streamlit_app.py

import streamlit as st
from PIL import Image
import os
import shutil

# ------------------------------
# Streamlit Config (must be first)
# ------------------------------
st.set_page_config(page_title="Image Categorization Demo", layout="wide")

# ------------------------------
# Load YOLO classification model
# ------------------------------
@st.cache_resource
def load_model():
    from ultralytics import YOLO
    model = YOLO("yolov8m-cls.pt")  # replace with your trained model
    return model

model = load_model()

# ------------------------------
# Helper: manage temp folder
# ------------------------------
TEMP_FOLDER = "temfolder"

def prepare_temp_folder():
    if os.path.exists(TEMP_FOLDER):
        shutil.rmtree(TEMP_FOLDER)
    os.makedirs(TEMP_FOLDER)

def cleanup_temp_folder():
    if os.path.exists(TEMP_FOLDER):
        shutil.rmtree(TEMP_FOLDER)

# ------------------------------
# Streamlit UI
# ------------------------------
st.title("Image Categorization Demo")

with st.form("upload_form", clear_on_submit=True):
    uploaded_files = st.file_uploader(
        "Upload one or more images", 
        type=["jpg", "jpeg", "png"], 
        accept_multiple_files=True
    )

    col1, col2 = st.columns([1, 1])
    submit = col1.form_submit_button("🚀 Submit for Classification")
    refresh = col2.form_submit_button("🔄 Refresh")

if refresh:
    cleanup_temp_folder()
    st.success("🧹 Uploads cleared!")

if submit:
    if not uploaded_files:
        st.warning("⚠️ Please upload at least one image before submitting.")
    else:
        total_files = len(uploaded_files)
        st.write(f"🔍 Classifying **{total_files}** images...")

        # Prepare clean folder
        prepare_temp_folder()

        results_by_class = {}

        progress = st.progress(0)          # progress bar
        status_text = st.empty()           # placeholder for progress text

        for idx, file in enumerate(uploaded_files, start=1):
            # Save uploaded file into temfolder
            img_path = os.path.join(TEMP_FOLDER, file.name)
            with open(img_path, "wb") as f:
                f.write(file.read())

            # Run YOLO classification
            results = model(img_path)
            pred_class = results[0].names[results[0].probs.top1]

            # Group images by predicted class
            if pred_class not in results_by_class:
                results_by_class[pred_class] = []
            results_by_class[pred_class].append(img_path)

            # Update progress bar + text
            percent = int((idx / total_files) * 100)
            progress.progress(idx / total_files)
            status_text.text(f"Processing {idx}/{total_files} images ({percent}%)")

        st.success("✅ Classification complete!")

        # ------------------------------
        # Show gallery grouped by class
        # ------------------------------
        for cls, img_list in results_by_class.items():
            st.subheader(f"📂 Category: **{cls}** ({len(img_list)})")
            cols = st.columns(4)  # show 4 images per row
            for i, img_path in enumerate(img_list):
                with cols[i % 4]:
                    st.image(Image.open(img_path), use_column_width=True)

        # Cleanup after displaying
        cleanup_temp_folder()