import gradio as gr
import tensorflow as tf
from tensorflow.keras.models import load_model
import numpy as np
from tensorflow.keras.preprocessing import image


def predict_image(input_image, threshold=0.5):
    # Load and preprocess the input image
    img = image.load_img(input_image, target_size=(224, 224))
    img = image.img_to_array(img)
    img = np.expand_dims(img, axis=0)
    img = img / 255.0  # Normalize the pixel values (if the model expects it)

    # Make a prediction
    # Load the saved model
    loaded_model = load_model('tumor_model.h5')
    predictions = loaded_model.predict(img)

    # Assuming it's a binary classification model
    predicted_probability = predictions[0][0]

    # Determine the class based on the threshold
    if predicted_probability > threshold:
        class_name = 'tumor'
    else:
        class_name = 'no tumor'

    return f'Predicted Class: {class_name}'

iface = gr.Interface(fn=predict_image, inputs="image", outputs="text", live=True)
iface.launch(share=True)