_CURR, dtype: float64 

ValueCounts: SK_ID_CURR
100002   0.000
337664   0.000
337661   0.000
337660   0.000
337659   0.000
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [1 0]

Median   : 0.0, 
Variance: 0.07421191849651897, 

Describe: count   307511.000
mean         0.081
std          0.272
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: TARGET, dtype: float64 

ValueCounts: TARGET
0   91.927
1    8.073
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Cash loans

Unique: ['Cash loans' 'Revolving loans']

ValueCounts: NAME_CONTRACT_TYPE
Cash loans        90.479
Revolving loans    9.521
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: F

Unique: ['M' 'F' 'XNA']

ValueCounts: CODE_GENDER
F     65.834
M     34.164
XNA    0.001
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: N

Unique: ['N' 'Y']

ValueCounts: FLAG_OWN_CAR
N   65.989
Y   34.011
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Y

Unique: ['Y' 'N']

ValueCounts: FLAG_OWN_REALTY
Y   69.367
N   30.633
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [ 0  1  2  3  4  7  5  6  8  9 11 12 10 19 14]

Median   : 0.0, 
Variance: 0.5214592938640343, 

Describe: count   307511.000
mean         0.417
std          0.722
min          0.000
25%          0.000
50%          0.000
75%          1.000
max         19.000
Name: CNT_CHILDREN, dtype: float64 

ValueCounts: CNT_CHILDREN
0   70.037
1   19.875
2    8.699
3    1.209
4    0.140
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 135000.0
Median   : 147150.0, 
Variance: 56227386501.174484, 

Describe: count      307511.000
mean       168797.919
std        237123.146
min         25650.000
25%        112500.000
50%        147150.000
75%        202500.000
max     117000000.000
Name: AMT_INCOME_TOTAL, dtype: float64 

ValueCounts: AMT_INCOME_TOTAL
135000.000   11.626
112500.000   10.087
157500.000    8.636
180000.000    8.038
90000.000     7.311
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 450000.0
Median   : 513531.0, 
Variance: 161998825566.80032, 

Describe: count    307511.000
mean     599026.000
std      402490.777
min       45000.000
25%      270000.000
50%      513531.000
75%      808650.000
max     4050000.000
Name: AMT_CREDIT, dtype: float64 

ValueCounts: AMT_CREDIT
450000.000   3.157
675000.000   2.887
225000.000   2.654
180000.000   2.388
270000.000   2.355
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 12,
Mode: 9000.0
Median   : 24903.0, 
Variance: 210068421.35962632, 

Describe: count   307499.000
mean     27108.574
std      14493.737
min       1615.500
25%      16524.000
50%      24903.000
75%      34596.000
max     258025.500
Name: AMT_ANNUITY, dtype: float64 

ValueCounts: AMT_ANNUITY
9000.000    2.076
13500.000   1.793
6750.000    0.741
10125.000   0.662
37800.000   0.521
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 278,
Mode: 450000.0
Median   : 450000.0, 
Variance: 136490687205.5433, 

Describe: count    307233.000
mean     538396.207
std      369446.461
min       40500.000
25%      238500.000
50%      450000.000
75%      679500.000
max     4050000.000
Name: AMT_GOODS_PRICE, dtype: float64 

ValueCounts: AMT_GOODS_PRICE
450000.000   8.470
225000.000   8.229
675000.000   8.125
900000.000   5.018
270000.000   3.720
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 1292,
Mode: Unaccompanied

Unique: ['Unaccompanied' 'Family' 'Spouse, partner' 'Children' 'Other_A' nan
 'Other_B' 'Group of people']

ValueCounts: NAME_TYPE_SUITE
Unaccompanied     81.160
Family            13.111
Spouse, partner    3.713
Children           1.067
Other_B            0.578
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Working

Unique: ['Working' 'State servant' 'Commercial associate' 'Pensioner' 'Unemployed'
 'Student' 'Businessman' 'Maternity leave']

ValueCounts: NAME_INCOME_TYPE
Working                51.632
Commercial associate   23.289
Pensioner              18.003
State servant           7.058
Unemployed              0.007
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Secondary / secondary special

Unique: ['Secondary / secondary special' 'Higher education' 'Incomplete higher'
 'Lower secondary' 'Academic degree']

ValueCounts: NAME_EDUCATION_TYPE
Secondary / secondary special   71.019
Higher education                24.345
Incomplete higher                3.342
Lower secondary                  1.241
Academic degree                  0.053
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Married

Unique: ['Single / not married' 'Married' 'Civil marriage' 'Widow' 'Separated'
 'Unknown']

ValueCounts: NAME_FAMILY_STATUS
Married                63.878
Single / not married   14.778
Civil marriage          9.683
Separated               6.429
Widow                   5.232
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: House / apartment

Unique: ['House / apartment' 'Rented apartment' 'With parents'
 'Municipal apartment' 'Office apartment' 'Co-op apartment']

ValueCounts: NAME_HOUSING_TYPE
House / apartment     88.734
With parents           4.826
Municipal apartment    3.637
Rented apartment       1.587
Office apartment       0.851
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.035792
Median   : 0.01885, 
Variance: 0.00019130430983259252, 

Describe: count   307511.000
mean         0.021
std          0.014
min          0.000
25%          0.010
50%          0.019
75%          0.029
max          0.073
Name: REGION_POPULATION_RELATIVE, dtype: float64 

ValueCounts: REGION_POPULATION_RELATIVE
0.036   5.336
0.046   4.371
0.031   3.955
0.025   3.886
0.026   3.773
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: -13749
Median   : -15750.0, 
Variance: 19044396.778353307, 

Describe: count   307511.000
mean    -16036.995
std       4363.989
min     -25229.000
25%     -19682.000
50%     -15750.000
75%     -12413.000
max      -7489.000
Name: DAYS_BIRTH, dtype: float64 

ValueCounts: DAYS_BIRTH
-13749   0.014
-13481   0.014
-18248   0.013
-10020   0.013
-15771   0.013
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 365243
Median   : -1213.0, 
Variance: 19958842205.435524, 

Describe: count   307511.000
mean     63815.046
std     141275.767
min     -17912.000
25%      -2760.000
50%      -1213.000
75%       -289.000
max     365243.000
Name: DAYS_EMPLOYED, dtype: float64 

ValueCounts: DAYS_EMPLOYED
 365243   18.007
-200       0.051
-224       0.049
-230       0.049
-199       0.049
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: -1.0
Median   : -4504.0, 
Variance: 12410728.030428572, 

Describe: count   307511.000
mean     -4986.120
std       3522.886
min     -24672.000
25%      -7479.500
50%      -4504.000
75%      -2010.000
max          0.000
Name: DAYS_REGISTRATION, dtype: float64 

ValueCounts: DAYS_REGISTRATION
-1.000   0.037
-7.000   0.032
-6.000   0.031
-4.000   0.030
-2.000   0.030
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: -4053
Median   : -3254.0, 
Variance: 2278440.5674284254, 

Describe: count   307511.000
mean     -2994.202
std       1509.450
min      -7197.000
25%      -4299.000
50%      -3254.000
75%      -1720.000
max          0.000
Name: DAYS_ID_PUBLISH, dtype: float64 

ValueCounts: DAYS_ID_PUBLISH
-4053   0.055
-4095   0.053
-4046   0.052
-4417   0.052
-4256   0.051
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 1

Unique: [1 0]

Median   : 1.0, 
Variance: 3.251916191640975e-06, 

Describe: count   307511.000
mean         1.000
std          0.002
min          0.000
25%          1.000
50%          1.000
75%          1.000
max          1.000
Name: FLAG_MOBIL, dtype: float64 

ValueCounts: FLAG_MOBIL
1   100.000
0     0.000
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 1

Unique: [1 0]

Median   : 1.0, 
Variance: 0.147671271296676, 

Describe: count   307511.000
mean         0.820
std          0.384
min          0.000
25%          1.000
50%          1.000
75%          1.000
max          1.000
Name: FLAG_EMP_PHONE, dtype: float64 

ValueCounts: FLAG_EMP_PHONE
1   81.989
0   18.011
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.15962120697935117, 

Describe: count   307511.000
mean         0.199
std          0.400
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_WORK_PHONE, dtype: float64 

ValueCounts: FLAG_WORK_PHONE
0   80.063
1   19.937
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 1

Unique: [1 0]

Median   : 1.0, 
Variance: 0.0018631217575536686, 

Describe: count   307511.000
mean         0.998
std          0.043
min          0.000
25%          1.000
50%          1.000
75%          1.000
max          1.000
Name: FLAG_CONT_MOBILE, dtype: float64 

ValueCounts: FLAG_CONT_MOBILE
1   99.813
0    0.187
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [1 0]

Median   : 0.0, 
Variance: 0.20206872204588347, 

Describe: count   307511.000
mean         0.281
std          0.450
min          0.000
25%          0.000
50%          0.000
75%          1.000
max          1.000
Name: FLAG_PHONE, dtype: float64 

ValueCounts: FLAG_PHONE
0   71.893
1   28.107
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.0535029466256677, 

Describe: count   307511.000
mean         0.057
std          0.231
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_EMAIL, dtype: float64 

ValueCounts: FLAG_EMAIL
0   94.328
1    5.672
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 96391,
Mode: Laborers

Unique: ['Laborers' 'Core staff' 'Accountants' 'Managers' nan 'Drivers'
 'Sales staff' 'Cleaning staff' 'Cooking staff' 'Private service staff'
 'Medicine staff' 'Security staff' 'High skill tech staff'
 'Waiters/barmen staff' 'Low-skill Laborers' 'Realty agents' 'Secretaries'
 'IT staff' 'HR staff']

ValueCounts: OCCUPATION_TYPE
Laborers      26.140
Sales staff   15.206
Core staff    13.059
Managers      10.123
Drivers        8.812
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 2,
Mode: 2.0

Unique: [ 1.  2.  3.  4.  5.  6.  9.  7.  8. 10. 13. nan 14. 12. 20. 15. 16. 11.]

Median   : 2.0, 
Variance: 0.8293409204392903, 

Describe: count   307509.000
mean         2.153
std          0.911
min          1.000
25%          2.000
50%          2.000
75%          3.000
max         20.000
Name: CNT_FAM_MEMBERS, dtype: float64 

ValueCounts: CNT_FAM_MEMBERS
2.000   51.497
1.000   22.063
3.000   17.106
4.000    8.031
5.000    1.131
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 2

Unique: [2 1 3]

Median   : 2.0, 
Variance: 0.2591155142160186, 

Describe: count   307511.000
mean         2.052
std          0.509
min          1.000
25%          2.000
50%          2.000
75%          2.000
max          3.000
Name: REGION_RATING_CLIENT, dtype: float64 

ValueCounts: REGION_RATING_CLIENT
2   73.813
3   15.717
1   10.470
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 2

Unique: [2 1 3]

Median   : 2.0, 
Variance: 0.2527445242643602, 

Describe: count   307511.000
mean         2.032
std          0.503
min          1.000
25%          2.000
50%          2.000
75%          2.000
max          3.000
Name: REGION_RATING_CLIENT_W_CITY, dtype: float64 

ValueCounts: REGION_RATING_CLIENT_W_CITY
2   74.626
3   14.263
1   11.111
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: TUESDAY

Unique: ['WEDNESDAY' 'MONDAY' 'THURSDAY' 'SUNDAY' 'SATURDAY' 'FRIDAY' 'TUESDAY']

ValueCounts: WEEKDAY_APPR_PROCESS_START
TUESDAY     17.528
WEDNESDAY   16.889
MONDAY      16.492
THURSDAY    16.452
FRIDAY      16.369
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 10

Unique: [10 11  9 17 16 14  8 15  7 13  6 12 19  3 18 21  4  5 20 22  1  2 23  0]

Median   : 12.0, 
Variance: 10.665660320665818, 

Describe: count   307511.000
mean        12.063
std          3.266
min          0.000
25%         10.000
50%         12.000
75%         14.000
max         23.000
Name: HOUR_APPR_PROCESS_START, dtype: float64 

ValueCounts: HOUR_APPR_PROCESS_START
10   12.267
11   12.107
12   11.132
13   10.068
14    9.002
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.014914876209068216, 

Describe: count   307511.000
mean         0.015
std          0.122
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: REG_REGION_NOT_LIVE_REGION, dtype: float64 

ValueCounts: REG_REGION_NOT_LIVE_REGION
0   98.486
1    1.514
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.04819158950907743, 

Describe: count   307511.000
mean         0.051
std          0.220
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: REG_REGION_NOT_WORK_REGION, dtype: float64 

ValueCounts: REG_REGION_NOT_WORK_REGION
0   94.923
1    5.077
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.039005704439295116, 

Describe: count   307511.000
mean         0.041
std          0.197
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: LIVE_REGION_NOT_WORK_REGION, dtype: float64 

ValueCounts: LIVE_REGION_NOT_WORK_REGION
0   95.934
1    4.066
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.07206205892600395, 

Describe: count   307511.000
mean         0.078
std          0.268
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: REG_CITY_NOT_LIVE_CITY, dtype: float64 

ValueCounts: REG_CITY_NOT_LIVE_CITY
0   92.183
1    7.817
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.17734528517578357, 

Describe: count   307511.000
mean         0.230
std          0.421
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: REG_CITY_NOT_WORK_CITY, dtype: float64 

ValueCounts: REG_CITY_NOT_WORK_CITY
0   76.955
1   23.045
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.14731519424670028, 

Describe: count   307511.000
mean         0.180
std          0.384
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: LIVE_CITY_NOT_WORK_CITY, dtype: float64 

ValueCounts: LIVE_CITY_NOT_WORK_CITY
0   82.045
1   17.955
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Business Entity Type 3
ValueCounts: ORGANIZATION_TYPE
Business Entity Type 3   22.110
XNA                      18.007
Self-employed            12.491
Other                     5.425
Medicine                  3.640
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 660,
Mode: 0.2858978721410488
Median   : 0.5659614260608526, 
Variance: 0.0365039828228688, 

Describe: count   306851.000
mean         0.514
std          0.191
min          0.000
25%          0.392
50%          0.566
75%          0.664
max          0.855
Name: EXT_SOURCE_2, dtype: float64 

ValueCounts: EXT_SOURCE_2
0.286   0.235
0.262   0.136
0.265   0.112
0.160   0.105
0.265   0.100
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 60965,
Mode: 0.746300213050371
Median   : 0.5352762504724826, 
Variance: 0.03796432636328682, 

Describe: count   246546.000
mean         0.511
std          0.195
min          0.001
25%          0.371
50%          0.535
75%          0.669
max          0.896
Name: EXT_SOURCE_3, dtype: float64 

ValueCounts: EXT_SOURCE_3
0.746   0.592
0.714   0.533
0.694   0.518
0.671   0.483
0.653   0.468
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 150007,
Mode: 0.9871
Median   : 0.9816, 
Variance: 0.0035074009635889924, 

Describe: count   157504.000
mean         0.978
std          0.059
min          0.000
25%          0.977
50%          0.982
75%          0.987
max          1.000
Name: YEARS_BEGINEXPLUATATION_AVG, dtype: float64 

ValueCounts: YEARS_BEGINEXPLUATATION_AVG
0.987   2.737
0.986   2.660
0.986   2.648
0.980   2.618
0.987   2.612
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 153020,
Mode: 0.1667
Median   : 0.1667, 
Variance: 0.020920931965718682, 

Describe: count   154491.000
mean         0.226
std          0.145
min          0.000
25%          0.167
50%          0.167
75%          0.333
max          1.000
Name: FLOORSMAX_AVG, dtype: float64 

ValueCounts: FLOORSMAX_AVG
0.167   40.051
0.333   20.654
0.042    9.450
0.375    5.130
0.125    4.514
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 150007,
Mode: 0.9871
Median   : 0.9816, 
Variance: 0.004169987074137329, 

Describe: count   157504.000
mean         0.977
std          0.065
min          0.000
25%          0.977
50%          0.982
75%          0.987
max          1.000
Name: YEARS_BEGINEXPLUATATION_MODE, dtype: float64 

ValueCounts: YEARS_BEGINEXPLUATATION_MODE
0.987   2.724
0.987   2.649
0.986   2.646
0.980   2.609
0.981   2.592
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 153020,
Mode: 0.1667

Unique: [0.0833 0.2917    nan 0.1667 0.3333 0.6667 0.375  0.0417 0.25   0.4583
 0.2083 0.125  0.     0.5833 0.625  0.9167 0.9583 0.5417 1.     0.4167
 0.875  0.7083 0.75   0.5    0.7917 0.8333]

Median   : 0.1667, 
Variance: 0.020652393543979256, 

Describe: count   154491.000
mean         0.222
std          0.144
min          0.000
25%          0.167
50%          0.167
75%          0.333
max          1.000
Name: FLOORSMAX_MODE, dtype: float64 

ValueCounts: FLOORSMAX_MODE
0.167   42.430
0.333   22.249
0.042   10.108
0.375    5.386
0.125    4.704
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 150007,
Mode: 0.9871
Median   : 0.9816, 
Variance: 0.003587688764103631, 

Describe: count   157504.000
mean         0.978
std          0.060
min          0.000
25%          0.977
50%          0.982
75%          0.987
max          1.000
Name: YEARS_BEGINEXPLUATATION_MEDI, dtype: float64 

ValueCounts: YEARS_BEGINEXPLUATATION_MEDI
0.987   2.739
0.986   2.696
0.986   2.666
0.987   2.627
0.980   2.613
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 153020,
Mode: 0.1667
Median   : 0.1667, 
Variance: 0.02104444200909498, 

Describe: count   154491.000
mean         0.226
std          0.145
min          0.000
25%          0.167
50%          0.167
75%          0.333
max          1.000
Name: FLOORSMAX_MEDI, dtype: float64 

ValueCounts: FLOORSMAX_MEDI
0.167   41.172
0.333   21.541
0.042    9.601
0.375    5.312
0.125    4.582
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 148431,
Mode: 0.0
Median   : 0.0688, 
Variance: 0.01154815111163752, 

Describe: count   159080.000
mean         0.103
std          0.107
min          0.000
25%          0.041
50%          0.069
75%          0.128
max          1.000
Name: TOTALAREA_MODE, dtype: float64 

ValueCounts: TOTALAREA_MODE
0.000   0.366
0.057   0.155
0.055   0.145
0.055   0.143
0.056   0.143
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 145755,
Mode: No

Unique: ['No' nan 'Yes']

ValueCounts: EMERGENCYSTATE_MODE
No    98.561
Yes    1.439
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 1021,
Mode: 0.0

Unique: [  2.   1.   0.   4.   8.  10.  nan   7.   3.   6.   5.  12.   9.  13.
  11.  14.  22.  16.  15.  17.  20.  25.  19.  18.  21.  24.  23.  28.
  26.  29.  27.  47. 348.  30.]

Median   : 0.0, 
Variance: 5.764746958955638, 

Describe: count   306490.000
mean         1.422
std          2.401
min          0.000
25%          0.000
50%          0.000
75%          2.000
max        348.000
Name: OBS_30_CNT_SOCIAL_CIRCLE, dtype: float64 

ValueCounts: OBS_30_CNT_SOCIAL_CIRCLE
0.000   53.480
1.000   15.917
2.000    9.726
3.000    6.631
4.000    4.615
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 1021,
Mode: 0.0

Unique: [ 2.  0.  1. nan  3.  4.  5.  6.  7. 34.  8.]

Median   : 0.0, 
Variance: 0.19953948681282563, 

Describe: count   306490.000
mean         0.143
std          0.447
min          0.000
25%          0.000
50%          0.000
75%          0.000
max         34.000
Name: DEF_30_CNT_SOCIAL_CIRCLE, dtype: float64 

ValueCounts: DEF_30_CNT_SOCIAL_CIRCLE
0.000   88.526
1.000    9.243
2.000    1.737
3.000    0.389
4.000    0.083
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 1021,
Mode: 0.0

Unique: [  2.   1.   0.   4.   8.  10.  nan   7.   3.   6.   5.  12.   9.  13.
  11.  14.  21.  15.  22.  16.  20.  25.  17.  19.  18.  24.  23.  28.
  29.  27.  47. 344.  30.  26.]

Median   : 0.0, 
Variance: 5.663463994080467, 

Describe: count   306490.000
mean         1.405
std          2.380
min          0.000
25%          0.000
50%          0.000
75%          2.000
max        344.000
Name: OBS_60_CNT_SOCIAL_CIRCLE, dtype: float64 

ValueCounts: OBS_60_CNT_SOCIAL_CIRCLE
0.000   53.726
1.000   15.945
2.000    9.712
3.000    6.596
4.000    4.550
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 1021,
Mode: 0.0

Unique: [ 2.  0.  1. nan  3.  5.  4.  7. 24.  6.]

Median   : 0.0, 
Variance: 0.131254626645024, 

Describe: count   306490.000
mean         0.100
std          0.362
min          0.000
25%          0.000
50%          0.000
75%          0.000
max         24.000
Name: DEF_60_CNT_SOCIAL_CIRCLE, dtype: float64 

ValueCounts: DEF_60_CNT_SOCIAL_CIRCLE
0.000   91.592
1.000    7.126
2.000    1.034
3.000    0.195
4.000    0.044
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 1,
Mode: 0.0
Median   : -757.0, 
Variance: 683612.2742421023, 

Describe: count   307510.000
mean      -962.859
std        826.808
min      -4292.000
25%      -1570.000
50%       -757.000
75%       -274.000
max          0.000
Name: DAYS_LAST_PHONE_CHANGE, dtype: float64 

ValueCounts: DAYS_LAST_PHONE_CHANGE
0.000    12.251
-1.000    0.914
-2.000    0.754
-3.000    0.573
-4.000    0.418
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 4.227326079183737e-05, 

Describe: count   307511.000
mean         0.000
std          0.007
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_2, dtype: float64 

ValueCounts: FLAG_DOCUMENT_2
0   99.996
1    0.004
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 1

Unique: [1 0]

Median   : 1.0, 
Variance: 0.20589084885614972, 

Describe: count   307511.000
mean         0.710
std          0.454
min          0.000
25%          0.000
50%          1.000
75%          1.000
max          1.000
Name: FLAG_DOCUMENT_3, dtype: float64 

ValueCounts: FLAG_DOCUMENT_3
1   71.002
0   28.998
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 8.129155979353892e-05, 

Describe: count   307511.000
mean         0.000
std          0.009
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_4, dtype: float64 

ValueCounts: FLAG_DOCUMENT_4
0   99.992
1    0.008
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.014886494471071949, 

Describe: count   307511.000
mean         0.015
std          0.122
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_5, dtype: float64 

ValueCounts: FLAG_DOCUMENT_5
0   98.489
1    1.511
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.08030189665614886, 

Describe: count   307511.000
mean         0.088
std          0.283
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_6, dtype: float64 

ValueCounts: FLAG_DOCUMENT_6
0   91.194
1    8.806
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.00019182686767647077, 

Describe: count   307511.000
mean         0.000
std          0.014
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_7, dtype: float64 

ValueCounts: FLAG_DOCUMENT_7
0   99.981
1    0.019
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.07475414850818889, 

Describe: count   307511.000
mean         0.081
std          0.273
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_8, dtype: float64 

ValueCounts: FLAG_DOCUMENT_8
0   91.862
1    8.138
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.0038806309936899315, 

Describe: count   307511.000
mean         0.004
std          0.062
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_9, dtype: float64 

ValueCounts: FLAG_DOCUMENT_9
0   99.610
1    0.390
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 2.276296919164723e-05, 

Describe: count   307511.000
mean         0.000
std          0.005
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_10, dtype: float64 

ValueCounts: FLAG_DOCUMENT_10
0   99.998
1    0.002
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.0038967636747465713, 

Describe: count   307511.000
mean         0.004
std          0.062
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_11, dtype: float64 

ValueCounts: FLAG_DOCUMENT_11
0   99.609
1    0.391
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 6.5038112332850705e-06, 

Describe: count   307511.000
mean         0.000
std          0.003
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_12, dtype: float64 

ValueCounts: FLAG_DOCUMENT_12
0   99.999
1    0.001
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.003512662405688219, 

Describe: count   307511.000
mean         0.004
std          0.059
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_13, dtype: float64 

ValueCounts: FLAG_DOCUMENT_13
0   99.647
1    0.353
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.0029278669255157, 

Describe: count   307511.000
mean         0.003
std          0.054
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_14, dtype: float64 

ValueCounts: FLAG_DOCUMENT_14
0   99.706
1    0.294
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.0012082533473030956, 

Describe: count   307511.000
mean         0.001
std          0.035
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_15, dtype: float64 

ValueCounts: FLAG_DOCUMENT_15
0   99.879
1    0.121
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.009829564925667255, 

Describe: count   307511.000
mean         0.010
std          0.099
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_16, dtype: float64 

ValueCounts: FLAG_DOCUMENT_16
0   99.007
1    0.993
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.00026658688860710627, 

Describe: count   307511.000
mean         0.000
std          0.016
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_17, dtype: float64 

ValueCounts: FLAG_DOCUMENT_17
0   99.973
1    0.027
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.00806372320835816, 

Describe: count   307511.000
mean         0.008
std          0.090
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_18, dtype: float64 

ValueCounts: FLAG_DOCUMENT_18
0   99.187
1    0.813
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.0005947484523376042, 

Describe: count   307511.000
mean         0.001
std          0.024
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_19, dtype: float64 

ValueCounts: FLAG_DOCUMENT_19
0   99.940
1    0.060
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.0005070432225544667, 

Describe: count   307511.000
mean         0.001
std          0.023
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_20, dtype: float64 

ValueCounts: FLAG_DOCUMENT_20
0   99.949
1    0.051
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.00033483626685604196, 

Describe: count   307511.000
mean         0.000
std          0.018
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_21, dtype: float64 

ValueCounts: FLAG_DOCUMENT_21
0   99.967
1    0.033
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 41519,
Mode: 0.0

Unique: [ 0. nan  1.  2.  3.  4.]

Median   : 0.0, 
Variance: 0.007030676341450841, 

Describe: count   265992.000
mean         0.006
std          0.084
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          4.000
Name: AMT_REQ_CREDIT_BUREAU_HOUR, dtype: float64 

ValueCounts: AMT_REQ_CREDIT_BUREAU_HOUR
0.000   99.389
1.000    0.586
2.000    0.021
3.000    0.003
4.000    0.000
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 41519,
Mode: 0.0

Unique: [ 0. nan  1.  3.  2.  4.  5.  6.  9.  8.]

Median   : 0.0, 
Variance: 0.012267203055661324, 

Describe: count   265992.000
mean         0.007
std          0.111
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          9.000
Name: AMT_REQ_CREDIT_BUREAU_DAY, dtype: float64 

ValueCounts: AMT_REQ_CREDIT_BUREAU_DAY
0.000   99.440
1.000    0.486
2.000    0.040
3.000    0.017
4.000    0.010
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 41519,
Mode: 0.0

Unique: [ 0. nan  1.  3.  2.  4.  5.  6.  8.  7.]

Median   : 0.0, 
Variance: 0.04189589838652147, 

Describe: count   265992.000
mean         0.034
std          0.205
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          8.000
Name: AMT_REQ_CREDIT_BUREAU_WEEK, dtype: float64 

ValueCounts: AMT_REQ_CREDIT_BUREAU_WEEK
0.000   96.791
1.000    3.086
2.000    0.075
3.000    0.022
4.000    0.013
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 41519,
Mode: 0.0

Unique: [ 0. nan  1.  2.  6.  5.  3.  7.  9.  4. 11.  8. 16. 12. 14. 10. 13. 17.
 24. 19. 15. 23. 18. 27. 22.]

Median   : 0.0, 
Variance: 0.8390603897599506, 

Describe: count   265992.000
mean         0.267
std          0.916
min          0.000
25%          0.000
50%          0.000
75%          0.000
max         27.000
Name: AMT_REQ_CREDIT_BUREAU_MON, dtype: float64 

ValueCounts: AMT_REQ_CREDIT_BUREAU_MON
0.000   83.549
1.000   12.462
2.000    2.025
3.000    0.749
4.000    0.405
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 41519,
Mode: 0.0

Unique: [  0.  nan   1.   2.   4.   3.   8.   5.   6.   7. 261.  19.]

Median   : 0.0, 
Variance: 0.6305243726283989, 

Describe: count   265992.000
mean         0.265
std          0.794
min          0.000
25%          0.000
50%          0.000
75%          0.000
max        261.000
Name: AMT_REQ_CREDIT_BUREAU_QRT, dtype: float64 

ValueCounts: AMT_REQ_CREDIT_BUREAU_QRT
0.000   80.986
1.000   12.730
2.000    5.418
3.000    0.646
4.000    0.179
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 41519,
Mode: 0.0

Unique: [ 1.  0. nan  2.  4.  5.  3.  8.  6.  9.  7. 10. 11. 13. 16. 12. 25. 23.
 15. 14. 22. 17. 19. 18. 21. 20.]

Median   : 1.0, 
Variance: 3.4942637902290734, 

Describe: count   265992.000
mean         1.900
std          1.869
min          0.000
25%          0.000
50%          1.000
75%          3.000
max         25.000
Name: AMT_REQ_CREDIT_BUREAU_YEAR, dtype: float64 

ValueCounts: AMT_REQ_CREDIT_BUREAU_YEAR
0.000   26.994
1.000   23.837
2.000   18.870
3.000   12.642
4.000    7.787
Name: proportion, dtype: float64 



------------------------------------------------------------------

DAYS_COLS = curr_appl_data1.filter(like="DAYS").columns  # select all cols that contain DAYS
FLAG_COLS = curr_appl_data1.filter(like="FLAG").columns  # select all cols that contain FLAG

FLOORSMAX_COLS = curr_appl_data1.columns[curr_appl_data1.columns.str.contains("FLOORSMAX")]                 # select all cols that contains FLOORSMAX
BEGINEXPLUATATION_COLS = curr_appl_data1.columns[curr_appl_data1.columns.str.contains("BEGINEXPLUATATION")] # select all cols that contains BEGINEXPLUATATION
AMT_REQ_CREDIT_BUREAU_COLS = curr_appl_data1.columns[curr_appl_data1.columns.str.contains("AMT_REQ_CREDIT_BUREAU")] # select all cols that contains AMT_REQ_CREDIT_BUREAU

# curr_appl_data1["FLOORSMAX"] = curr_appl_data1[FLOORSMAX_COLS].sum(axis=1, skipna=False) / len(FLOORSMAX_COLS)
# curr_appl_data1["YEARS_BEGINEXPLOITATION"] = curr_appl_data1.loc[:, BEGINEXPLUATATION_COLS].sum(axis=1, skipna=False) / len(BEGINEXPLUATATION_COLS)
curr_appl_data1["FLOORSMAX"] = curr_appl_data1[FLOORSMAX_COLS].mean(axis=1, skipna=False)
curr_appl_data1["YEARS_BEGINEXPLOITATION"] = curr_appl_data1.loc[:, BEGINEXPLUATATION_COLS].mean(axis=1, skipna=False)

• The above features are generally considred additional features ,
• In order to minimize the no of cols, clubbed similar cols & created single feature col (considered using mean among Min/Max/Mean/Median, interative approach)

_ = [curr_appl_data1[col].fillna(curr_appl_data1[col].mode()[0], inplace=True) for col in AMT_REQ_CREDIT_BUREAU_COLS]
curr_appl_data1["FLOORSMAX"] = curr_appl_data1["FLOORSMAX"].fillna( curr_appl_data1["FLOORSMAX"].median())
curr_appl_data1["EXT_SOURCE_2"] = curr_appl_data1["EXT_SOURCE_2"].fillna( curr_appl_data1["EXT_SOURCE_2"].median())
curr_appl_data1["EXT_SOURCE_3"] = curr_appl_data1["EXT_SOURCE_3"].fillna( curr_appl_data1["EXT_SOURCE_3"].median())
curr_appl_data1["TOTALAREA_MODE"] = curr_appl_data1["TOTALAREA_MODE"].fillna( curr_appl_data1["TOTALAREA_MODE"].mean())
curr_appl_data1["NAME_TYPE_SUITE"] = curr_appl_data1["NAME_TYPE_SUITE"].fillna( curr_appl_data1["NAME_TYPE_SUITE"].mode()[0])
curr_appl_data1["CNT_FAM_MEMBERS"] = curr_appl_data1["CNT_FAM_MEMBERS"].fillna( curr_appl_data1["CNT_FAM_MEMBERS"].mode()[0])
curr_appl_data1["EMERGENCYSTATE_MODE"] = curr_appl_data1["EMERGENCYSTATE_MODE"].fillna( curr_appl_data1["EMERGENCYSTATE_MODE"].mode()[0])
curr_appl_data1["DAYS_LAST_PHONE_CHANGE"] = curr_appl_data1["DAYS_LAST_PHONE_CHANGE"].fillna(curr_appl_data1["DAYS_LAST_PHONE_CHANGE"].median())
curr_appl_data1["YEARS_BEGINEXPLOITATION"] = curr_appl_data1["YEARS_BEGINEXPLOITATION"].fillna(curr_appl_data1["YEARS_BEGINEXPLOITATION"].mean())
curr_appl_data1["OBS_30_CNT_SOCIAL_CIRCLE"] = curr_appl_data1["OBS_30_CNT_SOCIAL_CIRCLE"].fillna(curr_appl_data1["OBS_30_CNT_SOCIAL_CIRCLE"].mode()[0])
curr_appl_data1["OBS_60_CNT_SOCIAL_CIRCLE"] = curr_appl_data1["OBS_60_CNT_SOCIAL_CIRCLE"].fillna(curr_appl_data1["OBS_60_CNT_SOCIAL_CIRCLE"].mode()[0])
curr_appl_data1["DEF_30_CNT_SOCIAL_CIRCLE"] = curr_appl_data1["DEF_30_CNT_SOCIAL_CIRCLE"].fillna(curr_appl_data1["DEF_30_CNT_SOCIAL_CIRCLE"].mode()[0])
curr_appl_data1["DEF_60_CNT_SOCIAL_CIRCLE"] = curr_appl_data1["DEF_60_CNT_SOCIAL_CIRCLE"].fillna(curr_appl_data1["DEF_60_CNT_SOCIAL_CIRCLE"].mode()[0])
# curr_appl_data1 = curr_appl_data1[~curr_appl_data1[["AMT_ANNUITY", "AMT_GOODS_PRICE"]].isna().any(axis=1)]
curr_appl_data1["AMT_ANNUITY"] = curr_appl_data1["AMT_ANNUITY"].fillna( curr_appl_data1["AMT_ANNUITY"].mean())
curr_appl_data1["AMT_GOODS_PRICE"] = curr_appl_data1["AMT_GOODS_PRICE"].fillna( curr_appl_data1["AMT_GOODS_PRICE"].mean())

• since there are many null values among cols, tried to impute it based on stats for the specific columns.
• for categorical/discrete columns, considered Mode if the frequency of the topmost value is above 50%.
• for numerical columns - considered Mean, Median, based on distribution and skewness.
• AMT_ANNUITY, AMT_GOODS_PRICE are considered critical features,
• To impute such features, we need to identify whether it is MAR, MNAR, MCAR. also to analyse dependency cols and distribution set to finetune the imputation.
• for this analysis, since the missing values are negligible (>200) values , we considered it to be of less impact to the outcome of the analysis,
• To keep the dataset as it is, we therefore decided to fill null values with the mean of the column, instead of dropping it.

curr_appl_data1 = curr_appl_data1.drop(FLOORSMAX_COLS, axis=1)
curr_appl_data1 = curr_appl_data1.drop(BEGINEXPLUATATION_COLS, axis=1)

• FLOORSMAX_COLS, BEGINEXPLUATATION_COLS collections are featured engineered and hence reduntant, therefore we dropped those columns

# curr_appl_data1["EMERGENCYSTATE_MODE"] = curr_appl_data1["EMERGENCYSTATE_MODE"].replace(to_replace=["Yes", "No"], value=[1, 0])

curr_appl_data1["FLAG_OWN_CAR"] = curr_appl_data1["FLAG_OWN_CAR"].replace(to_replace=["Y", "N"], value=["Yes", "No"])
curr_appl_data1["FLAG_OWN_REALTY"] = curr_appl_data1["FLAG_OWN_REALTY"].replace(to_replace=["Y", "N"], value=["Yes", "No"])

curr_appl_data1["NAME_TYPE_SUITE"] = curr_appl_data1["NAME_TYPE_SUITE"].replace(to_replace="Spouse, partner", value="Spouse")
curr_appl_data1["NAME_TYPE_SUITE"] = curr_appl_data1["NAME_TYPE_SUITE"].replace(to_replace=["Other_A", "Other_B"], value=["Others", "Others"])
curr_appl_data1["NAME_HOUSING_TYPE"] = curr_appl_data1["NAME_HOUSING_TYPE"].replace(to_replace="House / apartment", value="House")
curr_appl_data1["NAME_FAMILY_STATUS"] = curr_appl_data1["NAME_FAMILY_STATUS"].replace(to_replace="Single / not married", value="Single")
curr_appl_data1["NAME_EDUCATION_TYPE"] = curr_appl_data1["NAME_EDUCATION_TYPE"].replace(to_replace="Secondary / secondary special", value="Secondary")

• Some categorical column values are more detailed, therefore replaced with more apt and common terms

curr_appl_data1[DAYS_COLS] = (abs(curr_appl_data1[DAYS_COLS]) / 365).astype(int)

• DAYS_COLS - Is a list of days type columns, represented in days count,
• There were inconsistencies, therefore made all values absolute, then converted it to years count by dividing it with 365.

days_to_years = {i: i.replace("DAYS", "YEARS") for i in DAYS_COLS}
curr_appl_data1 = curr_appl_data1.rename(columns=days_to_years)
curr_appl_data1.rename(columns={'YEARS_BIRTH': 'AGE_IN_YEARS'})
YEARS_COLS = curr_appl_data1.filter(like="YEARS").columns

	SK_ID_CURR	TARGET	NAME_CONTRACT_TYPE	CODE_GENDER	FLAG_OWN_CAR	FLAG_OWN_REALTY	CNT_CHILDREN	AMT_INCOME_TOTAL	AMT_CREDIT	AMT_ANNUITY	AMT_GOODS_PRICE	NAME_TYPE_SUITE	NAME_INCOME_TYPE	NAME_EDUCATION_TYPE	NAME_FAMILY_STATUS	NAME_HOUSING_TYPE	REGION_POPULATION_RELATIVE	AGE_IN_YEARS	YEARS_EMPLOYED	YEARS_REGISTRATION	YEARS_ID_PUBLISH	FLAG_MOBIL	FLAG_EMP_PHONE	FLAG_WORK_PHONE	FLAG_CONT_MOBILE	FLAG_PHONE	FLAG_EMAIL	OCCUPATION_TYPE	CNT_FAM_MEMBERS	REGION_RATING_CLIENT	REGION_RATING_CLIENT_W_CITY	WEEKDAY_APPR_PROCESS_START	HOUR_APPR_PROCESS_START	REG_REGION_NOT_LIVE_REGION	REG_REGION_NOT_WORK_REGION	LIVE_REGION_NOT_WORK_REGION	REG_CITY_NOT_LIVE_CITY	REG_CITY_NOT_WORK_CITY	LIVE_CITY_NOT_WORK_CITY	ORGANIZATION_TYPE	EXT_SOURCE_2	EXT_SOURCE_3	TOTALAREA_MODE	EMERGENCYSTATE_MODE	OBS_30_CNT_SOCIAL_CIRCLE	DEF_30_CNT_SOCIAL_CIRCLE	OBS_60_CNT_SOCIAL_CIRCLE	DEF_60_CNT_SOCIAL_CIRCLE	YEARS_LAST_PHONE_CHANGE	FLAG_DOCUMENT_2	FLAG_DOCUMENT_3	FLAG_DOCUMENT_4	FLAG_DOCUMENT_5	FLAG_DOCUMENT_6	FLAG_DOCUMENT_7	FLAG_DOCUMENT_8	FLAG_DOCUMENT_9	FLAG_DOCUMENT_10	FLAG_DOCUMENT_11	FLAG_DOCUMENT_12	FLAG_DOCUMENT_13	FLAG_DOCUMENT_14	FLAG_DOCUMENT_15	FLAG_DOCUMENT_16	FLAG_DOCUMENT_17	FLAG_DOCUMENT_18	FLAG_DOCUMENT_19	FLAG_DOCUMENT_20	FLAG_DOCUMENT_21	AMT_REQ_CREDIT_BUREAU_HOUR	AMT_REQ_CREDIT_BUREAU_DAY	AMT_REQ_CREDIT_BUREAU_WEEK	AMT_REQ_CREDIT_BUREAU_MON	AMT_REQ_CREDIT_BUREAU_QRT	AMT_REQ_CREDIT_BUREAU_YEAR	FLOORSMAX	YEARS_BEGINEXPLOITATION
0	100002	1	Cash loans	M	No	Yes	0	202500.000	406597.500	24700.500	351000.000	Unaccompanied	Working	Secondary	Single	House	0.019	25	1	9	5	1	1	0	1	1	0	Laborers	1.000	2	2	WEDNESDAY	10	0	0	0	0	0	0	Business Entity Type 3	0.263	0.139	0.015	No	2.000	2.000	2.000	2.000	3	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	1.000	0.083	0.972
1	100003	0	Cash loans	F	No	No	0	270000.000	1293502.500	35698.500	1129500.000	Family	State servant	Higher education	Married	House	0.004	45	3	3	0	1	1	0	1	1	0	Core staff	2.000	1	1	MONDAY	11	0	0	0	0	0	0	School	0.622	0.535	0.071	No	1.000	0.000	1.000	0.000	2	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	0.000	0.292	0.985
2	100004	0	Revolving loans	M	Yes	Yes	0	67500.000	135000.000	6750.000	135000.000	Unaccompanied	Working	Secondary	Single	House	0.010	52	0	11	6	1	1	1	1	1	0	Laborers	1.000	2	2	MONDAY	9	0	0	0	0	0	0	Government	0.556	0.730	0.103	No	0.000	0.000	0.000	0.000	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	0.000	0.167	0.978
3	100006	0	Cash loans	F	No	Yes	0	135000.000	312682.500	29686.500	297000.000	Unaccompanied	Working	Secondary	Civil marriage	House	0.008	52	8	26	6	1	1	0	1	0	0	Laborers	2.000	2	2	WEDNESDAY	17	0	0	0	0	0	0	Business Entity Type 3	0.650	0.535	0.103	No	2.000	0.000	2.000	0.000	1	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	0.000	0.167	0.978
4	100007	0	Cash loans	M	No	Yes	0	121500.000	513000.000	21865.500	513000.000	Unaccompanied	Working	Secondary	Single	House	0.029	54	8	11	9	1	1	0	1	0	0	Core staff	1.000	2	2	THURSDAY	11	0	0	0	0	1	1	Religion	0.323	0.535	0.103	No	0.000	0.000	0.000	0.000	3	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	0.000	0.167	0.978
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
307506	456251	0	Cash loans	M	No	No	0	157500.000	254700.000	27558.000	225000.000	Unaccompanied	Working	Secondary	Separated	With parents	0.033	25	0	23	5	1	1	0	1	0	0	Sales staff	1.000	1	1	THURSDAY	15	0	0	0	0	0	0	Services	0.682	0.535	0.290	No	0.000	0.000	0.000	0.000	0	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	0.000	0.556	0.984
307507	456252	0	Cash loans	F	No	Yes	0	72000.000	269550.000	12001.500	225000.000	Unaccompanied	Pensioner	Secondary	Widow	House	0.025	56	1000	12	11	1	0	0	1	1	0	NaN	1.000	2	2	MONDAY	8	0	0	0	0	0	0	XNA	0.116	0.535	0.021	No	0.000	0.000	0.000	0.000	0	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	0.000	0.083	0.973
307508	456253	0	Cash loans	F	No	Yes	0	153000.000	677664.000	29979.000	585000.000	Unaccompanied	Working	Higher education	Separated	House	0.005	41	21	18	14	1	1	0	1	0	1	Managers	1.000	3	3	THURSDAY	9	0	0	0	0	1	1	School	0.536	0.219	0.797	No	6.000	0.000	6.000	0.000	5	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1.000	0.000	0.000	1.000	0.000	1.000	0.167	0.982
307509	456254	1	Cash loans	F	No	Yes	0	171000.000	370107.000	20205.000	319500.000	Unaccompanied	Commercial associate	Secondary	Married	House	0.005	32	13	7	2	1	1	0	1	0	0	Laborers	2.000	2	2	WEDNESDAY	9	0	0	0	1	1	0	Business Entity Type 1	0.514	0.661	0.009	No	0.000	0.000	0.000	0.000	0	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	0.000	0.042	0.977
307510	456255	0	Cash loans	F	No	No	0	157500.000	675000.000	49117.500	675000.000	Unaccompanied	Commercial associate	Higher education	Married	House	0.046	46	3	14	1	1	1	1	1	1	0	Laborers	2.000	1	1	THURSDAY	20	0	0	0	0	1	1	Business Entity Type 3	0.709	0.114	0.072	No	0.000	0.000	0.000	0.000	2	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	2.000	0.000	1.000	0.375	0.988

307511 rows × 77 columns

• Renamed DAYS columns to YEARS , since the values are already converted

curr_appl_na_pct1 = check_cols_null_pct(curr_appl_data1)
curr_appl_na_pct1
show_stats(curr_appl_data1, curr_appl_data1.columns)

OCCUPATION_TYPE               31.346
SK_ID_CURR                     0.000
FLAG_DOCUMENT_2                0.000
FLAG_DOCUMENT_8                0.000
FLAG_DOCUMENT_7                0.000
FLAG_DOCUMENT_6                0.000
FLAG_DOCUMENT_5                0.000
FLAG_DOCUMENT_4                0.000
FLAG_DOCUMENT_3                0.000
YEARS_LAST_PHONE_CHANGE        0.000
FLAG_DOCUMENT_10               0.000
DEF_60_CNT_SOCIAL_CIRCLE       0.000
OBS_60_CNT_SOCIAL_CIRCLE       0.000
DEF_30_CNT_SOCIAL_CIRCLE       0.000
OBS_30_CNT_SOCIAL_CIRCLE       0.000
EMERGENCYSTATE_MODE            0.000
TOTALAREA_MODE                 0.000
EXT_SOURCE_3                   0.000
FLAG_DOCUMENT_9                0.000
FLAG_DOCUMENT_11               0.000
ORGANIZATION_TYPE              0.000
FLAG_DOCUMENT_21               0.000
FLOORSMAX                      0.000
AMT_REQ_CREDIT_BUREAU_YEAR     0.000
AMT_REQ_CREDIT_BUREAU_QRT      0.000
AMT_REQ_CREDIT_BUREAU_MON      0.000
AMT_REQ_CREDIT_BUREAU_WEEK     0.000
AMT_REQ_CREDIT_BUREAU_DAY      0.000
AMT_REQ_CREDIT_BUREAU_HOUR     0.000
FLAG_DOCUMENT_20               0.000
FLAG_DOCUMENT_12               0.000
FLAG_DOCUMENT_19               0.000
FLAG_DOCUMENT_18               0.000
FLAG_DOCUMENT_17               0.000
FLAG_DOCUMENT_16               0.000
FLAG_DOCUMENT_15               0.000
FLAG_DOCUMENT_14               0.000
FLAG_DOCUMENT_13               0.000
EXT_SOURCE_2                   0.000
LIVE_CITY_NOT_WORK_CITY        0.000
TARGET                         0.000
YEARS_EMPLOYED                 0.000
REGION_POPULATION_RELATIVE     0.000
NAME_HOUSING_TYPE              0.000
NAME_FAMILY_STATUS             0.000
NAME_EDUCATION_TYPE            0.000
NAME_INCOME_TYPE               0.000
NAME_TYPE_SUITE                0.000
AMT_GOODS_PRICE                0.000
AMT_ANNUITY                    0.000
AMT_CREDIT                     0.000
AMT_INCOME_TOTAL               0.000
CNT_CHILDREN                   0.000
FLAG_OWN_REALTY                0.000
FLAG_OWN_CAR                   0.000
CODE_GENDER                    0.000
NAME_CONTRACT_TYPE             0.000
YEARS_BIRTH                    0.000
YEARS_REGISTRATION             0.000
REG_CITY_NOT_WORK_CITY         0.000
YEARS_ID_PUBLISH               0.000
REG_CITY_NOT_LIVE_CITY         0.000
LIVE_REGION_NOT_WORK_REGION    0.000
REG_REGION_NOT_WORK_REGION     0.000
REG_REGION_NOT_LIVE_REGION     0.000
HOUR_APPR_PROCESS_START        0.000
WEEKDAY_APPR_PROCESS_START     0.000
REGION_RATING_CLIENT_W_CITY    0.000
REGION_RATING_CLIENT           0.000
CNT_FAM_MEMBERS                0.000
FLAG_EMAIL                     0.000
FLAG_PHONE                     0.000
FLAG_CONT_MOBILE               0.000
FLAG_WORK_PHONE                0.000
FLAG_EMP_PHONE                 0.000
FLAG_MOBIL                     0.000
YEARS_BEGINEXPLOITATION        0.000
dtype: float64

Total Nulls: 0,
Mode: 100002
Median   : 278202.0, 
Variance: 10565820148.159698, 

Describe: count   307511.000
mean    278180.519
std     102790.175
min     100002.000
25%     189145.500
50%     278202.000
75%     367142.500
max     456255.000
Name: SK_ID_CURR, dtype: float64 

ValueCounts: SK_ID_CURR
100002   0.000
337664   0.000
337661   0.000
337660   0.000
337659   0.000
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [1 0]

Median   : 0.0, 
Variance: 0.07421191849651897, 

Describe: count   307511.000
mean         0.081
std          0.272
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: TARGET, dtype: float64 

ValueCounts: TARGET
0   91.927
1    8.073
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Cash loans

Unique: ['Cash loans' 'Revolving loans']

ValueCounts: NAME_CONTRACT_TYPE
Cash loans        90.479
Revolving loans    9.521
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: F

Unique: ['M' 'F' 'XNA']

ValueCounts: CODE_GENDER
F     65.834
M     34.164
XNA    0.001
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: No

Unique: ['No' 'Yes']

ValueCounts: FLAG_OWN_CAR
No    65.989
Yes   34.011
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Yes

Unique: ['Yes' 'No']

ValueCounts: FLAG_OWN_REALTY
Yes   69.367
No    30.633
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [ 0  1  2  3  4  7  5  6  8  9 11 12 10 19 14]

Median   : 0.0, 
Variance: 0.5214592938640343, 

Describe: count   307511.000
mean         0.417
std          0.722
min          0.000
25%          0.000
50%          0.000
75%          1.000
max         19.000
Name: CNT_CHILDREN, dtype: float64 

ValueCounts: CNT_CHILDREN
0   70.037
1   19.875
2    8.699
3    1.209
4    0.140
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 135000.0
Median   : 147150.0, 
Variance: 56227386501.174484, 

Describe: count      307511.000
mean       168797.919
std        237123.146
min         25650.000
25%        112500.000
50%        147150.000
75%        202500.000
max     117000000.000
Name: AMT_INCOME_TOTAL, dtype: float64 

ValueCounts: AMT_INCOME_TOTAL
135000.000   11.626
112500.000   10.087
157500.000    8.636
180000.000    8.038
90000.000     7.311
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 450000.0
Median   : 513531.0, 
Variance: 161998825566.80032, 

Describe: count    307511.000
mean     599026.000
std      402490.777
min       45000.000
25%      270000.000
50%      513531.000
75%      808650.000
max     4050000.000
Name: AMT_CREDIT, dtype: float64 

ValueCounts: AMT_CREDIT
450000.000   3.157
675000.000   2.887
225000.000   2.654
180000.000   2.388
270000.000   2.355
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 9000.0
Median   : 24903.0, 
Variance: 210060223.83415946, 

Describe: count   307511.000
mean     27108.574
std      14493.455
min       1615.500
25%      16524.000
50%      24903.000
75%      34596.000
max     258025.500
Name: AMT_ANNUITY, dtype: float64 

ValueCounts: AMT_ANNUITY
9000.000    2.076
13500.000   1.793
6750.000    0.741
10125.000   0.662
37800.000   0.521
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 450000.0
Median   : 450000.0, 
Variance: 136367294759.62889, 

Describe: count    307511.000
mean     538396.207
std      369279.426
min       40500.000
25%      238500.000
50%      450000.000
75%      679500.000
max     4050000.000
Name: AMT_GOODS_PRICE, dtype: float64 

ValueCounts: AMT_GOODS_PRICE
450000.000   8.462
225000.000   8.221
675000.000   8.117
900000.000   5.013
270000.000   3.716
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Unaccompanied

Unique: ['Unaccompanied' 'Family' 'Spouse' 'Children' 'Others' 'Group of people']

ValueCounts: NAME_TYPE_SUITE
Unaccompanied   81.239
Family          13.056
Spouse           3.697
Children         1.062
Others           0.857
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Working

Unique: ['Working' 'State servant' 'Commercial associate' 'Pensioner' 'Unemployed'
 'Student' 'Businessman' 'Maternity leave']

ValueCounts: NAME_INCOME_TYPE
Working                51.632
Commercial associate   23.289
Pensioner              18.003
State servant           7.058
Unemployed              0.007
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Secondary

Unique: ['Secondary' 'Higher education' 'Incomplete higher' 'Lower secondary'
 'Academic degree']

ValueCounts: NAME_EDUCATION_TYPE
Secondary           71.019
Higher education    24.345
Incomplete higher    3.342
Lower secondary      1.241
Academic degree      0.053
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Married

Unique: ['Single' 'Married' 'Civil marriage' 'Widow' 'Separated' 'Unknown']

ValueCounts: NAME_FAMILY_STATUS
Married          63.878
Single           14.778
Civil marriage    9.683
Separated         6.429
Widow             5.232
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: House

Unique: ['House' 'Rented apartment' 'With parents' 'Municipal apartment'
 'Office apartment' 'Co-op apartment']

ValueCounts: NAME_HOUSING_TYPE
House                 88.734
With parents           4.826
Municipal apartment    3.637
Rented apartment       1.587
Office apartment       0.851
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.035792
Median   : 0.01885, 
Variance: 0.00019130430983259252, 

Describe: count   307511.000
mean         0.021
std          0.014
min          0.000
25%          0.010
50%          0.019
75%          0.029
max          0.073
Name: REGION_POPULATION_RELATIVE, dtype: float64 

ValueCounts: REGION_POPULATION_RELATIVE
0.036   5.336
0.046   4.371
0.031   3.955
0.025   3.886
0.026   3.773
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 38
Median   : 43.0, 
Variance: 142.9122920578924, 

Describe: count   307511.000
mean        43.436
std         11.955
min         20.000
25%         34.000
50%         43.000
75%         53.000
max         69.000
Name: YEARS_BIRTH, dtype: float64 

ValueCounts: YEARS_BIRTH
38   2.885
37   2.861
39   2.852
40   2.804
36   2.801
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 1000
Median   : 6.0, 
Variance: 145902.7541267848, 

Describe: count   307511.000
mean       185.022
std        381.972
min          0.000
25%          2.000
50%          6.000
75%         15.000
max       1000.000
Name: YEARS_EMPLOYED, dtype: float64 

ValueCounts: YEARS_EMPLOYED
1000   18.007
1      10.354
2       9.641
0       9.074
3       8.165
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0
Median   : 12.0, 
Variance: 93.06154614411118, 

Describe: count   307511.000
mean        13.169
std          9.647
min          0.000
25%          5.000
50%         12.000
75%         20.000
max         67.000
Name: YEARS_REGISTRATION, dtype: float64 

ValueCounts: YEARS_REGISTRATION
0    5.482
1    5.065
2    4.877
3    4.225
12   4.214
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 11

Unique: [ 5  0  6  9  1 10  2  8 13  3  4  7 12 11 14 16 15 17 19]

Median   : 8.0, 
Variance: 17.094215357427704, 

Describe: count   307511.000
mean         7.713
std          4.135
min          0.000
25%          4.000
50%          8.000
75%         11.000
max         19.000
Name: YEARS_ID_PUBLISH, dtype: float64 

ValueCounts: YEARS_ID_PUBLISH
11   14.396
12   12.508
10    6.934
13    6.686
7     6.256
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 1

Unique: [1 0]

Median   : 1.0, 
Variance: 3.251916191640975e-06, 

Describe: count   307511.000
mean         1.000
std          0.002
min          0.000
25%          1.000
50%          1.000
75%          1.000
max          1.000
Name: FLAG_MOBIL, dtype: float64 

ValueCounts: FLAG_MOBIL
1   100.000
0     0.000
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 1

Unique: [1 0]

Median   : 1.0, 
Variance: 0.147671271296676, 

Describe: count   307511.000
mean         0.820
std          0.384
min          0.000
25%          1.000
50%          1.000
75%          1.000
max          1.000
Name: FLAG_EMP_PHONE, dtype: float64 

ValueCounts: FLAG_EMP_PHONE
1   81.989
0   18.011
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.15962120697935117, 

Describe: count   307511.000
mean         0.199
std          0.400
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_WORK_PHONE, dtype: float64 

ValueCounts: FLAG_WORK_PHONE
0   80.063
1   19.937
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 1

Unique: [1 0]

Median   : 1.0, 
Variance: 0.0018631217575536686, 

Describe: count   307511.000
mean         0.998
std          0.043
min          0.000
25%          1.000
50%          1.000
75%          1.000
max          1.000
Name: FLAG_CONT_MOBILE, dtype: float64 

ValueCounts: FLAG_CONT_MOBILE
1   99.813
0    0.187
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [1 0]

Median   : 0.0, 
Variance: 0.20206872204588347, 

Describe: count   307511.000
mean         0.281
std          0.450
min          0.000
25%          0.000
50%          0.000
75%          1.000
max          1.000
Name: FLAG_PHONE, dtype: float64 

ValueCounts: FLAG_PHONE
0   71.893
1   28.107
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.0535029466256677, 

Describe: count   307511.000
mean         0.057
std          0.231
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_EMAIL, dtype: float64 

ValueCounts: FLAG_EMAIL
0   94.328
1    5.672
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 96391,
Mode: Laborers

Unique: ['Laborers' 'Core staff' 'Accountants' 'Managers' nan 'Drivers'
 'Sales staff' 'Cleaning staff' 'Cooking staff' 'Private service staff'
 'Medicine staff' 'Security staff' 'High skill tech staff'
 'Waiters/barmen staff' 'Low-skill Laborers' 'Realty agents' 'Secretaries'
 'IT staff' 'HR staff']

ValueCounts: OCCUPATION_TYPE
Laborers      26.140
Sales staff   15.206
Core staff    13.059
Managers      10.123
Drivers        8.812
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 2.0

Unique: [ 1.  2.  3.  4.  5.  6.  9.  7.  8. 10. 13. 14. 12. 20. 15. 16. 11.]

Median   : 2.0, 
Variance: 0.8293356781151717, 

Describe: count   307511.000
mean         2.153
std          0.911
min          1.000
25%          2.000
50%          2.000
75%          3.000
max         20.000
Name: CNT_FAM_MEMBERS, dtype: float64 

ValueCounts: CNT_FAM_MEMBERS
2.000   51.497
1.000   22.063
3.000   17.105
4.000    8.031
5.000    1.131
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 2

Unique: [2 1 3]

Median   : 2.0, 
Variance: 0.2591155142160186, 

Describe: count   307511.000
mean         2.052
std          0.509
min          1.000
25%          2.000
50%          2.000
75%          2.000
max          3.000
Name: REGION_RATING_CLIENT, dtype: float64 

ValueCounts: REGION_RATING_CLIENT
2   73.813
3   15.717
1   10.470
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 2

Unique: [2 1 3]

Median   : 2.0, 
Variance: 0.2527445242643602, 

Describe: count   307511.000
mean         2.032
std          0.503
min          1.000
25%          2.000
50%          2.000
75%          2.000
max          3.000
Name: REGION_RATING_CLIENT_W_CITY, dtype: float64 

ValueCounts: REGION_RATING_CLIENT_W_CITY
2   74.626
3   14.263
1   11.111
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: TUESDAY

Unique: ['WEDNESDAY' 'MONDAY' 'THURSDAY' 'SUNDAY' 'SATURDAY' 'FRIDAY' 'TUESDAY']

ValueCounts: WEEKDAY_APPR_PROCESS_START
TUESDAY     17.528
WEDNESDAY   16.889
MONDAY      16.492
THURSDAY    16.452
FRIDAY      16.369
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 10

Unique: [10 11  9 17 16 14  8 15  7 13  6 12 19  3 18 21  4  5 20 22  1  2 23  0]

Median   : 12.0, 
Variance: 10.665660320665818, 

Describe: count   307511.000
mean        12.063
std          3.266
min          0.000
25%         10.000
50%         12.000
75%         14.000
max         23.000
Name: HOUR_APPR_PROCESS_START, dtype: float64 

ValueCounts: HOUR_APPR_PROCESS_START
10   12.267
11   12.107
12   11.132
13   10.068
14    9.002
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.014914876209068216, 

Describe: count   307511.000
mean         0.015
std          0.122
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: REG_REGION_NOT_LIVE_REGION, dtype: float64 

ValueCounts: REG_REGION_NOT_LIVE_REGION
0   98.486
1    1.514
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.04819158950907743, 

Describe: count   307511.000
mean         0.051
std          0.220
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: REG_REGION_NOT_WORK_REGION, dtype: float64 

ValueCounts: REG_REGION_NOT_WORK_REGION
0   94.923
1    5.077
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.039005704439295116, 

Describe: count   307511.000
mean         0.041
std          0.197
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: LIVE_REGION_NOT_WORK_REGION, dtype: float64 

ValueCounts: LIVE_REGION_NOT_WORK_REGION
0   95.934
1    4.066
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.07206205892600395, 

Describe: count   307511.000
mean         0.078
std          0.268
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: REG_CITY_NOT_LIVE_CITY, dtype: float64 

ValueCounts: REG_CITY_NOT_LIVE_CITY
0   92.183
1    7.817
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.17734528517578357, 

Describe: count   307511.000
mean         0.230
std          0.421
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: REG_CITY_NOT_WORK_CITY, dtype: float64 

ValueCounts: REG_CITY_NOT_WORK_CITY
0   76.955
1   23.045
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.14731519424670028, 

Describe: count   307511.000
mean         0.180
std          0.384
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: LIVE_CITY_NOT_WORK_CITY, dtype: float64 

ValueCounts: LIVE_CITY_NOT_WORK_CITY
0   82.045
1   17.955
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Business Entity Type 3
ValueCounts: ORGANIZATION_TYPE
Business Entity Type 3   22.110
XNA                      18.007
Self-employed            12.491
Other                     5.425
Medicine                  3.640
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.2858978721410488
Median   : 0.5659614260608526, 
Variance: 0.03643133076657983, 

Describe: count   307511.000
mean         0.515
std          0.191
min          0.000
25%          0.393
50%          0.566
75%          0.663
max          0.855
Name: EXT_SOURCE_2, dtype: float64 

ValueCounts: EXT_SOURCE_2
0.286   0.234
0.566   0.215
0.262   0.136
0.265   0.112
0.160   0.105
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.5352762504724826
Median   : 0.5352762504724826, 
Variance: 0.030532570833502897, 

Describe: count   307511.000
mean         0.516
std          0.175
min          0.001
25%          0.417
50%          0.535
75%          0.636
max          0.896
Name: EXT_SOURCE_3, dtype: float64 

ValueCounts: EXT_SOURCE_3
0.535   20.081
0.746    0.475
0.714    0.428
0.694    0.415
0.671    0.387
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.10254666268544127
Median   : 0.10254666268544127, 
Variance: 0.0059740116766552115, 

Describe: count   307511.000
mean         0.103
std          0.077
min          0.000
25%          0.067
50%          0.103
75%          0.103
max          1.000
Name: TOTALAREA_MODE, dtype: float64 

ValueCounts: TOTALAREA_MODE
0.103   48.269
0.000    0.189
0.057    0.080
0.055    0.075
0.055    0.074
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: No

Unique: ['No' 'Yes']

ValueCounts: EMERGENCYSTATE_MODE
No    99.243
Yes    0.757
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.0

Unique: [  2.   1.   0.   4.   8.  10.   7.   3.   6.   5.  12.   9.  13.  11.
  14.  22.  16.  15.  17.  20.  25.  19.  18.  21.  24.  23.  28.  26.
  29.  27.  47. 348.  30.]

Median   : 0.0, 
Variance: 5.75230052383172, 

Describe: count   307511.000
mean         1.418
std          2.398
min          0.000
25%          0.000
50%          0.000
75%          2.000
max        348.000
Name: OBS_30_CNT_SOCIAL_CIRCLE, dtype: float64 

ValueCounts: OBS_30_CNT_SOCIAL_CIRCLE
0.000   53.634
1.000   15.864
2.000    9.693
3.000    6.609
4.000    4.599
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.0

Unique: [ 2.  0.  1.  3.  4.  5.  6.  7. 34.  8.]

Median   : 0.0, 
Variance: 0.19894504078331582, 

Describe: count   307511.000
mean         0.143
std          0.446
min          0.000
25%          0.000
50%          0.000
75%          0.000
max         34.000
Name: DEF_30_CNT_SOCIAL_CIRCLE, dtype: float64 

ValueCounts: DEF_30_CNT_SOCIAL_CIRCLE
0.000   88.564
1.000    9.212
2.000    1.731
3.000    0.388
4.000    0.082
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.0

Unique: [  2.   1.   0.   4.   8.  10.   7.   3.   6.   5.  12.   9.  13.  11.
  14.  21.  15.  22.  16.  20.  25.  17.  19.  18.  24.  23.  28.  29.
  27.  47. 344.  30.  26.]

Median   : 0.0, 
Variance: 5.651195211134213, 

Describe: count   307511.000
mean         1.401
std          2.377
min          0.000
25%          0.000
50%          0.000
75%          2.000
max        344.000
Name: OBS_60_CNT_SOCIAL_CIRCLE, dtype: float64 

ValueCounts: OBS_60_CNT_SOCIAL_CIRCLE
0.000   53.880
1.000   15.892
2.000    9.680
3.000    6.574
4.000    4.535
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.0

Unique: [ 2.  0.  1.  3.  5.  4.  7. 24.  6.]

Median   : 0.0, 
Variance: 0.13085195710647549, 

Describe: count   307511.000
mean         0.100
std          0.362
min          0.000
25%          0.000
50%          0.000
75%          0.000
max         24.000
Name: DEF_60_CNT_SOCIAL_CIRCLE, dtype: float64 

ValueCounts: DEF_60_CNT_SOCIAL_CIRCLE
0.000   91.620
1.000    7.103
2.000    1.031
3.000    0.194
4.000    0.044
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [ 3  2  1  6  4  0  7  5  8  9 10 11]

Median   : 2.0, 
Variance: 4.812206075852008, 

Describe: count   307511.000
mean         2.225
std          2.194
min          0.000
25%          0.000
50%          2.000
75%          4.000
max         11.000
Name: YEARS_LAST_PHONE_CHANGE, dtype: float64 

ValueCounts: YEARS_LAST_PHONE_CHANGE
0   30.064
1   18.744
4   12.377
2   12.338
3    9.733
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 4.227326079183737e-05, 

Describe: count   307511.000
mean         0.000
std          0.007
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_2, dtype: float64 

ValueCounts: FLAG_DOCUMENT_2
0   99.996
1    0.004
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 1

Unique: [1 0]

Median   : 1.0, 
Variance: 0.20589084885614972, 

Describe: count   307511.000
mean         0.710
std          0.454
min          0.000
25%          0.000
50%          1.000
75%          1.000
max          1.000
Name: FLAG_DOCUMENT_3, dtype: float64 

ValueCounts: FLAG_DOCUMENT_3
1   71.002
0   28.998
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 8.129155979353892e-05, 

Describe: count   307511.000
mean         0.000
std          0.009
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_4, dtype: float64 

ValueCounts: FLAG_DOCUMENT_4
0   99.992
1    0.008
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.014886494471071949, 

Describe: count   307511.000
mean         0.015
std          0.122
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_5, dtype: float64 

ValueCounts: FLAG_DOCUMENT_5
0   98.489
1    1.511
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.08030189665614886, 

Describe: count   307511.000
mean         0.088
std          0.283
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_6, dtype: float64 

ValueCounts: FLAG_DOCUMENT_6
0   91.194
1    8.806
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.00019182686767647077, 

Describe: count   307511.000
mean         0.000
std          0.014
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_7, dtype: float64 

ValueCounts: FLAG_DOCUMENT_7
0   99.981
1    0.019
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.07475414850818889, 

Describe: count   307511.000
mean         0.081
std          0.273
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_8, dtype: float64 

ValueCounts: FLAG_DOCUMENT_8
0   91.862
1    8.138
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.0038806309936899315, 

Describe: count   307511.000
mean         0.004
std          0.062
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_9, dtype: float64 

ValueCounts: FLAG_DOCUMENT_9
0   99.610
1    0.390
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 2.276296919164723e-05, 

Describe: count   307511.000
mean         0.000
std          0.005
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_10, dtype: float64 

ValueCounts: FLAG_DOCUMENT_10
0   99.998
1    0.002
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.0038967636747465713, 

Describe: count   307511.000
mean         0.004
std          0.062
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_11, dtype: float64 

ValueCounts: FLAG_DOCUMENT_11
0   99.609
1    0.391
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 6.5038112332850705e-06, 

Describe: count   307511.000
mean         0.000
std          0.003
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_12, dtype: float64 

ValueCounts: FLAG_DOCUMENT_12
0   99.999
1    0.001
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.003512662405688219, 

Describe: count   307511.000
mean         0.004
std          0.059
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_13, dtype: float64 

ValueCounts: FLAG_DOCUMENT_13
0   99.647
1    0.353
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.0029278669255157, 

Describe: count   307511.000
mean         0.003
std          0.054
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_14, dtype: float64 

ValueCounts: FLAG_DOCUMENT_14
0   99.706
1    0.294
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.0012082533473030956, 

Describe: count   307511.000
mean         0.001
std          0.035
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_15, dtype: float64 

ValueCounts: FLAG_DOCUMENT_15
0   99.879
1    0.121
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.009829564925667255, 

Describe: count   307511.000
mean         0.010
std          0.099
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_16, dtype: float64 

ValueCounts: FLAG_DOCUMENT_16
0   99.007
1    0.993
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.00026658688860710627, 

Describe: count   307511.000
mean         0.000
std          0.016
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_17, dtype: float64 

ValueCounts: FLAG_DOCUMENT_17
0   99.973
1    0.027
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.00806372320835816, 

Describe: count   307511.000
mean         0.008
std          0.090
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_18, dtype: float64 

ValueCounts: FLAG_DOCUMENT_18
0   99.187
1    0.813
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.0005947484523376042, 

Describe: count   307511.000
mean         0.001
std          0.024
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_19, dtype: float64 

ValueCounts: FLAG_DOCUMENT_19
0   99.940
1    0.060
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.0005070432225544667, 

Describe: count   307511.000
mean         0.001
std          0.023
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_20, dtype: float64 

ValueCounts: FLAG_DOCUMENT_20
0   99.949
1    0.051
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.00033483626685604196, 

Describe: count   307511.000
mean         0.000
std          0.018
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          1.000
Name: FLAG_DOCUMENT_21, dtype: float64 

ValueCounts: FLAG_DOCUMENT_21
0   99.967
1    0.033
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.0

Unique: [0. 1. 2. 3. 4.]

Median   : 0.0, 
Variance: 0.006086204557259623, 

Describe: count   307511.000
mean         0.006
std          0.078
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          4.000
Name: AMT_REQ_CREDIT_BUREAU_HOUR, dtype: float64 

ValueCounts: AMT_REQ_CREDIT_BUREAU_HOUR
0.000   99.471
1.000    0.507
2.000    0.018
3.000    0.003
4.000    0.000
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.0

Unique: [0. 1. 3. 2. 4. 5. 6. 9. 8.]

Median   : 0.0, 
Variance: 0.010616648120285148, 

Describe: count   307511.000
mean         0.006
std          0.103
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          9.000
Name: AMT_REQ_CREDIT_BUREAU_DAY, dtype: float64 

ValueCounts: AMT_REQ_CREDIT_BUREAU_DAY
0.000   99.516
1.000    0.420
2.000    0.034
3.000    0.015
4.000    0.008
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.0

Unique: [0. 1. 3. 2. 4. 5. 6. 8. 7.]

Median   : 0.0, 
Variance: 0.03637714618090649, 

Describe: count   307511.000
mean         0.030
std          0.191
min          0.000
25%          0.000
50%          0.000
75%          0.000
max          8.000
Name: AMT_REQ_CREDIT_BUREAU_WEEK, dtype: float64 

ValueCounts: AMT_REQ_CREDIT_BUREAU_WEEK
0.000   97.224
1.000    2.669
2.000    0.065
3.000    0.019
4.000    0.011
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.0

Unique: [ 0.  1.  2.  6.  5.  3.  7.  9.  4. 11.  8. 16. 12. 14. 10. 13. 17. 24.
 19. 15. 23. 18. 27. 22.]

Median   : 0.0, 
Variance: 0.7341235021786728, 

Describe: count   307511.000
mean         0.231
std          0.857
min          0.000
25%          0.000
50%          0.000
75%          0.000
max         27.000
Name: AMT_REQ_CREDIT_BUREAU_MON, dtype: float64 

ValueCounts: AMT_REQ_CREDIT_BUREAU_MON
0.000   85.770
1.000   10.779
2.000    1.751
3.000    0.647
4.000    0.350
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.0

Unique: [  0.   1.   2.   4.   3.   8.   5.   6.   7. 261.  19.]

Median   : 0.0, 
Variance: 0.5536237845454758, 

Describe: count   307511.000
mean         0.230
std          0.744
min          0.000
25%          0.000
50%          0.000
75%          0.000
max        261.000
Name: AMT_REQ_CREDIT_BUREAU_QRT, dtype: float64 

ValueCounts: AMT_REQ_CREDIT_BUREAU_QRT
0.000   83.553
1.000   11.012
2.000    4.687
3.000    0.558
4.000    0.155
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.0

Unique: [ 1.  0.  2.  4.  5.  3.  8.  6.  9.  7. 10. 11. 13. 16. 12. 25. 23. 15.
 14. 22. 17. 19. 18. 21. 20.]

Median   : 1.0, 
Variance: 3.444070410791478, 

Describe: count   307511.000
mean         1.643
std          1.856
min          0.000
25%          0.000
50%          1.000
75%          3.000
max         25.000
Name: AMT_REQ_CREDIT_BUREAU_YEAR, dtype: float64 

ValueCounts: AMT_REQ_CREDIT_BUREAU_YEAR
0.000   36.851
1.000   20.619
2.000   16.322
3.000   10.936
4.000    6.736
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.1667
Median   : 0.1667, 
Variance: 0.01126328020998967, 

Describe: count   307511.000
mean         0.196
std          0.106
min          0.000
25%          0.167
50%          0.167
75%          0.167
max          1.000
Name: FLOORSMAX, dtype: float64 

ValueCounts: FLOORSMAX
0.167   69.852
0.333   10.352
0.042    4.735
0.375    2.531
0.125    2.188
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.9775174983915752
Median   : 0.9775174983915752, 
Variance: 0.0018895990628744477, 

Describe: count   307511.000
mean         0.978
std          0.043
min          0.000
25%          0.978
50%          0.978
75%          0.982
max          1.000
Name: YEARS_BEGINEXPLOITATION, dtype: float64 

ValueCounts: YEARS_BEGINEXPLOITATION
0.978   48.781
0.987    1.316
0.986    1.264
0.987    1.255
0.980    1.234
Name: proportion, dtype: float64 



------------------------------------------------------------------

• Apart from OCCUPATION_TYPE column all other columns are either dropped, or imputed
• OCCUPATION_TYPE column still has 31% missing values, upon analysis , it is decided that it cannot be replaced only with MODE, since there is no significant representation of categories.
• Also the column cannot be dropped since it is a critical feature, therefore we decided to identify whether it is MAR, MCAR, MNAR.
• So we keep the column as it is until later stage.

curr_appl_data1["AMT_CREDIT_BINS"] = pd.cut(curr_appl_data1['AMT_CREDIT'], bins=[0,200000,400000,600000,800000,1000000,10000000], labels=["0-200K","200-400k","400-600k","600-800k","800-1M","1M+"])
curr_appl_data1["YEARS_EMPLOYED_BINS"] = pd.cut(curr_appl_data1['YEARS_EMPLOYED'], bins=[-100,10,20,30,40,50,60,1000], labels=["0-10","10-20","20-30","30-40","40-50","50-60","60+"])
curr_appl_data1['AGE_Category'] = pd.cut(curr_appl_data1['YEARS_BIRTH'], [0,30,40,50,60,200], labels=["<30","30-40","40-50","50-60","60+"])
curr_appl_data1

	SK_ID_CURR	TARGET	NAME_CONTRACT_TYPE	CODE_GENDER	FLAG_OWN_CAR	FLAG_OWN_REALTY	CNT_CHILDREN	AMT_INCOME_TOTAL	AMT_CREDIT	AMT_ANNUITY	AMT_GOODS_PRICE	NAME_TYPE_SUITE	NAME_INCOME_TYPE	NAME_EDUCATION_TYPE	NAME_FAMILY_STATUS	NAME_HOUSING_TYPE	REGION_POPULATION_RELATIVE	YEARS_BIRTH	YEARS_EMPLOYED	YEARS_REGISTRATION	YEARS_ID_PUBLISH	FLAG_MOBIL	FLAG_EMP_PHONE	FLAG_WORK_PHONE	FLAG_CONT_MOBILE	FLAG_PHONE	FLAG_EMAIL	OCCUPATION_TYPE	CNT_FAM_MEMBERS	REGION_RATING_CLIENT	REGION_RATING_CLIENT_W_CITY	WEEKDAY_APPR_PROCESS_START	HOUR_APPR_PROCESS_START	REG_REGION_NOT_LIVE_REGION	REG_REGION_NOT_WORK_REGION	LIVE_REGION_NOT_WORK_REGION	REG_CITY_NOT_LIVE_CITY	REG_CITY_NOT_WORK_CITY	LIVE_CITY_NOT_WORK_CITY	ORGANIZATION_TYPE	EXT_SOURCE_2	EXT_SOURCE_3	TOTALAREA_MODE	EMERGENCYSTATE_MODE	OBS_30_CNT_SOCIAL_CIRCLE	DEF_30_CNT_SOCIAL_CIRCLE	OBS_60_CNT_SOCIAL_CIRCLE	DEF_60_CNT_SOCIAL_CIRCLE	YEARS_LAST_PHONE_CHANGE	FLAG_DOCUMENT_2	FLAG_DOCUMENT_3	FLAG_DOCUMENT_4	FLAG_DOCUMENT_5	FLAG_DOCUMENT_6	FLAG_DOCUMENT_7	FLAG_DOCUMENT_8	FLAG_DOCUMENT_9	FLAG_DOCUMENT_10	FLAG_DOCUMENT_11	FLAG_DOCUMENT_12	FLAG_DOCUMENT_13	FLAG_DOCUMENT_14	FLAG_DOCUMENT_15	FLAG_DOCUMENT_16	FLAG_DOCUMENT_17	FLAG_DOCUMENT_18	FLAG_DOCUMENT_19	FLAG_DOCUMENT_20	FLAG_DOCUMENT_21	AMT_REQ_CREDIT_BUREAU_HOUR	AMT_REQ_CREDIT_BUREAU_DAY	AMT_REQ_CREDIT_BUREAU_WEEK	AMT_REQ_CREDIT_BUREAU_MON	AMT_REQ_CREDIT_BUREAU_QRT	AMT_REQ_CREDIT_BUREAU_YEAR	FLOORSMAX	YEARS_BEGINEXPLOITATION	AMT_CREDIT_BINS	YEARS_EMPLOYED_BINS	AGE_Category
0	100002	1	Cash loans	M	No	Yes	0	202500.000	406597.500	24700.500	351000.000	Unaccompanied	Working	Secondary	Single	House	0.019	25	1	9	5	1	1	0	1	1	0	Laborers	1.000	2	2	WEDNESDAY	10	0	0	0	0	0	0	Business Entity Type 3	0.263	0.139	0.015	No	2.000	2.000	2.000	2.000	3	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	1.000	0.083	0.972	400-600k	0-10	<30
1	100003	0	Cash loans	F	No	No	0	270000.000	1293502.500	35698.500	1129500.000	Family	State servant	Higher education	Married	House	0.004	45	3	3	0	1	1	0	1	1	0	Core staff	2.000	1	1	MONDAY	11	0	0	0	0	0	0	School	0.622	0.535	0.071	No	1.000	0.000	1.000	0.000	2	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	0.000	0.292	0.985	1M+	0-10	40-50
2	100004	0	Revolving loans	M	Yes	Yes	0	67500.000	135000.000	6750.000	135000.000	Unaccompanied	Working	Secondary	Single	House	0.010	52	0	11	6	1	1	1	1	1	0	Laborers	1.000	2	2	MONDAY	9	0	0	0	0	0	0	Government	0.556	0.730	0.103	No	0.000	0.000	0.000	0.000	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	0.000	0.167	0.978	0-200K	0-10	50-60
3	100006	0	Cash loans	F	No	Yes	0	135000.000	312682.500	29686.500	297000.000	Unaccompanied	Working	Secondary	Civil marriage	House	0.008	52	8	26	6	1	1	0	1	0	0	Laborers	2.000	2	2	WEDNESDAY	17	0	0	0	0	0	0	Business Entity Type 3	0.650	0.535	0.103	No	2.000	0.000	2.000	0.000	1	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	0.000	0.167	0.978	200-400k	0-10	50-60
4	100007	0	Cash loans	M	No	Yes	0	121500.000	513000.000	21865.500	513000.000	Unaccompanied	Working	Secondary	Single	House	0.029	54	8	11	9	1	1	0	1	0	0	Core staff	1.000	2	2	THURSDAY	11	0	0	0	0	1	1	Religion	0.323	0.535	0.103	No	0.000	0.000	0.000	0.000	3	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	0.000	0.167	0.978	400-600k	0-10	50-60
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
307506	456251	0	Cash loans	M	No	No	0	157500.000	254700.000	27558.000	225000.000	Unaccompanied	Working	Secondary	Separated	With parents	0.033	25	0	23	5	1	1	0	1	0	0	Sales staff	1.000	1	1	THURSDAY	15	0	0	0	0	0	0	Services	0.682	0.535	0.290	No	0.000	0.000	0.000	0.000	0	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	0.000	0.556	0.984	200-400k	0-10	<30
307507	456252	0	Cash loans	F	No	Yes	0	72000.000	269550.000	12001.500	225000.000	Unaccompanied	Pensioner	Secondary	Widow	House	0.025	56	1000	12	11	1	0	0	1	1	0	NaN	1.000	2	2	MONDAY	8	0	0	0	0	0	0	XNA	0.116	0.535	0.021	No	0.000	0.000	0.000	0.000	0	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	0.000	0.083	0.973	200-400k	60+	50-60
307508	456253	0	Cash loans	F	No	Yes	0	153000.000	677664.000	29979.000	585000.000	Unaccompanied	Working	Higher education	Separated	House	0.005	41	21	18	14	1	1	0	1	0	1	Managers	1.000	3	3	THURSDAY	9	0	0	0	0	1	1	School	0.536	0.219	0.797	No	6.000	0.000	6.000	0.000	5	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1.000	0.000	0.000	1.000	0.000	1.000	0.167	0.982	600-800k	20-30	40-50
307509	456254	1	Cash loans	F	No	Yes	0	171000.000	370107.000	20205.000	319500.000	Unaccompanied	Commercial associate	Secondary	Married	House	0.005	32	13	7	2	1	1	0	1	0	0	Laborers	2.000	2	2	WEDNESDAY	9	0	0	0	1	1	0	Business Entity Type 1	0.514	0.661	0.009	No	0.000	0.000	0.000	0.000	0	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	0.000	0.042	0.977	200-400k	10-20	30-40
307510	456255	0	Cash loans	F	No	No	0	157500.000	675000.000	49117.500	675000.000	Unaccompanied	Commercial associate	Higher education	Married	House	0.046	46	3	14	1	1	1	1	1	1	0	Laborers	2.000	1	1	THURSDAY	20	0	0	0	0	1	1	Business Entity Type 3	0.709	0.114	0.072	No	0.000	0.000	0.000	0.000	2	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	2.000	0.000	1.000	0.375	0.988	600-800k	0-10	40-50

307511 rows × 80 columns

curr_appl_data1[curr_appl_data1['OCCUPATION_TYPE'].isna()]['NAME_INCOME_TYPE'].value_counts(normalize=True)

NAME_INCOME_TYPE
Pensioner              0.574
Working                0.259
Commercial associate   0.128
State servant          0.039
Unemployed             0.000
Student                0.000
Businessman            0.000
Maternity leave        0.000
Name: proportion, dtype: float64

Majority of the null values are under Pensioner category, therefore it is highly acceptable to replace Null values in Occupation Types with Pensioner as category

curr_appl_data1["OCCUPATION_TYPE"] = curr_appl_data1["OCCUPATION_TYPE"].fillna('Pensioner')

Univariate and Outlier analysis

curr_target_train_0 = curr_appl_data1.loc[curr_appl_data1['TARGET'] == 0]
curr_target_train_1 = curr_appl_data1.loc[curr_appl_data1['TARGET'] == 1]     

• The Requirement is to analyze the features based on Target variable, we therefore segmented it into two parts:
• Train 1 - All the data that is client with payment difficulties
• Train 0 - All the other set, that doesn't fall under

• As part of univariate analysis, we will generate plots and charts for all the columns depending on the data type. The inferences are explained below

curr_desc_cols = curr_appl_data1.describe().columns
dtype_dict = classify_feature_dtype(curr_appl_data1,curr_desc_cols)
curr_dtype_dict = dtype_dict['float_ts']

# univariate_plots(curr_appl_data1.head(60), curr_appl_data1.select_dtypes(object).columns, t0=curr_target_train_0, t1=curr_target_train_1, ftype="category")
# print("--------------------------------------------------------------------------")
# univariate_plots(curr_appl_data1.head(60), dtype_dict['int_cat'], t0=curr_target_train_0, t1=curr_target_train_1, ftype="category")
# print("--------------------------------------------------------------------------")
# univariate_plots(curr_appl_data1.head(60), dtype_dict['float_ts'], ftype="non_categorical")

univariate_plots(curr_appl_data1, curr_appl_data1.select_dtypes(object).columns, t0 = curr_target_train_0, t1 = curr_target_train_1, ftype = "category")
print("--------------------------------------------------------------------------")
univariate_plots(curr_appl_data1, dtype_dict['int_cat'], t0 = curr_target_train_0, t1 = curr_target_train_1, ftype = "category")
print("--------------------------------------------------------------------------")
univariate_plots(curr_appl_data1, dtype_dict['float_ts'], ftype = "non_categorical")

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

--------------------------------------------------------------------------

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

--------------------------------------------------------------------------

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

• Univariate and Outlier analysis:
• Important Observations

• Name_Contract Type:

  • In the given dataset, Majority of the loans are cash based loans.
  • Clients find Cash based loans attractive.

• Code Gender:

  • Majority of the clients are Females,
  • Females interest towards loan application is higher compared to Males.
  • They are more responsible towards timely repayment of the installments

• Flag Own Car:

  • Majority of the applicants do not own car.

• Flag Own Realty:

  • Majority of the applicants own a property.

• Name_Income_Type:

  • Majority of the applicants are working citizens.

• Name_education_type:

  • Majority of the applicants requesting for loan have completed secondary education

• Name_Family_Status:

  • Most applicants requesting for loan are Married.

• Housing Type:

  • Majority of the applicants own a house

• Occupation Type:

  • Most applications are laborers.
  • The possibility of their loan getting rejected are very high.

• Target Feature

  • Majority of the applicants do not have any payment difficulties

• Flag_Mobile:

  • Almost everyone owns a mobile.

• Flag_columns:

  • Majority of the flag columns are mostly one category, therefore not much insights can be derived.

• CNT_children:

  • There are outliers in cnt_children feature.
  • Many applicants have either one or two childrens

• AMT_INcome:

  • There are outliers in the income feature.

• AMT_Credit:

  • There are outlier in the AMT_credit feature.
  • Binnig or capping needs to be done.

• AMT_Annuity:

  • Majority of the applicants have similar annuity
  • skewed distribution

• Years_Birth

  • Majority if the population is between 30 -40 age

• Years employed

  • There are outliers

• Years Registration:

  • Majority are applicants have recently changed their registrations.

• CNT_Family Members:

  • There are outliers
  • Most applicants have between 1 to 5 family members

• Most clients applied for the loan between 10-13 hrs, which is predictable.

• Years employed:

  • Outliers are present

• Ext source 2:

  • Most clients have around 0.6 rating

• Ext source 3:

  • Most clients have around 0.5 rating

Data Cleaning - Previous Application Dataset

prev_appl_na_pct = check_cols_null_pct(prev_appl_data)
prev_appl_data1 = prev_appl_data.loc[:, (prev_appl_na_pct < 50)]
prev_appl_data1.columns

Index(['SK_ID_PREV', 'SK_ID_CURR', 'NAME_CONTRACT_TYPE', 'AMT_ANNUITY',
       'AMT_APPLICATION', 'AMT_CREDIT', 'AMT_GOODS_PRICE',
       'WEEKDAY_APPR_PROCESS_START', 'HOUR_APPR_PROCESS_START',
       'FLAG_LAST_APPL_PER_CONTRACT', 'NFLAG_LAST_APPL_IN_DAY',
       'NAME_CASH_LOAN_PURPOSE', 'NAME_CONTRACT_STATUS', 'DAYS_DECISION',
       'NAME_PAYMENT_TYPE', 'CODE_REJECT_REASON', 'NAME_TYPE_SUITE',
       'NAME_CLIENT_TYPE', 'NAME_GOODS_CATEGORY', 'NAME_PORTFOLIO',
       'NAME_PRODUCT_TYPE', 'CHANNEL_TYPE', 'SELLERPLACE_AREA',
       'NAME_SELLER_INDUSTRY', 'CNT_PAYMENT', 'NAME_YIELD_GROUP',
       'PRODUCT_COMBINATION', 'DAYS_FIRST_DRAWING', 'DAYS_FIRST_DUE',
       'DAYS_LAST_DUE_1ST_VERSION', 'DAYS_LAST_DUE', 'DAYS_TERMINATION',
       'NFLAG_INSURED_ON_APPROVAL'],
      dtype='object')

• Since there are many columns that have null values and are more than 50% it is generally not recommended to impute it,
• Therefore we drop all columns gt 50%

show_stats(prev_appl_data1, prev_appl_data1.columns)
prev_appl_na_pct1 = check_cols_null_pct(prev_appl_data1)
prev_appl_na_pct1

Total Nulls: 0,
Mode: 1000001
Median   : 1923110.5, 
Variance: 283660585606.6597, 

Describe: count   1670214.000
mean    1923089.135
std      532597.959
min     1000001.000
25%     1461857.250
50%     1923110.500
75%     2384279.750
max     2845382.000
Name: SK_ID_PREV, dtype: float64 

ValueCounts: SK_ID_PREV
2030495   0.000
1035848   0.000
1526498   0.000
2148893   0.000
2437429   0.000
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 187868
Median   : 278714.5, 
Variance: 10570888003.009909, 

Describe: count   1670214.000
mean     278357.174
std      102814.824
min      100001.000
25%      189329.000
50%      278714.500
75%      367514.000
max      456255.000
Name: SK_ID_CURR, dtype: float64 

ValueCounts: SK_ID_CURR
187868   0.005
265681   0.004
173680   0.004
242412   0.004
206783   0.004
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Cash loans

Unique: ['Consumer loans' 'Cash loans' 'Revolving loans' 'XNA']

ValueCounts: NAME_CONTRACT_TYPE
Cash loans        44.758
Consumer loans    43.656
Revolving loans   11.565
XNA                0.021
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 372235,
Mode: 2250.0
Median   : 11250.0, 
Variance: 218511584.1819954, 

Describe: count   1297979.000
mean      15955.121
std       14782.137
min           0.000
25%        6321.780
50%       11250.000
75%       20658.420
max      418058.145
Name: AMT_ANNUITY, dtype: float64 

ValueCounts: AMT_ANNUITY
2250.000    2.455
11250.000   1.077
6750.000    1.036
9000.000    0.963
22500.000   0.917
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.0
Median   : 71046.0, 
Variance: 85719989263.5158, 

Describe: count   1670214.000
mean     175233.860
std      292779.762
min           0.000
25%       18720.000
50%       71046.000
75%      180360.000
max     6905160.000
Name: AMT_APPLICATION, dtype: float64 

ValueCounts: AMT_APPLICATION
0.000        23.494
45000.000     2.864
225000.000    2.607
135000.000    2.435
450000.000    2.329
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 1,
Mode: 0.0
Median   : 80541.0, 
Variance: 101489786307.34668, 

Describe: count   1670213.000
mean     196114.021
std      318574.617
min           0.000
25%       24160.500
50%       80541.000
75%      216418.500
max     6905160.000
Name: AMT_CREDIT, dtype: float64 

ValueCounts: AMT_CREDIT
0.000        20.163
45000.000     2.099
225000.000    1.263
450000.000    1.195
135000.000    1.121
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 385515,
Mode: 45000.0
Median   : 112320.0, 
Variance: 99474988758.27444, 

Describe: count   1284699.000
mean     227847.279
std      315396.558
min           0.000
25%       50841.000
50%      112320.000
75%      234000.000
max     6905160.000
Name: AMT_GOODS_PRICE, dtype: float64 

ValueCounts: AMT_GOODS_PRICE
45000.000    3.723
225000.000   3.390
135000.000   3.165
450000.000   3.030
90000.000    2.286
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: TUESDAY

Unique: ['SATURDAY' 'THURSDAY' 'TUESDAY' 'MONDAY' 'FRIDAY' 'SUNDAY' 'WEDNESDAY']

ValueCounts: WEEKDAY_APPR_PROCESS_START
TUESDAY     15.275
WEDNESDAY   15.268
MONDAY      15.181
FRIDAY      15.091
THURSDAY    14.914
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 11

Unique: [15 11  7  9  8 10 12 13 14 16  6  4  5 19 17 18 20 22 21  3  1  2 23  0]

Median   : 12.0, 
Variance: 11.11574247090721, 

Describe: count   1670214.000
mean         12.484
std           3.334
min           0.000
25%          10.000
50%          12.000
75%          15.000
max          23.000
Name: HOUR_APPR_PROCESS_START, dtype: float64 

ValueCounts: HOUR_APPR_PROCESS_START
11   11.539
12   11.135
10   10.878
13   10.313
14    9.443
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Y

Unique: ['Y' 'N']

ValueCounts: FLAG_LAST_APPL_PER_CONTRACT
Y   99.493
N    0.507
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 1

Unique: [1 0]

Median   : 1.0, 
Variance: 0.003520005148666215, 

Describe: count   1670214.000
mean          0.996
std           0.059
min           0.000
25%           1.000
50%           1.000
75%           1.000
max           1.000
Name: NFLAG_LAST_APPL_IN_DAY, dtype: float64 

ValueCounts: NFLAG_LAST_APPL_IN_DAY
1   99.647
0    0.353
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: XAP

Unique: ['XAP' 'XNA' 'Repairs' 'Everyday expenses' 'Car repairs'
 'Building a house or an annex' 'Other' 'Journey'
 'Purchase of electronic equipment' 'Medicine' 'Payments on other loans'
 'Urgent needs' 'Buying a used car' 'Buying a new car'
 'Buying a holiday home / land' 'Education' 'Buying a home' 'Furniture'
 'Buying a garage' 'Business development' 'Wedding / gift / holiday'
 'Hobby' 'Gasification / water supply' 'Refusal to name the goal'
 'Money for a third person']

ValueCounts: NAME_CASH_LOAN_PURPOSE
XAP            55.242
XNA            40.589
Repairs         1.423
Other           0.934
Urgent needs    0.504
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Approved

Unique: ['Approved' 'Refused' 'Canceled' 'Unused offer']

ValueCounts: NAME_CONTRACT_STATUS
Approved       62.075
Canceled       18.939
Refused        17.404
Unused offer    1.583
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: -245
Median   : -581.0, 
Variance: 606996.2907461969, 

Describe: count   1670214.000
mean       -880.680
std         779.100
min       -2922.000
25%       -1300.000
50%        -581.000
75%        -280.000
max          -1.000
Name: DAYS_DECISION, dtype: float64 

ValueCounts: DAYS_DECISION
-245   0.146
-238   0.143
-210   0.142
-273   0.141
-196   0.139
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Cash through the bank

Unique: ['Cash through the bank' 'XNA' 'Non-cash from your account'
 'Cashless from the account of the employer']

ValueCounts: NAME_PAYMENT_TYPE
Cash through the bank                       61.881
XNA                                         37.563
Non-cash from your account                   0.491
Cashless from the account of the employer    0.065
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: XAP

Unique: ['XAP' 'HC' 'LIMIT' 'CLIENT' 'SCOFR' 'SCO' 'XNA' 'VERIF' 'SYSTEM']

ValueCounts: CODE_REJECT_REASON
XAP      81.013
HC       10.492
LIMIT     3.334
SCO       2.243
CLIENT    1.583
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 820405,
Mode: Unaccompanied

Unique: [nan 'Unaccompanied' 'Spouse, partner' 'Family' 'Children' 'Other_B'
 'Other_A' 'Group of people']

ValueCounts: NAME_TYPE_SUITE
Unaccompanied     59.892
Family            25.095
Spouse, partner    7.892
Children           3.714
Other_B            2.074
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Repeater

Unique: ['Repeater' 'New' 'Refreshed' 'XNA']

ValueCounts: NAME_CLIENT_TYPE
Repeater    73.719
New         18.043
Refreshed    8.122
XNA          0.116
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: XNA

Unique: ['Mobile' 'XNA' 'Consumer Electronics' 'Construction Materials'
 'Auto Accessories' 'Photo / Cinema Equipment' 'Computers' 'Audio/Video'
 'Medicine' 'Clothing and Accessories' 'Furniture' 'Sport and Leisure'
 'Homewares' 'Gardening' 'Jewelry' 'Vehicles' 'Education'
 'Medical Supplies' 'Other' 'Direct Sales' 'Office Appliances' 'Fitness'
 'Tourism' 'Insurance' 'Additional Service' 'Weapon' 'Animals'
 'House Construction']

ValueCounts: NAME_GOODS_CATEGORY
XNA                    56.927
Mobile                 13.454
Consumer Electronics    7.279
Computers               6.333
Audio/Video             5.954
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: POS

Unique: ['POS' 'Cash' 'XNA' 'Cards' 'Cars']

ValueCounts: NAME_PORTFOLIO
POS     41.373
Cash    27.635
XNA     22.286
Cards    8.681
Cars     0.025
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: XNA

Unique: ['XNA' 'x-sell' 'walk-in']

ValueCounts: NAME_PRODUCT_TYPE
XNA       63.684
x-sell    27.319
walk-in    8.997
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Credit and cash offices

Unique: ['Country-wide' 'Contact center' 'Credit and cash offices' 'Stone'
 'Regional / Local' 'AP+ (Cash loan)' 'Channel of corporate sales'
 'Car dealer']

ValueCounts: CHANNEL_TYPE
Credit and cash offices   43.106
Country-wide              29.618
Stone                     12.698
Regional / Local           6.498
Contact center             4.269
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: -1
Median   : 3.0, 
Variance: 50800450.2636309, 

Describe: count   1670214.000
mean        313.951
std        7127.443
min          -1.000
25%          -1.000
50%           3.000
75%          82.000
max     4000000.000
Name: SELLERPLACE_AREA, dtype: float64 

ValueCounts: SELLERPLACE_AREA
-1    45.663
 0     3.624
 50    2.239
 30    2.061
 20    2.026
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: XNA

Unique: ['Connectivity' 'XNA' 'Consumer electronics' 'Industry' 'Clothing'
 'Furniture' 'Construction' 'Jewelry' 'Auto technology' 'MLM partners'
 'Tourism']

ValueCounts: NAME_SELLER_INDUSTRY
XNA                    51.234
Consumer electronics   23.845
Connectivity           16.527
Furniture               3.464
Construction            1.783
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 372230,
Mode: 12.0
Median   : 12.0, 
Variance: 212.2058722166199, 

Describe: count   1297984.000
mean         16.054
std          14.567
min           0.000
25%           6.000
50%          12.000
75%          24.000
max          84.000
Name: CNT_PAYMENT, dtype: float64 

ValueCounts: CNT_PAYMENT
12.000   24.889
6.000    14.674
0.000    11.170
10.000   10.929
24.000   10.614
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: XNA

Unique: ['middle' 'low_action' 'high' 'low_normal' 'XNA']

ValueCounts: NAME_YIELD_GROUP
XNA          30.967
middle       23.083
high         21.155
low_normal   19.285
low_action    5.511
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 346,
Mode: Cash

Unique: ['POS mobile with interest' 'Cash X-Sell: low' 'Cash X-Sell: high'
 'Cash X-Sell: middle' 'Cash Street: high' 'Cash'
 'POS household without interest' 'POS household with interest'
 'POS other with interest' 'Card X-Sell' 'POS mobile without interest'
 'Card Street' 'POS industry with interest' 'Cash Street: low'
 'POS industry without interest' 'Cash Street: middle'
 'POS others without interest' nan]

ValueCounts: PRODUCT_COMBINATION
Cash                          17.127
POS household with interest   15.787
POS mobile with interest      13.215
Cash X-Sell: middle            8.616
Cash X-Sell: low               7.800
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 673065,
Mode: 365243.0
Median   : 365243.0, 
Variance: 7906075654.949019, 

Describe: count   997149.000
mean    342209.855
std      88916.116
min      -2922.000
25%     365243.000
50%     365243.000
75%     365243.000
max     365243.000
Name: DAYS_FIRST_DRAWING, dtype: float64 

ValueCounts: DAYS_FIRST_DRAWING
365243.000   93.712
-228.000      0.012
-224.000      0.012
-212.000      0.012
-223.000      0.012
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 673065,
Mode: 365243.0
Median   : -831.0, 
Variance: 5248259146.994171, 

Describe: count   997149.000
mean     13826.269
std      72444.870
min      -2892.000
25%      -1628.000
50%       -831.000
75%       -411.000
max     365243.000
Name: DAYS_FIRST_DUE, dtype: float64 

ValueCounts: DAYS_FIRST_DUE
365243.000   4.076
-334.000     0.077
-509.000     0.076
-208.000     0.075
-330.000     0.075
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 673065,
Mode: 365243.0
Median   : -361.0, 
Variance: 11418425883.920301, 

Describe: count   997149.000
mean     33767.774
std     106857.035
min      -2801.000
25%      -1242.000
50%       -361.000
75%        129.000
max     365243.000
Name: DAYS_LAST_DUE_1ST_VERSION, dtype: float64 

ValueCounts: DAYS_LAST_DUE_1ST_VERSION
365243.000   9.413
9.000        0.072
8.000        0.071
0.000        0.071
5.000        0.070
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 673065,
Mode: 365243.0
Median   : -537.0, 
Variance: 22394348853.11387, 

Describe: count   997149.000
mean     76582.403
std     149647.415
min      -2889.000
25%      -1314.000
50%       -537.000
75%        -74.000
max     365243.000
Name: DAYS_LAST_DUE, dtype: float64 

ValueCounts: DAYS_LAST_DUE
365243.000   21.182
-245.000      0.066
-188.000      0.065
-239.000      0.064
-167.000      0.064
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 673065,
Mode: 365243.0
Median   : -499.0, 
Variance: 23501968241.562744, 

Describe: count   997149.000
mean     81992.344
std     153303.517
min      -2874.000
25%      -1270.000
50%       -499.000
75%        -44.000
max     365243.000
Name: DAYS_TERMINATION, dtype: float64 

ValueCounts: DAYS_TERMINATION
365243.000   22.656
-233.000      0.079
-170.000      0.077
-184.000      0.077
-163.000      0.077
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 673065,
Mode: 0.0

Unique: [ 0.  1. nan]

Median   : 0.0, 
Variance: 0.2219674704465378, 

Describe: count   997149.000
mean         0.333
std          0.471
min          0.000
25%          0.000
50%          0.000
75%          1.000
max          1.000
Name: NFLAG_INSURED_ON_APPROVAL, dtype: float64 

ValueCounts: NFLAG_INSURED_ON_APPROVAL
0.000   66.743
1.000   33.257
Name: proportion, dtype: float64 



------------------------------------------------------------------

NAME_TYPE_SUITE               49.120
DAYS_FIRST_DRAWING            40.298
DAYS_TERMINATION              40.298
DAYS_LAST_DUE                 40.298
DAYS_LAST_DUE_1ST_VERSION     40.298
DAYS_FIRST_DUE                40.298
NFLAG_INSURED_ON_APPROVAL     40.298
AMT_GOODS_PRICE               23.082
AMT_ANNUITY                   22.287
CNT_PAYMENT                   22.286
PRODUCT_COMBINATION            0.021
AMT_CREDIT                     0.000
WEEKDAY_APPR_PROCESS_START     0.000
HOUR_APPR_PROCESS_START        0.000
NAME_CONTRACT_TYPE             0.000
AMT_APPLICATION                0.000
NAME_YIELD_GROUP               0.000
NAME_SELLER_INDUSTRY           0.000
SELLERPLACE_AREA               0.000
CHANNEL_TYPE                   0.000
NAME_PRODUCT_TYPE              0.000
NAME_PORTFOLIO                 0.000
NAME_GOODS_CATEGORY            0.000
NAME_CLIENT_TYPE               0.000
SK_ID_CURR                     0.000
CODE_REJECT_REASON             0.000
NAME_PAYMENT_TYPE              0.000
DAYS_DECISION                  0.000
NAME_CONTRACT_STATUS           0.000
NAME_CASH_LOAN_PURPOSE         0.000
NFLAG_LAST_APPL_IN_DAY         0.000
FLAG_LAST_APPL_PER_CONTRACT    0.000
SK_ID_PREV                     0.000
dtype: float64

prev_appl_data1.select_dtypes(object)
prev_appl_data1.isna().sum()

	NAME_CONTRACT_TYPE	WEEKDAY_APPR_PROCESS_START	FLAG_LAST_APPL_PER_CONTRACT	NAME_CASH_LOAN_PURPOSE	NAME_CONTRACT_STATUS	NAME_PAYMENT_TYPE	CODE_REJECT_REASON	NAME_TYPE_SUITE	NAME_CLIENT_TYPE	NAME_GOODS_CATEGORY	NAME_PORTFOLIO	NAME_PRODUCT_TYPE	CHANNEL_TYPE	NAME_SELLER_INDUSTRY	NAME_YIELD_GROUP	PRODUCT_COMBINATION
0	Consumer loans	SATURDAY	Y	XAP	Approved	Cash through the bank	XAP	NaN	Repeater	Mobile	POS	XNA	Country-wide	Connectivity	middle	POS mobile with interest
1	Cash loans	THURSDAY	Y	XNA	Approved	XNA	XAP	Unaccompanied	Repeater	XNA	Cash	x-sell	Contact center	XNA	low_action	Cash X-Sell: low
2	Cash loans	TUESDAY	Y	XNA	Approved	Cash through the bank	XAP	Spouse, partner	Repeater	XNA	Cash	x-sell	Credit and cash offices	XNA	high	Cash X-Sell: high
3	Cash loans	MONDAY	Y	XNA	Approved	Cash through the bank	XAP	NaN	Repeater	XNA	Cash	x-sell	Credit and cash offices	XNA	middle	Cash X-Sell: middle
4	Cash loans	THURSDAY	Y	Repairs	Refused	Cash through the bank	HC	NaN	Repeater	XNA	Cash	walk-in	Credit and cash offices	XNA	high	Cash Street: high
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
1670209	Consumer loans	WEDNESDAY	Y	XAP	Approved	Cash through the bank	XAP	NaN	Refreshed	Furniture	POS	XNA	Stone	Furniture	low_normal	POS industry with interest
1670210	Consumer loans	TUESDAY	Y	XAP	Approved	Cash through the bank	XAP	Unaccompanied	New	Furniture	POS	XNA	Stone	Furniture	middle	POS industry with interest
1670211	Consumer loans	MONDAY	Y	XAP	Approved	Cash through the bank	XAP	Spouse, partner	Repeater	Consumer Electronics	POS	XNA	Country-wide	Consumer electronics	low_normal	POS household with interest
1670212	Cash loans	WEDNESDAY	Y	XNA	Approved	Cash through the bank	XAP	Family	Repeater	XNA	Cash	x-sell	AP+ (Cash loan)	XNA	low_normal	Cash X-Sell: low
1670213	Cash loans	SUNDAY	Y	XNA	Approved	Cash through the bank	XAP	Family	Repeater	XNA	Cash	x-sell	AP+ (Cash loan)	XNA	middle	Cash X-Sell: middle

1670214 rows × 16 columns

SK_ID_PREV                          0
SK_ID_CURR                          0
NAME_CONTRACT_TYPE                  0
AMT_ANNUITY                    372235
AMT_APPLICATION                     0
AMT_CREDIT                          1
AMT_GOODS_PRICE                385515
WEEKDAY_APPR_PROCESS_START          0
HOUR_APPR_PROCESS_START             0
FLAG_LAST_APPL_PER_CONTRACT         0
NFLAG_LAST_APPL_IN_DAY              0
NAME_CASH_LOAN_PURPOSE              0
NAME_CONTRACT_STATUS                0
DAYS_DECISION                       0
NAME_PAYMENT_TYPE                   0
CODE_REJECT_REASON                  0
NAME_TYPE_SUITE                820405
NAME_CLIENT_TYPE                    0
NAME_GOODS_CATEGORY                 0
NAME_PORTFOLIO                      0
NAME_PRODUCT_TYPE                   0
CHANNEL_TYPE                        0
SELLERPLACE_AREA                    0
NAME_SELLER_INDUSTRY                0
CNT_PAYMENT                    372230
NAME_YIELD_GROUP                    0
PRODUCT_COMBINATION               346
DAYS_FIRST_DRAWING             673065
DAYS_FIRST_DUE                 673065
DAYS_LAST_DUE_1ST_VERSION      673065
DAYS_LAST_DUE                  673065
DAYS_TERMINATION               673065
NFLAG_INSURED_ON_APPROVAL      673065
dtype: int64

prev_appl_data1["DAYS_LAST_DUE"] = prev_appl_data1["DAYS_LAST_DUE"].fillna(prev_appl_data1["DAYS_LAST_DUE"].median())
prev_appl_data1["DAYS_FIRST_DUE"] = prev_appl_data1["DAYS_FIRST_DUE"].fillna(prev_appl_data1["DAYS_FIRST_DUE"].median())
prev_appl_data1["DAYS_TERMINATION"] = prev_appl_data1["DAYS_TERMINATION"].fillna(prev_appl_data1["DAYS_TERMINATION"].median())
prev_appl_data1["DAYS_FIRST_DRAWING"] = prev_appl_data1["DAYS_FIRST_DRAWING"].fillna(prev_appl_data1["DAYS_FIRST_DRAWING"].median())
prev_appl_data1["DAYS_LAST_DUE_1ST_VERSION"] = prev_appl_data1["DAYS_LAST_DUE_1ST_VERSION"].fillna(prev_appl_data1["DAYS_LAST_DUE_1ST_VERSION"].median())

prev_appl_data1["NFLAG_INSURED_ON_APPROVAL"] = prev_appl_data1["NFLAG_INSURED_ON_APPROVAL"].fillna(prev_appl_data1["NFLAG_INSURED_ON_APPROVAL"].mode()[0])
prev_appl_data1["NAME_TYPE_SUITE"] = prev_appl_data1["NAME_TYPE_SUITE"].fillna(prev_appl_data1["NAME_TYPE_SUITE"].mode()[0])

• Impute all missing values with median of the columns
• Imputed categorical columns with the mode of that column.

DAYS_COLS = prev_appl_data1.filter(like="DAYS").columns
prev_appl_data1[DAYS_COLS] = (abs(prev_appl_data1[DAYS_COLS])).astype(int)

• DAYS_COLS - Is a list of days type columns, represented in days count,
• There were inconsistencies, therefore made all values absolute

prev_appl_data1["AMT_ANNUITY"].describe()
prev_appl_data1["AMT_GOODS_PRICE"].describe()
show_stats(prev_appl_data1, ['AMT_ANNUITY','AMT_GOODS_PRICE'])

count   1297979.000
mean      15955.121
std       14782.137
min           0.000
25%        6321.780
50%       11250.000
75%       20658.420
max      418058.145
Name: AMT_ANNUITY, dtype: float64

count   1284699.000
mean     227847.279
std      315396.558
min           0.000
25%       50841.000
50%      112320.000
75%      234000.000
max     6905160.000
Name: AMT_GOODS_PRICE, dtype: float64

Total Nulls: 372235,
Mode: 2250.0
Median   : 11250.0, 
Variance: 218511584.1819954, 

Describe: count   1297979.000
mean      15955.121
std       14782.137
min           0.000
25%        6321.780
50%       11250.000
75%       20658.420
max      418058.145
Name: AMT_ANNUITY, dtype: float64 

ValueCounts: AMT_ANNUITY
2250.000    2.455
11250.000   1.077
6750.000    1.036
9000.000    0.963
22500.000   0.917
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 385515,
Mode: 45000.0
Median   : 112320.0, 
Variance: 99474988758.27444, 

Describe: count   1284699.000
mean     227847.279
std      315396.558
min           0.000
25%       50841.000
50%      112320.000
75%      234000.000
max     6905160.000
Name: AMT_GOODS_PRICE, dtype: float64 

ValueCounts: AMT_GOODS_PRICE
45000.000    3.723
225000.000   3.390
135000.000   3.165
450000.000   3.030
90000.000    2.286
Name: proportion, dtype: float64 



------------------------------------------------------------------

• After looking at the stats of these columns, it is better to impute with median of these columns.

# prev_appl_data1 = prev_appl_data1[~prev_appl_data1["AMT_ANNUITY"].isna()]
# prev_appl_data1 = prev_appl_data1[~prev_appl_data1["AMT_GOODS_PRICE"].isna()]

prev_appl_data1["AMT_ANNUITY"] = prev_appl_data1["AMT_ANNUITY"].fillna(prev_appl_data1["AMT_ANNUITY"].median())
prev_appl_data1["AMT_GOODS_PRICE"] = prev_appl_data1["AMT_GOODS_PRICE"].fillna(prev_appl_data1["AMT_GOODS_PRICE"].median())

show_stats(prev_appl_data1, prev_appl_data1.select_dtypes(object).columns )

Total Nulls: 0,
Mode: Cash loans

Unique: ['Consumer loans' 'Cash loans' 'Revolving loans' 'XNA']

ValueCounts: NAME_CONTRACT_TYPE
Cash loans        44.758
Consumer loans    43.656
Revolving loans   11.565
XNA                0.021
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: TUESDAY

Unique: ['SATURDAY' 'THURSDAY' 'TUESDAY' 'MONDAY' 'FRIDAY' 'SUNDAY' 'WEDNESDAY']

ValueCounts: WEEKDAY_APPR_PROCESS_START
TUESDAY     15.275
WEDNESDAY   15.268
MONDAY      15.181
FRIDAY      15.091
THURSDAY    14.914
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Y

Unique: ['Y' 'N']

ValueCounts: FLAG_LAST_APPL_PER_CONTRACT
Y   99.493
N    0.507
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: XAP

Unique: ['XAP' 'XNA' 'Repairs' 'Everyday expenses' 'Car repairs'
 'Building a house or an annex' 'Other' 'Journey'
 'Purchase of electronic equipment' 'Medicine' 'Payments on other loans'
 'Urgent needs' 'Buying a used car' 'Buying a new car'
 'Buying a holiday home / land' 'Education' 'Buying a home' 'Furniture'
 'Buying a garage' 'Business development' 'Wedding / gift / holiday'
 'Hobby' 'Gasification / water supply' 'Refusal to name the goal'
 'Money for a third person']

ValueCounts: NAME_CASH_LOAN_PURPOSE
XAP            55.242
XNA            40.589
Repairs         1.423
Other           0.934
Urgent needs    0.504
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Approved

Unique: ['Approved' 'Refused' 'Canceled' 'Unused offer']

ValueCounts: NAME_CONTRACT_STATUS
Approved       62.075
Canceled       18.939
Refused        17.404
Unused offer    1.583
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Cash through the bank

Unique: ['Cash through the bank' 'XNA' 'Non-cash from your account'
 'Cashless from the account of the employer']

ValueCounts: NAME_PAYMENT_TYPE
Cash through the bank                       61.881
XNA                                         37.563
Non-cash from your account                   0.491
Cashless from the account of the employer    0.065
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: XAP

Unique: ['XAP' 'HC' 'LIMIT' 'CLIENT' 'SCOFR' 'SCO' 'XNA' 'VERIF' 'SYSTEM']

ValueCounts: CODE_REJECT_REASON
XAP      81.013
HC       10.492
LIMIT     3.334
SCO       2.243
CLIENT    1.583
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Unaccompanied

Unique: ['Unaccompanied' 'Spouse, partner' 'Family' 'Children' 'Other_B' 'Other_A'
 'Group of people']

ValueCounts: NAME_TYPE_SUITE
Unaccompanied     79.593
Family            12.769
Spouse, partner    4.016
Children           1.890
Other_B            1.055
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Repeater

Unique: ['Repeater' 'New' 'Refreshed' 'XNA']

ValueCounts: NAME_CLIENT_TYPE
Repeater    73.719
New         18.043
Refreshed    8.122
XNA          0.116
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: XNA

Unique: ['Mobile' 'XNA' 'Consumer Electronics' 'Construction Materials'
 'Auto Accessories' 'Photo / Cinema Equipment' 'Computers' 'Audio/Video'
 'Medicine' 'Clothing and Accessories' 'Furniture' 'Sport and Leisure'
 'Homewares' 'Gardening' 'Jewelry' 'Vehicles' 'Education'
 'Medical Supplies' 'Other' 'Direct Sales' 'Office Appliances' 'Fitness'
 'Tourism' 'Insurance' 'Additional Service' 'Weapon' 'Animals'
 'House Construction']

ValueCounts: NAME_GOODS_CATEGORY
XNA                    56.927
Mobile                 13.454
Consumer Electronics    7.279
Computers               6.333
Audio/Video             5.954
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: POS

Unique: ['POS' 'Cash' 'XNA' 'Cards' 'Cars']

ValueCounts: NAME_PORTFOLIO
POS     41.373
Cash    27.635
XNA     22.286
Cards    8.681
Cars     0.025
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: XNA

Unique: ['XNA' 'x-sell' 'walk-in']

ValueCounts: NAME_PRODUCT_TYPE
XNA       63.684
x-sell    27.319
walk-in    8.997
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Credit and cash offices

Unique: ['Country-wide' 'Contact center' 'Credit and cash offices' 'Stone'
 'Regional / Local' 'AP+ (Cash loan)' 'Channel of corporate sales'
 'Car dealer']

ValueCounts: CHANNEL_TYPE
Credit and cash offices   43.106
Country-wide              29.618
Stone                     12.698
Regional / Local           6.498
Contact center             4.269
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: XNA

Unique: ['Connectivity' 'XNA' 'Consumer electronics' 'Industry' 'Clothing'
 'Furniture' 'Construction' 'Jewelry' 'Auto technology' 'MLM partners'
 'Tourism']

ValueCounts: NAME_SELLER_INDUSTRY
XNA                    51.234
Consumer electronics   23.845
Connectivity           16.527
Furniture               3.464
Construction            1.783
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: XNA

Unique: ['middle' 'low_action' 'high' 'low_normal' 'XNA']

ValueCounts: NAME_YIELD_GROUP
XNA          30.967
middle       23.083
high         21.155
low_normal   19.285
low_action    5.511
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 346,
Mode: Cash

Unique: ['POS mobile with interest' 'Cash X-Sell: low' 'Cash X-Sell: high'
 'Cash X-Sell: middle' 'Cash Street: high' 'Cash'
 'POS household without interest' 'POS household with interest'
 'POS other with interest' 'Card X-Sell' 'POS mobile without interest'
 'Card Street' 'POS industry with interest' 'Cash Street: low'
 'POS industry without interest' 'Cash Street: middle'
 'POS others without interest' nan]

ValueCounts: PRODUCT_COMBINATION
Cash                          17.127
POS household with interest   15.787
POS mobile with interest      13.215
Cash X-Sell: middle            8.616
Cash X-Sell: low               7.800
Name: proportion, dtype: float64 



------------------------------------------------------------------

Univariate and Outlier analysis - Prev Application

• As part of univariate analysis, we will generate plots and charts for all the columns depending on the data type. The inferences are explained below

prev_desc_cols = prev_appl_data1.describe().columns
prev_dtype_dict = dtype_dict = classify_feature_dtype(prev_appl_data1,prev_desc_cols)
prev_dtype_dict = dtype_dict['float_ts']

# univariate_plots(prev_appl_data1.head(60), prev_appl_data1.select_dtypes(object).columns, ftype="category")
# print("--------------------------------------------------------------------------")
# univariate_plots(prev_appl_data1.head(60), dtype_dict['int_cat'], ftype="category")
# print("--------------------------------------------------------------------------")
# univariate_plots(prev_appl_data1.head(60), dtype_dict['float_ts'], ftype="non_categorical")

univariate_plots(prev_appl_data1, prev_appl_data1.select_dtypes(object).columns, ftype="category")
print("--------------------------------------------------------------------------")
univariate_plots(prev_appl_data1, dtype_dict['int_cat'], ftype="category")
print("--------------------------------------------------------------------------")
univariate_plots(prev_appl_data1, dtype_dict['float_ts'], ftype="non_categorical")

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

--------------------------------------------------------------------------

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

--------------------------------------------------------------------------

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

• NAME_CONTRACT_TYPE - for prev applications majority are split between casloans and consumer loans

• WEEKDAY_APPR_PROCESS_START - looks like applications are processed on Sundays , which looks like a outlier

• CASH Loan purpose

  • Majority provided a reason as XAP,
  • the second highest is XNA, therefore this needs to be captured correctly for better insights

• NAME_CONTRACT_STATUS

  • Majority of the Applications were approved

• CODE_REJECT_RREASON

  • Again the code reject reason is not properly catured,
  • this is having greater impacts on our analysis

• Name client type

  • MAjority of the clients have reapplied for loan, which is good sign for credit agencies

• Name Good Category

  • Majority provided a reason as XNA,

• Name Product Type

  • Majority provided a reason as XNA,

• Name Seller Industry

  • Majority provided a reason as XNA,

• Name YIELD Group

  • Majority provided a reason as XNA,

• AMT_ANNUITY , AMT_APPLICATION, AMT_Credit, CNT_PAYMENT, DAYS _Termination, DAYS_LAST_DUE

  • There are outliers

show_stats(prev_appl_data1, prev_appl_data1.columns)
check_cols_null_pct(prev_appl_data1)

Total Nulls: 0,
Mode: 1000001
Median   : 1923110.5, 
Variance: 283660585606.6597, 

Describe: count   1670214.000
mean    1923089.135
std      532597.959
min     1000001.000
25%     1461857.250
50%     1923110.500
75%     2384279.750
max     2845382.000
Name: SK_ID_PREV, dtype: float64 

ValueCounts: SK_ID_PREV
2030495   0.000
1035848   0.000
1526498   0.000
2148893   0.000
2437429   0.000
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 187868
Median   : 278714.5, 
Variance: 10570888003.009909, 

Describe: count   1670214.000
mean     278357.174
std      102814.824
min      100001.000
25%      189329.000
50%      278714.500
75%      367514.000
max      456255.000
Name: SK_ID_CURR, dtype: float64 

ValueCounts: SK_ID_CURR
187868   0.005
265681   0.004
173680   0.004
242412   0.004
206783   0.004
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Cash loans

Unique: ['Consumer loans' 'Cash loans' 'Revolving loans' 'XNA']

ValueCounts: NAME_CONTRACT_TYPE
Cash loans        44.758
Consumer loans    43.656
Revolving loans   11.565
XNA                0.021
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 11250.0
Median   : 11250.0, 
Variance: 173646877.76482412, 

Describe: count   1670214.000
mean      14906.506
std       13177.514
min           0.000
25%        7547.096
50%       11250.000
75%       16824.026
max      418058.145
Name: AMT_ANNUITY, dtype: float64 

ValueCounts: AMT_ANNUITY
11250.000   23.123
2250.000     1.908
6750.000     0.805
9000.000     0.748
22500.000    0.713
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.0
Median   : 71046.0, 
Variance: 85719989263.5158, 

Describe: count   1670214.000
mean     175233.860
std      292779.762
min           0.000
25%       18720.000
50%       71046.000
75%      180360.000
max     6905160.000
Name: AMT_APPLICATION, dtype: float64 

ValueCounts: AMT_APPLICATION
0.000        23.494
45000.000     2.864
225000.000    2.607
135000.000    2.435
450000.000    2.329
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 1,
Mode: 0.0
Median   : 80541.0, 
Variance: 101489786307.34668, 

Describe: count   1670213.000
mean     196114.021
std      318574.617
min           0.000
25%       24160.500
50%       80541.000
75%      216418.500
max     6905160.000
Name: AMT_CREDIT, dtype: float64 

ValueCounts: AMT_CREDIT
0.000        20.163
45000.000     2.099
225000.000    1.263
450000.000    1.195
135000.000    1.121
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 112320.0
Median   : 112320.0, 
Variance: 78883944532.07816, 

Describe: count   1670214.000
mean     201181.535
std      280862.857
min           0.000
25%       67500.000
50%      112320.000
75%      180405.000
max     6905160.000
Name: AMT_GOODS_PRICE, dtype: float64 

ValueCounts: AMT_GOODS_PRICE
112320.000   23.084
45000.000     2.864
225000.000    2.607
135000.000    2.435
450000.000    2.331
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: TUESDAY

Unique: ['SATURDAY' 'THURSDAY' 'TUESDAY' 'MONDAY' 'FRIDAY' 'SUNDAY' 'WEDNESDAY']

ValueCounts: WEEKDAY_APPR_PROCESS_START
TUESDAY     15.275
WEDNESDAY   15.268
MONDAY      15.181
FRIDAY      15.091
THURSDAY    14.914
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 11

Unique: [15 11  7  9  8 10 12 13 14 16  6  4  5 19 17 18 20 22 21  3  1  2 23  0]

Median   : 12.0, 
Variance: 11.11574247090721, 

Describe: count   1670214.000
mean         12.484
std           3.334
min           0.000
25%          10.000
50%          12.000
75%          15.000
max          23.000
Name: HOUR_APPR_PROCESS_START, dtype: float64 

ValueCounts: HOUR_APPR_PROCESS_START
11   11.539
12   11.135
10   10.878
13   10.313
14    9.443
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Y

Unique: ['Y' 'N']

ValueCounts: FLAG_LAST_APPL_PER_CONTRACT
Y   99.493
N    0.507
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 1

Unique: [1 0]

Median   : 1.0, 
Variance: 0.003520005148666215, 

Describe: count   1670214.000
mean          0.996
std           0.059
min           0.000
25%           1.000
50%           1.000
75%           1.000
max           1.000
Name: NFLAG_LAST_APPL_IN_DAY, dtype: float64 

ValueCounts: NFLAG_LAST_APPL_IN_DAY
1   99.647
0    0.353
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: XAP

Unique: ['XAP' 'XNA' 'Repairs' 'Everyday expenses' 'Car repairs'
 'Building a house or an annex' 'Other' 'Journey'
 'Purchase of electronic equipment' 'Medicine' 'Payments on other loans'
 'Urgent needs' 'Buying a used car' 'Buying a new car'
 'Buying a holiday home / land' 'Education' 'Buying a home' 'Furniture'
 'Buying a garage' 'Business development' 'Wedding / gift / holiday'
 'Hobby' 'Gasification / water supply' 'Refusal to name the goal'
 'Money for a third person']

ValueCounts: NAME_CASH_LOAN_PURPOSE
XAP            55.242
XNA            40.589
Repairs         1.423
Other           0.934
Urgent needs    0.504
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Approved

Unique: ['Approved' 'Refused' 'Canceled' 'Unused offer']

ValueCounts: NAME_CONTRACT_STATUS
Approved       62.075
Canceled       18.939
Refused        17.404
Unused offer    1.583
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 245
Median   : 581.0, 
Variance: 606996.2907461969, 

Describe: count   1670214.000
mean        880.680
std         779.100
min           1.000
25%         280.000
50%         581.000
75%        1300.000
max        2922.000
Name: DAYS_DECISION, dtype: float64 

ValueCounts: DAYS_DECISION
245   0.146
238   0.143
210   0.142
273   0.141
196   0.139
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Cash through the bank

Unique: ['Cash through the bank' 'XNA' 'Non-cash from your account'
 'Cashless from the account of the employer']

ValueCounts: NAME_PAYMENT_TYPE
Cash through the bank                       61.881
XNA                                         37.563
Non-cash from your account                   0.491
Cashless from the account of the employer    0.065
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: XAP

Unique: ['XAP' 'HC' 'LIMIT' 'CLIENT' 'SCOFR' 'SCO' 'XNA' 'VERIF' 'SYSTEM']

ValueCounts: CODE_REJECT_REASON
XAP      81.013
HC       10.492
LIMIT     3.334
SCO       2.243
CLIENT    1.583
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Unaccompanied

Unique: ['Unaccompanied' 'Spouse, partner' 'Family' 'Children' 'Other_B' 'Other_A'
 'Group of people']

ValueCounts: NAME_TYPE_SUITE
Unaccompanied     79.593
Family            12.769
Spouse, partner    4.016
Children           1.890
Other_B            1.055
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Repeater

Unique: ['Repeater' 'New' 'Refreshed' 'XNA']

ValueCounts: NAME_CLIENT_TYPE
Repeater    73.719
New         18.043
Refreshed    8.122
XNA          0.116
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: XNA

Unique: ['Mobile' 'XNA' 'Consumer Electronics' 'Construction Materials'
 'Auto Accessories' 'Photo / Cinema Equipment' 'Computers' 'Audio/Video'
 'Medicine' 'Clothing and Accessories' 'Furniture' 'Sport and Leisure'
 'Homewares' 'Gardening' 'Jewelry' 'Vehicles' 'Education'
 'Medical Supplies' 'Other' 'Direct Sales' 'Office Appliances' 'Fitness'
 'Tourism' 'Insurance' 'Additional Service' 'Weapon' 'Animals'
 'House Construction']

ValueCounts: NAME_GOODS_CATEGORY
XNA                    56.927
Mobile                 13.454
Consumer Electronics    7.279
Computers               6.333
Audio/Video             5.954
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: POS

Unique: ['POS' 'Cash' 'XNA' 'Cards' 'Cars']

ValueCounts: NAME_PORTFOLIO
POS     41.373
Cash    27.635
XNA     22.286
Cards    8.681
Cars     0.025
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: XNA

Unique: ['XNA' 'x-sell' 'walk-in']

ValueCounts: NAME_PRODUCT_TYPE
XNA       63.684
x-sell    27.319
walk-in    8.997
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Credit and cash offices

Unique: ['Country-wide' 'Contact center' 'Credit and cash offices' 'Stone'
 'Regional / Local' 'AP+ (Cash loan)' 'Channel of corporate sales'
 'Car dealer']

ValueCounts: CHANNEL_TYPE
Credit and cash offices   43.106
Country-wide              29.618
Stone                     12.698
Regional / Local           6.498
Contact center             4.269
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: -1
Median   : 3.0, 
Variance: 50800450.2636309, 

Describe: count   1670214.000
mean        313.951
std        7127.443
min          -1.000
25%          -1.000
50%           3.000
75%          82.000
max     4000000.000
Name: SELLERPLACE_AREA, dtype: float64 

ValueCounts: SELLERPLACE_AREA
-1    45.663
 0     3.624
 50    2.239
 30    2.061
 20    2.026
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: XNA

Unique: ['Connectivity' 'XNA' 'Consumer electronics' 'Industry' 'Clothing'
 'Furniture' 'Construction' 'Jewelry' 'Auto technology' 'MLM partners'
 'Tourism']

ValueCounts: NAME_SELLER_INDUSTRY
XNA                    51.234
Consumer electronics   23.845
Connectivity           16.527
Furniture               3.464
Construction            1.783
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 372230,
Mode: 12.0
Median   : 12.0, 
Variance: 212.2058722166199, 

Describe: count   1297984.000
mean         16.054
std          14.567
min           0.000
25%           6.000
50%          12.000
75%          24.000
max          84.000
Name: CNT_PAYMENT, dtype: float64 

ValueCounts: CNT_PAYMENT
12.000   24.889
6.000    14.674
0.000    11.170
10.000   10.929
24.000   10.614
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: XNA

Unique: ['middle' 'low_action' 'high' 'low_normal' 'XNA']

ValueCounts: NAME_YIELD_GROUP
XNA          30.967
middle       23.083
high         21.155
low_normal   19.285
low_action    5.511
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 346,
Mode: Cash

Unique: ['POS mobile with interest' 'Cash X-Sell: low' 'Cash X-Sell: high'
 'Cash X-Sell: middle' 'Cash Street: high' 'Cash'
 'POS household without interest' 'POS household with interest'
 'POS other with interest' 'Card X-Sell' 'POS mobile without interest'
 'Card Street' 'POS industry with interest' 'Cash Street: low'
 'POS industry without interest' 'Cash Street: middle'
 'POS others without interest' nan]

ValueCounts: PRODUCT_COMBINATION
Cash                          17.127
POS household with interest   15.787
POS mobile with interest      13.215
Cash X-Sell: middle            8.616
Cash X-Sell: low               7.800
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 365243
Median   : 365243.0, 
Variance: 4793060111.321972, 

Describe: count   1670214.000
mean     351569.514
std       69231.930
min           2.000
25%      365243.000
50%      365243.000
75%      365243.000
max      365243.000
Name: DAYS_FIRST_DRAWING, dtype: float64 

ValueCounts: DAYS_FIRST_DRAWING
365243   96.246
228       0.007
224       0.007
212       0.007
223       0.007
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 831
Median   : 831.0, 
Variance: 3150557682.1309857, 

Describe: count   1670214.000
mean       9856.863
std       56129.829
min           2.000
25%         752.000
50%         831.000
75%        1132.000
max      365243.000
Name: DAYS_FIRST_DUE, dtype: float64 

ValueCounts: DAYS_FIRST_DUE
831      40.331
365243    2.434
334       0.046
509       0.046
208       0.045
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 361
Median   : 361.0, 
Variance: 7050978795.539533, 

Describe: count   1670214.000
mean      21138.662
std       83970.107
min           0.000
25%         361.000
50%         361.000
75%         996.000
max      365243.000
Name: DAYS_LAST_DUE_1ST_VERSION, dtype: float64 

ValueCounts: DAYS_LAST_DUE_1ST_VERSION
361      40.339
365243    5.620
2         0.082
1         0.081
8         0.081
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 537
Median   : 537.0, 
Variance: 14674120897.696123, 

Describe: count   1670214.000
mean      46875.043
std      121136.786
min           2.000
25%         537.000
50%         537.000
75%        1518.000
max      365243.000
Name: DAYS_LAST_DUE, dtype: float64 

ValueCounts: DAYS_LAST_DUE
537      40.328
365243   12.646
245       0.039
188       0.039
239       0.038
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 499
Median   : 499.0, 
Variance: 15539251588.869661, 

Describe: count   1670214.000
mean      50055.597
std      124656.534
min           2.000
25%         499.000
50%         499.000
75%        1544.000
max      365243.000
Name: DAYS_TERMINATION, dtype: float64 

ValueCounts: DAYS_TERMINATION
499      40.330
365243   13.526
233       0.047
170       0.046
184       0.046
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.0

Unique: [0. 1.]

Median   : 0.0, 
Variance: 0.15912835746123868, 

Describe: count   1670214.000
mean          0.199
std           0.399
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: NFLAG_INSURED_ON_APPROVAL, dtype: float64 

ValueCounts: NFLAG_INSURED_ON_APPROVAL
0.000   80.145
1.000   19.855
Name: proportion, dtype: float64 



------------------------------------------------------------------

CNT_PAYMENT                   22.286
PRODUCT_COMBINATION            0.021
AMT_CREDIT                     0.000
NAME_YIELD_GROUP               0.000
NAME_PORTFOLIO                 0.000
NAME_PRODUCT_TYPE              0.000
CHANNEL_TYPE                   0.000
SELLERPLACE_AREA               0.000
NAME_SELLER_INDUSTRY           0.000
SK_ID_PREV                     0.000
NAME_CLIENT_TYPE               0.000
DAYS_FIRST_DRAWING             0.000
DAYS_FIRST_DUE                 0.000
DAYS_LAST_DUE_1ST_VERSION      0.000
DAYS_LAST_DUE                  0.000
DAYS_TERMINATION               0.000
NAME_GOODS_CATEGORY            0.000
NAME_TYPE_SUITE                0.000
SK_ID_CURR                     0.000
CODE_REJECT_REASON             0.000
NAME_PAYMENT_TYPE              0.000
DAYS_DECISION                  0.000
NAME_CONTRACT_STATUS           0.000
NAME_CASH_LOAN_PURPOSE         0.000
NFLAG_LAST_APPL_IN_DAY         0.000
FLAG_LAST_APPL_PER_CONTRACT    0.000
HOUR_APPR_PROCESS_START        0.000
WEEKDAY_APPR_PROCESS_START     0.000
AMT_GOODS_PRICE                0.000
AMT_APPLICATION                0.000
AMT_ANNUITY                    0.000
NAME_CONTRACT_TYPE             0.000
NFLAG_INSURED_ON_APPROVAL      0.000
dtype: float64

prev_appl_data1['CNT_PAYMENT']= prev_appl_data1['CNT_PAYMENT'].fillna(prev_appl_data1['CNT_PAYMENT'].mode()[0])
prev_appl_data1['PRODUCT_COMBINATION'] = prev_appl_data1['PRODUCT_COMBINATION'].fillna(prev_appl_data1['PRODUCT_COMBINATION'].mode()[0])

Data Imbalance

# curr_appl_data1['TARGET'].head(10)
target_ratio_one = (curr_appl_data1['TARGET'] == 1).sum()
target_ratio_zero = (curr_appl_data1['TARGET'] == 0).sum()
target_ratio = target_ratio_zero/target_ratio_one
target_ratio

curr_appl_data1['TARGET'].value_counts(normalize=True).plot.bar()

11.387150050352467

<Axes: xlabel='TARGET'>

Merged/Combined Data Analysis

combined_appl_data = pd.merge(left=curr_appl_data1, right=prev_appl_data1, on="SK_ID_CURR", how='inner', suffixes=[None,'_right'])
combined_appl_data.head()

	SK_ID_CURR	TARGET	NAME_CONTRACT_TYPE	CODE_GENDER	FLAG_OWN_CAR	FLAG_OWN_REALTY	CNT_CHILDREN	AMT_INCOME_TOTAL	AMT_CREDIT	AMT_ANNUITY	AMT_GOODS_PRICE	NAME_TYPE_SUITE	NAME_INCOME_TYPE	NAME_EDUCATION_TYPE	NAME_FAMILY_STATUS	NAME_HOUSING_TYPE	REGION_POPULATION_RELATIVE	YEARS_BIRTH	YEARS_EMPLOYED	YEARS_REGISTRATION	YEARS_ID_PUBLISH	FLAG_MOBIL	FLAG_EMP_PHONE	FLAG_WORK_PHONE	FLAG_CONT_MOBILE	FLAG_PHONE	FLAG_EMAIL	OCCUPATION_TYPE	CNT_FAM_MEMBERS	REGION_RATING_CLIENT	REGION_RATING_CLIENT_W_CITY	WEEKDAY_APPR_PROCESS_START	HOUR_APPR_PROCESS_START	REG_REGION_NOT_LIVE_REGION	REG_REGION_NOT_WORK_REGION	LIVE_REGION_NOT_WORK_REGION	REG_CITY_NOT_LIVE_CITY	REG_CITY_NOT_WORK_CITY	LIVE_CITY_NOT_WORK_CITY	ORGANIZATION_TYPE	EXT_SOURCE_2	EXT_SOURCE_3	TOTALAREA_MODE	EMERGENCYSTATE_MODE	OBS_30_CNT_SOCIAL_CIRCLE	DEF_30_CNT_SOCIAL_CIRCLE	OBS_60_CNT_SOCIAL_CIRCLE	DEF_60_CNT_SOCIAL_CIRCLE	YEARS_LAST_PHONE_CHANGE	FLAG_DOCUMENT_2	FLAG_DOCUMENT_3	FLAG_DOCUMENT_4	FLAG_DOCUMENT_5	FLAG_DOCUMENT_6	FLAG_DOCUMENT_7	FLAG_DOCUMENT_8	FLAG_DOCUMENT_9	FLAG_DOCUMENT_10	FLAG_DOCUMENT_11	FLAG_DOCUMENT_12	FLAG_DOCUMENT_13	FLAG_DOCUMENT_14	FLAG_DOCUMENT_15	FLAG_DOCUMENT_16	FLAG_DOCUMENT_17	FLAG_DOCUMENT_18	FLAG_DOCUMENT_19	FLAG_DOCUMENT_20	FLAG_DOCUMENT_21	AMT_REQ_CREDIT_BUREAU_HOUR	AMT_REQ_CREDIT_BUREAU_DAY	AMT_REQ_CREDIT_BUREAU_WEEK	AMT_REQ_CREDIT_BUREAU_MON	AMT_REQ_CREDIT_BUREAU_QRT	AMT_REQ_CREDIT_BUREAU_YEAR	FLOORSMAX	YEARS_BEGINEXPLOITATION	AMT_CREDIT_BINS	YEARS_EMPLOYED_BINS	AGE_Category	SK_ID_PREV	NAME_CONTRACT_TYPE_right	AMT_ANNUITY_right	AMT_APPLICATION	AMT_CREDIT_right	AMT_GOODS_PRICE_right	WEEKDAY_APPR_PROCESS_START_right	HOUR_APPR_PROCESS_START_right	FLAG_LAST_APPL_PER_CONTRACT	NFLAG_LAST_APPL_IN_DAY	NAME_CASH_LOAN_PURPOSE	NAME_CONTRACT_STATUS	DAYS_DECISION	NAME_PAYMENT_TYPE	CODE_REJECT_REASON	NAME_TYPE_SUITE_right	NAME_CLIENT_TYPE	NAME_GOODS_CATEGORY	NAME_PORTFOLIO	NAME_PRODUCT_TYPE	CHANNEL_TYPE	SELLERPLACE_AREA	NAME_SELLER_INDUSTRY	CNT_PAYMENT	NAME_YIELD_GROUP	PRODUCT_COMBINATION	DAYS_FIRST_DRAWING	DAYS_FIRST_DUE	DAYS_LAST_DUE_1ST_VERSION	DAYS_LAST_DUE	DAYS_TERMINATION	NFLAG_INSURED_ON_APPROVAL
0	100002	1	Cash loans	M	No	Yes	0	202500.000	406597.500	24700.500	351000.000	Unaccompanied	Working	Secondary	Single	House	0.019	25	1	9	5	1	1	0	1	1	0	Laborers	1.000	2	2	WEDNESDAY	10	0	0	0	0	0	0	Business Entity Type 3	0.263	0.139	0.015	No	2.000	2.000	2.000	2.000	3	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	1.000	0.083	0.972	400-600k	0-10	<30	1038818	Consumer loans	9251.775	179055.000	179055.000	179055.000	SATURDAY	9	Y	1	XAP	Approved	606	XNA	XAP	Unaccompanied	New	Vehicles	POS	XNA	Stone	500	Auto technology	24.000	low_normal	POS other with interest	365243	565	125	25	17	0.000
1	100003	0	Cash loans	F	No	No	0	270000.000	1293502.500	35698.500	1129500.000	Family	State servant	Higher education	Married	House	0.004	45	3	3	0	1	1	0	1	1	0	Core staff	2.000	1	1	MONDAY	11	0	0	0	0	0	0	School	0.622	0.535	0.071	No	1.000	0.000	1.000	0.000	2	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	0.000	0.292	0.985	1M+	0-10	40-50	1810518	Cash loans	98356.995	900000.000	1035882.000	900000.000	FRIDAY	12	Y	1	XNA	Approved	746	XNA	XAP	Unaccompanied	Repeater	XNA	Cash	x-sell	Credit and cash offices	-1	XNA	12.000	low_normal	Cash X-Sell: low	365243	716	386	536	527	1.000
2	100003	0	Cash loans	F	No	No	0	270000.000	1293502.500	35698.500	1129500.000	Family	State servant	Higher education	Married	House	0.004	45	3	3	0	1	1	0	1	1	0	Core staff	2.000	1	1	MONDAY	11	0	0	0	0	0	0	School	0.622	0.535	0.071	No	1.000	0.000	1.000	0.000	2	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	0.000	0.292	0.985	1M+	0-10	40-50	2636178	Consumer loans	64567.665	337500.000	348637.500	337500.000	SUNDAY	17	Y	1	XAP	Approved	828	Cash through the bank	XAP	Family	Refreshed	Furniture	POS	XNA	Stone	1400	Furniture	6.000	middle	POS industry with interest	365243	797	647	647	639	0.000
3	100003	0	Cash loans	F	No	No	0	270000.000	1293502.500	35698.500	1129500.000	Family	State servant	Higher education	Married	House	0.004	45	3	3	0	1	1	0	1	1	0	Core staff	2.000	1	1	MONDAY	11	0	0	0	0	0	0	School	0.622	0.535	0.071	No	1.000	0.000	1.000	0.000	2	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	0.000	0.292	0.985	1M+	0-10	40-50	2396755	Consumer loans	6737.310	68809.500	68053.500	68809.500	SATURDAY	15	Y	1	XAP	Approved	2341	Cash through the bank	XAP	Family	Refreshed	Consumer Electronics	POS	XNA	Country-wide	200	Consumer electronics	12.000	middle	POS household with interest	365243	2310	1980	1980	1976	1.000
4	100004	0	Revolving loans	M	Yes	Yes	0	67500.000	135000.000	6750.000	135000.000	Unaccompanied	Working	Secondary	Single	House	0.010	52	0	11	6	1	1	1	1	1	0	Laborers	1.000	2	2	MONDAY	9	0	0	0	0	0	0	Government	0.556	0.730	0.103	No	0.000	0.000	0.000	0.000	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	0.000	0.167	0.978	0-200K	0-10	50-60	1564014	Consumer loans	5357.250	24282.000	20106.000	24282.000	FRIDAY	5	Y	1	XAP	Approved	815	Cash through the bank	XAP	Unaccompanied	New	Mobile	POS	XNA	Regional / Local	30	Connectivity	4.000	middle	POS mobile without interest	365243	784	694	724	714	0.000

show_stats(combined_appl_data, combined_appl_data.columns)
check_cols_null_pct(combined_appl_data)

Total Nulls: 0,
Mode: 265681
Median   : 278992.0, 
Variance: 10570263760.57637, 

Describe: count   1413701.000
mean     278481.288
std      102811.788
min      100002.000
25%      189364.000
50%      278992.000
75%      367556.000
max      456255.000
Name: SK_ID_CURR, dtype: float64 

ValueCounts: SK_ID_CURR
265681   0.005
173680   0.005
242412   0.005
206783   0.005
389950   0.005
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [1 0]

Median   : 0.0, 
Variance: 0.07906159747213361, 

Describe: count   1413701.000
mean          0.087
std           0.281
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: TARGET, dtype: float64 

ValueCounts: TARGET
0   91.345
1    8.655
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Cash loans

Unique: ['Cash loans' 'Revolving loans']

ValueCounts: NAME_CONTRACT_TYPE
Cash loans        92.460
Revolving loans    7.540
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: F

Unique: ['M' 'F' 'XNA']

ValueCounts: CODE_GENDER
F     67.563
M     32.433
XNA    0.004
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: No

Unique: ['No' 'Yes']

ValueCounts: FLAG_OWN_CAR
No    66.292
Yes   33.708
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Yes

Unique: ['Yes' 'No']

ValueCounts: FLAG_OWN_REALTY
Yes   72.440
No    27.560
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [ 0  1  2  3  4  7  5  6  8  9 11 12 10 19 14]

Median   : 0.0, 
Variance: 0.5145843604747418, 

Describe: count   1413701.000
mean          0.405
std           0.717
min           0.000
25%           0.000
50%           0.000
75%           1.000
max          19.000
Name: CNT_CHILDREN, dtype: float64 

ValueCounts: CNT_CHILDREN
0   71.080
1   19.003
2    8.519
3    1.220
4    0.136
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 135000.0
Median   : 157500.0, 
Variance: 39431384073.85981, 

Describe: count     1413701.000
mean       173316.044
std        198573.372
min         25650.000
25%        112500.000
50%        157500.000
75%        207000.000
max     117000000.000
Name: AMT_INCOME_TOTAL, dtype: float64 

ValueCounts: AMT_INCOME_TOTAL
135000.000   11.492
112500.000    9.665
157500.000    9.071
180000.000    8.473
225000.000    7.190
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 450000.0
Median   : 508495.5, 
Variance: 148161352000.59268, 

Describe: count   1413701.000
mean     587553.673
std      384917.331
min       45000.000
25%      270000.000
50%      508495.500
75%      807984.000
max     4050000.000
Name: AMT_CREDIT, dtype: float64 

ValueCounts: AMT_CREDIT
450000.000   3.157
225000.000   2.725
675000.000   2.613
270000.000   1.973
900000.000   1.862
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 13500.0
Median   : 24930.0, 
Variance: 194622195.8703764, 

Describe: count   1413701.000
mean      27017.022
std       13950.706
min        1615.500
25%       16821.000
50%       24930.000
75%       34542.000
max      225000.000
Name: AMT_ANNUITY, dtype: float64 

ValueCounts: AMT_ANNUITY
13500.000   1.396
9000.000    1.329
10125.000   0.514
6750.000    0.483
20250.000   0.481
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 225000.0
Median   : 450000.0, 
Variance: 124676554792.54468, 

Describe: count   1413701.000
mean     527727.717
std      353095.674
min       40500.000
25%      238500.000
50%      450000.000
75%      679500.000
max     4050000.000
Name: AMT_GOODS_PRICE, dtype: float64 

ValueCounts: AMT_GOODS_PRICE
225000.000   8.214
450000.000   8.112
675000.000   7.129
900000.000   4.528
270000.000   3.285
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Unaccompanied

Unique: ['Unaccompanied' 'Family' 'Spouse' 'Children' 'Others' 'Group of people']

ValueCounts: NAME_TYPE_SUITE
Unaccompanied   81.906
Family          13.034
Spouse           3.300
Children         0.984
Others           0.690
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Working

Unique: ['Working' 'State servant' 'Commercial associate' 'Pensioner' 'Unemployed'
 'Student' 'Maternity leave']

ValueCounts: NAME_INCOME_TYPE
Working                51.219
Commercial associate   22.740
Pensioner              19.349
State servant           6.679
Unemployed              0.009
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Secondary

Unique: ['Secondary' 'Higher education' 'Incomplete higher' 'Lower secondary'
 'Academic degree']

ValueCounts: NAME_EDUCATION_TYPE
Secondary           73.417
Higher education    22.130
Incomplete higher    3.198
Lower secondary      1.214
Academic degree      0.041
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Married

Unique: ['Single' 'Married' 'Civil marriage' 'Widow' 'Separated']

ValueCounts: NAME_FAMILY_STATUS
Married          64.434
Single           13.203
Civil marriage   10.126
Separated         6.461
Widow             5.775
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: House

Unique: ['House' 'Rented apartment' 'With parents' 'Municipal apartment'
 'Office apartment' 'Co-op apartment']

ValueCounts: NAME_HOUSING_TYPE
House                 89.478
With parents           4.358
Municipal apartment    3.645
Rented apartment       1.400
Office apartment       0.808
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.035792
Median   : 0.01885, 
Variance: 0.00017814293869714113, 

Describe: count   1413701.000
mean          0.021
std           0.013
min           0.000
25%           0.010
50%           0.019
75%           0.029
max           0.073
Name: REGION_POPULATION_RELATIVE, dtype: float64 

ValueCounts: REGION_POPULATION_RELATIVE
0.036   5.217
0.025   4.559
0.026   4.087
0.046   3.998
0.029   3.951
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 39
Median   : 43.0, 
Variance: 141.66155639435718, 

Describe: count   1413701.000
mean         44.214
std          11.902
min          20.000
25%          34.000
50%          43.000
75%          54.000
max          69.000
Name: YEARS_BIRTH, dtype: float64 

ValueCounts: YEARS_BIRTH
39   2.905
38   2.871
40   2.836
37   2.813
43   2.773
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 1000
Median   : 6.0, 
Variance: 154164.41352542056, 

Describe: count   1413701.000
mean        198.550
std         392.638
min           0.000
25%           2.000
50%           6.000
75%          17.000
max        1000.000
Name: YEARS_EMPLOYED, dtype: float64 

ValueCounts: YEARS_EMPLOYED
1000   19.352
1       9.566
2       9.120
3       8.048
0       7.594
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0
Median   : 12.0, 
Variance: 94.58338443246285, 

Describe: count   1413701.000
mean         13.217
std           9.725
min           0.000
25%           5.000
50%          12.000
75%          20.000
max          67.000
Name: YEARS_REGISTRATION, dtype: float64 

ValueCounts: YEARS_REGISTRATION
0    5.594
1    5.059
2    4.859
3    4.202
12   4.154
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 11

Unique: [ 5  0  6  9  1 10  2  8 13  3  7 12 11  4 14 15 17 16 19]

Median   : 9.0, 
Variance: 17.0505687512264, 

Describe: count   1413701.000
mean          7.824
std           4.129
min           0.000
25%           4.000
50%           9.000
75%          11.000
max          19.000
Name: YEARS_ID_PUBLISH, dtype: float64 

ValueCounts: YEARS_ID_PUBLISH
11   14.803
12   12.820
10    7.013
13    6.923
9     6.340
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 1

Unique: [1]

Median   : 1.0, 
Variance: 0.0, 

Describe: count   1413701.000
mean          1.000
std           0.000
min           1.000
25%           1.000
50%           1.000
75%           1.000
max           1.000
Name: FLAG_MOBIL, dtype: float64 

ValueCounts: FLAG_MOBIL
1   100.000
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 1

Unique: [1 0]

Median   : 1.0, 
Variance: 0.15609508527748012, 

Describe: count   1413701.000
mean          0.806
std           0.395
min           0.000
25%           1.000
50%           1.000
75%           1.000
max           1.000
Name: FLAG_EMP_PHONE, dtype: float64 

ValueCounts: FLAG_EMP_PHONE
1   80.644
0   19.356
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.14904043394825833, 

Describe: count   1413701.000
mean          0.182
std           0.386
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: FLAG_WORK_PHONE, dtype: float64 

ValueCounts: FLAG_WORK_PHONE
0   81.774
1   18.226
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 1

Unique: [1 0]

Median   : 1.0, 
Variance: 0.0007541873628204551, 

Describe: count   1413701.000
mean          0.999
std           0.027
min           0.000
25%           1.000
50%           1.000
75%           1.000
max           1.000
Name: FLAG_CONT_MOBILE, dtype: float64 

ValueCounts: FLAG_CONT_MOBILE
1   99.925
0    0.075
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [1 0]

Median   : 0.0, 
Variance: 0.20285697911891543, 

Describe: count   1413701.000
mean          0.283
std           0.450
min           0.000
25%           0.000
50%           0.000
75%           1.000
max           1.000
Name: FLAG_PHONE, dtype: float64 

ValueCounts: FLAG_PHONE
0   71.712
1   28.288
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.06718311789093928, 

Describe: count   1413701.000
mean          0.072
std           0.259
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: FLAG_EMAIL, dtype: float64 

ValueCounts: FLAG_EMAIL
0   92.757
1    7.243
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Pensioner

Unique: ['Laborers' 'Core staff' 'Accountants' 'Managers' 'Pensioner' 'Drivers'
 'Sales staff' 'Cleaning staff' 'Private service staff' 'Medicine staff'
 'Security staff' 'Cooking staff' 'High skill tech staff'
 'Waiters/barmen staff' 'Low-skill Laborers' 'Realty agents' 'Secretaries'
 'IT staff' 'HR staff']

ValueCounts: OCCUPATION_TYPE
Pensioner     32.368
Laborers      17.776
Sales staff   10.707
Core staff     8.366
Managers       6.716
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 2.0

Unique: [ 1.  2.  3.  4.  5.  6.  9.  7.  8. 10. 13. 14. 12. 20. 15. 16. 11.]

Median   : 2.0, 
Variance: 0.8112221242618761, 

Describe: count   1413701.000
mean          2.151
std           0.901
min           1.000
25%           2.000
50%           2.000
75%           3.000
max          20.000
Name: CNT_FAM_MEMBERS, dtype: float64 

ValueCounts: CNT_FAM_MEMBERS
2.000   52.886
1.000   21.414
3.000   16.493
4.000    7.895
5.000    1.142
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 2

Unique: [2 1 3]

Median   : 2.0, 
Variance: 0.24579750875361692, 

Describe: count   1413701.000
mean          2.060
std           0.496
min           1.000
25%           2.000
50%           2.000
75%           2.000
max           3.000
Name: REGION_RATING_CLIENT, dtype: float64 

ValueCounts: REGION_RATING_CLIENT
2   75.065
3   15.450
1    9.486
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 2

Unique: [2 1 3]

Median   : 2.0, 
Variance: 0.23877207124135386, 

Describe: count   1413701.000
mean          2.038
std           0.489
min           1.000
25%           2.000
50%           2.000
75%           2.000
max           3.000
Name: REGION_RATING_CLIENT_W_CITY, dtype: float64 

ValueCounts: REGION_RATING_CLIENT_W_CITY
2   75.981
3   13.889
1   10.129
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: TUESDAY

Unique: ['WEDNESDAY' 'MONDAY' 'THURSDAY' 'SUNDAY' 'SATURDAY' 'FRIDAY' 'TUESDAY']

ValueCounts: WEEKDAY_APPR_PROCESS_START
TUESDAY     17.587
WEDNESDAY   16.708
MONDAY      16.587
THURSDAY    16.289
FRIDAY      16.287
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 10

Unique: [10 11  9 17 16 14  8 15  7 13  6 12 19  3 18 21  4  5 20 22  1  2 23  0]

Median   : 12.0, 
Variance: 10.446994996371068, 

Describe: count   1413701.000
mean         11.984
std           3.232
min           0.000
25%          10.000
50%          12.000
75%          14.000
max          23.000
Name: HOUR_APPR_PROCESS_START, dtype: float64 

ValueCounts: HOUR_APPR_PROCESS_START
10   12.541
11   12.148
12   11.256
13    9.989
9     9.249
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.011927518818607137, 

Describe: count   1413701.000
mean          0.012
std           0.109
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: REG_REGION_NOT_LIVE_REGION, dtype: float64 

ValueCounts: REG_REGION_NOT_LIVE_REGION
0   98.793
1    1.207
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.042033808859841176, 

Describe: count   1413701.000
mean          0.044
std           0.205
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: REG_REGION_NOT_WORK_REGION, dtype: float64 

ValueCounts: REG_REGION_NOT_WORK_REGION
0   95.603
1    4.397
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.03474850480857744, 

Describe: count   1413701.000
mean          0.036
std           0.186
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: LIVE_REGION_NOT_WORK_REGION, dtype: float64 

ValueCounts: LIVE_REGION_NOT_WORK_REGION
0   96.395
1    3.605
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.06818767668337616, 

Describe: count   1413701.000
mean          0.074
std           0.261
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: REG_CITY_NOT_LIVE_CITY, dtype: float64 

ValueCounts: REG_CITY_NOT_LIVE_CITY
0   92.639
1    7.361
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.1727984473888169, 

Describe: count   1413701.000
mean          0.222
std           0.416
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: REG_CITY_NOT_WORK_CITY, dtype: float64 

ValueCounts: REG_CITY_NOT_WORK_CITY
0   77.785
1   22.215
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.14376331213040563, 

Describe: count   1413701.000
mean          0.174
std           0.379
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: LIVE_CITY_NOT_WORK_CITY, dtype: float64 

ValueCounts: LIVE_CITY_NOT_WORK_CITY
0   82.594
1   17.406
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Business Entity Type 3
ValueCounts: ORGANIZATION_TYPE
Business Entity Type 3   21.567
XNA                      19.352
Self-employed            12.952
Other                     5.201
Medicine                  3.684
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.2858978721410488
Median   : 0.5630476928116712, 
Variance: 0.03648315730123363, 

Describe: count   1413701.000
mean          0.511
std           0.191
min           0.000
25%           0.389
50%           0.563
75%           0.661
max           0.855
Name: EXT_SOURCE_2, dtype: float64 

ValueCounts: EXT_SOURCE_2
0.286   0.199
0.262   0.143
0.566   0.129
0.265   0.120
0.265   0.111
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.5352762504724826
Median   : 0.5352762504724826, 
Variance: 0.03231313722977966, 

Describe: count   1413701.000
mean          0.497
std           0.180
min           0.001
25%           0.381
50%           0.535
75%           0.621
max           0.896
Name: EXT_SOURCE_3, dtype: float64 

ValueCounts: EXT_SOURCE_3
0.535   17.362
0.746    0.406
0.671    0.400
0.555    0.391
0.694    0.367
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.10254666268544127
Median   : 0.10254666268544127, 
Variance: 0.005621824784519625, 

Describe: count   1413701.000
mean          0.101
std           0.075
min           0.000
25%           0.066
50%           0.103
75%           0.103
max           1.000
Name: TOTALAREA_MODE, dtype: float64 

ValueCounts: TOTALAREA_MODE
0.103   47.927
0.000    0.194
0.056    0.082
0.056    0.080
0.055    0.080
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: No

Unique: ['No' 'Yes']

ValueCounts: EMERGENCYSTATE_MODE
No    99.199
Yes    0.801
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.0

Unique: [  2.   1.   0.   4.   8.   7.   3.   6.   5.  12.   9.  10.  13.  11.
  14.  22.  16.  15.  17.  20.  25.  19.  18.  21.  24.  23.  28.  26.
  29.  27.  47. 348.  30.]

Median   : 0.0, 
Variance: 6.395559820743489, 

Describe: count   1413701.000
mean          1.541
std           2.529
min           0.000
25%           0.000
50%           0.000
75%           2.000
max         348.000
Name: OBS_30_CNT_SOCIAL_CIRCLE, dtype: float64 

ValueCounts: OBS_30_CNT_SOCIAL_CIRCLE
0.000   52.068
1.000   15.640
2.000    9.778
3.000    6.813
4.000    4.813
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.0

Unique: [ 2.  0.  1.  3.  4.  5.  6.  7. 34.  8.]

Median   : 0.0, 
Variance: 0.21662998541204448, 

Describe: count   1413701.000
mean          0.154
std           0.465
min           0.000
25%           0.000
50%           0.000
75%           0.000
max          34.000
Name: DEF_30_CNT_SOCIAL_CIRCLE, dtype: float64 

ValueCounts: DEF_30_CNT_SOCIAL_CIRCLE
0.000   87.858
1.000    9.696
2.000    1.844
3.000    0.459
4.000    0.117
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.0

Unique: [  2.   1.   0.   4.   8.   7.   3.   6.   5.  12.   9.  10.  13.  11.
  14.  21.  15.  22.  16.  20.  25.  17.  19.  18.  24.  23.  28.  29.
  27.  47. 344.  30.  26.]

Median   : 0.0, 
Variance: 6.286009401273656, 

Describe: count   1413701.000
mean          1.523
std           2.507
min           0.000
25%           0.000
50%           0.000
75%           2.000
max         344.000
Name: OBS_60_CNT_SOCIAL_CIRCLE, dtype: float64 

ValueCounts: OBS_60_CNT_SOCIAL_CIRCLE
0.000   52.301
1.000   15.677
2.000    9.776
3.000    6.778
4.000    4.756
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.0

Unique: [ 2.  0.  1.  3.  5.  4.  7. 24.  6.]

Median   : 0.0, 
Variance: 0.1433917755687901, 

Describe: count   1413701.000
mean          0.108
std           0.379
min           0.000
25%           0.000
50%           0.000
75%           0.000
max          24.000
Name: DEF_60_CNT_SOCIAL_CIRCLE, dtype: float64 

ValueCounts: DEF_60_CNT_SOCIAL_CIRCLE
0.000   91.042
1.000    7.527
2.000    1.127
3.000    0.229
4.000    0.065
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [ 3  2  1  6  4  0  7  5  8  9 10 11]

Median   : 2.0, 
Variance: 4.562196038297019, 

Describe: count   1413701.000
mean          2.536
std           2.136
min           0.000
25%           1.000
50%           2.000
75%           4.000
max          11.000
Name: YEARS_LAST_PHONE_CHANGE, dtype: float64 

ValueCounts: YEARS_LAST_PHONE_CHANGE
0   23.541
4   16.586
1   16.458
2   12.767
3   11.880
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 7.07313619368962e-05, 

Describe: count   1413701.000
mean          0.000
std           0.008
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: FLAG_DOCUMENT_2, dtype: float64 

ValueCounts: FLAG_DOCUMENT_2
0   99.993
1    0.007
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 1

Unique: [1 0]

Median   : 1.0, 
Variance: 0.19308926654498165, 

Describe: count   1413701.000
mean          0.739
std           0.439
min           0.000
25%           0.000
50%           1.000
75%           1.000
max           1.000
Name: FLAG_DOCUMENT_3, dtype: float64 

ValueCounts: FLAG_DOCUMENT_3
1   73.856
0   26.144
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 7.638943857510745e-05, 

Describe: count   1413701.000
mean          0.000
std           0.009
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: FLAG_DOCUMENT_4, dtype: float64 

ValueCounts: FLAG_DOCUMENT_4
0   99.992
1    0.008
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.01350839549939064, 

Describe: count   1413701.000
mean          0.014
std           0.116
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: FLAG_DOCUMENT_5, dtype: float64 

ValueCounts: FLAG_DOCUMENT_5
0   98.630
1    1.370
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.08142636131957906, 

Describe: count   1413701.000
mean          0.089
std           0.285
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: FLAG_DOCUMENT_6, dtype: float64 

ValueCounts: FLAG_DOCUMENT_6
0   91.058
1    8.942
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.0002128711269718906, 

Describe: count   1413701.000
mean          0.000
std           0.015
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: FLAG_DOCUMENT_7, dtype: float64 

ValueCounts: FLAG_DOCUMENT_7
0   99.979
1    0.021
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.06888615057472505, 

Describe: count   1413701.000
mean          0.074
std           0.262
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: FLAG_DOCUMENT_8, dtype: float64 

ValueCounts: FLAG_DOCUMENT_8
0   92.557
1    7.443
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.003172301239052705, 

Describe: count   1413701.000
mean          0.003
std           0.056
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: FLAG_DOCUMENT_9, dtype: float64 

ValueCounts: FLAG_DOCUMENT_9
0   99.682
1    0.318
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 5.587860603198018e-05, 

Describe: count   1413701.000
mean          0.000
std           0.007
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: FLAG_DOCUMENT_10, dtype: float64 

ValueCounts: FLAG_DOCUMENT_10
0   99.994
1    0.006
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.001699725391659478, 

Describe: count   1413701.000
mean          0.002
std           0.041
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: FLAG_DOCUMENT_11, dtype: float64 

ValueCounts: FLAG_DOCUMENT_11
0   99.830
1    0.170
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 1.414725309908651e-06, 

Describe: count   1413701.000
mean          0.000
std           0.001
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: FLAG_DOCUMENT_12, dtype: float64 

ValueCounts: FLAG_DOCUMENT_12
0   100.000
1     0.000
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.001780788588738822, 

Describe: count   1413701.000
mean          0.002
std           0.042
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: FLAG_DOCUMENT_13, dtype: float64 

ValueCounts: FLAG_DOCUMENT_13
0   99.822
1    0.178
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.0014268329032509246, 

Describe: count   1413701.000
mean          0.001
std           0.038
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: FLAG_DOCUMENT_14, dtype: float64 

ValueCounts: FLAG_DOCUMENT_14
0   99.857
1    0.143
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.0006567090023151083, 

Describe: count   1413701.000
mean          0.001
std           0.026
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: FLAG_DOCUMENT_15, dtype: float64 

ValueCounts: FLAG_DOCUMENT_15
0   99.934
1    0.066
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.005970150089362391, 

Describe: count   1413701.000
mean          0.006
std           0.077
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: FLAG_DOCUMENT_16, dtype: float64 

ValueCounts: FLAG_DOCUMENT_16
0   99.399
1    0.601
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.00014215987961972622, 

Describe: count   1413701.000
mean          0.000
std           0.012
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: FLAG_DOCUMENT_17, dtype: float64 

ValueCounts: FLAG_DOCUMENT_17
0   99.986
1    0.014
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.0041784741202988175, 

Describe: count   1413701.000
mean          0.004
std           0.065
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: FLAG_DOCUMENT_18, dtype: float64 

ValueCounts: FLAG_DOCUMENT_18
0   99.580
1    0.420
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.0003987939383243309, 

Describe: count   1413701.000
mean          0.000
std           0.020
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: FLAG_DOCUMENT_19, dtype: float64 

ValueCounts: FLAG_DOCUMENT_19
0   99.960
1    0.040
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.00036698695901663507, 

Describe: count   1413701.000
mean          0.000
std           0.019
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: FLAG_DOCUMENT_20, dtype: float64 

ValueCounts: FLAG_DOCUMENT_20
0   99.963
1    0.037
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0

Unique: [0 1]

Median   : 0.0, 
Variance: 0.00022418399786824068, 

Describe: count   1413701.000
mean          0.000
std           0.015
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: FLAG_DOCUMENT_21, dtype: float64 

ValueCounts: FLAG_DOCUMENT_21
0   99.978
1    0.022
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.0

Unique: [0. 1. 2. 3. 4.]

Median   : 0.0, 
Variance: 0.005932991534707245, 

Describe: count   1413701.000
mean          0.005
std           0.077
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           4.000
Name: AMT_REQ_CREDIT_BUREAU_HOUR, dtype: float64 

ValueCounts: AMT_REQ_CREDIT_BUREAU_HOUR
0.000   99.473
1.000    0.508
2.000    0.017
3.000    0.002
4.000    0.000
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.0

Unique: [0. 1. 3. 2. 4. 5. 6. 9.]

Median   : 0.0, 
Variance: 0.010039349307693738, 

Describe: count   1413701.000
mean          0.006
std           0.100
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           9.000
Name: AMT_REQ_CREDIT_BUREAU_DAY, dtype: float64 

ValueCounts: AMT_REQ_CREDIT_BUREAU_DAY
0.000   99.495
1.000    0.455
2.000    0.027
3.000    0.012
4.000    0.006
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.0

Unique: [0. 1. 3. 2. 4. 5. 6. 8. 7.]

Median   : 0.0, 
Variance: 0.04051775033978972, 

Describe: count   1413701.000
mean          0.034
std           0.201
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           8.000
Name: AMT_REQ_CREDIT_BUREAU_WEEK, dtype: float64 

ValueCounts: AMT_REQ_CREDIT_BUREAU_WEEK
0.000   96.782
1.000    3.110
2.000    0.070
3.000    0.017
4.000    0.011
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.0

Unique: [ 0.  1.  2.  5.  3.  7.  9.  4. 11.  8.  6. 16. 12. 14. 10. 13. 17. 24.
 19. 15. 23. 18. 27. 22.]

Median   : 0.0, 
Variance: 0.8590375176640723, 

Describe: count   1413701.000
mean          0.266
std           0.927
min           0.000
25%           0.000
50%           0.000
75%           0.000
max          27.000
Name: AMT_REQ_CREDIT_BUREAU_MON, dtype: float64 

ValueCounts: AMT_REQ_CREDIT_BUREAU_MON
0.000   83.704
1.000   12.312
2.000    2.036
3.000    0.737
4.000    0.393
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.0

Unique: [  0.   1.   2.   4.   3.   8.   5.   6.   7. 261.]

Median   : 0.0, 
Variance: 0.7711375890103405, 

Describe: count   1413701.000
mean          0.320
std           0.878
min           0.000
25%           0.000
50%           0.000
75%           0.000
max         261.000
Name: AMT_REQ_CREDIT_BUREAU_QRT, dtype: float64 

ValueCounts: AMT_REQ_CREDIT_BUREAU_QRT
0.000   78.751
1.000   12.597
2.000    7.243
3.000    0.977
4.000    0.347
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.0

Unique: [ 1.  0.  2.  4.  5.  3.  8.  6.  9.  7. 10. 11. 13. 16. 12. 25. 23. 15.
 14. 22. 17. 19. 18. 21. 20.]

Median   : 2.0, 
Variance: 5.147256027837513, 

Describe: count   1413701.000
mean          2.575
std           2.269
min           0.000
25%           1.000
50%           2.000
75%           4.000
max          25.000
Name: AMT_REQ_CREDIT_BUREAU_YEAR, dtype: float64 

ValueCounts: AMT_REQ_CREDIT_BUREAU_YEAR
0.000   22.540
2.000   16.713
1.000   15.606
3.000   14.889
4.000   10.984
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.1667
Median   : 0.1667, 
Variance: 0.010712743464371973, 

Describe: count   1413701.000
mean          0.194
std           0.104
min           0.000
25%           0.167
50%           0.167
75%           0.167
max           1.000
Name: FLOORSMAX, dtype: float64 

ValueCounts: FLOORSMAX
0.167   69.894
0.333   10.404
0.042    4.874
0.375    2.500
0.125    2.266
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.9775174983915752
Median   : 0.9775174983915752, 
Variance: 0.002071626816730406, 

Describe: count   1413701.000
mean          0.977
std           0.046
min           0.000
25%           0.978
50%           0.978
75%           0.982
max           1.000
Name: YEARS_BEGINEXPLOITATION, dtype: float64 

ValueCounts: YEARS_BEGINEXPLOITATION
0.978   48.432
0.987    1.319
0.986    1.270
0.980    1.261
0.987    1.256
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 200-400k

Unique: ['400-600k', '1M+', '0-200K', '200-400k', '600-800k', '800-1M']
Categories (6, object): ['0-200K' < '200-400k' < '400-600k' < '600-800k' < '800-1M' < '1M+']

ValueCounts: AMT_CREDIT_BINS
200-400k   26.851
400-600k   22.229
1M+        15.415
600-800k   14.226
0-200K     11.473
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0-10

Unique: ['0-10', '60+', '20-30', '10-20', '30-40', '40-50']
Categories (7, object): ['0-10' < '10-20' < '20-30' < '30-40' < '40-50' < '50-60' < '60+']

ValueCounts: YEARS_EMPLOYED_BINS
0-10    66.419
60+     19.352
10-20   10.768
20-30    2.647
30-40    0.758
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 30-40

Unique: ['<30', '40-50', '50-60', '30-40', '60+']
Categories (5, object): ['<30' < '30-40' < '40-50' < '50-60' < '60+']

ValueCounts: AGE_Category
30-40   26.579
40-50   24.657
50-60   22.668
<30     15.256
60+     10.840
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 1000001
Median   : 1922698.0, 
Variance: 283785628415.63196, 

Describe: count   1413701.000
mean    1922744.331
std      532715.335
min     1000001.000
25%     1461346.000
50%     1922698.000
75%     2384012.000
max     2845381.000
Name: SK_ID_PREV, dtype: float64 

ValueCounts: SK_ID_PREV
1038818   0.000
2734700   0.000
2622697   0.000
1634932   0.000
2177866   0.000
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Cash loans

Unique: ['Consumer loans' 'Cash loans' 'Revolving loans' 'XNA']

ValueCounts: NAME_CONTRACT_TYPE_right
Cash loans        44.335
Consumer loans    44.228
Revolving loans   11.415
XNA                0.022
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 11250.0
Median   : 11250.0, 
Variance: 173283112.1855973, 

Describe: count   1413701.000
mean      14840.322
std       13163.704
min           0.000
25%        7406.055
50%       11250.000
75%       16747.965
max      418058.145
Name: AMT_ANNUITY_right, dtype: float64 

ValueCounts: AMT_ANNUITY_right
11250.000   22.583
2250.000     1.865
6750.000     0.818
9000.000     0.750
22500.000    0.718
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.0
Median   : 70870.5, 
Variance: 86213992304.8011, 

Describe: count   1413701.000
mean     175243.594
std      293622.193
min           0.000
25%       19750.500
50%       70870.500
75%      180000.000
max     5850000.000
Name: AMT_APPLICATION, dtype: float64 

ValueCounts: AMT_APPLICATION
0.000        23.011
45000.000     2.836
225000.000    2.612
135000.000    2.469
450000.000    2.320
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 1,
Mode: 0.0
Median   : 80595.0, 
Variance: 102068269559.04918, 

Describe: count   1413700.000
mean     196354.086
std      319481.251
min           0.000
25%       24880.500
50%       80595.000
75%      215640.000
max     4509688.500
Name: AMT_CREDIT_right, dtype: float64 

ValueCounts: AMT_CREDIT_right
0.000        19.564
45000.000     2.060
225000.000    1.272
450000.000    1.193
135000.000    1.142
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 112320.0
Median   : 112320.0, 
Variance: 79534684972.55249, 

Describe: count   1413701.000
mean     200655.235
std      282018.944
min           0.000
25%       66375.000
50%      112320.000
75%      180000.000
max     5850000.000
Name: AMT_GOODS_PRICE_right, dtype: float64 

ValueCounts: AMT_GOODS_PRICE_right
112320.000   22.604
45000.000     2.836
225000.000    2.612
135000.000    2.468
450000.000    2.321
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: WEDNESDAY

Unique: ['SATURDAY' 'FRIDAY' 'SUNDAY' 'THURSDAY' 'TUESDAY' 'MONDAY' 'WEDNESDAY']

ValueCounts: WEEKDAY_APPR_PROCESS_START_right
WEDNESDAY   15.245
TUESDAY     15.212
MONDAY      15.174
FRIDAY      15.093
THURSDAY    14.926
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 11

Unique: [ 9 12 17 15  5 14 13 11  8 10 18 16 19  7  6  4 21 20  3  2 22  1 23  0]

Median   : 12.0, 
Variance: 11.098717981459602, 

Describe: count   1413701.000
mean         12.479
std           3.331
min           0.000
25%          10.000
50%          12.000
75%          15.000
max          23.000
Name: HOUR_APPR_PROCESS_START_right, dtype: float64 

ValueCounts: HOUR_APPR_PROCESS_START_right
11   11.549
12   11.143
10   10.879
13   10.316
14    9.452
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Y

Unique: ['Y' 'N']

ValueCounts: FLAG_LAST_APPL_PER_CONTRACT
Y   99.483
N    0.517
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 1

Unique: [1 0]

Median   : 1.0, 
Variance: 0.003555212075985266, 

Describe: count   1413701.000
mean          0.996
std           0.060
min           0.000
25%           1.000
50%           1.000
75%           1.000
max           1.000
Name: NFLAG_LAST_APPL_IN_DAY, dtype: float64 

ValueCounts: NFLAG_LAST_APPL_IN_DAY
1   99.643
0    0.357
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: XAP

Unique: ['XAP' 'XNA' 'Other' 'Payments on other loans' 'Buying a used car'
 'Repairs' 'Education' 'Buying a new car' 'Everyday expenses' 'Medicine'
 'Car repairs' 'Urgent needs' 'Buying a holiday home / land'
 'Building a house or an annex' 'Furniture' 'Journey'
 'Purchase of electronic equipment' 'Wedding / gift / holiday'
 'Buying a home' 'Business development' 'Gasification / water supply'
 'Buying a garage' 'Hobby' 'Money for a third person'
 'Refusal to name the goal']

ValueCounts: NAME_CASH_LOAN_PURPOSE
XAP            55.665
XNA            40.132
Repairs         1.423
Other           0.950
Urgent needs    0.512
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Approved

Unique: ['Approved' 'Canceled' 'Refused' 'Unused offer']

ValueCounts: NAME_CONTRACT_STATUS
Approved       62.679
Canceled       18.352
Refused        17.358
Unused offer    1.611
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 245
Median   : 582.0, 
Variance: 613935.1991857325, 

Describe: count   1413701.000
mean        880.367
std         783.540
min           1.000
25%         271.000
50%         582.000
75%        1313.000
max        2922.000
Name: DAYS_DECISION, dtype: float64 

ValueCounts: DAYS_DECISION
245   0.154
238   0.153
210   0.149
224   0.146
252   0.145
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Cash through the bank

Unique: ['XNA' 'Cash through the bank' 'Non-cash from your account'
 'Cashless from the account of the employer']

ValueCounts: NAME_PAYMENT_TYPE
Cash through the bank                       62.439
XNA                                         36.992
Non-cash from your account                   0.502
Cashless from the account of the employer    0.067
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: XAP

Unique: ['XAP' 'LIMIT' 'HC' 'SCO' 'SCOFR' 'VERIF' 'CLIENT' 'XNA' 'SYSTEM']

ValueCounts: CODE_REJECT_REASON
XAP      81.031
HC       10.326
LIMIT     3.379
SCO       2.309
CLIENT    1.611
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Unaccompanied

Unique: ['Unaccompanied' 'Family' 'Spouse, partner' 'Children' 'Other_A' 'Other_B'
 'Group of people']

ValueCounts: NAME_TYPE_SUITE_right
Unaccompanied     79.388
Family            12.866
Spouse, partner    4.073
Children           1.917
Other_B            1.066
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Repeater

Unique: ['New' 'Repeater' 'Refreshed' 'XNA']

ValueCounts: NAME_CLIENT_TYPE
Repeater    73.402
New         18.359
Refreshed    8.130
XNA          0.109
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: XNA

Unique: ['Vehicles' 'XNA' 'Furniture' 'Consumer Electronics' 'Mobile'
 'Audio/Video' 'Construction Materials' 'Gardening'
 'Photo / Cinema Equipment' 'Computers' 'Clothing and Accessories'
 'Homewares' 'Medical Supplies' 'Other' 'Jewelry' 'Office Appliances'
 'Tourism' 'Auto Accessories' 'Sport and Leisure' 'Medicine' 'Weapon'
 'Direct Sales' 'Fitness' 'Insurance' 'Additional Service' 'Education'
 'Animals']

ValueCounts: NAME_GOODS_CATEGORY
XNA                    56.392
Mobile                 13.705
Consumer Electronics    7.412
Computers               6.361
Audio/Video             6.048
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: POS

Unique: ['POS' 'Cash' 'XNA' 'Cards' 'Cars']

ValueCounts: NAME_PORTFOLIO
POS     41.908
Cash    27.705
XNA     21.731
Cards    8.629
Cars     0.027
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: XNA

Unique: ['XNA' 'x-sell' 'walk-in']

ValueCounts: NAME_PRODUCT_TYPE
XNA       63.666
x-sell    27.261
walk-in    9.072
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Credit and cash offices

Unique: ['Stone' 'Credit and cash offices' 'Country-wide' 'Regional / Local'
 'AP+ (Cash loan)' 'Contact center' 'Channel of corporate sales'
 'Car dealer']

ValueCounts: CHANNEL_TYPE
Credit and cash offices   42.466
Country-wide              29.926
Stone                     12.981
Regional / Local           6.541
Contact center             4.166
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: -1
Median   : 4.0, 
Variance: 59214281.81824192, 

Describe: count   1413701.000
mean        314.988
std        7695.082
min          -1.000
25%          -1.000
50%           4.000
75%          85.000
max     4000000.000
Name: SELLERPLACE_AREA, dtype: float64 

ValueCounts: SELLERPLACE_AREA
-1    45.030
 0     3.585
 50    2.242
 30    2.100
 20    2.057
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: XNA

Unique: ['Auto technology' 'XNA' 'Furniture' 'Consumer electronics' 'Connectivity'
 'Construction' 'Clothing' 'Industry' 'Tourism' 'Jewelry' 'MLM partners']

ValueCounts: NAME_SELLER_INDUSTRY
XNA                    50.599
Consumer electronics   24.171
Connectivity           16.860
Furniture               3.464
Construction            1.798
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 12.0

Unique: [24. 12.  6.  4. 48.  0. 18. 10. 30.  8. 16. 36.  5. 60. 42. 14. 54. 20.
  3. 15. 72.  9.  7. 11. 22. 17. 84. 13. 29. 59. 53. 40. 23. 26. 28. 32.
 34. 38. 19. 66. 35. 33. 39. 44. 21. 47. 41. 45.]

Median   : 12.0, 
Variance: 168.87129923304957, 

Describe: count   1413701.000
mean         15.171
std          12.995
min           0.000
25%          10.000
50%          12.000
75%          18.000
max          84.000
Name: CNT_PAYMENT, dtype: float64 

ValueCounts: CNT_PAYMENT
12.000   41.202
6.000    11.513
10.000    8.668
0.000     8.629
24.000    8.260
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: XNA

Unique: ['low_normal' 'middle' 'XNA' 'high' 'low_action']

ValueCounts: NAME_YIELD_GROUP
XNA          30.360
middle       22.987
high         21.650
low_normal   19.444
low_action    5.559
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: Cash

Unique: ['POS other with interest' 'Cash X-Sell: low' 'POS industry with interest'
 'POS household with interest' 'POS mobile without interest' 'Card Street'
 'Card X-Sell' 'Cash X-Sell: high' 'Cash' 'Cash Street: high'
 'Cash X-Sell: middle' 'POS mobile with interest'
 'POS household without interest' 'POS industry without interest'
 'Cash Street: low' 'Cash Street: middle' 'POS others without interest']

ValueCounts: PRODUCT_COMBINATION
Cash                          16.652
POS household with interest   16.019
POS mobile with interest      13.500
Cash X-Sell: middle            8.491
Cash X-Sell: low               7.823
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 365243
Median   : 365243.0, 
Variance: 4829581043.088714, 

Describe: count   1413701.000
mean     351460.352
std       69495.187
min           2.000
25%      365243.000
50%      365243.000
75%      365243.000
max      365243.000
Name: DAYS_FIRST_DRAWING, dtype: float64 

ValueCounts: DAYS_FIRST_DRAWING
365243   96.216
228       0.008
224       0.008
212       0.008
220       0.008
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 831
Median   : 831.0, 
Variance: 3111196020.1185093, 

Describe: count   1413701.000
mean       9743.652
std       55778.096
min           2.000
25%         738.000
50%         831.000
75%        1146.000
max      365243.000
Name: DAYS_FIRST_DUE, dtype: float64 

ValueCounts: DAYS_FIRST_DUE
831      39.724
365243    2.402
334       0.047
299       0.046
208       0.046
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 361
Median   : 361.0, 
Variance: 7021672901.056603, 

Describe: count   1413701.000
mean      21052.263
std       83795.423
min           0.000
25%         361.000
50%         361.000
75%        1010.000
max      365243.000
Name: DAYS_LAST_DUE_1ST_VERSION, dtype: float64 

ValueCounts: DAYS_LAST_DUE_1ST_VERSION
361      39.733
365243    5.595
1         0.082
8         0.081
10        0.081
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 537
Median   : 537.0, 
Variance: 14814814480.760283, 

Describe: count   1413701.000
mean      47395.172
std      121716.123
min           2.000
25%         537.000
50%         537.000
75%        1542.000
max      365243.000
Name: DAYS_LAST_DUE, dtype: float64 

ValueCounts: DAYS_LAST_DUE
537      39.720
365243   12.789
239       0.040
188       0.040
245       0.040
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 499
Median   : 499.0, 
Variance: 15729348068.356016, 

Describe: count   1413701.000
mean      50774.791
std      125416.698
min           2.000
25%         499.000
50%         499.000
75%        1570.000
max      365243.000
Name: DAYS_TERMINATION, dtype: float64 

ValueCounts: DAYS_TERMINATION
499      39.721
365243   13.723
156       0.046
233       0.046
170       0.045
Name: proportion, dtype: float64 



------------------------------------------------------------------
Total Nulls: 0,
Mode: 0.0

Unique: [0. 1.]

Median   : 0.0, 
Variance: 0.15961934771683145, 

Describe: count   1413701.000
mean          0.199
std           0.400
min           0.000
25%           0.000
50%           0.000
75%           0.000
max           1.000
Name: NFLAG_INSURED_ON_APPROVAL, dtype: float64 

ValueCounts: NFLAG_INSURED_ON_APPROVAL
0.000   80.063
1.000   19.937
Name: proportion, dtype: float64 



------------------------------------------------------------------

AMT_CREDIT_right                   0.000
TARGET                             0.000
AMT_ANNUITY_right                  0.000
NAME_CONTRACT_TYPE_right           0.000
SK_ID_PREV                         0.000
AGE_Category                       0.000
YEARS_EMPLOYED_BINS                0.000
AMT_CREDIT_BINS                    0.000
YEARS_BEGINEXPLOITATION            0.000
FLOORSMAX                          0.000
AMT_REQ_CREDIT_BUREAU_YEAR         0.000
AMT_REQ_CREDIT_BUREAU_QRT          0.000
AMT_REQ_CREDIT_BUREAU_MON          0.000
AMT_REQ_CREDIT_BUREAU_WEEK         0.000
AMT_REQ_CREDIT_BUREAU_DAY          0.000
AMT_REQ_CREDIT_BUREAU_HOUR         0.000
FLAG_DOCUMENT_21                   0.000
FLAG_DOCUMENT_20                   0.000
FLAG_DOCUMENT_19                   0.000
FLAG_DOCUMENT_18                   0.000
FLAG_DOCUMENT_17                   0.000
FLAG_DOCUMENT_16                   0.000
FLAG_DOCUMENT_15                   0.000
FLAG_DOCUMENT_14                   0.000
FLAG_DOCUMENT_13                   0.000
FLAG_DOCUMENT_12                   0.000
FLAG_DOCUMENT_11                   0.000
AMT_APPLICATION                    0.000
SK_ID_CURR                         0.000
AMT_GOODS_PRICE_right              0.000
NAME_PRODUCT_TYPE                  0.000
DAYS_TERMINATION                   0.000
DAYS_LAST_DUE                      0.000
DAYS_LAST_DUE_1ST_VERSION          0.000
DAYS_FIRST_DUE                     0.000
DAYS_FIRST_DRAWING                 0.000
PRODUCT_COMBINATION                0.000
NAME_YIELD_GROUP                   0.000
CNT_PAYMENT                        0.000
NAME_SELLER_INDUSTRY               0.000
SELLERPLACE_AREA                   0.000
CHANNEL_TYPE                       0.000
NAME_PORTFOLIO                     0.000
WEEKDAY_APPR_PROCESS_START_right   0.000
NAME_GOODS_CATEGORY                0.000
NAME_CLIENT_TYPE                   0.000
NAME_TYPE_SUITE_right              0.000
CODE_REJECT_REASON                 0.000
NAME_PAYMENT_TYPE                  0.000
DAYS_DECISION                      0.000
NAME_CONTRACT_STATUS               0.000
NAME_CASH_LOAN_PURPOSE             0.000
NFLAG_LAST_APPL_IN_DAY             0.000
FLAG_LAST_APPL_PER_CONTRACT        0.000
HOUR_APPR_PROCESS_START_right      0.000
FLAG_DOCUMENT_10                   0.000
FLAG_DOCUMENT_9                    0.000
FLAG_DOCUMENT_8                    0.000
NAME_FAMILY_STATUS                 0.000
FLAG_PHONE                         0.000
FLAG_CONT_MOBILE                   0.000
FLAG_WORK_PHONE                    0.000
FLAG_EMP_PHONE                     0.000
FLAG_MOBIL                         0.000
YEARS_ID_PUBLISH                   0.000
YEARS_REGISTRATION                 0.000
YEARS_EMPLOYED                     0.000
YEARS_BIRTH                        0.000
REGION_POPULATION_RELATIVE         0.000
NAME_HOUSING_TYPE                  0.000
NAME_EDUCATION_TYPE                0.000
OCCUPATION_TYPE                    0.000
NAME_INCOME_TYPE                   0.000
NAME_TYPE_SUITE                    0.000
AMT_GOODS_PRICE                    0.000
AMT_ANNUITY                        0.000
AMT_CREDIT                         0.000
AMT_INCOME_TOTAL                   0.000
CNT_CHILDREN                       0.000
FLAG_OWN_REALTY                    0.000
FLAG_OWN_CAR                       0.000
CODE_GENDER                        0.000
NAME_CONTRACT_TYPE                 0.000
FLAG_EMAIL                         0.000
CNT_FAM_MEMBERS                    0.000
FLAG_DOCUMENT_7                    0.000
TOTALAREA_MODE                     0.000
FLAG_DOCUMENT_6                    0.000
FLAG_DOCUMENT_5                    0.000
FLAG_DOCUMENT_4                    0.000
FLAG_DOCUMENT_3                    0.000
FLAG_DOCUMENT_2                    0.000
YEARS_LAST_PHONE_CHANGE            0.000
DEF_60_CNT_SOCIAL_CIRCLE           0.000
OBS_60_CNT_SOCIAL_CIRCLE           0.000
DEF_30_CNT_SOCIAL_CIRCLE           0.000
OBS_30_CNT_SOCIAL_CIRCLE           0.000
EMERGENCYSTATE_MODE                0.000
EXT_SOURCE_3                       0.000
REGION_RATING_CLIENT               0.000
EXT_SOURCE_2                       0.000
ORGANIZATION_TYPE                  0.000
LIVE_CITY_NOT_WORK_CITY            0.000
REG_CITY_NOT_WORK_CITY             0.000
REG_CITY_NOT_LIVE_CITY             0.000
LIVE_REGION_NOT_WORK_REGION        0.000
REG_REGION_NOT_WORK_REGION         0.000
REG_REGION_NOT_LIVE_REGION         0.000
HOUR_APPR_PROCESS_START            0.000
WEEKDAY_APPR_PROCESS_START         0.000
REGION_RATING_CLIENT_W_CITY        0.000
NFLAG_INSURED_ON_APPROVAL          0.000
dtype: float64

# combined_desc_cols = combined_appl_data.describe().columns
# dtype_dict = classify_feature_dtype(combined_appl_data, combined_desc_cols)
combined_appl_data.iloc[(combined_appl_data.SK_ID_CURR.value_counts() > 1).index, :].sort_index()

	SK_ID_CURR	TARGET	NAME_CONTRACT_TYPE	CODE_GENDER	FLAG_OWN_CAR	FLAG_OWN_REALTY	CNT_CHILDREN	AMT_INCOME_TOTAL	AMT_CREDIT	AMT_ANNUITY	AMT_GOODS_PRICE	NAME_TYPE_SUITE	NAME_INCOME_TYPE	NAME_EDUCATION_TYPE	NAME_FAMILY_STATUS	NAME_HOUSING_TYPE	REGION_POPULATION_RELATIVE	YEARS_BIRTH	YEARS_EMPLOYED	YEARS_REGISTRATION	YEARS_ID_PUBLISH	FLAG_MOBIL	FLAG_EMP_PHONE	FLAG_WORK_PHONE	FLAG_CONT_MOBILE	FLAG_PHONE	FLAG_EMAIL	OCCUPATION_TYPE	CNT_FAM_MEMBERS	REGION_RATING_CLIENT	REGION_RATING_CLIENT_W_CITY	WEEKDAY_APPR_PROCESS_START	HOUR_APPR_PROCESS_START	REG_REGION_NOT_LIVE_REGION	REG_REGION_NOT_WORK_REGION	LIVE_REGION_NOT_WORK_REGION	REG_CITY_NOT_LIVE_CITY	REG_CITY_NOT_WORK_CITY	LIVE_CITY_NOT_WORK_CITY	ORGANIZATION_TYPE	EXT_SOURCE_2	EXT_SOURCE_3	TOTALAREA_MODE	EMERGENCYSTATE_MODE	OBS_30_CNT_SOCIAL_CIRCLE	DEF_30_CNT_SOCIAL_CIRCLE	OBS_60_CNT_SOCIAL_CIRCLE	DEF_60_CNT_SOCIAL_CIRCLE	YEARS_LAST_PHONE_CHANGE	FLAG_DOCUMENT_2	FLAG_DOCUMENT_3	FLAG_DOCUMENT_4	FLAG_DOCUMENT_5	FLAG_DOCUMENT_6	FLAG_DOCUMENT_7	FLAG_DOCUMENT_8	FLAG_DOCUMENT_9	FLAG_DOCUMENT_10	FLAG_DOCUMENT_11	FLAG_DOCUMENT_12	FLAG_DOCUMENT_13	FLAG_DOCUMENT_14	FLAG_DOCUMENT_15	FLAG_DOCUMENT_16	FLAG_DOCUMENT_17	FLAG_DOCUMENT_18	FLAG_DOCUMENT_19	FLAG_DOCUMENT_20	FLAG_DOCUMENT_21	AMT_REQ_CREDIT_BUREAU_HOUR	AMT_REQ_CREDIT_BUREAU_DAY	AMT_REQ_CREDIT_BUREAU_WEEK	AMT_REQ_CREDIT_BUREAU_MON	AMT_REQ_CREDIT_BUREAU_QRT	AMT_REQ_CREDIT_BUREAU_YEAR	FLOORSMAX	YEARS_BEGINEXPLOITATION	AMT_CREDIT_BINS	YEARS_EMPLOYED_BINS	AGE_Category	SK_ID_PREV	NAME_CONTRACT_TYPE_right	AMT_ANNUITY_right	AMT_APPLICATION	AMT_CREDIT_right	AMT_GOODS_PRICE_right	WEEKDAY_APPR_PROCESS_START_right	HOUR_APPR_PROCESS_START_right	FLAG_LAST_APPL_PER_CONTRACT	NFLAG_LAST_APPL_IN_DAY	NAME_CASH_LOAN_PURPOSE	NAME_CONTRACT_STATUS	DAYS_DECISION	NAME_PAYMENT_TYPE	CODE_REJECT_REASON	NAME_TYPE_SUITE_right	NAME_CLIENT_TYPE	NAME_GOODS_CATEGORY	NAME_PORTFOLIO	NAME_PRODUCT_TYPE	CHANNEL_TYPE	SELLERPLACE_AREA	NAME_SELLER_INDUSTRY	CNT_PAYMENT	NAME_YIELD_GROUP	PRODUCT_COMBINATION	DAYS_FIRST_DRAWING	DAYS_FIRST_DUE	DAYS_LAST_DUE_1ST_VERSION	DAYS_LAST_DUE	DAYS_TERMINATION	NFLAG_INSURED_ON_APPROVAL
100002	125444	1	Cash loans	F	No	Yes	0	180000.000	1350000.000	39465.000	1350000.000	Family	Pensioner	Secondary	Married	House	0.031	55	1000	29	10	1	0	0	1	0	0	Pensioner	2.000	2	2	SATURDAY	8	0	0	0	0	0	0	XNA	0.722	0.521	0.103	No	1.000	0.000	1.000	0.000	6	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	0.000	0.167	0.978	1M+	60+	50-60	1678168	Consumer loans	13229.460	53950.500	46435.500	53950.500	SUNDAY	15	Y	1	XAP	Approved	2512	Cash through the bank	XAP	Family	Repeater	Mobile	POS	XNA	Country-wide	212	Connectivity	4.000	low_normal	POS mobile with interest	365243	2481	2391	2391	2386	1.000
100003	125445	0	Cash loans	F	No	Yes	1	72000.000	354276.000	20466.000	292500.000	Unaccompanied	Working	Secondary	Civil marriage	House	0.009	43	3	17	10	1	1	0	1	0	0	Cooking staff	3.000	2	2	WEDNESDAY	13	0	0	0	1	1	0	Business Entity Type 3	0.559	0.585	0.083	No	1.000	0.000	1.000	0.000	6	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	1.000	2.000	0.167	0.972	200-400k	0-10	40-50	1217532	Consumer loans	2358.990	18391.500	20214.000	18391.500	WEDNESDAY	10	Y	1	XAP	Approved	2443	Cash through the bank	XAP	Other_B	New	Mobile	POS	XNA	Country-wide	61	Connectivity	12.000	high	POS mobile with interest	365243	2412	2082	2112	2110	1.000
100004	125445	0	Cash loans	F	No	Yes	1	72000.000	354276.000	20466.000	292500.000	Unaccompanied	Working	Secondary	Civil marriage	House	0.009	43	3	17	10	1	1	0	1	0	0	Cooking staff	3.000	2	2	WEDNESDAY	13	0	0	0	1	1	0	Business Entity Type 3	0.559	0.585	0.083	No	1.000	0.000	1.000	0.000	6	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	1.000	2.000	0.167	0.972	200-400k	0-10	40-50	1443027	Cash loans	10332.720	90000.000	95940.000	90000.000	WEDNESDAY	13	Y	1	XNA	Approved	1022	Cash through the bank	XAP	Unaccompanied	Repeater	XNA	Cash	x-sell	Credit and cash offices	-1	XNA	12.000	middle	Cash X-Sell: middle	365243	992	662	752	744	1.000
100006	125445	0	Cash loans	F	No	Yes	1	72000.000	354276.000	20466.000	292500.000	Unaccompanied	Working	Secondary	Civil marriage	House	0.009	43	3	17	10	1	1	0	1	0	0	Cooking staff	3.000	2	2	WEDNESDAY	13	0	0	0	1	1	0	Business Entity Type 3	0.559	0.585	0.083	No	1.000	0.000	1.000	0.000	6	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	1.000	2.000	0.167	0.972	200-400k	0-10	40-50	2833540	Consumer loans	3276.180	14580.000	15300.000	14580.000	SATURDAY	12	Y	1	XAP	Approved	1390	Cash through the bank	XAP	Children	Repeater	Photo / Cinema Equipment	POS	XNA	Country-wide	51	Connectivity	6.000	high	POS mobile with interest	365243	1359	1209	1209	1200	0.000
100007	125446	0	Cash loans	F	No	No	0	225000.000	1687266.000	62658.000	1575000.000	Family	State servant	Secondary	Married	House	0.014	59	6	27	11	1	1	0	1	0	0	Pensioner	2.000	2	2	SATURDAY	5	0	0	0	0	0	0	Other	0.675	0.691	0.103	No	1.000	1.000	1.000	1.000	3	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	1.000	0.167	0.978	1M+	0-10	50-60	1201611	Consumer loans	22873.815	249714.000	249714.000	249714.000	MONDAY	12	Y	1	XAP	Approved	1447	XNA	XAP	Family	New	Audio/Video	POS	XNA	Regional / Local	134	Consumer electronics	12.000	low_action	POS household without interest	365243	1416	1086	1086	1083	0.000
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
456251	215478	0	Cash loans	F	No	Yes	0	144000.000	948582.000	27864.000	679500.000	Unaccompanied	Working	Secondary	Separated	House	0.025	59	39	10	13	1	1	0	1	0	0	Pensioner	1.000	2	2	SATURDAY	10	0	0	0	0	0	0	Industry: type 7	0.192	0.700	0.103	No	0.000	0.000	0.000	0.000	7	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	5.000	0.167	0.978	800-1M	30-40	50-60	2103396	Consumer loans	20482.110	221985.000	199786.500	221985.000	SATURDAY	13	Y	1	XAP	Approved	427	Cash through the bank	XAP	Unaccompanied	Refreshed	Furniture	POS	XNA	Stone	50	Furniture	12.000	middle	POS industry with interest	365243	393	63	333	328	0.000
456252	215478	0	Cash loans	F	No	Yes	0	144000.000	948582.000	27864.000	679500.000	Unaccompanied	Working	Secondary	Separated	House	0.025	59	39	10	13	1	1	0	1	0	0	Pensioner	1.000	2	2	SATURDAY	10	0	0	0	0	0	0	Industry: type 7	0.192	0.700	0.103	No	0.000	0.000	0.000	0.000	7	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	5.000	0.167	0.978	800-1M	30-40	50-60	2396415	Cash loans	44178.750	454500.000	472500.000	454500.000	THURSDAY	15	Y	1	XNA	Approved	324	Cash through the bank	XAP	Unaccompanied	Repeater	XNA	Cash	x-sell	Credit and cash offices	-1	XNA	12.000	low_normal	Cash X-Sell: low	365243	831	361	537	499	0.000
456253	215478	0	Cash loans	F	No	Yes	0	144000.000	948582.000	27864.000	679500.000	Unaccompanied	Working	Secondary	Separated	House	0.025	59	39	10	13	1	1	0	1	0	0	Pensioner	1.000	2	2	SATURDAY	10	0	0	0	0	0	0	Industry: type 7	0.192	0.700	0.103	No	0.000	0.000	0.000	0.000	7	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	5.000	0.167	0.978	800-1M	30-40	50-60	1263974	Consumer loans	22313.070	139455.000	125509.500	139455.000	SATURDAY	13	Y	1	XAP	Approved	336	XNA	XAP	Unaccompanied	Repeater	Audio/Video	POS	XNA	Country-wide	261	Consumer electronics	6.000	low_normal	POS household with interest	365243	305	155	185	181	0.000
456254	215478	0	Cash loans	F	No	Yes	0	144000.000	948582.000	27864.000	679500.000	Unaccompanied	Working	Secondary	Separated	House	0.025	59	39	10	13	1	1	0	1	0	0	Pensioner	1.000	2	2	SATURDAY	10	0	0	0	0	0	0	Industry: type 7	0.192	0.700	0.103	No	0.000	0.000	0.000	0.000	7	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	5.000	0.167	0.978	800-1M	30-40	50-60	2511765	Cash loans	11250.000	0.000	0.000	112320.000	THURSDAY	15	Y	1	XNA	Canceled	324	XNA	XAP	Unaccompanied	Repeater	XNA	XNA	XNA	Credit and cash offices	-1	XNA	12.000	XNA	Cash	365243	831	361	537	499	0.000
456255	215478	0	Cash loans	F	No	Yes	0	144000.000	948582.000	27864.000	679500.000	Unaccompanied	Working	Secondary	Separated	House	0.025	59	39	10	13	1	1	0	1	0	0	Pensioner	1.000	2	2	SATURDAY	10	0	0	0	0	0	0	Industry: type 7	0.192	0.700	0.103	No	0.000	0.000	0.000	0.000	7	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.000	0.000	0.000	0.000	0.000	5.000	0.167	0.978	800-1M	30-40	50-60	2351113	Cash loans	11250.000	0.000	0.000	112320.000	WEDNESDAY	14	Y	1	XNA	Canceled	164	XNA	XAP	Unaccompanied	Repeater	XNA	XNA	XNA	Credit and cash offices	-1	XNA	12.000	XNA	Cash	365243	831	361	537	499	0.000

291057 rows × 112 columns

comb_target_train_0 = combined_appl_data.loc[combined_appl_data['TARGET'] == 0]
comb_target_train_1 = combined_appl_data.loc[combined_appl_data['TARGET'] == 1]  

combined_desc_cols = combined_appl_data.describe().columns
FLAG_COLS = combined_appl_data.filter(like="FLAG").columns
combined_desc_cols = list(set(combined_desc_cols) - set(FLAG_COLS))
combined_dtype_dict = classify_feature_dtype(combined_appl_data, combined_desc_cols)
combined_dtype_floats = combined_dtype_dict['float_ts']
combined_desc_cols

['YEARS_EMPLOYED',
 'FLOORSMAX',
 'YEARS_BEGINEXPLOITATION',
 'SELLERPLACE_AREA',
 'REG_REGION_NOT_WORK_REGION',
 'DAYS_TERMINATION',
 'YEARS_ID_PUBLISH',
 'DAYS_FIRST_DRAWING',
 'CNT_FAM_MEMBERS',
 'DEF_30_CNT_SOCIAL_CIRCLE',
 'CNT_CHILDREN',
 'CNT_PAYMENT',
 'AMT_GOODS_PRICE',
 'AMT_ANNUITY',
 'AMT_REQ_CREDIT_BUREAU_MON',
 'AMT_REQ_CREDIT_BUREAU_QRT',
 'AMT_ANNUITY_right',
 'DAYS_LAST_DUE',
 'AMT_GOODS_PRICE_right',
 'OBS_60_CNT_SOCIAL_CIRCLE',
 'AMT_REQ_CREDIT_BUREAU_DAY',
 'AMT_CREDIT',
 'DEF_60_CNT_SOCIAL_CIRCLE',
 'HOUR_APPR_PROCESS_START',
 'REG_CITY_NOT_LIVE_CITY',
 'LIVE_REGION_NOT_WORK_REGION',
 'SK_ID_CURR',
 'AMT_INCOME_TOTAL',
 'HOUR_APPR_PROCESS_START_right',
 'OBS_30_CNT_SOCIAL_CIRCLE',
 'REGION_POPULATION_RELATIVE',
 'AMT_REQ_CREDIT_BUREAU_WEEK',
 'AMT_CREDIT_right',
 'SK_ID_PREV',
 'REG_CITY_NOT_WORK_CITY',
 'DAYS_LAST_DUE_1ST_VERSION',
 'YEARS_LAST_PHONE_CHANGE',
 'TARGET',
 'REGION_RATING_CLIENT_W_CITY',
 'AMT_REQ_CREDIT_BUREAU_HOUR',
 'YEARS_BIRTH',
 'REG_REGION_NOT_LIVE_REGION',
 'AMT_REQ_CREDIT_BUREAU_YEAR',
 'AMT_APPLICATION',
 'EXT_SOURCE_3',
 'YEARS_REGISTRATION',
 'TOTALAREA_MODE',
 'DAYS_DECISION',
 'EXT_SOURCE_2',
 'LIVE_CITY_NOT_WORK_CITY',
 'REGION_RATING_CLIENT',
 'DAYS_FIRST_DUE']

# univariate_plots(combined_appl_data.head(60), combined_appl_data.select_dtypes(include=['category','object']).columns, t0 = comb_target_train_0, t1 = comb_target_train_1, ftype="category")
# print("--------------------------------------------------------------------------")
# univariate_plots(combined_appl_data.head(60), combined_dtype_dict['int_cat'], t0 = comb_target_train_0, t1 = comb_target_train_1, ftype="category")
# print("--------------------------------------------------------------------------")
# univariate_plots(combined_appl_data.head(60), combined_dtype_dict['float_ts'], ftype="non_categorical")

univariate_plots(combined_appl_data, combined_appl_data.select_dtypes(include=['category','object']).columns, t0 = comb_target_train_0, t1 = comb_target_train_1, ftype="category")
print("--------------------------------------------------------------------------")
univariate_plots(combined_appl_data, combined_dtype_dict['int_cat'], t0 = comb_target_train_0, t1 = comb_target_train_1, ftype="category")
print("--------------------------------------------------------------------------")
univariate_plots(combined_appl_data, combined_dtype_dict['float_ts'], ftype="non_categorical")

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

--------------------------------------------------------------------------

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

--------------------------------------------------------------------------

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

Bivariate Analysis

plt.figure(figsize=(8,5))
sns.countplot(x='NAME_EDUCATION_TYPE', hue='NAME_CONTRACT_STATUS', data=comb_target_train_0, palette='viridis', stat="percent")
plt.xticks(rotation=0)
plt.show()

<Figure size 800x500 with 0 Axes>

<Axes: xlabel='NAME_EDUCATION_TYPE', ylabel='percent'>

([0, 1, 2, 3, 4],
 [Text(0, 0, 'Higher education'),
  Text(1, 0, 'Secondary'),
  Text(2, 0, 'Incomplete higher'),
  Text(3, 0, 'Lower secondary'),
  Text(4, 0, 'Academic degree')])

plt.figure(figsize=(8,5))
grouped_df = (curr_appl_data1.groupby(["NAME_CONTRACT_TYPE"])["TARGET"].value_counts(normalize=True)*100).reset_index()
sns.barplot(x='NAME_CONTRACT_TYPE', y='proportion', hue='TARGET', data=grouped_df, palette='magma')
plt.xticks(rotation=0)
plt.show()

<Figure size 800x500 with 0 Axes>

<Axes: xlabel='NAME_CONTRACT_TYPE', ylabel='proportion'>

([0, 1], [Text(0, 0, 'Cash loans'), Text(1, 0, 'Revolving loans')])

# sns.pairplot(curr_appl_data1[curr_dtype_dict])
# sns.pairplot(prev_appl_data1[prev_dtype_dict])

bivariate_plots(df=combined_appl_data, col='NAME_EDUCATION_TYPE', hue='NAME_CONTRACT_STATUS', train0=comb_target_train_0, train1=comb_target_train_1 )
bivariate_plots(df=combined_appl_data, col='NAME_CLIENT_TYPE', hue='NAME_CONTRACT_STATUS', train0=comb_target_train_0, train1=comb_target_train_1 )

<Figure size 640x480 with 0 Axes>

<Figure size 640x480 with 0 Axes>

Client who are Secondary educated have mostly applied for the loan, They have the greater chance of paying their installments on time without difficulties among the given categories.

Most loan applicants are Repeaters Also Most loan applicants who do not have difficulty in paying their installments are Repeaters among the given categories.

sk_id_tg = combined_appl_data.groupby(by=['SK_ID_CURR'])['TARGET'].value_counts() # grouped all sku ids wrt target
sk_id_tg_df = sk_id_tg[sk_id_tg > 2].reset_index()

sk_id_tg_df['count_bins'] = pd.cut(sk_id_tg_df['count'], bins=[0,10,20,30,40,50,60,70], labels=['>10','10-20','20-30','30-40','40-50','50-60','60+'])

fig, axes = plt.subplots(1, 2, figsize=(20, 6));
sns.barplot(x='count_bins', y='count', hue='TARGET',  data=sk_id_tg_df, palette='magma', ax=axes[0]);
axes[0].set_title("Barplot Analysis of SK_ID's that have more than 1 application wrt TARGET"); plt.xticks(rotation=45); 
axes[0].plot([5.8], [70], marker='o', color='red',  markersize=10, linestyle='--', linewidth=2)

sns.countplot(x='count_bins', hue='TARGET',  data=sk_id_tg_df, palette='magma', ax=axes[1])
axes[1].set_title("Countplot Analysis of SK_ID's that have more than 1 application wrt TARGET"); plt.xticks(rotation=45); 
fig.subplots_adjust(wspace=0.5, hspace=0.3);

plt.tight_layout();
plt.subplots_adjust(top=0.85);
plt.show();
plt.clf();

<Figure size 640x480 with 0 Axes>

sk_id_tg[sk_id_tg > 60]
sk_id_tg[sk_id_tg > 60].count()

SK_ID_CURR  TARGET
173680      0         72
198355      0         63
206783      0         67
238250      0         61
242412      0         68
265681      0         73
280586      0         61
345161      0         62
382179      0         64
389950      0         64
446486      0         62
Name: count, dtype: int64

11

• Using the combined data (application.csv , prev_application.csv)
• Having looked at the graph, it was very surprising to see the count_bins 60+ category,
• Those sku's have no payment difficulties. there were 11 of them.
• Further analysis may give out the reason for such outcomes.

Correlation Matrix

import numpy as np
corr = comb_target_train_0[comb_target_train_0.select_dtypes(exclude=['category','object']).columns].corr()
mask = np.zeros_like(corr)
mask[np.triu_indices_from(mask)] = True
f, ax = plt.subplots(figsize=(11, 9))
with sns.axes_style("white"):
    ax = sns.heatmap(corr, mask=mask, vmax=.3, square=True)

import numpy as np
corr = comb_target_train_1[comb_target_train_1.select_dtypes(exclude=['category','object']).columns].corr()
mask = np.zeros_like(corr)
mask[np.triu_indices_from(mask)] = True
f, ax = plt.subplots(figsize=(11, 9))
with sns.axes_style("white"):
    ax = sns.heatmap(corr, mask=mask, vmax=.3, square=True)

Heatmap correlation of all numerical features It is more or less similar for both targets AMT_CREDIT is high for Yongsters 'YEARS_BIRTH' AMT_CREDIT is high for low 'CNT_CHILDREN'

Top 10 Correlations

corr_0 = curr_target_train_0.corr(numeric_only=True).abs()
corr_0 = corr_0.unstack()
correlation_0 = corr_0.sort_values()
correlation_0 = corr_0.dropna()
# correlation_0

correlation_0 = correlation_0[correlation_0 != 1.0]
correlation_target_zero = correlation_0.reset_index()
correlation_target_zero.sort_values(by=0, ascending=False).head(10)

	level_0	level_1	0
507	YEARS_EMPLOYED	FLAG_EMP_PHONE	1.000
752	FLAG_EMP_PHONE	YEARS_EMPLOYED	1.000
2014	OBS_60_CNT_SOCIAL_CIRCLE	OBS_30_CNT_SOCIAL_CIRCLE	0.999
1891	OBS_30_CNT_SOCIAL_CIRCLE	OBS_60_CNT_SOCIAL_CIRCLE	0.999
313	AMT_GOODS_PRICE	AMT_CREDIT	0.987
190	AMT_CREDIT	AMT_GOODS_PRICE	0.987
1196	REGION_RATING_CLIENT_W_CITY	REGION_RATING_CLIENT	0.950
1134	REGION_RATING_CLIENT	REGION_RATING_CLIENT_W_CITY	0.950
78	CNT_CHILDREN	CNT_FAM_MEMBERS	0.879
1055	CNT_FAM_MEMBERS	CNT_CHILDREN	0.879

corr_1 = curr_target_train_1.corr(numeric_only=True).abs()
corr_1 = corr_1.unstack()
correlation_1 = corr_1.sort_values()
correlation_1 = corr_1.dropna()
# correlation_1

correlation_1 = correlation_1[correlation_1 != 1.0]
correlation_target_one = correlation_1.reset_index()
correlation_target_one.sort_values(by=0, ascending=False).head(10)

	level_0	level_1	0
474	YEARS_EMPLOYED	FLAG_EMP_PHONE	1.000
646	FLAG_EMP_PHONE	YEARS_EMPLOYED	1.000
1712	OBS_30_CNT_SOCIAL_CIRCLE	OBS_60_CNT_SOCIAL_CIRCLE	0.998
1827	OBS_60_CNT_SOCIAL_CIRCLE	OBS_30_CNT_SOCIAL_CIRCLE	0.998
293	AMT_GOODS_PRICE	AMT_CREDIT	0.983
178	AMT_CREDIT	AMT_GOODS_PRICE	0.983
1003	REGION_RATING_CLIENT	REGION_RATING_CLIENT_W_CITY	0.957
1061	REGION_RATING_CLIENT_W_CITY	REGION_RATING_CLIENT	0.957
929	CNT_FAM_MEMBERS	CNT_CHILDREN	0.885
73	CNT_CHILDREN	CNT_FAM_MEMBERS	0.885

a = combined_appl_data[["AMT_INCOME_TOTAL","AMT_CREDIT","AMT_ANNUITY","AMT_GOODS_PRICE"]].corr()
sns.heatmap(a, annot=True, cmap="viridis")

<Axes: >

Heatmap correlation of critical features

# curr_appl_data1.to_csv('curr.csv')
# prev_appl_data1.to_csv('prev.csv')

pairplot = sns.pairplot(curr_appl_data1[['AMT_CREDIT','AMT_ANNUITY', 'AMT_GOODS_PRICE', 'CNT_FAM_MEMBERS', 'CNT_CHILDREN', 'YEARS_REGISTRATION', 'YEARS_BIRTH', 'AMT_REQ_CREDIT_BUREAU_QRT', 'AMT_REQ_CREDIT_BUREAU_YEAR']])
plt.show();
# Rotate the x-axis labels
# for ax in pairplot.axes.flat:
#     ax.set_xlabel(ax.get_xlabel(), rotation=90,  labelpad=10)
#     ax.set_ylabel(ax.get_ylabel(), rotation=0, ha='right', labelpad=10)

Some of the high linear relationships observed as below: • 'AMT_CREDIT','AMT_ANNUITY','AMT_GOODS_PRICE’, • Therefore More the price of the goods, higher the credit amount • 'CNT_FAM_MEMBERS','CNT_CHILDREN’, • 'YEARS_REGISTRATION','YEARS_BIRTH’, • 'AMT_CREDIT vs 'AMT_REQ_CREDIT_BUREAU_YEAR’,

Related Posts

• Telecom-churn-Case-Study 13 Nov 2024

• Lead-Scoring-Case-Study 13 Oct 2024

• Bike Sharing Case Study 13 Sep 2024