import gradio as gr
from PIL import Image
import io
import numpy as np
import tensorflow as tf
from tensorflow.keras.applications.resnet50 import ResNet50
from tensorflow.keras.preprocessing.image import img_to_array
from tensorflow.keras.applications.resnet50 import preprocess_input, decode_predictions

# Load the ResNet50 model
model = ResNet50(weights='imagenet')

def process_image(image):
    try:
        if image is None:
            return "No image provided."

        # Convert Gradio image to a PIL Image
        img = Image.fromarray(image)

        # Resize the image
        img = img.resize((224, 224))  # ResNet50 input size

        # Convert to numpy array
        x = img_to_array(img)

        # Expand dimensions
        x = np.expand_dims(x, axis=0)

        # Normalize input
        x = preprocess_input(x)

        return x

    except Exception as e:
        error_message = f"Error processing image: {str(e)}"
        print(error_message)  # Print full error message for debugging
        return error_message

def extract_carbs_from_response(response):
    return 30.0  # Placeholder for carb estimate

def calculate_insulin_dose(current_insulin_level, carb_estimate):
    return current_insulin_level + (carb_estimate * 0.5)

def main():
    def insulin_dosage_predict(current_insulin_level, food_image):
        try:
            if current_insulin_level <= 0:
                return "Insulin level must be greater than zero."

            processed_image = process_image(food_image)
            if isinstance(processed_image, str):
                return processed_image

            # Get predictions (you can implement actual inference with the model)
            preds = model.predict(processed_image)
            decoded_preds = decode_predictions(preds, top=1)[0][0]
            food_item = decoded_preds[1]
            response = f"Predicted food item: {food_item}"

            carb_estimate = extract_carbs_from_response(response)
            insulin_dose = calculate_insulin_dose(current_insulin_level, carb_estimate)

            return f"""
            Nutritional Information:
            {response}

            Insulin Dose Recommendation:
            Current insulin level: {current_insulin_level} units
            Estimated carbs intake: {carb_estimate:.2f} grams
            Recommended insulin dose: {insulin_dose:.2f} units
            """
        except Exception as e:
            return f"An error occurred: {str(e)}"

    iface = gr.Interface(
        fn=insulin_dosage_predict,
        inputs=[
            gr.Number(label="Current Insulin Level"),
            gr.Image(label="Food Image"),
        ],
        outputs=gr.Textbox(),
        title="Insulin Dosage Calculator",
        description="Enter your current insulin level and upload an image of your food."
    )

    iface.launch()

if __name__ == "__main__":
    main()