import streamlit as st
import pickle
import pandas as pd
import numpy as np
from PIL import Image

with open('model.pkl', 'rb') as model:
    model = pickle.load(model)

def run():
  
    #Title
    st.title('🔍 Forest Fire & Smoke Prediction')

    images_by_class = {
        "": [
            "fire/fire1.jpg",
            "non-fire/non-fire1.png",
            "smoke/smoke4.png"
        ]
    }

    # Display in grid
    for cls, image_paths in images_by_class.items():
        st.subheader(cls.capitalize())
        cols = st.columns(len(image_paths))
        for i, path in enumerate(image_paths):
            img = Image.open(path)
            with cols[i]:
                st.image(img, use_container_width=True)

    # Upload image
    uploaded_file = st.file_uploader(
        "Upload your own image",
        type=["jpg", "jpeg", "png"],
        label_visibility="visible"
    )

    # Upload file
    if uploaded_file is not None:
        # Show image
        img = Image.open(uploaded_file).convert('RGB')
        st.image(img, caption="Uploaded Image", use_container_width=True)

        # Preprocess
        img_resized = img.resize((150, 150))
        img_array = np.expand_dims(np.array(img_resized) / 255.0, axis=0)

        # Load model
        model = tf.keras.models.load_model("best_model.h5")
        class_names = ["fire", "non-fire", "smoke"]

        # # Predict
        # prediction = model.predict(img_array)
        # predicted_label = class_names[np.argmax(prediction)]
        # confidence = np.max(prediction)

        # # Show Prediction
        # st.success(f"✅ Prediction: **{predicted_label}**")
        # st.write(f"Confidence: **{confidence:.2%}**")

if __name__ == '__main__':
    run()