model_predict.py

from input_schemas import ModelInput
import pandas as pd
import numpy as np
from sklearn.preprocessing import StandardScaler, LabelEncoder
import joblib  # or pickle

# Load your trained Ridge model
model = joblib.load("./ridge_best_model.pkl")

def predict_output(input: ModelInput):
    df = pd.DataFrame([input.dict()])

    categorical_cols = df.select_dtypes(include=['object']).columns.tolist()
    bool_cols = df.select_dtypes(include=['bool']).columns.tolist()

    for col in categorical_cols:
        le = LabelEncoder()
        df[col] = le.fit_transform(df[col])

    df[bool_cols] = df[bool_cols].astype(int)

    numeric_cols = df.select_dtypes(include=[np.number]).columns
    scaler = StandardScaler()
    df[numeric_cols] = scaler.fit_transform(df[numeric_cols])

    prediction = model.predict(df)

    return float(prediction[0])