app.py

import gradio as gr
import joblib
import numpy as np
# Load models
model_flavor = joblib.load("random_forest_model_flavor.pkl")
model_topping = joblib.load("random_forest_model_topping.pkl")
model_drink = joblib.load("random_forest_model_drink.pkl")

# Load encoders
encoder_flavor = joblib.load("encoder_flavor.pkl")
encoder_topping = joblib.load("encoder_topping.pkl")
encoder_drink = joblib.load("encoder_drink.pkl")
input_encoders = joblib.load("input_encoders.pkl")

# Prediction function
def predict_merienda(mood, weather, craving_level, last_meal, budget):
    features = [mood, weather, craving_level, last_meal, budget]
    encoded = [input_encoders[col].transform([val])[0] for col, val in zip(input_encoders.keys(), features)]
    encoded_np = np.array(encoded).reshape(1, -1)
    pred_flavor = encoder_flavor.inverse_transform(model_flavor.predict(encoded_np))[0]
    pred_topping = encoder_topping.inverse_transform(model_topping.predict(encoded_np))[0]
    pred_drink = encoder_drink.inverse_transform(model_drink.predict(encoded_np))[0]
    return pred_flavor, pred_topping, pred_drink

# Dropdown options
mood_list = input_encoders["mood"].classes_.tolist()
weather_list = input_encoders["weather"].classes_.tolist()
craving_list = input_encoders["craving_level"].classes_.tolist()
last_meal_list = input_encoders["last_meal"].classes_.tolist()
budget_list = input_encoders["budget"].classes_.tolist()
# Gradio Interface
iface = gr.Interface(
    fn=predict_merienda,
    inputs=[
        gr.Dropdown(mood_list, label="Mood"),
        gr.Dropdown(weather_list, label="Weather"),
        gr.Dropdown(craving_list, label="Craving Level"),
        gr.Dropdown(last_meal_list, label="Last Meal"),
        gr.Dropdown(budget_list, label="Budget"),
    ],
    outputs=[
        gr.Text(label="Recommended Flavor"),
        gr.Text(label="Recommended Topping"),
        gr.Text(label="Recommended Drink"),
    ],
    title=" Merienda Classifier",
    description="Predicts the best pancit canton flavor, topping, and drink based on your current situation"
    )

import gradio as gr

# Optional custom CSS for prettier background and fonts
custom_css = """
body {
    background: linear-gradient(to right, #fbc2eb, #a6c1ee);
    font-family: 'Poppins', sans-serif;
}
h1 {
    text-align: center;
    color: #6D214F;
}
"""

# Your function to classify merienda (replace with your actual model code)
def classify_merienda(image):
    return "Pancit"  # <-- Replace this with your prediction logic

# Build the app
with gr.Blocks(theme=gr.themes.Soft(), css=custom_css) as demo:
    gr.Markdown("# 🍽️ Welcome to the Merienda Classifier!")
    gr.Markdown("Upload a picture of your favorite **merienda** and let the AI guess what it is!")

    with gr.Row():
        with gr.Column():
            input_image = gr.Image(type="pil", label="Upload your Merienda")
            classify_button = gr.Button("Classify!")
        with gr.Column():
            output_label = gr.Label(label="Prediction")

    classify_button.click(classify_merienda, inputs=[input_image], outputs=[output_label])



iface.launch()