import tensorflow as tf
from tensorflow.keras.models import load_model
import numpy as np
import cv2
import os
from flask import Flask, request, jsonify
from werkzeug.utils import secure_filename

# Initialize Flask app
app = Flask(_name_)

# Load the trained Keras model
MODEL_PATH = "model.weights.h5"
model = load_model(MODEL_PATH)

# Define allowed extensions for image upload
ALLOWED_EXTENSIONS = {'png', 'jpg', 'jpeg'}

def allowed_file(filename):
    return '.' in filename and filename.rsplit('.', 1)[1].lower() in ALLOWED_EXTENSIONS

def preprocess_image(image_path):
    """ Preprocess the uploaded image to match the model's input shape """
    img = cv2.imread(image_path)
    img = cv2.resize(img, (224, 224))  # Resize to model's expected input size
    img = img / 255.0  # Normalize pixel values
    img = np.expand_dims(img, axis=0)  # Add batch dimension
    return img

@app.route('/predict', methods=['POST'])
def predict():
    """ API endpoint to predict melanoma from an uploaded image """
    if 'file' not in request.files:
        return jsonify({"error": "No file uploaded"}), 400

    file = request.files['file']
    if file.filename == '':
        return jsonify({"error": "No selected file"}), 400

    if file and allowed_file(file.filename):
        filename = secure_filename(file.filename)
        file_path = os.path.join("uploads", filename)
        file.save(file_path)

        # Preprocess and predict
        image = preprocess_image(file_path)
        prediction = model.predict(image)
        os.remove(file_path)  # Remove the file after prediction

        # Assuming model outputs a probability (0 = not melanoma, 1 = melanoma)
        result = "Melanoma" if prediction[0][0] > 0.5 else "Not Melanoma"

        return jsonify({"prediction": result, "confidence": float(prediction[0][0])})

    return jsonify({"error": "Invalid file format"}), 400

if _name_ == '_main_':
    os.makedirs("uploads", exist_ok=True)  # Ensure upload directory exists
    app.run(host='0.0.0.0', port=5000, debug=True)