import numpy as np
import tensorflow as tf

from PIL import Image
from tensorflow.keras.models import load_model
from tensorflow.keras.preprocessing.image import img_to_array
import gradio as gr

# Load the Keras model
model = load_model('model.keras')
classnames = np.array(['Normal', 'Pneumonia'])

# Define image preprocessing
def transform(img):
    img = img.resize((224, 224))  # Resize to match the input size expected by the model
    img_array = img_to_array(img)  # Convert PIL image to numpy array
    img_array = img_array / 255.0  # Normalize pixel values
    img_array = np.expand_dims(img_array, axis=0)  # Add batch dimension
    return img_array

# Define the classify function
def classify(path):
    img = Image.open(path)
    processed_img = transform(img)
    result = model.predict(processed_img)[0]
    index = np.argmax(result)  # Get the index of the highest probability
    predict = str(classnames[index])
    return predict

# Create the Gradio interface
with gr.Blocks(theme=gr.themes.Default(primary_hue=gr.themes.colors.blue, secondary_hue=gr.themes.colors.blue, neutral_hue=gr.themes.colors.zinc)) as demo:
    img_path = gr.Image(label="Input Image", type="filepath", height=512, width=512)
    output = gr.Textbox(label="Output")
    clear_btn = gr.ClearButton([img_path, output], variant="stop")
    img_path.upload(classify, inputs=img_path, outputs=output)
demo.launch()