app.py

import streamlit as st
from transformers import Qwen2VLForConditionalGeneration, AutoProcessor
from qwen_vl_utils import process_vision_info
import torch
from PIL import Image
import numpy as np
from datetime import datetime
import os

# Function to convert numpy array to image and save it
def array_to_image_path(image_array):
    img = Image.fromarray(np.uint8(image_array))
    timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
    filename = f"image_{timestamp}.png"
    img.save(filename)
    full_path = os.path.abspath(filename)
    return full_path

# Model and processor initialization
model = Qwen2VLForConditionalGeneration.from_pretrained("Qwen/Qwen2-VL-2B-Instruct", trust_remote_code=True, torch_dtype="auto").eval()
processor = AutoProcessor.from_pretrained("Qwen/Qwen2-VL-2B-Instruct", trust_remote_code=True)

user_prompt = '<|user|>\n'
assistant_prompt = '<|assistant|>\n'
prompt_suffix = "<|end|>\n"

# Streamlit App
st.title("Qwen2-VL-2B Demo")

# Upload image
uploaded_file = st.file_uploader("Upload an image...", type=["jpg", "png"])
if uploaded_file is not None:
    image = Image.open(uploaded_file).convert("RGB")
    st.image(image, caption="Uploaded Image", use_column_width=True)

    # User input
    text_input = st.text_input("Enter your question:")

    if st.button("Generate"):
        image_path = array_to_image_path(np.array(image))
        prompt = f"{user_prompt}<|image_1|>\n{text_input}{prompt_suffix}{assistant_prompt}"

        messages = [
            {
                "role": "user",
                "content": [
                    {"type": "image", "image": image_path},
                    {"type": "text", "text": text_input},
                ],
            }
        ]

        # Prepare inputs for inference
        text = processor.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
        image_inputs, video_inputs = process_vision_info(messages)
        inputs = processor(text=[text], images=image_inputs, videos=video_inputs, padding=True, return_tensors="pt")
        inputs = inputs.to("cpu")


        # Model inference
        generated_ids = model.generate(**inputs, max_new_tokens=128)
        generated_ids_trimmed = [out_ids[len(in_ids):] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)]
        output_text = processor.batch_decode(generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False)

        # Display the generated output
        st.write("Generated Response:", output_text[0])