app.py

import pandas as pd
import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt
from sklearn.model_selection import train_test_split
from sklearn.neighbors import KNeighborsClassifier
import gradio as gr
# Load data
nexus_bank = pd.read_csv('nexus_bank_dataa.csv')
# Preprocessing
X = nexus_bank[['salary', 'dependents']]
y = nexus_bank['defaulter']
# Train-test split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.15, random_state=90)
# Model training
knn_classifier = KNeighborsClassifier()
knn_classifier.fit(X_train, y_train)
# Prediction function
def predict_defaulter(salary, dependents):
    input_data = [[salary, dependents]]
    knn_predict = knn_classifier.predict(input_data)
    return "Yes! It's a Defaulter" if knn_predict[0] == 1 else "No! It's not a Defaulter"
# Interface
interface = gr.Interface(
    fn=predict_defaulter,
    inputs=["number", "number"],
    outputs="text",
    title="Defaulter Prediction",
    description="This app predicts whether an individual is likely to default based on their salary and number of dependents. Input the respective values and get instant predictions."
)
# Launch the interface
interface.launch()