End to end machine learning model deployment using flask

machine_learning_deployment_6.py

# Number of customers by loan status and the dependents

# Data aggregation between loan status and dependents
df_viz_2 = df_train.groupby(['Loan_Status', 'Dependents'])['Loan_ID'].count().reset_index(name = 'Total')
# Map the loan status
df_viz_2['Loan_Status'] = df_viz_2['Loan_Status'].map(
    {
        0: 'Not default',
        1: 'Default'
    }
)

# Data viz
plotnine.options.figure_size = (8, 4.8)
(
    ggplot(
        data = df_viz_2
    )+
    geom_bar(
        aes(
            x = 'Dependents',
            y = 'Total',
            fill = 'Loan_Status'
        ),
        stat = 'identity',
        position = 'fill',
        width = 0.5
    )+
    labs(
        title = 'The composition of loan status by the dependents',
        fill = 'Loan status'
    )+
    xlab(
        'Dependents'
    )+
    ylab(
        'Frequency'
    )+
    scale_x_discrete(
        limits = ['0', '1', '2', '3+']
    )+
    scale_fill_manual(
        values = ['#981220','#80797c'],
        labels = ['Default', 'Not Default']
    )+
    theme_minimal()
)