Update app.py

app.py

@@ -14,23 +14,6 @@ from array import *
 
 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)
@@ -38,17 +21,6 @@ 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)
@@ -70,21 +42,11 @@ 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']]
@@ -100,8 +62,6 @@ 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'])
@@ -169,26 +129,6 @@ 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":
@@ -243,6 +183,6 @@ def pred(Gender, Marital_Status, Dependents, Education, Self_Employed, Loan_Amou
 
 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"],
+  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(["Urban", "Semi Urban", "Rural"]), "text"],
   outputs="text")
 iface.launch(inline=False)