app.py

import gradio as gr
from tensorflow.keras.models import load_model
import cv2
import numpy as np

# Load the trained model
model = load_model("classifier-resnet.keras")

def preprocess_image(image):
    """Preprocess the input image for model prediction"""
    img = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)  # Convert to RGB
    img = cv2.resize(img, (256, 256))  # Resize to match model input shape
    img = img / 255.0  # Normalize pixel values (0-1)
    img = np.expand_dims(img, axis=0)  # Add batch dimension
    return img

def predict(image):
    """Predict if the image contains a brain tumor"""
    img = preprocess_image(image)
    pred = model.predict(img)
    result = "🧠 Tumor Detected" if pred[0][0] > 0.5 else "✅ No Tumor Detected"
    return result

# Create Gradio Interface
interface = gr.Interface(
    fn=predict, 
    inputs=gr.Image(type="numpy"), 
    outputs="text",
    title="Brain Tumor Detection",
    description="Upload an MRI scan to detect brain tumors."
)

# Launch the Gradio app
interface.launch()