End to end machine learning model deployment using flask

machine_learning_deployment_11.py

# Data partitioning

# Unique values of Loan_Status
df_concat['Loan_Status'].value_counts()

# Training set
df_train = df_concat[df_concat['Loan_Status'].isin([0, 1])].reset_index(drop = True)
print('Dimension data: {} rows and {} columns'.format(len(df_train), len(df_train.columns)))
df_train.head()

# Testing set
df_test = df_concat[df_concat['Loan_Status'].isin([999])].reset_index(drop = True)
print('Data dimension: {} rows and {} columns'.format(len(df_test), len(df_test.columns)))
df_test.head()

# Data partitioning >>> training set into training and validation
df_train_final = df_train.reset_index(drop = True)
X = df_train_final[df_train_final.columns[~df_train_final.columns.isin(['Loan_Status'])]]
y = df_train_final['Loan_Status']

# Training = 70% and validation = 30%
X_train, X_val, y_train, y_val = train_test_split(X , y, test_size = 0.3, random_state = 42)
print('Data dimension of training set   :', X_train.shape)
print('Data dimension of validation set :', X_val.shape)

# Testing set
X_test = df_test[df_test.columns[~df_test.columns.isin(['Loan_Status'])]]
print('Data dimension of testing set    :', X_test.shape)