app.py

import torch
import torchvision.transforms as T
from PIL import Image
from transformers import AutoModel, AutoTokenizer
import gradio as gr
import logging

# Setup logging
logging.basicConfig(level=logging.INFO, format="%(asctime)s - %(levelname)s - %(message)s")

# Device Configuration
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

# ImageNet normalization values
IMAGENET_MEAN = (0.485, 0.456, 0.406)
IMAGENET_STD = (0.229, 0.224, 0.225)

def build_transform(input_size):
    """Build preprocessing pipeline for images."""
    transform = T.Compose([
        T.Lambda(lambda img: img.convert("RGB") if img.mode != "RGB" else img),
        T.Resize((input_size, input_size), interpolation=T.InterpolationMode.BICUBIC),
        T.ToTensor(),
        T.Normalize(mean=IMAGENET_MEAN, std=IMAGENET_STD),
    ])
    return transform

def preprocess_image(image, input_size=448):
    """Preprocess the image to the required format."""
    transform = build_transform(input_size)
    tensor_image = transform(image).unsqueeze(0).to(torch.float32 if device == "cpu" else torch.bfloat16).to(device)
    return tensor_image

# Load the model and tokenizer
logging.info("Loading model from Hugging Face Hub...")
model_path = "OpenGVLab/InternVL2_5-1B"
model = AutoModel.from_pretrained(
    model_path,
    torch_dtype=torch.bfloat16 if torch.cuda.is_available() else torch.float32,
    trust_remote_code=True,
).to(device).eval()

tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True, use_fast=False)

# Add the `<image>` token if missing
if "<image>" not in tokenizer.get_vocab():
    tokenizer.add_tokens(["<image>"])
    model.resize_token_embeddings(len(tokenizer))  # Resize model embeddings

assert "<image>" in tokenizer.get_vocab(), "Error: `<image>` token is missing from tokenizer vocabulary."

def describe_image(image):
    """Generate a description for the uploaded image."""
    try:
        pixel_values = preprocess_image(image, input_size=448)
        prompt = "<image>\nExtract text from the image, respond with only the extracted text."

        response = model.chat(
            tokenizer=tokenizer,
            pixel_values=pixel_values,
            question=prompt,
            history=None,
            return_history=False,
            generation_config=dict(max_new_tokens=512, do_sample=True)
        )
        return response
    except Exception as e:
        logging.error(f"Error during processing: {e}")
        return f"Error: {e}"

# Gradio Interface
interface = gr.Interface(
    fn=describe_image,
    inputs=gr.Image(type="pil"),
    outputs=gr.Textbox(label="Extracted Text", lines=10, interactive=False),
    title="Image to Text",
    description="Upload an image to extract text using the pretrained model.",
)

if __name__ == "__main__":
    interface.launch(server_name="0.0.0.0", server_port=7860)