app.py

import gradio as gr
import tensorflow as tf
import joblib
import numpy as np
import zipfile
import os
from huggingface_hub import hf_hub_download

# Hugging Face repository ID
repo_id = "CaxtonEmeraldS/CholesterolConcentrationPredictor"  # Replace with your actual repo name

# Unzip models only once
unzip_dir = "unzipped_models"
if not os.path.exists(unzip_dir):
    print("Downloading and extracting model zip file...")
    zip_path = hf_hub_download(repo_id=os.path.join('spaces', repo_id), filename="Models.zip")  # Replace with your actual uploaded ZIP filename
    with zipfile.ZipFile(zip_path, 'r') as zip_ref:
        zip_ref.extractall(unzip_dir)
    print("Extraction complete.")

# Load linear models
# linear_rgb_path = os.path.join(unzip_dir, "linear_models/linear_rgb.joblib")
# linear_grey_path = os.path.join(unzip_dir, "linear_models/linear_grey.joblib")

# linear_rgb = joblib.load(linear_rgb_path)
# linear_grey = joblib.load(linear_grey_path)

def predict(r, g, b, activation, seed, neurons):
    try:
        X = np.array([[r, g, b]])
        # grey = 0.2989 * r + 0.5870 * g + 0.1140 * b

        # # Linear predictions
        # lin_pred_rgb = linear_rgb.predict(X)[0]
        # lin_pred_grey = linear_grey.predict([[grey]])[0]

        # Load corresponding ANN model
        keras_path = os.path.join(unzip_dir, f"{activation}/seed_{seed}/model_{neurons}.keras")
        if not os.path.exists(keras_path):
            raise FileNotFoundError(f"Model not found: {keras_path}")
        
        model = tf.keras.models.load_model(keras_path)
        ann_pred = model.predict(X)[0][0]

        return ann_pred, lin_pred_rgb, lin_pred_grey
    
    except Exception as e:
        return f"Error: {str(e)}", "", ""

iface = gr.Interface(
    fn=predict,
    inputs=[
        gr.Number(label="R"), 
        gr.Number(label="G"), 
        gr.Number(label="B"),
        gr.Textbox(label="Activation (folder name)"),
        gr.Number(label="Seed (folder name)"),
        gr.Number(label="Neurons (model number)")
    ],
    outputs=[
        gr.Text(label="ANN Model Prediction"),
        gr.Text(label="Linear RGB Prediction"),
        gr.Text(label="Linear Grey Prediction"),
    ],
    title="ANN vs Linear Model Predictor",
    description="Dynamically load models from Hugging Face repo and predict."
)

if __name__ == "__main__":
    iface.launch()