import joblib
import pandas as pd
from sklearn.preprocessing import LabelEncoder

# Load the trained model
model = joblib.load('time_slot_model.joblib')

# Load the LabelEncoder used in training (if applicable)
# label_encoder = joblib.load('label_encoder.joblib')  # Uncomment if you used a LabelEncoder

def preprocess_input(user_pref):
    # Example preprocessing; adjust based on your actual preprocessing steps
    user_input = pd.get_dummies(pd.Series([user_pref]), prefix='Pref')
    return user_input

def predict(user_pref):
    # Preprocess the input
    user_input = preprocess_input(user_pref)
    
    # Ensure the input has the same columns as the training data
    user_input = user_input.reindex(columns=model.feature_importances_, fill_value=0)
    
    # Make a prediction
    prediction = model.predict(user_input)
    return prediction[0]

# Example usage:
# user_pref = 'noon'  # Replace with actual user preference
# prediction = predict(user_pref)
# print(f"Predicted next rescheduled time: {prediction}")