app.py

pip install seaborn
import pandas as pd
import numpy as np
import seaborn as sns
import matplotlib as plt
import matplotlib.pyplot as plt
from sklearn import preprocessing
from sklearn.preprocessing import LabelEncoder
import gradio as gr
from array import *


#from google.colab import drive
#drive.mount('/content/drive')

df_train = pd.read_csv("train_ctrUa4K.csv") #Reading the dataset in a dataframe using Pandas

df_train.head()

df_train.describe()

df_train.shape

df_train.info()

df_train.isnull().sum()

print(df_train['Gender'].value_counts())
print(df_train['Married'].value_counts())
print(df_train['Dependents'].value_counts())
print(df_train['Self_Employed'].value_counts())
print(df_train['Credit_History'].value_counts())
print(df_train['Property_Area'].value_counts())

df_train['Gender'].fillna("Male", inplace = True)
df_train['Married'].fillna("Yes", inplace = True)
df_train['Dependents'].fillna("0", inplace = True)
df_train['Self_Employed'].fillna("No", inplace = True)
df_train['Credit_History'].fillna(1.0, inplace = True)
df_train.isnull().sum()

duplicate=df_train.duplicated()
print(duplicate.sum())
df_train[duplicate]

fig, ax = plt.subplots(3, 2, figsize = (10, 7))
sns.boxplot(x= df_train["ApplicantIncome"], ax = ax[0,0])
sns.distplot(df_train['ApplicantIncome'], ax = ax[0,1])
sns.boxplot(x= df_train["CoapplicantIncome"], ax = ax[1,0])
sns.distplot(df_train['CoapplicantIncome'], ax = ax[1,1])
sns.boxplot(x= df_train["Loan_Amount_Term"], ax = ax[2,0])
sns.distplot(df_train['Loan_Amount_Term'], ax = ax[2,1])

def remove_outlier(col):
  sorted(col)
  Q1, Q3=col.quantile([0.25, 0.75])
  IQR=Q3-Q1
  lower_range=Q1-(1.5*IQR)
  upper_range=Q3+(1.5*IQR)
  return lower_range, upper_range

low_AI, high_AI=remove_outlier(df_train['ApplicantIncome'])
df_train['ApplicantIncome']=np.where(df_train['ApplicantIncome']>high_AI, high_AI, df_train['ApplicantIncome'])
df_train['ApplicantIncome']=np.where(df_train['ApplicantIncome']<low_AI, low_AI, df_train['ApplicantIncome'])

low_CI, high_CI=remove_outlier(df_train['CoapplicantIncome'])
df_train['CoapplicantIncome']=np.where(df_train['CoapplicantIncome']>high_CI, high_CI, df_train['CoapplicantIncome'])
df_train['CoapplicantIncome']=np.where(df_train['CoapplicantIncome']<low_CI, low_CI, df_train['CoapplicantIncome'])

low_LAT, high_LAT=remove_outlier(df_train['Loan_Amount_Term'])
df_train['Loan_Amount_Term']=np.where(df_train['Loan_Amount_Term']>high_LAT, high_LAT, df_train['Loan_Amount_Term'])
df_train['Loan_Amount_Term']=np.where(df_train['Loan_Amount_Term']<low_LAT, low_LAT, df_train['Loan_Amount_Term'])

df_train.boxplot(column=['ApplicantIncome'])
plt.show()

df_train.boxplot(column=['CoapplicantIncome'])
plt.show()

df_train.boxplot(column=['Loan_Amount_Term'])
plt.show()

df_train.isnull().sum()

df_train['Loan_Amount_Term'].fillna(360, inplace = True)

table = df_train.pivot_table(values='LoanAmount', index='Self_Employed' ,columns='Education', aggfunc=np.median)
table

def val(x):
 return table.loc[x['Self_Employed'],x['Education']]

df_train['LoanAmount'].fillna(df_train[df_train['LoanAmount'].isnull()].apply(val, axis=1), inplace=True)

df_train['Total_income']=df_train['ApplicantIncome']+df_train['CoapplicantIncome']

df_train.head()

df=df_train

label_encoder = preprocessing.LabelEncoder()
df['Gender']= label_encoder.fit_transform(df['Gender'])

df

df['Married']= label_encoder.fit_transform(df['Married'])
df['Education']= label_encoder.fit_transform(df['Education'])
df['Self_Employed']= label_encoder.fit_transform(df['Self_Employed'])
df['Property_Area']= label_encoder.fit_transform(df['Property_Area'])
df['Dependents']= label_encoder.fit_transform(df['Dependents'])

df.head()

x=df_train[['Gender','Married','Dependents','Education','Self_Employed', 'LoanAmount','Loan_Amount_Term','Credit_History','Property_Area', 'Total_income']]

y=df_train[['Loan_Status']]

from sklearn.model_selection import train_test_split
x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.2, random_state=4)

"""LOGISTIC REGRESSION"""

from sklearn.metrics import classification_report, confusion_matrix
import itertools
def plot_confusion_matrix(cm, classes,
                          normalize=False,
                          title='Confusion matrix',
                          cmap=plt.cm.Blues):
    """
    This function prints and plots the confusion matrix.
    Normalization can be applied by setting `normalize=True`.
    """
    if normalize:
        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]
        print("Normalized confusion matrix")
    else:
        print('Confusion matrix, without normalization')

    print(cm)

    plt.imshow(cm, interpolation='nearest', cmap=cmap)
    plt.title(title)
    plt.colorbar()
    tick_marks = np.arange(len(classes))
    plt.xticks(tick_marks, classes, rotation=45)
    plt.yticks(tick_marks, classes)
    fmt = '.2f' if normalize else 'd'
    thresh = cm.max() / 2.
    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):
        plt.text(j, i, format(cm[i, j], fmt),
                 horizontalalignment="center",
                 color="white" if cm[i, j] > thresh else "black")

    plt.tight_layout()
    plt.ylabel('True label')
    plt.xlabel('Predicted label')

from sklearn.model_selection import GridSearchCV,RandomizedSearchCV
from sklearn.linear_model import LogisticRegression
#from sklearn.metrics import confusion_matrix
parametersLR={ 'penalty' : ['l1', 'l2', 'elasticnet', 'none'],
              'C': [1, 0.5, 0.1, 0.01],
              'fit_intercept': [True, False],
              'solver' : ['newton-cg', 'lbfgs', 'liblinear', 'sag', 'saga'],
              'random_state':[10, 50, 100, 'none']
}
LR = LogisticRegression()
#r = RandomizedSearchCV(LR,parametersLR)
g=GridSearchCV(LR, parametersLR)
g.fit(x_train, y_train)

ypred = g.predict(x_test)
ypred

print (classification_report(y_test, ypred))

l = {'Gender': [1], 
       'Married': [0],
       'Dependents':[0],
       'Education':[0],
       'Self_Employed':[0],
       'LoanAmount':[130],
       'Loan_Amount_Term':[360],
       'Credit_History':[1],
       'Property_Area':[2],
       'Total_income':[5849]
       }
df=pd.DataFrame(l)
ans = g.predict(df)
ans2 = ans.tolist()
ans2[0]
df

def pred(Gender, Marital_Status, Dependents, Education, Self_Employed, Loan_Amount, Credit_History, Property_Area, Total_Income):
  if Gender == "Male":
    gen=1
  elif Gender =="Female":
    gen=0
  if Marital_Status=="Married":
    m=1
  elif Marital_Status=="Unmarried":
    m=0
  if Dependents=="0":
    d=0
  elif Dependents=="1":
    d=1
  elif Dependents=="2":
    d=2
  elif Dependents=="3+":
    d=3
  if Education=="Educated":
    e=1
  elif Education == "Uneducated":
    e=0
  if Self_Employed=="Yes":
    se=1
  elif Self_Employed=="No":
    se=0
  if Credit_History=="1":
    ch=1
  elif Credit_History=="0":
    ch=0
  if Property_Area=="0":
    pa=0
  elif Property_Area=="1":
    pa=1
  elif Propert_Area=="2":
    pa=2

  l = {'Gender': [gen], 
       'Married': [m],
       'Dependents':[d],
       'Education':[e],
       'Self_Employed':[se],
       'LoanAmount':[Loan_Amount],
       'Loan_Amount_Term':[360],
       'Credit_History':[ch],
       'Property_Area':[pa],
       'Total_income':[Total_Income]
       }
  df=pd.DataFrame(l)
  ans = g.predict(df)
  ans2 = ans.tolist()
  if ans2[0]=="Y":
    return "Loan Status: Approved!" 
  elif ans2[0]=="N":
    return "Loan Status: Disapproved"

iface = gr.Interface(
  fn=pred, 
  inputs=[gr.inputs.Radio(["Male", "Female"]), gr.inputs.Radio(["Married", "Unmarried"]),gr.inputs.Radio(["0", "1","2", "3+"]), gr.inputs.Radio(["Educated", "Uneducated"]), gr.inputs.Radio(["Yes", "No"]), "text", gr.inputs.Radio(["1", "0"]), gr.inputs.Radio(["0", "1", "2"]), "text"],
  outputs="text")
iface.launch(inline=False)