risk_model.py

import os
import joblib
import numpy as np
import pandas as pd
from train_model import train_and_save_model, MODEL_PATH, LABEL_ENCODER_PATH

if not (os.path.exists(MODEL_PATH) and os.path.exists(LABEL_ENCODER_PATH)):
    train_and_save_model()

model = joblib.load(MODEL_PATH)
le = joblib.load(LABEL_ENCODER_PATH)

def predict_risk(temperature, duration):
    # Create a DataFrame with the same feature names used in training
    input_df = pd.DataFrame([[temperature, duration]], columns=['Max_Temperature', 'Duration'])
    pred_encoded = model.predict(input_df)[0]
    pred_proba = model.predict_proba(input_df)[0]
    pred_label = le.inverse_transform([pred_encoded])[0]
    confidence = pred_proba[pred_encoded] * 100
    return pred_label, confidence