import random

import gradio as gr
import joblib
import numpy as np
import xgboost

model = joblib.load("xgb.pkl")


def predict(*args):
    input_data = []

    for i in args:
        input_data.append(float(i))

    input_data = np.asarray(input_data)
    # reshape the array as we are predicting for one instance
    input_data_reshaped = input_data.reshape(1, -1)

    prediction = model.predict(input_data_reshaped)

    if prediction[0] == 0:
        return "The Credit Score is Good"
    elif prediction[0] == 1:
        return "The Credit Score is Poor"
    else:
        return "The Credit Score is Standard"


with gr.Blocks() as app:
    gr.Markdown(
        """
    **Credit Score Classification**"""
    )
    with gr.Row():
        with gr.Column():
            Annual_Income = gr.TextArea(label="Annual Income")
            Monthly_Inhand_Salary = gr.TextArea(label="Monthly Inhand Salary")
            Interest_Rate = gr.TextArea(label="Interest Rate")
            Num_of_Loan = gr.Slider(
                label="Number of Loans", minimum=1, maximum=10, step=1, randomize=True
            )
            Delay_from_due_date = gr.Slider(
                label="Number of Delayed Days",
                minimum=-100,
                maximum=100,
                step=1,
                randomize=True,
            )
            Num_of_Delayed_Payment = gr.Slider(
                label="Number of Delayed Payments",
                minimum=1,
                maximum=10,
                step=1,
                randomize=True,
            )
            Credit_Mix = gr.Dropdown(
                label="Credit Mix (Bad: 0, Don't Have: 1, Good: 2, Standard: 3)",
                choices=[0, 1, 2, 3],
                value=lambda: random.choice([0, 1, 2, 3]),
            )
            Outstanding_Debt = gr.TextArea(label="Outstanding Debt")
            Credit_Utilization_Ratio = gr.TextArea(label="Credit Utilization Ratio")
            Payment_of_Min_Amount = gr.Dropdown(
                label="Payment of Minimum Amount (NM: 0, No: 1, Yes: 2)",
                choices=[0, 1, 2],
                value=lambda: random.choice([0, 1, 2]),
            )
            Total_EMI_per_month = gr.TextArea(label="Total Equated Monthly Installment")
            Amount_invested_monthly = gr.TextArea(label="Amount Invested Monthly")
            Monthly_Balance = gr.TextArea(label="Monthly Balance")
            Credit_History_Age_In_Years = gr.TextArea(label="Credit History in Years")
            StudentLoan = gr.Dropdown(
                label="Student Loan (Don't Have: 0, Have: 1)",
                choices=[0, 1],
                value=lambda: random.choice([0, 1]),
            )
            MortgageLoan = gr.Dropdown(
                label="Mortage Loan (Don't Have: 0, Have: 1)",
                choices=[0, 1],
                value=lambda: random.choice([0, 1]),
            )
            PersonalLoan = gr.Dropdown(
                label="Personal Loan (Don't Have: 0, Have: 1)",
                choices=[0, 1],
                value=lambda: random.choice([0, 1]),
            )
            DebtConsolidationLoan = gr.Dropdown(
                label="Debt Consolidation Loan (Don't Have: 0, Have: 1)",
                choices=[0, 1],
                value=lambda: random.choice([0, 1]),
            )
            Credit_BuilderLoan = gr.Dropdown(
                label="Credit Builder Loan (Don't Have: 0, Have: 1)",
                choices=[0, 1],
                value=lambda: random.choice([0, 1]),
            )
            HomeEquityLoan = gr.Dropdown(
                label="Home Equity Loan (Don't Have: 0, Have: 1)",
                choices=[0, 1],
                value=lambda: random.choice([0, 1]),
            )
            NotSpecified = gr.Dropdown(
                label="Unspecified Loan (Don't Have: 0, Have: 1)",
                choices=[0, 1],
                value=lambda: random.choice([0, 1]),
            )
            AutoLoan = gr.Dropdown(
                label="Auto Loan (Don't Have: 0, Have: 1)",
                choices=[0, 1],
                value=lambda: random.choice([0, 1]),
            )
            PaydayLoan = gr.Dropdown(
                label="Payday Loan (Don't Have: 0, Have: 1)",
                choices=[0, 1],
                value=lambda: random.choice([0, 1]),
            )
        with gr.Column():
            label = gr.Label()
            with gr.Row():
                predict_btn = gr.Button(value="Predict")
            predict_btn.click(
                predict,
                inputs=[
                    Annual_Income,
                    Monthly_Inhand_Salary,
                    Interest_Rate,
                    Num_of_Loan,
                    Delay_from_due_date,
                    Num_of_Delayed_Payment,
                    Credit_Mix,
                    Outstanding_Debt,
                    Credit_Utilization_Ratio,
                    Payment_of_Min_Amount,
                    Total_EMI_per_month,
                    Amount_invested_monthly,
                    Monthly_Balance,
                    Credit_History_Age_In_Years,
                    StudentLoan,
                    MortgageLoan,
                    PersonalLoan,
                    DebtConsolidationLoan,
                    Credit_BuilderLoan,
                    HomeEquityLoan,
                    NotSpecified,
                    AutoLoan,
                    PaydayLoan,
                ],
                outputs=[label],
            )

app.launch()