import time
import streamlit as st
import numpy as np
from PIL import Image
import urllib.request
import io
from utils import *

# Initialize labels and model
labels = gen_labels()
model = model_arc()  # Assuming this function initializes and returns a trained model

# Streamlit UI
st.markdown('''
    <div style="padding-bottom: 20px; padding-top: 20px; padding-left: 5px; padding-right: 5px">
    <center><h1>Garbage Segregation</h1></center>
    </div>
''', unsafe_allow_html=True)

st.markdown('''
    <div>
    <center><h3>Please upload Waste Image to find its Category</h3></center>
    </div>
''', unsafe_allow_html=True)

opt = st.selectbox("How do you want to upload the image for classification?", 
                   ('Please Select', 'Upload image via link', 'Upload image from device'))

# Image processing based on user selection
image = None
if opt == 'Upload image from device':
    file = st.file_uploader('Select', type=['jpg', 'png', 'jpeg'])
    if file:
        try:
            image = Image.open(io.BytesIO(file.read())).resize((256, 256), Image.LANCZOS)
        except Exception as e:
            st.error(f"Error reading the file: {e}")

elif opt == 'Upload image via link':
    img_url = st.text_input('Enter the Image Address')
    if st.button('Submit'):
        try:
            response = urllib.request.urlopen(img_url)
            image = Image.open(response).resize((256, 256), Image.LANCZOS)
        except ValueError:
            st.error("Please Enter a valid Image Address!")

try:
  if image is not None:
    st.image(image, width = 300, caption = 'Uploaded Image')
    if st.button('Predict'):
        img = preprocess(image)

        model = model_arc()
        #model.load_weights("classify_model.h5")

        prediction = model.predict(img[np.newaxis, ...])
        st.info('Hey! The uploaded image has been classified as " {} waste " '.format(labels[np.argmax(prediction[0], axis=-1)]))
except Exception as e:
  st.info(e)
  pass